Peptides. Vol. 11. pp. 989-993.
0 Pergamon Press plc, 1990. Printed in the U.S.A.
0196-9781/90 $3.00 + .@I
Galanin in Porcine Vagal Sensory Nerves: Immunohistochemical and Immunochemical Study C. PHILIPPE,*’ J. C. CUBER,? A. BOSSHARD,t 0. RAMPIN,* J. P. LAPLACE” AND J. A. CHAYVIALLEt “Station de Physiologie de la Nutrition, INRA, 78350 Jouy-en-Josas, France tINSERM, U45, Hdpital E. Herriot, 49437 Lyon cedex 03, France Received
20 February
1990
0. RAMPIN, J. P. LAPLACE AND J. A. CHAYVIALLE. Galanin inporcine vagal study. PEPTIDES ll(5) 989-993, 1990.-In this work, the presence of galanin was examined by immunohistochemistry, radioimmunoassay and high performance liquid chromatography (HPLC) in porcine nodose ganglia, mainly constituted of cell bodies from the vagal sensory neurons. Galanin-like immunoreactivity (Gal-LI) was revealed in 10 to 15% of the total cell bodies by the indirect immunofluorescent technique of Coons. For comparison, a positive staining was revealed in a few cell bodies of the submucous plexus and in fibers located in the different layers of the ileum. The extractable Gal-L1 content in nodose ganglia was 7.2 kO.8 pmolig wet tissue, which represents a concentration about nine times lower than that found in the ileum. HPLC of extractable material revealed a predominant peak which coeluted with the synthetic peptide. We propose that, in pigs, galanin may play a role in the transmission of visceral information through the vagal afferences. PHILIPPE,
C., J. C. CUBER. A. BOSSHARD,
sensory nerves: lmmunohisrochemical
Galanin
Nodose ganglia
and immunochemical
Pig
Visceral afferences
METHOD
VAGAL sensory neurons relay information from receptors located in the gastrointestinal wall, lung and heart to the brain stem. Their
Materials
corresponding cell bodies are localized in the nodose ganglia. Several peptides, such as substance P, calcitonin gene-related peptide, cholecystokinin, vasoactive intestinal peptide, somatostatin, and neurokinin A have already been identified in the vagus nerves and in the nodose ganglia of the rat (13,14). Galanin is a 29 amino acid peptide which was first isolated from the upper small intestine of the pig (25). It is present along the gastrointestinal (2, 17, 20), respiratory (4) and urogenital (24) tracts in different mammalian species, and has various biological effects on the gastrointestinal motility (10, 11, 21), the gastric and endocrine pancreatic secretions (1, 16, 21). Galanin is also widely distributed in the central nervous system of mammals (4, 5, 19, 20, 22) and is found especially in motor nuclei of the brain stem (18) and in cells and fibers of the solitary tract nucleus in rat (22,23). Little is known about its possible role in the transmission of visceral information carried by vagal afferences. In the present study, we identified galanin in porcine nodose ganglia. For this purpose, we used immunohistochemistry and a new specific radioimmunoassay (RIA) for galanin combined with high performance liquid chromatography (HPLC). In addition, the same study was performed on gastrointestinal tissues used as a reference.
‘Requests France.
for reprints should be addressed
to Catherine
Philippe,
Synthetic porcine galanin was supplied by Bachem (Bunbendorf, Switzerland). Halothane was purchased from Coopers veterinaire S.A. (Meaux, France). Thiopenthal sodique was obtained from Rhone Merieux (Lyon, France) and pentobarbital sodique from Sanofi (Torcy, France). I-Ethyl-3-(3-dimethylaminopropyl) carbodiimide, lactoperoxidase and bovine serum albumin (fraction V) were from Sigma Chemical (St. Louis, MO). The fluoresceinisothiocyanate (FITC) conjugated goat anti-rabbit IgG was obtained from NORDIC (Tilburg, The Netherlands). ‘?odine, supplied as sodium iodide, was purchased from CIS International (Gif sur Yvette, France). p-Benzoquinone was from Merck (Darmstadt, West Germany). The Waters Associates liquid chromatography apparatus consisted of a Rheodyne injector, two model 5 10 pumps and a model 680 solvent programmer.
Immunohistochemistry Large white venous injection intubation was using halothane
C.R.J.J.,
989
L.E.P.S.D.,
pigs (40-50 kg) were anesthetized with an intraof pentobarbital and thiopental. An endotracheal then performed and anesthesia was maintained (1.5 to 2%). Nodose ganglia were identified as a
Bltiment
405, Domaine
de Vilvert,
78350 Jouy-en-Josas,
990
PHILIPPE ET AL.
FIG. 1. Immunofluorescence micrographs of different layers of the ileum wall in the pig after incubation with galanin antiserum 51162C. (A) Galanin-like immunoreactive fibers can be observed around the crypts of Lieberkfihn (C) in lamina muscularis mucosae (m.m.) and submucous plexus (s.p.). Note one immunoreactive cell body in the submucous plexus (arrow). (B) Some fluorescent fibers are seen in the longitudinal muscle layer (1.m.). Bar= 25 ~m. thickening of the cervical vagus nerve from which emerged the superior laryngeal nerve. The afferent and the efferent vagal fiber trunks were segmented and dissociated and nodose ganglia were isolated. Part of these were frozen in liquid nitrogen and stored at 20°C for later peptide extraction. Freshly removed nodose ganglia and full thickness ileal wall samples were immersed in an ice-cold phosphate buffer saline solution (PBS) (0.1 M, pH 7.4), containing 0.4% p-benzoquinone for 2 hours at 4°C. They were repeatedly rinsed in PBS containing -
10% sucrose at 4°C and frozen in liquid nitrogen. Sections 15 p~m thick were sliced with a Bright cryostat. Immunoreactive galanin was detected with the indirect immunofluorescent technique of Coons (7). Antiserum (As) 51162C raised in rabbit against synthetic porcine galanin was diluted 1/100 in PBS. Sections were incubated with this primary antiserum for 1 hour in a humid chamber at room temperature. After three washes with PBS, sections were incubated for 1 hour with the secondary antiserum, i.e., antirabbit IgG coupled to FITC diluted 1/100 in PBS.
