Up-regulation of nicotinamide phosphoribosyltransferase and increase of NAD+ levels by glucose restriction extend replicative lifespan of human fibroblast Hs68 cells
Abstract Calorie restriction (CR) extends lifespan in a remarkable range of organisms. However, the mechanisms of CR related to the longevity effects are not fully elucidated to date. Using human fibroblast Hs68 (Hs68) cells cultured at a lower level of medium glucose (i.e., glucose restriction; GR) to mimic CR, we investigated the crucial role of nicotinamide phos- phoribosyltransferase (Nampt), nicotinamide adenine dinucleotide (NAD?), and nicotinamide (NAM) in GR-extended replicative lifespan of Hs68 cells. We found that GR extended the lifespan of Hs68 cells, in parallel to significantly increased expression of Na- mpt, intracellular NAD? levels, and SIRT1 activities, and to significantly decreased NAM levels. The lifespan-extending effects of GR were profoundly diminished by FK866 (a noncompetitive inhibitor of Nampt) and blocked by sirtinol (a noncompetitive inhibitor of sirtuins). However, the steady-state intra- cellular NAM level (averaged 2.5 lM) was much lower than the IC50 of NAM on human SIRT1 (about 50 lM). All these results suggest that up-regulation of Nampt play an important role in GR-extended lifespan of Hs68 cells by increasing the intracellular NAD? levels followed by activating SIRT1 activity in Hs68 cells. In contrast, the role of NAM depletion is limited.
Keywords : Calorie restriction · Nicotinamide phosphoribosyltransferase · NAD? · Nicotinamide · FK866 · Hs68 cells
Introduction
Calorie restriction (CR) extends lifespan in a remark- able range of organisms, including yeast, worms, flies, mice, rat, and perhaps in monkeys and humans (Colman and Anderson 2011; Everitt and Le Couteur 2007). Nicotinamide adenine dinucleotid (NAD?), an active form of vitamin B3, is known as an essential cofactor regulating numerous cellular metabolic path- ways (Bogan and Brenner 2008; Sauve 2008; Lin and Guarente 2003). Recently, NAD? has been recognized as a substrate for a growing number of NAD?- dependent enzymes, including yeast Sir2 and human SIRT1 in the sirtuin family (Sauve et al. 2006; Donmez and Guarente 2010; Haigis and Sinclair 2010). As Sir2 and SIRT1 are NAD?-dependent deacetylases, it has been hypothesized that CR could stimulate sirtuin activity and then extend the lifespan of organisms by increasing the levels of intracellular NAD?, a theory known as the NAD?-fluctuation model (Lin and Guarente 2003; Anderson et al. 2003b; Lu and Lin 2010). However, the NAD?-fluctuation theory in yeast has been strongly challenged because CR does not affect intracellular NAD? levels in the caloric- restricted yeast (Anderson et al. 2003b; Lin et al. 2001, 2004) or in the genetic mimics of CR (Anderson et al. 2002). Recently, we reported that a CR mimetic, 2-deoxyglucose (2-DG), can extend the replicative lifespan of Hs68 cells by increasing intracellular NAD? and SIRT1 activity (Yang et al. 2011). These results are in contrast to those found in yeast and support the NAD?-fluctuation theory in human cells. Alternatively, a nicotinamide (NAM)-depletion model for sirtuin regulation in yeast has been proposed (Anderson et al. 2003a). NAM is a non-competitive inhibitor of sirtuins, with an IC50 of about 50 lM for human SIRT1 (Yang et al. 2006). The NAM deple- tion-model states that CR eliminates the negative feedback inhibition on Sir2 activity by NAM via positive regulation of a gene PNC1, which expresses an enzyme nicotinamidase that catalyzes the reaction of NAM to nicotinic acid (Lu and Lin 2010; Anderson et al. 2003a; Yang et al. 2006). However, no homo- logues of PNC1 have been found in vertebrates (Imai 2010). Another point to be considered is that the intracellular concentration NAM must be higher than the IC50 value on sirtuins, if NAM does play a critical role as an endogenous inhibitor of sirtuins. However, the intracellular concentration of NAM is still unknown to date. Although vertebrates lack any obvious homolog of the yeast PNC1, accumulated data support that nicotinamide phosphoribosytrans- ferase (Nampt) may play a similar role, as may pnc1 protein in mammalian cells (Yang et al. 2006; Imai 2010; van der Veer et al. 2007; Borradaile and Pickering 2009). Nampt is the rate-limiting enzyme in the NAD? synthesis salvage pathway, which catalyzes NAM into nicotinamide mononucleotide (NMN) which is then converted to NAD? by nicotin- amide mononucleotide adenylyltransferase (Bogan and Brenner 2008). Theoretically, up-regulation of Nampt would increase intracellular NAD? and decrease NAM levels simultaneously.
