Clusterin facilitates COMMD1 and I-kappaB degradation to enhance NF-kappaB activity in prostate cancer cells

Abstract

Secretory clusterin (sCLU) is a stress-activated, cytoprotective chaperone that confers broad-spectrum cancer treatment resistance, and its targeted inhibitor (OGX-011) is currently in phase II trials for prostate, lung, and breast cancer. However, the molecular mechanisms by which sCLU inhibits treatment-induced apoptosis in prostate cancer remain incompletely defined. We report that sCLU increases NF-kappaB nuclear translocation and transcriptional activity by serving as a ubiquitin-binding protein that enhances COMMD1 and I-kappaB proteasomal degradation by interacting with members of the SCF-betaTrCP E3 ligase family. Knockdown of sCLU in prostate cancer cells stabilizes COMMD1 and I-kappaB, thereby sequestrating NF-kappaB in the cytoplasm and decreasing NF-kappaB transcriptional activity. Comparative microarray profiling of sCLU-overexpressing and sCLU-knockdown prostate cancer cells confirmed that the expression of many NF-kappaB-regulated genes positively correlates with sCLU levels. We propose that elevated levels of sCLU promote prostate cancer cell survival by facilitating degradation of COMMD1 and I-kappaB, thereby activating the canonical NF-kappaB pathway.

Figure 1. sCLU is a COMMD1 and ubiquitin partner in prostate cancer cells

A, sCLU interacts with COMMD1. Total proteins from PC-3, LN_mock and LN_sCLU were co-IP with anti-CLU or normal IgG followed by western blot using COMMD1 antibody. The inverse experiment was performed using COMMD1 antibody for co-IP. CLU, COMMD1 and vinculin were used for western blot (sCLU is detected as a 60-kDa glycosylated presecretory form of sCLU and is referred as psCLU in all western blot as proposed by Trougakos (43)). B, sCLU co-localizes with COMMD1 and ubiquitin. Immunofluorescence was performed in PC-3, LN_mock and LN_sCLU using CLU, COMMD1, and ubiquitin antibodies and DAPI for nuclei staining. C, sCLU associates with ubiquitin. Total proteins were incubated with ubiquitin-agarose matrix or Nickel agarose; and bound protein and input were used for western blots with CLU and Hsp90 antibodies. D, sCLU enhances ubiquitinated proteins. Total proteins (50ug) from LN_mock and LN_sCLU or PC-3 treated with sCLU or Scr siRNA were monitored for cleavage of the Suc-LLVY-AMC. Fluorescence was quantified using a spectrofluorometer (Fluoroskan Ascent FL, Thermo Labsystem). Data is expressed as mean ± SE. *, statistical significance (p<0.05) from 3 biological replicates.

Figure 1. sCLU is a COMMD1 and ubiquitin partner in prostate cancer cells

A, sCLU interacts with COMMD1. Total proteins from PC-3, LN_mock and LN_sCLU were co-IP with anti-CLU or normal IgG followed by western blot using COMMD1 antibody. The inverse experiment was performed using COMMD1 antibody for co-IP. CLU, COMMD1 and vinculin were used for western blot (sCLU is detected as a 60-kDa glycosylated presecretory form of sCLU and is referred as psCLU in all western blot as proposed by Trougakos (43)). B, sCLU co-localizes with COMMD1 and ubiquitin. Immunofluorescence was performed in PC-3, LN_mock and LN_sCLU using CLU, COMMD1, and ubiquitin antibodies and DAPI for nuclei staining. C, sCLU associates with ubiquitin. Total proteins were incubated with ubiquitin-agarose matrix or Nickel agarose; and bound protein and input were used for western blots with CLU and Hsp90 antibodies. D, sCLU enhances ubiquitinated proteins. Total proteins (50ug) from LN_mock and LN_sCLU or PC-3 treated with sCLU or Scr siRNA were monitored for cleavage of the Suc-LLVY-AMC. Fluorescence was quantified using a spectrofluorometer (Fluoroskan Ascent FL, Thermo Labsystem). Data is expressed as mean ± SE. *, statistical significance (p<0.05) from 3 biological replicates.