GALANIN IN PORCINE NODOSE GANGLIA
991
FIG. 2. Immunofluorescence micrograph of porcine nodose ganglia after incubation with As 51162C. Few immunoreactive cell bodies are seen. No afferent sensory fibers were immunoreactive for galanin. Bar = 25 p,m. Sections were examined with a fluorescent microscope and photographed. To test the specificity of the immunostaining, sections were incubated with As 51162C that had been preabsorbed with 5.10-3 M galanin for 1 hour at room temperature.
Tissue Extractions of Galanin Tissue samples (full thickness wall) from the gastrointestinal
Radioimmunoassay
~ l-golonin (I/dilution) 16 8 4 2 I ' ' ' I I 50.
~
/d jejunal
extroct
'9 2,o
~o .,qo
40,
50
N.G. extroct I00 200
i :50, i
8
_
20.
n
I0.
O.
tract (stomach to colon) were taken and rinsed with ice-cold isotonic saline. About 100 mg of each sample were immediately homogenized in 10 vol. (v/w) ice-cold 2 M acetic acid with a polytron homogenizer for 5 min, then maintained at 100°C for 10 min. The homogenates were centrifuged at 2000 r.p.m, for 20 min at 4°C. The supernatant was lyophilized and reconstituted extemporally in the buffer used for the RIA of galanin. Frozen nodose ganglia were weighed and extracted as above.
I I0 I0 Frnol golanin per tube
FIG. 3. Inhibition of binding of iodinated galanin by antiserum 51162C in the presence of serial concentrations of synthetic porcine galanin (O), of two jejunal extracts (A,A) and of two nodose ganglia extracts (,,E3). The self-displacement study was performed with serial dilutions of iodinated galanin (©).
Galanin was measured using a specific RIA. The antisera were produced in rabbits immunized repeatedly with synthetic porcine galanin conjugated to bovine serum albumin with carbodiimide (12 moles galanin/1 mole BSA). One antiserum 51162C was selected for RIA and was used at a final dilution of 1/20,000. Synthetic porcine galanin was radioiodinated by the lactoperoxidase method. Five p~g (1.6 nmol) of galanin in 10 txl ammonium acetate (0.8 M, pH 5.5) were incubated with 0.24 nmol (0.5 mCi) carrier free [125I]-Na, 5 ~g lactoperoxidase and 5 ILl hydrogen peroxide 0.06% for 15 min. Iodinated galanin was purified by reverse phase HPLC using a p,-Bondapak C-18 column. The elution was performed with a 25-min linear gradient from 20% to 35% acetonitrile-0.05% trifluoroacetic acid (TFA) at a flow rate of 1.5 ml/min. ['25I]-Galanin was eluted in one peak with a retention time of 20 min corresponding to 31.2% acetonitrile. Assays were performed in 800 ixl phosphate buffer (PB) (0.05 M, pH 7.5) containing 4 mM EDTA, 166 KIU/ml aprotinin, 5 mM sodium azide and 2% porcine plasma (RIA buffer). Standard galanin (3 to 300 fmole in 50 ILl PB) or unknown samples (50 to 200 ILl) were incubated with 50 txl As 51162C at 4°C for 24 hours. Approximately 2000 cpm of ['2sI]-galanin in 50 ~1 PB were added and kept at 4°C for 48 hours. Then 50 bd donkey anti-rabbit serum (diluted 1/15 in PB) and 50 ILl nonimmune rabbit serum (1% in PB) were added and incubated at 4°C for 15 hours. Precipitates were centrifuged at 3000 r.p.m, for 1 hour at 4°C. The
992
PHILIPPE ET AL.
j
©
8
0
!
"6
%
!
|
III
I
•
®
E x
!
RESULTS
15,
I0-
g E
5.
.E
0
I
!-
| "t5
(.9
I I I I
®
15.
-8o
.40
IO.
DISCUSSION
/
.10 0
!
0
I0
A few galanin-like immunoreactive cell bodies could be identified with As 51162C in the submucous plexus but not in the myenteric plexus at the ileum level. Galanin-like immunoreactivity (Gal-LI) could be revealed in fibers present in all the different layers of the ileum, i.e., mucosae, muscularis mucosae, submucous plexus, circular muscle, myenteric plexus and longitudinal muscle (Fig. 1). Some of these fibers were seen around blood vessels. In the nodose ganglia, some cell bodies were immunoreactive for galanin and the visualization, at this level, showed homogeneous staining of the whole cell body (Fig. 2). Ten to 15% of the total cell bodies were estimated to be immunoreactive, but no afferent fibers could be detected with As 51162C. The extractable Gal-LI from tissues was quantified by RIA with As 51162C. Displacement curves obtained from serial dilutions of jejunal and nodose ganglia extracts were parallel to that of synthetic porcine galanin (Fig. 3). In nodose ganglia, Gal-LI concentration was 7.2 + 0.8 pmol/g wet tissue. For comparison, Gal-LI contents increased from 9.1 _+3.1 pmol/g wet tissue in the stomach to 61.7 _+ 14.3 pmol/g wet tissue in the ileum, and decreased to 40.8 _+ 11.2 pmol/g wet tissue in the colon. The HPLC profile of jejunal extract exhibited a single peak with the same retention time (41 rain corresponding to 30.5% acetonitrile) as the synthetic porcine galanin. Nodose ganglia extract HPLC profile was similar and provided one major peak coeluting with standard galanin (Fig. 4). A minor peak eluted at 35 min corresponding to 26.9% acetonitrile. It represented about 11% of the total immunoreactivity.
!
20
!
