The nanomolar sensing of nicotinamide adenine dinucleotide in human plasma using a cycling assay in albumin modified simulated body fluids

Abstract

Nicotinamide adenine dinucleotide (NAD), a prominent member of the pyridine nucleotide family, plays a pivotal role in cell-oxidation protection, DNA repair, cell signalling and central metabolic pathways, such as beta oxidation, glycolysis and the citric acid cycle. In particular, extracellular NAD+ has recently been demonstrated to moderate pathogenesis of multiple systemic diseases as well as aging. Herein we present an assaying method, that serves to quantify extracellular NAD+ in human heparinised plasma and exhibits a sensitivity ranging from the low micromolar into the low nanomolar domain. The assay achieves the quantification of extracellular NAD+ by means of a two-step enzymatic cycling reaction, based on alcohol dehydrogenase. An albumin modified revised simulated body fluid was employed as standard matrix in order to optimise enzymatic activity and enhance the linear behaviour and sensitivity of the method. In addition, we evaluated assay linearity, reproducibility and confirmed long-term storage stability of extracellular NAD+ in frozen human heparinised plasma. In summary, our findings pose a novel standardised method suitable for high throughput screenings of extracellular NAD+ levels in human heparinised plasma, paving the way for new clinical discovery studies.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Overview of the spectroscopic techniques. (a) Comparison of the assay dynamics in different standard matrices utilising the absorbance unit (AU) readings obtained from DEPC water, revised simulated body fluid (r-SBF), revised simulated body fluid adjusted with albumin (r-SBFA) and human plasma, all spiked with 50 μL of 376.8 nM β-NAD standard (S2). (b) Fluorescence scan of the autofluorescence of the Master Mix (MM) and a NAD+ sample as prepared in the actual assay reaction, scanned from λex1 = 280 nm − λex2 = 850 nm in steps of Δλ = 2 nm. The  resorufin signal is indicated at λ = 590 nm in the detail view. (c) Enzyme dependent assay kinetics for β-NAD standards S1–S6 including a heparinised plasma sample and the blank of a given run using the fluorimetric method. (d) Enzyme dependent assay kinetics for β-NAD standards S1–S6 including a heparinised plasma sample and the blank of a given run using the colorimetric method. (e) Fluorescence scan of the autofluorescence of the Master Mix (MM) and β-NAD standard S1 as prepared in the actual assay reaction (S1 + r-SBFA + MM), scanned from λex1 = 280 nm − λex2 = 850 nm in steps of Δλ = 2 nm. (f) Visual representation of the resorufin quenching effect that occurs when a minuscule amount of heparinised plasma sample is added to the β-NAD standard S1.

Figure 2

NAD+cycling principle. Schematic representation of the enzymatic alcohol dehydrogenae (ADH) cycling principle used to measure eNAD+, involving phenazine methosulfate (PMS) as the primary and 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) as the secondary redox indicator dye. The method was inspired by the findings of Rhodes et al. and adjusted with insights from the research of Zhu et al. and O’Reilly et al..

Figure 3

NAD+assay and storage. (a) Enzyme dependent assay kinetics for β-NAD standards S1–S6 and the blank of a given run during the assay validation: S1 (753.6 nM), S2 (376.8 nM), S3 (188.4 nM), S4 (94.2 nM), S5 (47.1 nM), S6 (23.5 nM), whilst standards S7 (11.8 nM) and S8 (5.6 nM) were omitted. (b) Relative reaction velocities (vR) of standards S1–S8 during min 5–40 of the assay reaction. n = 8. (c) Relative reaction accelerations (aR) of standards S1-S8 during min 5–40 of the assay reaction. n = 8. (d) Measured NAD+ concentrations in human heparinised plasma stored at −80 °C for a given amount of time. No statistical significance was found when comparing the measured timepoints to the baseline (d = 0). n = 6. Statistics: Two-tailed, unpaired t-test with the confidence limits of CL = 99%, as well as a two-way ANOVA (without repeated measures) adjusted with Tukey’s multiple comparisons test featuring CL = 99%. A significance level of p < 0.01 was applied to reject the null hypothesis. All error bars are given in terms of ±SD.

Figure 4

NAD+assay linearity and enzyme kinetics. (a) The ratio of the slope (vR) of two sequential β-NAD standards, (n) and neighbour (n + 1), is given by γn between min 5–25 of the assay reaction. n = 8. (b) The Lineweaver-Burk Plot constructed for averaged, blank corrected, vR data from eight Runs including S1–S6. n = 8. Statistics: All error bars are given in terms of ±SD.

Figure 5

Albumin dependency of the assay. (a) Comparison of the calibration working curves not through the origin (nTTO, dotted) and to through the origin (TTO, solid) with respect to a varying albumin concentration in the β-NAD standards S1 (753.6 nM) through S6 (23.5 nM) of 0 g/L for the lowermost trace, 10 g/L, 20 g/L, 30 g/L and 40 g/L for the uppermost trace. n = 3. (b) Comparison of the slopes, m, of the calibration working curves obtained by linear regression analyses nTTO (dark) and to TTO (light) for different albumin concentrations. n = 3. (c) Variance of the y-axis intercept, vn, of the regression lines nTTO with respect to varying albumin concentrations. n = 3. (d) Predicted pooled human heparinised sample eNAD+ concentrations using calibration working curves constructed nTTO (dark) and to TTO (light) with respect to varying albumin concentrations. A regression fit over the physiological range of albumin concentrations yielded y = −2.84 xnML/g + 236.5 nM (R2 = 0.9785) for the nTTO and y′ = −0.34 xnML/g + 194 nM (R2 = 0.7706) for the TTO approach. n = 3. Statistics: Two-tailed, unpaired t-test with the confidence limits of CL = 99%, where a significance level of p < 0.01 was applied to reject the null hypothesis. **p < 0.01. All error bars are given in terms of ±SD. The dotted line represents the physiological range of albumin concentrations.

Figure 6

NAD+assay reproducibility. (a) Mean relative reaction velocities (vR) for β-NAD standards S1 (753.6 nM) through S8 (5.6 nM) and the blank measured between min 5–25 of the assay reaction. n = 8. (b) The ratio (εn) of standard deviations (σn), to the difference in relative reaction velocity (vR) of two sequential β-NAD standards, (n) and neighbour (n + 1), represented by ψn. n = 8. (c) Calibration working curve constructed from the average slopes (vR) of β-NAD standards S1 (753.6 nM) through S6 (23.5 nM), obtained from the eight-fold assay repetition. n = 8. (d) Blood samples were taken from patients scheduled to undergo hernioplasty and analysed for their eNAD+ concentration. Statistics: n = 10. Among the analysed, there were 8 male and 2 female patients, where the mean age averaged (52.1 ± 15.8) years, ranging from 29 to 75 years. The underlying diseases were inguinal hernia (n = 7), epigastric hernia (n = 1), hiatal hernia (n = 1) and umbilical hernia (n = 1). The overall average is displayed. Statistics: Two-way ANOVA (without repeated measures) adjusted with Tukey’s multiple comparisons test with CL = 99%. A significance level of p < 0.01 was applied to reject the null hypothesis. ****p < 0.0001. All error bars are given in terms of ±SD.