Analysis of a brain-specific isozyme. Expression and chromatin structure of the rat aldolase C gene and transgenes

Hypoxia and oxygenation induce a metabolic switch between pentose phosphate pathway and glycolysis in glioma stem-like cells

Fluctuations in oxygen tension during tissue remodeling impose a major metabolic challenge in human tumors. Stem-like tumor cells in glioblastoma, the most common malignant brain tumor, possess extraordinary metabolic flexibility, enabling them to initiate growth even under non-permissive conditions. We identified a reciprocal metabolic switch between the pentose phosphate pathway (PPP) and glycolysis in glioblastoma stem-like (GS) cells. Expression of PPP enzymes is upregulated by acute oxygenation but downregulated by hypoxia, whereas glycolysis enzymes, particularly those of the preparatory phase, are regulated inversely. Glucose flux through the PPP is reduced under hypoxia in favor of flux through glycolysis. PPP enzyme expression is elevated in human glioblastomas compared to normal brain, especially in highly proliferative tumor regions, whereas expression of parallel preparatory phase glycolysis enzymes is reduced in glioblastomas, except for strong upregulation in severely hypoxic regions. Hypoxia stimulates GS cell migration but reduces proliferation, whereas oxygenation has opposite effects, linking the metabolic switch to the "go or grow" potential of the cells. Our findings extend Warburg's observation that tumor cells predominantly utilize glycolysis for energy production, by suggesting that PPP activity is elevated in rapidly proliferating tumor cells but suppressed by acute severe hypoxic stress, favoring glycolysis and migration to protect cells against hypoxic cell damage.

Hypoxia results in an HIF-1-dependent induction of brain-specific aldolase C in lung epithelial cells

Aldolase C (EC 4.1.2.13) is a brain-specific aldolase isoform and a putative target of the transcription factor hypoxia-inducible factor (HIF)-1. We identified aldolase C as a candidate hypoxia-regulated gene in mouse lung epithelial (MLE) cells using differential display. We show that the message accumulates in a robust fashion when MLE cells are exposed to 1% oxygen and is inversely related to oxygen content. Induction in hypoxia is dependent on protein synthesis. We localized a hypoxia-responsive element (HRE) in the aldolase C promoter using a series of deletion and heterologous expression studies. The HRE overlaps with a region of the proximal aldolase C promoter that is also related to its brain-specific expression. The HRE contains an Arnt (HIF-1beta) and an HIF-1alpha site. We show that induction in hypoxia is dependent on the HIF-1 site and that HIF-1alpha protein is present, by gel-shift assay, within nuclear complexes of MLE cells in hypoxia. Aldolase C mRNA expression is developmentally regulated in the fetal lung, rapidly downregulated in the newborn lung at birth, and inducible in the adult lung when exposed to hypoxia. This pattern of regulation is not seen in the brain. This preservation of this HRE in the promoters of four other species suggests that aldolase C may function as a stress-response gene.

Purkinje cell expression of the mouse aldolase C gene in transgenic mice is directed by an upstream regulatory element

We have sought to understand the regulation of the expression pattern of aldolase C (Zebrin II) in cerebellar Purkinje cells. Normally, aldolase C is expressed in a series of sagittal stripes of Purkinje cells interrupted by stripes of little or no expression. Genomic aldolase C:LacZ fusion genes with 1.8 kb of sequence 5' to the transcription start site drive CNS expression of LacZ only in astrocytes and cells of the pia mater. If the 5' portion of the transgene is extended to a full 5.0 kb, expression is reliably observed in Purkinje cells, yet none of the astrocyte expression is lost. We broke the additional 3.0 kb into 1.0 kb fragments and tested each for Purkinje cell enhancer activity when appended to the original 1.8 kb construct. We show that the 886 bp region from nucleotide -2796 to -3682 (relative to the start of transcription) contains virtually all of the Purkinje cell enhancer activity. However, neither the full 5.0 kb nor the 886 bp region directed a striped expression pattern, as is seen for the endogenous gene. Taken together, our study localizes a Purkinje cell enhancer to a small 5' region of the aldolase C gene and illustrates that the element(s) responsible for the normal anatomically complex pattern of aldolase C expression are separate from those conferring cell-type specificity. The relationship of these findings to previous work in other laboratories is discussed.

