Small interfering RNAs (siRNAs) are still rarely used in transgenic vertebrates to knockdown gene expression. Indeed, the in vitro highly efficient expression vectors for siRNA based on the use of polymerase III promoters as U6 or H1 gene promoters for expression in cells are often silenced in vivo. In this thesis, various shRNA expressing vectors have been used in cultured cells and in transgenic mice to inhibit the IE (Immediate Early gene) mRNA from the pig pseudo-rabies virus responsible for Aujeszky disease. The amount and the sequence of produced siRNA were studied by qPCR. In transiently transfected CHO cells, vectors containing the U6 gene promoter (U6-shRNA) were by far the most efficient to inhibit IE mRNA, mainly due to the very high level of produced siRNA. Besides, two constructs encompassing a RNA polymerase II promoter (EF-1α) upstream of either the miR30 skeleton containing IE sequence or the IE sequence bordered by 5T induced a significant but moderate inhibition. This could be related to the low siRNA production by these two constructs. Notably, despite the fact that the sequence of the siRNA was shorter than expected with the miR30 construct, this did not affect its efficiency. The three constructs were then used to generate transgenic mice. Transgenic mice harbouring an intact transgene were obtained only with the miR30 construct. The absence of U6-shRNA or 5T transgenic mice could result from a potent off-targeting effect of the siRNA sequence produced with these two constructs that was not the case with the miR30 construct producing shorter siRNA. These transgenic mice were subjected to viral infection. Although the level of miRNA production was low within the cells and tissues of these transgenic animals some of these mice were capable to resist the viral infection by Aujeszky virus. The second part of this work was to select new target sequence within the different regions on the IE mRNA sequence. Several constructs were evaluated among which two sequences were capable to induce inhibition of reporter gene expression. These two new sequences which are targeted the 5'UTR region of IE mRNA and the 3'UTR region of IE mRNA are currently under further investigation for preparation of transgenic mice. The data suggest that (1) the efficiency of a siRNA producing construct is determined by its sequence in an only predictable manner (2) the efficiency of a siRNA is higher when the targeted region of the mRNA is not in secondary structure (3) a potent and optimal siRNA effect may occur even at low concentration of siRNA when it is well targeted on mRNA sequence and to the amount of the siRNA, (4) off-targeting activity of siRNA may occur even at low concentration but it may be increased at high concentration (5) a siRNA to be used in transgenic animals should be designed in such a way to be potent even at low concentration to reduce the side effects.