Fibroblasts have a finite life in cell culture (the so- called Hayflick limit) (Hayflick and Moorehead 1961), which generally contains 25 mM (4.5 g/L) glucose in the medium (Li et al. 2010). Studies have demonstrated that fibroblasts cultured under the glucose restriction (GR) condition, i.e., at lower levels of glucose in the medium, can extend the replicative lifespan of several types of human fibroblasts including WI-38, IMR-90 and MRC-5 cells (Li et al. 2010; Li and Tollefsbol 2011). In this study, we aimed to elucidate the role of Nampt, NAD?, and NAM in GR-extended lifespan of human fibroblast Hs68 cells. We hypothesized that GR could up-regulate Nampt expression and thus increase intracellular NAD? levels as well as could deplete NAM levels, thereby activating SIRT1 protein and extending cell lifespan. Herein, we measured the intracellular levels of NAD? and NAM and the expression of Nampt in response to GR. We also used FK866 (a noncompetitive inhibitor of Nampt) (van der Veer et al. 2007; Hasmann and Schemainda 2003; Takeuchi et al. 2014) and sirtinol (a noncompetitive inhibitor of sirtuins) (Ota et al. 2006; Grozinger et al. 2001) to confirm the proposed mechanisms of the GR- extended lifespan of Hs68 cells. The IC50 values of FK866 on Nampt inhibition and sirtinol on SIRT1 inhibition are about 51 nM (Takeuchi et al. 2014) and 131 lM (Grozinger et al. 2001), respectively.
Materials and methods
Chemicals
All chemicals used were of analytical grade. NaCl, KCl, NaOH, MgCl2, Na2HPO4, KH2PO4, NaHCO3,Na2CO3, dimethylsulfoxide (DMSO), methanol, per- chloric acid and formaldehyde were purchased from Merck (Darmstadt, Germany). NAM, NAD?, 6-methyl nicotinamide (6-MN), 3-(4,5-dimethylthia- zol-2-yl)-2,5-diphenyltetrazolium (MTT), alcohol dehydrogenase (ADH), N-ethyldibenzopyrazine ethyl sulfate (commonly referred to as phenaxine ethosul- fate), Tirs-HCl, D-(?)glucose, protease inhibitor cocktail and other chemicals were obtained from Sigma (St. Louis, MO, USA).
Cell culture
The human fibroblast Hs68 cells (ATCC® CRL- 1635TM) were purchased from the Cell Culture Center of the Food Industry Research and Develop- ment Institute (Hsinchu, Taiwan) and were regularly cultured in Dulbecco’s modified Eagle medium (DMEM, Gibcol, Grand Island, NY, USA) with 25 mM (4.5 mg/mL) glucose in 75-cm2 flasks with 10 % fetal bovine serum and 1.0 mM pyruvate at 37 °C in a humidified incubator under 5 % CO2. For the GR condition, DMEM (Gibcol, Grand Island, NY, USA) without glucose was used, to which was added different concentrations (e.g., 5.5, 1.1 and 0.55 mM) of glucose. Those glucose concentrations in the medium lower than normal glucose concen- tration of 25 mM were considered as GR. Hs68 is one of a series of human foreskin fibroblast lines developed at the Naval Biosciences Laboratory (NBL) in Oakland, CA. Hs68 was obtained from an apparently normal Caucasian newborn male with a finite lifespan.
Assays of cytotoxicity
Cytotoxicity was assayed using MTT colorimetric assay (Yang et al. 2011). The cells were cultured in 12-well plates until confluence and incubated with tested agents for 24 h. After removing the cultured medium by aspiration, a volume (0.5 mL) of DMEM medium (300 lg MTT/mL) was added to the cells. After incubation for 2 h, the medium was aspirated, and the formazan was extracted in 0.5 mL of DMSO. A volume (100 lL) of the extract was transferred into a 96-well microplate, and the absorbance was measured at 570 nm using a microplate reader (Micro Quant, Bio-Tek, USA).