Diverse human aldolase C gene promoter regions are required to direct specific LacZ expression in the hippocampus and Purkinje cells of transgenic mice

Aldolase C is selectively expressed in the hippocampus and Purkinje cells in adult mammalian brain. The gene promoter regions governing cell-specific aldolase C expression are obscure. We show that aldolase C messenger expression in the hippocampus is restricted to CA3 neurons. The human distal promoter region (-200/-1200 bp) is essential for beta-galactosidase (beta-gal) expression in CA3 neurons and drives high stripe-like beta-gal expression in Purkinje cells. The 200 bp proximal promoter region is sufficient to drive low brain-specific and stripe-like beta-gal expression in Purkinje cells. Thus, the human aldolase C gene sequences studied drive endogenous-like expression in the brain.

Structure of human brain fructose 1,6-(bis)phosphate aldolase: linking isozyme structure with function

Fructose-1,6-(bis)phosphate aldolase is a ubiquitous enzyme that catalyzes the reversible aldol cleavage of fructose-1,6-(bis)phosphate and fructose 1-phosphate to dihydroxyacetone phosphate and either glyceral-dehyde-3-phosphate or glyceraldehyde, respectively. Vertebrate aldolases exist as three isozymes with different tissue distributions and kinetics: aldolase A (muscle and red blood cell), aldolase B (liver, kidney, and small intestine), and aldolase C (brain and neuronal tissue). The structures of human aldolases A and B are known and herein we report the first structure of the human aldolase C, solved by X-ray crystallography at 3.0 A resolution. Structural differences between the isozymes were expected to account for isozyme-specific activity. However, the structures of isozymes A, B, and C are the same in their overall fold and active site structure. The subtle changes observed in active site residues Arg42, Lys146, and Arg303 are insufficient to completely account for the tissue-specific isozymic differences. Consequently, the structural analysis has been extended to the isozyme-specific residues (ISRs), those residues conserved among paralogs. A complete analysis of the ISRs in the context of this structure demonstrates that in several cases an amino acid residue that is conserved among aldolase C orthologs prevents an interaction that occurs in paralogs. In addition, the structure confirms the clustering of ISRs into discrete patches on the surface and reveals the existence in aldolase C of a patch of electronegative residues localized near the C terminus. Together, these structural changes highlight the differences required for the tissue and kinetic specificity among aldolase isozymes.

Distinctive Gene Expression Profiles by cDNA Microarrays in Endometrioid and Serous Carcinomas of the Endometrium

Endometrial carcinomas are classified by their morphology into two major subtypes. Endometrioid carcinomas (type I) are generally estrogen dependent, well-differentiated, superficially invasive, and have a good outcome. Serous carcinomas (type II) are hormone independent, frequently deeply invasive and widely metastatic, and have a poor prognosis. Microarray technology and analysis allows us to determine if the global gene expression profiles of these two subtypes correlate with their morphologic phenotype. Fresh tissue from 18 endometrial carcinomas was studied: 7 well-, 2 moderately, and one poorly differentiated endometrioid, 4 serous carcinomas, and 4 high-grade mixed endometrioid-serous carcinomas. Labeled cDNA probes were synthesized (Cy5 for tumor, Cy3 for reference) and applied to microarrays containing 18,098 cDNA clones or ESTs. A pool of equal amounts of total RNA from each tumor served as the reference RNA. By unsupervised cluster analysis, the endometrioid carcinomas clustered together and were separate from the serous carcinomas. The high-grade mixed carcinomas clustered with the serous carcinomas. Using a statistical algorithm based on gene expression pattern and conducting a supervised analysis of the two defined groups, we have identified 315 genes that statistically differentiate type I from type II endometrial carcinomas. In addition to corroborating the predicted overexpression of known markers (e.g., ras and catenin in endometrioid carcinomas), the cDNA microarray technique has revealed novel alterations in gene expression relevant to cell cycle, cell adhesion, signal transduction, apoptosis, and tumor progression not previously implicated in endometrial carcinomas. For serous carcinomas, these include aldolase, desmoplakin, integrin-linked kinase, PKC, and metallopeptidase. In conclusion, the gene expression profiles of type I and type II endometrial carcinomas are different. Refinement of these profiles will permit more accurate diagnostic tumor classification and the development of prognosis assays.

