Quantcast
Channel: kimberly christen
Viewing all articles
Browse latest Browse all 60

Effects of Vacor on Cellular NAD Homeostasis

0
0

We next attempted to characterize the temporal relationship be- tween NAD depletion and death in SH-SY5Y cells exposed to Vacor. Notably, NAD depletion preceded cell death by some hours when evaluated by means of propidium iodide staining (Figure 2A). Cell fractionation experiments showed that 1 hr after Vacor exposure, NAD shortage already occurred in the cytosol but not in mitochondria (Figure 2B). In keeping with this, we found that mitochondrial membrane potential was unaltered in SH-SY5Y cells 1 hr after the Vacor challenge (Figures S2A and S2B), a time point corresponding to an almost 50% depletion of the whole cellular dinucleotide content. Likewise, oxygen consumption was unaffected in cells exposed for 1 hr to Vacor, being evident, however, 3 hr after the challenge (Figures 2C and 2D). Also consistent with sparing of mitochondria homeosta- sis at early time points after Vacor exposure, depletion of NADH, which is more concentrated in the organelles than in cytosol, was delayed compared with that of NAD (Figure 2E).

Even though the structural analogy of Vacor with prototypical NAMPT inhibitors FK866 and GMX1778 (Figure S1A) suggested an involvement of NAMPT in the metabolic effects of Vacor, kinetics of both NAD drop and cell death were much more rapid in cells exposed to Vacor than in those challenged with FK866 or GMX1778 (Figures 2F and S2C). This important finding sug- gested that NAMPT inhibition cannot exclusively explain Vacor’s effects. In keeping with this, NMN, the enzymatic product of NAMPT, fully prevented NAD drop by FK866 or GMX1778 but not that prompted by Vacor (Figure 2G). Collectively, different hints indicated that the impact of Vacor on NAD metabolome was different from that prompted by NAMPT inhibitors, notwith- standing its structural analogy with this class of drugs. To further add to the discrepancy, we found that both NAMPT inhibitors FK866 and GMX1778 entirely prevented initial NAD depletion in Vacor-exposed cells (Figure 2H). This finding prompted us to carefully evaluate the impact of Vacor on the two enzymes of the metabolic nicotinamide salvage pathway.

Vacor Is an Enzymatic Substrate of NAMPT

Disregarding the different kinetics of NAD depletion between Vacor and the two prototypical NAMPT inhibitors, the apparently paradoxical finding that FK866 and GMX1778 counteracted Vacor-dependent NAD drop further suggested an interaction between Vacor and NAMPT. We therefore first evaluated the ef- fects of Vacor on recombinant human NAMPT activity and found that the chemical indeed inhibited the enzyme with an IC50 of 400 ± 28 nM (Figure 3A). This finding, together with the notion that compounds undergoing phosphoribosylation cause potent NAMPT inhibition (Sampath et al., 2015), prompted us to verify whether NAMPT can use Vacor as a substrate. We therefore incubated the recombinant human enzyme in the presence of PRPP, Vacor, and ATP and found, by means of high=perform- ance liquid chromatography (HPLC) analysis, a time-dependent formation of a reaction product (Figure 3B).

The formation of the product was dependent on the presence of PRPP (Figure S3A), suggesting its identity as phosphoribosyl Vacor (VMN). This assumption was confirmed by liquid chromatography-mass spectrometry (LC-MS) analysis of the incubated enzymatic mixture containing Vacor, which revealed the formation of a molecule with a molecular mass identical to that of VMN (Fig- ure 3C). Interestingly, although in the NAMPT-catalyzed reaction NMN synthesis is coupled to a facultative ATP hydrolysis that causes a 6-fold increase of the kcat for NMN synthesis (Burgos and Schramm, 2008), we found that NAMPT had the same activ- ity toward Vacor in the presence or absence of ATP (Figure 3Bc). Still, Vmax of VMN formation was comparable with that of NMN (28 ± 3 nmol/min/mg versus 30 ± 2 nmol/min/mg). The apparent KM for Vacor was lower than the sensitivity threshold of our HPLC assay (i.e., lower than 2 mM), so we were not able to make a com- parison with the KM for nicotinamide, which is in the nanomolar range (Burgos and Schramm, 2008). Still, the finding that Vacor inhibits NAMPT activity with IC50 values in the nanomolar range when nicotinamide is present at 200 nM (see STAR Methods), a concentration very close to the KM value, strongly suggests that the enzyme might use Vacor and nicotinamide with a compara- ble catalytic efficiency.

Although the very low KM value of NAMPT for nicotinamide precluded a deeper analysis of the Vacor inhibi- tion mechanism, we were able to analyze the inhibition mecha- nism with respect to PRPP. In line with prior work (Burgos and Schramm, 2008), we found a KM value for PRPP of 5 ± 2.0 mM. Under this condition, Vacor behaved as a competitive inhibitor with a Ki of 1.56 ± 0.8 mM (Figure S3B). Next, a docking study was undertaken to gather information on the molecular interac- tions between NAMPT and Vacor. To this end, we used the 3D coordinates of human NAMPT (PDB: 4L4L, high resolution of 2.12 A ̊) cocrystallized with the phosphoribosylated inhib- itor 6-({4-[(3,5-difluorophenyl)sulfonyl]benzyl}carbamoyl)-1-(5- O-phosphono-b-D-ribofuranosyl)imidazo[1,2-a]pyridin-1-ium (Oh et al., 2014). Remarkably, when Vacor and VMN were docked in the catalytic pocket of human NAMPT, their best putative candidate positions were similar to that of the prototypical NAMPT inhibitor FK866 (Figure 3D). In particular, we found that the pyridyl ring of Vacor makes two p-p interactions with aromatic amino acids Tyr18 and Phe193, and that hydrogen bonds are possible between Vacor ureidic group and Asp219 (Figure S3C).

Importantly, we also found that the pyridyl moiety of Vacor may acquire a peculiar binding position similar to that of the imidazole ring of the above cocrystallized inhibitor (Oh et al., 2014) (Figures 3E and S3D), thereby corroborating LC-MS data showing phosphoribose condensation with Vacor. As discovered in other crystals (PDB: 2E5C and 2E5D), this binding mode is also similar of that of the natural substrate nicotinamide and close to the binding site for PRPP (Takahashi et al., 2010). These results taken together were consistent with HPLC and LC-MS data indicating that Vacor can behave as an NAMPT substrate, undergoing phosphoribosylation to form VMN and cause enzymatic inhibition.


Viewing all articles
Browse latest Browse all 60

Latest Images

Trending Articles





Latest Images