Fig 2: R45H mutation increases NADK activity. (A) The mutation frequency (mutated cases/total cases) of NADK across 22 human tissue types from the COSMIC database. (B) The distribution of substitution missense mutation positions of NADK from the COSMIC database. (C) The methylation levels of NADK WT and R45H mutant proteins were quantified by immunoblotting with anti-MME-R antibody. (D–F) NADK enzymatic assay with purified NADK WT and R45H mutant proteins from HEK 293T cells (n = 3). Michaelis–Menten curve (D), Kcat (E), and relative catalytic efficiency (F) are presented. (G) NADK WT and R45H mutant plasmids were transfected into HEK293T cells. The cell lysates were cross-linked with BS3. NADK input and cross-linked protein samples were detected by immunoblot analyses. Data in (D–F) are presented as mean ± standard deviation; one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

Fig 3: R45H mutation promotes tumor growth by reducing reactive oxygen species (ROS) levels. (A, B) H520 cells with stable sgRNA-mediated knockout of NADK were stably reconstituted with NADK WT or R45H mutant. The relative abundance of NADP+ (A) and NADPH (B) quantified from these reconstituted H520 cell lines was presented (n = 3). (C) The reconstituted H520 cells were incubated with the ROS assay working solution at 37 °C for 30 min. Relative ROS level was quantified by flow cytometry (n = 3). (D, E) The reconstituted H520 cells were labeled with 10 μM EdU for 30 min, and EdU-positive cells were examined by immunostaining test (D). EdU-positive cells are presented in (E) (n = 5). Scale bars: 50 μm. (F) The reconstituted H520 cells were exposed to indicated concentrations of cisplatin for 96 h, and cell viability assay was performed (n = 4). The IC50 of cisplatin was calculated. (G–I) Mice were subcutaneously injected with 1 × 106 of the indicated H520 cells on both sides of the armpit (n = 5). When tumors were visible, mice were treated with cisplatin (2 mg/kg) every 3 days for 2 weeks. The tumors were harvested after an additional 7 days of growth (G). Relative tumor volume (H) and tumor weight ratio (cisplatin/ctrl) (I) are presented. Data in (A–F, I) are presented as mean ± standard deviation; one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Data in (H) are presented as mean ± standard deviation; two-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

Fig 4: N-terminal region regulates NADK conformation. (A) Comparison of the cryo-EM structure (green, pdb: 8KGC) and the crystal structure (purple, pdb: 3PFN) of tetrameric NADK. (B) Close-up view of the NADK dimer involving the N-terminal regulation. Two pairs of anti-parallel β strands, the first β1 strand in the N-terminal domain (NTD), the linker anti-parallel β strands (β6 and β19), and the C-terminal αC helix are labeled. (C) NADK dimer presented in ribbon, highlighting the GGDG, GxxG, and the Gly-rich motifs. The arrows indicate the rotation of NTD around the linker anti-parallel β strands. (D) The indicated NADK full-length (FL) and N-terminal truncation mutant (ΔN95) plasmids were introduced into HEK293T cells with endogenous NADK knocked out. HA-NADK was eluted in size exclusion fractions on the same column in identical settings. Indicated fractions of cell lysates were used for immunoblot analyses.

Fig 5: NADK tetramer is required for its enzyme activity. (A) Tetramer interface of NADK. C349 and R356 on the interface, and the hydrophobic interaction and salt bridges are highlighted and labeled. (B) NADK WT and indicated mutant plasmids were transfected into HEK293T cells. The cell lysates were cross-linked with 0.2 mM BS3 and terminated with 50 mM Tris–HCl (pH 8.0). NADK input and cross-linked fractions were detected by immunoblot analyses. (C–E) NADK enzymatic assay with purified NADK WT and indicated mutant proteins from E. coli BL21 strain (n = 3). Michaelis–Menten curve (C), Kcat (D), and relative catalytic efficiency (E) are presented. (F) ITC binding curves for NADK WT and indicated mutant proteins with substrate molecule NAD+. The Kd values, when measurable, are shown in the figure, while non-measurable values are indicated by NA. Data in (C–E) are presented as mean ± standard deviation; one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.