Peptides,Vol. 11, pp. 863-867. ~ Pergamon Press plc, 1990. Printed in the U.S.A.
0196-9781/90 $3.00 + .00
BRIEF COMMUNICATION
Peptidergic Innervation of the Major Salivary Glands of the Ferret G. T O B I N , * A. L U T S , t F. S U N D L E R t A N D J. E K S T R O M * ~ : 1
Departments of *Physiology and ~Medical Cell Research, Lund University and ~Department of Pharmacology, Gothenburg University, Sweden R e c e i v e d 4 January 1990
TOBIN, G., A. LUTS, F. SUNDLER AND J. EKSTROM. Peptidergic innervation of the major salivary glands of the ferret. PEPTIDES 11(4) 863-867, 1990.--In parotid, sublingual and submandibular glands of the ferret, morphological correlates were looked for, using immunocytochemistry, to previous physiological findings showing parasympathetic "atropine-resistant" salivary secretion and neuropeptide-evoked salivation in this species. Nerve fibers storing VIP were numerous in association with acini, ducts and blood vessels, while the number of fibers storing substance P was moderate and those containing CGRP and galanin few; also the number of NPY-containing fibers was low around acini and ducts but relatively high around vessels. Sympathectomy eliminated all NPY- and almost all dopamine [3-hydroxylase-containing fibers. Parasympathectomy of the parotid gland resulted in a total loss of the VIP-containing fibers, and a profound reduction in the number of substance P- and CGRP-containing fibers. Salivary glands DBH
Denervation
Immunocytochemistry
Substance P
VIP
CGRP
Galanin
NPY
THE ferret belongs to the species in which stimulation of the parasympathetic innervation of the salivary glands gives rise to "atropine-resistant" vasodilatation in the glands (2) as well as "atropine-resistant" secretion of saliva (6,7). In the salivary glands of this species exogenous substance P (SP) and vasoactive intestinal peptide (VIP) are secretagogues (6). In the present study, not only nerve fibers containing SP and VIP but also fibers containing calcitonin gene-related peptide (CGRP), enkephalin, galanin, gastrin-releasing peptide (GRP), neuropeptide Y (NPY) and somatostatin were looked for in the parotid, sublingual and submandibular glands of the ferret.
immunocytochemistry; the antibodies used are described elsewhere (11). The specificity of the immunostaining was tested by adding excess amount of the antigen to the antibodies (10-100 I~g of synthetic peptide per ml of diluted antibody). As a marker for adrenergic nerve fibers we used an antiserum against dopamine [3-hydroxylase (11). To reveal coexistence of neuropeptides in one and the same neuron a simultaneous double immunostaining technique was used (1,13). In the present study, peptide antibodies raised in guinea pigs (11) were paired with antibodies raised in rabbits and the labels were fluorescein isothiocyanate (FITC) and tetramethyl rhodamine isothiocyanate (TRITC), respectively (1).
METHOD
RESULTS
In pentobarbitone-anesthetized ferrets (45 mg/kg IP): a) the auriculo-temporal nerve was cut unilaterally aiming at parasympathetic denervation of the parotid gland (n = 2); b) the chordalingual nerve was cut unilaterally and the chorda nerve fibers were dissected deep into the hilus of the submandibular gland, where two distinct ganglia were identified, and removed together with nerve strands aiming at parasympathetic denervation (n = 4); and c) the superior cervical ganglion was removed unilaterally aiming at sympathetic denervation (n = 2). The animals were killed by an IP overdose of pentobarbitone 1-2 weeks later. The salivary glands of both sides were removed, weighed and processed for
The observations are summarized in Table 1 (GRP, somatostatin and enkephalin were not found) and visualized in Figs. 1 (parotid glands) and 2 (submandibular glands). In the submandibular gland, nerve cell bodies containing SP were regularly seen in ganglia located in the hilus region outside the parenchyma as well as in smaller ganglia within the parenchyma. These ganglia also contained numerous VIP-immunoreactive nerve cell bodies. In fact, such nerve cell bodies constituted the majority of the cell bodies in the ganglia and clearly outnumbered those storing SP. Double immunostaining showed that the majority of the CGRPcontaining nerve fibers stored also SP. Further, double immuno-
1Requests for reprints should be addressed to J. Ekstr6m, Department of Physiology, Lund University, S61vegatan 19, 223 62 Lund, Sweden.
