Fig 2: Cas9 enzyme is ubiquitylated and targeted for proteasomal degradation upon expression in human cells.(A) Immunoprecipitation of Cas9 expressed in HEK293 cells reveals Cas9 modification by ubiquitin. HEK293 cells were transfected with doxycycline (Dox)-inducible FLAG-tagged Cas9 along with His-tagged ubiquitin (His–Ub). Pull-down of FLAG-Cas9 was followed by Western blot analysis with a ubiquitin antibody, which yielded a smear pattern corresponding to ubiquitylated Cas9 (Cas9–Ub, highlighted by a bracket). (B) Verification of Cas9 ubiquitylation by His pull-down. HEK293 cells were transfected with FLAG–Cas9 along with His–Ub. Nickel-purified His–Ub conjugates were probed with a FLAG antibody to detect ubiquitylated Cas9 forms (highlighted by a bracket and asterisks). (C) Demonstration of Cas9 ubiquitylation by endogenous ubiquitin. HEK293 cells were transfected with doxycycline (Dox)-inducible FLAG-tagged Cas9 only. Pull-down of FLAG-Cas9 was followed by Western blot analysis with a ubiquitin antibody, as in (A). ± MG132 samples were run on the same gel and then subjected to Western blot and enhanced chemiluminescence on the same membrane. (D) Proximity ligation (Duolink) assays probe Cas9 modification by endogenous ubiquitin and physical interaction with the proteasomes, using an antibody against the PSMA5 subunit. Assays were performed in HEK293 cells stably expressing FLAG–Cas9. Z-stack projections are shown. Nuclei were stained with DAPI. Quantifications of positive Duolink signals are shown in the graph. n > 25 cells per experiment. Data represent mean value from two experiments per condition, including the negative controls using a single antibody of a given Duolink pair, ± SEM (****P < 0.0001, unpaired t test). MG132: proteasome inhibitor. Representative confocal images of the negative controls are shown in Fig S3B. For simplicity, P-values between the following groups were not included in the graph: PCas9-Ubiquitin & Cas9 < 0.0001, PCas9-Ubiquitin & Ubiquitin < 0.0001, PCas9-Ubiquitin (+MG132) & Cas9 < 0.0001, PCas9-Ubiquitin (+MG132) & Ubiquitin < 0.0001, PCas9-PSMA5 & Cas9 < 0.0001, PCas9-PSMA5 & PSMA5 < 0.0001, PCas9-PSMA5 (+MG132) & Cas9 < 0.0001, PCas9-PSMA5 (+MG132) & PSMA5 < 0.0001.

Fig 3: Galangin-induced TRAIL-mediate apoptosis is associated with up-regulation of proteasome activity.(a) Caki cells were treated with 30 μM galangin for 24 h. After lysis of cells, the proteasome activity was determined as described in the Materials and Methods. As a positive control, 1 μM MG132 was used. (b) Caki cells were treated with indicated concentrations of galangin for 24 h. Protein levels of PSMA5 and PSMD4 were determined by western blotting. Actin was used as a loading control. (cropped, full-length blots are in Supplementary Fig. S8). (c) Caki cells were treated with 30 μM galangin for the indicated time periods and loaded with a H2DCF-DA fluorescent dye. H2DCF-DA fluorescence intensity was detected by flow cytometry. P.C. (Positive control; 20 μM shogaol) (d) Caki cells were pretreated with 5 mM NAC and 2 mM GEE for 30 min, and then stimulated with 30 μM galangin plus 50 ng/ml TRAIL for 24 h. Apoptosis was analyzed as the sub G1 population by FACS analysis. The protein expression levels of PARP and actin were determined by Western blotting. The level of actin was used as a loading control. (cropped, full-length blots are in Supplementary Fig. S8).

Fig 4: Summary of the results identified in this manuscript through which NF-?B2 and RelB regulate p53 dependent cellular senescence in primary human NHD fibroblasts.Depletion of NF-?B2 and RelB leads to a decrease in Rb phosphorylation. Unphosphorylated Rb represses E2F transcriptional activity and consequently inhibits EZH2 expression (which is negatively regulated by p53). However, this occurs through distinct pathways. Depletion of NF-?B2 leads to down regulation of CDK4 and CDK6, which are known to phosphorylate Rb directly. RelB depletion leads to an increase in p53 and p21 protein stability as a consequence of loss of expression of genes such as PSMA5 and ANAPC1. Other gene targets may be involved in these processes. NF-?B subunits also bind directly to the EZH2 promoter and this may also contribute towards its regulation. As a consequence of this pathway, down regulation of EZH2 results in numerous changes in gene expression, including p53 dependent repression of a number of gene targets. These include inhibition of RACGAP1 expression, resulting in Rac1/Cdc42 dependent induction of reactive oxygen species (ROS). Together with other changes, such as suppression of DEK and ultimately induction of p14ARF and p16INK4a, the ultimate consequence of NF-?B2 and RelB siRNA depletion and subsequent loss of EZH2 expression is p53 dependent cell senescence. Note, RACGAP1 activity as well as various inducers/suppressors or senescence may also be regulated by p53. Many of these genes may also be direct targets of NF-?B. Not shown is that oxidative stress is required to drive the basal level p53 activity seen in these cells.

Fig 5: RelB controls Rb phosphorylation, EZH2 expression and senescence through PSMA5 induced regulation of p21WAF1 and p53 protein stability.(A) PSMA5 and ANAPC1 regulate p21WAF1 and p53 protein stability. Western blot analysis of NHD fibroblasts treated with the indicated siRNAs. (B & C) RelB regulates PSMA5 expression. Whole cell protein lysates (B) or RNA (C) was prepared from NHD fibroblasts treated with the indicated siRNAs and western blot or Q-PCR analysis of PSMA5 expression was performed. (D) RelB regulates ANAPC1 expression. RNA was prepared from NHD fibroblasts treated with the indicated siRNAs and Q-PCR analysis of ANAPC1 was performed. (E) siRNA mediated knock down of PSMA5 results in loss of Rb phosphorylation. Western blot analysis of NHD fibroblasts treated with the indicated siRNAs. (F) siRNA mediated knock down of PSMA5 results in loss of EZH2 expression. RNA was prepared from NHD fibroblasts treated with the indicated siRNAs and Q-PCR analysis of EZH2 was performed. Psma5: (*** p=0.001) Anapc1: (* p=0.05). (G & H) siRNA mediated knock down of PSMA5 (G) or ANAPC1 (H) induces p53 dependent cellular senescence. NHD fibroblasts were transfected with the listed siRNAs and analyzed for senescence by ß-galactosidase staining after 7 days.