NAADP (nicotinic acid–adenine dinucleotide phosphate) is a derivative of NADP (nicotinamide–adenine dinucleotide phosphate), which differs by the presence of a nicotinic acid instead of a nicotinamide moiety. This small structural difference makes NAADP one of the most powerful second messengers known, able to mobilize intracellular Ca2+ in a wide range of cellular models, ranging from invertebrates to mammals. Despite this, our understanding of NAADP homoeostasis, metabolism and physiological action is still limited. A new report by Vasudevan and colleagues in this issue of the Biochemical Journal provides important new data by describing a new synthetic activity in sperm cells which may turn out to represent the most physiologically relevant route to this second messenger.
NAADP (nicotinic acid–adenine dinucleotide phosphate) is one of the most powerful second messengers known, able to mobilize intracellular Ca2+ in a wide range of cellular models, ranging from invertebrates to mammals . It activates a Ca2+-release pathway distinct from those activated by IP3 (inositol 1,4,5-trisphosphate) and cADPR (cADP-ribose)  and may engage a distinct non-ER (endoplasmic reticulum) Ca2+ pool. The latter appears to be acidic, and, in sea urchin egg homogenates, ‘reserve pool’ lysosome-like granules have been identified as a Ca2+ store which is sensitive only to NAADP . Similarly, NAADP, as well as cADPR, but not IP3, releases Ca2+ from a VAMP2 (vesicle-associated membrane protein 2)-positive compartment, which is likely to correspond to insulin granules in pancreatic β-cells .
NAADP has up to now been thought to be synthesized by one or more members of the ADP-ribosyl cyclase family, which includes a cyclase from Aplysia californica (the first characterized)  and the cell-surface antigens CD38  and CD157 , in response to undefined signals. The reaction leading to the formation of NAADP is known as base-exchange, where, utilizing NADP (nicotinamide–adenine dinucleotide phosphate) as substrate and at acidic pH, the enzyme catalyses the exchange of nicotinamide with nicotinic acid  (Figure 1).
Members of the family of the ADP-ribosyl cyclases catalyse two main reactions
The above enzymes are defined as multifunctional because they are able to catalyse distinct reactions , i.e. both base-exchange (to generate NAADP), cyclization (to generate cADPR) (Figure 1) and product hydrolysis. These reactions appear to be differentially regulated, and show distinct responses to pH (cADPR is preferentially synthesized over NAADP at neutral pH, whereas the base-exchange reaction predominates at acidic pH ). Moreover, in pancreatic β-cells, millimolar concentrations of ATP inhibit the hydrolytic activity of CD38 towards cADPR , whereas Zn2+ ions are able to enhance CD38 cyclic activity . In sea urchins, cGMP stimulates cADPR synthesis , whereas cAMP stimulates NAADP synthesis . Finally, a recent study  suggests that there are enzymes other than CD38 in mammalian tissues that are able to produce NAADP via mechanisms different from the base-exchange reaction (deamination of NADP or phosphorylation of NAAD).
A new study by Vasudevan and colleagues  reported in this issue of the Biochemical Journal suggests the presence, in sea urchin sperm, of a new NAADP synthase. By comparing the conversion of NADP into NAADP in intact sperm with that in permeabilized sperm, the authors demonstrate that the enzyme (in common with members of the family of ADP-ribosyl cyclases) is situated largely on the plasma membrane and possesses an activity which is pH-dependent. The important finding of the new study is that the enzyme displays only base-exchange, and not cyclase, activity (Figure 1), suggesting that it may represent the ‘genuine’ NAADP synthetic enzyme, at least in sea urchin sperm. Secondly, the finding that the NAADP synthase is regulated by Ca2+ provides evidence for a new regulatory positive-feedback loop during physiological stimulation. Indeed, the ‘bell-shaped’ Ca2+ dose–response curve of the new activity may mean that the balance of NAADP synthesis and breakdown are likely to be under close control by cytosolic Ca2+ concentration. Thus, at least in mouse brain homogenates, NAADP is dephosphorylated to NAAD via a mechanism that is also Ca2+-dependent . It also follows that, in response to stimuli which raise cytosolic Ca2+ through other means (e.g. glucose-induced depolarization in the case of β-cells), NAADP increases  may be as much a consequence as a cause of an initial Ca2+ rise. Nevertheless, NAADP-induced Ca2+ release may nonetheless serve as a trigger in this system to generate local Ca2+ increases, possibly close to secretory granules, through Ca2+-induced Ca2+ release .
The new study raises several important new perspectives. A conundrum surrounding the mechanisms of NAADP synthesis arises from the localization of the known synthetic enzymes. CD38, the only molecule thought previously to produce NAADP in mammalian tissues, is an ectoenzyme, with the catalytic domain situated at its C-terminus, in the extracellular space, where the acidic pH used to catalyse the base-exchange reaction is unlikely to pertain. In common with CD38, the localization of the new activity on the extracellular surface, leads to a ‘topological paradox’ (Figure 2). This localization obviously requires that a specific transport mechanism exists both for substrate (NADP) efflux and for product (NAADP) influx; of note, a transport mechanism for NAADP uptake has recently been found on the plasma membrane of RBL (rat basophilic leukaemia) cells , and may suggest that NAADP could be synthesized extracellularly (and also, incidentally, raises the possibility of a paracrine/autocrine role for NAADP). However, it is probably more likely that these transport mechanisms are chiefly involved in transporting substrate/product across the membrane of an acidic internal organelle (lyso-/endo-some, secretory granule, etc.) in which the new NAADP synthetic activity (or a cyclase such as CD38) is located in a low-pH environment. Despite the more limited pH-dependency of the new enzyme, which still displays significant activity at neutral pH , unlike the cyclases , the inhibitory effect of physiological Ca2+ concentrations outside the cell (approx. 1 mM) seem likely to completely suppress base-exchange activity at the cell surface. Nevertheless, it is conceivable that Ca2+ action on the enzyme may occur only via a cytosolic domain, in which case activity may still be present in the face of high extracellular Ca2+. Fortunately, this possibility can be readily tested using intact sperm.
Although the study of Vasudevan and colleagues  is important, it is clear that much remains to be done. First, molecular identification of the NAADP synthetic enzyme activity is needed and should allow an assessment of its expression and overall contribution to NAADP synthesis in sperm and in other cell types: RNA interference or gene-knockout approaches in different species and cell types are likely to be illuminating. Studies are also required to explain how the activity is regulated in response to cellular stimulation. Such analyses, as well as complementary studies to identify the receptor(s) for NAADP at the molecular level , are eagerly awaited.
Work in our laboratory is funded by the Wellcome Trust, MRC (Medical Research Council), European Union (Consortium ‘SaveBeta’), the National Institutes of Health and Diabetes UK. E. A. B. receives a Ph.D. studentship from Imperial College.