Apolipoprotein CIII (apoCIII), a lipid-binding protein involved in the transport of triglycerides and cholesterol in the plasma, is synthesized primarily in the liver and the intestine. A cis-acting regulatory element, C3P, located at -90 to -66 upstream from the apoCIII gene transcriptional start site (+1), is necessary for maximal expression of the apoCIII gene in human hepatoma (HepG2) and intestinal carcinoma (Caco2) cells.This report shows that three members of the steroid receptor superfamily of transcription factors, hepatocyte nuclear factor 4 (HNF-4), apolipoprotein Al regulatory protein 1 (ARP-1), and Ear3/COUP-TF, act at the C3P site. HNF-4 activates apoCIII gene expression in HepG2 and Caco2 cells, while ARP-1 and Ear3/COUP-TF repress its expression in the same cells. HNF-4 activation is abolished by increasing amounts of ARP-1 or Ear3/COUP-TF, and repression by ARP-1 or Ear3/COUP-TF is alleviated by increasing amounts of HNF-4. HNF-4 and ARP-1 bind with similar affinities to the C3P site, suggesting that their opposing transcriptional effects may be mediated by direct competition for DNA binding. HNF-4 and ARP-1 mRNAs are present within the same cells in the liver and intestine, and protein extracts from hepatic tissue, HepG2, and Caco2 cells contain significantly more HNF-4 than ARP-1 or Ear3/COUP-TF binding activities. These findings suggest that the transcription of the apoCIII gene in vivo is dependent, at least in part, upon the intracellular balance of these positive and negative regulatory factors.Apolipoprotein CIII (apoCIII) is a major protein constituent of the triglyceride-rich lipoproteins, very low density lipoproteins, and chylomicrons, and it appears to play an important role in their metabolism by inhibiting the hydrolysis of triglycerides by lipoprotein lipase (4,12,56) and inhibiting the removal of chylomicrons and triglyceride-rich lipoproteins by hepatocytes (44,53,59). ApoCIII plasma levels are often elevated in hypertriglyceridemic individuals (5, 45). Furthermore, overexpression of apoCIII in transgenic mice results in profound hypertriglyceridemia (21). However, the mechanism whereby apoCIII influences triglyceride metabolism has not been clearly defined.

In situ hybridization showed that all fetal hepatocytes contain glutamine synthetase (GS) mRNA but that in adult mouse liver, only a single cell layer surrounding the central veins contains GS mRNA. A shift from the fetal to the adult pattern begins within a few days of birth and is complete within 12 days of birth. Since the total GS mRNA and the transcription rate of the single GS gene are similar at birth and in adults, we conclude that there is a generalized reduction in GS transcription for most hepatocytes and a sharp rise in GS transcription for the immediate pericentral cells. This may be a case of positional regulation of specific gene transcription in apparentdy a single cell lineage.All liver parenchymal cells, or hepatocytes, are thought to be part of the same cell lineage. In the mouse, this lineage arises on day 9 or 10 of gestation when endodermal cells bud from the ventral side of the gut tube and contact mesenchymal cells from the cardiac region (13). Differentiation of hepatocytes is characterized by the coordinated expression of many secreted proteins, such as those found in serum and enzymes active in gluconeogenesis, the urea cycle, or detoxification (16). A good many of these products, serum albumin being the classic example (2, 21), appear to be produced equally by all hepatocytes in accord with the idea of there being a single cell lineage for all hepatocytes. However, it has also been recognized for many years that the concentration of some specialized hepatic proteins was not uniform throughout the liver acini (9, 16). For example, the periportal region is richer in enzymes active in gluconeogenesis (16), while the pericentral area has a higher level of enzymes active in detoxification (17,20,23,28). In the past, this heterogeneity has been attributed largely to gradients of metabolites that exist across the hepatic acini (12, 15) which might be expected to generate gradients of protein concentrations. However, recent examinations have revealed sharp demarcations in the distribution of some specific proteins and mRNAs so that all-or-none positional effects on gene expression seem possible (1,2,9,21). This is perhaps most dramatically illustrated by the presence in adult mouse (2) and rat (10) liver of glutamine synthetase (GS) in a single-cell layer surrounding the central vein. Such a sharp pattern of distribution cannot be explained simply by gradients of metabolites and therefore may represent positional regulation of gene expression within the same cell lineage.We previously showed by immunofluorescence that most mouse fetal hepatocytes stain with antiserum to GS and that not until 10 to 12 days after birth is the adult pattern of GS distribution established (2). In this paper, we used in situ hybridization, mRNA concentration studies, and transcription rate analysis to show that at least part of the regulation for this position-specific expression is at the transcriptional level. A repression of transcription in most adult hepatocytes and an increased transcription in the pericentr...

The mRNA encoding the mouse homolog of

No abstract

To identify the kinds of cells in the brain that express the yes proto-oncogene, we examined chicken brains by using immunofluorescent staining and in situ hybridization. Both approaches showed that the highest level of the yes gene product was in cerebellar Purkinje cells. In addition, we analyzed Purkinje cell degeneration (pcd) mutant mice. The level of yes mRNA in cerebella of pcd mutants was four times lower than that found in cerebella of normal littermates. Our studies point to Purkinje cells as an attractive model for functional studies of the yes protein.The cellular yes (c-yes) gene was identified as a homolog of its oncogenic form found in genomes of Y73 and Esh avian sarcoma viruses (9,19,45,47). It encodes a 62-kilodalton protein that is myristylated, associated with membranes, and cifically in the molecular layer, whereas other parts of the brain showed three-to six-times-lower levels (43).We undertook this study to identify at the cellular level specific areas of the brain in which the yes gene is expressed,