30
,05'06'0
I
Time in rain
FIG. 4. High pressure liquid chromatography profiles of synthetic porcine galanin (A), jejunal extract (B) and nodose ganglia extract (C) obtained as described in the Method section. The dashed line represents the gradient characteristics which are the same for the three separations.
supernatants were removed by aspiration and the pellets counted in a gamma counter for 2 min each. Cross-reactivity of As 51162C with calcitonin gene-related peptide, cholecystokinin, dynorphin, gastrin, gastric inhibitory peptide, gastrin-related peptide, neurotensin, motilin, Met-enkephalin, pancreatic polypeptide, peptide YY, physalaemin, somatostatin, secretin, substance P, and vasoactive intestinal polypeptide was less than 0.1%. The sensitivity and EDso of the assay were 4 fmole/tube and 40 fmole/tube, respectively. Intraassay and interassay coefficients of variation were calculated and found to be less than 10%. High Performance Liquid Chromatography Lyophilized tissue extracts were reconstituted in 2.5 ml 10% acetonitrile-0.05% TFA and 2 ml were applied onto a ix-Bondapak C-18 column (3.9 × 300 mm). The column was eluted for 6 min with 10% acetonitrile-0.05% TFA followed by a 56-min linear gradient from 10% to 40% acetonitrile-0.05% TFA throughout. The flow rate was 1.5 ml/min. One-min fractions were collected. Acetonitrile was eliminated with a speed vacuum evaporator. All fractions were adjusted to 1 ml with RIA buffer; 200 I.d of each fraction were assayed for galanin. The recovery of the procedure was 74 ± 6% (n = 8).
The present work was performed on pigs. This species was a good model for several reasons. Cell bodies from the afferent vagal neurons could be easily isolated. They were concentrated in two nodose ganglia which were readily detected in the vagus nerve at the cervical level and separated from the efferent vagal fibers trunk. They allowed quantitative studies since one ganglion weighed between 20 and 30 mg. Immunohistochemistry, performed with As 51162C, showed the presence of Gal-L1 in some cell bodies of the vagal sensor s' neurons, however, none could be detected in fibers arising from these cell bodies. Gal-LI was also present in cell bodies and fibers of the ileum in agreement with the results of Melander et al. (17) in the pig. Gal-LI-positive cell bodies and fibers have also been shown in the gastrointestinal tract tissues by Ekblad et al. (9) and Ch'ng et al. (5) in the rat and by Gonda et al. (12) in the dog. Along the gastrointestinal tract, the ileum contained the highest amount of Gal-LI as determined by RIA. However, only a few cell bodies and fibers were revealed with As 51162C. Since the concentration of Gal-LI in vagal sensory neurons was nine times lower than in the ileum, less material could be revealed. Indeed, nodose ganglia are mainly constituted of cell bodies which reduce the probability to have galanin-positive fibers. The staining density of Gal-LI containing cell bodies was homogeneous in nodose ganglia and in the enteric nervous system. The staining density of Gal-LI containing fibers in the enteric nervous system was heterogeneous. As a peptide, galanin or its precursor molecules is distributed in the whole nerve cell (15). On the contrary, the heterogenous distribution of the fluorescence in fibers can be related to the existence of varicosities. These varicosities are associated with a neuromediation process (8) and stock large concentration of peptides. At present, no synapse has been described near the afferent neurons of the vagus nerve at the level of the nodose ganglia. This suggests that afferent vagal fibers, at this level, do not possess varicosities, which can explain the
GALANIN IN PORCINE NODOSE GANGLIA
993
absence of fluorescence on these fibers. Gal-LI contents in nodose ganglia extracts, measured by RIA, were close to those found in gastric extracts. In these latter tissues, our results were similar to those of Bauer et al. (2). However, the comparison with the GaI-LI concentration found in the gastrointestinal tract is not totally satisfactory. The nature of these two tissues is different. The proportion of nervous tissues in the gastrointestinal tract, as related to the total tissue weight, is lower than the proportion of the nervous tissues in nodose ganglia. Thus, Gal-LI is probably more concentrated in the enteric nervous system than in the vagal afferences, but it is underestimated by this protocol. Nevertheless, this value is between the values found by C h ' n g et al. (6) in the porcine hypothalamus and spinal cord. These latter tissues are closer to nodose ganglia in composition. HPLC analysis of Gal-LI in extractable material demonstrated the existence of galanin in both digestive tissues and nodose ganglia. One major peak of immunoreactive galanin coeluting with synthetic porcine galanin was revealed after HPLC analysis with As 51162C. In contrast, Bauer et al. (2) reported two peaks of equimolar proportion of immunoreactive galanin in digestive tissues. Since we used an acid extraction procedure comparable to Bauer's one, the discrepancy in the results does not seem to be the consequence of artifactual events occurring during the extraction of immunoreactive material. One explanation can be the different specificity of the antiserum used in this study. However, another minor peak of immunoreactive galanin was revealed in vagal sensory neurons which might correspond to the molecular variant
found by Bauer et al. (2) in the digestive tract. Our results are in fair agreement to those obtained by C h ' n g et al. (6) in porcine hypothalamic extracts with the same antiserum as that used by Bauer et al. (2). The present study demonstrates the existence of galanin in some cell bodies of vagal sensory neurons. The origin of these afferent vagal neurons remains unknown. Cheung et al. (4) have detected very few galanin-like immunoreactive fibers in the porcine lung, while Melander et al. (17) have reported a wide distribution of these fibers along the gastrointestinal tract tissues which is in agreement with our results. Taken together, these data suggest that the galanin-positive cell bodies of porcine nodose ganglia may be the afferent vagal neurons coming from the gastrointestinal tract tissues. Additional work is required to verify this hypothesis. Since Gal-LI is detected 1) in the gastrointestinal tract and particularly in the enteric nervous system, 2) in the rat cells and fibers of the solitari tract nucleus, rich in vagal sensory endings (22,23), 3) in cell bodies located in nodose ganglia as demonstrated in the present study, it may be speculated that galanin is a modulatory substance in the transmission of visceral ascending information through the vagus nerve. ACKNOWLEDGEMENTS The excellent technical assistance of T. Gibard, F. Levenez and M.-A. Dechelette is gratefully acknowledged.