Figure 2. sCLU induces ubiquitination and proteasome-dependent degradation of COMMD1

A, COMMD1 levels are negatively regulated by sCLU. Left Panel - LNCaP cells were transiently transfected with increasing amounts of sCLU cDNA as indicated. After 48h, expression levels of COMMD1 were determined by western blot. PC-3 cells were treated with 10nM sCLU or Scr siRNA and western blot was performed using COMMD1, CLU. Right Panel - COMMD1 mRNA levels are not affected by sCLU. RNA was extracted from LN_mock, and LN_sCLU or PC-3 cells treated sCLU or Scr siRNA. COMMD1 mRNA levels were analyzed using RT-PCR (MM S3) with actin as a control (upper panel) and by northern blot with CLU and 28S as a controls (lower panel). B, sCLU levels affect COMMD1 stability. LN_mock, LN_sCLU (left panel) or PC-3 cells were treated with 10nM sCLU or Scr siRNA (right panel) followed by 10 µmol/L cycloheximide with DMSO as a control. COMMD1 protein levels were measured by Western blot analysis. C, COMMD1 protein levels are regulated by proteasomal degradation. LN_mock or LN_sCLU cells were treated with proteasome inhibitor (bortezomib) and COMMD1 levels assessed by western blot using COMMD1 antibodies. D, Effect of sCLU on COMMD1 ubiquitination. LNCaP cells were co-transfected with either empty vector or sCLU in parallel with COMMD1, ubiquitin WT (upper left panel), or mutated ubiquitin (K48R) (upper right panel). After 48 h, cells were treated +/− MG132 for 6 hrs, proteins extracted in RIPA buffer, and pull down assay using Nickel-agarose was performed followed by western blot using ubiquitin, COMMD1 and CLU (input) antibodies. LNCaP cells were co-transfected with ubiquitin +/− sCLU and COMMD1 as indicated. After 48h, cells were treated with MG132 for 6 h. Cells lysates were prepared with 8M Urea buffer; co-IP using COMMD1 (lower left panel) or ubiquitin (lower right panel) antibodies; and western blots (co-IP and Input) were performed using ubiquitin COMMD1 and CLU antibodies.

Figure 2. sCLU induces ubiquitination and proteasome-dependent degradation of COMMD1

A, COMMD1 levels are negatively regulated by sCLU. Left Panel - LNCaP cells were transiently transfected with increasing amounts of sCLU cDNA as indicated. After 48h, expression levels of COMMD1 were determined by western blot. PC-3 cells were treated with 10nM sCLU or Scr siRNA and western blot was performed using COMMD1, CLU. Right Panel - COMMD1 mRNA levels are not affected by sCLU. RNA was extracted from LN_mock, and LN_sCLU or PC-3 cells treated sCLU or Scr siRNA. COMMD1 mRNA levels were analyzed using RT-PCR (MM S3) with actin as a control (upper panel) and by northern blot with CLU and 28S as a controls (lower panel). B, sCLU levels affect COMMD1 stability. LN_mock, LN_sCLU (left panel) or PC-3 cells were treated with 10nM sCLU or Scr siRNA (right panel) followed by 10 µmol/L cycloheximide with DMSO as a control. COMMD1 protein levels were measured by Western blot analysis. C, COMMD1 protein levels are regulated by proteasomal degradation. LN_mock or LN_sCLU cells were treated with proteasome inhibitor (bortezomib) and COMMD1 levels assessed by western blot using COMMD1 antibodies. D, Effect of sCLU on COMMD1 ubiquitination. LNCaP cells were co-transfected with either empty vector or sCLU in parallel with COMMD1, ubiquitin WT (upper left panel), or mutated ubiquitin (K48R) (upper right panel). After 48 h, cells were treated +/− MG132 for 6 hrs, proteins extracted in RIPA buffer, and pull down assay using Nickel-agarose was performed followed by western blot using ubiquitin, COMMD1 and CLU (input) antibodies. LNCaP cells were co-transfected with ubiquitin +/− sCLU and COMMD1 as indicated. After 48h, cells were treated with MG132 for 6 h. Cells lysates were prepared with 8M Urea buffer; co-IP using COMMD1 (lower left panel) or ubiquitin (lower right panel) antibodies; and western blots (co-IP and Input) were performed using ubiquitin COMMD1 and CLU antibodies.

