Sensory nerve fibers containing calcitonin gene-related peptide in gastrocnemius, latissimus dorsi and erector spinae muscles and thoracolumbar fascia in mice.

Chronic pain is a significant burden and much is attributed to back muscles. Back muscles and their associated fasciae make important and distinct contributions to back pain. Peptidergic nociceptors innervating these structures contribute to central transmission and pain modulation by peripheral and central actions. Plastic changes that augment and prolong pain are exhibited by neurons containing calcitonin gene-related peptide (CGRP) following muscle injury. Subpopulations of neurons containing this peptide have been identified in dorsal root ganglia but the distribution of their fibers in skeletal muscles and associated fasciae has not been fully documented. This study used multiple-labeling immunofluorescence and retrograde axonal tracing to identify dorsal root ganglion cells associated with muscle, and to characterize the distribution and density of their nerve fibers in mouse gastrocnemius and back muscles and in the thoracolumbar fascia. Most nerve fibers in these tissues contained CGRP and two major subpopulations of neurons were found: those containing CGRP and substance P (SP) and those containing CGRP but not SP. Innervation density was three times higher in the thoracolumbar fascia than in muscles of the back. These studies show mouse back and leg muscles are predominantly innervated by neurons containing CGRP, an important modulator of pain signal transmission. There are two distinct populations of neurons containing this peptide and their fibers were three times more densely distributed in the thoracolumbar fascia than back muscles.

MicroRNA-143 expression in dorsal root ganglion neurons.

The unpleasant sensory and emotional experience of pain is initiated by excitation of primary afferent nociceptive neurons. Nerve damage or inflammation induces changes in nociceptive DRG neurons which contribute to both peripheral and central sensitization of pain-sensitive pathways. Recently, blockade of microRNA synthesis has been found to modulate the response of nociceptive neurons to inflammatory stimuli. However, little is known about the contributions of individual miRNAs to painful conditions. We compared miRNA expression in mouse sensory neurons and focussed on the localisation and control of miR-143. Using miRNA-arrays we compared the microRNA expression profile of intact lumbar DRG with one-day-old DRG cultures and found that nine miRNAs including miR-143 showed lower expression levels in cultures. Subsequent RT-qPCR confirmed array data and in-situ hybridisation localised miR-143 in the cytosol of sensory DRG neurons in situ and in vitro. Analysis of microbead-enriched neuron cultures showed significantly higher expression levels of miR-143 in isolectin B4 (I-B4) binding sensory neurons compared with neurons in the I-B4 negative flow-through fraction. In animal models of peripheral inflammation (injection of Complete Freund's Adjuvant, CFA) and nerve damage (transection of the sciatic nerve), we found that expression levels of miR-143 were significantly lower in DRGs ipsilateral to CFA injection or after nerve damage. Taken together, our data demonstrate for the first time miR-143 expression in nociceptive neurons. Since expression levels of miR-143 were higher in I-B4 positive neurons and declined in response to inflammation but not axotomy, miR-143 could selectively contribute to mRNA regulation in specific populations of nociceptors.

The cholinergic system in rat testis is of non-neuronal origin.

The cholinergic system consists of acetylcholine (ACh), its synthesising enzyme, choline acetyltransferase (CHAT), transporters such as the high-affinity choline transporter (SLC5A7; also known as ChT1), vesicular ACh transporter (SLC18A3; also known as VAChT), organic cation transporters (SLC22s; also known as OCTs), the nicotinic ACh receptors (CHRN; also known as nAChR) and muscarinic ACh receptors. The cholinergic system is not restricted to neurons but plays an important role in the structure and function of non-neuronal tissues such as epithelia and the immune system. Using molecular and immunohistochemical techniques, we show in this study that non-neuronal cells in the parenchyma of rat testis express mRNAs for Chat, Slc18a3, Slc5a7 and Slc22a2 as well as for the CHRN subunits in locations completely lacking any form of innervation, as demonstrated by the absence of protein gene product 9.5 labelling. We found differentially expressed mRNAs for eight α and three ß subunits of CHRN in testis. Expression of the α7-subunit of CHRN was widespread in spermatogonia, spermatocytes within seminiferous tubules as well as within Sertoli cells. Spermatogonia and spermatocytes also expressed the α4-subunit of CHRN. The presence of ACh in testicular parenchyma (TP), capsule and isolated germ cells could be demonstrated by HPLC. Taken together, our results reveal the presence of a non-neuronal cholinergic system in rat TP suggesting a potentially important role for non-neuronal ACh and its receptors in germ cell differentiation.

