Strength of functional coupling can be measured in several ways, and each type of measurement has a different threshold for detection and may depend on different sets of variables and have different limitations. The most rigorous type of measurement is that of junctional conductance, which is most simply achieved on mammalian cell pairs using the dual whole cell voltage clamp technique. Although this method is exquisitely sensitive at low coupling strengths (being able to resolve junctional currents flowing through single channels), modifications are required to measure junctional conductances above about 10 nS, owing to series resistance errors. What is measured here is the total current carried by small, mobile, intracellular ions. The simplest method for determining coupling strength is to measure diffusion of a tracer molecule such as Lucifer Yellow or neurobiotin after their intracellular injection, or loading from extracellular space either by scrape-loading or loading one population of cells with dye esters and then mixing these with unloaded cells (see Chapter 12). These methods are most sensitive to large changes in well coupled systems, and we generally see little or no dye coupling in Cx43-expressing cells when junctional conductance is 2 nS or less. What is measured using this technique is permeability to molecules of specific charge and of large size, although the coculture method is also dependent on rate of gap junction formation.
Decreases in functional coupling can be due to altered effective size or charge selectivity of the channel, decreased percentage of time the channel is open, or decreased number of channels functionally present. For few uncoupling agents have detailed enough studies been done thus far to conclude just how coupling strength is reduced, although as considered below, it is clear that certain effectors (such as transjunctional voltage and intracellular acidification or halothane treatment) act through largely independent mechanisms.
Quantitative analysis of connexin-specific permeabilities to fluorescent dyes and to ions of different sizes have supported the concept that connexins show differential permeabilities that cannot be predicted only on size considerations, but strongly depend on the charge of the probe, cell volume, cell input resistance, junctional capacity and binding of the dye to axoplasmic elements (Lucifer Yellow [LY] binds irreversibly to organelles). In addition, since pore diameters vary significantly among homotypic channels and pore diameter can directly affect ionic permeability depending on the relative strength of the electrostatic surface charge of the pore and the distance to the permeant ion or molecule, precise determinations should be performed for each cell type in each specific condition. Several issues that remain to be rigorously determined include: (1) whether subconductance states differ in selective permeability when compared to the main state; and (2) what number of channels represents the threshold for dye permeation? Without these answers, experimental data regarding the lack of detection of dye transfer and uncoupling effects may be misinterpreted.
If a few channels effectively modulate a specific effect, partial blockage of the total number of channels would prevent dye transfer among neighboring cells and decrease the magnitude but not prevent the expression of the effect. For example, we have recently showed that bystander killing of adjacent cells by 20 ^M ganciclovir (GCV) is dependent on cell communication via gap junctions; HSV-tk transfected cells are killed in vitro within 5 d of treatment with GCV and transmit this toxicity to adjacent cells lacking HSV-tk. In short, the extent of cell death and sensitivity to GCV depend on the degree of connexin expression in transfectants and also depend on which connexin is expressed in the cocultured cells. Our results support the notion of the bystander effect is dependent on the strength of cell communication via gap junctions; even when the transfer of the dye LY could not be detected among weakly coupled cells, cell death was significantly assessed by the fluorescence-activated cell sorting (FACS) technique (68-70). Thus, the lack of detection of dye transfer could have been misinterpreted as lack of involvement of gap junctions in the bystander effect. Similarly, the results obtained by treating cells with partial uncouplers would be misinterpreted if evaluated only by dye transfer assays.
The use of the terms efficacy and potency has been confused in our field. Efficacy (or intrinsic activity) is related to the magnitude of the effect (i.e., blockade). Potency represents the dependency of the uncoupling effect on its concentration, the concentration of the drug at which it is half-maximally effective. Thus, the efficacy of a specific uncoupler is not necessarily related to its potency as an antagonist . For example, 2-3 mM halothane and 1 mM octanol inhibit gj by 100%, in contrast to 50 ^M oleic acid that inhibits gj by 80%; the efficacies and potencies in regard to their uncoupling effects are octanol ~ halothane >>>> oleic acid (efficacy) and oleic acid >>> octanol > halothane (potency).
As summarized in Table 1, agents that have been found to reduce junctional conductance may to some extent be grouped according to hypothesized mode of action. For example, intracellular acidification sufficient to close gap junction channels may be achieved by membrane permeant weak acids, by application of nigericin in the presence of high K+ and low extracellular pH, by metabolic inhibition with dinitrophenol and by sufficiently high concentrations of nitrobenzyl esters. Likewise, several lipophilic compounds uncouple cells, including doxylstearic acids, heptanol and octanol, halothane, and oleic acid. Arachidonic acid may exert is uncoupling effect in certain types through direct action or through cyclooxygenase or lipoxygenase metabolites. For other agents, such as dieldrin, DDT and phorbol esters (TPA), there is a phenomeno-logical correlation with their action as tumor promoters, but whether TPA (and the endogenous protein kinase C activator diacylglycerol) exert their action through an accompanying phosphorylation of the connexin proteins remains controversial. Finally, many of the agents used to close gap junction channels have no known mechanism of action, and it remains to be determined whether effects are directly on the channel protein or through other effects on the cell. In the sections that follow, we have briefly summarized some of the information available regarding modes of action of uncoupling agents.
Was this article helpful?