Nicotinamide phosphoribosyltransferase (NAMPT)

Immunology and Cell Biology (2013), 1–9 & 2013 Australasian Society for Immunology Inc. All rights reserved 0818-9641/13 www.nature.com/icb

ORIGINAL ARTICLE

Nicotinamide phosphoribosyltransferase (NAMPT) activity is essential for survival of resting lymphocytes Maria Pittelli1,2,3, Leonardo Cavone1,2,3, Andrea Lapucci1,2, Claudia Oteri1,2, Roberta Felici1,2, Elena Niccolai1,2,3, Amedeo Amedei1,2,3 and Alberto Chiarugi1,2 NAD biosynthesis is emerging as a key regulator of immune cell functions. Accordingly, inhibitors of the NAD-synthesizing enzyme nicotinamide phosphoribosyltransferase (NAMPT) have anti-inflammatory effects, counteract hematological malignancies and are being tested in clinical trials. Still, their effect on different cell types still waits to be fully investigated. Here we show that the NAMPT inhibitor FK866 induces NAD depletion in various mouse organs but selectively causes dramatic atrophy of the spleen red pulp. Accordingly, in cultured mouse lymphocytes exposed to FK866, NAD contents drop to 50% of basal values within 2 days, a condition sufficient to prompt complete cell death. Cultures of human lymphocytes are more resistant to FK866 and sustain a 50% NAD reduction for 5 days before dying. Death of both cell types can be prevented by different NAD precursors, indicating critical NAD homeostasis in lymphocytes. Indeed, inhibition of the NAD-consuming enzyme poly(ADP-ribose) polimerase-1 suffices to prevent FK866-induced NAD depletion and death of both lymphocyte types. Poly(ADP-ribose) polymerase-1-null lymphocytes also undergo lower NAD depletion and reduced cell death when exposed to the drug. At variance with other cell types, neither apoptosis nor autophagy are exclusively responsible for lymphocyte death by FK866, consistent with a general impairment of lymphocyte homeostasis following NAD depletion. Data demonstrate a unique sensitivity of resting lymphocytes to NAD-depleting agents, providing new hints of relevance to lymphocyte biology and therapeutic interventions with NAMPT inhibitors. Immunology and Cell Biology (2013) 0, 000–000. doi:10.1038/icb.2013.85 Keywords: FK866; NAD; NAMPT; lymphocyte

During the recent several years we witnessed to a renewed interest in the biosynthesis of NAD.1 Thanks to the identification of several NAD-hydrolyzing enzymes it is now well appreciated that pyridine nucleotides, besides being classic, key cofactors of metabolic redox reactions, also undergo irreversible hydrolysis and transformation into molecules with pleiotypic roles in cellular homeostasis and disease pathogenesis.2 Because of ongoing degradation of cellular NAD, continuous dinucleotide resynthesis is necessary in cells to warrant maintenance of homeostasis and survival.3 In mammals, the ‘NAD rescue pathway’ is the shortest metabolic route to allow NAD resynthesis. It stems from nicotinamide (Nam), the byproduct of NAD-hydrolyzing enzymes, which is converted by nicotinamide phosphoribosyltransferase (NAMPT) into nicotinamide mononucleotide (NMN). The latter then undergoes direct conversion by NMN adenylyltransferases (NMNATs) into NAD. Additional NAD precursors are tryptophan, nicotinic acid and nicotinamide riboside (NR) that lead to NAD formation through the kynurenine, Priess-Handler and NR kinase pathways, respectively.4 Several lines of evidence demonstrate that NAMPT catalyzes the rate-limiting step of NAD resynthesis and therefore finely tunes NAD

Q2

homeostasis.5 It has been repeatedly demonstrated that NAMPT has key roles in a wide array of cell functions such as oncogenic transformation, immune activation and circadian rhythm regulation.5 This has been possible because of the availability of potent and selective inhibitors of NAMPT such as FK866 (also known as APO866) and CHS-828 (also known as GMX1777). Of note, these drugs already reached the clinical arena and have been tested in several clinical trials for different solid and hematological malignancies.6,7 Although the exact clinical relevance of NAMPTblocking drugs needs to be clearly established, mechanisms leading to cancer cell death upon NAMPT inhibition have been ascribed to activation of apoptosis or autophagy.8–11 Intriguingly, several reports indicate that inhibition of NAMPT selectively kill cancer cells, a feature in line with the apparent safety profile of FK866 or GMX1777 in humans.6,7 Because doses higher than those used in first clinical trials might be required for optimal tumor killing or anti-inflammatory purposes, it is possible that the real safety profile of NAMPT inhibitors has not been clearly defined. In this regard, it has been reported that lymphopenia occurs in patients enrolled in clinical trials evaluating

1Section of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Florence, Italy and 2Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy 3These authors contributed equally to this work. Correspondence: M Pittelli, Section of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Viale Pieraccini 6, Florence 50139, Italy. E-mail: [email protected] Received 30 July 2013; revised 28 October 2013; accepted 28 October 2013

