Multi-kinase inhibitors can associate with heat shock proteins through their NH2-termini by which they suppress chaperone function

Abstract

We performed proteomic studies using the GRP78 chaperone-inhibitor drug AR-12 (OSU-03012) as bait. Multiple additional chaperone and chaperone-associated proteins were shown to interact with AR-12, including: GRP75, HSP75, BAG2; HSP27; ULK-1; and thioredoxin. AR-12 down-regulated in situ immuno-fluorescence detection of ATP binding chaperones using antibodies directed against the NH2-termini of the proteins but only weakly reduced detection using antibodies directed against the central and COOH portions of the proteins. Traditional SDS-PAGE and western blotting assessment methods did not exhibit any alterations in chaperone detection. AR-12 altered the sub-cellular distribution of chaperone proteins, abolishing their punctate speckled patterning concomitant with changes in protein co-localization. AR-12 inhibited chaperone ATPase activity, which was enhanced by sildenafil; inhibited chaperone - chaperone and chaperone - client interactions; and docked in silico with the ATPase domains of HSP90 and of HSP70. AR-12 combined with sildenafil in a GRP78 plus HSP27 -dependent fashion to profoundly activate an eIF2α/ATF4/CHOP/Beclin1 pathway in parallel with inactivating mTOR and increasing ATG13 phosphorylation, collectively resulting in formation of punctate toxic autophagosomes. Over-expression of [GRP78 and HSP27] prevented: AR-12 -induced activation of ER stress signaling and maintained mTOR activity; AR-12 -mediated down-regulation of thioredoxin, MCL-1 and c-FLIP-s; and preserved tumor cell viability. Thus the inhibition of chaperone protein functions by AR-12 and by multi-kinase inhibitors very likely explains why these agents have anti-tumor effects in multiple genetically diverse tumor cell types.

Keywords: ATPase; OSU-03012; chaperones; pazopanib; sorafenib.

Figure 1. Assessing chaperone expression by immuno-fluorescence and SDS PAGE / western blotting generates divergent data after OSU-03012 or sorafenib treatment, Part 1

(A and B) HuH7 cells and HT1080 were treated with vehicle, OSU-03012 (0–3.0 μM) and/or sildenafil (2 μM) for 2 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of GRP78, HSP70 and HSP90. The relative fluorescence intensity value from 40 different cells from each condition was determined using Hermes system software (+/− SEM). (C) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2 μM) and/or sildenafil (2 μM) for 2 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of GRP78, HSP70 and HSP90. (D) As indicated, GBM5, GBM6 and GBM12 cells were transfected with either an empty vector plasmid (CMV) or a plasmid to express dominant negative eIF2α S51A. Twenty four h after transfection cells were treated with vehicle, OSU-03012 (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of LC3 (ATG8) and Beclin1 (ATG6).

Figure 2. Assessing chaperone expression by immuno-fluorescence and SDS PAGE/western blotting generates divergent data after OSU-03012 or sorafenib treatment, Part 2

(A) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) and/or sildenafil (2 μM) for 6 h after which: (a) in 96 well plates, cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of ERK2 and JNK1; (b) cells were lysed with bromophenol blue buffer and subjected to SDS PAGE followed by immuno-blotting to detect the expression levels of ERK2 and JNK1. (B) GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2 μM) and/or sildenafil (2 μM) for 2 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of GRP78 and HSP90, using antibodies that recognize epitopes in the NH₂-terminus; COOH-terminus; and in the middle of the proteins, as well as our previously utilized “usual” antibody. (C) Left: GBM5 cells were transfected with a scrambled siRNA molecule or siRNA molecules to knock down the expression of BAG2, GRP75, HSP75, HSP27 or GRP78. Twenty four h after transfection cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of BAG2, GRP75, HSP75, HSP27 or GRP78, as indicated. Right: GBM5 and GBM12 cells were transfected with an empty vector plasmid (CMV) or plasmids to express GRP78 or HSP27. Twenty four h after transfection cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP27 or GRP78, as indicated. (D) Left: GBM5 and GBM12 cells were transfected with a scrambled siRNA molecule or siRNA molecules to knock down the expression of HSP90, GRP78 or HSP70. Twenty four h after transfection cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP90, HSP70 or GRP78, as indicated. Right: GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) or sorafenib (2 μM) and/or sildenafil (2 μM) for 6 h. Cells were then lysed with bromophenol blue buffer and subjected to SDS PAGE followed by immuno-blotting to detect the expression levels of GRP78, HSP70, HSP90 and GAPDH.

