Cloning and characterization of the mouse glucokinase gene locus and identification of distal liver-specific DNase I hypersensitive sites.

We cloned and characterized an 83-kb fragment of mouse genomic DNA containing the entire glucokinase (GK) gene. The 11 exons of the gene span a total distance of 49 kb, with exons 1 beta and 1L being separated by 35 kb. A total of 25,266 bp of DNA sequence information was determined: from approximately -9.2 to approximately +15 kb (24,195 bp), relative to the hepatocyte transcription start site, and from -335 to +736 bp (1071 bp), relative to the transcription start site in beta cells. These sequences revealed that mouse GK is > 94% identical to rat and human GK. Mouse hepatic GK mRNA is regulated by fasting and refeeding, as also occurs in the rat. Alignment of the upstream and downstream promoter regions of the mouse, rat, and human genes revealed several evolutionarily conserved regions that may contribute to transcriptional regulation. However, fusion gene studies in transgenic mice indicate that the conserved regions near the transcription start site in hepatocytes are themselves not sufficient for position-independent expression in liver. Analysis of the chromatin structure of a 48-kb region of the mouse gene using DNase I revealed eight liver-specific hypersensitive sites whose locations ranged from 0.1 to 36 kb upstream of the liver transcription start site. The availability of a single, contiguous DNA fragment containing the entire mouse GK gene should allow further studies of cell-specific expression of GK to be performed.

Analysis of upstream glucokinase promoter activity in transgenic mice and identification of glucokinase in rare neuroendocrine cells in the brain and gut.

A transgene consisting of an upstream glucokinase (GK) promoter fragment linked to coding sequences of the human growth hormone gene was expressed in certain neuroendocrine cells of the pancreas, pituitary, brain, gut, thyroid, and lungs of mice. In pancreas, the transgene was expressed in a nonuniform manner among beta cells and in a variable but substantial fraction of the other islet cell types. In pituitary, it was expressed in corticotropes, and in brain, it was expressed in cells of the medial hypothalamus. Within the gut transgene expression was detected in a subset of enteroendocrine cells of the stomach and duodenal epithelium, some of which also exhibited glucagon-like polypeptide-1 immunoreactivity. In thyroid, transgene expression was observed in C cells of neonatal animals, whereas in the lung, it was expressed among rare endocrine cells of the bronchopulmonary mucosa. RNA polymerase chain reaction analysis of human growth hormone mRNA corroborated the tissue-specific transgene expression pattern. Prompted by the finding of transgene expression in specific neuroendocrine cells, we sought to determine whether GK mRNA and GK itself was also expressed in the brain and gut, tissues not previously associated with the expression of this enzyme. Using rat tissues, GK mRNA was detected by RNA polymerase chain reaction in both the brain and intestine and was localized to specific cells in the hypothalamus and enteric mucosa by in situ hybridization. A high Km glucose phosphorylating activity was detected from isolated rat jejunal enterocytes that displayed a chromatographic elution profile identical to hepatic GK. GK immunoreactivity was detected in cells of the medial hypothalamus with many of the same cells also displaying GLUT2 immunoreactivity. Together, these studies provide evidence for upstream GK promoter activity, GK mRNA, and GK itself in certain neuroendocrine cells outside the pancreatic islet and lead us to suggest that GK may play a broader role in glucose sensing by neuroendocrine cells than was thought previously.

Repair of alkali-labile sites within the mitochondrial DNA of RINr 38 cells after exposure to the nitrosourea streptozotocin.

Studies were initiated to investigate whether mechanisms exist within mitochondria to repair damage incurred by mitochondrial DNA after exposure to alkylating toxins. A clonal isolate from a rat insulinoma cell line was utilized to measure the formation and repair of alkali-labile sites within the mitochondrial genome after exposure to the alkylating antibiotic streptozotocin. Alkali-labile sites were formed in mitochondrial DNA in a dose-dependent fashion. Eight hours after exposure to the toxin, 55% of the lesions were removed. The level of repair increased to 70% after 24 h. In comparison, only 46% of N7-methylguanines were removed across the entire cellular genome. These studies demonstrate that streptozotocin causes appreciable mitochondrial DNA damage in a dose-dependent manner and provide the first evidence that a repair mechanism for alkali-labile sites is present within the mitochondrion.