Proteins binding to the liver-specific pyruvate kinase gene promoter. A unique combination of known factors

DNase-I hypersensitivity analysis of the L-type pyruvate kinase gene in rats and transgenic mice

The rat L-type pyruvate kinase gene possesses two promoters located 500 bp apart. The L' promoter is specific to erythroid cells. The L promoter is specific to liver and is regulated by diet and hormones; positively by glucose and insulin and negatively by glucagon via cAMP. The DNA elements involved in this tissue-specific and hormone-regulated gene expression are located within 3.2 kbp of 5' flanking region as previously demonstrated by transgenic mice analysis [Tremp, G. L., Boquet, D., Ripoche, M. A., Cognet, M., Yu-Chun, L., Jami, J., Kahn, A. and Daegelen, D. (1989) J. Biol. Chem. 264, 19,904-19,910]. Moreover, we have observed in these mice that gene expression was dependent on the transgene copy number and independent of the integration site. We present here DNase-I-hypersensitivity analysis of the endogenous rat L-type pyruvate kinase gene and of two transgene constructs in relation to development, tissue differentiation, nutritional and hormonal status. In rats, two groups of proximal sites were detected on the endogenous gene; hypersensitive site (HSS) HSS-1 in adult liver and HSS-A in fetal liver (a major erythropoietic tissue). Both groups are probably related to the transcriptional initiation complexes at either the L or L' promoter. Two other distal groups were detected; HSS-2 at -3 kbp (with respect to the liver-specific cap site) in adult liver and HSS-B around -4 kbp in fetal liver. These sites are thought to correspond to activating sequences; in adult liver, deletion of a fragment encompassing HSS-2 provokes a dramatic reduction of transcription starting at the L promoter of the transgene. In adult liver, HSS-1 appears to be a transcription-associated site, being greatly weakened in fasted rats, while HSS-2 is transcription independent. The pattern of DNase-I hypersensitivity is similar for the rat endogenous gene and for the complete rat transgene; the liver-specific HSS-1 and HSS-2 are present and the intensity of the sites is correlated to the number of integrated copies. Interestingly, HSS-1 is still detectable and its intensity remains proportional to the number of integrated copies in a truncated transgene with HSS-2 deletion, while this transgene is very weakly (but nevertheless tissue-specifically) expressed. These results strongly suggest that each transgene copy possesses a complete set of specific nucleoprotein complexes and that, with or without HSS-2, the DNA is in a potentially active configuration.

Functional characterization of the promoter of pp63, a gene encoding a natural inhibitor of the insulin receptor tyrosine kinase

PP63 is a liver specific phosphorylated glycoprotein encoded by a single copy gene, which has the property of inhibiting both autophosphorylation and tyrosine kinase activity of the insulin receptor. In this study, we have analyzed the structure activity relationship of the pp63 gene promoter. Five protein binding sites were found in the proximal 5' flanking region of the gene (-223 to +4). Using oligonucleotides as competitors and purified recombinant C/EBP in footprinting and gel retardation assays, we identified two typical C/EBP sites (X1 and X3) plus a heterogenous, C/EBP-NF1 like site (X5), separated by two classical NF1 binding sites (X2 and X4). C/EBP or the related proteins were predominantly involved in supporting cell-free transcription. Occupancy of the first high affinity C/EBP site conferred almost maximal promoter efficiency, in vitro. However, this pp63 promoter activity remained very low as compared to that in intact hepatocytes. In these cells, occupancy of the first C/EBP (X1) and NF1 (X2) sites was already required for achieving a weak transcriptional activity. The use of the second C/EBP site (X3) strongly enhanced transcription, up to 60-70% of the maximum, whereas occupancy of the two more distal sites (X4 and X5) was necessary to fully activate the promoter. Thus, the strength of the promoter as well as the liver specific expression of pp63 gene appear to result from the interplay of several DNA-protein complexes involving mainly C/EBP and/or related proteins as well as the ubiquitous NF1 factor(s), rather than from the interaction of a more liver specific trans-acting factor with the promoter.

Cis-regulation of the L-type pyruvate kinase gene promoter by glucose, insulin and cyclic AMP

The glucose/insulin response element of the L-pyruvate kinase gene is a perfect palindrome located from nt -168 to -144 with respect to the cap site. This element (L4) is partially homologous to MLTF binding sites. Its full efficiency requires cooperation with a contiguous binding site for HNF4, termed L3 and located from nt -145 to -125. In the presence of the L4 element contiguous to L3, cyclic AMP inhibits activity of the L-PK promoter while in its absence, or when the normal L4-L3 contiguity is modified, cyclic AMP behaves as a transcriptional activator that does not seem to be sequence-specific. Therefore, we propose that the mechanism of inhibition of the L-PK gene by cyclic AMP requires precise interactions between the nucleoprotein complex built up at sites L4 and L3 and other components of the L-PK transcription initiation complex.