REFERENCES 1. Ahren, B.; Rorsman, P.; Berggren, P-O. Galanin and the endocrine pancreas. FEBS Lett. 229(2):233-237; 1988. 2. Bauer, F. E.; Adrian, T. E.; Christofides, N. D.; Ferri, G-L.; Yanaihara, N.; Polak, J. M.; Bloom, S. R. Distribution and molecular heterogeneity of galanin in human, pig, guinea pig, and rat gastrointestinal tracts. Gastroenterology 91:877-883; 1986. 3. Beal, M. F.; Gabriel, S. M.; Swartz, K. J.; MacGarvey, U. M. Distribution of galanin-like immunoreactivity in baboon brain. Peptides 9:847-851; 1988. 4. Cheung, A.; Polak, J. M.; Bauer, F. E.; Cadieux, A.; Christofides, N. D.; Springall, D. R.; Bloom, S. R. Distribution of galanin immunoreactivity in the respiratory tract of pig, guinea pig, rat, and dog. Thorax 40:889-896; 1985. 5. Ch'ng, J. L. C.; Christofides, N. D.; Suzki, H.; Yiangou, Y.; Stephano, A.; Tatemoto, K.; Polak, J. M.; Bloom, S. R. Distribution of galanin immunoreactivity (GIR) in the central nervous system and gastrointestinal tract. Dig. Dis. Sci. 29:17S; 1985. 6. Ch'ng, J. L. C.; Christofides, N. D.; Anand, P.; Gibson, S. J.; Allen, Y. S.; Su, H. C.; Tatemoto, K.; Morrison, J. F. B.; Polak, J. M.; Bloom, S. R. Distribution of galanin immunoreactivity in the central nervous system and the responses of galanin-containing neuronal pathways to injury. Neuroscience 16(2):343-354; 1985. 7. Coons, A. H. Fluorescent antibody methods. In: Danielli, J. F., ed. General cytochemical methods. New York: Academic Press; 1958: 394-422. 8. Cooper, J. R.; Bloom, F. E. The biochemical basis of neuropharmacology. Oxford: University Press; 1982:295-296. 9. Ekblad, E.; Rrkaeus, A.; Hakanson, R.; Sundler, F. Galanin nerve fibers in the rat gut: Distribution, origin and projections. Neuroscience 16(2):355-363; 1985. 10. Ekblad, E.; Hakanson, R.; Sundler, R.; Wahlestedt, C. Galanin: Neuromodulatory and direct contractile effects on smooth muscle preparations. Br. J. Pbarmacol. 86:241-246; 1985. 11. Fox, J. E. T.; McDonald, T. J.; Kostolanska, F.; Tatemoto, K. Galanin: An inhibitory neural peptide of the canine small intestine. Life Sci. 39:103-110; 1986. 12. Gonda, T.; Daniel, E. E.; McDonald, T. J.; Fox, J. E. T.; Brooks, B. D.; Oki, M. Distribution and function of enteric GAL-IR nerves in dogs: Comparison with VIP. Am. J. Physiol. 256:G884-G896; 1989.
13. Green, T.; Dockray, G. J. Calcitonin gene-related peptide and substance P in afferents to the upper gastrointestinal tract in the rat. Neurosci. Len. 76:151-156; 1987. 14. Helke, C. J.; Hill, K. M. Immunohistochemical study of neuropeptides in vagal and glossopharyngeal afferent neurons in the rat. Neuroscience 26(2):539-551; 1988. 15. Hrkfelt, T.; Johansson, O.; Ljungdahl, A.; Lundberg, J. M.; Schultzberg, M. Peptidergic neurones. Nature 284(10):515-521; 1980. 16. Kwok, Y. N.; Verchere, C. B.; Mclntosh, C. H. S.; Brown, J. C. Effect of galanin on endocrine secretions from the isolated perfused rat stomach and pancreas. Eur. J. Pharmacol. 145:49-54; 1988. 17. Melander, T.; Hrkfelt, T.; Rrkaeus, A.; Fahrenkrug, J.; Tatemoto, K.; Mutt, V. Distribution of galanin-like immunoreactivity in the gastrointestinal tract of several mammalian species. Cell Tissue Res. 239:253-270; 1985. 18. Moore, R. Y. Cranial motor neurons contain either galanin or calcitonin gene-related peptidelike immunoreactivity. J. Comp. Neurol. 282:512-522; 1989. 19. Palkovits, M.; Rrkaeus, A.; Antoni, F. A.; Kiss, A. Galanin in the hypothalamo-hypophyseal system. Neuroendocrinology 46:417-423; 1987. 20. Rrkaeus, A.; Melander, T.; Hrkfelt, T.; Lundberg, J. M.; Tatemoto, K.; Carlquist, M.; Mutt, V. A galanin-like peptide in the central nervous system and intestine of the rat. Neurosci. Lett. 47:161-166; 1984. 21. Rrkaeus, A. Galanin: A newly isolated biologically active neuropeptide. Trends Neurosci. 10(4):158-164; 1987. 22. Skofitsch, G.; Jacobowitz, D. M. Immunohistochemical mapping of galanin-like neurons in the rat central nervous system. Peptides 6:509-546; 1985. 23. Skofitsch, G.; Jacobowitz, D. M. Quantitative distribution of galaninlike immunoreactivity in the rat central nervous system. Peptides 7:609-613; 1986. 24. Stjemquist, M.; Ekblad, E.; Owman, Ch.; Sundler, F. Immunocytochemical localization of galanin in the rat male and female genital tracts and motor effects in vitro. Regul. Pept. 20:335-343; 1988. 25. Tatemoto, K.; Rrkaeus, A.; Jornvall, H.; McDonald, T. J.; Mutt, V. Galanin--A novel biologically active peptide from porcine intestine. FEBS Len. 164(1):124-128; 1983.
