The play is still being written on opening day: postnatal maturation of enteric neurons may provide an opening for early life mischief

Clinical experience with the aganglionoses of Chagas (acquired) and Hirschsprung's (congenital) diseases makes clear that the life-supporting motility of the bowel depends on an intact enteric nervous system (ENS). Although the CNS innervates the gut, the ENS, uniquely, can regulate intestinal behaviour independently of CNS input. To enable the ENS to do so, evolution has made it large and complex with microcircuits that are still imperfectly understood; nevertheless, most enteric neurons have been phenotypically identified (Furness, 2006) and the basic outline of their ontogeny is known (Gershon, 2010).

Enteric neurons can broadly be classified, according to their shape and electrophysiological properties, as Dogiel type I (S) or Dogiel type II (AH) cells. These two groups, however, accommodate many others when neurotransmitter content or other molecular properties are used for classification. Only a single long process extends from Dogiel type I (S) neurons and they readily display fast excitatory postsynaptic potentials (fEPSPs) in response to fibre tract stimulation. In contrast, Dogiel type II (AH) neurons extend many long neurites, manifest a characteristic hyperpolarizing afterpotential (the AH) and infrequently exhibit fEPSPs when fibre tracts are stimulated.

Developmentally, enteric neurons are progeny of immigrants (Gershon, 2010). Their predecessors migrate to the bowel from vagal, rostral truncal and sacral levels of the neural crest. These predecessors are heterogeneous; although some are determined before emigrating from the neuraxis, most are proliferating, multipotent and responsive to microenvironmental cues encountered while migrating to and within the gut. A cacophony of genes encoding essential transcription factors, growth factors and receptors has been discovered, which, when deficient, cause enteric aganglionoses. Guidance molecules and the extracellular matrix also play roles in directing émigrés from the crest to correct destinations in the gut. The forces, however, that organize this molecular cacophony and turn it into the symphony that provides mammals with a functioning ENS remain unknown. In fact, there has been little or no investigation of the transition that crest-derived precursor cells undergo to give rise to mature type I (S) and type II (AH) neurons, synaptic connections or microcircuits. The technical difficulty involved in measuring the activity of neurons in the developing bowel is daunting. The publication of Foong et al. (2012) in this issue of The Journal of Physiology, however, has finally opened this mysterious transitional period to physiological analysis. These undaunted investigators successfully employed intracellular recording to investigate electrophysiological properties, synaptic transmission and morphology of enteric neurons during early postnatal life.

Myogenic ‘ripples’ and action potentials in neurons have been detected before the ENS assumes control of gastrointestinal activity (Hao et al. 2011). Since the ENS of a newborn must be sufficiently mature to permit oral feeding, it is not surprising that Foong et al. (2012) report that the two main classes of neuron are already present at P0 and P10–11. Both type I (S) and type II (AH) neurons, however, change substantially in electrophysiological properties and synaptic inputs during later development; moreover, the dendritic morphology of type I (S) neurons also changes. The work of Foong et al. thus documents that the postnatal period is one of developmental ferment. Enteric neurons do not spring, fully formed, into existence when a mammal is born; rather, ENS development is still a work in progress and, as such, is potentially subject to the vagaries that early life may unpredictably bring to bear upon it.

The discovery that the physiology of enteric neurons is modified postnatally is one that is likely to be fruitful, not only in giving physiologists something new to work on, but in providing insights into the pathophysiology of functional gastrointestinal motility disorders, such as irritable bowel syndrome (IBS). These conditions are often thought to be of CNS origin because they share a high degree of co-morbidity with psychoneuroses. This co-morbidity, however, could as easily be due to an effect the ENS exerts on the CNS as the reverse; nevertheless, functional bowel disease often begins early in life and can be traced to infection, stress or abuse, which can alter activities of enteric neurons. The demonstration that postnatal ENS physiology is still being modified, is compatible with the likelihood that it is plastic. The activities of early-born enteric neurons may therefore affect the fates of their late-born confederates.

Neurotransmitters, like growth factors, affect ENS development. It is thus necessary to add neurotransmitters to the list of microenvironmental factors to which the development of enteric neurons is in thrall. Enteric serotonergic neurons arise early but their birth is dependent on transiently catecholaminergic precursor cells (Li et al. 2010); moreover, enteric dopaminergic neurons, which develop late, are dependent on 5-HT from enteric serotonergic neurons (Li et al. 2011). Crest-derived enteric neuronal precursors evidently listen to molecular advice provided, not only by non-neuronal neighbours, but by precociously developing colleagues from the crest. This process is likely to be reflected in the sequential birthdays of enteric neurons (Pham et al. 1991). Some, which are Ascl1-dependent, are born early and co-exist with still dividing precursors. It is thus plausible that epigenetic alterations in the enteric microenvironment during neurogenesis can cause subtle but lasting changes in the ENS. This concept extends beyond fetal life because enteric neurogenesis, as well as physiological change, continues postnatally (Pham et al. 1991) and even in mature animals (Liu et al. 2009). Psychosocial trauma, stress and inflammation alter neuronal activity and, if that activity affects the subsequent generation of neurons, and the changing physiology that Foong et al. (2012) have observed, neuronal regulation of neuronal development would enable psychological and infectious/inflammatory events to sculpt the ENS.