Release, neuronal effects and removal of extracellular β-nicotinamide adenine dinucleotide (β-NAD⁺) in the rat brain

Abstract

Recent evidence supports an emerging role of β-nicotinamide adenine dinucleotide (β-NAD(+) ) as a novel neurotransmitter and neuromodulator in the peripheral nervous system -β-NAD(+) is released in nerve-smooth muscle preparations and adrenal chromaffin cells in a manner characteristic of a neurotransmitter. It is currently unclear whether this holds true for the CNS. Using a small-chamber superfusion assay and high-sensitivity high-pressure liquid chromatography techniques, we demonstrate that high-K(+) stimulation of rat forebrain synaptosomes evokes overflow of β-NAD(+) , adenosine 5'-triphosphate, and their metabolites adenosine 5'-diphosphate (ADP), adenosine 5'-monophosphate, adenosine, ADP-ribose (ADPR) and cyclic ADPR. The high-K(+) -evoked overflow of β-NAD(+) is attenuated by cleavage of SNAP-25 with botulinum neurotoxin A, by inhibition of N-type voltage-dependent Ca(2+) channels with ω-conotoxin GVIA, and by inhibition of the proton gradient of synaptic vesicles with bafilomycin A1, suggesting that β-NAD(+) is likely released via vesicle exocytosis. Western analysis demonstrates that CD38, a multifunctional protein that metabolizes β-NAD(+) , is present on synaptosomal membranes and in the cytosol. Intact synaptosomes degrade β-NAD(+) . 1,N (6) -etheno-NAD, a fluorescent analog of β-NAD(+) , is taken by synaptosomes and this uptake is attenuated by authentic β-NAD(+) , but not by the connexin 43 inhibitor Gap 27. In cortical neurons local applications of β-NAD(+) cause rapid Ca(2+) transients, likely due to influx of extracellular Ca(2+) . Therefore, rat brain synaptosomes can actively release, degrade and uptake β-NAD(+) , and β-NAD(+) can stimulate postsynaptic neurons, all criteria needed for a substance to be considered a candidate neurotransmitter in the brain.

Figure 1. Protocol for isolation of synaptosomes from rat forebrain

Different velocities of cold ultracentrifugation and Percoll gradient centrifugation were used to obtain intact synaptosomes (P3). In some experiments synaptosomal membranes were hypotonically disrupted and supernatant (S5) was used as intrasynaptosomal content as described in Methods.

Figure 2. Diagram of major enzymatic pathways for the degradation of β-NAD⁺ and ATP

CD38 exhibits NAD glycohydrolase (NADase), ADP-ribosyl cyclase and cADPR hydrolaze activities and degrades β-NAD⁺ to ADPR and cADPR. ADPR is degraded to AMP by nucleotide pyrophosphatases (NPPs). AMP in turn is degraded to ADO by 5′-nucleotidase. ATP is degraded by a family of ecto-nucleoside triphosphate diphosphohydrolases (E-NTPDases) to ADP and AMP, and then to ADO by 5′-nucleotidases.

Figure 3. Spontaneous release and high K⁺-evoked release of adenine purines

KCl 25 mmol/L (closed circles, n=7) or 65 mmol/L (open circles, n=13) evoked overflow of ATP (A), ADP (B), AMP (C), ADO (D), and β-NAD+ADPR+cADPR (E) in rat isolated forebrain synaptosomes. The sum of all purines (i.e., ATP+ADP+AMP+β-NAD⁺+ADPR+cADPR+ADO) is also shown (F). Data are averaged for each individual purine as well as for the sum of all purines (total purines) and are expressed as the mean ± SE. *P<0.05 vs. controls at 0 min. One-way ANOVA followed by post-hoc Bonferroni’s multiple comparison tests. Note the sustained pattern of the overflow of β-NAD⁺ and total purines.

Figure 4. Effects of botulinum neurotoxin A (BoNT/A), bafilomycin A1 and ω-conotxin GVIA (ω-Ctx GVIA,) on spontaneous and high-K⁺-evoked release of adenine purines

(AD) Original chromatograms of superfusates from synaptosomes pretreated with either KBH solution (controls, n=13), BoNT/A (300 nmol/L for 2 h, n=6), bafilomycin (0.1 μmol/L for 30 min, n=8) or ω-Ctx GVIA (20–50 nmol/L for 30 min, n=6) and superfused with 25 mmol/L K⁺ for 5 minutes. Superfusate samples were collected at 0, 1, 2, and 5 minute of superfusion with high K⁺-solution. The peak eluting at ~11.5 minute is formed by β-NAD⁺+ADPR+cADPR; this peak is labeled as β-NAD⁺ to reflect the primary compound in the β-NAD⁺+ADPR+cADPR mixture. Scales apply to each groups of chromatograms. LU, luminescence units. (E, F) Data are averaged for β-NAD⁺ and total purines and are expressed as the mean ± SE. In contrast to controls the high K⁺ solution failed to evoke an increased release of purine nucleotides in the presence of BoNT/A, bafilomycin A1 or ω-Ctx GVIA. *P<0.05 vs. controls at 0 minutes. One-way ANOVA followed by post-hoc Bonferroni’s multiple comparison tests.

