Meta‑analysis Of NAD(P)(H) Quantifcation Results Exhibits Variability Across Mammalian Tissues Ⅱ

The effect of premortem versus post‑mortem tissue collection on NAD(P)(H) levels.

We predicted that there may be some observable difference in NAD(P)(H) redox state that is dependent on tissue harvesting procedures. To answer this question, we examined the effect of pre- versus post-mortem tissue collection procedures on NAD(P)(H) redox status. For this analysis, we compared values from rat liver tissues given the larger number of studies defining their sacrifice protocol. When disclosed, all the reviewed studies stated that the tissue samples were kept cold and immediately extracted on ice or frozen at − 80 °C prior to NAD(P)(H) measurements. This analysis determined that there was no significant difference in NAD+ concentrations in rat liver (Fig. 6a) between tissues extracted before or after euthanasia. However, the reported NADH levels in rat liver are significantly higher in post-mortem samples (Fig. 6b). No diferences in total NAD(H) levels (Fig. 6c) were observed but the NAD+/NADH ratio was lower in tissues collected following euthanasia (Fig. 6d), which is consistent with the variation observed in NADH levels. NADP+, NADPH, and the NADP+/NADPH ratio results in rat liver grouped by tissue harvest time point are shown in Supplementary Fig. 9; however, the insufficient number of post-mortem samples does not allow us to interpret the effect of tissue harvest timepoint on NADP(H) levels. In mouse liver, there was no difference between pre-and post-mortem NAD+ levels, however, the number of reporting studies was limiting (Fig. 6e). No NADH, NADP+, or NADPH post-mortem data in mouse liver was available for comparison.

Click Here To Get Hebrs Cistanche 

Discussion 

Many recent studies have demonstrated that NAD(P)(H) regulation is essential for cellular homeostasis and redox status. Data obtained from rodents have led to studies examining human blood NAD+ levels, as a less invasive indicator of whole-body NAD+ levels. Recent clinical studies demonstrated that NAD+ levels were reduced in disease states23,47 and elevated following NAD+ boosting strategies23,44,48,49. We performed a meta-analysis of all studies published between 1946 and June 20, 2021, that contained quantitative data for NAD(P)(H) levels in mammalian species with a special focus on mice, rats, and humans as they represent the most studied species in the biomedical field. This analysis was done in part to determine the mean standard level of NAD(P)(H) concentrations in normal mammalian tissues given the well-known variance in these measures across studies. We hope that these data can be used to stimulate standardized protocols in the field of NAD+ research and to promote the use of NAD(P)(H) levels and redox ratios as reliable biomarkers for disease and treatment regimens. 

Despite considerable variability in measures across rodent tissues, this meta-analysis displayed similar mean NAD+ levels across tissues with some exceptions. Interestingly, mouse skeletal muscle exhibited lower median NAD+ levels (Fig. 2a) than other highly metabolic tissues (i.e., liver, kidney, heart, and brain). This may indicate different NAD(H) redox statuses in skeletal muscle or may point to greater issues with metabolite extraction in fibrous tissues. However, these observations could not be verified in humans due to a lack of sampled tissues beyond blood and muscle.

There are many potential factors that could affect the accuracy of physiological NAD(P)(H) measurements in tissue samples. When reported, all studies described tissues as being harvested and directly submerged into liquid nitrogen, or kept on ice before freezing or processing in order to preserve the heat-sensitive fractions (NAD+ and NADP+). Te effect of pH on the various NAD(P)(H) fractions, such as the acid-labile nature of the reduced NAD(P)H forms, has led to the use of pH-neutral extraction solvents by the majority of the studies. However, given the rapid interconversion of the reduced and oxidized metabolites and the effect of anoxia on NAD(H)50–53, it is also important to ensure rapid extraction of tissue metabolites and the appropriate procedures for quenching cellular NAD(P)(H)-redox/consumption enzymes, such as through deproteinization. Along these lines, our analysis of pre- versus post-mortem tissue collection indicates that post-mortem analysis favors the reduction of NAD+. The reduction of tissue NAD+ has been previously described in vivo in the brain, heart, and liver of rats exposed to an extended 2.5-h hypoxic atmosphere54, an environment that may be partially recapitulated by extended periods of post-sacrifice tissue collection before snap freezing. This may indicate that extracting tissues while under anesthetic or prioritizing the immediate harvest of tissues intended to be used for metabolite analysis post-sacrifice, should occur to obtain representative NAD(P)(H) measurements. 