863
864
TOBIN, LUTS, SUNDLER AND EKSTROM
TABLE 1 RELATIVE FREQUENCY OF PEPTIDE- AND DBH-CONTAINING NERVE FIBERS AROUND ACIN1, DUCTS AND BLOOD VESSELS IN PAROTID, SUBMANDIBULARAND SUBLINGUALGLANDS OF THE FERRET Parotid Gland
VIP SP CGRP Galanin NPY DBH
Submandibular Gland
Sublingual Gland
Acini
Ducts
Vessels
Acini
Ducts
Vessels
Acini
Ducts
Vessels
+++ ++ +* (+) (+) +++
+++ +* (+) (+) (+) +++
+++ ++ + (+) ++ +++
+++ +++ + (+) (+) +++
+++ +++ + (+) (+) +++
+++ +++ + (+) ++ +++
+++ + (+) (+) (+) ++
+++ ++ (+) (+) (+) ++
+++ ++ (+) (+) ++ +++
*Mostly within nerve bundles. Nerve fiber frequency: (+) very few; + few; + + moderate number; + + + numerous.
staining also revealed that although many fibers stored either VIP or SP, a minor population of nerve fibers around ducts, blood vessels and occasionally around acini in all three glands stored both VIP and SP. Coexistence of SP and VIP was observed also in a few nerve cell bodies in the hilus ganglia of the submandibular gland. Nerve cell bodies in the submandibular gland were surrounded by nerve fibers containing galanin, NPY and DBH. No ganglia were found in the sublingual or parotid glands. Cutting the parasympathetic postganglionic nerve of the parotid gland resulted in an almost total loss of VIP-containing nerve fibers. The number of SP-containing nerve fibers was reduced as was the number of CGRP-containing fibers. However, the reduction was not as marked as that of VIP-containing nerve fibers. The number of NPY-, DBH- and galanin-containing nerve fibers seemed unaffected by the denervation procedure. Parasympathetic postganglionic denervation of the submandibular glands was attempted. However, neither the density nor the distribution of any of the neuropeptide-containing nerve fiber populations or the DBHcontaining nerve fibers seemed affected. Furthermore, many ganglia remained within the parenchyma. Removal of the superior cervical ganglion eliminated the NPY-containing nerve fibers and almost all the DBH-containing nerve fibers in the parotid, submandibular and sublingual glands. Those DBH-containing nerve fibers persisting were mainly found around blood vessels. Occasionally such persisting fibers occurred also around acini and ducts. Nerve fibers containing VIP, SP, CGRP and galanin were seemingly unaffected. DISCUSSION The present study offers morphological correlates to our earlier physiological findings. SP injected intravenously elicits a copious flow of saliva from the parotid and submandibular glands in the ferret (6), while VIP evokes protein release but no flow of saliva (7). VIP in combination with SP potentiates the responses both with respect to fluid and protein secretion (7). It is therefore of interest that VIP and SP were found to coexist in some nerve fibers within the glands. Stimulation of the parasympathetic innervation releases SP, VIP and CGRP (Bloom, Edwards, Ekstr6m and
Tobin, unpublished observation). In the presence of muscarinic receptor blockade, the secretion of saliva in response to parasympathetic stimulation that persists is abolished by tachykinin antagonists (5). In the ferret CGRP-containing nerve fibers also appeared close to secretory elements albeit in small numbers. This peptide elicits no flow of saliva from the glands in this species (Ekstr6m and Tobin, unpublished observation) as is the case in rat salivary glands (4). However, it enhances the secretory effects of parasympathomimetic drugs in the rat and, further, it releases proteins (4). A majority of the CGRP-containing nerve fibers did also store SP; in the rat salivary glands, this type of nerve is sensitive to capsaicin treatment, suggesting a sensory origin (3). A few scattered fibers containing galanin were observed. The functional significance of galanin in the ferret salivary glands is unknown. Atropine-resistant vasodilatation in salivary glands upon stimulation of the parasympathetic innervation is a well-known phenomenon, and VIP is thought to be a mediator (12). Not unexpectedly, blood vessels (10) in the ferret salivary glands were innervated not only by VlP-containing fibers but also by fibers storing SP and CGRP. Of the latter two peptides, CGRP is known to be a particularly potent