[CANCER RESEARCH 63, 2109–2117, May 1, 2003]

Redox Regulation of the G1 to S Phase Transition in the Mouse Embryo FibroblastCell Cycle1

Sarita G. Menon, Ehab H. Sarsour, Douglas R. Spitz, Ryuji Higashikubo, Mary Sturm, Hannah Zhang, andPrabhat C. Goswami2

Free Radical and Radiation Biology Program, Department of Radiation Oncology, University of Iowa, Iowa City, Iowa 52242 [S. G. M., E. H. S., D. R. S., M. S., H. Z., P. C. G.,and Department of Radiation Oncology, Washington University Medical School, St. Louis, Missouri [R. H.]

ABSTRACT

The hypothesis that intracellular oxidation/reduction (redox) reactionsregulate the G0-G1 to S-phase transition in the mouse embryonic fibro-blast cell cycle was investigated. Intracellular redox state was modulatedwith a thiol-antioxidant, N-acetyl-L-cysteine (NAC), and cell cycle pro-gression was measured using BrdUrd pulse-chase and flow cytometricanalysis. Treatment with NAC for 12 h resulted in an �6-fold increase inintracellular low-molecular-weight thiols and a decrease in the MFI of anoxidation-sensitive probe, dihydrofluorescein diacetate, indicating a shiftin the intracellular redox state toward a more reducing environment.NAC-induced alterations in redox state caused selective delays in progres-sion from G0-G1 to S phase in serum-starved cells that were serumstimulated to reenter the cell cycle as well as to inhibit progression fromG1 to S phase in asynchronous cultures with no significant alterations inS phase, and G2�M transits. NAC treatment also showed a 70% decreasein cyclin D1 protein levels and a 3–4-fold increase in p27 protein levels,which correlated with decreased retinoblastoma protein phosphorylation.Cells released from the NAC treatment showed a transient increase indihydrofluorescein fluorescence and oxidized glutathione content between0 and 8 h after release, indicating a shift in intracellular redox state to amore oxidizing environment. These changes in redox state were followedby an increase in cyclin D1, a decrease in p27, retinoblastoma proteinhyperphosphorylation and subsequent entry into S phase by 8–12 h afterthe removal of NAC. These results support the hypothesis that a redoxcycle within the mammalian cell cycle might provide a mechanistic linkbetween the metabolic processes early in G1 and the activation of G1-regulatory proteins in preparation for the entry of cells into S phase.

INTRODUCTION

In recent years, increasing evidence has suggested that intracellularoxidation-reduction (redox) reactions play a critical role in the regu-lation of several physiological processes including cell proliferation(1), senescence (2), differentiation (3), and apoptosis (4). ROS3 in-cluding superoxide, hydrogen peroxide, hydroxyl radical, singlet mo-lecular oxygen, and organic hydroperoxides are believed to be con-tinuously generated intracellularly as byproducts of O2 metabolismand have traditionally been thought of as unwanted and toxic byprod-ucts of living in an aerobic environment (5, 6). However, results fromseveral studies suggest that the metabolic production of ROS is tightlyregulated and serves a physiological function during mitogenic stim-

ulation of cultured cells (1, 5–9). It has been proposed that theendogenous production of ROS operates as a key signaling process inthe cascade of events leading to cell proliferation after stimulationwith platelet-derived growth factor (7), EGF (8), cytokines and anti-gen receptors (9), and oncogenic Ras (10). Although these studiesprovide evidence for a possible role of redox signaling in the transi-tion of quiescent cells to proliferation during mitogenic stimulation, itis currently unknown whether changes in intracellular redox stateinfluence progression through specific phases of the cell cycle and theidentity of specific cell-cycle-regulatory processes that may be redox-regulated during the normal cell cycle.

Cellular growth is a tightly regulated sequence of transitions fromG0-G1 to S phase to G2 to M phases involving the sequential activa-tion of CDK activities (11–14). Progression from G0-G1 to S phase islargely regulated by the D-type cyclins (in particular cyclin D1) inassociation with the CDKs CDK4/6 (14). This involves promoting thesynthesis and stability of the cyclin subunit as well as decreasing thelevels of CDK inhibitors. Cyclin D1/CDK4,6 kinase complex isactivated on removal of inhibitory serine and threonine phosphatesfrom CDK by Cdc25A phosphatase. Active cyclin D1/CDK4,6 kinasecomplex partially phosphorylates the Rb protein, which causes therelease of the E2F family of proteins, initiating the transcription ofE2F-mediated gene expression during the transition from G1 to Sphase. The inhibition of cyclin D1 expression prevents the transitionof cells from G1 to S phase, whereas ectopic expression of cyclin D1accelerates G1 progression (15, 16). Thus, it is possible that changesin intracellular redox state could provide a mechanistic link couplingnecessary metabolic processes early in G1 and the activation ofG1-regulatory proteins, leading to the redox-sensitive progressionfrom G1 to S phase.

