Proc. NatI. Acad. Sci. USA Vol. 77, No. 3, pp. 1651-1655, March 1980 Neurobiology Vasoactive intestinal polypeptide in cholinergic neurons of exocrine glands: Functional significance of coexisting transmitters for vasodilation and secretion (acetylcholinesterase/atropine-resistant vasodilation) JAN M. LUNDBERG*, ANDERS ANGGARDt, JAN FAHRENKRUG§, TOMAS HOKFELT*, AND VIKTOR MUTTf Departments of *Histology, tPharmacology, and tBiochemistry, Karolinska Institutet, Stockholm, Sweden; and §Departments of Clinical Chemistry, Glostrup and Bispebjerg Hospitals, Copenhagen, Denmark Communicated by Rolf Luft, October 18, 1979 ABSTRACT By a combination of the indirect immunoflu- in acetylcholinesterase (acetylcholine acetylhydrolase, EC orescence technique with acetylcholinesterase (acetylcholine 3.1.1.7) (AcChoE)-rich neurons innervating several exocrine acetylhydrolase, EC 3.1.1.7) staining, it was shown that vaso- active intestinal polypeptide (VIP) is present in cholinergic glands. Furthermore, pharmacological effects of VIP and (acetylcholinesterase-rich) neurons involved in control of se- AcCho as well as of various blocking agents, including VIP cretion and vasodilation in exocrine glands of cat. The sub- antiserum, on submandibular salivary secretion and vasodilation mandibular salivary gland was used as a functional model. have been studied. Release of immunoreactive VIP into the Preganglionic nerve stimulation induced an atropine-resistant, venous outflow from the submandibular gland after nerve hexamethonium-sensitive vasodilation and release of VIP into stimulation was the venous outflow from the gland and an atropine- and hexa- also demonstrated. methonium-sensitive secretion. Infusion of VIP antiserum re- duced both the vasodilation and secretion. Infusion of VIP MATERIALS AND METHODS caused vasodilation only, whereas acetylcholine caused both Ten cats (body weight 2-4 kg) of both sexes were used for im- vasodilation and secretion. Simultaneous infusion of VIP and munohistochemical studies. In two of these cats acetylcholine in low doses resulted in a marked potentiation of the superior both vasodilation and secretion. The present morphological and cervical ganglion was excised unilaterally 14 days before they functional data support the following hypothesis for regulation were killed. In four other cats vinblastine sulfate (Sigma) (1% of vasodilation and secretion in exocrine glands. Preganglionic in 0.9% NaCl) was locally applied to the submandibular and cholinergic nerves activate, via nicotinic receptors, postgan- sphenopalatine ganglia and to pancreas 24 hr before perfusion glionic neurons, causing concomitant release from the same with ice-cold 10% (vol/vol) formalin (21). These ganglia and nerve endings of two coexisting putative transmitters, acetyl- choline and VIP. Acetylcholine produces mainly secretion by parts of the submandibular gland, the nasal mucosa, tongue, a muscarinic action and VIP causes mainly vasodilation, but lung, and pancreas were then processed for indirect immuno- the two substances seem to cooperate directly or indirectly in histochemical studies according to Coons (22) as described (see both types of response. Thus, the coexistence of two putative ref. 18) by use of a VIP antibody (5603-5; for detail about this neurotransmitters, VIP and acetylcholine, in one neuron may antiserum, see refs. 23 and 24) and AcChoE staining (18, 25, explain the dual effector response (i.e., the cholinergic secretion In some and the atropine-resistant vasodilation) caused by nerve stim- 26). cases, immunohistochemical analysis for VIP and ulation in exocrine glands. AcChoE staining were performed on the same section. Alter- natively, adjacent sections were processed for the respective The vasodilation induced by nerve stimulation in exocrine procedure. The latter approach was used because Triton X-100 glands-e.g., submandibular salivary gland (1), tongue (2), (added to VIP antiserum to enhance penetration of antibodies) sweat glands (3), and nasal mucosa (4)-is largely atropine re- inactivated AcChoE (18, 27). For details and validity of pro- sistant, whereas secretion easily can be blocked by muscarinic cedures and for controls, see ref. 18. antagonists. The occurrence of a subpopulation of vasodilatory Physiological experiments were done in 15 cats anesthetized nerves was first suggested, but later a kinin mechanism was with chloralose (50 mg/kg) plus urethane (100 mg/kg) intra- suggested to be responsible for this vasodilation (5-9). The venously. The trachea was cannulated. The blood pressure was importance of a kallikrein-kinin activation has, however, been recorded via a femoral artery. Rectal temperature was kept questioned (see refs. 10 and 11). Another possibility has recently constant at 