Muscarinic receptor subtypes in cilia-driven transport and airway epithelial development.

Ciliary beating of airway epithelial cells drives the removal of mucus and particles from the airways. Mucociliary transport and possibly airway epithelial development are governed by muscarinic acetylcholine receptors but the precise roles of the subtypes involved are unknown. This issue was addressed by determining cilia-driven particle transport, ciliary beat frequency, and the composition and ultrastructural morphology of the tracheal epithelium in M1-M5 muscarinic receptor gene-deficient mice. Knockout of M3 muscarinic receptors prevented an increase in particle transport speed and ciliary beat frequency in response to muscarine. Furthermore, the ATP response after application of muscarine was blunted. Pretreatment with atropine before application of muscarine restored the response to ATP. Additional knockout of the M2 receptor in these mice partially restored the muscarine effect, most likely through the M1 receptor, and normalised the ATP response. M1, M4 and M5 receptor-deficient mice exhibited normal responses to muscarine. None of the investigated mutant mouse strains had any impairment of epithelial cellular structure or composition. In conclusion, M3 receptors stimulate whereas M2 receptors inhibit cilia-driven particle transport. The M1 receptor increases cilia-driven particle transport if the M3 and M2 receptors are missing. None of the receptors is necessary for epithelial development.

Reduced expression of nicotinic alpha subunits 3, 7, 9 and 10 in lesional and nonlesional atopic dermatitis skin but enhanced expression of alpha subunits 3 and 5 in mast cells.

BACKGROUND: The skin cholinergic signalling system is modulated in atopic dermatitis (AD). OBJECTIVES: To investigate of the role of nicotinic acetylcholine receptors (nAChRs) in the pathogenesis of AD. METHODS: We investigated the expression and localization of nAChR alpha subunits in AD by quantitative reverse transcription-polymerase chain reaction and immunohistochemistry of biopsies from lesional and nonlesional areas of AD skin and of skin biopsies from healthy control persons. RESULTS: Our data demonstrate the presence of mRNA and protein of the nAChR alpha subunits 3, 5, 7, 9 and 10 in keratinocytes and mast cells in healthy and AD skin. Expression of the alpha subunits 3, 7, 9 and 10 was generally reduced in the skin of patients with AD whereas mast cells in AD but not in healthy skin showed alpha3 and alpha5 subunit immunoreactivity. Differences in the subunit mRNA levels between lesional and nonlesional skin were obtained for the alpha subunits 3, 9 and 10 with higher levels of alpha3 but lower levels of alpha10 subunit mRNA in lesional areas. No differences in the expression of the alpha subunits was found between the groups of extrinsic, intrinsic or mixed AD types, between genders and between smokers and nonsmokers. CONCLUSIONS: This supports the idea that the cholinergic system is dysregulated independently from inflammation in AD and that inflammation further modulates individual nAChR subunits.

Hypoxic increase in nitric oxide generation of rat sensory neurons requires activation of mitochondrial complex II and voltage-gated calcium channels.

Recently, we have demonstrated that sensory neurons of rat lumbar dorsal root ganglia (DRG) respond to hypoxia with an activation of endothelial nitric oxide (NO) synthase (eNOS) resulting in enhanced NO production associated with mitochondria which contributes to resistance against hypoxia. Extracellular calcium is essential to this effect. In the present study on rat DRG slices, we set out to determine what types of calcium channels operate under hypoxia, and which upstream events contribute to their activation, thereby focusing upon mitochondrial complex II. Both the metallic ions Cd2+ and Ni2+, known to inhibit voltage-gated calcium channels and T-type channels, respectively, and verapamil and nifedipine, typical blocker of L-type calcium channels completely prevented the hypoxic neuronal NO generation. Inhibition of complex II by thenoyltrifluoroacetone at the ubiquinon binding site or by 3-nitropropionic acid at the substrate binding site largely diminished hypoxic-induced NO production while having an opposite effect under normoxia. An additional blockade of voltage-gated calcium channels entirely abolished the hypoxic response. The complex II inhibitor malonate inhibited both normoxic and hypoxic NO generation. These data show that complex II activity is required for increased hypoxic NO production. Since succinate dehydrogenase activity of complex II decreased at hypoxia, as measured by histochemistry and densitometry, we propose a hypoxia-induced functional switch of complex II from succinate dehydrogenase to fumarate reductase, which subsequently leads to activation of voltage-gated calcium channels resulting in increased NO production by eNOS.

