Article Figures & Data
Figures
- Fig. 1.
Overview of high-throughput screening data. (A) Hit triage from HTS library screen (10,804 compounds), to compounds that are cytotoxic toward parental KB-3-1 cells (1362 compounds), to putative P-gp substrates (90 compounds). (B) Summary of percentage of cytotoxic compounds (orange) vs. nontoxic compounds (blue), both cumulative and represented in each library screened. (C) Area-under-the-curve heatmap of compound activity for screen hits, where red intensity represents magnitude of AUC: Deep red is strongest cytotoxicity, and white represents no cytotoxicity.
- Fig. 2.
Summary of P-gp substrate analysis. (A) Summary of data analysis. For each compound, data are displayed as “loss of signal from baseline,” where −100% represents total cell killing. KB-3-1 cells are sensitive (black), KB-8-5-11 cells are resistant (red) but sensitized when P-gp inhibitor tariquidar is added (blue), left. To perform HTS data analysis, AUCs were determined for the parent and resistant lines, and the difference in AUC (termed ΔAUC) was used to identify putative substrates. ΔAUC1 is the difference in AUC in the KB-3-1 (black line) sensitive cells compared with KB-8-5-11 (red line) resistant cells (center); ΔAUC2 is the difference in AUC of KB-8-5-11 resistant cells in the absence (red line) and presence (blue line) of tariquidar (right). Comparison of the AUC for substrates calculated using ΔAUC1 and ΔAUC2, assessing (B) rank order of substrates (where largest ΔAUC is strongest substrate) and (C) orthogonal data generated testing P-gp substrates against HEK cells. (D) Unsupervised clustering of compound activity against parent and resistant lines (with and without tariquidar), where deep red represents greatest cytotoxicity.
- Fig. 3.
Representative examples of known (A–C) and newly identified (D–F) substrates of P-gp as determined by the screen.
- Fig. 4.
Effects of novel P-gp substrates on ATPase activity. The vanadate-sensitive activity of P-gp was determined as outlined in Materials and Methods. Basal P-gp ATPase activity was compared with activity in the presence of 0.1, 1, or 10 μM concentrations of the substrates (A) ispinesib, (B) KW-2478, (C) GSK-690693, (D) AT7519 or (E) gedatolisib. Verapamil (F) is shown as a positive control. Graphs depict average values from three independent experiments (error bars ± S.D.). Significance was determined by a one-way analysis of variance followed by a Dunnet test for multiple comparisons. Asterisks denote significant difference from the control, where P < 0.05 (*); P < 0.01 (**).
- Fig. 5.
Substrate synergy with P-gp inhibitors. Sample combination of elacridar and paclitaxel tested in 10 × 10 dose-response matrices with KB-3-1 (A and B) and KB-8-5-11 (C and D). (A and C) Left, percent response of cell viability where red = cell death, black = cell survival. (A and C) Right, matrix plot in ΔBliss model, magenta = synergy. (B and D) Dose-response curves extracted from synergy blocks for paclitaxel with increasing concentrations of elacridar. Dose-response curves for (E) tariquidar and (F) elacridar for all substrates tested (listed in Table 2). Values in parentheses are the log of the molar concentration of tariquidar or elacridar needed to achieve maximum cell killing when combined with the noted drugs.
- Fig. 6.
Inhibition of P-gp- and ABCG2-mediated transport by P-gp substrates. P-gp-overexpressing MDR-19 cells (A) or ABCG2-overexpressing R-5 cells (B) were incubated with 0.5 μg/ml rhodamine 123 or 5 μM pheophorbide A, respectively, in the absence or presence a specific inhibitor (3 μM Valspodar for P-gp and 10 μM fumitremorgin C for ABCG2) or 25 μM gedatolisib, KW-2478, ispinesib, GSK-690693, AT9283, or AT7519 for 30 minutes after which the medium was removed and replaced with substrate-free medium continuing without or with the inhibitor. Cells incubated with substrates alone are shown by the red histogram, cells incubated with the substrate and specific inhibitor are shown by the teal histogram, and cells incubated with 25 μM concentrations of the test compounds are noted in the key. Results from one of two experiments are shown.
Tables
- TABLE 1
Cross-resistance profile for novel P-gp substrates in P-gp- or ABCG2-expressing cells
All compounds were tested in triplicate. Results presented are mean GI50 values ± S.D.
Compound Structure GI50 pcDNA GI50 MDR-19 RRa P-gp GI50 R-5 RRa ABCG2 μM μM μM AT7519 
0.61 ± 0.07 66.7 ± 28.2 110 2.3 ± 0.2 4 AT9283 2.7 ± 0.1 34.8 ± 6.8 13 60.8 ± 1.9 22 Ispinesib 0.044 ± 0.037 2.8 ± 0.5 64 0.14 ± 0.09 3 Gedatolisib 0.028 ± 0.007 47.2 ± 12.8 1672 2.0 ± 0.3 70 GSK-690693 0.19 ± 0.03 56.2 ± 18.8 290 11.6 ± 1.8 60 KW-2478 0.48 ± 0.04 129.7 ± 27.5 270 4.5 ± 0.2 9 ↵a Relative resistance (RR) value is the ratio of the GI50 (50% growth inhibitory concentration) values of P-gp- or ABCG2-expressing cells (MDR-19 or R-5) to the GI50 value of pcDNA cells.
Compound Mechanism of Action Mithramycin A Alcohol dehydrogenase inhibitor Danusertib Aurora-A/B/C kinase inhibitor Tozasertib Aurora-A/B/C kinase inhibitor PHA-793887 CDK1,2,3,4,5 inhibitor Romidepsin Histone deacetylase (HDAC) inhibitor JNK-IN-7 JNK inhibitor PKI-587 mTOR inhibitor CB 300919 NAMPT inhibitor AST-487 RET kinase inhibitor Sepantronium bromide Survivin inhibitor Docetaxel Tubulin polymerization promoter Paclitaxel Tubulin polymerization promoter Vinorelbine Tubulin polymerization inhibitor RO495 Tyk 2 inhibitor RA-XII N/A (Natural product) CDK, cyclin-dependent kinase.
Data Supplement
- Supplemental Table -
Supplementary Table 1 - P-gp substrates identified.
- Supplemental Table -
