Fig 1: QA can bind to the ATP−binding site of Hsp90 and inhibit Hsp90 activity. (A) The expression level of proteins with QA treatment. A549 cells were treated with QA (2.5 and 5 μM) for 48 h. Protein levels were analyzed by western blotting. The 17−AAG (4 and 8 μM) was used as a positive control. (B) Effect of QA on A549 cell migration. Scratched A549 cells were treated with QA (0.25, 0.5, 1 μM) for 0, 24, and 48 h; 17−AAG (0.5 μM) was used as a positive control. (C) Inhibitory effect of compounds on Hsp90α N−ATPase. Hsp90α N−ATPase activity of the compounds at 100 μM was tested by the malachite green phosphate assay with 2.5 μg Hsp90α N−ATPase. After 3 h, the reaction was stopped, and the OD620 value was measured. (D) A competitive binding assay employing FITC−labeled geldanamycin and Hsp90α N−ATPase was performed. We used 40 nM GM−FITC, 80 nM Hsp90 N−ATPase, and QA at the indicated concentrations (100, 50, 25, 12.5, 6.25, 3.12, and 1.56 μM).
Fig 2: Binding model of QA on Topo II and Hsp90. (A,B) Binding model of QA on Topo II (PDB ID:1ZXM). (C) Overlay of QA and ANP on Topo II. QA and ANP are shown in cyan and magenta, respectively. (D,E) Binding model of QA on Hsp90 (PDB ID:3T1K). (F) Overlay of QA and ANP on Hsp90. QA and ANP are shown in cyan and magenta, respectively.
Fig 3: Hsp90 protein binds PS/cEt or PS/LNA ASOs but not PS/MOE ASO. (A) Affinity selection was performed using biotinylated gapmer PS-ASOs with different 2′-modifications in the wings. Proteins were eluted using non-biotinylated PS-ASOs with the same chemical composition. Isolated proteins were separated by SDS-PAGE, and visualized by silver staining (upper panel), or detected by Western assay for a duplicate gel (lower panel) using antibodies specific to Hsp90α or HSP90β. Ku70 protein served as a control for loading. The protein band containing Hsp90 is indicated for the silver stained gel. (B) Western analysis for Hsp90 protein co-selected using a biotinylated PS/cEt ASO, and eluted by competition using PS/cEt ASOs with two different sequences that target PTEN or NCL1 mRNAs, or using a PS/MOE ASO targeting NCL1. The Hsp90 antibody recognizes both α and β isoforms. Ku70 protein was detected and served as a control. (C) Affinity selection using PS/cEt ASOs that have the same sequence and chemistry but tagged with biotin at either 5′ or 3′ end. Co-isolated Hsp90 and Ku70 were detected by western assay. (D) Affinity selection was carried out using either a single stranded PS/cEt ASO or a duplex formed using the same ASO and a complementary 2′-O-methylated oligonucleotide. Co-isolated proteins were directly separated on SDS-PAGE and analyzed by Western assay for Hsp90 and Ku70 proteins. (E) Affinity selection was conducted using either a PS/MOE ASO 386652 or a PS/cEt ASO 586183. After washing, bound proteins were first eluted by RNase I (5 U/μl) or DNase I (5 U/μl) treatment for 30 min at RT, followed by elution using 9M Urea from the same beads for 30 min at RT. The eluted proteins were precipitated and analyzed by Western assay for Hsp90 protein. (F) Silver staining of one μg of the recombinant Hsp90α protein (Abcam, ab80369). (G) Interaction of the recombinant Hsp90α protein with PS/cEt ASO was determined by affinity selection using biotinylated PS-ASOs and 3 μg of purified Hsp90 protein. After wash, the beads were directly loaded on a 4–12% SDS-PAGE, and Hsp90 protein was detected by Western assay. The ASO ID numbers in each panel are shown. The above experiments were repeated at least three times and representative results are shown. (H) Membrane binding assay for Hsp90α protein and different PS-ASOs, as described in Materials and Methods. For each ASO, triplicate binding reactions were performed. The molar ratio between Hsp90a protein and ASO was given below the wells. An example of a double-filter binding for PS/cEt and PS/LNA ASOs is shown in the middle panel for protein-bound ASOs captured using a Hybond ECL membrane, and the lower panel for unbound ASOs captured using a Hybond-N+ nylon membrane. The signal intensity for the ASOs was quantified and the binding curves for (I) PS/cEt and (J) PS/LNA ASOs were plotted using Prism. The calculated kds are given. The error bars represent standard deviations from three experiments.
Fig 4: Plot of RMSD (in ångstrom) for the Hsp90-LZY228 complex during 50 ns MD simulation (A). The binding model of Hsp90-LZY228 complex was shown by PyMoL (B).
Fig 5: Evaluation of LZY228 as the inhibitor of Hsp90 in cellular systems.
Supplier Page from Abcam for Recombinant human Hsp90 alpha protein (Active) (His tag C-Terminus)