Meeting ReportsMeeting Report

Emerging Roles of NAD+ and Its Metabolites in Cell Signaling

See allHide authors and affiliations

Science Signaling  10 Feb 2009:
Vol. 2, Issue 57, pp. mr1
DOI: 10.1126/scisignal.257mr1
A report on the NAD2008 symposium, Hamburg, Germany, 14 to 17 September 2008.


Nicotinamide adenine dinucleotide (NAD+) is the universal currency of energy metabolism and electron transfer. Recent studies indicate that apart from its role as a coenzyme, NAD+ and its metabolites also function in cell signaling pathways; for example, they are substrates for nucleotide-metabolizing enzymes and ligands for extra- and intracellular receptors and ion channels. Moreover, the NAD+ and NAD+ phosphate metabolites adenosine 5′-diphosphoribose (ADP-ribose), cyclic ADP-ribose, and nicotinic acid adenine dinucleotide phosphate (NAADP) have emerged as key second messengers in Ca2+ signaling. A symposium in Hamburg, Germany, brought together 120 researchers from various fields, who were all engaged in the molecular characterization of the key players of NAD+ signaling (

Nicotinamide adenine dinucleotide (NAD+) is a key metabolite in energy metabolism and electron transfer. Most organisms synthesize NAD+ from the vitamin B3 (niacin), which includes both nicotinic acid and nicotinamide. Although the redox conversions of NAD+ are reversible, recent discoveries indicate that it is constantly degraded in various signaling pathways (Fig. 1) (13). NAD glycohydrolases, now also known as ADP-ribosyl cyclases (ARCs), convert NAD+ and NAD+ phosphate (NADP) to adenosine 5′-diphosphoribose (ADP-ribose or ADPR), cyclic ADP-ribose (cADPR), and nicotinic acid adenine dinucleotide phosphate (NAADP), all of which have emerged as key second messengers in Ca2+ signaling by acting as ligands for various Ca2+ channels (4, 5). ADP-ribosyl transferases (ARTs) attach the ADP-ribose moiety of NAD+ to specific acceptors, generally proteins, thereby regulating the biological activity of the target (6, 7). Similarly, poly(ADP-ribose) polymerases (PARPs) generate polymers of ADP-ribose from multiple NAD+ molecules and affix them to target proteins (8). Both mono– and poly–ADP-ribosylation are reversible, and the demodification reactions are catalyzed by ADP-ribosyl hydrolases (ARHs) and poly(ADP-ribose) glycohydrolases (PARGs), respectively (6). Furthermore, NAD+-dependent protein deacetylases transfer the acetyl group from lysine residues onto ADP-ribose derived from NAD+. This reaction is carried out by several sirtuins, a protein family whose founding member is the yeast silence information regulator 2 (SIR2) (9), although some sirtuins have been shown to exhibit ADP-ribosyl transferase activity in addition to deacetylase activity. A symposium from 14 to 17 September 2008 sponsored by the Deutsche Forschungsgemeinschaft brought together 120 researchers in Hamburg, Germany, including 42 invited lecturers and 60 poster presenters from various fields, who were all engaged in the molecular characterization of the key players of NAD+ signaling.

Fig. 1

Nicotinamide adenine dinucleotide (NAD+) and its metabolites in cell signaling. ADP-ribosyl transferases (ARTs), poly(ADP-ribose) polymerases (PARPs), sirtuins (SIRs), and ADP-ribosyl cyclases (ARCs) use the high energy released by breaking the glycosidic bond between nicotinamide and ADP-ribose in NAD+ (red line). ARTs, PARPs, and SIRs transfer the ADP-ribose (ADPR) moiety onto acceptors, such as proteins, ADPR polymers, DNA, sugars, small molecules, and water, whereas ADP-ribosyl hydrolases (ARHs) and poly(ADPR) glycohydrolases (PARGs) remove the ADPR moiety from acceptors. ARCs catalyze multiple reactions, including hydrolysis of NAD+ to ADPR, as well as the cyclization of ADPR or the substitution of nicotinamide in NAD phosphate [NADP+; which is generated by the phosphorylation of NAD+ by NAD+ kinase (NADK)] with nicotinic acid, thereby generating the Ca2+ mobilizing second messengers cyclic ADPR (cADPR) and nicotinic acid adenine dinucleotide phosphate (NAADP), respectively. NAD+ can act as a ligand for the P2Y11 purinoreceptor, ADPR- and poly(ADP-ribosyl)ated proteins as ligands for macro domain-containing proteins, cADPR as a ligand for the ryanodine receptor (RyR), and NAADP as a ligand for a putative receptor in acidic stores. The NAD2008 Hamburg meeting focused on ARTs, ARCs, and Ca2+ mobilization (highlighted in yellow).

