Fig 2: USP15 de-ubiquitinates and stabilizes HECTD1 in LN-229The de-ubiquitination assay was performed in cells transfected with an expression plasmid for HA-Ubiquitin that 40 hours after transfection were treated for 8 hours with the proteasome inhibitor MG132 (5µM). The cells were lyzed and used for IP using an antibody against the endogenous HECTD1, or an AB against rabbit IgG in the negative control. The immunoprecipitated proteins were then analyzed by Western blot using an anti HA antibody to detect ubiquitination. (a) LN-229 cells stably expressing USP15, the mutant USP15C/S, or the empty vector control, respectively. Lanes 1 to 3 correspond to the input controls. The USP15 wt overexpressing LN-229 cells show lower ubiquitination levels of HECTD1 than the vector control. The mutant USP15-C/S overexpressing cells displayed the highest ubiquitination of HECTD1. (b) De-ubiquitination assay in LN-229 depleted for endogenous USP15 by siRNA against USP15 or the respective si-control. LN-229 cells depleted for endogenous USP15 display higher ubiquitination of HECTD1 than the sicontrol cells.

Fig 3: USP15 knockdown leads to decreased HECTD1 protein levels(a) USP15 knockdown with three siRNAs in LN-229 and (b) two siRNAs in LN-428.

Fig 4: USP15 exerts an inhibitory effect on WNT pathway activity in GBM cell lines(a) Axin2 is canonical WNT-pathway target gene and its expression was measured as read-out for WNT-pathway activity. qRT-PCR of Axin2 relative to GAPDH was peformed in LN-229 clones stably expressing USP15 (blue), the functional mutant USP15C/S (green), or the empty vector control cells (black). (b) Expression of USP15 relative to GAPDH. (c) Canonical WNT pathway activity was measured with the TCF/LEF luciferase reporter system (TOPflash/FOPflash) in LN-428 clones stably expressing USP15 (blue), USP15C/S (green), or the empty vector control. Histograms are representative of two independent experiments (data shown as mean +/-SD). In both cell lines overexpression of USP15 attenuated WNT-pathway activity, while the functional mutant USP15C/S displayed an activating effect. The experiments are normalized to the empty vector control (eV).

Fig 5: Effect of USP15 on cell growth in LN-229(a) Structure of the wildtype USP15 construct with the V5 protein tag, and the respective catalytic mutant USP15C298S. (b) Protein expression by Western blot using an anti-V5 AB, loading control tubulin for the selected clones from LN-229 and LN-428 with stable expression of USP15, or USP15C298S (USPC/S), or empty vector controls. (c) Growth of USP15, or USP15C298S (USPC/S) overexpressing clones, and respective empty vector control (pIRES2/EGFP) followed over 5 days in culture. The error bars indicate standard deviations of triplicate samples. (d) Adhesion-independent growth in soft agar, representative images, and quantification of total colony numbers. Empty vector control, eV, black; USP15, blue; USP15C/S, green. Histogram shows data as mean +/-SD.