Potassium-Hydroxide-Based Extraction of Nicotinamide Adenine Dinucleotides from Biological Samples Offers Accurate Assessment of Intracellular Redox Status

Abstract

The reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) is a primary electron donor for both antioxidant enzymes, such as glutathione reductase, and pro-oxidant enzymes, such as NADPH oxidases that produce reactive oxygen species (ROS) and nitric oxide synthases that generate nitric oxide which act as signaling molecules. Monitoring NADPH levels, NADPH/NADP⁺ ratio, and especially distinguishing from NADH, provides vital information about cellular redox status, energy generation, survival, lineage specification, and death pathway selection. NADPH detection is key to understanding metabolic reprogramming in cancer, aging, and cardiovascular, hormonal, neurodegenerative, and autoimmune diseases. Liquid chromatography combined with mass spectrometry (LC-MS) is crucial for NADPH detection in redox signaling because it offers the high sensitivity, specificity, and comprehensive profiling needed to quantify this vital but labile redox cofactor in complex biological samples. Using hepatoma cell lines, liver tissues, and primary hepatocytes from mice lacking transaldolase or nicotinamide nucleotide transhydrogenase, or having lupus, this study demonstrates that accurate measurement of NADPH depends on its preservation in reduced form which can be optimally achieved by extraction of metabolites in alkaline solution, such as 0.1 M potassium hydroxide (KOH) in comparison to 80% methanol (MeOH) alone or 40:40:20 methanol/acetonitrile/formic acid solution. While KOH extraction coupled with hydrophilic interaction liquid chromatography (HILIC) and mass spectrometry most reliably detects NADPH, NADP, NADH, NAD, polyamines, and polyols, MeOH extraction is best suited for detection of glutathione and overall discrimination between complex metabolite extracts. This study therefore supports performing parallel KOH and MeOH extractions to enable comprehensive metabolomic analysis of redox signaling.

Keywords: NAD; NADH; NADP; NADPH; alkaline extraction; formic acid extraction; hepatocellular carcinoma; hepatocyte; liver; lupus; methanol extraction; transaldolase.

Figure 1

KOH extraction is superior to MeOH extraction for detection of reduced and oxidized pyridine nucleotides in hepatocellular carcinoma (HCC) cell lines. Extraction with 80% methanol (MeOH) and 0.1 M KOH (KOH) were compared for detection of reduced and oxidized pyridine nucleotides in 1875 TALKO, vT, and C4 TAL WT hepatoma cells. NADPH/NADP⁺ (panel (A)), NADPH (panel (B)), NADP⁺ (panel (C)), NADPH, NADP⁺ combined (panel (D)), NADH/NAD⁺ (panel (E)), NADH (panel (F)), NAD⁺ (panel (G)), and NADH and NAD⁺ combined (panel (H)) were measured by LC-MS. p values reflect comparison based on five independent experiments. *, statistical analysis were done with 2-way ANOVA corrected for multiple comparisons with one exception that was done by t-test (*).

Figure 2

Effect of NNT deficiency on reduced and oxidized pyridine nucleotide content in the liver and heart. KOH and MeOH extractions were performed using liver (panel (A)) and heart tissues (panel (B)) from six, 10-month-old, age-matched male NNT/WT and NNT/Mut mice. p values reflect analysis with one-way ANOVA with Sidak correction for multiple comparisons.

Figure 3

KOH extraction is superior to FA or MeOH extraction for detection of reduced pyridine nucleotides from 1875 TALKO hepatocytes. Extraction with KOH, FA, and MeOH were compared for detection of reduced and oxidized pyridine nucleotides in 1875 TALKO. NADPH/NADP⁺ (panel (A)), NADPH (panel (B)), NADP⁺ (panel (C)), NADPH, NADP⁺ combined (panel (D)), NADH/NAD⁺ (panel (E)), NADH (panel (F)), NAD⁺ (panel (G)), and NADH and NAD⁺ combined (panel (H)) were measured by LC-MS. p values reflect comparison based on five independent experiments.

Figure 4

Comparative analyses of KOH, FA, and MeOH extractions for assessing the redox status of pyridine nucleotides and glutathione in hepatocytes from lupus-prone mice. The performance of metabolite extractions was compared in primary hepatocytes isolated from livers of five-month old, age-matched female lupus-prone mice carrying wild-type (B6.TC; n = 12) or constitutively active Rab4A alleles (B6.TC/Rab4A^Q72L, n = 18) and mice also lacking Rab4A in T cells (B6.TC/Rab4A^Q72L-KO, n = 9) [55]. NADPH/NADP⁺ (panel (A)), NADPH (panel (B)), NADP⁺ (panel (C)), NADH/NAD⁺ (panel (D)), NADH (panel (E)), NAD⁺ (panel (F)), GSH/GSSG (panel (G)), GSH (panel (H)), and GSSG (panel (I)) were measured by LC-MS. p values reflect analysis with one-way ANOVA with Sidak correction for multiple comparisons.

Figure 5

NADPH predominates over NADP in A549 cells. Metabolites were extracted from A549 human lung adenocarcinoma cells with 0.1 M KOH on ice and measured by HPLC using an ultraviolet detector [25]. (A), Detection of pyridine nucleotide standards at 259 nm absorbance by HPLC using photodiode array detector. Standard solutions, A549 extracts, and A549 cell extracts spiked with NADPH, NADP⁺, NADH, and NAD⁺ standards were analyzed separately. NADP⁺ was undetectable in 2 × 10⁷ cells. (B), Quantitation of NADPH, NADH, and NAD⁺ per 10⁶ A549 cells. Results and p values reflect 8 independent experiments.