Determination of intracellular NAD? levels
Intracellular NAD? was determined by an acid extraction method followed by an enzymatic cycling method (Yang et al. 2011). Briefly, cells were added 500 lL of an ice-cold solution of 7.0 % perchloric acid. After centrifugation (120009g, 10 min), the supernatant was neutralized with 3 N NaOH and 1 M phosphate buffer. The protein pellet was reconstituted by adding proper amount of 1 N NaOH (equal to the volume of supernatant with the neutralized solution), and concentrations of protein were determined by using a commercial kit later (Novagen, USA). The level of NAD? was determined using a 96-well microplate, in which 100-lL reaction mixture per well was used. The reaction mixture contained 10 lmol Tris-HCl (pH 8.0), 0.4 lmol PES, 0.05 lmol MTT, 0.03 mg ADH and 60 lmol ethanol with 20 lL cell extracts. The rate of increase of absorbance at 570 nm within 10 min was detected by a microplate reader (Micro Quant, Bio-Tek, USA) and calibrated using commercial NAD? dissolved in the neutralized acid solution as standard. NAD? level (fmole/lg protein) was calculated as: NAD? concentration of sample (fmole/mL)/protein concentration of sample (lg/mL).
Determination of intracellular NAM levels
Intracellular levels of NAM were analyzed by an LC/ MS/MS system with multiple reaction monitoring analysis mode. Sample was prepared by modifications from the method proposed by Stratford and Dennis (Stratford and Dennis 1992). Briefly, 1 9 106 cells were collected and were extracted by adding 1 mL of methanol which contained 10 nmole of 6-methylnic- otinamide (6-MN) as internal standard. The extracts
were dried by N2 gas at 37 °C, reconstituted by adding 200 lL H2O, and then analyzed by using LC/MS/MS, as described elsewhere (Lee et al. 2012). The gradient chromatographic separation was conducted by using reverse-phase HPLC under the following conditions: column, phenomenex Synergi Polar-RP (4-lm parti- cle size, 150 mm 9 4.6 mm; Phenomenex, Torrance, CA, USA); flow rate, 0. 5 mL/min; injection volume, 20 lL; mobile phase A, of water with 0.1 % formic acid; and mobile phase B, methanol with 0.1 % formic acid. The gradient program was as follows: initial, 10 % mobile phase B; 0–2.5 min, from 10 to 90 % mobile phase B; 2.5–3.5 min, 90 % mobile phase B; 3.6–7 min, 10 % mobile phase B. The precursor ions for NAM and 6-MN were m/z 123.0 and 137.0, respectively. The quantitative ion for NAM was m/z 80.0 and for 6-MN was m/z 94.1. The qualitative ions for NAM were m/z 78.1 and 53.2, and for 6-MN was m/z 92.1. Other instrument conditions were referred to the preliminary publication (Lee et al. 2012).
Determination of Nampt expression
After various treatments, cell lysates were collected in microcetrifuge tubes by using CytoBusterTM Protein Extraction Reagent (Novagen, USA) containing 1 % protease inhibitor cocktail (CalBiochem, Germany). Protein concentrations were determined using the BCA protein assay kit (Novagen, USA). Equal amounts of protein were resolved by SDS-PAGE, transferred onto PVDF membrane (Bio-Red, USA), and immunoblotted for Nampt (cell signaling, Denmark) and b-actin (Novus, USA). Protein was visualized using the ECL kit (Amersham Biosciences, UK). Nampt expressions (fold of control) were calculated as: (Nampt/b-actin of sample)/(Nampt/b-actin of control).
Assay of replicative lifespan
Cells were serially cultured in a 10-cm dish with 1.0 9 105 cells per dish in 10-mL cultured medium. The cells were subcultured exactly once per week and the cell numbers were counted by Trypan blue exclusion assay. The population doubling levels (PDLs) were calculated as log2 (Nt/No) (Yang et al. 2011), where Nt and No were defined as the total count of cells at the time of harvesting and seeding, respectively. The cumulative PDLs (CPDs) were obtained by summation of total PDL before a given passage time. Cells were considered senescent, when the Nt/No ratios were less than 1.5 for two consecutive passages. As the precise level of CPDs of the cells was not specified by the supplier, we defined CPDs at the initial passage as zero, and we used the additional CPDs to represent the doubling levels after the initial passage.