Development of cerebellar modules: extrinsic control of late-phase zebrin II pattern and the exploration of rat/mouse species differences

The vertebrate cerebellum is divided into a characteristic set of 13 parasagittal "bands" or modules that are revealed in many different domains-ranging from patterns of gene and protein expression to the organization of afferent input. We have used the expression of Zebrin II/aldolase C in Purkinje cells as a marker of these bands and have discovered several new features of their regulation. We find that appearance of the banded expression of aldolase C during development differs between rat and mouse. In agreement with previous reports there is, in rat, a transient period during which all Purkinje cells are positive for aldolase C expression. By contrast, in mouse, the pattern emerges in its adult (banded) form from the earliest postnatal times. This species difference is found in both mRNA and protein expression. There also appears to be a transition that occurs in vivo between postnatal days 8 and 10. Slice cultures established from cerebella at the younger age do not develop a complete banding pattern, even after 6 days in culture. Slice cultures established from postnatal day 10 mice develop the full pattern within 2 days. This difference cannot be overcome by manipulating the levels of neuronal activity in the cultures. Thus some event must occur in vivo that "releases" the adult pattern and allows it to be realized in the more artificial situation of the slice culture. Taken together the results offer a more complete picture of the regulation of the aldolase C gene in cerebellar Purkinje cells and suggest important species differences in its developmental expression pattern.

Neuronal expression of enhanced green fluorescent protein directed by 5' flanking sequences of the rat aldolase C gene in transgenic mice

The rat aldolase C gene encodes a glycolytic enzyme strongly expressed in adult brain. We previously reported that a combination of distal and proximal 5' flanking sequences, the A + C + 0.8 kilobase (kb) pairs fragments, ensured high brain-specific expression in vivo (Skala et al. 1998). We show here that the expression pattern conferred by these sequences, when placed in front of the chloramphenicol acetyltransferase (CAT) or the enhanced green fluorescent protein (EGFP) reporter genes in transgenic mice, is similar to the distribution of the endogenous mRNA and protein. Double immunostaining for neuronal or glial cell-specific markers and for the EGFP protein indicates that the A + C + 0.8 kb genomic sequences from the rat aldolase C gene direct a predominant expression in neuronal cells of adult brain.

Novel target sequences for Pax-6 in the brain-specific activating regions of the rat aldolase C gene

Upstream activating sequences of the rat aldolase C gene are shown here to confer brain-specific expression in transgenic mice. In addition to binding sites described previously for the brain-expressed POU proteins Brn-1 and Brn-2 (Skala, H., Porteu, A., Thomas, M., Szajnert, M. F., Okazawa, H., Kahn, A., and Phan-Dinh-Tuy, F. (1998) J. Biol. Chem. 273, 31806-31814), we have identified two novel DNA elements critical for an interaction with a brain-specific, high affinity DNA-binding protein. Characterization of this binding protein showed it to be sensitive to thiol oxidation and stable to heat at 100 degrees C. This protein was purified on the basis of its thermostability and its selective adsorption to streptavidin magnetic particles via a biotinylated multimer of its target DNA binding site. Liquid chromatography coupled to tandem mass spectrometry analysis, binding competition with consensus oligonucleotides, and antibody supershift assays led to its identification as the homeodomain paired protein Pax-6. This result suggests that the brain-specific aldolase C gene could constitute a new target for the transcription factor Pax-6, which is implicated increasingly in neurogenesis.

