Transgenic mice expressing a human apolipoprotein[a] allele.

The most important determinant of plasma levels of Lp[a] are sequence differences at the highly polymorphic apolipoprotein[a] (apo[a]) locus. To define the sequences that mediate the regulation of apo[a] expression, we cloned a 370 kb DNA fragment that included a 130 kb apo[a] gene, and 40 kb 5'- and 200 kb 3'-flanking region from an individual with high plasma levels of Lp[a] using a YAC vector. This genomic clone was used to generate transgenic mice. In the YAC-apo[a] transgenic mouse, apo[a] was only expressed in the liver, as it is in humans. The mean serum level of apo[a] in 4-week-old YAC-apo[a] transgenic mice was 20 mg/dl. In the female mice the levels of apo[a] varied over a 1.5-fold range during the 4-day estrus cycle and the levels correlated directly with serum progesterone levels. The serum levels of apo[a] decreased to almost undetectable level in male mice after puberty and this decrease was reversed by castration. Ingestion of a high-fat diet resulted in a approximately 100-fold fall in hepatic apo[a] mRNA levels and >60-fold decrease in serum apo[a] levels. To delimit the control elements that mediate tissue-specific and sex hormone-responsive apo[a] transcription, we derived a reporter YAC in which 40 kb of 5' flanking sequences from the cloned apo[a] allele were linked to a luciferase reporter gene. Analysis of four independent transgenic lines revealed no hepatic luciferase expression, suggesting that important cis -acting elements located outside the apo[a] 5'-flanking region are necessary for in vivo expression of apo[a].

Identification of vascular endothelial growth factor (VEGF) receptor-2 (Flk-1) promoter/enhancer sequences sufficient for angioblast and endothelial cell-specific transcription in transgenic mice.

The vascular endothelial growth factor (VEGF) receptor-2 (Flk-1) is the first endothelial receptor tyrosine kinase to be expressed in angioblast precursors, and its function is essential for the differentiation of endothelial cells and hematopoietic precursors. We have identified cis-acting regulatory elements of the murine Flk-1 gene that mediate endothelium-specific expression of a LacZ reporter gene in transgenic mice. Sequences within the 5'-flanking region of the Flk-1 gene, in combination with sequences located in the first intron, specifically targeted transgene expression to angioblasts and endothelial cells of transgenic mice. The intronic regulatory sequences functioned as an autonomous endothelium-specific enhancer. Sequences of the 5'-flanking region contributed to a strong, uniform, and reproducible transgene expression and were stimulated by the transcription factor HIF-2alpha. The Flk-1 gene regulatory elements described in this study should allow the elucidation of the molecular mechanisms involved in endothelial cell differentiation and angiogenesis.

Differential downregulation of vascular endothelial growth factor by dexamethasone in normoxic and hypoxic rat glioma cells.

Vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) is a mitogen and chemotactic factor for endothelial cells in vitro and an angiogenesis and vascular permeability factor in vivo. Due to its properties, VEGF is a candidate for both angiogenesis and vascular permeability/oedema induction which typically occur in glioblastomas. In this study we test the hypothesis that the antioedema effect of dexamethasone is mediated by downregulation of VEGF or VEGF receptor expression. VEGF mRNA and protein levels of two rat glioma cells lines, C6 and GS-9L, were determined after incubation with dexamethasone under normoxic and hypoxic conditions. In normoxic C6 and GS9L cells, we observed 50-60% downregulation of VEGF mRNA by dexamethasone (P=0.015 and P=0. 01, respectively). This effect was dependent on glucocorticoid-receptor (GR) function. The inhibitory effect of dexamethasone on VEGF gene expression by tumour cells was markedly reduced by hypoxia which suggests that the upregulation of VEGF driven by hypoxia overcomes the effect of the dexamethasone. Dexamethasone did not alter VEGFR-2 mRNA levels in human umbilical endothelial cells. In a subcutaneous glioma tumour model, we observed only a 15% decrease in VEGF mRNA expression in dexamethasone treated animals (n = 12) compared with controls animals (P = 0.24). We conclude that dexamethasone may decrease brain tumour-associated oedema by reduction of VEGF expression in tumour cells. However, the highly reduced activity on hypoxic tumour cells suggests that dexamethasone efficacy may be limited by hypoxia in rapidly growing tumours.

Reporter gene analysis of four DNaseI hypersensitive sites in the plasminogen/apolipoprotein(a) intergenic region.

We have previously described four DNaseI hypersensitive sites (DH 1 to DH4) in the 40-kb intergenic region between the plasminogen gene and the apo(a) gene. Here, we wanted to analyse whether any of these sites, located 4, 21, 28 and 34 kb upstream of the apo(a) transcriptional start site, would act as an enhancer on a minimal apo(a) promoter. Starting from a cloned, highly expressed apo(a) allele, we obtained four fragments comprising the DHI to DH4 sites, respectively. These fragments were cloned in both orientations into a luciferase reporter gene plasmid comprising a minimal apo(a) promoter (-100 to +141 with respect to the transcriptional start site). Our results from transfection studies with the resulting series of reporter gene plasmids into liver (HepG2) and non-liver (HeLa) cells suggest that the four DH sites from the selected apo(a) allele do not provide a strong, liver-specific enhancer activity.

Characterization of the endothelium-specific murine vascular endothelial growth factor receptor-2 (Flk-1) promoter.

Flk-1, a high-affinity signaling receptor for vascular endothelial growth factor (VEGF), is strongly and specifically expressed on endothelial cells during embryonic development of the vascular system and during tumor angiogenesis. Disruption of Flk-1 gene function has recently been shown to prevent completely endothelial cell differentiation during murine embryonic development. To gain insights into the mechanisms that regulate the endothelium-specific Flk-1 expression, we have isolated the 5'-flanking region of the murine Flk-1 gene. RNase protection and primer extension analyses revealed a single transcriptional start site located 299 bp upstream from the translational start site in an initiator-like pyrimidine-rich sequence. The 5'-flanking region is rich in GC residues and lacks a typical TATA or CAAT box. A luciferase reporter construct containing a fragment from nucleotides -1900 to +299 showed strong endothelium-specific activity in transfected bovine aortic endothelial cells. Deletion analyses revealed that endothelium-specific Flk-1 expression is stimulated by the 5'-untranslated region of the first exon, which contains an activating element between nucleotides +137 and +299. In addition, two endothelium-specific negative regulatory elements were identified between nucleotides -4100 and -623. Two strong general activating elements were present in the region between nucleotides -96 and -37, which contains one potential NF kappa B and three potential AP-2 binding sites. This study shows that Flk-1 expression in endothelial cells is mainly regulated by an endothelium-specific activating element in the long 5'-untranslated region of the first exon and by negative regulatory elements located further upstream.

Alternative splicing of human cyclin E.

Cyclin E is a regulatory subunit of the cdc2-related protein kinase cdk2, which is activated shortly before S-phase entry, thus defining it as a G1 cyclin. We report here the existence of a 43 kDa splice variant of human cyclin E, termed cyclin Es, which lacks 49 amino acids within the cyclin box compared to the known 48 kDa cyclin E. Cyclin Es is expressed at approximately 1/10 of the level of full-length cyclin E in several cell lines analysed. The two cyclin E forms differ functionally in that cyclin E, but not cyclin Es, is able to complex with cdk2, to activate the histone H1, pRb and p107 in vitro kinase activity of cdk2 and to rescue a triple CLN mutation in S. cerevisiae. Cyclin Es is the first splice variant of a cell cycle regulatory protein to be described. Our findings also indicate that the cyclin box in cyclin E mediates the interaction with cdk2.