Nicotinamide phosphoribosyltransferase activity M Pittelli et al

Q1 2

the antineoplastic effects of FK866. This finding, together with evidence that NAMPT is upregulated upon lymphocyte activation,12 prompted us to explore the role of NAD metabolisms and the effects of NAMPT inhibition in lymphocytes in vitro and in vivo. RESULTS Effect of FK866 on NAD content in mouse organs Despite the use of FK866 in patients enrolled in clinical trials, a clear picture of the impact of FK866 on NAD content in different organs in vivo is lacking. We therefore measured NAD content in organs of mice treated with FK866 at doses of 25 and 50 mg kg 1 b.i.d. for 3 days. We found that the dinucleotide contents decreased in a dosedependent manner in the brain, blood, lung, kidney and liver. The brain showed the highest NAD content and also the highest NAD depletion upon FK866 treatment. Mouse cotreatment with the NAMPT substrate Nam (200 mg kg 1) completely prevented NAD depletion in organs of mice receiving FK866 at 50 mg kg 1, in keeping with the competitive nature of NAMPT inhibition by FK866.7 Surprisingly, splenic NAD contents were not affected by FK866 (Figure 1a). However, when the weight of organs from vehicle- or drug-treated mice was compared after 3 days of treatment, only the spleen showed a marked weight decrease that was prevented in Namcotreated mice (Figure 1b). As shown in Figure 1c, the spleen of mice treated with FK866 appeared pale and markedly atrophic, whereas those of animals receiving FK866 together with Nam were identical to those of vehicle-injected mice. Histological evaluation of the spleen showed that treatment with FK866 dramatically reduced red pulp with increased eosinophilic staining compared with the spleen from vehicle-injected mice (Figures 1d and e). The histological changes were completely prevented by cotreatment with Nam (200 mg kg 1) (Figure 1f). Higher magnifications of white pulp confirmed dramatic loss of lymphocytes in the spleen of FK866-treated mice but not in those also receiving Nam (Figures 1g–i). In light of the selective weight reduction of the spleen and red pulp shrinkage, we then investigated in vitro the effect of FK866 on mouse and human lymphocytes. Effect of FK866 on NAD content and survival of mouse and human lymphocytes In an attempt to reproduce in vitro the model of the spleen atrophy by FK866, we exposed mouse and human lymphocytes in vitro to a drug concentration of 100 nM able to fully inhibit NAMPT.7 Fluorescenceactivated cell sorting analysis of propidium iodide-negative mouse splenocytes showed that they were 44±9% CD3-positive T cells and 53±5% CD45R-positive B cells (Figure 2a). A 5±2% of CD11bpositive monocytes was also present (not shown). Under these conditions, cellular NAD contents decreased in a time-dependent manner in mouse cells, reaching 54±5 and 9±4% of control after 2 and 3 days, respectively (Figure 2b). Interestingly, human lymphocytes from peripheral blood mononucleated cells showed a similar time-dependent NAD depletion during the first 2 days of drug exposure, then NAD content stabilized at about 50% of basal values for 3 days, and underwent abrupt collapse at day 6 (Figure 2b). We then evaluated the impact of NAD depletion on cell survival evaluating propidium iodide permeability of the plasma membrane by means of the cytofluorimetric analysis. We found that NAD depletion correlated to a time-dependent loss of lymphocyte viability (Figure 2c). Specifically, mouse cells showed a complete loss of the plasma membrane integrity at day 2 (Figure 2d), a time point at which a 54% of the basal NAD pool was still present (see Figure 2b). Conversely, in spite of equivalent NAD loss, human lymphocytes remained propidium iodide negative at day 2. Cell death only Immunology and Cell Biology

developed at day 3, and then linearly increased to reach the maximum at day 6 (Figure 2d). We therefore next analyzed the FK866’s effects on human cells activated with phytohaemagglutinin (Figure 2e) and mouse B and T cells activated with lipopolysaccharide or anti-CD3/ CD28 monoclonal antibodies (Figure 2f), respectively, and found an identical drug sensitivity (Figure 2d). The latter was similar in B and T cells both in vitro (Figure 2d) and in mice treated with the drug for 3 days (not shown). We also compared the sensitivity to FK866 of mouse splenocytes, peripheral blood or naı¨ve lymphocytes. As shown in Figure 2g, the three cell types showed almost the same sensitivity to FK866-induced cell death. Together, these findings indicate that, although with differences, mouse and human resting lymphocytes are sensitive to the toxic effects of FK866, in keeping with the splenic atrophy of mice exposed to the drug. Data, however, are in apparent contrast with prior work showing that FK866 cytotoxicity spares resting human lymphocytes and kills those activated with phytohaemagglutinin.13 We then reasoned that if FK866 kills lymphocytes via NAD depletion, then NAD precursors should be able to counteract the drug’s effects. By means of quantitative PCR, we first evaluated whether mouse and human lymphocytes express enzymes are able to metabolize the different NAD precursors. We found that mRNAs for NAMPT, nicotinic acid phosphoribosyltransferase (NAPRT), nicotinamide riboside kinase-1 (NRK1), NMNAT1 and indoleamine dioxygenase (IDO) were well detectable in these cells whereas, in comparison, levels of tryptophan 2 3-dioxygenase transcripts were much lower (Figure 3a). Given that it is currently unknown whether NAD shortage triggers homeostatic responses converging on transcriptional activation of genes coding for NAD (re)synthesis, we quantified transcript levels for NAMPT, NAPRT, NRK1, NMNAT and IDO in mouse and human lymphocytes up to 3 days after FK866 exposure. Figure 3b shows that mRNA contents remained stable up to 6 h in mouse cells and then decrease over time consistent with ongoing cell death. No transcript changes were detected in human lymphocytes up to 36 h (Figure 3c). When the effect of NAD precursors was investigated in mouse lymphocytes, we found that Nam, NR and NMN added to the culture medium fully prevented cell death occurring 2 days after FK866 exposure. Conversely, NAD and nicotinic acid only partially prevented cell demise, and kynurenine had not effect (Figure 3d). At variance with mouse lymphocytes, all the NAD precursors efficiently counteracted death of human cells induced by a 4-day exposure to the NAMPT inhibitor. In particular, nicotinic acid fully prevented cytotoxicity at 1 mM, and kynurenine prevented cell death by about 75% when used at a similar concentration. More importantly, NAD, Nam, NR and NMN increased cell survival above 100% (Figure 3e). The ability of kynurenine to rescue FK866 cytotoxicity in human but not mouse lymphocytes indicated that, as described for other cell types,14 mouse cells do not have a functional de-novo NAD synthesis from tryptophan, likely because of lack of expression of enzymes downstream to IDO. Together, these findings indicated that cell killing by FK866 was causally related to NAD depletion and not to non-specific effects. In addition, the ability of some NAD precursors to increase survival of human lymphocytes above control values suggested that they were able to counteract spontaneous cell death of these cells in cultures. Role of PARP-1 in FK866-dependent lymphocyte NAD depletion and death To ascertain the cause of NAD depletion in lymphocytes exposed to FK866, we focused on poly(ADP-ribose) polymerase (PARP)-1, the