Human aldolase C gene expression is regulated by adenosine 3',5'-cyclic monophosphate (cAMP) in PC12 cells

We have examined the effects of an adenosine 3',5'-cyclic monophosphate (cAMP) analog on human aldolase C gene expression in the rat pheochromocytoma cell line PC12. Incubation for 4 h with 500 microM 8-Br-cAMP increased aldolase C mRNA expression 2.5-fold and the expression was still above basal level 24 h later. Using transient transfection experiments we demonstrate that the distal element D in the promoter region of the human aldolase C gene, which binds a transcriptional activator (NGFI-B), is involved in this regulation. NGFI-B mRNA and protein expression were promptly (15 min) increased after 8-Br-cAMP treatment and precedes aldolase C mRNA increase (30 min). After 4 h of 8-Br-cAMP treatment, the binding of NGFI-B protein to the distal element D in the distal promoter region was increased twofold and this correlates with the increased expression of the clone that contains distal element D. These results indicate that the distal element D in the promoter region of the human aldolase C gene is the target of a cAMP-dependent regulation pathway.

Bidirectional activity and orientation-dependent specificity of the rat aldolase C promoter in transgenic mice

We previously reported that the rat aldolase C 115 bp promoter is sufficient to ensure the brain specific expression of the chloramphenicol acetyltransferase reporter gene in transgenic mice. We identify in a further reduced 84 bp promoter several putative binding sites for the transcriptional factors Sp1, USF, AP1, and AP2. Deletion or mutation of these partially overlapping binding sites results in inactivation of the cognate transgenes. Moreover, we show that the 115 bp sequence is able to direct bidirectional transcription in vivo but, surprisingly, transcriptional activity in the opposite direction is no more brain specific.

Association of 5' CpG demethylation and altered chromatin structure in the promoter region with transcriptional activation of the multidrug resistance 1 gene in human cancer cells

Selection of human cells for resistance to vincristine or doxorubicin often induces overexpression of the multidrug resistance 1 gene (MDR1), which encodes the cell surface P-glycoprotein, as a result of gene amplification or transcriptional activation. However, the precise mechanism underlying such transcriptional activation of MDR1 remains unclear. The relation between methylation status of CpG sites in the MDR1 promoter region and transcriptional activation of MDR1 has now been investigated. The P-glycoprotein-overexpressing, multidrug-resistant KB/VJ300 and KB-C1 cells, which were established from human cancer KB3-1 cells, were examined; MDR1 is transcriptionally activated but not amplified in KB/VJ300 cells, whereas it is amplified in KB-C1 cells. Determination of the methylation status revealed that the MDR1 promoter region was hypomethylated in KB/VJ300 and KB-C1 cells, but hypermethylated in KB3-1 cells. Prior treatment of KB3-1 cells with the DNA methyltransferase inhibitor 5-aza-2'-deoxycytidine resulted in a 90-fold increase in the frequency of vincristine-resistance. Of three lines, KB/CdR-1, KB/CdR-2, and KB/CdR-3, established from KB3-1 cells after exposure to 5-aza-2'-deoxycytidine, MspI/HpaII sites in the MDR1 promoter region were hypomethylated in KB/CdR-1 and KB/CdR-2 cells, but not in KB/CdR-3 cells. MDR1 mRNA expression was detected in KB/CdR-1 and KB/CdR-2 cells, but not in KB/CdR-3 cells. The binding of YB-1 and Sp1, transcription factors implicated in MDR1 expression, in the MDR1 promoter was not affected by the methylation status of a neighboring CpG sites. The MDR1 promoter region in KB/VJ300 cells showed an increased sensitivity to DNase I compared with that in KB3-1 cells, suggesting an altered chromatin structure. The methylation status of the promoter region may plays an important role in MDR1 overexpression and in acquisition of the P-glycoprotein-mediated multidrug resistance phenotype.

