Discovery and characterization of novel small-molecule inhibitors targeting nicotinamide phosphoribosyltransferase

Abstract

Nicotinamide phosphoribosyltransferase (NAMPT) is a promising anticancer target. Using high throughput screening system targeting NAMPT, we obtained a potent NAMPT inhibitor MS0 (China Patent ZL201110447488.9) with excellent in vitro activity (IC50 = 9.87 ± 1.15 nM) and anti-proliferative activity against multiple human cancer cell lines including stem-like cancer cells. Structure-activity relationship studies yielded several highly effective analogues. These inhibitors specifically bound NAMPT, rather than downstream NMNAT. We provided the first chemical case using cellular thermal shift assay to explain the difference between in vitro and cellular activity; MS7 showed best in vitro activity (IC50 = 0.93 ± 0.29 nM) but worst cellular activity due to poor target engagement in living cells. Site-directed mutagenesis studies identified important residues for NAMPT catalytic activity and inhibitor binding. The present findings contribute to deep understanding the action mode of NAMPT inhibitors and future development of NAMPT inhibitors as anticancer agents.

Figure 1. Discovery of a novel NAMPT inhibitor MS0 from the chemical library screen.

Schematic illustration of discovering a novel NAMPT inhibitor MS0 by HTS in a chemical library containing 24434 small-molecule compounds. Error bars represent the s.e. of experimental triplicates.

Figure 2. Identification of MS0 as an antitumor compound.

(A) Effect of MS0 and its analogue 733 on NAD level of HepG2 cells at 0 ∼ 100 μM after 24 hours treatment. (B) Concentration response curve of MS0 on NAD level of HepG2 cells after 24 hours treatment. (C) ITC titration of MS0 into NMNAT does not show detectable interaction. (D) Effect of MS0 at 0 ~ 10 μM on HepG2 cell viability (CCK-8 assay after 24 hours treatment). (E) Effect of 10 μM MS0 on HepG2 cell viability (CCK-8 assay after 24, 36, 48 or 72 hours treatment). (F) IC₅₀ of MS0 inhibition on various human cancer cell lines (SRB assay after 72 hours treatment). **P < 0.01 vs. Control. Error bars in (A), (B) and (D-F) represent the s.e. of experimental triplicates.

Figure 3. Inhibitory potency (IC₅₀) for designed and synthesized MS0 analogues on NAMPT activity.

MS0 and 46 novel analogues were synthesized and examined. Error bars in (B) and (C) represent the s.e. of experimental triplicates.

Figure 4. Inhibitory potency (IC₅₀) for MS0 analogues on human cancer cell lines and stem-like cancer cell line.

(A) After initial screen in three human cancer cell lines (HepG2, A549 and HCT116), 11 MS0 analogues were selected for IC₅₀ determination in HepG2 cells using SRB assay after 72 hours treatment. (B) Further comparison was performed for the potency of 4 MS0 analogues in HepG2 cells using CCK-8 assay after 48 hours treatment. **P < 0.01 vs. MS0. (C) Similar effects were observed for these 4 MS0 analogues in a stem-like human hepatoma cell line Huh7-C. *P < 0.05, **P < 0.01 vs. MS0. Error bars in A-C represent the s.e. of experimental triplicates.

Figure 5. Monitoring target binding of MS0 analogues in cell lysate and in intact cell.

(A) Representative Western blot of cellular thermal shift assay (CETSA) in cell lysate for NAMPT target with MS1 (at 100 μM). (B) CESTA melt curves in cell lysate for NAMPT target with MS0, MS1 and MS34 (all at 100 μM). (C) Representative Western blot of CETSA in intact cell for NAMPT target with MS1 (at 10 μM). (D) CESTA melt curves in intact cell for NAMPT target with MS0, MS1 and MS34 (all at 10 μM).

Figure 6. Comparison for target engagement of MS0 analogues in cell lysate and in intact cell.

(A) CESTA melt curves in cell lysate for NAMPT target with MS7 (at 100 μM). (B, C) ITDRF_CETSA at 62 °C in cell lysate (B) or in intact cell (C) for NAMPT target with MS1 or MS7.

Figure 7. Binding mode study of MS0 analogues with NAMPT.

(A-B) The potential binding mode of MS0 (yellow stick) with NAMPT. (C) The comparison of MS0 binding conformation in our docking model (yellow stick) with that in reported complex crystal structure (cyan stick). (D) Relative Kcat/Km in wild type NAMPT and 5 mutants. *P < 0.05, **P < 0.01 vs. NAMPT. (E-G) IC₅₀ of MS0 (E), MS1 (F) and MS34 (G) on wild type NAMPT and 3 mutants (H191A, A244S and I309Y). Error bars in (D-G) represent the s.e. of experimental triplicates.