Switch from Cap- to Factorless IRES-Dependent 0 and +1 Frame Translation during Cellular Stress and Dicistrovirus Infection

Wang, Qing S.; Jan, Eric

0 frame FLuc activity mediated by the IAPV [31] and CrPV IGR IRESs was compared to an empty bicistronic reporter construct ( Figure 3A ). As expected, mutations that inhibit IGR IRES translation, such as disrupting PKI basepairing (DPKI) ( Figure S1 , A) or both PKI and PKIII basepairing (DPKI/DPKIII (CAC6148-50GUG) in CrPV IGR IRES) resulted in significantly lower FLuc activity ( Figure 3A ). In in vitro translation experiments, these mutant IRESs are inactive [12] , however, in Drosophila cells using our transfection protocol, we observed ,20% residual translation ( Figure 3A ). Investigations into this phenomenon are ongoing but it is possible that these mutant IRESs may still adopt a core structure that can still drive residual IRES translation in vivo. It is most probable that the bulk luciferase activity detected is IRES-dependent. It has been proposed that the adjacent sequence downstream of IGR IRES is unstructured to allow the IRES to adopt its conformation for optimal translation [33] . To confirm whether sequences adjacent to the CrPV IGR IRES can affect IRES activity as shown in IAPV IGR IRES [25, 31] , we compared translational activities using a reporter RNA containing either the minimal core CrPV IGR IRES (nucleotides 6025-6231) or the IRES with sequences adjacent to the core IRES (nucleotides 5974-6372) ( Figure S2 , A). The FLuc enzymatic activity was not affected when the FLuc ORF was fused with the sequences adjacent to the core IRES ( Figure S2, A) . The CrPV IRES containing regions adjacent to the core IRES was 5 fold more active than the minimal core IRES in vivo and ,2 fold in vitro ( Figure S2, A and B) .