vasodilator agent; in salivary glands of the rabbit, its effect is additive to the vasodilator response evoked by acetylcholine (12). NPY was found particularly around blood vessels, but a few nerve fibers also occurred close to the secretory elements. Dopamine [3-hydroxylase (DBH) was used as an immunocytochemical marker for adrenergic nerves. DBH-containing nerve fibers showed a partly different pattern of distribution; there was a rich innervation not only of blood vessels but also of acini and ducts. This pattern of distribution, which is valid for carnivores but not for rodents (10,14), may favor the view that a separate population of sympathetic nerve fibers is responsible for vasoconstriction in the gland and another for secretory effects. Sympathetic nerve stimulation as well as adrenaline, injected intravenously, gives rise to a sparse protein rich secretion of saliva from the parotid and submandibular glands of the ferret (Ekstr6m and Tobin, unpublished observation). Parasympathetic denervation of the parotid gland eliminated virtually all the VIP-containing nerve fibers of the parotid glands. There was also a reduction in the number of SP- and CGRP-containing nerve fibers in this gland,
FACING PAGE FIG. 1. Parotid glands immunostained for VIP, SP, CGRP and NPY. Sections from control (A, C, E and F) and parasympathectomized (B and D) glands. In the control gland numerous VIP-containing nerve fibers are seen around acini (A); these fibers disappeared after parasympathectomy (B). Substance P-containing nerve fibers are seen around a duct (centrally in figure) and in small numbers close to acini (C); the number is reduced after parasympathectomy (remaining nerve fiber indicated by arrow, D). A few CGRP-containing nerve fibers occur close to acini (E). NPY-containing nerve fibers are seen in close association with small blood vessels (centrally in figure, F). (A, B, C and F, × 222; D and E, × 267,)
PEPTIDERGIC NERVES IN SALIVARY GLANDS
865
A
(3
L
F
866
T O B I N , LUTS, S U N D L E R A N D EKSTR(3M
FIG. 2. Submandibular glands immunostained for VIP, SP, CGRP, NPY and DBH. Numerous VIP-containing nerve fibers surround acini and ducts (A). Many nerve cell bodies in small ganglia within parenchyma display VIP immunofluorescence (B). Moderate number of SP-containing nerve fibers occur close to acini and ducts (C). CGRP-containing nerve fibers are seen within a nerve bundle close to a blood vessel (note autofluorescence in blood vessel intima, D). NPY-containing nerve fibers associated with a blood vessel [E). Numerous DBH-containing nerve fibers close to acini, ducts and blood vessels (centrally in figure, F). (A and C, x 222; B, D, E and F, × 267).
PEPTIDERGIC NERVES IN SALIVARY GLANDS
867
thus showing that the auriculo-temporal nerve seems to be the main pathway for VIP-containing nerve fibers, while nerve fibers containing SP and CGRP also reach the parotid gland via other pathways. It was not possible to demonstrate any changes in the innervation of the ferret submandibular glands after dissecting the chorda nerve deep into hilus. That this attempt to denervate met with no success was probably due to the fact that many ganglia remained within the parenchyma. Sympathetic denervation caused the disappearance of the NPY-containing nerve fibers. In the rat, on the other hand, sympathetic denervation eliminates only a proportion of the NPY-containing fibers, namely those around blood vessels [(9), cf. (14)]. The remaining fibers are probably parasympathetic and store in addition VIP and SP (14). The nature
of few DBH-containing nerve fibers remaining after sympathetic denervation in the ferret is unknown. In salivary glands the adrenergic innervation, as studied by the fluorescence method of Falck and Hillarp and more recently using immunocytochemistry, has been found to vary markedly (8,10). Of the ferret major salivary glands, the sublingual glands seemed to have the least dense adrenergic nerve supply as judged by the marker DBH. ACKNOWLEDGEMENTS This study was supported by grants from the Swedish Medical Council (project No. 4499 and 5927), Nordisk Insulinfond, Dir. A. P~hlsson's foundation and the Medical and Odontological Faculties in Lund.