New discoveries were presented in sessions focusing on NAD+ homeostasis, Ca2+ signaling functions of NAD+ metabolites, ADP-ribosyl cyclases, and ADP-ribosyl transferases. Mathias Ziegler (Bergen, Norway) presented new insights into NAD+ synthesis, its compartmentation, and relationships to NAD-mediated signaling (10). Nobumasa Hara (Izumo, Japan) highlighted the importance of nicotinic acid as a NAD+ precursor in humans (11). NAD+ and its metabolites can be released into the extracellular environment, according to evidence presented by Violeta Mutafova-Yambolieva (Reno, USA) (12). Eduardo Lazarowski (Chapel Hill, USA) suggested that a possible release mechanism could involve exocytic vesicles (13). Leukocytes appear to be particularly well equipped with NAD+ sensors, including the ARCs CD38 and CD157 (CD is cluster of differentiation antigen), the ART CD296, and the NAD-binding purinergic receptor P2Y11 (2, 14, 15). Antonio de Flora (Genova, Italy) highlighted how these different receptors and ectoenzymes (enzymes whose catalytic domain is located on the external side of the cell membrane) might functionally interact (15). Sirpa Jalkanen (Turku, Finland) presented experiments demonstrating the role of ectoenzymes in the migration of leukocytes (14). CD38 is a disease marker for human leukemias and myeloma and, in particular, a reliable negative prognostic marker in chronic lymphocytic leukemia (CLL). On the basis of these findings, Fabio Malavasi (Torino, Italy) proposed therapeutic strategies targeting CD38 in CLL (3). By using a CD38−/− mouse model, Frances Lund (Rochester, USA) demonstrated the importance of CD38 in NAD homeostasis and its influence on PARP-1 activity (16). Friedrich Haag (Hamburg, Germany) presented the identification of target proteins of ART2-mediated ADP-ribosylation (2). A differentiation-dependent role of NAD in the regulation of calcium elevation in monocytes was discussed by Sunna Hauschildt (Leipzig, Germany) (17).

CD38-related enzymes generate NAADP and cADPR in intracellular vesicles and on the cell surface (which are transported across the cell membrane into cells) that affect Ca2+ signaling in various settings. Antony Galione (Oxford, UK) presented the purification and identification of novel NAADP-binding proteins from sea urchin eggs (18). Luigia Santella (Napoli, Italy) showed that NAADP peaks before the Ca2+ wave generated by sperm fertilizing starfish eggs, indicating that NAADP is responsible for the initiation of the sperm-induced Ca2+ signal (19). Timothy Walseth (Minneapolis, USA) reported on the putative role of cADPR in causing hyperresponsiveness in inflamed airway smooth muscle cells (20). Andreas Guse (Hamburg, Germany) presented evidence supporting a role for the NAADP pathway in the activation of effector CD4+ T cells (21). Jaime Sancho (Armilla, Spain) demonstrated that CD38 localizes to the immunological synapse during the interaction of T cells with antigen-presenting cells (22). Haruhiro Higashida (Kanazawa, Japan) showed altered social behavior in CD38 knockout mice, linking reduced hypothalamic oxytocin release to decreased intracellular Ca2+ levels (23). A special feature of NAD 2008 was to bring together chemists interested in NAD analogs with cell biologists. Specific NAADP antagonists that hold promise as therapeutic tools for combating diverse human diseases were presented by Barry Potter (Bath, UK), Grant Churchill (Oxford, UK), and Andreas Guse (24). Barry Potter, Francis Schuber (Strasbourg, France), and Li-he Zhang (Beijing, China) reported on specific cADPR antagonists that have already fuelled research in different areas (25, 26). Francis Schuber introduced a high-throughput screening test suitable for an automated search of Schistosoma mansoni NAD-catabolizing enzyme, which is located on the surface of S. mansoni and could be a target for treatment of schistosomiasis (27). Santina Bruzzone (Genova, Italy) described how abscisic acid, which was originally identified as a plant stress hormone, is an endogenous stimulator of cADPR-mediated Ca2+ signaling and insulin release in human pancreatic β cells (28). The putative molecular target of NAADP is the subject of controversy: Martin Hohenegger (Vienna, Austria) and Andreas Guse presented evidence in favor of the type 1 ryanodine receptor as the NAADP receptor, whereas Antony Galione and Sandip Patel (London, UK) presented evidence in favor of a new class of channels localized to acidic stores (29).