Figure 3. Proteomic and cell biology analyses of AR-12 interacting proteins

(A) As described in the Methods, AR-12 conjugated via biotin to sepharose beads was used to capture proteins from a whole cell lysate and those proteins specifically associating with beads in an AR-12–dependent fashion were determined after proteolytic digestion in a mass spectrometer. (B) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which: (a) in 96 well plates, cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of HSP27 presented at 10X and 60X magnification; (b) cells were lysed with bromophenol blue buffer and subjected to SDS PAGE followed by immuno-blotting to detect the expression level of HSP27. (C) GBM5, GBM6 and GBM12 cells were treated with vehicle, OSU-03012 (0–3.0 μM) or sorafenib (0–3.0 μM) for 3 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of HSP27. The relative fluorescence intensity value from 40 different cells from each condition was determined using Hermes system software (+/− SEM). (D) GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2 μM) and/or sildenafil (2 μM) for 2 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of HSP27, using antibodies that recognize epitopes in the NH₂-terminus; COOH-terminus; and in the middle of the protein, as well as our previously used “usual” antibody. (E) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the total expression level of AKT and p38 MAPK, and the phosphorylation levels of AKT T308; p38 MAPK; mTOR S2448; HSP27 S15; HSP27 S78; and HSP27 S82.

Figure 4. HSP75 and GRP75 are AR-12 interacting proteins

(A) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM)/sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of HSP75 presented at 10X and 60X magnification. (B and C) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM)/sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of GRP75 presented at 10X and 60X magnification.

Figure 5. BAG2 is an AR-12 interacting protein

(A) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM)/sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of BAG2 presented at 10X and 60X magnification. (B and C) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM)/sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of BAG1 and BAG3 presented at 10X magnification. (D) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM)/sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of HIP presented at 10X magnification.

Figure 6. AR-12 disrupts the co-localization of chaperones with their regulatory proteins

(A and B) GBM5 and GBM12 cells were treated with vehicle control or with OSU-03012 (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the co-localization of HSP70 and BAG2; and HSP70 and HSP27; presented at 60X magnification. (C) GBM5 and GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of AhA1 presented at 10X magnification. (D) GBM12 cells were treated with vehicle control or with OSU-03012 (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the co-localization of HSP90 and AhA1 presented at 10X and 60X magnification.

Figure 7. Drug combination treatments reduce CDC37 Serine 13 phosphorylation

(A) GBM12 cells were treated for 6 h with OSU-03012 (2.0 μM) and sildenafil (2.0 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of CDC37 and HSP90 and the co-localization of CDC37 and HSP90 at 60X magnification. (B) GBM12 cells were treated with vehicle, OSU-03012 (2.0 μM) / sorafenib (2.0 μM) / pazopanib (2.0 μM) and/or sildenafil (2 μM) for 6 h after which cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression level of CDC37 and the Serine 13 phosphorylation level in CDC37 presented at 10X magnification.

Figure 8. The sub-cellular distribution and morphology of chaperone complexes in cells before and after treatment with [OSU-03012 + sildenafil], Part 1

(A–D) GBM14 cells were treated with either vehicle control or OSU-03012 (2 μM) and sildenafil (2 μM) for 6 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP90; GRP78; HSP70; and HSP40, presented at 60X magnification.

Figure 9. The sub-cellular distribution and morphology of chaperone complexes in cells before and after treatment with [OSU-03012 + sildenafil], Part 2

(A–C) GBM14 cells were treated with either vehicle control or OSU-03012 (2 μM) and sildenafil (2 μM) for 6 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of mitochondrial HSP70; HSP60; and HSP10, presented at 60X magnification.

Figure 10. Chaperone conformation is regulated by OSU-03012

GBM12 cells were treated with vehicle or OSU-03012 (2 μM); sorafenib (2 μM); regorafenib (2 μM); pazopanib (2 μM); AR-13 (2 μM); celecoxib (2 μM) for 2 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP90, GRP78 and HSP27, using antibodies that recognize epitopes in the NH₂-terminus; COOH-terminus; and in the middle of the proteins.

