The lungs are innervated by sympathetic and parasympathetic components of the autonomic nervous system. The role of autonomic nerves in the regulation of the blood flow in the lungs is comparatively poorly understood. Sensory, parasympathetic and sympathetic nerves regulate epithelial, vascular and glandular processes in the upper respiratory tract of humans. Pulmonary vasculature is innervated by sympathetic and parasympathetic nerves and non-adrenergic, non-cholinergic (NANC) neural pathways. Receptors for specific transmitters are present on endothelial cells and/or smooth muscle cells of the vasculature. Ganglia of sympathetic systems and parasympathetic systems are rather different. Parasympathetic ganglia are found along extrapulmonary and intrapulmonary airways and function as integrators and modulators of the autonomic tone in the airways.

Three neural networks are currently recognized in mammalian airways, namely the adrenergic, the cholinergic and the sensory efferent nerve systems. The nerves of these networks contain peptides, which mediate NANC neural effects. These peptides include vasoactive intestinal peptide (VIP) in parasympathetic nerves, neuropeptide Y (NPY) in sympathetic nerves and substance P, neurokinin A and calcitonin gene-related peptide (CGRP) in sensory nerves. Nitric oxide (NO) mediates vasodilator responses of a number of neurotransmitters, and inflammatory responses via receptors on endothelial cells.

Adult amphibians utilize different surfaces for respiration such as richly vascularized skins, lungs, the buccopharyngeal cavity and the gills in larval stages. Lungs are separated from the pharynx by a short tracheolarynx that is enriched with cartilage and muscles closing and opening the glottis (for a review, see Goniakowska-Witalinska, 2001). Their structure shows similarities with lung of lungfishes and polypterids and air-bladders of air-breathing fishes.

Sympathetic and parasympathetic fibers innervate lungs of amphibians. concluded that four types of innervation can be recognized in the amphibian lung: parasympathetic cholinergic, parasympathetic NANC, sympathetic adrenergic and sensorimotoric. However, the innervation of the lung of amphibians is not well studied and most studies have been carried out in Bufo marinus. Furthermore, there are no studies available of the neural regulation of secretion and ciliary activity in amphibian lungs. The mucociliary transport in the buccal cavity and esophagus is under cholinergic parasympathetic neural control.

In various anurans, nerve cell bodies are found abundantly in the wall of the lung and scattered along the vagosympathetic trunk. They are believed to be postganglionic neurons in vagal pathways. However, immunohistochemical markers have been tested, but so far did not provide positive results except for VIP, somatostatin and galanin that were demonstrated by using immunohistochemistry in nerve-cell bodies of Bufo marinus.

found NOS-immunopositive nerve-cell bodies in the lung of Rana temporaria. Moreover, colocalization of nNOS and NADPH-diaphorase (NADPH-d) activity in neurons of Cynops pyrrhogaster lung was reported by. The function of NO was correlated with inhibitory NANC neurotransmission in the pulmonary nervous system of urodeles.

Experimental studies of innervation patterns of visceral and vascular smooth muscles in lungs of a number of amphibian species have described various types of innervation, but so far immunohistochemical studies have not been performed. Despite the parasympathetic NANC innervation of both visceral and vascular smooth muscles in amphibian lungs and a possible involvement of VIP and NO in vagal inhibitory mechanisms, the nature of NANC innervation is still not clear. To study patterns of innervation in the respiratory tract of two frog species (Rana esculenta and Discoglossus pictus), immunohistochemistry was used to localize parasympathetic NANC innervation and adrenergic innervation of smooth muscles of the respiratory vasculature.

Ten specimens of the two frog species used in the present study were collected locally. Animals were maintained and killed in accordance with the guiding principles of animal care and experimentation and national laws and regulations. The Institutional Board of the Faculty of Science, Messina, Italy, approved the protocol.

Buccopharynx, larynx and lung of two species of the frog, Rana esculenta and Discoglossus pictus, were removed after anesthesia with a high dose of MS-222 (tricaine methanesulfonate; Argent, Redmond WA, USA) and immediately fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4, for 2–3 h, dehydrated and embedded in paraffin. Serial sections, 6- m thick, were cut in sagittal and transverse planes and mounted on glass slides for immunohistochemical staining. Sections were also stained with hematoxylin and eosin (H&E) to assess the general morphology.

For immunohistochemical staining, sections were processed for the indirect immunoperoxidase method as previously reported by using heterologous antisera. Details of the antisera are presented in Table 1. After inhibition of endogenous peroxidase activity using 1% H₂O₂ in PBS for 30 min, sections were incubated in the presence of primary antibodies overnight at 4°C in a moist chamber. Following incubation with primary antibodies, sections were treated with goat-anti-mouse IgG (dilution, 1:100; Chemicon, Temecula CA, USA) or a goat-anti-rabbit IgG peroxidase conjugate (dilution, 1:100; Sigma, St. Louis MO, USA). Peroxidase activity was visualized using 3,3?-diaminobenzidine tetrahydrochloride and H₂O₂ as substrates.

All antibodies were raised in rabbit except for the anti-tyrosine hydroxylase antibody and the antibody against calbindin D28K which were raised in mouse.

The specificity of immunostaining was established by substituting the first antibodies with normal serum or inhibition by preincubation of the primary antibodies with the respective pure antigens (10–100 Image g/ml).

Synthetic peptides and synthetic 5-hydroxytryptamine (5HT) creatinine sulfate were purchased from Sigma. Synthetic endothelin (ET)-big was obtained from Peninsula (Belmont CA, USA). The specificity of the nNOS antibodies was tested by preabsorption with the purified enzyme (50–100 ?g/ml) kindly donated by Dr. B. Mayer (Institute of Pharmacology and Toxicology, University of Graz, Austria). VIP (from chicken) was purchased from Peninsula.

The respiratory tract of Rana esculenta and Discoglossus pictus is connected with the alimentary tract via a larynx, which directly communicates with the lungs. The buccopharyngeal cavity is also considered as a respiratory surface and consists of a thick layer of connective tissue and respiratory epithelium with ciliated and mucous goblet cells. The larynx is supported by two pieces of lateral (arytenoid) cartilage, which are part of the vocal cords that are surrounded by cricoid cartilage. A fibromuscular layer is present between the arytenoid cartilage and epithelium. A laryngeal sphincter and dilator are present around the larynx in the cranial region. The laryngeal cavity is lined by pseudostratified ciliated mucous epithelium including the areas next to the aditus laryngis and the insertion of vocal cords into the laryngeal cavity.