Figure 3. Clusterin facilitates I-kB degradation

A, sCLU modulates total I-κB protein levels. Total protein from LN_mock or LN_sCLU (left panel) were extracted and I-κB protein levels assessed by western blot. PC-3 cells (right panel) were treated with 10nM sCLU or Scr siRNA and western blotting was performed using COMMD1, sCLU and vinculin antibodies (left panel). B, sCLU modulates I-κBα degradation. Left panels-LN_mock or LN_sCLU and PC-3 treated with 10nM sCLU or Scr siRNA were treated with 20ng/ml of TNF-α as indicated. Right Panel- PC-3 treated with 10nM sCLU or Scr siRNA and treated with either 20ng/ml of TNF-α for 30 min or MG132 for 3 hours followed by 30 min TNF-α treatment. Western blots were performed using I-κBα pI-κBα, sCLU and vinculin antibodies. C, sCLU knockdown stabilizes I-κB in cellulo. Hela -I-κB-Fluc cells were treated with 10 nM sCLU- or Scr-siRNA for 48hr. Cells were then treated with 20ng/ml TNF-α and imaged using an IVIS system with luciferin as a substrate (left panel). Photon count was plotted as a function of time after addition of TNF-α or vehicle and normalized at given time point as a fold of TNF α untreated cells (MM S4); values were expressed as a fold of initial value (right panel). D, sCLU regulates steady-state turnover of I-κBα. LN_mock or LN_sCLU were stimulated by a 10 min pulse of TNF-α and medium containing cyclohexamide (30uM) was added as indicated. Proteins were western blotted using I-κBα, CLU and vinculin antibodies.

Figure 4. sCLU regulates proteasomal-dependent degradation rates of I-κB

A, sCLU enhances proteasomal degradation of endogenous I-κBα. LNCaP cells were transfected with sCLU or empty vector, and MG132 added 3h before harvesting cells. Total lysates were analyzed by western blot using I-κBα CLU and vinculin antibodies. B, sCLU induces I-κBα ubiquitination. LNCaP cells were co-transfected with ubiquitin +/− sCLU and I-κB as indicated. After 48h, cells were treated with MG132. Total proteins were co-IP with I-κBα antibody. Western blots (co-IP and Input) were performed using ubiquitin or I-κBα and CLU antibodies (left panel). The inverse experiment was performed using ubiquitin antibody for co-IP and I-κB, CLU for western blot (right panel). C, CLU is found in a complex with I-κB, NF-κB and COMMD1. LNCaP cells were co-transfected with COMMD1, I-κB +/− sCLU as indicated. After 48h, cells were treated with TNF-α for 30 min. Co-IP was performed using COMMD1 antibody and western blot with I-κB, CLU and COMMD1 antibodies (upper panel). sCLU interacts with p65-NF-kB. LN_sCLU were treated with 20ng/ml TNF-α as indicated and proteins were Co-IP with p65-NF-κB antibody. A western blot was performed using sCLU or p65-NF-κB (middle panel). sCLU interacts with pI-κB. LN_sCLU cells were treated with or without MG132 for 3h prior to TNF-α as indicated. Proteins were co-IP with pI-κB antibody and western blot performed using CLU antibody (lower panel). D, sCLU interacts with cullin1: upper panel - LN_CLU2 cells were treated with or without TNF-α for 30 min and co-IP with Cul1 or IgG antibodies followed by western blot with CLU and Cul-1 antibodies. Lower panel - LNCaP cells were transfected with βTrCP-1 and sCLU and treated with TNF-α as indicated. Proteins were co-IP with Flag antibody and western blot performed using CLU and Flag antibodies.

Figure 5. Effect of sCLU on NF-kB transcription activity

A, sCLU enhances NF-κB transactivation. Left panel - LNCaP cells were transiently co-transfected with NF-κB-Luciferase and Renilla plasmids and sCLU plasmid. Total amount of plasmid DNA transfected was normalized to 1.65 ug/well by addition of empty vector. Right panel - LN_mock and LN_CLU2 were transiently transfected with NF-κB-Luciferase and Renilla plasmids. After 48 h, cells were harvested and luciferase activity determined. B, Left panel - TNF-α enhances the effect of sCLU on NF-κB activation: LN_mock and LN_sCLU were transfected with NF-κB-Luciferase plasmid. After 24 h, cells were treated with 20ng/ml of TNF-α for 24h and luciferase activity determined. Data represents the mean of at least 3 independent experiments performed in triplicate. Fold is measured relative to NF-κB activation in LN_mock without treatment. Right panel - sCLU accelerates NF-κB nuclear translocation after TNF-α. LN_mock and LN_sCLU were treated with +/− 20ng/ml TNF-α and immunofluorescence performed using p65-NF-κB antibody. C. sCLU knockdown decreases NF-κB activation. Left - PC3 cells were transfected simultaneously with NF-κB-Luciferase plasmid and sCLU or Scr siRNA. After 48 h, luciferase activity was determined. Triplicate luciferase assays repeated 3 times and reported as mean + SE. **, statistical significance, (p<0.001). Right - sCLU knockdown inhibits p65-NF-kB nuclear localization. PC-3 cells were transfected with 10nM sCLU or Scr siRNA, for 48h followed by +/− 20ng/ml TNF-α Immunofluorescence was performed using p65 NF-kB antibody and DAPI for nuclei staining.