Determination of cellular SIRT1 activity
Cellular SIRT1 activity was determined by using a commercialized kit method (Yang et al. 2011). For the determination, cell extracts were obtained by using a lysis buffer (50 mM Tris-HCl (pH 8), 125 mM NaCl, 1 mM DTT, 5 mM MgCl2, 1 mM EDTA, 10 % glycerol, and 0.1 % NP40 supplemented with 1 mM PMSF and protease inhibitors mix). SIRT1 activities were determined using the SIRT1 Direct Fluorescent Screening Assay Kit (Cayman, Ann Arbor, MI). In brief, 25 lL of assay buffer (50 mM Tris-HCl, pH 8.0, containing 137 mM NaCl, 2.7 mM KCl, and 1 mM MgCl2), proper volume of extracts (containing 30 lg protein), and 15 lL of substrate solution (i.e., assay buffer containing 125 lM peptide solution comprising amino acids 379–382 of human p53 conjugated to aminomethylcoumarin) were added to all wells. The plate was incubated in a shaker for 45 min. Then, 50 lL of Stop/Development solution was added to each well, and the mixture was incubated for 30 min at room temperature. The fluorescence of the plate was then determined using a fluorometer (Flx800, Bio-Tek, USA) at an excitation wavelength of 360 nm and an emission wavelength of 460 nm. Relative SIRT1 activities were calculated as: (fluorescent signal of sample/fluorescent signal of control) * 100 %.
Effect of GR on the levels of intracellular NAM
Cell extract was analyzed for the intracellular NAM levels by our LC/MS/MS system. As shown in Fig. 3a, the intracellular level of NAM in the control was about 13 pmol/106 cells and was significantly decreased with increasing incubation time. In addition, GR0.55 significantly decreased NAM level starting on day 1 and remained unchanged until day 5. A representative chromatograph was shown in Fig. 3b.
The steady-state intracellular levels of NAM in control (i.e., 13 pmol/106 cells) were further esti- mated. The total volume of the cells was calculated group mean comparison using SPSS v.10.0 software (SPSS Inc., Chicago, USA). P values less than 0.05 were considered statistically significant.
Fig. 2 Effects of glucose restriction (GR) on intracellular NAD? levels. Hs68 cells at passage 26 were incubated in the medium containing 25 mM glucose (control) or 5.5 mM (GR0.55) and 0.55 mM (GR0.55) glucose in 10-cm dish for 0–7 days. Values (mean ± SD, n = 3) not sharing a common letter at the same incubation time are significantly different (P \ 0.05).
Results
Effect of GR on the cells viability
The cytotoxic effects of different concentrations of glucose (25, 5.5, 1.1 and 0.55 mM) in the medium (i.e., different extents of GR) were tested. After incubation for 24 h, cell viabilities were detected by MTT method. As shown in Fig. 1, none of the indicated GR conditions had any cytotoxic effect on Hs68 cells.
Effect of GR on the levels of intracellular NAD?
As shown in Fig. 2, the intracellular NAD? levels increased in proportion to the extent of GR, i.e., intracellular NAD? levels in Hs68 cells increased with decreasing medium glucose concentrations. In addi- tion, the intracellular NAD? levels increased with increasing incubation time, when cells were cultured in GR conditions. Interestingly, the intracellular NAD? level of Hs68 cells incubated with 5.5 and 0.55 mM medium glucose was not significantly higher (P [ 0.05) than that of the control level, unless the incubation time was longer than or equal to 5 days.
Effect of GR on the expression of Nampt
To monitor the time and concentration-effects of GR on the expression of Nampt, we cultured cells under GR condition. As showed in Fig. 4a, b, the protein levels of Nampt were significantly increased under GR0.55 condition from day 1 to day 5. The expression of Nampt in cells treated with GR0.55 was about 1.19-, 1.48- and 1.47-fold on days 1, 3 and 5, respectively, as compared to day 0 of the control (Fig. 4a, b). The expression of Nampt also was significantly increased from day 1 to day 5 for the control cells, but all expression levels of Nampt were significantly lower than that treated with GR0.55 on the same day (P \ 0.05; Fig. 4b). In addition, the expression levels of Nampt were proportional to the extent of GR (Fig. 4c).