0 Pergamon Press plc, 1990. Printed in the U.S.A.
0196-9781/90 $3.00 + .@I
Galanin in Porcine Vagal Sensory Nerves: Immunohistochemical and Immunochemical Study C. PHILIPPE,*’ J. C. CUBER,? A. BOSSHARD,t 0. RAMPIN,* J. P. LAPLACE” AND J. A. CHAYVIALLEt “Station de Physiologie de la Nutrition, INRA, 78350 Jouy-en-Josas, France tINSERM, U45, Hdpital E. Herriot, 49437 Lyon cedex 03, France Received
20 February
1990
0. RAMPIN, J. P. LAPLACE AND J. A. CHAYVIALLE. Galanin inporcine vagal study. PEPTIDES ll(5) 989-993, 1990.-In this work, the presence of galanin was examined by immunohistochemistry, radioimmunoassay and high performance liquid chromatography (HPLC) in porcine nodose ganglia, mainly constituted of cell bodies from the vagal sensory neurons. Galanin-like immunoreactivity (Gal-LI) was revealed in 10 to 15% of the total cell bodies by the indirect immunofluorescent technique of Coons. For comparison, a positive staining was revealed in a few cell bodies of the submucous plexus and in fibers located in the different layers of the ileum. The extractable Gal-L1 content in nodose ganglia was 7.2 kO.8 pmolig wet tissue, which represents a concentration about nine times lower than that found in the ileum. HPLC of extractable material revealed a predominant peak which coeluted with the synthetic peptide. We propose that, in pigs, galanin may play a role in the transmission of visceral information through the vagal afferences. PHILIPPE,
C., J. C. CUBER. A. BOSSHARD,
sensory nerves: lmmunohisrochemical
Galanin
Nodose ganglia
and immunochemical
Pig
Visceral afferences
METHOD
VAGAL sensory neurons relay information from receptors located in the gastrointestinal wall, lung and heart to the brain stem. Their
Materials
corresponding cell bodies are localized in the nodose ganglia. Several peptides, such as substance P, calcitonin gene-related peptide, cholecystokinin, vasoactive intestinal peptide, somatostatin, and neurokinin A have already been identified in the vagus nerves and in the nodose ganglia of the rat (13,14). Galanin is a 29 amino acid peptide which was first isolated from the upper small intestine of the pig (25). It is present along the gastrointestinal (2, 17, 20), respiratory (4) and urogenital (24) tracts in different mammalian species, and has various biological effects on the gastrointestinal motility (10, 11, 21), the gastric and endocrine pancreatic secretions (1, 16, 21). Galanin is also widely distributed in the central nervous system of mammals (4, 5, 19, 20, 22) and is found especially in motor nuclei of the brain stem (18) and in cells and fibers of the solitary tract nucleus in rat (22,23). Little is known about its possible role in the transmission of visceral information carried by vagal afferences. In the present study, we identified galanin in porcine nodose ganglia. For this purpose, we used immunohistochemistry and a new specific radioimmunoassay (RIA) for galanin combined with high performance liquid chromatography (HPLC). In addition, the same study was performed on gastrointestinal tissues used as a reference.
‘Requests France.
for reprints should be addressed
to Catherine
Philippe,
Synthetic porcine galanin was supplied by Bachem (Bunbendorf, Switzerland). Halothane was purchased from Coopers veterinaire S.A. (Meaux, France). Thiopenthal sodique was obtained from Rhone Merieux (Lyon, France) and pentobarbital sodique from Sanofi (Torcy, France). I-Ethyl-3-(3-dimethylaminopropyl) carbodiimide, lactoperoxidase and bovine serum albumin (fraction V) were from Sigma Chemical (St. Louis, MO). The fluoresceinisothiocyanate (FITC) conjugated goat anti-rabbit IgG was obtained from NORDIC (Tilburg, The Netherlands). ‘?odine, supplied as sodium iodide, was purchased from CIS International (Gif sur Yvette, France). p-Benzoquinone was from Merck (Darmstadt, West Germany). The Waters Associates liquid chromatography apparatus consisted of a Rheodyne injector, two model 5 10 pumps and a model 680 solvent programmer.
Immunohistochemistry Large white venous injection intubation was using halothane
C.R.J.J.,
989
L.E.P.S.D.,
pigs (40-50 kg) were anesthetized with an intraof pentobarbital and thiopental. An endotracheal then performed and anesthesia was maintained (1.5 to 2%). Nodose ganglia were identified as a
Bltiment
405, Domaine
de Vilvert,
78350 Jouy-en-Josas,
990
PHILIPPE ET AL.
FIG. 1. Immunofluorescence micrographs of different layers of the ileum wall in the pig after incubation with galanin antiserum 51162C. (A) Galanin-like immunoreactive fibers can be observed around the crypts of Lieberkfihn (C) in lamina muscularis mucosae (m.m.) and submucous plexus (s.p.). Note one immunoreactive cell body in the submucous plexus (arrow). (B) Some fluorescent fibers are seen in the longitudinal muscle layer (1.m.). Bar= 25 ~m. thickening of the cervical vagus nerve from which emerged the superior laryngeal nerve. The afferent and the efferent vagal fiber trunks were segmented and dissociated and nodose ganglia were isolated. Part of these were frozen in liquid nitrogen and stored at 20°C for later peptide extraction. Freshly removed nodose ganglia and full thickness ileal wall samples were immersed in an ice-cold phosphate buffer saline solution (PBS) (0.1 M, pH 7.4), containing 0.4% p-benzoquinone for 2 hours at 4°C. They were repeatedly rinsed in PBS containing -
10% sucrose at 4°C and frozen in liquid nitrogen. Sections 15 p~m thick were sliced with a Bright cryostat. Immunoreactive galanin was detected with the indirect immunofluorescent technique of Coons (7). Antiserum (As) 51162C raised in rabbit against synthetic porcine galanin was diluted 1/100 in PBS. Sections were incubated with this primary antiserum for 1 hour in a humid chamber at room temperature. After three washes with PBS, sections were incubated for 1 hour with the secondary antiserum, i.e., antirabbit IgG coupled to FITC diluted 1/100 in PBS.
