The actions of neuropeptide Y and peptide YY on the hepatic arterial and portal vascular beds of the anaesthetized dog

Biliary epithelium: A neuroendocrine compartment in cholestatic liver disease

Hepatic fibrosis is characterized by abnormal accumulation of extracellular matrix (ECM) that can lead to ductopenia, cirrhosis, and even malignant transformation. In this review, we examine cholestatic liver diseases characterized by extensive biliary fibrosis such as primary sclerosing cholangitis (PSC), primary biliary cholangitis (PBC), polycystic liver disease (PLD), and MDR2-/- and BDL mouse models. Following biliary injury, cholangiocytes, the epithelial cells that line the bile ducts, become reactive and adopt a neuroendocrine phenotype in which they secrete and respond to neurohormones and neuropeptides in an autocrine and paracrine fashion. Emerging evidence indicates that cholangiocytes influence and respond to changes in the ECM and stromal cells in the microenvironment. For example, activated myofibroblasts and hepatic stellate cells are major drivers of collagen deposition and biliary fibrosis. Additionally, the liver is richly innervated with adrenergic, cholinergic, and peptidergic fibers that release neurohormones and peptides to maintain homeostasis and can be deranged in disease states. This review summarizes how cholangiocytes interact with their surrounding environment, with particular focus on how autonomic and sensory regulation affects fibrotic pathophysiology.

Neural connections between the hypothalamus and the liver

After receiving information from afferent nerves, the hypothalamus sends signals to peripheral organs, including the liver, to keep homeostasis. There are two ways for the hypothalamus to signal to the peripheral organs: by stimulating the autonomic nerves and by releasing hormones from the pituitary gland. In order to reveal the involvement of the autonomic nervous system in liver function, we focus in this study on autonomic nerves and neuroendocrine connections between the hypothalamus and the liver. The hypothalamus consists of three major areas: lateral, medial, and periventricular. Each area has some nuclei. There are two important nuclei and one area in the hypothalamus that send out the neural autonomic information to the peripheral organs: the ventromedial hypothalamic nucleus (VMH) in the medial area, the lateral hypothalamic area (LHA), and the periventricular hypothalamic nucleus (PVN) in the periventricular area. VMH sends sympathetic signals to the liver via the celiac ganglia, the LHA sends parasympathetic signals to the liver via the vagal nerve, and the PVN integrates information from other areas of the hypothalamus and sends both autonomic signals to the liver. As for the afferent nerves, there are two pathways: a vagal afferent and a dorsal afferent nerve pathway. Vagal afferent nerves are thought to play a role as sensors in the peripheral organs and to send signals to the brain, including the hypothalamus, via nodosa ganglia of the vagal nerve. On the other hand, dorsal afferent nerves are primary sensory nerves that send signals to the brain via lower thoracic dorsal root ganglia. In the liver, many nerves contain classical neurotransmitters (noradrenaline and acetylcholine) and neuropeptides (substance P, calcitonin gene-related peptide, neuropeptide Y, vasoactive intestinal polypeptide, somatostatin, glucagon, glucagon-like peptide, neurotensin, serotonin, and galanin). Their distribution in the liver is species-dependent. Some of these nerves are thought to be involved in the regulation of hepatic function as well as of hemodynamics. In addition to direct neural connections, the hypothalamus can affect metabolic functions by neuroendocrine connections: the hypothalamus-pancreas axis, the hypothalamus-adrenal axis, and the hypothalamus-pituitary axis. In the hypothalamus-pancreas axis, autonomic nerves release glucagon and insulin, which directly enter the liver and affect liver metabolism. In the hypothalamus-adrenal axis, autonomic nerves release catecholamines such as adrenaline and noradrenaline from the adrenal medulla, which also affects liver metabolism. In the hypothalamus-pituitary axis, release of glucocorticoids and thyroid hormones is stimulated by pituitary hormones. Both groups of hormones modulate hepatic metabolism. Taken together, the hypothalamus controls liver functions by neural and neuroendocrine connections.

Catecholamines inhibit the antigen-presenting capability of epidermal Langerhans cells

The sympathetic nervous system modulates immune function at a number of levels. Within the epidermis, APCs (Langerhans cells (LC)) are frequently anatomically associated with peripheral nerves. Furthermore, some neuropeptides have been shown to regulate LC Ag-presenting function. We explored the expression of adrenergic receptors (AR) in murine LC and assessed their functional role on Ag presentation and modulation of cutaneous immune responses. Both purified LC and the LC-like cell lines XS52-4D and XS106 expressed mRNA for the ARs alpha(1A) and beta(2). XS106 cells and purified LC also expressed beta(1)-AR mRNA. Treatment of murine epidermal cell preparations with epinephrine (EPI) or norepinephrine inhibited Ag presentation in vitro. Furthermore, pretreatment of epidermal cells with EPI or norepinephrine in vitro suppressed the ability of these cells to present Ag for elicitation of delayed-type hypersensitivity in previously immunized mice. This effect was blocked by use of the beta(2)-adrenergic antagonist ICI 118,551 but not by the alpha-antagonist phentolamine. Local intradermal injection of EPI inhibited the induction of contact hypersensitivity to epicutaneously administered haptens. Surprisingly, injection of EPI at a distant site also suppressed induction of contact hypersensitivity. Thus, catecholamines may have both local and systemic effects. We conclude that specific ARs are expressed on LC and that signaling through these receptors can decrease epidermal immune reactions.