Nicotinic receptor mediated stimulation of NO-generation in neurons of rat thoracic dorsal root ganglia.

The contribution of nicotinic acetylcholine receptors (nAChRs) to stimulation of NO-production was investigated in isolated rat dorsal root ganglion (DRG) neurons utilizing an NO-sensitive fluorescent indicator 4,5-diaminofluorescein-diacetate (DAF-2DA) and appropriate channel blockers. RT-PCR and immunohistochemical analysis of NOS isoforms in cultured neurons revealed the expression of eNOS in the vast majority of neurons, nNOS in about 5-10%, and iNOS only exceptionally. Application of nicotine resulted in an abrupt increase in DAF-2T fluorescence in 65% of neuronal cell bodies that was fully sensitive to the general nAChR antagonist mecamylamine. Methyllycaconitine reduced the number of nicotine-sensitive neurons and the extent of NO-generation. Thus, alpha7- and/or alpha9/10-nAChRs are required for nicotine-induced NO-production in a subpopulation of DRG neurons, and appear to be partially involved in the remaining, larger subpopulation.

Expression and distribution of calcitonin receptor-like receptor in human hairy skin.

Calcitonin gene-related peptide and adrenomedullin exert potent effects in skin but their cellular targets are unknown. This study aimed to identify the cellular location of calcitonin receptor-like receptor (CRLR) which is pharmacologically identical to CGRP receptor-1, a putative molecular target of CGRP and adrenomedullin. RT-PCR analysis of human hairy skin revealed the presence of CRLR mRNA and immunohistochemical analysis, employing a previously characterized polyclonal antibody raised to CRLR, provided novel evidence of the cellular distribution of CRLR. Extensive and specific CRLR-immunostaining was detected in arteriolar smooth muscle and venular endothelium and is consistent with CGRP's putative role in neurogenic inflammation. Novel targets for CGRP and/or adrenomedullin were identified, including capillary endothelium, hair follicles and sweat glands.

Immunohistochemical detection of calcitonin gene-related peptide receptor (CGRPR)-1 in the endothelium of human coronary artery and bronchial blood vessels.

The potent vasodilatory peptide, calcitonin gene-related peptide (CGRP) is present in the innervation of vascular tissue. The actions of CGRP occur via a receptor, CGRP receptor(R)-1, which is also a target for the cardioprotective peptide adrenomedullin. The human version of the pharmacologically-defined CGRPR-1 has been cloned but its distribution and cellular location is unknown. A rabbit antibody was generated to a synthetic peptide that corresponds to the C-terminus of human CGRPR-1 Immunochemical analysis of the human cell-line, SK-N-MC, which exhibits functional expression of the CGRPR-1 confirmed the antibody's specificity. The antiserum revealed specific staining in the endothelium of human coronary arteries. The vascular smooth muscle and ventricular myocardium were not immunoreactive. In bronchial blood vessels CGRPR-1-immunoreactivity was detected in the endothelium of the venules and not in the arterioles, which is particularly relevant for elucidating the putative role of CGRP in inflammation in this tissue.

Excitatory nicotinic and desensitizing muscarinic (M2) effects on C-nociceptors in isolated rat skin.

The actions of different cholinergic agonists and antagonists were investigated on nociceptive afferents using the rat skin-saphenous nerve preparation, in vitro. Nicotine was able to weakly excite C-nociceptors and to induce a mild sensitization to heat stimulation (in 77% of tested fibers) in a dose-dependent manner (10(-)6 to 10(-)5 m), but it caused no alteration in mechanical responsiveness tested with von Frey hairs. Muscarine did not induce a significant nociceptor excitation, but almost all fibers exhibited a marked desensitization to mechanical and heat stimuli in a dose-dependent manner (from 10(-)6 to 10(-)4 m). The muscarinic effects could be prevented by the general muscarinic antagonist scopolamine (10(-)5 m), by the M3 antagonist 1,1-dimethyl-4-diphenylacetoxypiperidium oxide (10(-)6 m) co-applied with the M2 antagonist gallamine (10(-)5 m), and by gallamine alone. As positive control we used the relatively M2-selective agonist arecaidine (10(-)6 to 10(-)5 m), obtaining a similar desensitizing effect as with muscarine. Finally, we performed an immunocytochemical study that demonstrated the presence of M2 but not M3 receptors in thin epidermal nerve fibers of the rat hairy skin. Altogether, these data demonstrate opposite effects of nicotinic and muscarinic receptor stimulation on cutaneous nociceptors. M2 receptor-mediated depression of nociceptive responsiveness may convey a therapeutic, i.e., analgesic or antinociceptive, potential.