NAD+-dependent ARTs posttranslationally attach ADP-ribose to specific amino acid residues in target proteins, which can have activating or inhibitory effects. Several potent bacterial toxins are ARTs. Klaus Aktories (Freiburg, Germany) described a mosquitocidal toxin (MTX) from Bacillus sphaericus, which ADP-ribosylates arginine residues in a 55-kD chaperone protein and in other proteins in insect cell lysates (30). Joseph Barbieri (Milwaukee, USA) showed that another promiscuous arginine-specific ADP-ribosyl transferase [Pseudomonas aeruginosa exoenzyme S (exoS)] ADP-ribosylates numerous targets in mammalian host cells, resulting in the inhibition of cell vesicle trafficking (31). Mariana Margarit (Ridgefield, USA) reported that ADP-ribosylation of cytoskeletal actin at arginine 177 by the Salmonella virulence protein SpvB does not lead to dramatic conformational changes in actin but rather blocks actin polymerization through steric disruption of contacts between filaments (32). Friedrich Koch-Nolte (Hamburg, Germany) presented results demonstrating that the toxin-related murine membrane-bound ectoenzyme ART2.2 ADP-ribosylates the P2X7 purinergic receptor at arginine 125, thereby activating the channel function of the receptor (33). Michel Seman (Rouen, France) demonstrated that activation of P2X7 by ADP-ribosylation induces apoptosis of regulatory T cells and plays an important role in shaping the repertoire of these cells (34). David Serreze (Bar Harbor, USA) presented studies in which ART2.2, CD38, or both were genetically ablated in the nonobese diabetic (NOD) mouse model of insulin-dependent diabetes mellitus. The results indicate that the elevated levels of extracellular NAD in the CD38 knockout mice dramatically enhance disease progression in an ART2.2-dependent manner (35). Michael Hottiger (Zürich, Switzerland) provided evidence that poly(ADP-ribose) polymerase (PARP-1) physically interacts with nuclear factor κB (NF-κB) and affects transcriptional regulation by NF-κB in primary macrophages (36). In his talk dedicated to Helmuth Hilz, professor emeritus in the Institute of Biochemistry and Molecular Biology, University-Hospital Hamburg-Eppendorf, Germany, and one of the pioneers of ADP-ribosylation research, Myron Jacobson (Tucson, USA) presented data indicating that differential splicing of trancripts for PARG can target PARG to multiple cell compartments, including mitochondria, and proposed that poly(ADP-ribose) signaling should no longer be considered only a nuclear event (37). John Denu (Madison, USA) discussed how some members of the sirtuin family of NAD-dependent protein deacetylases (e.g., Trypanosoma cruzi Sir2) can also catalyze ADP-ribosylation of histones (38). Keiji Wakabayashi (Tokyo, Japan) presented evidence that butterflies and shellfish produce toxin-related ARTs that ADP-ribosylate DNA instead of proteins (39). Fernando Goldbaum (Buenos Aires, Argentina) and Friedrich Koch-Nolte described new tools for blocking ART functions, which they generated from single-chain antibodies isolated from ART-immunized llamas (40). Daniela Corda (Chieti, Italy) presented evidence that the brefeldin A–induced ADP-ribosylation of the Golgi-regulator C-terminal binding protein (CtBP) [also known as brefeldin A–ADP ribosylated substrate (BARS)] occurs through an ART-independent mechanism, which abolishes the fissioning activity of BARS on the Golgi apparatus (41). ARHs reverse protein ADP-ribosylation, and Joel Moss (Bethesda, USA) described how mice lacking ARH1, the ARH isoform that catalyzes de–ADP-ribosylation of proteins that are ADP-ribosylated on arginine residues, are more likely than their wild-type counterparts to accumulate fluid in their intestines in response to the arginine-specific ART cholera toxin (42).