Q1

Nicotinamide phosphoribosyltransferase activity M Pittelli et al 3

Figure 1 Effect of FK866 on mouse organ NAD content and weight. Effect of FK866 injected i.p. b.i.d at 25 or 50 mg kg 1 or vehicle (V) for 3 days on mouse organ NAD content (a) and weight (b). The effect of a cotreatment with nicotinamide (Nam, 200 mg kg 1) is also shown in (a) and (b). (c) Representative appearance of the spleen of mice receiving vehicle, FK866 50 mg kg 1 with or w/o nicotinamide cotreatment as described in (a). Histological preparations of spleens from mice treated with vehicle (d, g), FK866 50 mg kg 1 (e, h) or FK866 50 mg kg 1 þ Nam 200 mg kg 1 (f, i) as described in (a). Columns represent the mean±s.e.m. of three (a, b) experiments conducted in triplicate (a) or quadruplicate (b). (c–i) representative images are shown. *Po0.05, **Po0.01, ***Po0.001 vs control or vehicle. Analysis of variance (ANOVA) plus Tukey’s post hoc test.

major NAD-consuming enzyme within the cell.15 We therefore first investigated whether FK866-induced NAD depletion in mouse lymphocytes was affected by pharmacological or genetic suppression of PARP-1. Interestingly, two structurally unrelated PARP-1 inhibitors such as PJ34 (20 mM) and PHE (30 mM) reduced NAD depletion in cultured mouse lymphocytes exposed to the drug. Accordingly, NAD loss in lymphocytes from PARP-1-null mice was lower than that occurring in wild-type cells (Figure 4a). The two PARP-1 inhibitors

also counteracted NAD loss in human lymphocytes exposed to FK866 (Figure 4a). Of note, a 24 h exposure to PHE or PJ34 (30 mM) increased basal NAD levels by 47±6 and 56±7% in human lymphocytes not exposed to FK866 (that is, under resting conditions, not shown). This latter finding suggests ongoing NAD consumption in cultured human lymphocytes, and is consistent with the ability of NAD precursors Nam, NR, NMN and NAD itself to protect cultured human lymphocytes from spontaneous cell death (see Figure 3e). Immunology and Cell Biology


Q1 4

Figure 2 Effects of FK866 on mouse and human NAD content and survival. (a) Cytofluorimetric analysis of mouse splenocytes stained with anti-CD45R and anti-CD3 (b). Effect of FK866 (100 nM) on NAD contents of cultured mouse and human lymphocytes. (c) Representative cytofluorimetric analysis of the effect of FK866 (100 nM) on survival of mouse lymphocytes after 48 h of incubation. (d) Effect of FK866 (100 nM) on survival of mouse B and T cells or human lymphocytes under resting conditions or activated with lipopolysaccharide, anti-CD/CD28 monoclonal antibodies or phytohaemagglutinin (PHA), respectively. (e) Evaluation of human PBLs activation upon PHA stimulation. (f) Representative cytofluorimetric analysis of mouse B and T-cell activation. (g) Comparison of the cytotoxic effects of FK866 (100 nM) on mouse splenocytes, naive T cells or blood lymphocytes. Points represent the mean±s.e.m. of four (b, d and f, g) experiments conducted in duplicate. In (a, c, e) representative plots are shown.

To further investigate the role of PARP-1 in FK866 lymphocyte killing, we next evaluated whether the FK866-induced cell demise could be reduced by inhibiting PARP-1. Of note, the two PARP inhibitors markedly reduced death of mouse and human cells in a concentrationdependent manner. Again, PARP-1-null lymphocytes were less sensitive to the toxic effects of FK866 than PARP-1-proficient cells (Figure 4b). Immunology and Cell Biology

Characterization of FK866-induced lymphocyte death Several reports demonstrate that FK866-induced cell death can be either of the apoptotic or autophagic type.8,9,11,16 Thus, intrigued by the different time-dependent sensitivity of mouse and human lymphocytes to FK866, we sought to gather information of the cell death pathways triggered by the drug in the two cell types. Caspase-3

Q1


Figure 3 Effects of NAD precursors on FK866-induced lymphocyte death. (a) Reverse transcription-PCR evaluation of transcript levels for NAMPT, NAPRT, NMNAT1, NRK1, IDO and tryptophan 2 3-dioxygenase (TDO) in mouse and human lymphocytes. Effect of FK866 (100 nM) on transcript levels of the above mentioned genes in mouse (b) and human (c) lymphocytes at different time of exposure. Effect of different NAD precursors, Nam, NR, NMN, nicotinic acid (NA) and KYN on FK866-dependent mouse (d) or human (e) lymphocyte death evaluated 48 h after exposure to the drug. (a–e) points represent the mean±s.e.m. of three experiments conducted in duplicate.