FK866 diminishes the effects of GR on cell lifespan and intracellular NAD?
We further examined whether the lifespan-extended effect of GR could be inhibited by co-treating cells with FK866. If indeed Nampt plays a role in the longevity of GR on Hs68 cells, the effects of GR- extended cell lifespan should be profoundly inhibited by FK866. Results reveal that both 10 and 100 nM of FK866 significantly inhibited GR-extended lifespan (Fig. 5a). Treatment with 100 nM FK866 also dra- matically decreased the cell density and had a more senescent morphology i.e., becoming an aging cell type, even for cells grown under GR condition, whereas treatment with 10 nM FK866 was much less effective on cell morphology and density (Fig. 5b).
The much lower cell density induced by FK866 was likely due to its senescent effect (i.e., cells became a non-divided situation), since the concentrations of FK866 used (B100 nM) did not induce any cytotox- icity to Hs68 cells (Fig. 5c). In addition, FK866 diminished lifespan-extending effect of GR, which was accompanied by declined intracellular NAD? levels. As shown in Fig. 5d, the intracellular NAD? levels in GR condition were significantly decreased by FK866. Interestingly, the lifespan-shortening and NAD?-depleting effects of 10 nM FK866 (a lower concentration than its IC50 on Nampt) were dramat- ically antagonized by the treatment with GR0.55 (Fig. 5a, b, d). Treatment with 10 nM FK866 for 5 days alone depleted NAD? dramatically; however, co-treatment of 10 nM FK866 with GR0.55 only decreased intracellular NAD? moderately.
The role of SIRT1 in GR-extended lifespan of Hs68 cells
As showed in Fig. 6a, the SIRT1 activities were proportional to the extent of GR. The relative activities of SIRT1 in cells treated with GR5.5 and GR0.55 were about 14 (P \ 0.05) and 36 % (P \ 0.05) higher, respectively, than those of control. The positive control, i.e., cells cultured in the medium containing 200 lM NAD?, was found to increase SIRT1 activity by about 40 % as compared to that of control (P \ 0.05). In addition, the SIRT1 protein expression detected by western blotting was not significantly affected by different extent of GR (data not shown). Furthermore, we found that the lifespan-extended effects of GR 0.55 were dramatically blocked by the co-treatment of both 10 and 100 lM sirtinol (Fig. 6b). Moreover, both 10 and 100 lM sirtinol shortened the cell lifespan, even for cells grown under GR0.55 condition. Sirtinol at the concentrations used (B100 lM) did not result in cytotoxicity in Hs68 cells (data not shown).
Discussion
In this study, we investigated the crucial role of lifespan of human Hs68 cells. We found that GR increased intracellular NAD? levels in a concentra- tion- and time-dependent manner, in parallel to increased lifespan of Hs68 cells. The intracellular concentration of NAM (2.5 lM) was decreased by GR, but we argue that this effect of GR is likely to have little meaning, as this steady-state level of NAM in cells is relatively low, as compared to the IC50 of NAM on human SIRT1 (50 lM) (Yang et al. 2006). The result suggests that the role of NAM as an endogenous inhibitor of sirtuins in Hs68 cells is limited. In addition, Nampt was up-regulated by GR0.55, with a 1.5-fold increase in protein expression on day 3, as compared to that on day 0 of the control. The use of through increased intracellular NAD? levels, which then activated SIRT1 in Hs68 cells.
Fig. 4 Effects of glucose restriction (GR) on the expression of Nampt protein in Hs68 cells. Cells were cultured in the medium containing 25 (control), and 0.55 mM (GR0.55) glucose for 0–5 days. The cell lysates were used for measurement of Nampt expression by western blotting method (a), and the data obtained above are further presented as a curve plot of Nampt expression vs. incubation time (b). The concentration effect of GR on Nampt expression was measured after incubation for 3 days (c). Values (mean ± SD, n = 3) not having a common letter are significantly different (P \ 0.05) among different treatments except for b where the data are compared at the same incubation time.
