NAMPT inhibitors exert their effect by decreasing intracellular NAD content via direct inhibition of NAMPT. Addition of excess nicotinamide or nicotinic acid abolishes this effect by increasing NAD content . To examine the effect of KN-93 and KN-92 on NAD content, THP-1 cells were incubated with KN-93 or KN-92 for 24 h as pretreatment, and NAMPT inhibitor was added and incubated for 5 h. In a previous study, exposure to an NAMPT in- hibitor for 5 h decreased NAD content to approximately 50% or 60% . In our present study, KN-93 or KN-92 did not have any effect on this decrease (Fig. 3).
KN-93 inhibits disruption of mitochondrial membrane potential induced by NAMPT inhibitors
As NAMPT inhibitors cause apoptosis via loss of mitochondrial membrane potential , the effect of KN-93 and KN-92 on mitochondrial membrane potential was also investigated. The reagent JC-10, a superior derivative of JC-1, was used in this assay to detect membrane potential as a fluorescence signal. As shown in Fig. 4, KN- 93 and KN-92 inhibited the loss of mitochondrial membrane po- tential, suggesting that the mechanism of action occurs upstream of the induction of mitochondrial membrane permeabilization. Sublingual
NAMPT inhibitors upregulate Bim and KN-92 does not affect this induction
The Bcl-2 family of proteins have either pro- or anti-apoptotic activities and control the permeabilization of the outer mitochondrial membrane to regulate the mitochondrial pathway of apoptosis. As the induction of the Bcl-2 family of proteins by NAMPT inhibitors has not been evaluated in detail, the induction of the main Bcl-2 family members was assessed via quantitative RT-PCR. As shown in Fig. 5, Bim was markedly upregulated by AS1604498 after 40 h, and KN-92 exerted no inhibition. Bim is a proapoptotic BH3 protein that func- tions as an ‘activator’  and is induced by ER stress , implying that the mechanism of NAMPT inhibition involves ER stress. In contrast, KN-92 is not related to this process.
L-type Ca2 þ channel blockade partially inhibits apoptosis in- duced by NAMPT inhibitors
Given that KN-93 and KN-92 are reported to inhibit L-type cal- cium channels , the effects of the L-type calcium channel blockers verapamil and nimodipine on NAMPT inhibitors were evaluated. Both verapamil and nimodipine exhibited moderate inhibitory ac- tivity up to 32 μM (Fig. 6). However, we were unable to evaluate higher concentrations due to the inhibition of normal cell prolifera- tion. IC50 values of KN-93 and KN-92 against CaV1.3 channels were approximately 1 μM . Although the effect on each subtype could not be determined, considering that nimodipine is reported to have an IC50 of 2.7 μM against CaV1.3 channels , the effect of KN-93 may be partially explained by the effect on L-type Ca2 þ channels.
KN-92 does not effectively inhibit apoptosis induced by other apoptosis-inducing agents
Given that our data demonstrated that the KN-93 series of compounds inhibit apoptosis induced by NAMPT inhibitors prior to mitochondrial membrane permeabilization, the ability of these compounds to inhibit apoptosis induced by other anti-cancer drugs such as etoposide, actinomycin D, AB-737 and TW-37 was investigated. As these agents rapidly induce apoptosis of THP-1, test compounds were added 24 h before addition of apoptosis- inducing agents and measurements were conducted 8h later (Fig. 7). KN-92 did not effectively inhibit the apoptosis induced by these agents. KN-93 exhibited stronger inhibition of etoposide and actinomycin D than KN-92. As Ant-AIP-II also exhibited weak in- hibition of these molecules, this difference might be due to the inhibition of CaMKII by KN-93. Interestingly, the effect of ABT-737, a BH-3 mimetic, was markedly enhanced by L-type Ca2 þ channel blockade. These data show that the inhibition of the KN-93 series compounds is specific to apoptosis induced by NAMPT inhibitors.
Several NAMPT inhibitors have been reported, a proportion of which have been evaluated in clinical trials. However, the precise downstream mechanism of apoptosis induced by decreases in NAD is still unknown. We found that KN-93 inhibited apoptosis of THP-1 cells induced by a NAMPT inhibitor independent of CaMKII activity. KN-93 and KN92 have L-type Ca2 þ channel blocking ac- tivity  and the L-type Ca2 þ channel blockers nimodipine and verapamil inhibited apoptosis in the present study. The inhibition of apoptosis exerted by KN-93 and KN92 might therefore be par- tially due to the blockade of L-type Ca2þ channels. To our knowledge, the relationship between Ca2 þ and apoptosis caused by NAMPT inhibitors has not been reported, although Magnone et al. reported that decreasing NAD in turn reduced ER Ca2 þ stores via reduction of ADP-ribose synthesized by CD38 in T lymphocytes . Our data imply that the L-type Ca2þ channel is related to Ca2 þ perturbation and apoptosis in this system. KN-93 was only partially effective when the agent was added 20 h after the addi- tion of a NAMPT inhibitor (data not shown). This implies that the L-type Ca2 þ channel exerts its effect relatively early in the apop- tosis process.
The Bcl-2 family of proteins is intimately involved in the apoptosis induced by NAMPT inhibitors. Bruzzone et al. reported that Bcl-2-overexpressing Jurkat cells survived against FK866 for up to 96 h . In the present study, we showed that Bim was upregulated by NAD depletion. Bim has been reported as a proa- poptotic BH3 protein induced by various types of ER stress, including thapsigargin-induced stimuli . ER stress might in- volve this process ER stress might involve this process of Bim upregulation by NAD depletion. Our data demonstrate that KN-93 inhibits the depolarization of mitochondrial membrane potential, but not Bim induction. This finding indicates that the induction of Bim alone does not result in apoptosis in this system. Previous reports have cited a relationship between Ca2þ signaling and apoptosis , involving calpain activation and mitochondrial permeability transition pore. Calpain is a Ca2 þ -dependent cysteine protease involved in apoptosis , and cleavage of Bcl-2 family proteins is also reported to enhance apoptosis [23,24]. Bim induction and calcium activation might synergistically induce apoptosis.
Recent reports have demonstrated that FK866 causes autop- hagy in a proportion of cancer cells [25,26]. Given that caspase activation was not observed in these cells, autophagy might be one type of cell death induced by NAMPT inhibition. Furthermore, Del Nagro et al. identified another type of cell death called “oncosis”, which involves necrosis due to NAD depletion . They showed that the rate at which ATP levels decrease following those of NAD is important for determining the type of cell death.