Figure 4 Effect of pharmacological or genetic suppression of PARP-1 on FK866-dependent NAD depletion and death in mouse or human lymphocytes. (a) Effect of pharmacological or genetic suppression of PARP-1 on FK866-dependent NAD depletion in mouse or human lymphocytes. PARP-1 inhibitors PHE (30 mM) and PJ34 (20 mM) or vehicle alone (V) were used. (b) Effect of pharmacological or genetic suppression of PARP-1 on FK866-dependent death in mouse or human lymphocytes. KO: PARP-1-null lymphocytes. Columns represent the mean±s.e.m. of three (a, b) experiments conducted in triplicate (a) or quadruplicate (b). *Po0.05, **Po0.01, ***Po0.001 vs vehicle. ANOVA plus Tukey’s post hoc test.

activity was measured as an index of activation of the apoptotic route in mouse and human lymphocytes undergoing cell death (that is, at 24 and 72 h for mouse and human cells, respectively). We found that

exposure to FK866 did not activate caspase-3 in mouse-resting lymphocytes (Figure 5a). Conversely, a fourfold increase of caspase3 activity was found in human lymphocytes challenged with FK866. Immunology and Cell Biology


Q1 6

Figure 5 Effects of FK866 on autophagic or apoptotic genes of mouse and human lymphocytes. (a) Evaluation of caspase-3 activity in mouse and human lymphocytes exposed to FK866 100 nM for 12 h and 24 h respectively. Staurosposrine (STP, 100 nM) is shown as positive controls.The effect of the caspase3 inhibitor Z-VAD (50 mM) is also shown. (b) Effect of Z-VAD (50 mM) on FK866-dependent lymphocyte killing. Effect of FK866 on expression of autophagy (CTSB, MAP1LC3A and WIPI1) or apoptotic (TP53, p21, BAX and MDM2) genes in mouse (100 nM, 12 h) (c) or human (d) lymphocytes (100 nM, 24 h). (e) Evaluation of ATP levels in human and mouse lymphocytes exposed to FK866 (100 nM). Columns represent the mean±s.e.m. of three (a–d) experiments conducted in triplicate.

Of note, this increase was prevented by concomitant exposure to the pan-caspase inhibitor Z-VAD-fmk (Figure 5a). However, when the effect of Z-VAD-fmk on FK866-induced cell death was investigated, we found that it was unable to reduce mouse or human lymphocyte death (Figure 5b). The possible contribution of apoptosis and autophagy to FK866 cytotoxicity was also investigated at the transcriptional level. Cea et al.16 report that in multiple myeloma cells undergoing FK866-depdenent cell death autophagic genes MAP1LC3A and WIPI1 are induced, whereas transcript levels of CTSB are reduced. We therefore quantified these transcripts by reverse transcription-PCR in mouse and human lymphocytes exposed to FK866 for 12 or 48 h, respectively. Figure 5c shows that these mRNAs did not undergo significant changes in mouse and human cells. Recent work by Thakur et al.11 reports that apoptotic factors such as p53 and p53-driven genes are induced in myeloid cells exposed to FK866. We therefore investigated whether this compound altered p53, p21, BAX and MDM2 transcripts in mouse and human lymphocytes. We found that these transcripts were not significantly increased in mouse cells, whereas a coherent, although modest, increase was detected in human ones (Figure 5d). Finally, we quantified ATP in Immunology and Cell Biology

mouse and human lymphocytes exposed to FK866. Data indicate that ATP contents decreased over time in both cell populations with kinetics consistent with loss of cell viability (compare Figure 5e with Figure 2d). DISCUSSION The present study demonstrates that cultured resting lymphocytes are exquisitely sensitive to the toxic effects of the NAMPT inhibitor FK866. The original finding that severe atrophy of spleen red pulp early occurs in mice treated with the drug in the absence of any overt atrophy in other organs confirms in vivo a remarkable and selective lymphocyte death following NAMPT inhibition. The ability of the NAMPT substrate Nam to fully restore organ NAD content and weight, on the one hand points to NAMPT inhibition as the sole mechanism of FK866 toxicity and on the other corroborates the hypothesis that NAD homeostasis is critical in lymphoid cells. The fact that the spleen NAD content expressed as moles per mg protein is not reduced by FK866 in spite of dramatic atrophy can be explained considering that NAD content is similar in FK866-sensitive and resistant-cells. At variance with prior work showing selective cell

