Regulation of Cx45 hemichannels mediated by extracellular and intracellular calcium

Connexin45 (Cx45) hemichannels (HCs) open in the absence of Ca(2+) and close in its presence. To elucidate the underlying mechanisms, we examined the role of extra- and intracellular Ca(2+) on the electrical properties of HCs. Experiments were performed on HeLa cells expressing Cx45 using electrical (voltage clamp) and optical (Ca(2+) imaging) methods. HCs exhibit a time- and voltage-dependent current (I(hc)), activating with depolarization and inactivating with hyperpolarization. Elevation of [Ca(2+)](o) from 20 nM to 2 μM reversibly decreases I(hc), decelerates its rate of activation, and accelerates its deactivation. Our data suggest that [Ca(2+)](o) modifies the channel properties by adhering to anionic sites in the channel lumen and/or its outer vestibule. In this way, it blocks the channel pore and reversibly lowers I(hc) and modifies its kinetics. Rapid lowering of [Ca(2+)](o) from 2 mM to 20 nM, achieved early during a depolarizing pulse, led to an outward I(hc) that developed with virtually no delay and grew exponentially in time paralleled by unaffected [Ca(2+)](i). A step increase of [Ca(2+)](i) evoked by photorelease of Ca(2+) early during a depolarizing pulse led to a transient decrease of I(hc) superimposed on a growing outward I(hc); a step decrease of [Ca(2+)](i) elicited by photoactivation of a Ca(2+) scavenger provoked a transient increase in I(hc). Hence, it is tempting to assume that Ca(2+) exerts a direct effect on Cx45 hemichannels.

Pitfalls when examining gap junction hemichannels: interference from volume-regulated anion channels

Human HeLa cells transfected with mouse connexin45 were used to explore the experimental conditions suitable to measure currents carried by gap junction hemichannels. Experiments were performed with a voltage-clamp technique and whole-cell recording. Lowering [Ca(2+)](o) from 2 mM to 20 nM evoked an extra current, I (m), putatively carried by Cx45 hemichannels. However, the variability of I (m) (size, voltage sensitivity, kinetics) suggested the involvement of other channels. The finding that growth medium in the incubator increased the osmolarity with time implied that volume-regulated anion channels (VRAC) may participate. This assumption was reinforced by the following observations. On the one hand, keeping [Ca(2+)](o) normal while the osmolarity of the extracellular solution was reduced from 310 to 290 mOsm yielded a current characteristic of VRAC; I (VRAC) activated/deactivated at negative/positive voltage, giving rise to the conductance functions g (VRAC,inst)=f(V (m)) (inst: instantaneous; V (m): membrane potential) and g (VRAC,ss)=f(V (m)) (ss: steady state). Moreover, it was reversibly inhibited by mibefradil, a Cl(-)channel blocker (binding constant K (d)=38 microM, Hill coefficient n=12), but not by the gap junction channel blocker 18alpha-glycyrrhetinic acid. On the other hand, minimizing the osmotic imbalance while [Ca(2+)](o) was reduced led to a current typical for Cx45 hemichannels; I (hc) activated/deactivated at positive/negative voltage. Furthermore, it was reversibly inhibited by 18alpha-glycyrrhetinic acid or palmitoleic acid, but not by mibefradil. Computations based on g (VRAC,ss)=f(V (m)) and g (hc,ss)=f(V (m)) indicated that the concomitant operation of both currents results in a bell-shaped conductance-voltage relationship. The functional implications of the data presented are discussed. Conceivably, VRAC and hemichannels are involved in a common signaling pathway.

Connexin37 protects against atherosclerosis by regulating monocyte adhesion

A genetic polymorphism in the human gene encoding connexin37 (CX37, encoded by GJA4, also known as CX37) has been reported as a potential prognostic marker for atherosclerosis. The expression of this gap-junction protein is altered in mouse and human atherosclerotic lesions: it disappears from the endothelium of advanced plaques but is detected in macrophages recruited to the lesions. The role of CX37 in atherogenesis, however, remains unknown. Here we have investigated the effect of deleting the mouse connexin37 (Cx37) gene (Gja4, also known as Cx37) on atherosclerosis in apolipoprotein E-deficient (Apoe(-/-)) mice, an animal model of this disease. We find that Gja4(-/-)Apoe(-/-) mice develop more aortic lesions than Gja4(+/+)Apoe(-/-) mice that express Cx37. Using in vivo adoptive transfer, we show that monocyte and macrophage recruitment is enhanced by eliminating expression of Cx37 in these leukocytes but not by eliminating its expression in the endothelium. We further show that Cx37 hemichannel activity in primary monocytes, macrophages and a macrophage cell line (H36.12j) inhibits leukocyte adhesion. This antiadhesive effect is mediated by release of ATP into the extracellular space. Thus, Cx37 hemichannels may control initiation of the development of atherosclerotic plaques by regulating monocyte adhesion. H36.12j macrophages expressing either of the two CX37 proteins encoded by a polymorphism in the human GJA4 gene show differential ATP-dependent adhesion. These results provide a potential mechanism by which a polymorphism in CX37 protects against atherosclerosis.

Structural and functional diversity of connexin genes in the mouse and human genome

Gap junctions are clustered channels between contacting cells through which direct intercellular communication via diffusion of ions and metabolites can occur. Two hemichannels, each built up of six connexin protein subunits in the plasma membrane of adjacent cells, can dock to each other to form conduits between cells. We have recently screened mouse and human genomic data bases and have found 19 connexin (Cx) genes in the mouse genome and 20 connexin genes in the human genome. One mouse connexin gene and two human connexin genes do not appear to have orthologs in the other genome. With three exceptions, the characterized connexin genes comprise two exons whereby the complete reading frame is located on the second exon. Targeted ablation of eleven mouse connexin genes revealed basic insights into the functional diversity of the connexin gene family. In addition, the phenotypes of human genetic disorders caused by mutated connexin genes further complement our understanding of connexin functions in the human organism. In this review we compare currently identified connexin genes in both the mouse and human genome and discuss the functions of gap junctions deduced from targeted mouse mutants and human genetic disorders.

Connexin genes in the mouse and human genome

Gap junctions serve for direct intercellular communication by docking of two hemichannels in adjacent cells thereby forming conduits between the cytoplasmic compartments of adjacent cells. Connexin genes code for subunit proteins of gap junction channels and are members of large gene families in mammals. So far, 17 connexin (Cx) genes have been described and characterized in the murine genome. For most of them, orthologues in the human genome have been found (see White and Paul 1999; Manthey et al. 1999; Teubner et al. 2001; Söhl et al. 2001). We have recently performed searches for connexin genes in murine and human gene libraries available at EMBL/Heidelberg, NCBI and the Celera company that have increased the number of identified connexins to 19 in mouse and 20 in humans. For one mouse connexin gene and two human connexin genes we did not find orthologues in the other genome. Here we present a short overview on distinct connexin genes which we found in the mouse and human genome and which may include all members of this gene family, if no further connexin gene will be discovered in the remaining non-sequenced parts (about 1-5%) of the genomes.