Weak mitochondrial targeting sequence determines tissue-specific subcellular localization of glutamine synthetase in liver and brain cells.

Evolution of the uricotelic system for ammonia detoxification required a mechanism for tissue-specific subcellular localization of glutamine synthetase (GS). In uricotelic vertebrates, GS is mitochondrial in liver cells and cytoplasmic in brain. Because these species contain a single copy of the GS gene, it is not clear how tissue-specific subcellular localization is achieved. Here we show that in chicken, which utilizes the uricotelic system, the GS transcripts of liver and brain cells are identical and, consistently, there is no difference in the amino acid sequence of the protein. The N-terminus of GS, which constitutes a 'weak' mitochondrial targeting signal (MTS), is sufficient to direct a chimeric protein to the mitochondria in hepatocytes and to the cytoplasm in astrocytes. Considering that a weak MTS is dependent on a highly negative mitochondrial membrane potential (DeltaPsi) for import, we examined the magnitude of DeltaPsi in hepatocytes and astrocytes. Our results unexpectedly revealed that DeltaPsi in hepatocytes is considerably more negative than that of astrocytes and that converting the targeting signal into 'strong' MTS abolished the capability to confer tissue-specific subcellular localization. We suggest that evolutional selection of weak MTS provided a tool for differential targeting of an identical protein by taking advantage of tissue-specific differences in DeltaPsi.

A single glutamine synthetase gene produces tissue-specific subcellular localization by alternative splicing.

Glutamine synthetase (GS) plays a key role in two major biochemical pathways: In liver GS catalyzes ammonia detoxification, whereas in neural tissues it also functions in recycling of the neurotransmitter glutamate. In most species the GS gene gives rise to a cytoplasmic protein in both liver and neural tissues. However, in species that utilize the ureosmotic or uricotelic system for ammonia detoxification, the enzyme is cytoplasmic in neural tissues, but mitochondrial in liver cells. Since most vertebrates have a single copy of the GS gene, it is not clear how tissue-specific subcellular localization is achieved. Here we show that in the ureosmotic elasmobranch, Squalus acanthias (spiny dogfish), two different GS transcripts are generated by tissue-specific alternative splicing. The liver transcript contains an alternative exon that is not present in the neural one. This exon leads to acquisition of an upstream in-frame start codon and formation of a mitochondrial targeting signal (MTS). Therefore, the liver product is targeted to the mitochondria while the neural one is retained in the cytoplasm. These findings present a mechanism in which alternative splicing of an MTS-encoding exon is used to generate tissue-specific subcellular localization.

Loss of E-cadherin-mediated cell-cell contacts activates a novel mechanism for up-regulation of the proto-oncogene c-Jun.

Loss of E-cadherin-mediated cell-cell contacts can elicit a signaling pathway that leads to acquisition of an invasive phenotype. Here, we show that at the receiving end of this pathway is the proto-oncogene c-Jun, a member of the activator protein-1 family of transcription factors that play a key role in stimulation of cell proliferation and tumor promotion. Cell separation or abrogation of E-cadherin-mediated cell-cell contacts both cause a dramatic increase in accumulation of the c-Jun protein. Unlike growth factors that enhance the expression of c-Jun by activating the transcription of the c-jun gene, the cell contact-dependent increase in c-Jun accumulation is not accompanied by a corresponding increase in c-Jun mRNA or c-Jun protein stability but rather in the translatability of the c-Jun transcript. Consistently, the increase in c-Jun accumulation is not dependent on activation of the mitogen-activated protein kinase or beta-catenin pathways but is mediated by signals triggered by the restructured cytoskeleton. Depolymerization of the cytoskeleton can mimic the effect of cell separation and cause a dramatic increase in c-Jun accumulation, whereas Taxol inhibits the cell contact-dependent increase. This novel mechanism of c-Jun regulation seems to underlie the robust overexpression of c-Jun in tumor cells of patients with colon carcinoma.