New mechanistic insights into NAD+-mediated signaling were provided by recently elucidated crystal structures (Fig. 2). Hon Cheung Lee (Hong Kong, China) presented the crystal structure of CD38 in complex with NAD+ and ADPR, revealing that polar interactions between the catalytic glutamate 226 and the substrate 2′,3′-OH groups are essential for initiating catalysis, whereas the reorientation of the product ADP-ribose could explain the product inhibition effect (43). Rod Merrill (Guelph, Canada) presented the first crystal structure of a complex of an ART [Pseudomonas exotoxin A (ETA)], its substrate NAD+, and its target protein elongation factor 2 (EF2), revealing the movement of two loops in the ART and providing glimpses of the likely transition state (44). Klaus Aktories (Freiburg, Germany) described the crystal structure of MTX, revealing allosteric inhibition of the NAD-binding crevice by a linker peptide (residues 265 to 285) that connects the N-terminal catalytic domain with the C-terminal ricinlike receptor binding domains. This peptide folds back into and sterically blocks the active site crevice (30). Manfred Weiss (Hamburg, Germany) provided the first crystal structure of ARH3 catalyzing the de–ADP-ribosylation of poly(ADP-ribosyl)ated target proteins. Docking studies indicated a possible binding mode of ADPR and a role for two Mg2+ ions at the bottom of the active site crevice in initiating catalysis (45). Fernando Bazan (San Francisco, USA) presented the results of sensitive fold recognition, docking, and modeling studies, revealing that the fold of human PARG is unrelated to the helical metalloenzyme structure of ARHs but instead features an α+β fold reminiscent of ADP-ribose monophosphatases, such as the Archaeoglobus fulgidus enyzme Af1521. As reported by Andreas Ladurner (Heidelberg, Germany), the crystal structure of Af1512 resembles that of the macro domain of a human histone protein (MacroH2A1), which seems to have lost enzymatic activity but retains high binding affinity to ADP-ribosylated and poly(ADP-ribosyl)ated proteins (46). Maria Di Girolamo (Chieti, Italy) reported the successful use of a catalytically inactive mutant of the Af1521 macro domain as a tool for purifying and identifying ADP-ribosylated proteins (47).

Fig. 2

Crystal structures of key players in NAD+ signaling. (A) Human CD38, an ARC, in complex with NAD [Protein Data Bank (PDB) identification (ID) code 2i65]. (B) Rat CD296, an ART, in complex with NAD (PDB ID 1og3). (C) Thermotoga maritima Sir2, a sirtuin (Sir) or NAD+-dependent deacetylase, in complex with NAD (PDB ID 2h4h). (D) Human ARH3, with two magnesium ions (in purple) (PDB ID 2foz), modeled with ADPR. (E) Archaeoglobus fulgidus Af1512, a macro domain in complex with ADPR (PDB ID 2bfq). (F) Yeast elongation factor 2 (EF2) with ADPR covalently attached to diphthamide, a modified histidine residue, at position 699 (PDB ID 1u2r).

The Hamburg symposium followed a rich tradition of meetings on the key enzymes of NAD metabolism, whose focus traditionally reflects the interests of the local organizers (48, 49). Although NAD 2008 featured sessions on ARTs and Ca2+ signaling, reflecting the interests of the Haag, Koch-Nolte, and Guse laboratories, the conference covered many aspects of NAD+ and NADP+ metabolism and signaling processes. The meeting was co-organized by Frances Lund and Mathias Ziegler. We hope that this symposium will represent the kick-off for future conferences that will bring together scientists interested in various aspects of this dynamic field. A followup meeting organized by Frances Lund and Timothy Walseth is planned for 2010 in the United States.


50.The symposium was supported by grants from the Deutsche Forschungsgemeinschaft and the Behörde für Wissenschaft und Forschung (Hamburg, Germany). We thank B. Rissiek for design and maintenance of the conference Web site ( and G. Kulms for excellent secretarial assistance.

References and Notes

View Abstract

Navigate This Article