Pas de résumé / The aim of the study was to evaluate in controlled retinal risk LED lighting conditions compared to other common household lighting such as CCFL or CFL, in standard lighting conditions, on different animal species pigmented and non-pigmented. At first, we characterized the conditions of adaptation before enlightenment. We found that housing conditions before the light exposure is a source of artifact. Indeed, the location of the cages in a ventilated cabinet and the period before stalling before illumination generates a variable response to retinal light toxicity. Then, we tested the relative sensitivity of the albino strains and pigmented. After 3 weeks of stabulation, the animals were exposed for 24 hours to cold white LED at a luminance of 6000 lux, with dilation of pupils, and the retinas were examined in a week. In these extreme illumination conditions, retinas showed a significant loss of photoreceptors in superior retina, not only in albino animals, but also in pigmented animals. In another experiment, we tested different luminance in cages provided for this purpose. We have used as control a compact fluorescent lighting at 500 lux, with a homogeneous on the floor of the cage. An illumination of 24 hours dilation was performed after the time of dark adaptation. A luminance of 500 lux is a classic condition of a good visual quality domestic lighting. Unlike a compact fluorescent lighting at 500 lux, white LEDs result in a significant loss of photoreceptor nuclei of retinal pigmented rats (LE) 500 lux with an increasing toxicity in function of the luminance of the LED lighting. Finally, to assess the effects prolonged exposure we exposed the rats for one week or one month, but in alternating illumination only during the day, and without dilated pupils (day / night cycle 12h / 12h, no dilated pupils) with LED different spectra. We have compared these lighting conditions to a compact fluorescent lighting at 500 lux and non-illuminated rats. After one week, only albino rats showed a loss of photoreceptors and only after exposure to blue LEDs. These results show that the blue LEDs are more toxic than the white LEDs confirming the effects of short wavelengths. After 1 month of illumination, a significant loss of photoreceptors is observed in the retinas of non-pigmented rats, not only with the blue LEDs, but also with green LEDs and cool white LEDs. An increase of the exposure time under standard conditions leads to a loss of photoreceptors accumulated suggesting a potentially toxic effect of LED light, not observed with a compact fluorescent lighting even luminance.

L’épissage en cis des précurseurs d’ARN messager (pre-ARNm) est une stratégie intéressante afin de réparer des gènes dont la régulation transcriptionnelle est déterminante pour la fonction de la protéine. Les mutations touchant le gène dysferline (DYSF) sont liées au développement de dystrophies musculaires: la dystrophie musculaire des ceintures de type 2B et la myopathie distale de Miyoshi. Une stratégie à modifier l’épissage en cis des pre-ARNm du gène DYSF est le procédé SMaRT (pour spliceosome-mediated mRNA trans-splicing), une technique de réparation de l'ARN messager au moyen d'un complexe de trans-épissage appelé PTM (pour pre-mRNA trans-splicing molecule). Le procédé SMaRT utilisant le complexe PTM permet le remplacement d’importantes portions de pre-ARNm tout en préservant l’intégrité totale du transcrit. Dans un soucis d’obtenir un trans-epissage efficace, seuls les introns codant pour les pre-ARNm de DYSF présentant de forts signaux d’épissage répartis de façon disparate ainsi que des tailles très différentes furent ciblées par les PTMs dans des myoblasts humains ne possédant pas de dysferlin. Le trans-épissage de deux introns ciblés du gène DYSF engendra une formation correcte de la protéine dysferlin dans des mutants DYSF-/- de souris. Les niveaux de protéine fonctionnelle furent toutefois modérés, mais similaires aux taux de récupération obtenus par des stratégies précédentes de trans-épissage ciblant d’autres gènes. Néanmoins, parmi les introns ciblés avec succès dans cette étude et dans des essais précédents, des critères concordants ont pu être identifiés afin de faciliter le choix des introns à cibler pour de futures stratégies de trans-épissage. / RNA-based therapy is an approach to cure genetic disorders with no intervention into endogenous spatiotemporal gene regulation. I established two approaches for the dysferlin gene, (i) spliceosome-mediated pre-mRNA trans-splicing (SmaRT) and (ii) exon-skipping, in order to rescue dysferlin mutations leading to Limb Girdle Muscular Dystrophy 2B and Miyoshi Myopathy. SmaRT permits the correction of numerous mutations of a gene by a single pre-mRNA trans-splicing molecule (PTM) by exchanging multiple exons of a gene for a healthy mRNA sequence. The PTM binds to intronic sequence and competes with the endogenous pre-mRNA for the binding of the spliceosome. I designed PTMs to exchange the 3’ part of the dysferlin messenger and determined two functioning PTMs bytransduction of human myoblasts and intramuscular injection in wild-type and DYSF-/- mice and could show dysferlin protein rescue in DYSF-/- mice.By exon-skipping exons carrying mutations can be excised from pre-mRNA in masking exon or intron internal sequences defining the exon in the splicing process. I employed antisense oligonucleotides (AONs) of tricyclo-DNA in order to excise dysferlin exon 32. It was shown to be particularly feasible for systemic application, making it suitable for diseases affecting different compartments of skeletal muscle and other organs. The dysferlin exon 32 has been shown to be dispensable for known functions of the dysferlin protein. I designed tc-DNA AONs leading to efficient skipping in patient myoblasts and in wild-type mice following intramuscular injection. I am collaborating to investigate effects of exon 32 skipping in an Exon-32-STOP mouse model.