38°C. The submandibular gland, its duct, and nerve been proposed: Immunohistochemical studies indicate that supply were exposed. The duct was cannulated with a poly- vasoactive intestinal polypeptide (VIP), a potent vasodilatory ethylene tube (PE 50) and salivation was measured with a drop agent (12, 13) with a putative transmitter function (14, 15), is recorder. The chorda-lingual nerve was cut, and the peripheral present in nerves in exocrine glands (16, 17). Furthermore in cut end was mounted to a bipolar platinum electrode. Pre- cats, VIP seems to be present in cholinergic neurons in several ganglionic stimulation was performed with square wave pulses exocrine glands (18-20). It was therefore suggested that these of either 2 or 15 Hz, 8 V, and a duration of 1 msec. To measure nerves release both acetylcholine (AcCho) and VIP, causing the venous outflow from the gland, we introduced a polyeth- secretion and vasodilation, respectively (18). ylene catheter (PE 160) into the external jugular vein after all In the present paper we report further evidence for this hy- branches, except the one draining the submandibular gland, pothesis by demonstrating the presence of a VIP-like peptide had been tied. Blood flow was measured by a drop recorder and blood was reinfused into the cephalic vein with a peristaltic The publication costs of this article were defrayed in part by page pump. Heparin (1000 IU/kg) was given intravenously. The charge payment. This article must therefore be hereby marked "ad- vertisement" in accordance with 18 U. S. C. §1734 solely to indicate Abbreviations: VIP, vasoactive intestinal polypeptide; AcCho, ace- this fact. tylcholine; AcChoE, acetylcholinesterase. 1651 1652 Neurobiology: Lundberg et al. Proc. Natl. Acad. Sci. USA 77 (1980) facial artery was cannulated for close intra-arterial infusions corresponding to the Golgi apparatus (Fig. 1A). Vinblastine into the carotid artery. All branches of the carotid artery were pretreatment caused a marked increase in the amount of flu- ligated except the one supplying the submandibular gland. VIP orescent material, which occupied the entire cytoplasm and (purified from porcine intestine by V. Mutt) and AcCho chlo- extended into the cell processes (Fig. 1B). AcChoE activity was ride (Sigma) were dissolved in 50 mM phosphate-buffered sa- localized diffusely in the cytoplasm. Furthermore, both VIP- line (pH 7.4) containing 0.5% bovine serum albumin and in- positive (Fig. 1C) and AcChoE-rich (Fig. iF) nerve terminals fused intra-arterially at a constant rate of 50 Al/min for 2 min. with a similar distribution pattern were seen around the gland Atropine (0.5 mg/kg) and hexamethonium (20 mg/kg) were acini and ducts. Both VIP-immunoreactive and AcChoE-rich given intravenously 20 min prior to the various experiments. nerve endings were also seen around small blood vessels, mainly Finally, undiluted VIP antiserum was infused intra-arterially arteries and arterioles, in the gland tissues and in the skeletal at a rate of 50 Ml/min starting 10 min prior to and during low- muscle of the tongue. No definite effect on VIP-immuno- frequency (2 Hz) stimulation. Normal rabbit serum infused reactive or AcChoE-rich nerve endings was seen after sympa- under similar conditions served as control. As a further control thectomy. None of the fluorescent structures described above for the specificity of the VIP antibody effects, the ipsilateral was seen after incubation with VIP control serum; i.e., VIP cervical symphatic trunk was isolated, transected, and stimu- antiserum pretreated with an excess of VIP (15 nmol per ml lated in a cranial direction during infusion of VIP antiserum antiserum diluted 1:10). No AcChoE activity was observed in and control serum. The concentration of VIP in plasma (1.2 ganglion cells or nerve fibers after incubation with medium ml/sample) from the submandibular venous outflow was containing the specific AcChoE inhibitor BW284C51. measured by radioimmunoassay (23, 24). To compensate for Supramaximal stimulation (15 Hz) of preganglionic fibers blood loss during sampling for radioimmunoassay, we infused in the chorda-lingual nerve resulted in a more than 5-fold in- Macrodex (Pharmacia) intravenously. crease in blood flow from the submandibular gland accompa- nied by a marked salivary secretion (Fig. 2 C and B). Simul- RESULTS taneously, a marked increase in plasma VIP levels in the venous effluent from the gland was observed (n = 4). A typical example The morphological studies revealed that most neurons in local is seen in Fig. 2E, in which VIP increased from 2 pmol/liter to submandibular ganglia and almost all neurons in the spheno- a peak of 60 pmol/liter after approximately 1.5 min of stimu- palatine ganglion of normal and vinblastine-treated cats