Q1


death of activated lymphocytes upon FK866 exposure,13 we now report that sensitivity to FK866 does not differ between resting and activated mouse or human lymphocytes. At present, we do not know the reasons of this inconsistency. Still, the fact that lymphopenia often occurs in cancer patients on FK86617 suggests that the drug exerts its cytotoxic effects even in human resting lymphocytes in situ. Our findings are also consistent with the notion that NAMPT activity is required for thymic development of mouse lymphocytes.12 In spite of similar sensitivity to the toxic effects of FK866, we show here that death of human lymphocytes is delayed by 48 h with respect to that of mouse cells. Further, mouse lymphocytes start dying when NAD contents reach B70% of resting values and completely loss the plasma membrane integrity when NAD pool drops to B50%. Conversely, human lymphocytes are less sensitive to NAD depletion and fully maintain membrane integrity in spite of a 50% NAD drop (compare Figures 2b and d). Loss of viability concomitant with maintenance of 50% of the NAD pool is in keeping with the ability of FK866 to selectively reduce nucleotide content in the nucleocytoplasmic but not in the mitochondrial compartment.18 These findings indicate different requirement for NAD-dependent homeostatic processes and/or that pathways responsible for execution of FK866dependent cell death differ between mouse and human lymphocytes. However, the fact that NAD precursors can fully counteract cell demise of both cell types indicates that NAD depletion due to NAMPT inhibition is the common and sole trigger of FK866-induced mouse and human lymphocyte death. In this light, it appears evident that signaling pathways prompted by NAD loss and leading to cytotoxicity do not exclusively depend on entity of NAD depletion but also on the metabolism of a given cell. Data on NAD precursors indicate additional important features of NAD homeostasis in lymphocytes. First, the ability of Nam, NR, NMN and NAD to increase survival above control levels suggests that NAD shortage is an unexpected, key trigger of cultured human lymphocytes. Whether NAD depletion also occurs in vivo and contributes to death or exhaustion of T and/or B cell is not known. Still, pharmacological strategies able to boost NAD biosynthesis might represent innovative approaches to treatment of immune deficiencies and/or boost immune responses. Second, lack of induction of IDO, NAMPT, NARPT, NMNAT or NRK upon FK866 exposure indicates that NAD depletion is not sensed by the cell as a trigger to re-establish NAD homeostasis. This is at odds with induction of IDO or NAMPT during neoplastic transformation or activation of immune cells,5,19 including lymphocytes.12 Lastly, the cytoprotective effect of exogenous NAD further supports the hypothesis that the nucleotide can cross the plasma membrane.9,18 Mechanisms contributing to NAD depletion in cells exposed to NAMPT inhibitors have been only partially addressed. Even though all the NAD-consuming enzymes might theoretically contribute to pyridine nucleotide depletion, PARP-1 is considered the main enzyme responsible for ongoing cellular NAD consumption.15 Indeed, PARP-1 activity is not invariantly related to DNA damage and may be due to DNA cruciform structures and hairpin.20,21 Consistently, genetic of pharmacological suppression of PARP-1 is associated to increased NAD levels in the resting cell or mice.22,23 The ability of PARP-1 inhibitors to almost completely prevent NAD shortage of mouse and human lymphocytes (Figure 4a) is in keeping with ongoing PARP-1 activity in these cells. Evidence that these drugs also afford remarkable and comparable protection of the two types of lymphocytes from FK866 cytotoxicity also indicates that PARP-1 is causally responsible for cell death. A large body of evidence demonstrates that PARP-1

hyperactivation is a powerful trigger of cell death.24 For instance, hyperactivation of PARP-1 by the alkylating agent 1-Methyl-3-nitro1-nitrosoguanidine prompts lymphocyte death, an effect prevented by pharmacological inhibition of the enzyme.12 Given that FK866 does not lead to PARP-1 activation (actually it causes reduced poly(ADPribosyl)ation),25 we reason that the cytoprotective effects of PARP-1 inhibitors should be ascribed to their ability to prevent basal PARP-1 activity-dependent NAD depletion in cells in which NAD resynthesis is compromised by NAMPT inhibition. Still, the evidence that protection afforded by PARP-1 inhibitors is more robust than that provided by genetic suppression of the enzyme (Figure 4b) suggests that the drugs inhibit additional PARPs involved in ongoing NAD consumption. This is in keeping with the lack of selectivity of the PARP inhibitors used toward PARP-1.26 As for the cell death machineries operating during FK866-induced lymphocyte demise, our data suggest that autophagy is not involved. Conversely, caspase-3 activation and induction of p53-driven genes in human lymphocyte exposed to FK866 suggest that the apoptotic program is triggered in these cells. These findings taken together, plus evidence that caspase-3 inhibition does not prevent death of human lymphocytes, indicate that NAD depletion likely prompts a general derangement of cellular homeostasis that cannot be prevented by inhibition of a single cell death pathway. Of course, besides the direct clinical implications of the present study for patients enrolled in clinical trials with NAMPT inhibitors, it will be important to unravel the biochemical basis of the high lymphocyte sensitivity to NAD depletion. In this regard, it is worth noting that FK866 depletes cytosolyc NAD, leaving unaffected the mitochondrial pool.25,27 This notion, along with evidence that upon NAMPT inhibition the glycolytic flux is blocked at the NADdependent glyceraldehyde 3-phosphate dehydrogenase step,27 suggests that impairment of the glycolytic flux by FK866 may be a major mechanism contributing to lymphocyte death. In keeping with this, even though these cells are endowed with sufficient metabolic plasticity to readily switch from a purely glycolytic (that is, fermentative) to a respiratory (that is, oxidative) metabolism,28 high glycolytic fluxes in the presence of oxygen (the so-called aerobic glycolysis) seem essential for lymphocyte survival, proliferation and resistance to metabolic stressors.29,30 Hence, dependence of lymphocytes on aerobic glycolysis might entirely or in part contribute to their high sensitivity to a metabolic stress such as FK866-dependent NAD depletion and glycolytic derangement. In conclusion, the present study furthers our understanding of NAD homeostasis in lymphocytes. It also provides information on the pharmacodynamic properties of NAMPT inhibitors that could be of clinical relevance to innovative treatment of neoplastic or inflammatory disorders.