M2-receptor subtype does not mediate muscarine-induced increases in [Ca(2+)](i) in nociceptive neurons of rat dorsal root ganglia.

Multiple muscarinic receptor subtypes are present on sensory neurons that may be involved in the modulation of nociception. In this study we focused on the presence of the muscarinic receptor subtypes, M2 and M3 (M2R, M3R), in adult rat lumbar dorsal root ganglia (DRG) at the functional ([Ca(2+)](i) measurement), transcriptional (RT-PCR), and translational level (immunohistochemistry). After 1 day in culture exposure of dissociated medium-sized neurons (20-35 micrometer diam) to muscarine was followed by rises in [Ca(2+)](i) in 76% of the neurons. The [Ca(2+)](i) increase was absent after removal of extracellular calcium and did not desensitize after repetitive application of the agonist. This rise in [Ca(2+)](i) may be explained by the expression of M3R, which can induce release of calcium from internal stores via inositoltrisphospate. Indeed the effect was antagonized by the muscarinic receptor antagonist atropine as well as by the M3R antagonist, 4-diphenylacetoxy-N-(2 chloroethyl)-piperidine hydrochloride (4-DAMP). The pharmacological identification of M3R was corroborated by RT-PCR of total RNA and single-cell RT-PCR, which revealed the presence of mRNA for M3R in lumbar DRG and in single sensory neurons. In addition, RT-PCR also revealed the expression of M2R, which did not seem to contribute to the calcium changes since it was not prevented by the M2 receptor antagonist, gallamine. Immunohistochemistry demonstrated the presence of M2R and M3R in medium-sized lumbar DRG neurons that also coexpressed binding sites for the lectin I-B4, a marker for mainly cutaneous nociceptors. The occurrence of muscarinic receptors in putative nociceptive I-B4-positive neurons suggests the involvement of these acetylcholine receptors in the modulation of processing of nociceptive stimuli.

Immunohistochemical localization of muscarinic receptors (M2) in the rat skin.

Acetylcholine (ACh) produces pain when applied to human skin and excites cutaneous mechanoreceptors and nerve terminals. These effects are partially mediated by activation of muscarinic receptors. The expression of muscarinic receptor subtype M2 has been shown in sensory neurons of rat dorsal root ganglia using reverse transcriptase polymerase chain reaction (RT-PCR), in situ hybridization and immunohistochemistry. The purpose of the present study was to determine whether these M2 receptors are targeted to the peripheral endings of sensory neurons in the rat skin. Double-staining histochemical procedures were employed using a specific antiserum to M2 receptors combined with either of the following neuronal markers: an antiserum to the neuropeptide substance P, an antiserum to the protein gene product 9.5, which is a marker for peripheral nerve fibres, and the histochemical marker of a subpopulation of unmyelinated C-fibre afferents, I-B4, the Bandeira simplicifolia-derived isolectin. The M2 receptor subtype was found on different populations of nerve fibres. In the nerve plexus at the epidermal-dermal junction, M2 receptors are mainly present on I-B4-positive axons but are absent on fibres with substance P immunoreactivity. Sweat glands receive M2-receptor-immunoreactive fibres that express neither I-B4 binding nor substance P immunoreactivity, whereas blood vessels of the deeper dermis are innervated by I-B4-positive nerve fibres that are immunoreactive for M2 receptors and substance P. In addition to axon profiles, keratinocytes and endothelial cells also exhibit M2 receptor immunoreactivity. The results show the presence of M2 receptors in neuronal and non-neuronal cells, suggesting multiple effects of acetylcholine in the skin.

Expression of the cholinergic gene locus in pulmonary arterial endothelial cells.

