Identification of a fat cell enhancer: analysis of requirements for adipose tissue-specific gene expression.

The molecular basis for adipose-specific gene expression is not known. To approach the problem of adipocyte gene expression, we have analyzed in detail the capacity of the 5'-flanking region of the adipocyte P2 (aP2) gene to direct cell-type specific gene expression. Although the proximal promoter containing AP-1 and C/EBP binding sites is capable of directing differentiation-dependent gene expression in cultured adipocytes, these constructs are essentially inactive in the tissues of transgenic mice. We found that -5.4 kb of the 5'-flanking region were required to direct heterologous gene (chloramphenicol acetyl transferase; CAT) expression to the adipose tissue of transgenic mice. By deletion analysis, we identified a 520 bp enhancer at -5.4 kb of the aP2 gene. We show that this enhancer can direct high levels of gene expression specifically to the adipose tissue of transgenic mice. This enhancer also functions in a differentiation-dependent manner in cultured adipocytes and cannot be transactivated in preadipocytes by C/EBP. Molecular analysis indicates that several cis- and trans- acting acting elements, though not C/EBP, contribute to the specificity and potency of this enhancer.     

Identification of multiple cis-acting elements mediating the induction of prostaglandin G/H synthase-2 by phorbol ester in murine osteoblastic cells

The tumor promoter phorbol 13-myristate 12-acetate (PMA), the best characterized protein kinase C agonist, frequently regulates gene expression via activation of Fos/Jun (AP-1) complexes. PMA rapidly and transiently induces prostaglandin G/H synthase-2 (PGHS-2) expression in murine osteoblastic MC3T3-E1 cells, but no functional AP-1 binding motifs in the 5'-flanking region have been identified. In MC3T3-E1 cells transfected with -371/+70 bp of the PGHS-2 gene fused to a luciferase reporter gene (Pluc), PMA stimulates luciferase activity up to eightfold. Computer analysis of the sequence of the PGHS-2 promoter region identified three potential AP-1 elements in the -371/+70 bp region, and deletion analysis suggested that the sequence 5'-aGAGTCA-3' at -69/-63 bp was most likely to mediate stimulation by PMA. Mutation of the putative AP-1 sequence reduces the ability of PMA to stimulate Pluc activity by 65%. On electrophoretic mobility shift analysis (EMSA), PMA induces binding to a PGHS-2 probe spanning this sequence, binding is blocked by an unlabeled AP-1 canonical sequence, and antibodies specific for c-Jun and c-Fos inhibit binding. Mutation of this AP-1 site also causes a small (22%) but significant reduction in the serum stimulation of Pluc activity in transiently transfected MC3T3-E1 cells. On EMSA, serum induces binding to a PGHS-2 probe spanning the AP-1 site, binding is blocked by an unlabeled AP-1 canonical sequence, and antibodies specific for c-Jun and c-Fos inhibit binding. Joint mutation of this AP-1 site and the nearby CRE site at -56/-52 bp, previously shown to mediate serum, v-src and PDGF induction of PGHS-2 in NIH-3T3 cells, blocks both PMA and serum induction of Pluc activity in MC3T3-E1 cells. Hence, the AP-1 and CRE binding sites are jointly but differentially involved in both the PMA and serum stimulation of PGHS-2 promoter activity.     

Identification of GATA2 and AP-1 Activator elements within the enhancer VNTR occurring in intron 5 of the human <Emphasis Type="Italic">SIRT3</Emphasis> gene

Human SIRT3 gene contains an intronic VNTR enhancer. A T > C transition occurring in the second repeat of each VNTR allele implies the presence/absence of a putative GATA binding motif. A partially overlapping AP-1 site, not affected by the transition, was also identified. Aims of the present study were: 1) to verify if GATA and AP-1 sites could bind GATA2 and c-Jun/c-Fos factors, respectively; 2) to investigate whether such sites modulate the enhancer activity of the SIRT3-VNTR alleles. DAPA assay proved that GATA2 and c-Jun/c-Fos factors are able to bind the corresponding sites. Moreover, co-transfection experiments showed that the over-expression of GATA2 and c-Jun/c-Fos factors boosts the VNTR enhancer activity in an allelic-specific way. Furthermore, we established that GATA2 and c-Jun/c-Fos act additively in modulating the SIRT3-VNTR enhancer function. Therefore, GATA2 and AP-1 are functional sites and the T S> C transition of the second VNTR repeat affects their activity.     

Identification of a decidua-specific enhancer on the human prolactin gene with two critical activator protein 1 (AP-1) binding sites

Deletion analysis of the human PRL promoter in endometrial stromal cells decidualized in vitro revealed a 536-bp enhancer located between nucleotide (nt) -2,040 to -1,505 in the 5'-flanking region. The 536-bp enhancer fragment ligated into a thymidine kinase (TK) promoter-luciferase reporter plasmid conferred enhancer activity in decidual-type cells but not nondecidual cells. DNase I footprint analysis of decidualized endometrial stromal cells revealed three protected regions, FP1-FP3. Transfection of overlapping 100-bp fragments of the 536-bp enhancer indicated that FP1 and FP3 each conferred enhancer activity. Gel shift assays indicated that both FP1 and FP3 bind activator protein 1 (AP-1), and JunD and Fra-2 are components of the AP-1 complex in decidual fibroblasts. Mutation of the AP-1 binding site in either FP1 or FP3 decreased enhancer activity by approximately 50%, while mutation of both sites almost completely abolished activity. Coexpression of the 536-bp enhancer and A-fos, a dominant negative to AP-1, decreased enhancer activity by approximately 70%. Conversely, coexpression of Fra-2 in combination with JunD or c-Jun and p300 increased enhancer activity 6- to 10-fold. Introduction of JunD and Fra-2 into nondecidual cells is sufficient to confer enhancer activity. JunD and Fra-2 protein expression was markedly increased in secretory phase endometrium and decidua of early pregnancy (high PRL content) compared with proliferative phase endometrium (no PRL). These investigations indicate that the 5'-flanking region of the human PRL gene contains a decidua-specific enhancer between nt -2,040/-1,505 and AP-1 binding sites within this enhancer region are critical for activity.