REFERENCES 1. Costa, M.; Furness, J. B.; Gibbins, I. L. Chemical coding of enteric neurons. Prog. Brain Res. 68:217-239; 1986. 2. Edwards, A. V.; EkstrSm, J.; Tobin, G. Responses to stimulation of the parasympathetic innervation to the submandibular gland in the anaesthetized ferret. J. Physiol. 414:29P; 1989. 3. Ekstr6m, J.; Ekman, R.; H~tkanson, R.; Luts, A.; Sundler, F.; Tobin, G. Effects of capsaicin pretreatment on neuropeptides and salivary secretion of rat parotid glands. Br. J. Pharmacol. 97:1031-1038; 1989. 4. Ekstr6m, J.; Ekman, R.; H~akanson, R.; Sj/Sgren, S.; Sundler, F. Calcitonin gene-related peptide in rat salivary glands: neuronal localization, depletion upon nerve stimulation, and effects on salivation in relation to substance P. Neuroscience 26:933-949; 1988. 5. Ekstr6m, J.; H~tkanson, R.; M~nsson, B.; Tobin, G. Tachykinininvolvement in parasympathetic nerve-evoked salivation of the ferret. Br. J. Pharmacol. 94:707-712; 1988. 6. Ekstr6m, J.; M~nsson, B.; Olgart, L.; Tobin, G. Non-adrenergic, non-cholinergic salivary secretion in the ferret. Q. J. Exp. Physiol. 73:163-173; 1988. 7. Ekstr6m, J.; Tobin, G. Secretion of protein from salivary glands in the ferret in response to vasoactive intestinal peptide. J. Physiol. 415: 131-141; 1989. 8. Garrett, J. R. Adventures with the autonomic nerves. Perspectives in salivary glandular innervation. Proc. R. Micr. Soc. 17:242-253; 1982.
9. Leblanc, G. G.; Landis, S. C. Target specificity of neuropeptide Y-immunoreactive cranial parasympathetic neurons. J. Neurosci. 8:146-155; 1988. 10. Lundberg, J. M.; Martling, C.-R.; H/kkfelt, T. Airways, oral cavity and salivary glands: classical transmitters and peptides in sensory and autonomic motor neurons. In: Bj/Srklund, A.; H6kfelt, T.; Owman, C., eds. Handbook of chemical neuroanatomy. Amsterdam: Elsevier Science Publishers B. V.; 1988:391 444. 11. Luts, A.; Sundler, F. Peptide-containing nerve fibers in the respiratory tract of the ferret. Cell Tissue Res. 258:259-267; 1989. 12. Smaje, L. H.; Edwards, A. V. Nonadrenergic innervation of salivary gland blood vessels. In: Burnstock, G.; Griffith, S. G., eds. Nonadrenergic innervation of blood vessels, vol. 2, Regional innervation. Boca Raton, FL: CRC Press; 1988:173-189. 13. Sundler, F.; Edvinsson, L.; Ekblad, E.; H/tkanson, R.; Uddman, R. Existence and coexistence of peptides in perivascular nerve fibres. In: Nobin, A.; Owman, C.; Arneklo-Nobin, B., eds. Neuronal messengers in vascular function. Amsterdam: Elsevier Science Publishers B. V.; 1987:341-353. 14. Sundler, F.; Ekblad, E.; Grunditz, T.; HLkanson, R.; Luts, A.; Uddman, R. NPY in peripheral non-adrenergic neurons. In: Mutt, V.; Fuxe, K.; H6kfelt, T.; Lundberg, J. M., eds. Neuropeptide Y. New York: Raven Press; 1989:93-102.