In the pulmonary vasculature of man, pig and guinea-pig, acetylcholine (ACh) exerts a relaxant effect by interacting with muscarinic receptors located on endothelial cells. The present experiments were designed to detect the endogenous source of ACh in the pulmonary vasculature. For this purpose, we investigated whether pulmonary artery endothelial cells contain elements of the "cholinergic gene locus", the ACh synthesising enzyme choline acetyltransferase (ChAT) and the vesicular ACh transporter (VAChT). ChAT mRNA was detected by means of reverse transcription polymerase chain reaction (RT-PCR) in endothelial cells of porcine pulmonary arteries freshly isolated and after 7 days in culture. ChAT protein was demonstrated in endothelial cells in vitro and in situ. ChAT immunoreactivity was present in endothelial cells freshly isolated and after 2, 4, 7, and 9 days in culture. Tissue sections from extra- and intraparenchymal pulmonary arteries of man, pig and guinea-pig expressed a mosaic pattern of ChAT-positive and -negative endothelial cells. VAChT mRNA was detected by RT-PCR in rat pulmonary artery and in endothelial cells isolated from human and porcine pulmonary trunk. The detection of VAChT and ChAT mRNA and the demonstration of ChAT protein in vitro and in situ suggest that the endothelium is an endogenous source of ACh in the pulmonary vasculature.

[Recruitment of immunocompetent cells in a transplanted lung is not impaired despite incompletel reinnervation]. / Das Rekruitment immunkompetenter Zellen in eine transplantierte Lunge ist trotz inkompletter Reinnervierung nicht behindert.

BACKGROUND: It is not clear whether surgical intervention during lung transplantation which includes cutting vegetative nerves, lymphatic vessels and bronchial arteries, leads to alterations in immune responses. Thus, it was studied in an animal model whether an induced pulmonary immune reaction after syngenic lung transplantation was impaired without the influence of immunosuppression and rejection. The recruitment of leukocytes and the status of reinnervation was examined. METHODS: Syngenic transplantation of the left lung was performed in Lewis rats without rejection and therefore without immunosuppressive therapy. In a subgroup of animals host and donor leukocytes were distinguished. An ovalbumin (OVA)-specific pulmonary immune response was induced four months after transplantation. Bronchoalveolar lavage (BAL) and interstitial leukocytes were examined using flow cytometry and immunocytology, comparing the right lung and the grafted left lung. Immunohistology was performed to detect nerve fibers on cryostat sections. RESULTS: An induced cellular inflammation was observed in the right host lung as well as in the grafted left lung. However, the CD4 T cell numbers in the BAL were increased in the left lung. Single donor-type leukocytes could still be observed four months after transplantation. A partial reinnervation was found. CONCLUSIONS: The recruitment of immune cells into the lung interstitium and bronchoalveolar space of grafted lungs is not impaired. The incomplete reinnervation has no influence on leukocyte recruitment.

Extrinsic and intrinsic cholinergic systems of the vascular wall.

Immunohistochemical, biochemical and functional studies have revealed two separate cholinergic systems in the arterial vascular wall. Endothelial cells represent the ubiquitous intrinsic, intimal system; they contain the acetylcholine-synthesizing enzyme, choline acetyltransferase, release a choline ester, and contain functional muscarinic receptors. Perivascular autonomic nerve fibres represent the extrinsic, adventitial system. These axons are not ubiquitous but show a highly selective distribution among and even within organs, and utilize co-mediators (NO, neuropeptides) in an organ-specific pattern. We put forward the hypothesis that the intrinsic, intimal system serves as a general regulator of basal vascular tone and wall structure responding to local, luminal stimuli, whereas the perivascular nerve fibres act on top of this basal tone by providing fine tuning in response to reflex activation due to systemic demands.

Muscarinic M2-receptors in rat thoracic dorsal root ganglia.

The occurrence and distribution of the muscarinic M2-receptor subtype (M2R) was investigated in rat thoracic dorsal root ganglia (DRG). Messenger RNA for M2R was demonstrated by RT-PCR in total RNA from DRG. Immunoreactivity to M2R-protein was localized to 26% of sensory neurons, the majority of them (85%) belonging to the size class of 25-40 microm in diameter. Double-labeling (immuno)histochemistry revealed that all M2R-immunoreactive neurons bind the lectin, I-B4, whereas they are generally devoid of substance P-immunoreactivity. These data show the presence of M2R on a subpopulation of presumably nociceptive primary afferent neurons, thereby extending previous pharmacological and electrophysiological studies that indicated a role of M2R and/or M4R in inhibition of calcium channel currents in rat sensory neurons (Wanke, E., Bianchi, L., Mantegazza, M., Guatteo, E., Macinelli, E. and Ferroni, A., Muscarinic regulation of Ca2+ currents in rat sensory neurons: channel and receptor types, dose-response relationships and cross-talk pathways. Eur. J. Neurosci., 6 (1994) 381-391).