Fig. 5 Effects of FK866 on replicative lifespan and NAD? levels in Hs68 cells treated with glucose restriction (GR). Cells at passage 24 were cultured serially in the medium containing 25 mM glucose (control) or 0.55 mM glucose (GR0.55) with or without 10 and 100 nM FK866 (F10 and F100). Cumulative growth curves were obtained (a). On day 21 during the serial.
FK866 to inhibit Nampt activity dramatically inhib- ited the effects of GR on Hs68 cells including the increase of intracellular NAD? levels and the exten- sion of cell lifespan. Furthermore, SIRT1 activities were proportionally increased by the increasing extent of GR, and the lifespan-extending effect of GR was blocked by sirtinol. All these results suggest that up- regulation of Nampt plays an important role in the GR- extending effect on lifespan in Hs68 cells, possibly culture, the cells were photographed under a microcope (b). The cytotoxicity of FK866 (0–100 nM) was determined after incubation for 24 h by the MTT method (c). The intracellular NAD? levels were measured after incubation for 5 days (d). Values (mean ± SD, n = 3) not sharing a common letter are significantly different (P \ 0.05).
Nampt is the rate-limiting enzyme in the salvage pathway of NAD? biosynthesis. Human Nampt was originally identified as pre-B-cell colony-enhancing factor (PBEF), a putative cytokine for early B cell proliferation and has been suggested to have insulin- mimetic activity like that of adipokine secreted by visceral fat (called visfastin) (Dahl et al. 2012). In this study, we found that GR time- and dose-dependently increased the expression of Nampt; besides, FK866 affected GR-induced longevity profoundly, suggest- ing that Nampt plays a critical role in response to GR in Hs68 cells. A similar result has been reported by Fulo et al. (2008) in which they showed that glucose restriction inhibits the differentiation of skeletal myoblasts through Nampt induction and activation of SIRT1. An animal study also showed that fasting increases the expression of Nampt in the liver of mice (Hayashida et al. 2010). Given these promising results, Nampt is likely to play a role in CR-induced effects and is a connector between GR and SIRT1 through modulating the intracellular levels of NAD?.
Fig. 6 The role of SIRT1 in glucose restriction (GR)-extended lifespan of Hs68 cells. a Effects of GR on SIRT 1 activity in Hs68 cells cultured in the medium containing 25 (control), 5.5 (GR5.5) and 0.55 mM (GR0.55) glucose, and 200 lM NAD? (N200; a positive control) for 5 days. Values (mean ± SD, n = 3) not having a common letter are significantly different (P \ 0.05). b, c Effects of sirtinol on GR-extended lifespan of Hs68 cells cultured under a GR0.55 condition with or without incubation of 10 and 100 lM sirtinol (S10 and S100). The cumulative growth curves were obtained (b). On day 7, the cells were photographed under a microcope to observe cell morphol- ogy and density.
To the best of our knowledge, there also has no published study in the literature that has analyzed the intracellular NAM level either in tissues in vivo or in cultured cells in vitro. It is worth noting that the precise NAM levels in cells are difficult to determine because intracellular NAM must be extracted by certain solvent or buffer before the analysis. This crucial process will dramatically dilute the concentra- tion of NAM. To overcome this problem, we used the total volume of the test cells to determine the dilution factor after the extraction. The precise NAM levels in cells were then estimated by correction with the dilution factor. As a result, we found that the intracellular level of NAM in Hs68 cells (averaged 2.5 lM) was much lower than the IC50 of NAM (50 lM) on SIRT1. Indeed, studies have revealed that the blood levels of NAM in humans are in the nano- molar levels (Horsman et al. 1993; Stratford et al. 1992; Dragovic et al. 1995; Hoskin et al. 1995), and are about 4.6 lM in blood plasma of mice (Hara et al.2011). All of these results suggest that the critical role of NAM depletion related to the calorie restriction is indeed questionable, and this issue has been discussed elsewhere (Revollo et al. 2004; David Adams and Klaidman 2007). In addition, the averaged protein content of 1 9 106 cells is about 2,000 lg. Thus,
13 pmol/1 9 106 cells is roughly equal to 6.5 fmole/ lg protein, which is only about 1/185 of the NAD? level in Hs68 cells.