Hypomethylation Status of CpG Sites at the Promoter Region and Overexpression of the Human MDR1 Gene in Acute Myeloid Leukemias

Selection of human cells for resistance to vincristine or doxorubicin often induces overexpression of the multidrug resistance 1 gene (MDR1), which encodes the cell surface P-glycoprotein, as a result of gene amplification or transcriptional activation. Moreover, overexpression of the MDR1 gene has been shown to be associated closely with clinical outcome in various hematological malignancies, including acute myeloid leukemia (AML). However, the precise mechanism underlying overexpression of the MDR1 gene during acquisition of drug resistance remains unclear. We recently described an inverse correlation between the methylation status of CpG sites at the promoter region and expression of the MDR1 gene in malignant cell lines. In this study, we expanded this analysis to 42 clinical AML samples. We adapted a quantitative reverse transcription-polymerase chain reaction (RT-PCR) assay for gene expression and a quantitative PCR after digestion by Hpa II for methylation status of the MDR1gene. We observed a statistically significant inverse correlation between methylation and MDR1 expression in clinical samples. The hypomethylation status of the MDR1 promoter region might be a necessary condition for MDR1 gene overexpression and establishment of P-glycoprotein–mediated multidrug resistance in AML patients.

Hypomethylation Status of CpG Sites at the Promoter Region and Overexpression of the Human MDR1 Gene in Acute Myeloid Leukemias

Selection of human cells for resistance to vincristine or doxorubicin often induces overexpression of the multidrug resistance 1 gene (MDR1), which encodes the cell surface P-glycoprotein, as a result of gene amplification or transcriptional activation. Moreover, overexpression of the MDR1 gene has been shown to be associated closely with clinical outcome in various hematological malignancies, including acute myeloid leukemia (AML). However, the precise mechanism underlying overexpression of the MDR1 gene during acquisition of drug resistance remains unclear. We recently described an inverse correlation between the methylation status of CpG sites at the promoter region and expression of the MDR1 gene in malignant cell lines. In this study, we expanded this analysis to 42 clinical AML samples. We adapted a quantitative reverse transcription-polymerase chain reaction (RT-PCR) assay for gene expression and a quantitative PCR after digestion by Hpa II for methylation status of the MDR1gene. We observed a statistically significant inverse correlation between methylation and MDR1 expression in clinical samples. The hypomethylation status of the MDR1 promoter region might be a necessary condition for MDR1 gene overexpression and establishment of P-glycoprotein–mediated multidrug resistance in AML patients.

Upstream elements involved in vivo in activation of the brain-specific rat aldolase C gene. Role of binding sites for POU and winged helix proteins

The rat aldolase C gene encodes a glycolytic enzyme strongly expressed in adult brain. We previously reported that a 115-base pair (bp) promoter fragment was able to ensure the brain-specific expression of the chloramphenicol acetyltransferase (CAT) reporter gene in transgenic mice, but only at a low level (Thomas, M., Makeh, I., Briand, P., Kahn, A., and Skala, H. (1993) Eur. J. Biochem. 218, 143-151). Here we show that in vivo activation of this promoter at a high level requires cooperation between an upstream 0.6-kilobase pair (kb) fragment and far upstream sequences. In the 0.6-kb region, a 28-bp DNA element is shown to include overlapping in vitro binding sites for POU domain regulatory proteins and for the Winged Helix hepatocyte nuclear factor-3beta factor. An hepatocyte nuclear factor-3beta-binding site previously described in the short proximal promoter fragment is also shown to interact in vitro with POU proteins, although with a lower affinity than the 28-bp motif. Additional binding sites for POU factors were detected in the upstream 0.6-kb sequences. Progressive deletion in this region resulted in decreased expression levels of the transgenes in mice, suggesting synergistic interactions between these multiple POU-binding sites. We propose that DNA elements characterized by a dual binding specificity for both POU domain and Winged Helix transcription factors could play an essential role in the brain-specific expression of the aldolase C gene and other neuronal genes.