Development of neuropeptide Y innervation in the liver

Hepatic neuropeptide Y (NPY) innervation was studied by immunohistochemistry in various mature vertebrates including the eel, carp, bullfrog, turtle, chicken, mouse, rat, guinea pig, dog, monkey, and human. In addition, an ontogenetic study on hepatic NPY was made in developing mice and guinea pigs. In all species examined except the eel, NPY-like immunoreactivity was detected in nerve fibers. In the carp, bullfrog, turtle, chicken, mouse, and rat, NPY-positive fibers were distributed around the wall of hepatic vessels and the bile duct of the Glisson's sheath. The density of NPY-positive fibers increased with evolution. However, in the guinea pig, dog, monkey, and human, numerous NPY-positive fibers were observed not only in the Glisson's sheath but also in the liver parenchyma. Positive fibers formed a dense network that surrounded the hepatocytes. The present immunoelectron microscopic study has confirmed that NPY-positive terminals are closely apposed to hepatocytes. Ontogenically, NPY-positive fibers were first found in the embryonic liver of 19-day-old mice. Positive fibers increased with age, and the highest peak was seen 1 week after birth. However, NPY-positive nerve fibers were present abundantly in Glisson's sheath and in the hepatic parenchyma of neonatal (3 and 7 days old) guinea pigs in a distribution similar to that in mature animals. This ontogenetic pattern suggests that NPY plays a certain role in the developing liver.

The absence of autonomic perivascular nerves in human colorectal liver metastases

The peptidergic/aminergic innervation of normal liver and tumour blood vessels was investigated in order to determine vascular control with a view to improving the efficacy of hepatic arterial cytotoxic infusion in the treatment of colorectal liver metastases. Selected areas of liver metastases and macroscopically normal liver from resection specimens (n = 13) were studied using light microscope immunohistochemistry for the presence of protein gene product 9.5 (PGP), vasoactive intestinal polypeptide (VIP), neuropeptide Y (NPY), calcitonin gene-related peptide (CGRP), substance P (SP) and tyrosine hydroxylase (TH). The ultrastructure of blood vessels supplying liver metastases and their perivascular innervation were also examined by transmission electron microscopy. In the normal liver, perivascular immunoreactive nerve fibres containing PGP, NPY and TH were observed around the interlobular blood vessels and along the sinusoids and the central vein of the hepatic lobule. The greatest density of immunoreactive nerve fibres was seen for PGP, followed (in decreasing order) by NPY and TH. VIP, SP and CGRP immunoreactivity was observed only in nerve bundles associated with the large interlobular blood vessels. In contrast, no perivascular immunoreactive nerves were observed in colorectal liver metastases. Electron microscopy confirmed the absence of perivascular nerves in liver metastases. In addition, it showed that the walls of these blood vessels were composed of a layer of endothelial cells surrounded by an incomplete or, very rarely in the periphery of the tumour, a complete, layer of synthetic phenotype of smooth muscle-like cells. These results imply that the blood vessels supplying liver metastases are bereft of normal neuronal regulation; whether there is a role for endothelial cell control of blood flow in these vessels is not yet known.

Phylogenetic and ontogenetic study of neuropeptide Y-containing nerves in the liver

The distribution of neuropeptide Y was investigated by light and electron microscopic immunohistochemistry in the liver of various vertebrates including the eel, carp, bullfrog, turtle, chicken, mouse, rat, guinea-pig, dog, monkey and human. The ontogenetic development of neuropeptide Y was also studied in the mouse liver. In all species examined except the eel, neuropeptide Y-like immunoreactivity was detected in nerve fibres. In the carp, bullfrog, turtle, chicken, mouse and rat, positive fibres were distributed around the wall of hepatic vessels and the bile duct of the Glisson's sheath. The density of the positive fibres increased with evolution. On the other hand, in the guinea-pig, dog monkey and human, numerous neuropeptide Y-positive fibres were observed not only in the Glisson's sheath but also in the liver parenchyma. Positive fibres formed a dense network to surround hepatocytes. The present immunoelectron microscopic study has confirmed that neuropeptide Y-positive terminals are closely apposing to hepatocytes. Ontogenetically, neuropeptide Y-positive fibres were first found in embryonic liver of 19-day-old mice. Positive fibres increased with age and the highest peak was seen one week after birth. This ontogenetic pattern has suggested that neuropeptide Y plays a certain role in developing liver.