GALANIN IN PORCINE NODOSE GANGLIA
991
FIG. 2. Immunofluorescence micrograph of porcine nodose ganglia after incubation with As 51162C. Few immunoreactive cell bodies are seen. No afferent sensory fibers were immunoreactive for galanin. Bar = 25 p,m. Sections were examined with a fluorescent microscope and photographed. To test the specificity of the immunostaining, sections were incubated with As 51162C that had been preabsorbed with 5.10-3 M galanin for 1 hour at room temperature.
Tissue Extractions of Galanin Tissue samples (full thickness wall) from the gastrointestinal
Radioimmunoassay
~ l-golonin (I/dilution) 16 8 4 2 I ' ' ' I I 50.
~
/d jejunal
extroct
'9 2,o
~o .,qo
40,
50
N.G. extroct I00 200
i :50, i
8
_
20.
n
I0.
O.
tract (stomach to colon) were taken and rinsed with ice-cold isotonic saline. About 100 mg of each sample were immediately homogenized in 10 vol. (v/w) ice-cold 2 M acetic acid with a polytron homogenizer for 5 min, then maintained at 100°C for 10 min. The homogenates were centrifuged at 2000 r.p.m, for 20 min at 4°C. The supernatant was lyophilized and reconstituted extemporally in the buffer used for the RIA of galanin. Frozen nodose ganglia were weighed and extracted as above.
I I0 I0 Frnol golanin per tube
FIG. 3. Inhibition of binding of iodinated galanin by antiserum 51162C in the presence of serial concentrations of synthetic porcine galanin (O), of two jejunal extracts (A,A) and of two nodose ganglia extracts (,,E3). The self-displacement study was performed with serial dilutions of iodinated galanin (©).
Galanin was measured using a specific RIA. The antisera were produced in rabbits immunized repeatedly with synthetic porcine galanin conjugated to bovine serum albumin with carbodiimide (12 moles galanin/1 mole BSA). One antiserum 51162C was selected for RIA and was used at a final dilution of 1/20,000. Synthetic porcine galanin was radioiodinated by the lactoperoxidase method. Five p~g (1.6 nmol) of galanin in 10 txl ammonium acetate (0.8 M, pH 5.5) were incubated with 0.24 nmol (0.5 mCi) carrier free [125I]-Na, 5 ~g lactoperoxidase and 5 ILl hydrogen peroxide 0.06% for 15 min. Iodinated galanin was purified by reverse phase HPLC using a p,-Bondapak C-18 column. The elution was performed with a 25-min linear gradient from 20% to 35% acetonitrile-0.05% trifluoroacetic acid (TFA) at a flow rate of 1.5 ml/min. ['25I]-Galanin was eluted in one peak with a retention time of 20 min corresponding to 31.2% acetonitrile. Assays were performed in 800 ixl phosphate buffer (PB) (0.05 M, pH 7.5) containing 4 mM EDTA, 166 KIU/ml aprotinin, 5 mM sodium azide and 2% porcine plasma (RIA buffer). Standard galanin (3 to 300 fmole in 50 ILl PB) or unknown samples (50 to 200 ILl) were incubated with 50 txl As 51162C at 4°C for 24 hours. Approximately 2000 cpm of ['2sI]-galanin in 50 ~1 PB were added and kept at 4°C for 48 hours. Then 50 bd donkey anti-rabbit serum (diluted 1/15 in PB) and 50 ILl nonimmune rabbit serum (1% in PB) were added and incubated at 4°C for 15 hours. Precipitates were centrifuged at 3000 r.p.m, for 1 hour at 4°C. The
992
PHILIPPE ET AL.
j
©
8
0
!
"6
%
!
|
III
I
•
®
E x
!
RESULTS
15,
I0-
g E
5.
.E
0
I
!-
| "t5
(.9
I I I I
®
15.
-8o
.40
IO.
DISCUSSION
/
.10 0
!
0
I0
A few galanin-like immunoreactive cell bodies could be identified with As 51162C in the submucous plexus but not in the myenteric plexus at the ileum level. Galanin-like immunoreactivity (Gal-LI) could be revealed in fibers present in all the different layers of the ileum, i.e., mucosae, muscularis mucosae, submucous plexus, circular muscle, myenteric plexus and longitudinal muscle (Fig. 1). Some of these fibers were seen around blood vessels. In the nodose ganglia, some cell bodies were immunoreactive for galanin and the visualization, at this level, showed homogeneous staining of the whole cell body (Fig. 2). Ten to 15% of the total cell bodies were estimated to be immunoreactive, but no afferent fibers could be detected with As 51162C. The extractable Gal-LI from tissues was quantified by RIA with As 51162C. Displacement curves obtained from serial dilutions of jejunal and nodose ganglia extracts were parallel to that of synthetic porcine galanin (Fig. 3). In nodose ganglia, Gal-LI concentration was 7.2 + 0.8 pmol/g wet tissue. For comparison, Gal-LI contents increased from 9.1 _+3.1 pmol/g wet tissue in the stomach to 61.7 _+ 14.3 pmol/g wet tissue in the ileum, and decreased to 40.8 _+ 11.2 pmol/g wet tissue in the colon. The HPLC profile of jejunal extract exhibited a single peak with the same retention time (41 rain corresponding to 30.5% acetonitrile) as the synthetic porcine galanin. Nodose ganglia extract HPLC profile was similar and provided one major peak coeluting with standard galanin (Fig. 4). A minor peak eluted at 35 min corresponding to 26.9% acetonitrile. It represented about 11% of the total immunoreactivity.
!
20
!
