The NAD metabolome of human cells - understanding a fundamental metabolic and signalling network

Biochemical processes in all organisms rely on a set of key molecules that carry important functions in metabolism and regulation. A number of these molecules are involved in bioenergetic processes that convert the ingredients of food into building blocks that can be used by the organism to develop and function normally. Moreover, recent research has shown that these molecules are also integral parts of control mechanisms governing important cellular and organismal events including gene expression, life span regulation, proliferation and cell-specific activities. Therefore, major diseases have now been associated with imbalances in the production and use of these biochemical regulators. NAD is among the most important metabolic and regulatory molecules known so far. It participates in a multitude of biochemical reactions and is a key player in a variety of cellular signalling pathways. In humans, NAD is synthesized from vitamin B3. Recent studies have shown that there are several alternative ways to produce NAD from vitamin B3 and its related compounds. Importantly, supplementation with novel B3 derivatives has impressive beneficial health effects in animals and humans. However, knowledge about the precise mechanisms and the regulation of NAD biosynthesis has only begun to emerge. This project aimed at deciphering the molecular mechanisms of NAD biosynthetic processes, in particular, their distribution in subcellular compartments. Another important goal was to develop model systems to study the cellular responses to induced NAD deficiency, thereby simulating aging- and disease-related processes. We addressed these challenges by performing comprehensive biochemical and cell biological analyses using state-of-the-art analytical instrumentation. The use of chemical tracers enabled quantitative assessment of NAD synthesis and NAD-dependent processes, thereby supplying the experimental data that are required to build a mathematical model of the entire NAD metabolome. The experimental part of the project was conducted at the University of Bergen and was supported by leading national and international experts who contributed with bioinformatics and experimental expertise and unique reagents. During the first phase of the project, we have developed genetically encoded sensors permitting to monitor NAD fluctuations in different parts of cells. These tools have enabled studies of organelle-specific generation of cellular NAD pools. We have extended this technology to study the dynamics of NAD concentrations in peroxisomes and the potential interaction with the mitochondrial NAD pool. This work was conducted in close collaboration with our partners in St. Petersburg. Another key methodology we have established is the accurate, mass spectrometry-based quantitative measurement of NAD and its metabolites in human cells. Thanks to our partner at the University of South Alabama, we could use isotope-labelled standard compounds for quantification. Moreover, these isotope-labelled precursors enabled the measurement of cellular NAD turnover rates. Using these unique technologies, we have discovered some fundamental principles of NAD metabolism in human cells including the interplay between mitochondrial and other cellular NAD pools and the adaptations under conditions of NAD depletion. Our results support the view that maintenance, in particular, of the mitochondrial NAD pool is essential for healthy cell physiology. Our project partners at the Arctic University Tromsø who are experienced bioinformaticians and systems biologists, have revealed exciting new insights into the evolution and dynamics of NAD metabolism. While most lower organisms preferentially use a rather inefficient pathway of NAD biosynthesis, animals, specifically mammals, have developed a highly active system of NAD biosynthesis and consumption that enables multiple NAD-dependent signalling pathways. That is, the diversification of NAD-dependent signalling processes has turned out as a major advantage in the evolution of mammals. Using the flux measurements conducted with isotope-labelled tracers, the mathematical model describing NAD metabolism developed at UiT could be refined and appropriately parameterized. This model can now simulate the dynamics of the NAD metabolome in human cells. Moreover, it can predict readjustments in this metabolic network that may be evoked by pathological processes or during aging. In conclusion, our studies have made major contributions to the development of highly advanced diagnostic and analytical tools for a medical and scientific field that has emerged as unusually promising for the treatment of age-related diseases. Combining experimental studies with mathematical modelling has enabled predictive simulations of this complex metabolic network. The application of these tools has led to fundamentally new insights into NAD biology, providing a basis for the development of targeted therapeutic approaches.

This project was conducted as an international collaboration of researchers in Bergen, Tromsø, St. Petersburg (Russia) and Mobile (USA). It has laid the ground for future joint projects. Our studies combined experimental (Bergen, St. Petersburg), computational (Tromsø) and chemical biology approaches (Mobile) demonstrating the importance and the mutual benefit from interdisciplinarity. The involvement of a wide range of competences was of great value for the early stage researchers who got an idea of the opportunities when working in a multidisciplinary team. Our results will have an impact on the field of NAD biology and for medical applications. For example, the techniques to quantify NAD metabolites will be used in clinical trials of NAD supplementation (NAD-PARK, NO-PARK) conducted at the clinical research center NeuroSysMed in Bergen. The mathematical model will be further developed to predict the potential outcome of therapeutic interventions.

NAD is an essential redox carrier and NAD-mediated signaling regulates fundamental processes including the circadian clock, gene expression, DNA repair, metabolic pathways and apoptosis. Perturbations in this network can cause pathologies such as diabetes, cancer and neurodegenerative disorders. Research has so far focused on individual routes, barely considering crosstalk, e.g. between synthesis and signaling. Based on the hypothesis that NAD metabolism represents an integrated network, this project will lay the ground to study the dynamics of this network as a whole. Two fundamental issues have remained that need to be resolved to understand the NAD metabolome. First, the biosynthetic route to mitochondrial NAD, likely the largest subcellular pool, is still unknown. Recent studies suggest that NMN is taken up and converted to NAD in the matrix. Using biochemical, genetic and pharmacological approaches, we will identify and characterize the mitochondrial NMN carrier. Successful completion of this task would be a huge step forward towards understanding NAD metabolism. Second, a peroxisomal activity might constitute a vital buffer for NMN to feed subcellular NAD pools. NUDT12 is a peroxisomal enzyme that cleaves NAD to NMN (and AMP), which can be released into the cytosol. Thus, characterization of NUDT12 and its influence on subcellular NAD pools is critical. Many pathways of NAD synthesis and signaling have been well characterized providing a large data reservoir to generate a comprehensive mathematical model. Availability of metabolic flux rates would profoundly strengthen these efforts. Therefore, using LC-MS- and NMR- based metabolomics we will establish flux rates for different branches of the network. These measurements will allow parameterization of an initial model of human NAD metabolism. Leading national and international scientists will support the project and contribute research tools, analytical methods and expertise in modeling approaches.