METHODS Animal treatment

Q7

CD1 male albino mice (20–25 g) (n ¼ 6 per group) were injected b.i.d. intraperitoneally with FK866 or vehicle alone for 3 days. Groups of mice also received subcutaneous injections of Nam. Animals were killed 72 h after first FK866 administration, organs were rapidly collected and stored at 80 1C. From each organ, few milligrams were collected and processed for NAD quantitation. Spleens were paraffin-embedded and transversal sections were stained with hematoxylin and eosin. Experiments were conducted in compliance with the Italian guidelines for animal care (DL 116/92) in application of the European Communities Council Directive (86/609/EEC) and were formally approved by the Animal Care Committee of the University of Florence. Immunology and Cell Biology


Q1 8

Cytofluorimetric analysis Control cells were stained with propidium iodide and negative cells were gated and counted. The type of leukocytes present within the gate was identified as described31 with anti-CD3 (T cells), anti-CD45R (B cells) or anti-CD11b (monocytes) antibodies and evaluated by means of cytofluorimetric analysis.

Cell preparation, culture and treatment

Q3

Peripheral blood mononucleated cells were isolated from the blood samples obtained from healthy donors by Ficoll Hypaque (Sigma-Aldrich, Milan, Italy) density gradient centrifugation. Briefly, 10 ml of whole blood were diluted with phosphate-buffered saline to a final volume of 35 ml. The sample was carefully layered over 15 ml of Ficoll Hypaque in a 50 ml conical tube and centrifuged at 1500 r.p.m. for 300 at room temperature. Peripheral blood mononucleated cells were recovered and washed three times before culturing. Mononuclear cells (107) per well were plated in six-well plates for 2 h in complete Roswell Park Memorial Institute medium to allow monocytes to adhere. Non-adhering mononuclear cells (PBLs) were harvested by washing with phosphate-buffered saline (Invitrogen Italia, Milan, Italy) and used for further experiments. As for mouse splenocytes, cells were prepared as described.31 Cells were maintained in complete RPMI 1640 medium (Gibco, Milan, Italy) supplemented with 10% of fetal bovine serum, 2 mM glutamine, 1% penicillin-streptomycin and 50 mM mercaptoethanol (Gibco, Milan, Italy). Cell cultures were exposed to FK866 (RTI International, Research Triangle Park, NC, USA) or other compounds directly dissolved in the culture medium.

Mouse B and T-cell purification Q4

B and T lymphocytes were separated by immunomagnetic cell sorting using specific labelling kit from Myltenyi Biotech following the manufacturer’s instruction. In indicated experiments, B cells were activated by lipopolysaccharide exposure (200 ng ml 1 per 48 h) and T cells were activated for 48 h by anti-CD3/CD28 covered beads following the manufacturer’s instruction.

T naive cell purification Naive CD4 þ CD25 CD62Lhi T cells were purified by immunomagnetic cell sorting using specific labelling kit from Myltenyi Biotech following the manufacturer’s instruction.

NAD measurement NAD contents were quantified by means of an enzymatic cycling procedure according to Shah et al.32 Briefly, cells grown in a 48-well plate were washed and resuspended in 50 ml of HClO4 (1 N) and then neutralized with an equal volume of 1 N KOH. After the addition of 100 ml Bicine (100 mM, pH 8), 50 ml of the cell extract was mixed with an equal volume of the Bicine buffer containing 23 ml ml 1 ethanol, 0.17 mg ml 1 of 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium, 0.57 mg ml 1 of phenazine ethosulfate and 10 mg of alcohol dehydrogenase. The mixture was kept at room temperature for 20 min and then absorbance at 550 nm was measured. A standard curve allowed quantification of NAD.

Total RNA extraction and quantitative reverse transcription-PCR Q5

Total RNA was isolated from human and mouse lymphocytes with RNeasy mini kit (Qiagen, Milan, Italy) and treated with RNase-free DNase (Ambion, USA). One microgram of RNA was retro-transcribed using iScript (Bio-Rad, USA) and amplified with specific primers. Human: NAMPT fw 50 -AGCTGTT CCTGAGGGCTTTGTC-30 and rv 50 -AGTGAGCAGATGCTCCTATGCC-30 , NMNAT1 fw 50 -TCCCATCACCAACATGCACC-30 and rv 50 -TGATGACCCGG TGATAGGCAG-30 , NAPRT1 fw 50 -CTGGCTGGAGTCAGTCCTCATC-30 and rv 50 -GGTCCTCGGTCAGCTTCATTC-30 , IDO fw 50 -TGGTGGAGGACATGC TGCTC-30 and rv 50 -TTGCAGATGGTAGCTCCTCAGG-30 , NRK1 fw 50 -GAA ACACCTCCCAAATTGCAGTGTC-30 and rv 50 -GGAATTTCCTCAGCACTTT CCTGG-30 , CTSB fw 50 -CCTGTCGGATGAGCTGGTCAAC-30 and rv 50 -GGC CATTGTTCCCGTGCATCG-30 , MAP1LC3A fw 50 -TCAGACCGGCCTTTCA AGCAGC-30 and rv 50 -GGCGCCGGATGATCTTGACC-30 , WIPI1 fw 50 -TCC ACGGAAGCAATGAAATCCCG-30 and rv 50 -CAGCAGCCTTTGCCGGTT CAG-30 . TP53 fw 50 -GGCCCATCCTCACCATCATCAC-30 and rv 50 -GAGCTG Immunology and Cell Biology

