cal activation of antimiRs could overcome this obstacle. To that end, we were able to synthesize light-inducible antimiRs by modifying oligonucleotides with photolabile protecting groups, so called "cages". We demonstrated that light-induced activation of caged antimiR-92a significantly improved angiogenic sprouting in vitro, consistent with previous studies in which conventional antimiRs were used. In further studies we were able to extend the use of caged antimiRs to an in vivo setting.

more »

... ermal injection of caged antimiR-92a with subsequent light irradiation in murine skin led to a marked down-regulation of miR-92a expression that was as efficient as non-caged antimiR-92a application (caged antimiR-92a with light activation: −98±1%, p<0.003; antimiR-92a: −99±0.1%, p<0.008), while treatment with caged antimiR-92a without light activation had no effect (caged antimiR-92a: +20±17%, p=0.5). However, in contrast to conventional constitutively active antimiR-92a, the local activation of caged antimiR-92a did not inhibit the expression of miR-92a for example in liver tissue (caged antimiR-92a with light activation: +37±31%, antimiR-92a: −40±14%). Since quality and rapidity of tissue repair is dependent on the angiogenic response, wound healing is a suitable model to study the therapeutic potential of the pro-angiogenic antimiR-92a. Indeed, we were able to show, that administration of caged antimiR-92a led to improved tissue repair in healing impaired db/db mice after light activation. Caged antimiR-92a treated wounds reveal an accelerated healing kinetics (caged antimiR-92a with light induction: −30±13% wound size 11 days post injury, p<0.05), accompanied by a dense and cell-rich granulation tissue in comparison to non-irradiated antimiR-92a (caged antimiR-92a with light induction: +71±36%, p<0.01). Mechanistically, we were able to show that improved wound healing upon caged antimiR-92a treatment was based on increased capillary formation (caged antimiR-92a with light induction: +74±28%, p<0.05) and a significant derepression of the miR-92a target Sirt1 in comparison to controls (caged antimiR-92a with light induction: +56±14%, p<0.01). In conclusion, these data provide first in vivo evidence and proof-of-concept to demonstrate the feasibility and efficiency of light-induced activation of antimiRs. Local activation of antimiRs allows for a tissue-specific targeting of antimiRs reducing putative adverse effects of antimiRs in other tissues. This strategy may be used for therapeutic targeting of miRs in cardiac or vascular disease.