Figure 3. FOXP3 is a direct target of miR15a/16.

(a) Sequences and introduced mutations of the miR15a and miR16 binding sites. The putative target sites for the miR15a seed sequence were identified by using TargetScan²⁵ and microRNA.org²⁶ software. The interaction between miR15a/16 and FOXP3 was predicted by three algorithms: miranda, PITA and RNA22. Annotation of the FOXP3 3′-UTR showing predicted binding sites of miR15a and miR16, their seed sequences and subsequent introduced mutations. Two sites were identified: binding site 1: 5′-GTGGTTCTAGACACCCCCTCCCCCATCATA-3′ (forward) and 5′-GTGGTTCTAGAGGCTCTCTGTGTTTTGGGGT-3′ (reverse) and; binding site 2: 5′-GTGGTTCTAGACCTACACAGAAGCAGCGTCA-3′ (forward) and 5′-GTGGTTCTAGAGATCAGGGCTCAGGGAATGG-3′ (reverse). Using a Quik-Change site-directed mutagenesis was identified in the following primers: binding site 1, 5′-CACAGAGCCTGCCTCACGCACAGATTACTTC-3′ (forward) and 5′-GAAGTAATCTGTGCGTGAGGCAGGCTCTGTG-3′ (reverse); and binding site 2, 5′-GAGGTCCCAACACGTCACACACACGGCCTG-3′ (forward) and 5′-CAGGCCGTGTGTGTGACGTGTTGGGACCTC-3′ (reverse). (b) miR15a/16 directly interact with Foxp3 at binding site 1 and binding site 2. In luciferase-based assay, miR15a/16 showed direct binding interaction with FOXP3 at binding site 1 and binding site 2. When these sites are specifically mutated, mir15a/16 lose their ability to bind to FOXP3. (c) ‘Toggle-switch’ role of miR15a/16 and FOXP3 function. Overexpression of miR15a/16 in Tregs can push their functional differentiation into a more conventional T-cell-like activity. On the other hand, silencing of miR15a/16 pathway in Tcon may lead to an inducible Treg-like function.