Article Text


Endothelins, pseudo-obstruction and Hirschsprung’s disease
  1. P J MILLA
  1. Gastroenterology Unit,
  2. Institute of Child Health,
  3. University of London,
  4. 30 Guildford Street,
  5. London WC1N 1EH, UK

Statistics from

See article on page246

It has been known since the 1950s that the enteric nervous system is formed from cells that arise from the neural crest.1 The enteric neurones mainly arise from the vagal neural crest of the developing hind brain and colonise the gut in a rostro caudal migration but some seem to arrive in the hind gut from the lumbosacral level via a caudo rostral wave of colonisation. The neural crest cells that migrate and colonise the gut are committed to become neuroblasts or neuronal support cells, glioblasts; however, differentiation into neurones and glial cells seems not to take place until they have reached their final resting places in the gut. Movement through the gut mesenchyme, survival in the gut and differentiation into mature cells is strongly influenced by contacts with the microenvironment which consists of other cells in the mesenchyme, neural crest, and the extracellular matrix. The extracellular matrix components provide directional clues to migrating neural crest cells and together with neighbouring cells provide some of the signals for crest cell differentiation. For example, the appearance of neural crest cells in the gut is preceded by expression of extracellular matrix molecules2 and other factors such as glial derived neurotropic factor (GDNF) ensure survival of committed neuroblasts.3 Thus defects of the neural crest cells themselves or alteration of the microenvironment of the migratory pathway may result in defects of development of the enteric nervous system. In humans this disordered development results in the most commonly presenting forms of chronic idiopathic intestinal pseudo-obstruction, congenital enteric neuromuscular disease. In Hirschsprung’s disease defects in at least two different cell signalling systems, ret/GDNF and endo- thelin-3/endothelin B receptor,4 cause the aganglionosis.

The endothelin system’s important role in the development of the enteric nervous system has become apparent in the past four years or so when mice with targeted disruption of endothelin B receptor (ETR-B) and endothelin-3 (ET-3)5-6 were found to have congenital distal intestinal aganglionosis.

The endothelins are a family of three peptides, endothelin-1, -2 and -3, coded for by distinct but related genes and act on cells via two G protein coupled receptors ETR-A and ETR-B. The endothelins are synthesised as much larger proproteins which are cleaved by an endothelin converting enzyme (ECE-1) to produce the active 21 amino acid peptide.7 Of the three endothelins it is ET-3 which is so important in the enteric nervous system and binding of ET-3 to ETR-B on vagal neural crest cells is required for colonisation of the hind gut.

Mutations of either ETR-B or ET-3 have been identified in several naturally occurring animal models of Hirschsprung’s disease, the piebald lethal mouse and the lethal spotted mouse4respectively. The ovaro-lethal white foal also has a significant mutation of ETR-B with a single amino acid substitution in the first transmembrane spanning domain of the ETR-B gene. The lethal spotted mouse carries a mutation in its ET-3 gene which prevents proteolytic activation of the peptide. Mutation analysis of children with Hirschsprung’s disease has shown that about 10% carry mutations of either ETR-B, ET-3 or ECE-1. The effects of these genetic defects is to curtail neural crest migration in the distal colon and this is associated with localised overexpression of extracellular matrix molecules.2 Using transgenic lines of mice which are either ETR-B deficient or ET-3 deficient, Kapur and colleagues8 have shown that in ETR-B deficient mice enteric nervous system precursors can colonise the murine hind gut when they are surrounded by wild type enteric nervous system precursors. Further wild type enteric nervous system precursors will fail to colonise the hind gut when surrounded by ETR-B deficient ones. This strongly suggests that the enteric nervous system precursors signal ETR-B activation to those nearby and that when this signal is of sufficient intensity an ETR-B deficient crest cell can develop normally. It is thus clear that the interaction between the migrating neural crest cells and the mesenchymal environment of the hind gut is of critical importance in achieving normal innervation of the colon. The mechanism of the terminal aganglionosis that occurs either in the absence of ET-3 or ETR-B however remains unclear.

Despite the increasing understanding of the role of endothelins in the developing enteric nervous system, little work has been done in normal mice or men regarding the timetable of activity or the spatial orientation of these molecules in the developing embryonal gut. On page 246 of this issue Leibl et al describe the temporal and spatial expression of ET-3 and ETR-B in CD1 mouse embryos. They show clearly that ETR-B is confined to migrating neural crest cells and ET-3 to mesenchymal cells initially of the caecum but with a gradient extending rostrally into the small intestine and caudally into the proximal colon. Interestingly by 14 days postcoitum the ETR-B mRNA signal in the colon was stronger than in the more proximal part of the gut at this or earlier stages, perhaps, suggesting that ETR-B is expressed by both vagal and sacral neural crest cells.

The present results add to the growing body of work emphasising the importance of the gut mesenchyme in determining regional identity along the gut primordium and also in the regulation of region specific innervation of the gastrointestinal tract. The mechanisms that regulate expression of ET-3 and ETR-B genes are currently unknown. It is clear however that the rostro caudal specification of the gastrointestinal tract is likely to involve a spatial, temporal and combinatorial patterns of expression of homeobox genes, the so called enteric hox code. In chick embryos there is clearly overlapping expression of the genes Hox A-9, -10 and -11 and we have recently produced some preliminary data demonstrating specific spatial, temporal and combinational expression patterns of hox genes A4, B4, D4, A5 and C59 in developing murine gut. The relation between caecum specific hox gene expression and ET3 and ETB-R is currently unknown but they are certainly candidate downstream molecules for these developmental control genes. A number of transgenic animal models provide evidence of the importance of homeobox genes in the control of morphogenesis of the gut and these include the “knock out” of ENX, causing increased innervation of the hind gut,10 and over expression of hox A4, resulting in megacolon.2 Thus this family of genes and their downstream targets are of importance within the genetic hierarchy of gut morphogenesis. Delineation of the genes comprising the enteric hox code, their downstream targets and their spatiotemporal patterns of expression is an essential and integral part of understanding the molecular events underlying the devastating diseases which cause pseudo-obstruction and Hirschsprung’s disease in humans. Such knowledge may enable antenatal diagnosis in some families and will be essential for the development of neuronal transplant strategies for the treatment of enteric neuropathic diseases.

See article on page246


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles