Liver-specific RNA metabolism in hepatoma cells: variations in transcription rates and mRNA levels

Antioxidant and Antisteatotic Activities of Fucoidan Fractions from Marine and Terrestrial Sources

Fucoidan is a fucose-rich sulfated polysaccharide typically found in the cell wall of marine algae but also recently isolated from terrestrial sources. Due to a variety of biological activities, including antioxidant properties, fucoidan exhibits an attractive therapeutic potential against a wide array of metabolic diseases associated with oxidative stress. We used FTIR, 1H NMR and 13C NMR spectroscopy to investigate the structural features of a fucoidan fraction extracted from the brown alga Cystoseira compressa (CYS). The antioxidant potential of CYS was measured by DPPH, ABTS and FRAP assays, which revealed a radical scavenging capacity that was confirmed in in vitro cellular models of hepatic and endothelial cells. The same antioxidant effects were observed for another fucoidan fraction previously identified in the terrestrial tree Eucalyptus globulus (EUC). Moreover, in hepatic cells, CYS and EUC exhibited a significant antisteatotic action, being able to reduce intracellular triglyceride content through the regulation of key genes of hepatic lipid metabolism. EUC exerted stronger antioxidant and antisteatotic effects as compared to CYS, suggesting that both marine and terrestrial sources should be considered for fucoidan extraction and therapeutic applications.

Ablation of Aquaporin-9 Ameliorates the Systemic Inflammatory Response of LPS-Induced Endotoxic Shock in Mouse

Septic shock is the most severe complication of sepsis, being characterized by a systemic inflammatory response following bacterial infection, leading to multiple organ failure and dramatically high mortality. Aquaporin-9 (AQP9), a membrane channel protein mainly expressed in hepatocytes and leukocytes, has been recently associated with inflammatory and infectious responses, thus triggering strong interest as a potential target for reducing septic shock-dependent mortality. Here, we evaluated whether AQP9 contributes to murine systemic inflammation during endotoxic shock. Wild type (Aqp9^+/+; WT) and Aqp9 gene knockout (Aqp9^−/−; KO) male mice were submitted to endotoxic shock by i.p. injection of lipopolysaccharide (LPS; 40 mg/kg) and the related survival times were followed during 72 h. The electronic paramagnetic resonance and confocal microscopy were employed to analyze the nitric oxide (NO) and superoxide anion (O₂⁻) production, and the expression of inducible NO-synthase (iNOS) and cyclooxigenase-2 (COX-2), respectively, in the liver, kidney, aorta, heart and lung of the mouse specimens. LPS-treated KO mice survived significantly longer than corresponding WT mice, and 25% of the KO mice fully recovered from the endotoxin treatment. The LPS-injected KO mice showed lower inflammatory NO and O₂⁻ productions and reduced iNOS and COX-2 levels through impaired NF-κB p65 activation in the liver, kidney, aorta, and heart as compared to the LPS-treated WT mice. Consistent with these results, the treatment of FaO cells, a rodent hepatoma cell line, with the AQP9 blocker HTS13268 prevented the LPS-induced increase of inflammatory NO and O₂⁻. A role for AQP9 is suggested in the early acute phase of LPS-induced endotoxic shock involving NF-κB signaling. The modulation of AQP9 expression/function may reveal to be useful in developing novel endotoxemia therapeutics.

The Nutraceutic Silybin Counteracts Excess Lipid Accumulation and Ongoing Oxidative Stress in an In Vitro Model of Non-Alcoholic Fatty Liver Disease Progression

Non-alcoholic fatty liver disease (NAFLD) is a major cause of liver-related morbidity and mortality. Oxidative stress and release of pro-inflammatory cytokines, such as tumor necrosis factor α (TNFα), are major consequences of hepatic lipid overload, which can contribute to progression of NAFLD to non-alcoholic steatohepatitis (NASH). Also, mitochondria are involved in the NAFLD pathogenesis for their role in hepatic lipid metabolism. Definitive treatments for NAFLD/NASH are lacking so far. Silybin, the extract of the milk thistle seeds, has previously shown beneficial effects in NAFLD. Sequential exposure of hepatocytes to high concentrations of fatty acids (FAs) and TNFα resulted in fat overload and oxidative stress, which mimic in vitro the progression of NAFLD from simple steatosis (SS) to steatohepatitis (SH). The exposure to 50 µM silybin for 24 h reduced fat accumulation in the model of NAFLD progression. The in vitro progression of NAFLD from SS to SH resulted in reduced hepatocyte viability, increased apoptosis and oxidative stress, reduction in lipid droplet size, and up-regulation of IκB kinase β-interacting protein and adipose triglyceride lipase expressions. The direct action of silybin on SS or SH cells and the underlying mechanisms were assessed. Beneficial action of silybin was sustained by changes in expression/activity of peroxisome proliferator-activated receptors and enzymes for FA oxidation. Moreover, silybin counteracted the FA-induced mitochondrial damage by acting on complementary pathways: (i) increased the mitochondrial size and improved the mitochondrial cristae organization; (ii) stimulated mitochondrial FA oxidation; (iii) reduced basal and maximal respiration and ATP production in SH cells; (iv) stimulated ATP production in SS cells; and (v) rescued the FA-induced apoptotic signals and oxidative stress in SH cells. We provide new insights about the direct protective effects of the nutraceutic silybin on hepatocytes mimicking in vitro NAFLD progression.

Multiscale cytometry and regulation of 3D cell cultures on a chip

Three-dimensional cell culture is emerging as a more relevant alternative to the traditional two-dimensional format. Yet the ability to perform cytometry at the single cell level on intact three-dimensional spheroids or together with temporal regulation of the cell microenvironment remains limited. Here we describe a microfluidic platform to perform high-density three-dimensional culture, controlled stimulation, and observation in a single chip. The method extends the capabilities of droplet microfluidics for performing long-term culture of adherent cells. Using arrays of 500 spheroids per chip, in situ immunocytochemistry and image analysis provide multiscale cytometry that we demonstrate at the population scale, on 10⁴ single spheroids, and over 10⁵ single cells, correlating functionality with cellular location within the spheroids. Also, an individual spheroid can be extracted for further analysis or culturing. This will enable a shift towards quantitative studies on three-dimensional cultures, under dynamic conditions, with implications for stem cells, organs-on-chips, or cancer research.3D cell culture is more relevant than the two-dimensional format, but methods for parallel analysis and temporal regulation of the microenvironment are limited. Here the authors develop a droplet microfluidics system to perform long-term culture of 3D spheroids, enabling multiscale cytometry of individual cells within the spheroid.

Liver-enriched transcription factors uncoupled from expression of hepatic functions in hepatoma cell lines

Among the liver-enriched transcription factors identified to date, only expression of hepatocyte nuclear factor 4 (HNF4) and hepatocyte nuclear factor 1 (HNF1) is in strict correlation with hepatic differentiation in cultured rat hepatoma cells. Indeed, differentiated hepatoma cells that stably express an extensive set of adult hepatic functions express liver-enriched transcription factors, while dedifferentiated cells that have lost expression of all these hepatic functions no longer express HNF4 and HNF1. We describe a new heritable phenotype, designated as uncoupled, in which there is a spontaneous dissociation between the expression of these transcription factors and that of the hepatic functions. Cells presenting this phenotype, isolated from differentiated hepatoma cells, cease to accumulate all transcripts coding for hepatic functions but nevertheless maintain expression of HNF4 and HNF1. Transitory transfection experiments indicate that these two factors present in these cells have transcriptional activity similar to that of differentiated hepatoma cells. Characterization of the appropriate intertypic cell hybrids demonstrates that this new phenotype is recessive to the dedifferentiated state and fails to be complemented by differentiated cells. These results indicate the existence of mechanisms that inhibit transcription of genes coding for hepatocyte functions in spite of the presence of functional HNF4 and HNF1. Cells of the uncoupled phenotype present certain properties of oval cells described for pathological states of the liver.

The bifunctional DCOH protein binds to HNF1 independently of its 4-alpha-carbinolamine dehydratase activity

HNF1 is a liver enriched atypical homeoprotein isolated from vertebrates which is involved in the transcriptional activation of liver, kidney, intestine and pancreas specific genes. HNF1 contains an N-terminal dimerisation and a POU-like domain both essential together with the homeodomain for DNA specific recognition. Using the yeast two-hybrid system we searched for proteins interacting with HNF1. We repeatedly obtained cDNA clones encoding DCOH/4-alpha-carbinolamine dehydratase, an enzyme involved in the oxidation of aromatic amino acids that was shown to bind to and stabilise HNF1 dimers. Using the yeast system, we show that the enzymatic activity of DCOH is not essential for HNF1 binding and that the HNF1 dimerisation domain is sufficient for DCOH binding. Furthermore we demonstrate that both proteins co-localise in co-transfected cells.

Differential effects of cell density on 5-lipoxygenase (5-LO), five-lipoxygenase-activating protein (FLAP) and interleukin-1 beta (IL-1 beta) expression in human neutrophils

We have analyzed the effect of cellular density of 5-Lipoxygenase (5-LO), 5-lipoxygenase-activating protein (FLAP) and interleukin-1 beta (IL-1 beta) gene expression in neutrophils from healthy subjects under culture conditions of low and high cell density. By using RT-PCR techniques, we have found that 5-LO mRNA accumulation decreased in cells cultured at high density, while FLAP mRNA is not affected. De novo 5-LO synthesis, as well as steady-state levels, were reduced in cells maintained at high density. In contrast, the high density conditions lead to the induction of IL-1 beta gene at the RNA and protein levels as measured by RT-PCR and by immunoprecipitation. These results suggest that cellular density plays a role in gene modulation when neutrophils are accumulating at an inflammatory site since neutrophils obtained from the synovial fluid of patients with RA exhibit a protein synthesis profile similar to that observed in peripheral blood neutrophils cultured at high density.

Regulation of human C-reactive protein gene expression by two synergistic IL-6 responsive elements

To study the mechanism of interleukin-6 (IL-6) induction of human C-reactive protein (CRP) gene expression, we have utilized a human hepatoma (PLC/PRF/5) cell culture system to analyze the trans-acting factors which bind to the 300 bp 5'-flanking region of human CRP gene. In vitro gel mobility shift analyses and methylation interference assays demonstrated that NFIL-6 alpha interacted with two IL-6 responsive elements, and HNF-1 alpha and HNF-3/Octamer-like factors interacted with the downstream IL-6 responsive element in the human CRP promoter. In vivo functional analysis by transient transfection of plasmid constructs containing site-specific mutations in one or two IL-6 responsive elements in the CRP promoter fused to a reporter gene, chloramphenicol acetyl transferase (CAT), demonstrated that the binding of NFIL-6 alpha to two IL-6 responsive elements resulted in synergistic induction of the gene. When HNF-1 alpha or HNF-3/Octamer-like factors were independently bound to their corresponding sites, they had either a positive or negative effect, respectively, on IL-6 inducible transcriptional activity.

Preproenkephalin promoter yields region-specific and long-term expression in adult brain after direct in vivo gene transfer via a defective herpes simplex viral vector

We have previously used a defective herpes simplex virus vector to express a foreign gene in the adult rat brain. One application of this technology would be the in vivo analysis of promoter function in brain after de novo transfer, which would allow the rapid generation of vectors with localized application in a broad range of mammalian species while avoiding influences of other nearby promoters. A 2.7-kb fragment of the rat preproenkephalin promoter was placed upstream of the bacterial lacZ gene in our herpes simplex virus amplicon. A restricted pattern of lacZ expression was observed in vivo, which follows previously observed patterns of endogenous preproenkephalin expression. These results, from the direct gene transfer into an adult animal brain for in vivo promoter analysis, demonstrate that sequence information that influences restricted expression of preproenkephalin is located within 2.7 kb upstream of transcriptional initiation. lacZ expression was also observed in rat brain for 2 months after direct transfer, and PCR analysis confirmed the continued presence of amplicon DNA in lacZ-positive sections. Restricted and long-term expression observed with an endogenous promoter has important implications for gene therapy using viral vectors.

Suppression of albumin enhancer activity by H-ras and AP-1 in hepatocyte cell lines

We demonstrated, using a transient transfection assay, that the albumin enhancer increased the expression of the albumin promoter in a highly differentiated, simian virus 40 (SV40)-immortalized hepatocyte cell line, CWSV1, but was not functional in two ras-transformed cell lines (NR3 and NR4) derived from CWSV1 by stable transfection with the T24ras oncogene. A transient cotransfection assay showed that T24ras and normal c-Ha-ras were each able to inhibit the activity of the albumin enhancer in an immortal hepatocyte cell line. DNase I footprinting and gel mobility shift assays demonstrated that the DNA binding activities specific to the albumin enhancer were not decreased in the ras-transformed cells. ras also did not diminish the expression of HNF1 alpha, C/EBP alpha, HNF3 alpha, HNF3 beta, or HNF3 gamma but did significantly increase AP-1 binding activity. Three AP-1 binding sites were identified within the albumin enhancer, and DNA binding activities specific to these AP-1 sites were induced in the ras-transformed hepatocytes. Subsequent functional assays showed that overexpression of c-jun and c-fos inhibited the activity of the albumin enhancer. Site-directed mutagenesis of the AP-1 binding sites in the albumin enhancer partially abrogated the suppressing effect of ras and c-jun/c-fos on the enhancer. These functional studies therefore supported the results of the structural studies with AP-1. We conclude that the activity of the albumin enhancer is subject to regulation by ras signaling pathways and that the effect of ras on the albumin enhancer activity may be mediated by AP-1.

Suppression of albumin enhancer activity by H-ras and AP-1 in hepatocyte cell lines

We demonstrated, using a transient transfection assay, that the albumin enhancer increased the expression of the albumin promoter in a highly differentiated, simian virus 40 (SV40)-immortalized hepatocyte cell line, CWSV1, but was not functional in two ras-transformed cell lines (NR3 and NR4) derived from CWSV1 by stable transfection with the T24ras oncogene. A transient cotransfection assay showed that T24ras and normal c-Ha-ras were each able to inhibit the activity of the albumin enhancer in an immortal hepatocyte cell line. DNase I footprinting and gel mobility shift assays demonstrated that the DNA binding activities specific to the albumin enhancer were not decreased in the ras-transformed cells. ras also did not diminish the expression of HNF1 alpha, C/EBP alpha, HNF3 alpha, HNF3 beta, or HNF3 gamma but did significantly increase AP-1 binding activity. Three AP-1 binding sites were identified within the albumin enhancer, and DNA binding activities specific to these AP-1 sites were induced in the ras-transformed hepatocytes. Subsequent functional assays showed that overexpression of c-jun and c-fos inhibited the activity of the albumin enhancer. Site-directed mutagenesis of the AP-1 binding sites in the albumin enhancer partially abrogated the suppressing effect of ras and c-jun/c-fos on the enhancer. These functional studies therefore supported the results of the structural studies with AP-1. We conclude that the activity of the albumin enhancer is subject to regulation by ras signaling pathways and that the effect of ras on the albumin enhancer activity may be mediated by AP-1.

HNF4 and HNF1 as well as a panel of hepatic functions are extinguished and reexpressed in parallel in chromosomally reduced rat hepatoma-human fibroblast hybrids

Rat hepatoma-human fibroblast hybrids of two independent lineages containing only 8-11 human chromosomes show pleiotropic extinction of thirteen out of fifteen hepatic functions examined. Reexpression of the entire group of functions most often occurs in a block, and except for one discordant subclone, correlates with loss of human chromosome 2. The extinguished cells and their reexpressing derivatives have been examined for the expression of seven liver-enriched transcription factors. C/EBP, LAP, DBP, HNF3, and vHNF1 expression are not systematically extinguished in parallel with the hepatic functions. However, HNF1 and HNF4 show a perfect correlation with phenotype: these factors are expressed only in the cells showing pleiotropic reexpression. Since recent evidence indicates that HNF4 controls HNF1 expression, it can be proposed that the HNF4 gene is the primary target of the pleiotropic extinguisher.

Tissue-specific, developmental, hormonal, and dietary regulation of rat phosphoenolpyruvate carboxykinase-human growth hormone fusion genes in transgenic mice

The cytosolic phosphoenolpyruvate carboxykinase (PEPCK) gene is expressed in multiple tissues and is regulated in a complex tissue-specific manner. To map the cis-acting DNA elements that direct this tissue-specific expression, we made transgenic mice containing truncated PEPCK-human growth hormone (hGH) fusion genes. The transgenes contained PEPCK promoter fragments with 5' endpoints at -2088, -888, -600, -402, and -207 bp, while the 3' endpoint was at +69 bp. Immunohistochemical analysis showed that the -2088 transgene was expressed in the correct cell types (hepatocytes, proximal tubular epithelium of the kidney, villar epithelium of the small intestine, epithelium of the colon, smooth muscle of the vagina and lungs, ductal epithelium of the sublingual gland, and white and brown adipocytes). Solution hybridization of hGH mRNA expressed from the transgenes indicated that white and brown fat-specific elements are located distally (-2088 to -888 bp) and that liver-, gut-, and kidney-specific elements are located proximally (-600 to +69 bp). However, elements outside of the region tested are necessary for the correct developmental pattern and level of PEPCK expression in kidney. Both the -2088 and -402 transgenes responded in a tissue-specific manner to dietary stimuli, and the -2088 transgene responded to glucocorticoid stimuli. Thus, different tissues utilize distinct cell-specific cis-acting elements to direct and regulate the PEPCK gene.

Tissue-specific, developmental, hormonal, and dietary regulation of rat phosphoenolpyruvate carboxykinase-human growth hormone fusion genes in transgenic mice

The cytosolic phosphoenolpyruvate carboxykinase (PEPCK) gene is expressed in multiple tissues and is regulated in a complex tissue-specific manner. To map the cis-acting DNA elements that direct this tissue-specific expression, we made transgenic mice containing truncated PEPCK-human growth hormone (hGH) fusion genes. The transgenes contained PEPCK promoter fragments with 5' endpoints at -2088, -888, -600, -402, and -207 bp, while the 3' endpoint was at +69 bp. Immunohistochemical analysis showed that the -2088 transgene was expressed in the correct cell types (hepatocytes, proximal tubular epithelium of the kidney, villar epithelium of the small intestine, epithelium of the colon, smooth muscle of the vagina and lungs, ductal epithelium of the sublingual gland, and white and brown adipocytes). Solution hybridization of hGH mRNA expressed from the transgenes indicated that white and brown fat-specific elements are located distally (-2088 to -888 bp) and that liver-, gut-, and kidney-specific elements are located proximally (-600 to +69 bp). However, elements outside of the region tested are necessary for the correct developmental pattern and level of PEPCK expression in kidney. Both the -2088 and -402 transgenes responded in a tissue-specific manner to dietary stimuli, and the -2088 transgene responded to glucocorticoid stimuli. Thus, different tissues utilize distinct cell-specific cis-acting elements to direct and regulate the PEPCK gene.

Recognition of the CDEI motif GTCACATG by mouse nuclear proteins and interference with the early development of the mouse embryo

We have reported previously (1) two unexpected consequences of the microinjection into fertilized mouse eggs of a recombinant plasmid designated p12B1, carrying a 343 bp insert of non-repetitive mouse DNA. Injected at very low concentrations, this plasmid could be established as an extrachromosomal genetic element. When injected in greater concentration, an early arrest of embryonic development resulted. In the present work, we have studied this toxic effect in more detail by microinjecting short synthetic oligonucleotides with sequences from the mouse insert. Lethality was associated with the nucleotide sequence GTCACATG, identical with the CDEl element of yeast centromeres. Development of injected embryos was arrested between the one-cell and the early morula stages, with abnormal structures and DNA contents. Electrophoretic mobility shift and DNAse foot-printing assays demonstrated the binding of mouse nuclear protein(s) to the CDEl-like box. Base changes within the CDEl sequence prevented both the toxic effects in embryos and the formation of protein complex in vitro, suggesting that protein binding at such sites in chromosomal DNA plays an important role in early development.

Hybrid cell lines constitute a potential reservoir of polarized cells: isolation and study of highly differentiated hepatoma-derived hybrid cells able to form functional bile canaliculi in vitro

A large number of hepatoma cell lines has been used to study expression and regulation of liver-specific function. However these cells, even the most differentiated, are morphologically far from hepatocytes. In no case is the typical hepatocyte cell polarity well maintained. Cell hybridization has been used as a potential means for turning on specific genes. From hybrids between well differentiated Fao rat hepatoma cells and WI 38 human fibroblasts, we have attempted to isolate segregated cells that are highly differentiated and polarized. Such cells, detected in aged cultures of only one hybrid (WIF12), were isolated by subcloning. One subclone, WIF12-1 was analyzed. Expression of liver-specific functions extinguished in the original hybrid is restored in all WIF12-1 cells at a very high level, similar to that of hepatocytes and 5-30 times higher that that of parental cells. Moreover human genes coding for liver-specific proteins (albumin, fibrinogen, and alcohol dehydrogenase) are actively expressed. WIF12-1 cells have acquired a polarized phenotype as attested by the presence of bile canaliculi between adjacent cells and by the asymmetrical localization of apical (Mg(2+)-ATPase, gamma-glutamyl transpeptidase) and basolateral membrane markers. The bile canaliculi formed are dynamic and functional structures, characterized by long periods of expansion followed by rapid contractions. The ability to polarize is a general and permanent property of WIF12-1 cells. These cells appear to constitute a valid model for the in vitro study of hepatocyte cell polarity, membrane domain formation and mechanisms of membrane protein sorting.

Functional characterization of a complex protein-DNA-binding domain located within the human immunodeficiency virus type 1 long terminal repeat leader region

Transcriptional trans activation of the human immunodeficiency virus type 1 (HIV-1) long terminal repeat (LTR) by the viral tat trans activator is mediated by an LTR-specific sequence located immediately 3' to the start of transcription initiation. We have used a range of molecular techniques to examine DNA-protein interactions that occur in the vicinity of this cis-acting sequence. Our results demonstrate the existence of a sequence-specific DNA-protein interaction involving the HIV-1 leader DNA and map this binding event to between -2 and +21 base pairs relative to the HIV-1 LTR transcription start site. Evidence suggesting that this interaction involves three distinct protein-DNA contact sites extending along one side of the DNA helix is presented. Mutation of these sites was found to ablate protein-DNA binding yet was observed to have no effect on either the basal or tat trans-activated level of HIV-1 LTR-specific gene expression. We therefore conclude that this DNA-protein interaction has a function distinct from the regulation of HIV-1 LTR-specific gene expression.

Positive and negative regulatory elements in the mouse albumin enhancer

Most enhancer elements so far described are comprised of multiple DNA binding sites for proteins that act to increase the rate of transcription. In this report we show that the far upstream mouse albumin enhancer element (-10.5 to -8.5 kilobase pairs) is a composite of at least three functional regions: a negative region that suppresses an otherwise positive, liver-specific enhancer element and a third region that by itself is inactive but in conjunction with the remaining elements overrides the effect of the negative region. The protein that binds to the overriding element is C/EBP, a liver-enriched transcription factor, whereas the newly identified protein that is responsible for full liver-specific enhancer activity is found in HeLa cell and spleen extracts as well as in extracts prepared from the liver. This latter protein may therefore be an example of a widely distributed protein that because of cell-specific modification or interaction has a tissue-specific, positive action on transcription.

Tissue-specific gene expression in mouse hepatocytes cultured in growth-restricting medium

Culture conditions which maintain hepatocytes in their in vivo state are not known. This hampers the study of liver gene expression and of direct responses of liver genes to hormonal stimulation. We argued that hepatocytes that were unable to divide might retain in vivo characteristics. We therefore plated mouse (BALB/c) hepatocytes on plastic dishes in medium lacking arginine and measured the levels and transcription rates of six tissue-specific mRNAs over a period of days. Alpha-fetoprotein mRNA began to accumulate at about 48 h of culture, and transcription could sometimes be detected after 72 h. The levels and transcription rates of four mRNAs (albumin, alpha-1-antitrypsin, apolipoprotein A1, and major urinary protein [MUP]) fell sharply. The rate of transcription of transferrin mRNA fell less rapidly, and its level remained high, partly due to its longer half-life. The overall pattern of gene expression in the plated cells did not exactly parallel that of either fetal or regenerating liver. The hepatocytes remained responsive to hormonal stimulation. Insulin and dexamethasone each tended to counteract changes in mRNA levels, for example, preventing the accumulation of alpha-fetoprotein mRNA. The effects of insulin were primarily due to changes in transcription rates. Bovine growth hormone and thyroxine elevated the levels of most of the mRNAs. Many of the effects of these hormones, when added singly, could not be ascribed to changes in transcription. The level of MUP mRNA was strongly affected by added hormones. The mRNA level at 5 days was increased by added insulin, dexamethasone, growth hormone, and thyroxine. In the presence of these three hormones, the decay in the transcription rate of the MUP genes was reduced about 10-fold. We conclude that hepatocytes plated under these nongrowing conditions can provide insights into the hormonal responsiveness of tissue-specific genes.

Conditional enhancement of liver-specific gene transcription

We sought to develop a cell line in which liver-specific transcription could be induced at will, to facilitate the study of factors that cause hepatocyte-specific transcription of the serum albumin gene in mice. We therefore created the H2.35 cell line from mouse hepatocytes infected with a temperature-sensitive strain of simian virus 40. During routine propagation at the permissive temperature, H2.35 cells exhibit extremely low levels of albumin transcription and mRNA. Albumin mRNA increases at least 100-fold when H2.35 cells are cultured at the restrictive temperature and in serum-free medium on a collagen substratum; the two latter conditions maintain the differentiated state of primary hepatocyte cultures. Although a major cause of the mRNA increase is posttranscriptional, the transcription rates of albumin and other liver-specific genes increase significantly. Transient-transfection experiments demonstrated that an induction of transcription is caused by activation of an albumin upstream sequence that was previously shown to enhance liver-specific transcription in transgenic mice. Thus, hepatocyte differentiation appears to be maintained in part by extracellular signals that stimulate the activity of a tissue-specific enhancer element.

Tissue-specific gene expression in mouse hepatocytes cultured in growth-restricting medium

Culture conditions which maintain hepatocytes in their in vivo state are not known. This hampers the study of liver gene expression and of direct responses of liver genes to hormonal stimulation. We argued that hepatocytes that were unable to divide might retain in vivo characteristics. We therefore plated mouse (BALB/c) hepatocytes on plastic dishes in medium lacking arginine and measured the levels and transcription rates of six tissue-specific mRNAs over a period of days. Alpha-fetoprotein mRNA began to accumulate at about 48 h of culture, and transcription could sometimes be detected after 72 h. The levels and transcription rates of four mRNAs (albumin, alpha-1-antitrypsin, apolipoprotein A1, and major urinary protein [MUP]) fell sharply. The rate of transcription of transferrin mRNA fell less rapidly, and its level remained high, partly due to its longer half-life. The overall pattern of gene expression in the plated cells did not exactly parallel that of either fetal or regenerating liver. The hepatocytes remained responsive to hormonal stimulation. Insulin and dexamethasone each tended to counteract changes in mRNA levels, for example, preventing the accumulation of alpha-fetoprotein mRNA. The effects of insulin were primarily due to changes in transcription rates. Bovine growth hormone and thyroxine elevated the levels of most of the mRNAs. Many of the effects of these hormones, when added singly, could not be ascribed to changes in transcription. The level of MUP mRNA was strongly affected by added hormones. The mRNA level at 5 days was increased by added insulin, dexamethasone, growth hormone, and thyroxine. In the presence of these three hormones, the decay in the transcription rate of the MUP genes was reduced about 10-fold. We conclude that hepatocytes plated under these nongrowing conditions can provide insights into the hormonal responsiveness of tissue-specific genes.

A variant nuclear protein in dedifferentiated hepatoma cells binds to the same functional sequences in the beta fibrinogen gene promoter as HNF-1

Normal liver and differentiated hepatoma cell lines contain a nuclear factor, HNF-1, which binds functional sequences within the promoters of the alpha and beta chains of fibrinogen and alpha 1-antitrypsin. In UV cross-linking studies we find that HNF-1 has an apparent mol. wt of 92 kd in differentiated hepatocytes. Nuclear extracts from a dedifferentiated hepatoma cell line, Fao flC2 (C2), selected on the basis of morphological and biochemical dedifferentiation from Fao contains a protein, vHNF, which binds to the same DNA sequence motif as HNF-1 but has an apparent mol. wt of 72 rather than 92 kd. Mixing experiments indicate that this variant nuclear factor does not arise from HNF-1 by proteolysis. Reversion to the differentiated phenotype in C2-Rev7 (Rev7), selected by growth in glucose-free media, results in the re-expression of many liver-specific functions including the fibrinogen genes. In Rev7, HNF-1 is indistinguishable from that in the original differentiated cell line Fao. Transfection studies and nuclear run-on experiments indicate that reduced expression of fibrinogen RNA in C2 relative to Fao is related to reduced transcription. vHNF but not HNF-1 is present in somatic hybrids between fibroblasts and liver cells which show extinction of liver specific traits and it can also be detected in normal tissue, predominantly in lung nuclear extracts. Since vHNF and HNF-1 are not co-expressed yet correlate with the non-hepatic and hepatic phenotype, respectively, we suggest that the expression of these variant forms reflects determination events in establishing the hepatic phenotype.

Regulation of albumin gene expression in a series of rat hepatocyte cell lines immortalized by simian virus 40 and maintained in chemically defined medium

A series of simian virus 40 (SV40)-immortalized hepatocyte cell lines were characterized for albumin production, the regulation of albumin production, and the expression of other liver-specific genes. This series of cell lines is particularly useful for studying the regulation of hepatocyte gene expression because the cell lines express liverlike levels of a number of liver-specific functions and do so while growing in a chemically defined medium. SV40-immortalized hepatocyte cell lines were derived from colonies of albumin-producing epithelial cells that arose after primary hepatocytes maintained in chemically defined medium were transfected with SV40 DNA. Some cell lines secreted albumin at levels equal to or greater than those secreted by freshly plated primary hepatocytes, and all but one line continued to produce albumin for more than 20 passages. The variation in albumin secretion among cell lines reflected differences in the amount of albumin produced per cell and not in the percentage of albumin-producing cells in each line. The characterization of selected cell lines showed that albumin production was regulated by cell density during the growth cycle. Albumin production in most cell lines was also regulated by dexamethasone; however, one cell line continued to produce high levels of albumin when the cells were grown in medium lacking dexamethasone, demonstrating that although glucocorticoid can induce albumin production in some cell lines, it is not required for high levels of albumin production by all cells in culture. Regulation of albumin production measured at the level of protein secretion was paralleled by changes in steady-state levels of a 2.3-kilobase albumin RNA. Albumin-producing SV40-immortalized hepatocytes secreted a variety of other plasma proteins, including transferrin, hemopexin, and the third component of complement. These cells also expressed tyrosine aminotransferase activity that was inducible by dexamethasone. Alpha-fetoprotein production was not detected in any of the cell lines examined.

Regulation of albumin gene expression in a series of rat hepatocyte cell lines immortalized by simian virus 40 and maintained in chemically defined medium

A series of simian virus 40 (SV40)-immortalized hepatocyte cell lines were characterized for albumin production, the regulation of albumin production, and the expression of other liver-specific genes. This series of cell lines is particularly useful for studying the regulation of hepatocyte gene expression because the cell lines express liverlike levels of a number of liver-specific functions and do so while growing in a chemically defined medium. SV40-immortalized hepatocyte cell lines were derived from colonies of albumin-producing epithelial cells that arose after primary hepatocytes maintained in chemically defined medium were transfected with SV40 DNA. Some cell lines secreted albumin at levels equal to or greater than those secreted by freshly plated primary hepatocytes, and all but one line continued to produce albumin for more than 20 passages. The variation in albumin secretion among cell lines reflected differences in the amount of albumin produced per cell and not in the percentage of albumin-producing cells in each line. The characterization of selected cell lines showed that albumin production was regulated by cell density during the growth cycle. Albumin production in most cell lines was also regulated by dexamethasone; however, one cell line continued to produce high levels of albumin when the cells were grown in medium lacking dexamethasone, demonstrating that although glucocorticoid can induce albumin production in some cell lines, it is not required for high levels of albumin production by all cells in culture. Regulation of albumin production measured at the level of protein secretion was paralleled by changes in steady-state levels of a 2.3-kilobase albumin RNA. Albumin-producing SV40-immortalized hepatocytes secreted a variety of other plasma proteins, including transferrin, hemopexin, and the third component of complement. These cells also expressed tyrosine aminotransferase activity that was inducible by dexamethasone. Alpha-fetoprotein production was not detected in any of the cell lines examined.

Expression of six mouse major urinary protein genes in the mammary, parotid, sublingual, submaxillary, and lachrymal glands and in the liver

Mouse major urinary proteins (MUPs) are encoded by a family of about 35 to 40 highly conserved genes. In the preceding paper (K. Shahan, M. Gilmartin, and E. Derman, Mol. Cell. Biol. 7:1938-1946, 1987), we presented the sequences of the most abundant MUP mRNAs in the liver (MUP I, II, and III) and in the lachrymal (MUP IV) and submaxillary (MUP V) glands. We have shown that these five mRNAs are coded by five distinct genes, MUP I through V. In the present communication, we examine the expression of MUP genes in all of the six tissues in which MUP mRNAs are synthesized, the mammary, parotid, sublingual, lachrymal, and submaxillary glands and the liver. We show that gene MUP II is expressed in the liver and in the mammary gland, that gene MUP IV is expressed in the lachrymal and parotid glands, and that gene MUP V is expressed in the submaxillary, sublingual, and lachrymal and parotid glands, and that gene MUP V is expressed in the submaxillary, sublingual, and lachrymal glands. Furthermore, we present evidence that in addition to genes MUP I through V, another gene, MUP VI, is expressed in BALB/c mice in the parotid gland. The tissue-specific synthesis of MUP mRNAs is thus brought about by two major mechanisms: the expression, in different tissues, of different members of the family and the expression of a single gene at various levels in different tissues. When a particular MUP gene is expressed in several tissues, transcripts of this gene initiate at the same site and are spliced and polyadenylated in the same manner.

Expression of six mouse major urinary protein genes in the mammary, parotid, sublingual, submaxillary, and lachrymal glands and in the liver

Mouse major urinary proteins (MUPs) are encoded by a family of about 35 to 40 highly conserved genes. In the preceding paper (K. Shahan, M. Gilmartin, and E. Derman, Mol. Cell. Biol. 7:1938-1946, 1987), we presented the sequences of the most abundant MUP mRNAs in the liver (MUP I, II, and III) and in the lachrymal (MUP IV) and submaxillary (MUP V) glands. We have shown that these five mRNAs are coded by five distinct genes, MUP I through V. In the present communication, we examine the expression of MUP genes in all of the six tissues in which MUP mRNAs are synthesized, the mammary, parotid, sublingual, lachrymal, and submaxillary glands and the liver. We show that gene MUP II is expressed in the liver and in the mammary gland, that gene MUP IV is expressed in the lachrymal and parotid glands, and that gene MUP V is expressed in the submaxillary, sublingual, and lachrymal and parotid glands, and that gene MUP V is expressed in the submaxillary, sublingual, and lachrymal glands. Furthermore, we present evidence that in addition to genes MUP I through V, another gene, MUP VI, is expressed in BALB/c mice in the parotid gland. The tissue-specific synthesis of MUP mRNAs is thus brought about by two major mechanisms: the expression, in different tissues, of different members of the family and the expression of a single gene at various levels in different tissues. When a particular MUP gene is expressed in several tissues, transcripts of this gene initiate at the same site and are spliced and polyadenylated in the same manner.

Cellular promoters incorporated into the adenovirus genome: effects of viral regulatory elements on transcription rates and cell specificity of albumin and beta-globin promoters

In the accompanying paper (Friedman et al., Mol. Cell. Biol. 6:3791-3797, 1986), hepatoma-specific expression of the rat albumin promoter within the adenovirus genome was demonstrated. However, the rate of transcription was very low compared with that of the endogenous chromosomal albumin gene. Here we show that in hepatoma cells the adenovirus E1A enhancer, especially in the presence of E1A protein, greatly stimulates transcription from the albumin promoter but not the mouse beta-globin promoter. This enhancer-dependent stimulation did not occur in myeloma cells in which a virus containing a immunoglobulin promoter and enhancer did function. These experiments suggest a limited distribution in cultured differentiated cells of cell-specific transcription factors. However, either the regulation of such cell-specific factors breaks down in other cultured cells, or strictly cell-specific factors are not at play in controlling cell-specific transcription, because HeLa cells could transcribe the albumin promoter from the same start site about 10% as well as hepatomas could and 293 cells could transcribe both albumin and globin promoters.

Expression of a human cytomegalovirus late gene is posttranscriptionally regulated by a 3'-end-processing event occurring exclusively late after infection

A phenomenon of posttranscriptional regulation has been previously identified in cytomegalovirus-infected human fibroblast cells (Wathen and Stinski, J. Virol. 41:462, 1982). A region typifying this phenomenon has been located within the large unique component of the viral genome (map units 0.408 to 0.423). Even though this transcriptional unit was highly transcribed at early times after infection, mRNAs from this region were only detectable on the polyribosomes after viral DNA replication. Thus, this region is believed to code for a late gene. Single-strand-specific nuclease mapping experiments of viral transcripts established that the transcriptional initiation sites and the 5' ends of a downstream exon were identical at early and late times. However, the late transcripts differed from the early transcripts by the processing of the 3' end of the viral RNAs. This involved either the removal of a distinct region of the transcript by the selection of an upstream cleavage and polyadenylation site or the differential splicing of the RNA molecule. The upstream cleavage and polyadenylation site was identified by nuclease mapping analyses and DNA sequencing. The 3'-end processing of these transcripts is necessary for the detection of these viral RNAs within the cytoplasm of the infected cell. We propose that human cytomegalovirus either codes for a factor(s) that is involved in the 3'-end-processing event at late times after infection or stimulates the synthesis of a host cell factor(s) involved in this complex regulatory event. This level of regulation may have an influence on the types of cells that permit productive cytomegalovirus replication.

Expression of a human cytomegalovirus late gene is posttranscriptionally regulated by a 3'-end-processing event occurring exclusively late after infection

A phenomenon of posttranscriptional regulation has been previously identified in cytomegalovirus-infected human fibroblast cells (Wathen and Stinski, J. Virol. 41:462, 1982). A region typifying this phenomenon has been located within the large unique component of the viral genome (map units 0.408 to 0.423). Even though this transcriptional unit was highly transcribed at early times after infection, mRNAs from this region were only detectable on the polyribosomes after viral DNA replication. Thus, this region is believed to code for a late gene. Single-strand-specific nuclease mapping experiments of viral transcripts established that the transcriptional initiation sites and the 5' ends of a downstream exon were identical at early and late times. However, the late transcripts differed from the early transcripts by the processing of the 3' end of the viral RNAs. This involved either the removal of a distinct region of the transcript by the selection of an upstream cleavage and polyadenylation site or the differential splicing of the RNA molecule. The upstream cleavage and polyadenylation site was identified by nuclease mapping analyses and DNA sequencing. The 3'-end processing of these transcripts is necessary for the detection of these viral RNAs within the cytoplasm of the infected cell. We propose that human cytomegalovirus either codes for a factor(s) that is involved in the 3'-end-processing event at late times after infection or stimulates the synthesis of a host cell factor(s) involved in this complex regulatory event. This level of regulation may have an influence on the types of cells that permit productive cytomegalovirus replication.

