Glycosaminoglycans and proteoglycans induce gap junction expression and restore transcription of tissue‐specific mRNAs in primary liver cultures

Normal rat hepatocytes maintained on tissue culture plastic and in serum‐supplemented medium lose their gap junctions within 12 hr and expression of their tissue‐specific functions within 24 to 72 hr. The gap junctions are lost via internalization and degradation, and the differentiated functions due to loss of synthesis and to rapid degradation of tissue‐specific mRNAs. Near normal levels of tissue‐specific mRNAs can be achieved by stabilization of the mRNAs but not by transcription (for most genes), if the cells are cultured in a serum‐free, hormonally defined medium and on substrata of tissue culture plastic, fibronectin or laminin, or on various purified collagens. The hormonally defined medium also extends the life‐span of the gap junctions to about 24 hr. Certain glycosaminoglycans, proteoglycans and anionic polysaccharides have proven to be potent inducers of gap junction expression and function, to increase abundance of tissue‐specific mRNAs, and to lower abundance of common gene mRNAs, a level of gap junctions and a pattern of gene expression similar to that in vivo. Addition to the hormonally defined medium of 10 μg per ml of hyaluronates, dermatan sulfates, bovine lung heparan sulfate, chondroitin 4‐sulfate or chondroitin 6‐sulfate resulted in a weak response in induction of gap junctions (5 to 15% of the cells became dye and electrically coupled) and in gene expression. An intermediate response in gap junction expression (30 to 50% coupled cells) and in gene expression was observed with 50 to 100 μg per ml of heparins or hyaluronates. The most extensive coupling (70 to 100%) and the strongest responses in gene expression were seen with proteoglycans, such as dermatan sulfate proteoglycan or chondroitin sulfate proteoglycan. Treatment of cultures with several forms of carrageenans, polymers of sulfated galactose, or dextran sulfates, polymers of sulfated glucose, also gave responses in levels of gap junctions that did not always correlate with responses in gene expression. Western blots demonstrated that levels of the main intrinsic gap junction polypeptide were consistent with the degree of electrical and dye coupling. The gene expression responses (like the gap junction levels) were dependent on dosage and length of time of exposure to heparins or carrageenans and were due to an increase (tissue‐specific genes) or a decrease (common genes) in the stability of their mRNAs, and, for some liver‐specific genes, to restoration of their transcriptional signals. Tissue‐specific genes did not respond coordinately to any of the glycosaminoglycans, proteolgycans or anionic polysaccharides tested. In summary, glycosaminoglycans and proteoglycans were found to be important regulators in the synthesis and the stability of tissue‐specific mRNAs, in the stability of common gene mRNAs and in expression of gap junction protein in cultures of normal rat hepatocytes. Of especial significance is the realization that glycosaminoglycans and proteoglycans are unique among the matrix components and hormones tested to date, in that they can restore transcriptional signals for tissue‐specific mRNAs in primary liver cultures.