30
,05'06'0
I
Time in rain
FIG. 4. High pressure liquid chromatography profiles of synthetic porcine galanin (A), jejunal extract (B) and nodose ganglia extract (C) obtained as described in the Method section. The dashed line represents the gradient characteristics which are the same for the three separations.
supernatants were removed by aspiration and the pellets counted in a gamma counter for 2 min each. Cross-reactivity of As 51162C with calcitonin gene-related peptide, cholecystokinin, dynorphin, gastrin, gastric inhibitory peptide, gastrin-related peptide, neurotensin, motilin, Met-enkephalin, pancreatic polypeptide, peptide YY, physalaemin, somatostatin, secretin, substance P, and vasoactive intestinal polypeptide was less than 0.1%. The sensitivity and EDso of the assay were 4 fmole/tube and 40 fmole/tube, respectively. Intraassay and interassay coefficients of variation were calculated and found to be less than 10%. High Performance Liquid Chromatography Lyophilized tissue extracts were reconstituted in 2.5 ml 10% acetonitrile-0.05% TFA and 2 ml were applied onto a ix-Bondapak C-18 column (3.9 × 300 mm). The column was eluted for 6 min with 10% acetonitrile-0.05% TFA followed by a 56-min linear gradient from 10% to 40% acetonitrile-0.05% TFA throughout. The flow rate was 1.5 ml/min. One-min fractions were collected. Acetonitrile was eliminated with a speed vacuum evaporator. All fractions were adjusted to 1 ml with RIA buffer; 200 I.d of each fraction were assayed for galanin. The recovery of the procedure was 74 ± 6% (n = 8).
The present work was performed on pigs. This species was a good model for several reasons. Cell bodies from the afferent vagal neurons could be easily isolated. They were concentrated in two nodose ganglia which were readily detected in the vagus nerve at the cervical level and separated from the efferent vagal fibers trunk. They allowed quantitative studies since one ganglion weighed between 20 and 30 mg. Immunohistochemistry, performed with As 51162C, showed the presence of Gal-L1 in some cell bodies of the vagal sensor s' neurons, however, none could be detected in fibers arising from these cell bodies. Gal-LI was also present in cell bodies and fibers of the ileum in agreement with the results of Melander et al. (17) in the pig. Gal-LI-positive cell bodies and fibers have also been shown in the gastrointestinal tract tissues by Ekblad et al. (9) and Ch'ng et al. (5) in the rat and by Gonda et al. (12) in the dog. Along the gastrointestinal tract, the ileum contained the highest amount of Gal-LI as determined by RIA. However, only a few cell bodies and fibers were revealed with As 51162C. Since the concentration of Gal-LI in vagal sensory neurons was nine times lower than in the ileum, less material could be revealed. Indeed, nodose ganglia are mainly constituted of cell bodies which reduce the probability to have galanin-positive fibers. The staining density of Gal-LI containing cell bodies was homogeneous in nodose ganglia and in the enteric nervous system. The staining density of Gal-LI containing fibers in the enteric nervous system was heterogeneous. As a peptide, galanin or its precursor molecules is distributed in the whole nerve cell (15). On the contrary, the heterogenous distribution of the fluorescence in fibers can be related to the existence of varicosities. These varicosities are associated with a neuromediation process (8) and stock large concentration of peptides. At present, no synapse has been described near the afferent neurons of the vagus nerve at the level of the nodose ganglia. This suggests that afferent vagal fibers, at this level, do not possess varicosities, which can explain the
GALANIN IN PORCINE NODOSE GANGLIA
993
absence of fluorescence on these fibers. Gal-LI contents in nodose ganglia extracts, measured by RIA, were close to those found in gastric extracts. In these latter tissues, our results were similar to those of Bauer et al. (2). However, the comparison with the GaI-LI concentration found in the gastrointestinal tract is not totally satisfactory. The nature of these two tissues is different. The proportion of nervous tissues in the gastrointestinal tract, as related to the total tissue weight, is lower than the proportion of the nervous tissues in nodose ganglia. Thus, Gal-LI is probably more concentrated in the enteric nervous system than in the vagal afferences, but it is underestimated by this protocol. Nevertheless, this value is between the values found by C h ' n g et al. (6) in the porcine hypothalamus and spinal cord. These latter tissues are closer to nodose ganglia in composition. HPLC analysis of Gal-LI in extractable material demonstrated the existence of galanin in both digestive tissues and nodose ganglia. One major peak of immunoreactive galanin coeluting with synthetic porcine galanin was revealed after HPLC analysis with As 51162C. In contrast, Bauer et al. (2) reported two peaks of equimolar proportion of immunoreactive galanin in digestive tissues. Since we used an acid extraction procedure comparable to Bauer's one, the discrepancy in the results does not seem to be the consequence of artifactual events occurring during the extraction of immunoreactive material. One explanation can be the different specificity of the antiserum used in this study. However, another minor peak of immunoreactive galanin was revealed in vagal sensory neurons which might correspond to the molecular variant
found by Bauer et al. (2) in the digestive tract. Our results are in fair agreement to those obtained by C h ' n g et al. (6) in porcine hypothalamic extracts with the same antiserum as that used by Bauer et al. (2). The present study demonstrates the existence of galanin in some cell bodies of vagal sensory neurons. The origin of these afferent vagal neurons remains unknown. Cheung et al. (4) have detected very few galanin-like immunoreactive fibers in the porcine lung, while Melander et al. (17) have reported a wide distribution of these fibers along the gastrointestinal tract tissues which is in agreement with our results. Taken together, these data suggest that the galanin-positive cell bodies of porcine nodose ganglia may be the afferent vagal neurons coming from the gastrointestinal tract tissues. Additional work is required to verify this hypothesis. Since Gal-LI is detected 1) in the gastrointestinal tract and particularly in the enteric nervous system, 2) in the rat cells and fibers of the solitari tract nucleus, rich in vagal sensory endings (22,23), 3) in cell bodies located in nodose ganglia as demonstrated in the present study, it may be speculated that galanin is a modulatory substance in the transmission of visceral ascending information through the vagus nerve. ACKNOWLEDGEMENTS The excellent technical assistance of T. Gibard, F. Levenez and M.-A. Dechelette is gratefully acknowledged.