GTGTTGTTGGGCAGTG-30 , p21 (CDKN1A) fw 50 -TGTCACCGAGACACC ACTGGA-30 and rv 50 -GGTCCACATGGTCTTCCTCTGC-30 , BAX fw 50 -TGG AGCTGCAGAGGATGATTGCC-30 and rv 50 -TGTCCAGCCCATGATGGTTCT GATC-30 , MDM2 fw 50 -ATTCTCCTGGCTCAGCCTCTGG-30 and rv 50 -TGGT GGCTCACATCTGCAATCC-30 , 18S ribosomal RNA fw 50 -CGGCTACCACAT CCAAGGAA-30 and rv 50 -GCTGGAATTACCGCGGCT-30 . Mouse: NAMPT fw 50 -TGGTTACAGAGGAGTCTCTTCG-30 and rv 50 -AAGCCGTTATGGTACTG TGC-30 , NMNAT1 fw 50 -AACAGGTGTGCCCAAGGTG-30 and rv 50 -CTCCAC AGCACATCGGACTC-30 , NAPRT1 fw 50 -CCAGGGCCTGCTAGATTCCTAC-30 and rv 50 -CAGACTCTAGCCAGGGCATCTG-30 , IDO1 fw 50 -ACAAGGGCTTC TTCCTCGTCTC-30 and rv 50 -AAACGTGTCTGGGTCCACAAAG-30 , NRK1 fw 50 -GACCAGCAGCCTTGGAAAGTGC-30 and rv 50 -CCACACGTGGCCATCG AAGTAC-30 , CTSB fw 50 -TGGACCCAAACTGCCAGGAAG-30 and rv 50 -CGT TGACTCGGCCATTGGTG-30 , MAP1LC3A fw 50 -CATCGAGCGCTACAAGGG TGAGA-30 and rv 50 -GATGTCAGCGATGGGTGTGGAG-30 , WIPI1 fw 50 -CTG GAGCGGCTACATGGGAAAG-30 and rv 50 -TAAAGGTGTCCGTCGGAGG AGG-30 . TP53 fw 50 -TGCCATGGAGGAGTCACAGTCG-30 and rv 50 -ACACTC GGAGGGCTTCACTTGG-30 , p21 (CDKN1A) fw 50 -TGGAGGGCAACTTCGT CTGGGA-30 and rv 50 -ATCTTCAGGCCGCTCAGACACC-30 , BAX fw 50 -TGG AGCTGCAGAGGATGATTGCTG-30 and rv 50 -TGTCCAGCCCATGATGGTTC TGATC-30 , MDM2 fw 50 -AGAGCCATGTGCTGAGGAGGA-30 and rv 50 -CATG GTTCGATGGCATTCAGGG-30 , 18S ribosomal RNA fw 50 -AAAACCAACC CGGTGAGCTCCCTC-30 and rv 50 -CTCAGGCTCCCTCTCCGGAATCG-30 . Real-time PCR was performed by SaoFast EwaGreen Supermix (Bio-Rad) and analyzed using the Rotor-Gene 3000 cycler system (Qiagen, Germany). PCR cycling parameters were: 95 1C for 30 s and 40 cycles of 95 1C for 5 s, 60 1C for 10 s.

Caspase-3 assay Human and mouse lymphocytes were challenged with the pan-caspase inhibitor N-Benzyloxycarbonyl-Val-Ala-Asp(O-Me) fluoromethyl ketone (Z-VAD-fmk) (Sigma-Aldrich) 50 mM for 24 h and 12 h, respectively. Cells were collected and analyzed using EnzChek Caspase-3 Assay kit (Invitrogen, Carlsbad, CA, USA) according to manufacturer’s instruction.

ACKNOWLEDGEMENTS This study was supported by grants from the University of Florence, the Italian Ministry of University and Scientific and Technological Research, Associazione Italiana Sclerosi Multipla (2009/R6) and Ente Cassa di Risparmio di Firenze.

1 Ziegler M, Niere M. NAD þ surfaces again. Biochem J 2004; 382: e5–e6. 2 Koch-Nolte F, Haag F, Guse AH, Lund F, Ziegler M. Emerging roles of NAD þ and its metabolites in cell signaling. Sci Signal 2009; 2: mr1. 3 Chiarugi A, Dolle C, Felici R, Ziegler M. The NAD metabolome–a key determinant of cancer cell biology. Nat Rev Cancer 2012; 12: 741–752. 4 Belenky P, Bogan KL, Brenner C. NAD þ metabolism in health and disease. Trends Biochem Sci 2007; 32: 12–19. 5 Garten A, Petzold S, Korner A, Imai S, Kiess W. Nampt: linking NAD biology, metabolism and cancer. Trends Endocrinol Metab 2009; 20: 130–138. 6 Montecucco F, Cea M, Bauer I, Soncini D, Caffa I, Lasiglie D et al. Nicotinamide phosphoribosyltransferase (NAMPT) inhibitors as therapeutics: rationales, controversies, clinical experience. Curr Drug Targets 2013; 14: 637–643. 7 Galli U, Travelli C, Massarotti A, Fakhfouri G, Rahimian R, Tron GC et al. Medicinal chemistry of nicotinamide phosphoribosyltransferase (NAMPT) inhibitors. J Med Chem 2013; 56: 6279–6296. 8 Hasmann M, Schemainda I. FK866, a highly specific noncompetitive inhibitor of nicotinamide phosphoribosyltransferase, represents a novel mechanism for induction of tumor cell apoptosis. Cancer Res 2003; 63: 7436–7442. 9 Billington RA, Genazzani AA, Travelli C, Condorelli F. NAD depletion by FK866 induces autophagy. Autophagy 2008; 4: 385–387. 10 Cagnetta A, Cea M, Calimeri T, Acharya C, Fulciniti M, Tai YT et al. Intracellular NAD þ depletion enhances bortezomib-induced anti-myeloma activity. Blood 2013; 122: 1243–1255. 11 Thakur BK, Dittrich T, Chandra P, Becker A, Kuehnau W, Klusmann JH et al. Involvement of p53 in the cytotoxic activity of the NAMPT inhibitor FK866 in myeloid leukemic cells. Int J Cancer 2013; 132: 766–774. 12 Rongvaux A, Galli M, Denanglaire S, Van Gool F, Dreze PL, Szpirer C et al. Nicotinamide phosphoribosyl transferase/pre-B cell colony-enhancing factor/visfatin is required for lymphocyte development and cellular resistance to genotoxic stress. J Immunol 2008; 181: 4685–4695.

