Emerging roles of connexin hemichannels in gastrointestinal and liver pathophysiology

Abstract

Connexin hemichannels have long been considered as mere structural precursors for gap junctions. In the last decade, it has become clear that they also act as individual channels, connecting the intracellular compartment and the extracellular environment. Impairement of connexin hemichannel functionality may result in disturbance of homeostasis, as exemplified in the current paper for the intestine and the liver. Research in this field still has a number of shortcomings, of which some are also discussed here.

Key Words: Hemichannel; Connexin; Pathophysiology

CONNEXIN CHANNELS: STRUCTURAL ASPECTS

The maintenance of tissue homeostasis is governed by the well-orchestrated interplay between three major communicative networks located at the extracellular, intracellular and intercellular level. Direct intercellular communication is mediated by gap junctions, which arise from the interaction of two hemichannels of neighbouring cells. Hemichannels, in turn, are composed of six connexin (Cx) units (Figure 1). The connexin family comprises as many as twenty isoforms in mammals. They are named according to their molecular weight and are expressed in a cell-specific way[1-5]. In the gastrointestinal tract, Cx40, Cx45 and particularly Cx43 are expressed by the intestinal smooth muscle cells and the interstitial cells of Cajal[6,7]. In the liver, hepatocytes abundantly produce Cx32 and small quantities of Cx26, while non-parenchymal hepatic cells mainly harbour Cx43[3,4].

Figure 1 Molecular architecture of gap junctions. Gap junctions are grouped in plaques at the cell membrane surface and are composed of twelve connexin proteins organized as two hexameric hemichannels. The connexin structure consists of four membrane-spanning domains, two extracellular loops, one cytoplasmic loop, one cytoplasmic amino tail and one cytoplasmic carboxy tail.

CONNEXIN CHANNELS: FUNCTIONAL ASPECTS

Gap junctions provide a pathway for the intercellular exchange of small and hydrophilic substances, including nucleotides (e.g. ATP) and ions (e.g. Ca²⁺). As numerous physiological processes are driven by these substances, gap junctional intercellular communication is considered as a key mechanism in the maintenance of tissue homeostasis[1-5]. Gastrointestinal gap junctions play a specific role in pacemaking and neurotransmission, and thus in the regulation of motility[7,8]. Hepatocellular gap junctions, on the other hand, essentially underlie most liver-specific functions, such as xenobiotic biotransformation and albumin secretion[3,4]. Gap junctions are also master regulators of cell growth and cell death[1,2]. In fact, a growing number of reports point to non-gap junctional functions for connexins in these events. Indeed, connexin proteins themselves can alter the production of critical homeostasis regulators, such as caspases and cyclins, irrespective of their channel properties. The mechanisms that govern these atypical connexin functions remain elusive, but may involve direct interaction with these regulatory molecules or the modulation of their gene transcription[1,2,9-11]. Another gap junction-independent cell signalling platform for connexins relates to hemichannels. For a long time, hemichannels have been considered as merely structural precursors for gap junctions, remaining closed until docking with counterparts from adjacent cells prior to gap junction formation. It has now become clear that connexin hemichannels also provide a pathway for communication, albeit between the intracellular compartment and the extracellular environment. The substances that travel through hemichannels are very similar to those that are intercellularly exchanged via gap junctions[1,2,9-12]. To make the picture even more complicated, a second set of gap junction-related proteins has been described in recent years, the pannexins, which are structurally similar but phylogenetically unrelated to the connexin family. At present, three pannexins, namely Panx1, Panx2 and Panx3, have been identified in humans and rodents, and they preferentially occur in a hemichannel configuration[1,13,14].

CONNEXIN HEMICHANNELS IN THE INTESTINE AND THE LIVER

Thus far, most attention has been paid to connexin hemichannels in nervous tissue as well as their involvement in cerebral ischemia, in which their opening results in cell death[1,15,16]. Nevertheless, a limited number of studies have addressed hemichannels in the gastrointestinal tract and the liver. In an attempt to characterize gap junctions in a human intestinal epithelial cell line, Clair and colleagues found that functional hemichannels composed of Cx26, Cx32 and Cx43 are abundantly present at the basal side of these polarized cells[17]. Research from the same group showed that the Cx26 hemichannels facilitate the pathogenesis induced by Shigella flexneri, the causative agent of bacillary dysentery that invades the colonic mucosa where it elicits an intense inflammatory reaction responsible for destruction of the mucosa. Thus, Shigella invasion induces the opening of Cx26 hemichannels, allowing extracellular release of ATP, which in turn favours bacterial dissemination[18]. In a recent in vivo study conducted by Guttman et al, increasing levels of Cx43 were observed in mouse colon infected with the diarrhea-causing Citrobacter rodentium, whereby unpaired hemichannels were formed at both the apical and the lateral membrane surface of the colonocytes. Using animals genetically deficient in Cx43, it was subsequently demonstrated that Cx43 hemichannel opening triggers water release during infectious diarrhea[19]. Our group was the first to report the occurrence of Cx32 hemichannels in hepatocytes. Using an in vitro model of Fas-mediated apoptosis, a cell death mode involved in most liver pathologies, we found that de novo synthesized Cx32 gathers in hemichannels during apoptosis, and that hemichannel opening is critical for the termination of the cell death response[20].

CONCLUSION

Although a large body of evidence is nowadays available, the concept of functional connexin hemichannels still remains controversial[1,21]. A major point of debate concerns the identity of the molecular constituents of hemichannels. Critics claim that most, if not all, of the functions that have been attributed to connexin hemichannels, can actually be ascribed to other channel types, consisting for example of pannexins[21,22]. In addition, it seems challenging to discriminate between the functionality of connexin hemichannels and that of gap junctions, especially in an in vivo environment[1,21]. These constraints mainly arise from the ubiquitous lack of appropriate testing approaches to unequivocally study connexin hemichannels[21,23], including pannexin hemichannel- and connexin hemichannel-specific inhibitors. Undoubtedly, this research field will become more appealing when such experimental tools are avaliable. Moreover, it can be anticipated that the basic knowledge that will be gained by doing so, will also open new perspectives for the development of new clinical (gastrointestinal and liver) therapies.