Noxa protein induction is necessary for cell death to occur following treatment with some cytotoxic cancer drugs, we set out to investigate if Noxa is regulated by microRNAs. Any given gene is generally predicted to be regulated by many different microRNAs. One major obstacle in microRNA research is that the numerous bioinformatic tools available for target prediction invariably give a large set of false positive results. Therefore, we made use of a luciferasebased screening method to pick out the most relevant microRNAs that target Noxa. Cloning the 39UTR of Noxa downstream of a luciferase reporter and introducing this construct into cells allowed us to determine to what degree the reporter activity is repressed in different tissues. This analysis was then complemented with luciferase experiments using deletion constructs that pinpointed the critical regulatory part of the 39UTR. Finally, the combined results were then compared with existing microRNA expression profiling data to identify candidate microRNA that might account for the differential luciferase activity. Using this screening system we identified miR-200c as a new regulator of Noxa. MiR-200c was shown to repress both basal and stressBYL719 induced Noxa protein expression. Surprisingly, enforced miR200c expression at the same time led to increased bortezomibinduced apoptosis. This apparent discrepancy was reconciled by the finding that in cells devoid of Noxa expression, miR-200c caused an even greater increase in apoptosis. These data suggest that miR-200c potentiates apoptosis induced by proteasomal inhibitors but that it concomitantly represses Noxa which leads to an attenuated apoptotic induction. The data in this study define miR-200c as a novel regulator of Noxa and more generally show that microRNA-induced phenotypes must always be viewed as the complex results of a large number of occurring individual microRNA:mRNA target interactions. This clinically used drug was chosen since it has been shown that Noxa induction is important for bortezomib-induced cell death. Treatment of HCT116 cells with clinically relevant doses of bortezomib led to a time- and dose-dependent induction of Noxa protein. As can be seen in Figure 5A, overexpression of miR-200c in HCT116 cells treated with bortezomib led to a downregulation of Noxa at all doses. Surprisingly, at the same time miR-200c overexpression resulted in increased bortezomib-induced apoptosis as assessed by immunoblotting for cleaved caspase 3 and cleaved PARP. In order to directly test how apoptosis induction is affected by miR-200c overexpression, Annexin V/PI staining was performed on HCT116 left untreated or treated with bortezomib. Again, in both cases miR-200c overexpression led to increased cell death, as compared to a scrambled pre-miR control oligonucleotide. A similar result was obtained in the 
HEK293 cell line. Also, this effect was not restricted to proteasome inhibition, as cells treated with the DNAdamaging drug doxorubicin showed increased apoptosis induction upon miR-200c overexpression as well. Since the effects of miR-200c on Noxa and cell death induced by bortezomib apparently contradict one another, we went on to examine the effect of miR-200c on apoptosis in a setting without Noxa expression. Therefore, we BEZ235 moa knocked down Noxa expression in bortezomib-treated HCT116 cells using siRNA oligos. Knockdown of Noxa led to an expected decrease in both Noxa protein levels and proteasome inhibitor-induced apoptosis as measured by Annexin V/PI staining. Interestingly, when Noxa was knocked down, miR-200c overexpression had an even more pronounced effect on apoptosis induction. Indeed, in cells transfected with control siRNA oligos.