0196-9781/90 $3.00 + .00
BRIEF COMMUNICATION
Peptidergic Innervation of the Major Salivary Glands of the Ferret G. T O B I N , * A. L U T S , t F. S U N D L E R t A N D J. E K S T R O M * ~ : 1
Departments of *Physiology and ~Medical Cell Research, Lund University and ~Department of Pharmacology, Gothenburg University, Sweden R e c e i v e d 4 January 1990
TOBIN, G., A. LUTS, F. SUNDLER AND J. EKSTROM. Peptidergic innervation of the major salivary glands of the ferret. PEPTIDES 11(4) 863-867, 1990.--In parotid, sublingual and submandibular glands of the ferret, morphological correlates were looked for, using immunocytochemistry, to previous physiological findings showing parasympathetic "atropine-resistant" salivary secretion and neuropeptide-evoked salivation in this species. Nerve fibers storing VIP were numerous in association with acini, ducts and blood vessels, while the number of fibers storing substance P was moderate and those containing CGRP and galanin few; also the number of NPY-containing fibers was low around acini and ducts but relatively high around vessels. Sympathectomy eliminated all NPY- and almost all dopamine [3-hydroxylase-containing fibers. Parasympathectomy of the parotid gland resulted in a total loss of the VIP-containing fibers, and a profound reduction in the number of substance P- and CGRP-containing fibers. Salivary glands DBH
Denervation
Immunocytochemistry
Substance P
VIP
CGRP
Galanin
NPY
THE ferret belongs to the species in which stimulation of the parasympathetic innervation of the salivary glands gives rise to "atropine-resistant" vasodilatation in the glands (2) as well as "atropine-resistant" secretion of saliva (6,7). In the salivary glands of this species exogenous substance P (SP) and vasoactive intestinal peptide (VIP) are secretagogues (6). In the present study, not only nerve fibers containing SP and VIP but also fibers containing calcitonin gene-related peptide (CGRP), enkephalin, galanin, gastrin-releasing peptide (GRP), neuropeptide Y (NPY) and somatostatin were looked for in the parotid, sublingual and submandibular glands of the ferret.
immunocytochemistry; the antibodies used are described elsewhere (11). The specificity of the immunostaining was tested by adding excess amount of the antigen to the antibodies (10-100 I~g of synthetic peptide per ml of diluted antibody). As a marker for adrenergic nerve fibers we used an antiserum against dopamine [3-hydroxylase (11). To reveal coexistence of neuropeptides in one and the same neuron a simultaneous double immunostaining technique was used (1,13). In the present study, peptide antibodies raised in guinea pigs (11) were paired with antibodies raised in rabbits and the labels were fluorescein isothiocyanate (FITC) and tetramethyl rhodamine isothiocyanate (TRITC), respectively (1).
METHOD
RESULTS
In pentobarbitone-anesthetized ferrets (45 mg/kg IP): a) the auriculo-temporal nerve was cut unilaterally aiming at parasympathetic denervation of the parotid gland (n = 2); b) the chordalingual nerve was cut unilaterally and the chorda nerve fibers were dissected deep into the hilus of the submandibular gland, where two distinct ganglia were identified, and removed together with nerve strands aiming at parasympathetic denervation (n = 4); and c) the superior cervical ganglion was removed unilaterally aiming at sympathetic denervation (n = 2). The animals were killed by an IP overdose of pentobarbitone 1-2 weeks later. The salivary glands of both sides were removed, weighed and processed for
The observations are summarized in Table 1 (GRP, somatostatin and enkephalin were not found) and visualized in Figs. 1 (parotid glands) and 2 (submandibular glands). In the submandibular gland, nerve cell bodies containing SP were regularly seen in ganglia located in the hilus region outside the parenchyma as well as in smaller ganglia within the parenchyma. These ganglia also contained numerous VIP-immunoreactive nerve cell bodies. In fact, such nerve cell bodies constituted the majority of the cell bodies in the ganglia and clearly outnumbered those storing SP. Double immunostaining showed that the majority of the CGRPcontaining nerve fibers stored also SP. Further, double immuno-
1Requests for reprints should be addressed to J. Ekstr6m, Department of Physiology, Lund University, S61vegatan 19, 223 62 Lund, Sweden.