Q6

Q1


13 Bruzzone S, Fruscione F, Morando S, Ferrando T, Poggi A, Garuti A et al. Catastrophic NAD þ depletion in activated T lymphocytes through Nampt inhibition reduces demyelination and disability in EAE. PLoS One 2009; 4: e7897. 14 Stone TW, Darlington LG. The kynurenine pathway as a therapeutic target in cognitive and neurodegenerative disorders. Br J Pharmacol 2013; 169: 1211–1227. 15 D’Amours D, Desnoyers S, Poirier GG. Poly(ADP-ribosyl)ation reactions in the regulation of nuclear functions. Biochem J 1999; 342: 249–268. 16 Cea M, Cagnetta A, Fulciniti M, Tai YT, Hideshima T, Chauhan D et al. Targeting NAD þ salvage pathway induces autophagy in multiple myeloma cells via mTORC1 and extracellular signal-regulated kinase (ERK1/2) inhibition. Blood 2012; 120: 3519–3529. 17 Holen K, Saltz LB, Hollywood E, Burk K, Hanauske AR. The pharmacokinetics, toxicities, and biologic effects of FK866, a nicotinamide adenine dinucleotide biosynthesis inhibitor. Invest New Drugs 2008; 26: 45–51. 18 Pittelli M, Felici R, Pitozzi V, Giovannelli L, Bigagli E, Cialdai F et al. Pharmacological effects of exogenous NAD on mitochondrial bioenergetics, DNA repair, and apoptosis. Mol Pharmacol 2011; 80: 1136–1146. 19 Munn DH, Mellor AL. Indoleamine 2,3-dioxygenase and tumor-induced tolerance. J Clin Invest 2007; 117: 1147–1154. 20 Potaman VN, Shlyakhtenko LS, Oussatcheva EA, Lyubchenko YL, Soldatenkov VA. Specific binding of poly(ADP-ribose) polymerase-1 to cruciform hairpins. J Mol Biol 2005; 348: 609–615. 21 Lonskaya I, Potaman VN, Shlyakhtenko LS, Oussatcheva EA, Lyubchenko YL, Soldatenkov VA. Regulation of poly(ADP-ribose) polymerase-1 by DNA structurespecific binding. J Biol Chem 2005; 280: 17076–17083. 22 Rapizzi E, Fossati S, Moroni F, Chiarugi A. Inhibition of poly(ADP-ribose) glycohydrolase by gallotannin selectively up-regulates expression of proinflammatory genes. Mol Pharmacol 2004; 66: 890–898.

23 Bai P, Canto C, Oudart H, Brunyanszki A, Cen Y, Thomas C et al. PARP-1 inhibition increases mitochondrial metabolism through SIRT1 activation. Cell Metab 2011; 13: 461–468. 24 Chiarugi A. PARP-1: killer or conspirator? The suicide hypothesis revisited. Trends Pharmacol Sci 2002; 23: 122–129. 25 Pittelli M, Formentini L, Faraco G, Lapucci A, Rapizzi E, Cialdai F et al. Inhibition of nicotinamide phosphoribosyltransferase: cellular bioenergetics reveals a mitochondrial insensitive NAD pool. J Biol Chem 2010; 285: 34106–34114. 26 Pellicciari R, Camaioni E, Costantino G, Formentini L, Sabbatini P, Venturoni F et al. On the way to selective PARP-2 inhibitors. Design, synthesis, and preliminary evaluation of a series of isoquinolinone derivatives. ChemMedChem 2008; 3: 914–923. 27 Tan B, Young DA, Lu ZH, Wang T, Meier TI, Shepard RL et al. Pharmacological inhibition of nicotinamide phosphoribosyltransferase (NAMPT), an enzyme essential for NAD þ biosynthesis, in human cancer cells: metabolic basis and potential clinical implications. J Biol Chem 2013; 288: 3500–3511. 28 Chang CH, Curtis JD, Maggi LB Jr., Faubert B, Villarino AV, O’Sullivan D et al. Posttranscriptional control of T cell effector function by aerobic glycolysis. Cell 2013; 153: 1239–1251. 29 Frauwirth KA, Thompson CB. Regulation of T lymphocyte metabolism. J Immunol 2004; 172: 4661–4665. 30 Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 2009; 324: 1029–1033. 31 Cavone L, Aldinucci A, Ballerini C, Biagioli T, Moroni F, Chiarugi A. PARP-1 inhibition prevents CNS migration of dendritic cells during EAE, suppressing the encephalitogenic response and relapse severity. Mult Scler 2011; 17: 794–807. 32 Shah T, Atwell M, Bird D, Lim CK, Peters TJ, Webster AD. Adenine nucleotide concentrations in peripheral blood lymphocytes from patients with common variable hypogammaglobulinaemia and B-cell chronic lymphatic leukaemia. J Clin Lab Immunol 1988; 26: 9–12.

Immunology and Cell Biology