A human binding site for transcription factor USF/MLTF mimics the negative regulatory element of human immunodeficiency virus type 1

Transcriptional regulation of the proviral form of the human immunodeficiency virus type 1 (HIV-1) is exerted by its 5' long terminal repeat (LTR), which contains recognition sites for several cell factors. By gel retardation and DNase I footprinting experiments we have identified a binding site for a human nuclear protein between nucleotides -152 to -174 upstream of transcription start site, in a region previously recognized as a negative regulator of transcription (negative regulatory element, NRE). The recognized sequence contains the dyad symmetry element CACGTG, which represents a binding motif, very conserved through evolution, present in a putative human DNA replication origin (pB48), in the upstream element of the major late promoter (MLP-UE) of adenovirus, and, as transcriptional element, upstream of many eukaryotic genes. Common binding activities exist in human nuclear extracts for pB48, MLP-UE and the HIV-1 LTR; at least three protein species recognize the LTR sequence, of 44 (corresponding to transcription factor USF/MLTF), 70, and 110 kDa, respectively. Chloramphenicol acetyltransferase assays suggest that the USF/MLTF binding site located in the HIV-1 LTR acts as a negative regulator of transcription, and that it contributes to the overall negative function exerted by the NRE. An oligonucleotide corresponding to another characterized human USF/MLTF binding site can functionally replace part of the activity of the NRE. This negative function is exerted both in presence or absence of tat transactivation, in different cell lines, and after PMA stimulation.

Interplay of an original combination of factors: C/EBP, NFY, HNF3, and HNF1 in the rat aldolase B gene promoter

The rat aldolase B 5' flanking region (nucleotides - 194 to +41) contains sufficient information for liver-specific expression. A detailed investigation of factors binding to the rat aldolase B 5' flanking region has allowed us to identify three distinct factors that filled different sites of this region (A, B, C). The liver-enriched C/EBP or related factors bind to box C, as demonstrated by the specific interaction with bacterially expressed C/EBP protein. Box B bearing the CCAAT sequence binds the ubiquitous factor NFY. Surprisingly, Box A is able to bind two liver enriched factors, namely HNF1 and HNF3. However, in the context of the intact promoter, as shown by footprinting competition experiments, HNF3 binds solely to this sequence. HNF3, but not HNF1 is a transcriptional activator as demonstrated in the in vitro transcription assay.

NFY or a related CCAAT binding factor can be replaced by other transcriptional activators for co-operation with HNF1 in driving the rat albumin promoter in vivo

Like many eukaryotic genes, the rat albumin promoter contains a CCAAT consensus motif at position -80. In transfected H4II hepatoma cells the strength of this promoter depends to a large extent on the integrity of a hepatic nuclear factor 1 (HNF1) binding site located at position -60 and to a lesser extent on the CCAAT element. However, if the affinity for HNF1 is reduced, the CCAAT-box becomes essential for high, and tissue specific, promoter activity. We wished to determine which, among the different CCAAT binding factors co-existing in eukaryotic cells, was responsible for this co-operativity with HNF1. To this end we prepared a series of mutants of the CCAAT sequence and compared their effects on albumin promoter activity in vivo and on the binding of different CCAAT binding factors in vitro. Our results strongly suggest that a ubiquitous factor NFY (also designated CBF, ACF, CP1) interacts with this CCAAT element in vivo. We propose that during development NFY could facilitate transcription of the albumin gene in hepatocytes when the concentration of HNF1 is limiting. This co-operativity in transcriptional activation is not due to strict co-operativity in DNA binding between the two proteins and is not limited to NFY or a closely related factor, as the CCAAT-box can be replaced by AP1, SP1 or E2 target sites without significantly affecting the final activity.

Alteration in L-type pyruvate kinase gene expression is not associated with the LF-B1 mRNA level

The relation of expression of the LF-B1 gene with the L-type pyruvate kinase (L-PK) mRNA level in rat liver and hepatoma cells was investigated. The L-PK mRNA level in rat liver changed after partial hepatectomy, during development and on intake of a high carbohydrate diet, while the level of LF-B1 mRNA remained unchanged or altered reciprocally. Dedifferentiated AH-130 cells, which did not express L-PK mRNA, expressed LF-B1 mRNA. These results suggest that transcription of the pyruvate kinase L gene is not simply regulated by the level of LF-B1 mRNA.

Liver-specific gene expression: A-activator-binding site, a promoter module present in vitellogenin and acute-phase genes

The A2 vitellogenin gene of Xenopus laevis, which is expressed liver specifically, contains an A-activator-binding site (AABS) that mediates high in vitro transcriptional activity in rat liver nuclear extracts. Footprint experiments with DNase I and gel retardation assays revealed the binding of several proteins to AABS. Using binding sites of known DNA-binding proteins as competitors in the gel retardation assay, we found that the transcription factor C/EBP and/or one of its "iso-binders" as well as LFB1/HNF1 bound AABS. These interactions were confirmed by in vitro transcription experiments using various oligonucleotides as competitors. However, saturating amounts of C/EBP- and LFB1/HNF1-binding sites as competitors only partially blocked AABS-mediated transcriptional activity. This finding implies that at least a third distinct transcription factor interacts with AABS. In vitro transcription experiments revealed that AABS was present not only in the closely related Xenopus A1 vitellogenin gene but also in acute-phase genes as a liver-specific regulatory element known to confer the interleukin-6 response. Both AABS and the interleukin-6 response element are promoter modules interacting with at least three distinct transcription factors, including C/EBP and LFB1/HNF1.