Genomic sequences of aldolase C (Zebrin II) direct lacZ expression exclusively in non-neuronal cells of transgenic mice

Aldolase C is regarded as the brain-specific form of fructose-1, 6-bisphosphate aldolase whereas aldolase A is regarded as muscle-specific. In situ hybridization of mouse central nervous system using isozyme-specific probes revealed that aldolase A and C are expressed in complementary cell types. With the exception of cerebellar Purkinje cells, aldolase A mRNA is found in neurons; aldolase C message is detected in astrocytes, some cells of the pia mater, and Purkinje cells. We isolated aldolase C genomic clones that span the entire protein coding region from 1.5 kb 5' to the transcription start site to 0.5 kb 3' to the end of the last exon. The bacterial gene, lacZ, was inserted in two different locations and the constructs tested in transgenic mice. When the protein coding sequences were replaced with lacZ, three of five transgenic lines expressed beta-galactosidase only in cells of the pia mater; one line also expressed in astrocyte-like cells. When lacZ was inserted into the final exon (and all structural gene sequences were retained) transgene expression was observed in astrocytes in all regions of the central nervous system as well as in pial cells. Thus, with the exception of Purkinje cell expression, the behavior of the full-length transgene mimics the endogenous aldolase C gene. The results with the shorter transgene suggest that additional enhancer elements exist within the intragenic sequences. The absence of Purkinje cell staining suggests that the cis elements required for this expression must be located outside of the sequences used in this study.

Maintenance of hypomethylation status and preferential expression of exogenous human MDR1/PGY1 gene in mouse L cells by YAC mediated transfer

Selection of cells for resistance to vincristine or doxorubicin often induces overexpression of the multidrug resistance (MDR) genes, which encode the cell surface P-glycoproteins, as a result of gene amplification, transcriptional activation, or mRNA stabilization. The LMD1 and LMD4 cell lines were established after the transfer into mouse L cells of two independent yeast artificial chromosome clones containing 300 and 850 kb, respectively, of the human MDR locus. The human MDR1/PGY1 gene, but not the endogenous mouse mdr1a and mdr1b genes, was overexpressed as a result of gene amplification and transcriptional activation in various sublines of LMD1 and LMD4 cells selected for resistance to vincristine. Then we asked why human MDR1/PGY1 gene, but not mouse relevant gene, was expressed. Determination of the methylation status of cytosine residues at Msp I/Hap II cleavage sites (CCGG) in the promoter regions of human MDR1/PGY1 and mouse mdr1a revealed hypomethylation and hypermethylation of the human and mouse genes, respectively in LMD1, LMD4, and their vincristine-resistant derivatives. Various vincristine-resistant sublines were also established after exposure of LMD1 cells for 48 h to 5-aza-2'-deoxycytidine, an inhibitor of DNA methyltransferase. These sublines exhibited overexpression of mouse mdr1a and mdr1b, but not of human MDR1/PGY1, as well as hypomethylation of the mouse mdr1a promoter region. Thus, the selective expression of human or mouse MDR genes in this cell system appears to be related to the methylation status of the respective gene promoter regions.

A (G+C)-rich motif in the aldolase C promoter functions as a constitutive transcriptional enhancer element