The actions of two sensory neuropeptides, substance P and calcitonin gene-related peptide, on the canine hepatic arterial and portal vascular beds

1. The two peptides, calcitonin gene-related peptide (CGRP) and substance P (SP) were administered individually as bolus injections into the separately perfused hepatic arterial and portal vascular beds of the anaesthetized dog to assess their actions and relative molar potencies at these sites. 2. CGRP caused an immediate dose-related increase in hepatic arterial flow when injected close-arterially, reflecting a fall in resistance. This vasodilator effect was slightly increased by the prior administration of the selective beta 2-adrenoceptor antagonist, ICI 118,551. 3. On a molar basis, CGRP was more potent as an hepatic arterial vasodilator than the non-selective beta-adrenoceptor agonist, isoprenaline (Iso). 4. Intra-portal injection of CGRP also evoked hepatic arterial vasodilatation unaccompanied by other cardiovascular changes. 5. CGRP in doses up to 10 nmol had no effect on portal vascular resistance when administered intra-portally. 6. SP evoked a rapid, dose-related increase in hepatic arterial flow when injected intra-arterially. The molar ED50 for this hepatic vasodilatation was 40.2 fmol, significantly less than the ED50 for either CGRP or Iso. SP was the most potent hepatic arterial vasodilator yet examined. The vasodilator effect of SP was slightly potentiated by prior beta 2-adrenoceptor blockade. 7. SP caused hepatic arterial vasodilatation when administered by intra-portal injection; its absolute and relative potency was much reduced. 8. SP when injected intra-portally caused a graded increase in hepatic portal inflow resistance. The molar potency for this portal vasoconstriction was significantly greater than that for noradrenaline (NA); however, the maximum increase in portal resistance was significantly less to SP than to NA.9. In view of the location of the peptides CGRP and SP within the afferent innervation of the liver, it is proposed that they play an important function in controlling the hepatic microvasculature in response to sensory stimuli, particularly those arising from changes in portal blood composition secondary to change in metabolic activity within the gastrointestinal tract (GIT).10. Since the peptides are released from the GIT into the hepatic portal inflow, they may modify hepatic arterial blood flow, the extent of which is related to events within the GIT.

An isolated dual-perfused rabbit liver preparation for the study of hepatic blood flow regulation

An original, isolated dual-perfused rabbit liver preparation was developed for investigations into mechanisms that control the hepatic vascular tone. The hepatic artery (HA) and portal vein (PV) were perfused at constant flows of 0.16 +/- 0.01 and 0.64 +/- 0.05 mL/g/min (n = 5), respectively. Responses of the hepatic arterial and portal venous vascular beds to noradrenaline (NA) were measured as changes in perfusion pressure. Noradrenaline injected directly into the hepatic artery and portal vein produced dose-dependent increases in pressure in the respective vascular beds, the maximum response in the hepatic arterial bed being two to three times greater than that in the portal venous bed. A restricted transmission of vasoconstrictor stimulus between the intrahepatic portal venous and hepatic arterial vasculature was demonstrated. The results demonstrate the suitability of the dual-perfused rabbit liver model for detailed studies of the control of hepatic vascular tone.

The actions of human atrial natriuretic factor on hepatic arterial and portal vascular beds of the anaesthetized dog

1. The vascular actions of atrial natriuretic factor (ANF) have been assessed with other vasoactive agents on the hepatic arterial and portal vascular beds of the anaesthetized dog. 2. Intra-arterial bolus injections of ANF (0.1-50 nmol) caused graded increases in hepatic arterial blood flow representing a vasodilatation of relatively short duration. Vasoconstriction was never observed. 3. The maximum increase in hepatic arterial blood was the same for ANF and isoprenaline (Iso) i.e. approximately 60-70% increase over control flow. 4. On a molar basis, ANF was less potent than Iso although over the higher dose range (10(-9)-10(-7) mol) its vasodilator activity exceeded that of the endogenous vasodilator adrenaline. 5. Intraportal bolus injections (1.0-50 nmol) of ANF did not alter portal inflow resistance since no changes in portal inflow pressure occurred when the portal circuit was perfused at constant inflow volume. 6. This differential action of ANF on the hepatic arterial and portal vascular beds may provide a change in total liver blood flow in favour of the arterial component. 7. ANF, by altering hepatic haemodynamics to favour formation of trans-sinusoidal fluid exchange, may provide a temporary expansion of the extravascular fluid reservoir to buffer any increased venous pressure. However, chronically elevated plasma levels of ANF would encourage the formation of ascitic fluid.