There is also the potential for various quantification methods, each having different limitations, to contribute to the overall intra-study variability of NAD(P)(H) measurements. In the last two decades, the most frequently used methods were enzyme cycling assays, LC–MS, and HPLC. Each of these methods are affected by metabolite extraction techniques, quantification parameters, and the implementation of proper quality controls. Using LC–MS, Lu et al. have shown that the interconversion between reduced and oxidized forms of NAD(P)(H) metabolites occurs at different rates in various extraction buffers or solvents with the acetonitrile:methanol: water with 0.1 M formic acid mixture yielding the highest recovery with the least interconversions31. However, when using LC–MS this interconversion can be monitored between individual samples of a study by spiking with internal controls such as NAD(P)(H) isotopes, which is an advantage of the LC–MS technique over that of HPLC and enzyme cycling assays31,33,55. Nonetheless, even with well-controlled LC–MS techniques, various factors can interfere with the measured signal, including matrix effects and variations in ionization efficiency. For example, some studies use 13C-labeled yeast extracts as internal standards in LC–MS-based metabolomics. However, spiking the extracted sample metabolite matrix with a 13C-labeled yeast extract metabolite matrix can have various consequences, such as ion suppression, which reduces the signal of labeled and/or unlabeled metabolites, and can lead to errors in the absolute quantification if not detected by a thorough quality assessment56. Also, although LC–MS techniques theoretically allow for the measurement of multiple NAD(P)(H) metabolites in one experiment, there are still limitations since optimal dilutions must be run if the metabolites of interest have large diferences in concentrations. Thus, despite their advantages over the simpler enzyme cycling assays, the complexity of LC–MS methods impose the necessity for thorough optimization. Recently new analytical methods, such as, NAD+ biosensors and imaging-based mass spectrometry, have been developed but there is still insufficient quantitative data generated from these techniques to include in a meta-analysis. Although, one study included in our meta-analysis used a paper-based bioluminescent biosensor to measure NAD+ levels in mouse liver and other sample types57. This study was included in the bioluminescent assays group due to the similar method of detection (Fig. 1a).

Methods

Data extraction. 

Numerical data for NAD(P)(H) metabolites were either obtained directly from the articles or extracted from article figures using a semi-automatic data extraction software (WebPlotDigitizer58). Additionally, where applicable, the tissue sampling time points relative to sacrifice and/or tissue/blood sampling methods (i.e. method of anesthesia and/or euthanasia, tissue handling, and storage temperature) were recorded. For animal models, the species, characteristics (e.g. strain, genotype, age, and weight), and environmental conditions (e.g. sleep cycle, type of diet, feeding frequency, and feeding state relative to tissue sampling) were recorded. Additionally, for all studies, treatments (e.g. pharmacological treatments, surgeries, irradiation, tumor induction, and other procedures applied to the study subjects) and all study group information (e.g. treatment or disease groups and corresponding controls) were extracted. Finally, the method of NAD(P)(H) quantification method (e.g. enzymatic assays, mass spectrometry-based, HPLC, NMR, bioluminescence) and publication specifics (i.e. publication title, reference code [DOI, Medline UI, PMID] or hyperlink and year of publication) were noted (Supplementary Material 2).

Data analysis. Prior to analysis, all concentrations were converted to nmol/g of tissue or nmol/g of proteins for tissue samples or nmol/ml for blood fractions. Due to the low number of results normalized to protein content, we only show results normalized to tissue weight and blood volume. Statistical analysis was performed with GraphPad Prism version 9.3.1 (GraphPad Sofware Inc., San Diego, California, USA, Data availability All data extracted and analyzed during this study are included in this published article and its supplementary information fles. Received: 29 April 2022; Accepted: 7 February 2023

References

10. Love, N. R. et al. NAD kinase controls animal NADP biosynthesis and is modulated via evolutionarily divergent calmodulin-dependent mechanisms. Proc. Natl. Acad. Sci. 112, 1386–1391 (2015).

Ask for more:

Email:wallence.suen@wecistanche.com whatsapp:+86 15292862950