REFERENCES 1. Ahren, B.; Rorsman, P.; Berggren, P-O. Galanin and the endocrine pancreas. FEBS Lett. 229(2):233-237; 1988. 2. Bauer, F. E.; Adrian, T. E.; Christofides, N. D.; Ferri, G-L.; Yanaihara, N.; Polak, J. M.; Bloom, S. R. Distribution and molecular heterogeneity of galanin in human, pig, guinea pig, and rat gastrointestinal tracts. Gastroenterology 91:877-883; 1986. 3. Beal, M. F.; Gabriel, S. M.; Swartz, K. J.; MacGarvey, U. M. Distribution of galanin-like immunoreactivity in baboon brain. Peptides 9:847-851; 1988. 4. Cheung, A.; Polak, J. M.; Bauer, F. E.; Cadieux, A.; Christofides, N. D.; Springall, D. R.; Bloom, S. R. Distribution of galanin immunoreactivity in the respiratory tract of pig, guinea pig, rat, and dog. Thorax 40:889-896; 1985. 5. Ch'ng, J. L. C.; Christofides, N. D.; Suzki, H.; Yiangou, Y.; Stephano, A.; Tatemoto, K.; Polak, J. M.; Bloom, S. R. Distribution of galanin immunoreactivity (GIR) in the central nervous system and gastrointestinal tract. Dig. Dis. Sci. 29:17S; 1985. 6. Ch'ng, J. L. C.; Christofides, N. D.; Anand, P.; Gibson, S. J.; Allen, Y. S.; Su, H. C.; Tatemoto, K.; Morrison, J. F. B.; Polak, J. M.; Bloom, S. R. Distribution of galanin immunoreactivity in the central nervous system and the responses of galanin-containing neuronal pathways to injury. Neuroscience 16(2):343-354; 1985. 7. Coons, A. H. Fluorescent antibody methods. In: Danielli, J. F., ed. General cytochemical methods. New York: Academic Press; 1958: 394-422. 8. Cooper, J. R.; Bloom, F. E. The biochemical basis of neuropharmacology. Oxford: University Press; 1982:295-296. 9. Ekblad, E.; Rrkaeus, A.; Hakanson, R.; Sundler, F. Galanin nerve fibers in the rat gut: Distribution, origin and projections. Neuroscience 16(2):355-363; 1985. 10. Ekblad, E.; Hakanson, R.; Sundler, R.; Wahlestedt, C. Galanin: Neuromodulatory and direct contractile effects on smooth muscle preparations. Br. J. Pbarmacol. 86:241-246; 1985. 11. Fox, J. E. T.; McDonald, T. J.; Kostolanska, F.; Tatemoto, K. Galanin: An inhibitory neural peptide of the canine small intestine. Life Sci. 39:103-110; 1986. 12. Gonda, T.; Daniel, E. E.; McDonald, T. J.; Fox, J. E. T.; Brooks, B. D.; Oki, M. Distribution and function of enteric GAL-IR nerves in dogs: Comparison with VIP. Am. J. Physiol. 256:G884-G896; 1989.
13. Green, T.; Dockray, G. J. Calcitonin gene-related peptide and substance P in afferents to the upper gastrointestinal tract in the rat. Neurosci. Len. 76:151-156; 1987. 14. Helke, C. J.; Hill, K. M. Immunohistochemical study of neuropeptides in vagal and glossopharyngeal afferent neurons in the rat. Neuroscience 26(2):539-551; 1988. 15. Hrkfelt, T.; Johansson, O.; Ljungdahl, A.; Lundberg, J. M.; Schultzberg, M. Peptidergic neurones. Nature 284(10):515-521; 1980. 16. Kwok, Y. N.; Verchere, C. B.; Mclntosh, C. H. S.; Brown, J. C. Effect of galanin on endocrine secretions from the isolated perfused rat stomach and pancreas. Eur. J. Pharmacol. 145:49-54; 1988. 17. Melander, T.; Hrkfelt, T.; Rrkaeus, A.; Fahrenkrug, J.; Tatemoto, K.; Mutt, V. Distribution of galanin-like immunoreactivity in the gastrointestinal tract of several mammalian species. Cell Tissue Res. 239:253-270; 1985. 18. Moore, R. Y. Cranial motor neurons contain either galanin or calcitonin gene-related peptidelike immunoreactivity. J. Comp. Neurol. 282:512-522; 1989. 19. Palkovits, M.; Rrkaeus, A.; Antoni, F. A.; Kiss, A. Galanin in the hypothalamo-hypophyseal system. Neuroendocrinology 46:417-423; 1987. 20. Rrkaeus, A.; Melander, T.; Hrkfelt, T.; Lundberg, J. M.; Tatemoto, K.; Carlquist, M.; Mutt, V. A galanin-like peptide in the central nervous system and intestine of the rat. Neurosci. Lett. 47:161-166; 1984. 21. Rrkaeus, A. Galanin: A newly isolated biologically active neuropeptide. Trends Neurosci. 10(4):158-164; 1987. 22. Skofitsch, G.; Jacobowitz, D. M. Immunohistochemical mapping of galanin-like neurons in the rat central nervous system. Peptides 6:509-546; 1985. 23. Skofitsch, G.; Jacobowitz, D. M. Quantitative distribution of galaninlike immunoreactivity in the rat central nervous system. Peptides 7:609-613; 1986. 24. Stjemquist, M.; Ekblad, E.; Owman, Ch.; Sundler, F. Immunocytochemical localization of galanin in the rat male and female genital tracts and motor effects in vitro. Regul. Pept. 20:335-343; 1988. 25. Tatemoto, K.; Rrkaeus, A.; Jornvall, H.; McDonald, T. J.; Mutt, V. Galanin--A novel biologically active peptide from porcine intestine. FEBS Len. 164(1):124-128; 1983.
