In vitro and in vivo effects of a multimerized alphas 1-casein enhancer on whey acidic protein gene promoter activity.

Experimental data obtained in previous works have led to postulate that enhancers increase the frequency of action of a linked promoter in a given cell and may have some insulating effects. The multimerized rabbit alpha s1-casein gene enhancer, the 6i multimer, was added upstream of the rabbit whey acidic protein gene (WAP) promoter (-6,300; +28 bp) fused to the firefly luciferase (luc) gene (6i WAP-luc construct). The 6i multimer increased reporter gene expression in mouse mammary HC11 cells. In transgenic mice, a very weak but significant increase was also observed. More noticeable, no silent lines were found when the 6i multimer was associated to the WAP-luc construct. This reflects the fact that the 6i multimer tends to prevent the silencing of the WAP-luc construct. After addition of the 5'HS4 insulator region from the chicken beta-globin locus upstream of the 6i multimer, similar luciferase levels were measured in 6i WAP-luc and 5'HS4 WAP-luc transgenic mice. Our present data and previous ones, which show that the 6i multimer has no insulating activity on a TK gene promoter construct indicate that the insulating activity of the 6i multimer is construct-dependent and not amplified by the 5'HS4 insulator.

Substitution of the alpha-lactalbumin transcription unit by a CAT cDNA within a BAC clone silenced the locus in transgenic mice without affecting the physically linked Cyclin T1 gene.

We recently reported that a goat bacterial artificial chromosome (BAC) clone conferred site-independent expression in transgenic mice of the two loci present within its insert, the ubiquitously expressed Cyclin T1 and the mammary specific alpha-lactalbumin (alphalac) genes. To assess if this vector could target mammary-restricted expression of cDNA, the CAT ORF was introduced by homologous recombination in Escherichia coli in place of the alphalac transcription unit. The insert of this modified BAC was injected into mice and three transgenic lines were derived. None of these lines expressed the CAT gene suggesting that the use of long genomic inserts is not sufficient to support the expression of intron-less transgenes. The physically linked goat Cyclin T1 locus was found to be active in all three lines. This observation reinforced the hypothesis that the two loci are localised in two separate chromatin domains.

The insulator effect of the 5'HS4 region from the beta-globin chicken locus on the rabbit WAP gene promoter activity in transgenic mice.

Previous studies have shown that the 5'HS4 DNaseI hypersensitive site of the chicken beta-globin locus is endowed with classic insulator activities: (i) it blocks the interaction between promoter and enhancers when it is inserted between them (ii) it confers expression of integrated foreign genes independent of their position in the chromatin. The aim of this present work was to determine whether the 5'HS4 element was able to stimulate the expression level and/or to increase the expression frequency of a luc+ reporter gene controlled by the rabbit WAP gene promoter. Two constructs with 5'HS4 insulator (p5'HS4-WAPluc) or without (pWAPluc) were introduced in mouse fertilised oocytes. All transgenic lines containing the 5'HS4 element (six lines) expressed the transgene whereas only two out of eight lines harbouring the pWAP-luc construct expressed the transgene to a significant level. Moreover, the mean level of expression was seven times higher in p5'HS4WAP-luc lines than in pWAP-luc lines. Even all these benefits on transgene expression, the 5'HS4 element did not confer a copy-dependent expression, did not decrease the ectopic expression of the reporter gene and did not decrease the variability of expression. Thus, the 5'HS4 element does not have all the properties of a perfect insulator on a construct containing the luc+ reporter gene controlled by the rabbit WAP promoter.

Position-independent and tissue-specific expression of porcine whey acidic protein gene from a bacterial artificial chromosome in transgenic mice.

Silencing of transgenes is a frequent event after the random integration of foreign DNA in the host genome following microinjection. Long genomic fragments are expected to contain all the regulatory elements necessary to induce an appropriate expression of transgenes. A bacterial artificial chromosome containing the porcine wap gene with approximately 145 and 5 kb of 5'- and 3'-flanking sequences, respectively, was microinjected into fertilized mouse ovocytes. In the six transgenic lines studied, expression was strictly specific to the mammary gland of lactating animals and was position-independent. Levels of exogenous porcine wap mRNA per copy compared favorably with the porcine wap mRNA yield in the mammary gland of a 9-day lactating pig. These findings suggest that this insert contained most if not all of the cis-acting elements involved in the full specific expression of the porcine wap gene. These elements constitute good candidates for directing the optimized expression of protein recombinant-encoding genes in the mammary gland of lactating animals.

Effect of the rabbit alphas1-casein gene distal enhancer on the expression of a reporter gene in vitro and in vivo.

Several gene constructs containing the firefly luciferase gene and the herpes simplex virus thymidine kinase gene promoter (TK) were used to evaluate the transcriptional activity of the distal enhancer (-3442, -3285) of the rabbit alphas1-casein gene. Six copies of the enhancer (6i) were added upstream of the TK-luciferase construct in the presence or absence of the chicken beta-globin 5'HS4 insulator. The activity of the constructs was tested by transient transfection in CHO cells and in rabbit primary mammary cell cultured on plastic or on floating collagen. Constructs were also tested in stably transfected mouse mammary HC11 cells. In all cell types the multimerized alphas1-casein enhancer strongly stimulated luciferase gene expression in the presence of lactogenic hormones. It was also sensitive to the extracellular matrix in rabbit primary mammary cells. The constructs were used to generate transgenic mice. The 6i TK transgenic animals expressed the luciferase gene at very low levels irrespectively of the physiological state. No preferential expression in the mammary gland was observed. Addition of 5'HS4 insulator to the 6i TK construct did not prevent silencing in most of the transgenic lines. However, two lines expressed high luciferase levels specifically in the mammary gland. Our data suggest that 6i may confer, when insulated properly, a higher and mammary-specific expression to the TK promoter.