Figure 5. Degradation of exogenous purine substrates during superfusion of synaptosomes

(A) Ecto-ATPase activity. Left panel: Original chromatograms of 50 nmol/L eATP in the absence of synaptosomes (−s) and after 1-minute contact with synaptosomes (+s); LU, luminescence units. Note the decrease in eATP and the increase in eADP, eAMP, and eADO in the (+s) samples. Right panel: Graphic representation of e-product (eADP+eAMP+eADO) formation in superfusate samples collected before (−s) and during 1-min perfusion of synaptosomes with eATP (+s). (B) NAD glycohydrolase activity. Left panel: Original chromatograms of 0.2 μmol/L eNAD in the absence (−s) and presence of synaptosomes (+s). Note the decrease in eNAD and the increase in eADPR formation in the (+s) samples. Right panel: Graphic representation of eADPR formation in superfusate samples collected in the absence (−s) or presence (+s) of synaptosomes. (C) ADP-ribosyl cyclase activity. Left panel: Original chromatograms of 0.2 mmol/L NGD in the absence (−s) and presence of synaptosomes (+s). Note the increase in cGDPR formation in the (+s)-samples. Right panel: Graphic representation of cGDPR formation in superfusate samples collected in the absence (−s) or presence (+s) of synaptosomes. *P<0.05, **P<0.01. Data represent mean ± SE from 4 experiments with each substrate, paired Student’s t-test.

Figure 6. Degradation of eNAD and expression of CD38

(A) NAD glycohydrolase activity determined by a 60-minute incubation of synaptosomes with 1 mmol/L eNAD. Original chromatograms of eNAD in the absence (−s) and presence of synaptosomes (+s); LU, luminescence units. Note the increased formation of eADPR and eADO in contact with synaptosomes. (B) Western analysis showed expression of CD38 (45 kDa) in intact synaptosomes and in the intrasynaptosomal compartment. Intrasynaptosomal compartment also expressed the synaptic vesicle markers synaptophysin (42 kDa) and chromogranin B (78 kDa).

Figure 7. Uptake of eNAD

Original chromatograms of intra-synaptosomal content of eNAD after 0, 1, 5, and 60-minute incubation of intact synaptosomes with 1 mmol/L eNAD alone (A) and in the presence of Gap 27 (100 μmol/L, n=2) (B). Incubation with eNAD caused appearance of eNAD and its metabolites eADPR and eADO in the intra-synaptosomal compartment. Gap 27 had no effect on the intrasynaptosomal levels of eNAD or eNAD⁺+eADPR+eADO (C, D). Simultaneous incubation of β-NAD⁺ (10 mmol/L) and eNAD (1 mmol/L) abolished the appearance of eNAD in the intrasynaptosomal compartment and reduced the intrasynaptosomal levels of eNAD+eADPR+eADO after 60-minute incubation, n=3 (E, F). Data represent mean ± SE, *P<0.05, unpaired Student’s t-test.

Figure 8. Intracellular Ca²⁺ increases in a cultured neurons following localized applications of β-NAD⁺

(A) Rapid Ca²⁺ increases in 6 neurons identified by number in the culture illustrated in the inset. β-NAD⁺ was applied via a brief (500 ms) pressure pulse from a microelectrode (E). Two consecutive responses are shown (2 min interstimulus interval), showing that responses in individual neurons are reproducible. (B) After 10 min of exposure to Ca²⁺-free buffer, responses to the same β-NAD⁺ stimuli were abolished. (C) For comparison with β-NAD⁺ responses, Ca²⁺ elevations are shown in the same neurons after restoration of extracellular Ca²⁺, and bath application of 40 mmol/L KCl. Scale bar: 30 μm.

Figure 9. Similar neuronal [Ca²⁺]_i transients with ADPR and ATP

(A) [Ca²⁺]_i increases in the same neurons, following sequential applications of β-NAD⁺ and ATP. The delivery micropipette was changed between β-NAD⁺ and ATP trials, during a 10 min interval indicated by the axis break. (B) Desensitization of P2X receptors by pre-exposure to αβ-methylene ATP (50 μmol/L, 10 min) reduced, but did not abolish β-NAD⁺ responses. (C) Comparison of responses to β-NAD⁺ and its metabolite ADPR in the same neurons. As in A, the delivery micropipette was changed between challenges with the two agonists. Identical pipette concentrations (i.e., 50 mmol/L) and delivery pulses (500 ms) were used for β-NAD⁺, ATP and ADPR.