The enzyme fructose-1,6-bisphosphate aldolase consists of three isozymes that are expressed in a tissue-specific manner. Using antibodies against aldolase B and C, it is shown that aldolase C is expressed in virtually all neuronal cell lines derived from the central and peripheral nervous system. Recently, experiments with transgenic mice indicated that a (G+C)-rich region of the aldolase C promoter might function as a neuron-specific control element of the rat aldolase C gene [Thomas, M., Makeh, I., Briand, P., Kahn, A. & Skala, H. (1993) Eur. J. Biochem. 218, 143-151). To functionally analyse this element, a plasmid consisting of four copies of this (G+C)-rich sequence, a TATA box, and the rabbit beta-globin gene as reporter was constructed. This plasmid was transfected into neuronal and nonneuronal cell lines and transcription was monitored by RNase protection mapping of the beta-globin mRNA. It is shown that the (G+C)-rich element of the aldolase C promoter directs transcription in neuronal as well as in nonneuronal cells. In contrast, the synapsin I promoter, used as a control for neuron-specific gene expression, directed transcription only in neuronal cells. In gel-retardation assays, two major DNA-protein complexes were detected with the (G+C)-rich element of the aldolase C promoter used as a DNA probe and nuclear extracts from brain and liver as a source for DNA-binding proteins. These DNA-proteins interactions could be impaired by a DNA probe that contained an Sp1-binding site, indicating that Sp1 or an Sp1-related factor binds to the aldolase C promoter (G+C)-rich element. This was confirmed by supershift analysis with antibodies specific for Sp1. The zinc finger transcription factor zif268/egr-1, also known to recognize a (G+C)-rich consensus site, did not, however, bind to the (G+C)-rich motif of the aldolase C promoter, nor could it stimulate transcription in transactivation assays from this control region. From these data, we conclude that the (G+C)-rich element of the aldolase C promoter functions as a constitutive transcriptional response element mediated by Sp1 and Sp1-related transcription factors.

Aldolase C/zebrin II and the regionalization of the cerebellum

The cerebellum is comprised of multiple bands of cells, each with characteristic afferent and efferent projections, and patterns of gene expression. The most studied example of a striped pattern of expression is the antigen recognized by monoclonal antibody antizebrin II. Zebrin II is expressed by subsets of Purkinje cells that form an array of parasagittal bands that extend rostrocaudally throughout the cerebellar cortex, separated by similar bands of Purkinje cells that do not express zebrin II. Recent cloning studies have revealed that the zebrin II antigen is the respiratory isoenzyme aldolase C. This article reviews the cellular and molecular compartmentation of the cerebellum together with the molecular biology of the aldolase C gene, and speculates on possible reasons for a striped pattern of expression.

Functional dissection of the brain-specific rat aldolase C gene promoter in transgenic mice. Essential role of two GC-rich boxes and an HNF3 binding site

The aldolase C gene product is a glycolytic isoenzyme specifically detected in brain. We have previously defined a short 115-base pair promoter fragment able to confer on a reporter chloramphenicol acetyltransferase (CAT) gene a specific expression in brain of transgenic mice. In this promoter fragment, two GC-rich regions (A/A' and B boxes) were detected by in vitro DNase1 footprinting experiments with brain, fibroblast, or liver nuclear extracts. Both A/A' and B boxes, sharing structural homology, are able to interact with Sp1, Krox20/Krox24 factors and with other proteins (Thomas, M., Makeh, I., Briand, P., Kahn, A., and Skala, H. (1993) Eur. J. Biochem. 218, 143-151). In this paper, we describe a new ubiquitous factor termed Ub able to bind the A/A' box. We also delimit a third element (box C) binding a hepatocyte-enriched protein displaced by a hepatocyte nuclear factor 3-specific oligonucleotide. The functional involvement of each binding site in brain-specific transcription of the aldolase C gene has been tested in transgenic mice carrying different mutant promoters cloned in front of the CAT gene. A promoter containing only box C was totally inactive, suggesting an essential role of the region containing A/A' and B boxes. However, mutations or deletions of either the A/A' or the B box have no significant effect on the CAT gene expression. We therefore hypothesize that the A/A' and B sites may be functionally redundant. Indeed, constructs harboring only one of these two boxes (A/A' or B) linked to the C box displayed a brain-specific CAT activity similar to that obtained with the wild-type promoter. Furthermore, a transgene with disruption of the C box, keeping intact the A/A' and B boxes, was totally inactive, suggesting a crucial role of the hepatocyte nuclear factor 3 binding site in activation of the aldolase C gene.