Figure 5. Effect of sCLU on NF-kB transcription activity

A, sCLU enhances NF-κB transactivation. Left panel - LNCaP cells were transiently co-transfected with NF-κB-Luciferase and Renilla plasmids and sCLU plasmid. Total amount of plasmid DNA transfected was normalized to 1.65 ug/well by addition of empty vector. Right panel - LN_mock and LN_CLU2 were transiently transfected with NF-κB-Luciferase and Renilla plasmids. After 48 h, cells were harvested and luciferase activity determined. B, Left panel - TNF-α enhances the effect of sCLU on NF-κB activation: LN_mock and LN_sCLU were transfected with NF-κB-Luciferase plasmid. After 24 h, cells were treated with 20ng/ml of TNF-α for 24h and luciferase activity determined. Data represents the mean of at least 3 independent experiments performed in triplicate. Fold is measured relative to NF-κB activation in LN_mock without treatment. Right panel - sCLU accelerates NF-κB nuclear translocation after TNF-α. LN_mock and LN_sCLU were treated with +/− 20ng/ml TNF-α and immunofluorescence performed using p65-NF-κB antibody. C. sCLU knockdown decreases NF-κB activation. Left - PC3 cells were transfected simultaneously with NF-κB-Luciferase plasmid and sCLU or Scr siRNA. After 48 h, luciferase activity was determined. Triplicate luciferase assays repeated 3 times and reported as mean + SE. **, statistical significance, (p<0.001). Right - sCLU knockdown inhibits p65-NF-kB nuclear localization. PC-3 cells were transfected with 10nM sCLU or Scr siRNA, for 48h followed by +/− 20ng/ml TNF-α Immunofluorescence was performed using p65 NF-kB antibody and DAPI for nuclei staining.

Figure 5. Effect of sCLU on NF-kB transcription activity

A, sCLU enhances NF-κB transactivation. Left panel - LNCaP cells were transiently co-transfected with NF-κB-Luciferase and Renilla plasmids and sCLU plasmid. Total amount of plasmid DNA transfected was normalized to 1.65 ug/well by addition of empty vector. Right panel - LN_mock and LN_CLU2 were transiently transfected with NF-κB-Luciferase and Renilla plasmids. After 48 h, cells were harvested and luciferase activity determined. B, Left panel - TNF-α enhances the effect of sCLU on NF-κB activation: LN_mock and LN_sCLU were transfected with NF-κB-Luciferase plasmid. After 24 h, cells were treated with 20ng/ml of TNF-α for 24h and luciferase activity determined. Data represents the mean of at least 3 independent experiments performed in triplicate. Fold is measured relative to NF-κB activation in LN_mock without treatment. Right panel - sCLU accelerates NF-κB nuclear translocation after TNF-α. LN_mock and LN_sCLU were treated with +/− 20ng/ml TNF-α and immunofluorescence performed using p65-NF-κB antibody. C. sCLU knockdown decreases NF-κB activation. Left - PC3 cells were transfected simultaneously with NF-κB-Luciferase plasmid and sCLU or Scr siRNA. After 48 h, luciferase activity was determined. Triplicate luciferase assays repeated 3 times and reported as mean + SE. **, statistical significance, (p<0.001). Right - sCLU knockdown inhibits p65-NF-kB nuclear localization. PC-3 cells were transfected with 10nM sCLU or Scr siRNA, for 48h followed by +/− 20ng/ml TNF-α Immunofluorescence was performed using p65 NF-kB antibody and DAPI for nuclei staining.

Figure 6. sCLU modulates NF-κB dependent genes

A, sCLU expression correlates with NF-kB dependent genes. Total RNA were extracted from LN_mock and LN_sCLU, and also from LN_sCLU cells treated with Scr- or CLU-siRNA. Northern analysis was performed using Sema3C, NGAL, sPLA2-IIa, and MIP3a probes. B, Validation of gene array at the protein level. Total proteins were extracted from LN_mock and LN_sCLU, and also from LN_sCLU (left panel) and PC-3 cells (right panel) treated with Scr- or CLU-siRNA, western blots were performed using CLU, MCP-1, MCP-2, cIAP₂, sPLA2-IIa, MIP-3 and vinculin antibodies. C, Schema illustrating ligand-independent mode of NF-κB transactivation involving the molecular chaperone, sCLU.