Fig 2: Effects of transient (siRNA-mediated) NNT silencing on NCI-H295R cell redox balance, respiration, proliferation, and viability. Bars represent means ± SEM values, unless stated otherwise. (A) GSH/GSSG ratio in NCI-H295R cells transfected with KD siRNA (96 hours posttransfection), normalized to the corresponding ratio of SCR siRNA-transfected cells. Significant suppression of the GSH/GSSG ratio in KD siRNA cells suggests higher intracellular oxidative stress. Bars represent medians ± interquartile range values. *P < 0.05 (n = 8 independent experiments). (B) Proliferation rates observed in siRNA-transfected NCI-H295R cells, 72 to 166 hours posttransfection. ***P < 0.001 (n = 14). (C) Caspase-3/caspase-7 activity ratio in KD siRNA cells to SCR siRNA-transfected cells, after standardization to cell numbers (120 hours posttransfection). *P < 0.05 (n = 8). (D) Seahorse XF24 analysis of cellular OCR at baseline and after successive application of three mitochondrial respiration inhibitors (166 hours posttransfection). Results were standardized to protein concentration. Bars represent medians ± interquartile range values. P > 0.05 (n = 4). (E) ECAR, surrogate marker of anaerobic glycolysis, standardized for protein concentration. P > 0.05 (n = 4). (F) Proliferation under low-dose chemically induced oxidative stress (paraquat 10 µM) in KD siRNA and SCR siRNA-transfected cells, normalized to corresponding cell proliferation without paraquat treatment. *P < 0.05 (n = 6). A&R, antimycin A plus rotenone; FCCP, carbonyl cyanide-p-trifluoromethoxyphenylhydrazone; O/mycin, oligomycin.

Fig 3: Response of NCI-H295R cells to NNT silencing in the acute (transient KD) and chronic (stable KD) setting, with proposed redox adaptation mechanisms. Acute NNT KD induces oxidative stress as predicted by the role of NNT as a major NADPH generator; enhanced steroidogenesis and polyamine catabolism further accentuate ROS accumulation, triggering apoptosis and a sharp decline in cell proliferation. With time (stable KD), cells manage to adapt removing damaged proteins and enhancing spermine synthesis as an alternative, NADPH-independent ROS scavenger. This restores redox homeostasis and abrogates the original proapoptotic effect, but cellular proliferation remains suppressed. Horizontal arrows represent paucity of change.

Fig 4: RNA-seq analysis flowchart and differential gene expression. (A) Flowchart of initial RNA-seq analysis of mouse adrenals. (B) Representative heat map of RNA-seq analysis for substrain-specific differentially expressed genes (between Nnt−/− and NntBAC) within mouse adrenals. Genes were clustered by Partek hierarchical clustering based on gene expression values. Normalisation was performed by genes shifted to mean of zero and scaled to s.d. of 1. Arbitrary signal intensity from RNA-seq data is represented by colours (red, higher expression, blue lower expression). (C) Venn diagram showing the number of differential genes in pairwise analyses between; Nnt+/+ vs Nnt−/− (187), Nnt−/− vs NntBAC (157) and Nnt+/+ vs NntBAC (141). Genes at the intersection of the pairwise analyses Nnt+/+ vs Nnt−/− and Nnt−/− vs NntBAC represent genes that are modulated by Nnt levels (39 + Nnt) (Table 1 and Supplementary Tables 3, 4, 5, 6, 7 and 8).

Fig 5: Detoxification of free radicals in the mitochondria. NNT encodes a protein, integral to the inner mitochondrial membrane, which under normal physiological conditions uses energy from the mitochondrial proton gradient to generate high concentrations of NADPH. This is required for many processes in the cell including the supply of reductive power to a network of antioxidant enzymes, specifically the glutathione (GSH/GSSG) and thioredoxin (Trx(SH)2/TrxS2) systems, to allow the detoxification of H2O2. Manganese superoxide dismutase (MnSOD) converts O2·- into H2O2 and protects ROS-sensitive proteins from oxidative damage. H2O2 is then removed by glutathione peroxidases (e.g. GPX1) or peroxiredoxins (e.g. PRDX3) using GSH and Trx(SH)2 as co-factors. GSH and Trx(SH)2 can be regenerated by glutathione reductase (GR) and thioredoxin reductase-2 (TXNRD2), respectively, using the reducing power from NADPH. Without NNT, the production of NADPH is compromised, causing the mitochondria to become more sensitive to oxidative stress. Enzymes underlined in red are affected by one or more mutations in FGD patients.