Gap junctions are ubiquitous in vertebrates and invertebrates with certain exceptions, such as mature skeletal muscle and spermatocytes. A number of different connexins have been identified todate which constitute a multigene family of related but distinct members, and connexins are expressed in temporally and spatially regulated manner.A great deal of effort has been expended to understand the structural components of gap junctions, their arrangement in channel formation, tissue distribution, and their regulation. These efforts have produced detailed information on the structure of the gap junction channels, electrical conduction properties, and different levels of regulatory controls ranging from the synthesis of the transcripts for connexin to the opening and closing of the channels. However, two distinct areas of research have shown meager results; these are (a) assembly ofconnexins into the hexameric connexon and ultimately in gap junction formation, and (b) specific role(s) of gap junctions in the cellular physiology of non-excitable cells. Progress in delineating the processes of assembly and intracellular transport has been dependent on the availability of molecular probes such as specific antibodies which have now become available (Allen, et al, 1990, Kumar and Gilula, 1992).The role of gap junctions in electrotonic coupling of neurons and synchronization of excitable cells such as cardiac and smooth muscle cells has been well established (reviewed by Bennett, 1973, Sheridan and Atkinson, 1985). However, their role in the non-excitable cells has long been a subject of speculations based on circumstantial evidence except for a finding that mutations in a Schwann cell connexin gene may lead to Charcot-Marie Tooth disease of the peripheral nervous system (Bergoffen, et al, 1993a, Bergoffen, et al, 1993b) which leads to nerve degeneration. In another recent report, Reaume et al. (1995) targeted a mutation in a connexin gene (Cx43) of mouse and observed cardiac malformation during embryonic development.Gap junctions also have been implicated in loss of growth control and tumor suppression (reviewed by Loewenstein, 1979, Bertram, 1990, Loewenstein and Rose, 1992). Correlative evidence for this hypothesis emanates from studies that showed that (a) many tumor cells are defective in gap junctional communication, (b) viral transformation of cultured cells results in the loss of gap junctions, and (c) tumor-promoter agents such as TPA inhibit gap junctional communication. It has been hypothesized (Loewenstein, 1979, Loewenstein and Rose, 1992) that growth regulatory molecules can diffuse through the gap junction channels and exert their growth regulatory effects on neighboring cells. Regulatory molecules such as calcium ions, inositol phosphate, or cyclic nucleotides, in theory, may pass from one cell to another through these channels, but the identity of these regulatory molecules has not been established. However, the idea of a direct role of gap junctional communication as a primary growth regulatory mechanism is controversial because of the following reasons; (a) many tumor cells and malignant cell lines do form gap junctions and exhibit intercellular communication (Bennett et al., 1991), (b) during embryonic development gap junctions are formed as early as the eight-cell stage, and yet cell proliferation during embryogenesis is more rapid than in many tumors, (c) "knock-out" mice lacking a connexin43 gene (a gap junction gene which is expressed as early as the eight-cell stage of embryogenesis) developed to birth and showed only cardiac malformations (Reaume et al., 1995). By experimentally causing the overexpression of Cx43 in the connexin-deficient C6 rat glioma cell line, it has been shown that the growth of cells decreased in cultures (Zhu et al., 1991). Thus, a direct relationship between gap junctional intercellular communication and the growth of C6 cells was proposed.The ubiquitous presence of gap junctions and their relatively conserved molecular structures indicate their importance in vertebrate physiology, yet the tissue and stage-specific expression of different connexins argues against them having a common role in different tissues. Moreover, the function of gap junctional communication may be supplementary to other cellular regulatory mechanisms because the loss of gap junctional potential does not affect cellular viability. That gap junctions are not indispensable has been shown by the development of"knock-out" mice to full-term (Reaume et al., 1995); the mutation in Cx43 which is expressed even at eight-cell stage in the mouse embryo did not affect embryo development other than to produce cardiac malformations. It is known that different connexins impart distinct electrophysiological properties to the gap junctional channels in which they occur (Bennett, et al, 1991, Bennett and Verselis, 1992). Therefore, it is possible that a multiplicity of connexins provide supplementary functions in gap junction channels, with certain "general" roles shared by all the channels and "specific" roles restricted to different connexin isoforms. A major problem in assessing the effects of gap junctions has been the lack of a reliable method of inhibiting gap junction formation in situ in cells. A number of pharmacological agents cause closure of these channels, or downregulation of connexin synthesis, but they tend to be non-specific in their actions. In the central nervous system (CNS) glia comprise the major population of cells.Gap junction formation and cell-to-cell coupling is known to exist in these cells (reviewed by Bennett, et al, 1991, Dermietzel and Spray, 1993). Metabolic cooperation and sharing of organic molecules and inorganic ions have been suggested to be the major roles of the gap junctions in astrocytes. A salient feature of gap junctional function in brain physiology is the buffeting of potassium ions during neuronal activity (Orkand et al.,1966; Karwoski et al., 1989). Astrocytes are believed to provide a sink for potassium ion efflux into the extracellular milieu (see reviews by Newman, et al, 1984, Dermietzel and Spray, 1993) inasmuch as they offset the local increase in potassium ions through the uptake and spatial redistribution of potassium ions from the extracellular space. Gap junctions may be important in reactive astrocytes during the process of astrogliosis possibly by facilitating cell-to-cell communication (Alonso and Privat, 1993). A role for gap junctions in the growth and proliferation of glial cells and their tumors also has been proposed (Zhu, et al, 1991, Zhu, et al, 1992, Naus, et al, 1992). A direct method of investigating the latter possible role of gap junctions would be to use "knock-out" mutations and antisense techniques. Transfection of cells to express antisense RNA could inhibit connexin expression in a direct and specific fashion so as to reveal the regulatory role of gap junctions in controlling cell growth and proliferation. Further studies on the gap junctions in this respect should be rewarding.