When either of these two factors were depleted by RNAi, otherwise NMD-resistant b-globin mRNAs with AUGproximal PTCs became susceptible to NMD. To test whether NMD inhibition induced by the tethered eIF4GI core domain requires the presence of eIF3f or eIF3h, we conducted the tethering assay in cells depleted for each of these factors. To this end, a plasmid expressing an shRNA either against eIF3f or eIF3h was co-transfected along with plasmids Etofibrate encoding the eIF4GI-MS2 fusion protein and the minim ter310 reporter construct A. Due to the lack of specific antibodies, the efficacy of the eIF3f and eIF3h knockdowns was assessed by measuring the corresponding mRNA levels. As a control, an shRNA that does not target any known human mRNA was expressed. Depleting either of the two eIF3 subunits did not affect the capacity of PABPC1-MS2 to suppress NMD. However, the reporter mRNA increase induced by tethering of the eIF4GI core domain was reduced by approximately 40�C50% when eIF3f or eIF3h was depleted as compared to the Homatropine Bromide control knockdown. Although the effect is moderate, this result indicates that eIF3 is involved in the PABPC1-independent mechanism by which eIF4GI antagonizes NMD, but does not play a role in the PABPC1-dependent pathway. Notably, our result is reminiscent of the results obtained by Peixeiro and colleagues. It should be noted that the reporter mRNA levels were much higher when PABPC1 was tethered than when the eIF4GI core domain was tethered. This is in contrast to the results of the standard tethering assays, where these two fusion proteins had a similar effect on the reporter. One difference is that in the knockdown experiments, the time window for the expression of the fusion protein and the reporter transcript is extended. It is therefore possible that the reporter mRNA reaches its steady-state level earlier when the core domain is tethered than when PABPC1 is tethered. In vitro experiments have previously shown that subunit e of eIF3 interacts directly with the eIF4GI core domain. The same study however also suggests a close association of eIF3f and eIF3h with eIF4GI.