Smooth muscle cells are the site of neurokinin-1 receptor localization in the arterial supply of the rat sciatic nerve.

The occurrence and distribution of the preferred receptor for the neuropeptide, substance P (SP), the neurokinin-1 receptor (NK1R) was investigated in the vascular supply of the rat sciatic nerve. Messenger RNA for NK1R was demonstrated by RT-PCR in the epineurial layer where the majority of small arteries and arterioles feeding the endoneurial vasculature are located. Immunoreactivity to NK1 R-protein was localized on the smooth muscle cells of these arterial vessels by means of immunofluorescence using a polyclonal NK1R antiserum. This muscular localization of NK1R explains the previously reported [Zochodne, D.W. and Ho, L.T., J. Physiol., 444 (1991) 615-630] moderate vasoconstrictor rather than vasodilator effects of SP in this vascular bed.

Rat cardiac neurons express the non-coding R-exon (exon 1) of the cholinergic gene locus.

Choline acetyltransferase (ChAT) and the vesicular acetylcholine transporter (VAChT) are both encoded by the cholinergic gene locus from which, in the rat, five different species of ChAT mRNA and three different species of VAChT mRNA are produced. So far, discrimination between mRNA subtypes has been possible only in CNS homogenates or in cell cultures. In this study, cardiac neurons were microdissected from frozen sections of rat heart using a u.v. laser and harvested using a micromanipulator. RT-PCR demonstrated the expression of the non-coding R-exon and splicing to R1-type mRNA in the majority of cardiac neurons. The technique presented here is the first to allow subtype analysis of cholinergic locus mRNA species in neurons in situ.

Stimulation of neuropeptide Y release in rat pheochromocytoma cells by nitric oxide.

Neuropeptide Y and nitric oxide (NO) synthase are colocalized in nervous tissues. We tested the hypothesis whether or not NO might be involved in the release of neuropeptide Y. Neuropeptide Y concentration in the supernatant of PC12 rat pheochromocytoma cells, shown to express NO synthase I by immunohistochemistry, rose threefold in a time- and dose-dependent manner following sodiumnitroprusside and 3-morpholinosydnonimine (SIN-1) incubation. Neuropeptide Y mRNA expression was induced by NO-donors as a function of incubation-time. Neuropeptide Y production rose fivefold with zaprinast, an inhibitor of the phosphodiesterase V and threefold with nerve growth factor (NGF). Combined application of zaprinast and NGF did not further increase neuropeptide Y production while combination of zaprinast and sodiumnitroprusside potentiated the NO effect on neuropeptide Y release. The data suggest that NO regulates neuropeptide Y secretion of PC12 pheochromocytoma cells on the mRNA level.

Innervation pattern of guinea pig pulmonary vasculature depends on vascular diameter.

The pulmonary vasculature is supplied by various neurochemically distinct types of nerve fibers, including sensory substance P-containing and autonomic noradrenergic, nitrergic, and cholinergic axons. Pharmacological experiments have suggested that various segments of the pulmonary vascular tree respond differently to the respective neuromediators. We, therefore aimed to determine histochemically and immunohistochemically for each of these neurochemically distinct perivascular axons their quantitative distribution along the vascular tree from the extrapulmonary trunks to the smallest intraparenchymal ramifications in control guinea pigs (n = 5). Generally, arterial innervation was more developed than that of veins. Along the arterial tree, noradrenergic and substance P-containing axons were ubiquitous from the pulmonary trunk to smallest intraparenchymal vessels, whereas nitrergic axons were practically restricted to large (> 700-microns) extrapulmonary arteries. Cholinergic axons were regularly present at arteries down to 100 microns in diameter and innervated two-thirds of small arteries (50-100 microns). The results demonstrate that the noradrenergic vasoconstrictor innervation extends throughout the pulmonary vascular system whereas the innervation pattern with various types of vasodilator fibres changes with vascular diameter, parallel to known pharmacological differences in cholinergic and nitrergic vasodilator effects.