863
864
TOBIN, LUTS, SUNDLER AND EKSTROM
TABLE 1 RELATIVE FREQUENCY OF PEPTIDE- AND DBH-CONTAINING NERVE FIBERS AROUND ACIN1, DUCTS AND BLOOD VESSELS IN PAROTID, SUBMANDIBULARAND SUBLINGUALGLANDS OF THE FERRET Parotid Gland
VIP SP CGRP Galanin NPY DBH
Submandibular Gland
Sublingual Gland
Acini
Ducts
Vessels
Acini
Ducts
Vessels
Acini
Ducts
Vessels
+++ ++ +* (+) (+) +++
+++ +* (+) (+) (+) +++
+++ ++ + (+) ++ +++
+++ +++ + (+) (+) +++
+++ +++ + (+) (+) +++
+++ +++ + (+) ++ +++
+++ + (+) (+) (+) ++
+++ ++ (+) (+) (+) ++
+++ ++ (+) (+) ++ +++
*Mostly within nerve bundles. Nerve fiber frequency: (+) very few; + few; + + moderate number; + + + numerous.
staining also revealed that although many fibers stored either VIP or SP, a minor population of nerve fibers around ducts, blood vessels and occasionally around acini in all three glands stored both VIP and SP. Coexistence of SP and VIP was observed also in a few nerve cell bodies in the hilus ganglia of the submandibular gland. Nerve cell bodies in the submandibular gland were surrounded by nerve fibers containing galanin, NPY and DBH. No ganglia were found in the sublingual or parotid glands. Cutting the parasympathetic postganglionic nerve of the parotid gland resulted in an almost total loss of VIP-containing nerve fibers. The number of SP-containing nerve fibers was reduced as was the number of CGRP-containing fibers. However, the reduction was not as marked as that of VIP-containing nerve fibers. The number of NPY-, DBH- and galanin-containing nerve fibers seemed unaffected by the denervation procedure. Parasympathetic postganglionic denervation of the submandibular glands was attempted. However, neither the density nor the distribution of any of the neuropeptide-containing nerve fiber populations or the DBHcontaining nerve fibers seemed affected. Furthermore, many ganglia remained within the parenchyma. Removal of the superior cervical ganglion eliminated the NPY-containing nerve fibers and almost all the DBH-containing nerve fibers in the parotid, submandibular and sublingual glands. Those DBH-containing nerve fibers persisting were mainly found around blood vessels. Occasionally such persisting fibers occurred also around acini and ducts. Nerve fibers containing VIP, SP, CGRP and galanin were seemingly unaffected. DISCUSSION The present study offers morphological correlates to our earlier physiological findings. SP injected intravenously elicits a copious flow of saliva from the parotid and submandibular glands in the ferret (6), while VIP evokes protein release but no flow of saliva (7). VIP in combination with SP potentiates the responses both with respect to fluid and protein secretion (7). It is therefore of interest that VIP and SP were found to coexist in some nerve fibers within the glands. Stimulation of the parasympathetic innervation releases SP, VIP and CGRP (Bloom, Edwards, Ekstr6m and
Tobin, unpublished observation). In the presence of muscarinic receptor blockade, the secretion of saliva in response to parasympathetic stimulation that persists is abolished by tachykinin antagonists (5). In the ferret CGRP-containing nerve fibers also appeared close to secretory elements albeit in small numbers. This peptide elicits no flow of saliva from the glands in this species (Ekstr6m and Tobin, unpublished observation) as is the case in rat salivary glands (4). However, it enhances the secretory effects of parasympathomimetic drugs in the rat and, further, it releases proteins (4). A majority of the CGRP-containing nerve fibers did also store SP; in the rat salivary glands, this type of nerve is sensitive to capsaicin treatment, suggesting a sensory origin (3). A few scattered fibers containing galanin were observed. The functional significance of galanin in the ferret salivary glands is unknown. Atropine-resistant vasodilatation in salivary glands upon stimulation of the parasympathetic innervation is a well-known phenomenon, and VIP is thought to be a mediator (12). Not unexpectedly, blood vessels (10) in the ferret salivary glands were innervated not only by VlP-containing fibers but also by fibers storing SP and CGRP. Of the latter two peptides, CGRP is known to be a particularly potent