Moreover, the NAD?-fluctuation model to regulate sirtuin activities in yeast has been strongly disputed because GR does not increase the NAD? level in yeast (Anderson et al. 2003b; Lin et al. 2001; Lin et al. 2004; Anderson et al. 2002). Our present data support the NAD?-fluctuation model, i.e., NAD? regulates sirtu- ins in human Hs68 cells responding to GR. It is unclear why GR increases the NAD? levels in human Hs68 cells but not in yeast. As sirtuins are a family of conserved genes in organisms, it is reasonable to expect that a universal mechanism for regulation of sirtuins may exist. In this study, we found that GR indeed did not significantly increase intracellular NAD? of Hs68 cells unless the incubation time was longer than or equal to 5 days. This result indicates that GR can only increase intracellular NAD? level of Hs68 cells after extended incubation time. However, we found that Nampt expression was significant increased after one day of GR, and there existed a 4-day gap between the first event (the increase in Nampt expression levels) and the last event (the increase in NAD? levels). The reason for this gap appears to be complicated. It is possible that the conversion of produced NMN to NAD? needs a period of time, or that the produced NAD? can readily be converted to other molecules such as NADH. It is worth noting that studies in yeast on the effect of GR on intracellular NAD? levels are usually conducted within a short period of incubation time (Lin and Guarente 2003; Anderson et al. 2003b; Lin et al. 2001). For example, Anderson et al. (2003b) used in vivo C13 NMR to monitor the fluctuation of intracellular NAD? and the monitoring time in their experiment was only 70 min.
It has been shown that FK866-induced NAD? depletion and cell death of NIH 3T3 fibroblasts can be restored by extracellular application of NAD? (Bil- lington et al. 2008). In the present study, FK866 effectively decreased intracellular NAD? levels and drastically shortened the replicative lifespan of Hs68 cells, suggesting that the lifespan-shortening effect of FK866 is primarily due to the decrease in intracellular NAD? levels via Nampt inhibition. It is interesting that this effect of FK866 was countervailed by GR, as we showed that the NAD?-depleting and the lifespan- shortening effects of 10 nM FK866 was drastically restored by co-treatment with GR0.55. The result suggests an important role of NAD? level in GR- extended cell lifespan. However, GR0.55 did not antagonize the effects of FK866, when the concentra- tion of FK866 was elevated to 100 nM. It is possible that GR increases the expression of Nampt, but the low concentration of FK866 (10nM) was not sufficient to inhibit the increased activity of Nampt. Thus, the intracellular NAD? was effectively maintained by GR and not drastically depleted by FK866. All these results suggest the crucial role of Nampt and NAD? in GR-extended lifespan of Hs68 cells. Moreover, NAD? is known to be synthesized by other pathways: starting from the NAD?-essential amino acid tryptophan and from nicotinic acid (Bogan and Brenner 2008). Therefore, further studies are warranted to investigate the effects of GR on the enzymes involved in other pathways of NAD? synthesis to increase intracellular NAD? level.
Alternatively, intracellular NAD? may also be increased by the limited supply of energy source of glucose during GR. However, our present finding that 100 nM FK866 could diminish the lifespan- extending effect of GR suggests that the role of GR to increase intracellular NAD? by glucose metabolism via glycol- ysis is limited. Furthermore, NADH also may play a crucial role in regulation of sirutin activity and CR- extended lifespan of yeast. As NADH is a competitive inhibitor of sirtuins, an increase in the NADH level can result in inhibition of sirtuin activities (Lin et al. 2004). However, it has been shown that the IC50 of NADH for Sir2 is 11 mM (Schmidt et al. 2004), which is much higher than the concentration of NADH in cells, i.e. 50–100 lM (Anderson et al. 2002; Srivastava and Bernhard 1987). Thus, the role of NADH on the regulation of sirtuin activities is likely very limited.
Conclusions
In summary, the results obtained from this study suggest that GR up-regulates Nampt expression, which then possibly increases intracellular levels of NAD? and activates SIRT1, thereby extending the lifespan of human fibroblast Hs68 cells. In contrast, the depletion of intracellular levels of NAM only plays a limited role in GR-extended lifespan of Hs68 cells.