Glioma Cell Growth Suppression is Associated with the Expression of Specific Secreted Protein Factors and the Selective Transfer of Endogenous Metabolites through Gap Junctions Composed of Different Connexins

Gary S. Goldberg1,6, Paul D. Lampe2, John F. Bechberger3, 4Youichi Tajima, 4Yutaka Sanai, 5Hiroshi Yamasaki, 6Hiroyuki Tsuda, 3Christian C.G. Naus, and 1Bruce J. Nicholson.

1Biological Sciences, Cooke Hall, State University of New York at Buffalo, Buffalo, NY 14260; 2Fred Hutchinson Cancer Research Center, Seattle, WA 98104; 3Cell Biology and Anatomy, Medical Science Bldg., University of Western Ontario, London, ONT N6A 5C1; 4Department of Biochemical Cell Research, Tokyo Metropolitan Institute of Medical Science, 3-18-22 Honkomagome, Bunkyo-ku, Tokyo 113; 5Unit of Multistage Carcinogenesis, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon cedex 08; 6Experimental Pathology and Chemotherapy, National Cancer Center Research Institute, 5-1-1 Tsukiji Chuo-ku, Tokyo 104.

The connexin family of proteins has evolved into at least 13 different members in mammals. The diverse array of biological processes associated with different connexins suggests that gap junction properties, including their permeability characteristics, may vary significantly with connexin composition. For example, C6 glioma cells express very low endogenous levels of Cx43 which, when augmented by transgene expression, suppresses the growth of these cells in vitro. In contrast to Cx43, however, expression of Cx32 does not alter C6 cell growth in vitro. Moreover, in an apparent dichotomy, transfer of fluorescent dye is actually several-fold higher between the Cx32 transfectants. To help resolve this issue, we have developed techniques to "capture", identify, and quantitate the transfer of physiologically relevant, endogenous transjunctional metabolites between these cells. In stark contrast to calcein, the rate of transfer of a significant portion of endogenous metabolites through Cx43 channels exceeds that through Cx32 channels. Diffusion of ADP, ATP, glutamate, and glutathione between the Cx43 transfectants approaches equilibrium within 20 minutes of communication. In particular, more ADP and/or ATP transfer between Cx43 than Cx32 transfectants. However, no significant preference is evident for the transfer of glutamate or glutathione through these connexins.

As mentioned above, transfection by Cx43, but not Cx32, inhibits the growth of C6 cells in vitro. The preferred passage of endogenous metabolites, including ADP and ATP, through Cx43 may help resolve this apparent contradiction. In one scenario, diffusion of these high-energy metabolites from metabolically active cells to less active neighbors may dilute their mitogenic potential, resulting in a slowing of cell growth. In another scenario, transjunctional molecules may directly target genes or proteins to affect cell growth and behavior. Cx43 induces contact growth inhibition, without affecting log phase growth rates, of C6 cells in vitro. This phenomenon is mediated by diffusable factors secreted from the transfected cells. Analysis of conditioned medium from these cells reveals that the expression and/or secretion of several proteins are modulated by Cx43. These include an oncofetal antigen implicated in integrin signaling, as well as a potent, but nontoxic factor which reduces the saturation density of transformed cells.

This work was supported by awards from PachTech Laboratories, International Union Against Cancer, Wendy Will Case Cancer Fund, and the Japanese Ministry of Health and Welfare Foundation for Promotion of Cancer Research to GSG.


 

Top of Bruce Nicholson's Page

Dept. of Biochemistry, UTHSCSA home page