vasodilator agent; in salivary glands of the rabbit, its effect is additive to the vasodilator response evoked by acetylcholine (12). NPY was found particularly around blood vessels, but a few nerve fibers also occurred close to the secretory elements. Dopamine [3-hydroxylase (DBH) was used as an immunocytochemical marker for adrenergic nerves. DBH-containing nerve fibers showed a partly different pattern of distribution; there was a rich innervation not only of blood vessels but also of acini and ducts. This pattern of distribution, which is valid for carnivores but not for rodents (10,14), may favor the view that a separate population of sympathetic nerve fibers is responsible for vasoconstriction in the gland and another for secretory effects. Sympathetic nerve stimulation as well as adrenaline, injected intravenously, gives rise to a sparse protein rich secretion of saliva from the parotid and submandibular glands of the ferret (Ekstr6m and Tobin, unpublished observation). Parasympathetic denervation of the parotid gland eliminated virtually all the VIP-containing nerve fibers of the parotid glands. There was also a reduction in the number of SP- and CGRP-containing nerve fibers in this gland,
FACING PAGE FIG. 1. Parotid glands immunostained for VIP, SP, CGRP and NPY. Sections from control (A, C, E and F) and parasympathectomized (B and D) glands. In the control gland numerous VIP-containing nerve fibers are seen around acini (A); these fibers disappeared after parasympathectomy (B). Substance P-containing nerve fibers are seen around a duct (centrally in figure) and in small numbers close to acini (C); the number is reduced after parasympathectomy (remaining nerve fiber indicated by arrow, D). A few CGRP-containing nerve fibers occur close to acini (E). NPY-containing nerve fibers are seen in close association with small blood vessels (centrally in figure, F). (A, B, C and F, × 222; D and E, × 267,)
PEPTIDERGIC NERVES IN SALIVARY GLANDS
865
A
(3
L
F
866
T O B I N , LUTS, S U N D L E R A N D EKSTR(3M
FIG. 2. Submandibular glands immunostained for VIP, SP, CGRP, NPY and DBH. Numerous VIP-containing nerve fibers surround acini and ducts (A). Many nerve cell bodies in small ganglia within parenchyma display VIP immunofluorescence (B). Moderate number of SP-containing nerve fibers occur close to acini and ducts (C). CGRP-containing nerve fibers are seen within a nerve bundle close to a blood vessel (note autofluorescence in blood vessel intima, D). NPY-containing nerve fibers associated with a blood vessel [E). Numerous DBH-containing nerve fibers close to acini, ducts and blood vessels (centrally in figure, F). (A and C, x 222; B, D, E and F, × 267).
PEPTIDERGIC NERVES IN SALIVARY GLANDS
867
thus showing that the auriculo-temporal nerve seems to be the main pathway for VIP-containing nerve fibers, while nerve fibers containing SP and CGRP also reach the parotid gland via other pathways. It was not possible to demonstrate any changes in the innervation of the ferret submandibular glands after dissecting the chorda nerve deep into hilus. That this attempt to denervate met with no success was probably due to the fact that many ganglia remained within the parenchyma. Sympathetic denervation caused the disappearance of the NPY-containing nerve fibers. In the rat, on the other hand, sympathetic denervation eliminates only a proportion of the NPY-containing fibers, namely those around blood vessels [(9), cf. (14)]. The remaining fibers are probably parasympathetic and store in addition VIP and SP (14). The nature
of few DBH-containing nerve fibers remaining after sympathetic denervation in the ferret is unknown. In salivary glands the adrenergic innervation, as studied by the fluorescence method of Falck and Hillarp and more recently using immunocytochemistry, has been found to vary markedly (8,10). Of the ferret major salivary glands, the sublingual glands seemed to have the least dense adrenergic nerve supply as judged by the marker DBH. ACKNOWLEDGEMENTS This study was supported by grants from the Swedish Medical Council (project No. 4499 and 5927), Nordisk Insulinfond, Dir. A. P~hlsson's foundation and the Medical and Odontological Faculties in Lund.