Liver-specific gene expression: A-activator-binding site, a promoter module present in vitellogenin and acute-phase genes

The A2 vitellogenin gene of Xenopus laevis, which is expressed liver specifically, contains an A-activator-binding site (AABS) that mediates high in vitro transcriptional activity in rat liver nuclear extracts. Footprint experiments with DNase I and gel retardation assays revealed the binding of several proteins to AABS. Using binding sites of known DNA-binding proteins as competitors in the gel retardation assay, we found that the transcription factor C/EBP and/or one of its "iso-binders" as well as LFB1/HNF1 bound AABS. These interactions were confirmed by in vitro transcription experiments using various oligonucleotides as competitors. However, saturating amounts of C/EBP- and LFB1/HNF1-binding sites as competitors only partially blocked AABS-mediated transcriptional activity. This finding implies that at least a third distinct transcription factor interacts with AABS. In vitro transcription experiments revealed that AABS was present not only in the closely related Xenopus A1 vitellogenin gene but also in acute-phase genes as a liver-specific regulatory element known to confer the interleukin-6 response. Both AABS and the interleukin-6 response element are promoter modules interacting with at least three distinct transcription factors, including C/EBP and LFB1/HNF1.

Positive and negative regulation of gene expression by insulin and glucagon: the model of L-type pyruvate kinase gene

L-type pyruvate kinase gene regulation is an excellent model of gene control by hormones and diet. In vivo and ex vivo experiments allowed us to established that thyroid hormones and glucocorticoids act on pyruvate kinase gene expression at the post-transcriptional level. In contrast, glucose and insulin together stimulate transcription of this gene while glucagon inhibits it. Insulin or glucose are individually inefficient and glucagon-dependent transcriptional inhibition seems to be dominant in insulin + glucose-dependent activation. A 14-kbp fragment encompassing the entire pyruvate kinase gene and 3.2-kbp of 5' flanking sequences is expressed in transgenic mice exactly like the endogenous gene; the 3.2-kbp upstream region is sufficient to confer this tissue-specific and hormone/diet-regulated expression to reporter genes. In vivo, DNAse I hypersensitivity analysis revealed the presence of 3 liver-specific groups of hypersensitive sites (HSS). The proximal sites, between + 1 and -183 bp with respect to the start site of transcription, were, in addition, transcription-dependent. The nature and functional role of proteins binding to this proximal upstream sequence were analyzed by in vitro binding and cell free transcription experiments. The existence of more upstream cis-acting elements was investigated by transient transfection assays using differentiated hepatoma cell lines and hepatocytes in primary culture. These experiments permitted the detection of an extinguisher active in hepatoma Hep G2 cells but not in hepatocytes, and of an activating element which could correspond to a distal HSS. Unfortunately, this investigation has not yet allowed us to determine with accuracy the DNA elements responsible for response to diet and hormones.

Cloning of human hepatic nuclear factor 1 (HNF1) and chromosomal localization of its gene in man and mouse

HNF1 is a transcription factor that is required for hepatocyte-specific expression of several genes, including albumin and fibrinogen. Rat HNF1-encoding cDNAs have recently been cloned, revealing that this factor is a distant member of the homeoprotein family. We have now isolated HNF1 clones from a human liver cDNA library by using a rat HNF1 cDNA-derived probe. The longest clone, HCL20, contains a sequence corresponding to the intact rat HNF1-coding region followed by a 3' nontranslated region and a poly(A) tail, hence representing an almost full-length HNF1 cDNA. Alignment of the human and rat sequences shows that HNF1 is highly conserved between the two species. The HNF1 gene was mapped by in situ hybridization and by RFLP analysis of interspecific mouse backcrosses to chromosomes 12q24.3 and 5F in human and mouse, respectively, establishing a new segmental homology between these two chromosomes.

A myosin-like dimerization helix and an extra-large homeodomain are essential elements of the tripartite DNA binding structure of LFB1

The transcription activator LFB1 is a major determinant of hepatocyte-specific expression of many genes. To study the mechanisms underlying LFB1 transcriptional selectivity, we have initiated its biochemical characterization. By in vitro complementation assays we have defined two distinct regions required for high levels of transcription, which resemble previously described activation domains. In contrast, the region of LFB1 necessary for DNA binding displays several novel features. The DNA binding domain is tripartite, including a homeodomain of unusual length (81 amino acids) and an N-terminal helix similar to part of myosin. This helical region mediates dimerization, which is shown to be essential for DNA binding.