Cellular promoters incorporated into the adenovirus genome: effects of viral regulatory elements on transcription rates and cell specificity of albumin and beta-globin promoters

In the accompanying paper (Friedman et al., Mol. Cell. Biol. 6:3791-3797, 1986), hepatoma-specific expression of the rat albumin promoter within the adenovirus genome was demonstrated. However, the rate of transcription was very low compared with that of the endogenous chromosomal albumin gene. Here we show that in hepatoma cells the adenovirus E1A enhancer, especially in the presence of E1A protein, greatly stimulates transcription from the albumin promoter but not the mouse beta-globin promoter. This enhancer-dependent stimulation did not occur in myeloma cells in which a virus containing a immunoglobulin promoter and enhancer did function. These experiments suggest a limited distribution in cultured differentiated cells of cell-specific transcription factors. However, either the regulation of such cell-specific factors breaks down in other cultured cells, or strictly cell-specific factors are not at play in controlling cell-specific transcription, because HeLa cells could transcribe the albumin promoter from the same start site about 10% as well as hepatomas could and 293 cells could transcribe both albumin and globin promoters.

Alteration of cellular gene expression in adenovirus transformed cells by post-transcriptional mechanisms

We have isolated cDNA clones complementary to human mRNAs that are expressed at elevated levels in 293 cells, adenovirus-transformed human embryonic kidney cells, as compared to a normal counterpart of this cell line. Approximately 200 clones out of 100,000 that were screened were positive; 40 of these were isolated, of which 31 were determined to be unique and were further characterized. Each clone detected a mRNA that was 5 to 50 times more abundant in 293 cells than in the non-transformed HEK cell line. For several of these transcripts, the elevated expression appeared to be a function of transformation since they were also high in other human tumor cell lines. Strikingly, we have found that post-transcriptional control is largely responsible for the regulation of the abundance of these mRNAs.

Dependence of liver-specific transcription on tissue organization

When the liver is disaggregated and hepatocytes are cultured as a cellular monolayer for 24 h, a sharp decline (80 to 99% decrease) in the transcription of most liver-specific mRNAs, but not common mRNAs, occurs (Clayton and Darnell, Mol. Cell. Biol. 2:1552-1561, 1983). A wide variety of culture conditions involving various hormones and substrates and cocultivation with other cells failed to sustain high rates of liver-specific mRNA synthesis in cultured hepatocytes, although they continued to synthesize common mRNAs at normal or elevated rates. In contrast, when slices of intact mouse liver tissue were placed in culture, the transcription of liver-specific genes was maintained at high levels (20 to 100% of normal liver). Furthermore, we found that cells in the liver could be disengaged and immediately reengaged in a tissue-like structure by perfusing the liver with EDTA followed by serum-containing culture medium. Slices of reengaged liver continued to transcribe tissue-specific mRNA sequences at significantly higher rates after 24 h in culture than did individual cells isolated by EDTA perfusion followed by culturing as a monolayer. Therefore we conclude that a mature tissue structure plays an important role in the maintenance of maximum tissue-specific transcription in liver cells.

Transcriptional and posttranscriptional control of immunoglobulin mRNA production during B lymphocyte development

A variety of cell lines representing different maturation stages of the B lymphocyte were used to analyse developmental changes in the transcriptional pattern through the mu-delta locus and the relationship between mu mRNA accumulation and transcriptional activity. As anticipated from earlier studies, we observed that RNA polymerase loading in the region between the mu m cleavage/poly A addition site and the delta 1 exon is markedly decreased in IgM secreting cells compared to cells bearing surface IgM or surface IgM and IgD. In several IgM secreting hybridomas, transcriptional termination mainly occurred downstream of the first mu m exon. Thus, the predominance of mu s-terminated transcripts in these cells would appear to be principally determined by RNA processing events, most likely by more efficient cleavage at the mu s poly A site and/or less efficient splicing of the C mu and mu m exons. In two plasmacytoma lines, polymerase unloading between the mu s and mu m sites also contributed significantly to the high mu s mRNA phenotype. Our results further indicate that posttranscriptional regulation is largely responsible for the greatly increased accumulation of mu mRNA in the IgM secretors. Interestingly, the sterile-mu RNA components do not seem to be subject to this posttranscriptional regulation.

Dependence of liver-specific transcription on tissue organization

When the liver is disaggregated and hepatocytes are cultured as a cellular monolayer for 24 h, a sharp decline (80 to 99% decrease) in the transcription of most liver-specific mRNAs, but not common mRNAs, occurs (Clayton and Darnell, Mol. Cell. Biol. 2:1552-1561, 1983). A wide variety of culture conditions involving various hormones and substrates and cocultivation with other cells failed to sustain high rates of liver-specific mRNA synthesis in cultured hepatocytes, although they continued to synthesize common mRNAs at normal or elevated rates. In contrast, when slices of intact mouse liver tissue were placed in culture, the transcription of liver-specific genes was maintained at high levels (20 to 100% of normal liver). Furthermore, we found that cells in the liver could be disengaged and immediately reengaged in a tissue-like structure by perfusing the liver with EDTA followed by serum-containing culture medium. Slices of reengaged liver continued to transcribe tissue-specific mRNA sequences at significantly higher rates after 24 h in culture than did individual cells isolated by EDTA perfusion followed by culturing as a monolayer. Therefore we conclude that a mature tissue structure plays an important role in the maintenance of maximum tissue-specific transcription in liver cells.