REFERENCES 1. Costa, M.; Furness, J. B.; Gibbins, I. L. Chemical coding of enteric neurons. Prog. Brain Res. 68:217-239; 1986. 2. Edwards, A. V.; EkstrSm, J.; Tobin, G. Responses to stimulation of the parasympathetic innervation to the submandibular gland in the anaesthetized ferret. J. Physiol. 414:29P; 1989. 3. Ekstr6m, J.; Ekman, R.; H~tkanson, R.; Luts, A.; Sundler, F.; Tobin, G. Effects of capsaicin pretreatment on neuropeptides and salivary secretion of rat parotid glands. Br. J. Pharmacol. 97:1031-1038; 1989. 4. Ekstr6m, J.; Ekman, R.; H~akanson, R.; Sj/Sgren, S.; Sundler, F. Calcitonin gene-related peptide in rat salivary glands: neuronal localization, depletion upon nerve stimulation, and effects on salivation in relation to substance P. Neuroscience 26:933-949; 1988. 5. Ekstr6m, J.; H~tkanson, R.; M~nsson, B.; Tobin, G. Tachykinininvolvement in parasympathetic nerve-evoked salivation of the ferret. Br. J. Pharmacol. 94:707-712; 1988. 6. Ekstr6m, J.; M~nsson, B.; Olgart, L.; Tobin, G. Non-adrenergic, non-cholinergic salivary secretion in the ferret. Q. J. Exp. Physiol. 73:163-173; 1988. 7. Ekstr6m, J.; Tobin, G. Secretion of protein from salivary glands in the ferret in response to vasoactive intestinal peptide. J. Physiol. 415: 131-141; 1989. 8. Garrett, J. R. Adventures with the autonomic nerves. Perspectives in salivary glandular innervation. Proc. R. Micr. Soc. 17:242-253; 1982.
9. Leblanc, G. G.; Landis, S. C. Target specificity of neuropeptide Y-immunoreactive cranial parasympathetic neurons. J. Neurosci. 8:146-155; 1988. 10. Lundberg, J. M.; Martling, C.-R.; H/kkfelt, T. Airways, oral cavity and salivary glands: classical transmitters and peptides in sensory and autonomic motor neurons. In: Bj/Srklund, A.; H6kfelt, T.; Owman, C., eds. Handbook of chemical neuroanatomy. Amsterdam: Elsevier Science Publishers B. V.; 1988:391 444. 11. Luts, A.; Sundler, F. Peptide-containing nerve fibers in the respiratory tract of the ferret. Cell Tissue Res. 258:259-267; 1989. 12. Smaje, L. H.; Edwards, A. V. Nonadrenergic innervation of salivary gland blood vessels. In: Burnstock, G.; Griffith, S. G., eds. Nonadrenergic innervation of blood vessels, vol. 2, Regional innervation. Boca Raton, FL: CRC Press; 1988:173-189. 13. Sundler, F.; Edvinsson, L.; Ekblad, E.; H/tkanson, R.; Uddman, R. Existence and coexistence of peptides in perivascular nerve fibres. In: Nobin, A.; Owman, C.; Arneklo-Nobin, B., eds. Neuronal messengers in vascular function. Amsterdam: Elsevier Science Publishers B. V.; 1987:341-353. 14. Sundler, F.; Ekblad, E.; Grunditz, T.; HLkanson, R.; Luts, A.; Uddman, R. NPY in peripheral non-adrenergic neurons. In: Mutt, V.; Fuxe, K.; H6kfelt, T.; Lundberg, J. M., eds. Neuropeptide Y. New York: Raven Press; 1989:93-102.