Figure 11. Modulation of HSP70 NH₂-terminal detection by OSU-03012 and other drugs

(A) GBM12 cells were treated with vehicle or sorafenib (2 μM); regorafenib (2 μM); OSU-03012 (2 μM); AR-13 (2 μM); pazopanib (2 μM) for 2 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP70 using antibodies raised to detect the NH₂-terminus, the central region, and COOH terminus of the protein. (B) GBM14 and GBM12 cells were treated with vehicle or sorafenib (2 μM); regorafenib (2 μM); pazopanib (2 μM) for 2 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of the indicated chaperone proteins.

Figure 12. Co-localization of the NH₂-terminal and COOH-terminal directed HSP70 antibodies

(A) GBM12 cells were treated with vehicle or [OSU-03012 (0-1.0 μM) + sildenafil (2.0 μM)] or with [AR-13 (1 μM) + sildenafil (2 μM)] for 6 h. Cells were then fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the total expression and the co-localization of c-MYC and HSPH1/p105. (B) GBM12 cells were treated with vehicle or [OSU-03012 (2.0 μM) + sildenafil (2.0 μM)] for 30 min. Cells were lysed and portions subjected to immuno-precipitation for c-MYC or for HSPH1/p105, followed by SDS PAGE and western blotting to determine the co-precipitation of HSPH1/p105 with c-MYC and of c-MYC with HSPH1/p105.

Figure 13. Sildenafil enhances the ATPase inhibitory effects of regorafenib, sorafenib, pazopanib, AR-13 and OSU-03012

(A) A GST-HSP90 NH₂-terminal fragment containing the ATP binding domain of the chaperone was synthesized in E. coli and purified from other bacterial proteins using glutathione sepharose. The GST-HSP90 NH₂-terminal fragment protein was not eluted off the sepharose beads. Equal portions of beads were immediately aliquoted into individual wells in a 96 well plate. Beads were resuspended in kinase reaction buffer containing vehicle control; OSU-03012; sorafenib tosylate; regorafenib; pazopanib; AR-13; celecoxib (30 nM; 100 nM; 300 nM; 1 μM) in triplicate, and incubated for 30 min at 37°C. The reaction was started by addition of ATP-lite substrate. The plate was removed from the incubator and placed into a Vector 3 plate reader to determine the luminescence of the reactions under each treatment condition (n = 3 (× 3) +/− SEM). (B and C) GBM12 cells were transfected with a plasmid to express HSP70-GFP or to express FLAG-tagged HSP90. Twenty four h after transfection cells were treated with vehicle control or sildenafil (2 μM) for 1 h. Chaperone proteins were immuno-precipitated using their tags in the presence of phosphatase inhibitors. Equal portions of precipitate sepharose beads were immediately aliquoted into individual wells in a 96 well plate. Beads were resuspended in ATPase reaction buffer containing vehicle control; OSU-03012; sorafenib tosylate; regorafenib; pazopanib; AR-13; (30 nM; 100 nM; 300 nM; 1 μM) in triplicate, and incubated for 30 min at 37°C. The reaction was started by addition of ATP-lite substrate. The plate was removed from the incubator and placed into a Vector 3 plate reader to determine the luminescence of the reactions under each treatment condition (n = 3 (× 3) +/− SEM).

Figure 14. Afatinib nor ruxolitinib have ATPase inhibitory effects against HSP90 or HSP70

(A) GBM12 cells were transfected with a plasmid to express HSP70-GFP or to express FLAG-tagged HSP90. Twenty four h after transfection chaperone proteins were immuno-precipitated using their tags. Equal portions of precipitate sepharose beads were immediately aliquoted into individual wells in a 96 well plate. Beads were resuspended in ATPase reaction buffer containing vehicle control; ruxolitinib; or afatinib (500 nM; 1000 nM; 1500 nM; 2 μM) in triplicate, and incubated for 30 min at 37°C. The reaction was started by addition of ATP-lite substrate. The plate was removed from the incubator and placed into a Vector 3 plate reader to determine the luminescence of the reactions under each treatment condition (n = 3 (× 3) +/− SEM). (B) GBM12 cells were treated with vehicle or [afatinib (1 μM) and ruxolitinib (1 μM)] for 6 h. Cells were fixed in place and permeabilized using 0.5% Triton X100. Immuno-fluorescence was performed to detect the expression levels of HSP70 and of HSP90 using antibodies raised to detect the NH₂-termini and COOH termini of the proteins.