The present study was designed to test the hypothesis that theintracellular redox state has a specific regulatory role in cell cycleprogression from G0-G1 to S phase and to begin to identify thespecific cell-cycle-regulatory processes amenable to redox regulation.Results from this study show that progression from (a) G1 to S phasein asynchronously growing exponential cultures, and (b) G0-G1 to Sphase in serum-stimulated quiescent cells and cells released from aNAC-induced G1 arrest, is associated with a transient increase inpro-oxidant production that appears to mediate the activation of theG1-regulatory proteins. Furthermore, the inhibition of these signalsusing the thiol-reducing agent NAC blocks entry into S phase as wellas blocking the activation of G1-regulatory proteins. These resultssupport the hypothesis that redox-sensitive signaling in early G1

controls the progression into S phase via the activation of G1-regula-tory proteins.

MATERIALS AND METHODS

Cell Culture and Synchronization. MEF clones [12(1), wild type p53phenotype] and [10(1), mutant p53 phenotype] were a gracious gift from Dr.Arnold Levine’s laboratory Princeton University, Princeton, New Jersey, USAand were cultured in DMEM supplemented with 10% fetal bovine serum,nonessential amino acids (Life Technologies, Inc.), and antibiotics (penicillinand streptomycin). Cells were cultured at 37°C in a humidified incubator with

Received 8/27/02; accepted 3/5/03.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by NIH Grant CA69593 and a departmental support (to P. C. G.), and NIHGrant HL51469 (to D. R. S.) and CA66081.

2 To whom requests for reprints should be addressed, at Free Radical and RadiationBiology Program, Department of Radiation Oncology, University of Iowa MedicalSchool, B180 Medical Laboratories, Iowa City, Iowa 52242. Phone: (319) 384-4666; Fax:(319) 335-8039; E-mail: goswamip@mail.medicine.uiowa.edu.

3 The abbreviations used are: ROS, reactive oxygen species; BrdUrd, bromodeoxyuri-dine; NAC, N-acetyl-L-cysteine; NPM, N-(1-pyrenyl) maleimide; MFI, mean fluorescenceintensity; CDK, cyclin-dependent kinase; Rb, retinoblastoma protein; MEF, mouse em-bryonic fibroblast; BSO, buthionine-(S,R)-sulfoximine; HPLC, high-performance liquidchromatography; GSH, reduced glutathione; GSSG, oxidized glutathione; �GGC, �-glutamyl cysteine; DFH-DA, dihydrofluorescein diacetate; PI, propidium iodide;FACS, fluorescence-activated cell sorter.

2109

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

5% CO2. Stock solutions of NAC (Sigma Chemicals) were adjusted to pH 7.0with sodium bicarbonate, and appropriate aliquots were added to the cell-culture medium. The toxicity of NAC was determined by single-cell colony-forming assay, and no toxicity was observed in the concentration ranges (1–20mM) used in this study (data not shown). Asynchronously growing cell cultureswere incubated simultaneously with NAC and 1 mM BSO, an inhibitor of�-glutamylcysteine synthase, for the inhibition of glutathione synthesis.

Measurement of Intracellular Glutathione and NAC Levels. Intracellu-lar GSH and GSSG as well as NAC levels were assayed using previouslypublished spectrophotometric and HPLC assays (17, 18). Monolayer cells werescrape-harvested in PBS at 4°C and centrifuged, and the cell pellet was storedfrozen at �80°C. Cell pellets were homogenized in 50 mM potassium phos-phate buffer (pH 7.8) containing 1.34 mM diethylenetriaminepenta-acetic acid.Total glutathione content was determined in sulfosalicylic acid extracts (5%SSA, w/v) by the method of Anderson (19). GSH and GSSG were distin-guished by the addition of 2 �l of a 1:1 mixture of 2-vinylpyridine and ethanolper 30 �l of sample, followed by incubation at room temperature for 1.5 hbefore the addition of the sulfosalicylic acid. Amounts of individual thiols werenormalized per mg protein in whole cell extract.

HPLC (HPLC system; Shimadzu Scientific Instruments, Inc., Columbia,MD) with fluorescent detection was used to determine intracellular low-molecular-weight thiol levels after previously published assays (17). Fiftypmol to 100 fmol of standard thiols (NAC, GSH, �GGC, and cysteine) werederivatized with NPM and resolved using a 15-cm C18 Reliasil column(Column Engineering, Ontario, CA) to generate standard curves. Retentiontimes of individual thiols were determined to be 7.7 min (NAC), 20.1 min(GSH), 24.7 min (�GGC), and 30.9 min (cysteine) using this system. Totalcellular protein extracts prepared from control and NAC-treated cells weretreated with NPM and were analyzed by HPLC. All of the biochemicaldeterminations were normalized to the protein content of whole-cell homoge-nates by using the method of Lowry et al. (20).

Measurement of Intracellular Prooxidant Production. MEF cells weretrypsinized and pelleted at 200–400 � g for 5 min and the pellets wereresuspended in 1 ml of PBS supplemented with 5.5 mM glucose at 37°C. Cellswere stained with 10 �M DFH-DA (Molecular Probes, Eugene, OR) for 15 minat 37°C. DFH-DA stock solutions were prepared in DMSO, and cells that werestained with 0.2% DMSO alone were included as a negative control forDFH-DA staining. DFH-DA is a low-molecular-weight nonfluorescent andnonpolar compound that enters cells freely. On entering the cell, the intracel-lular esterases cleave the diacetate bond, resulting in a polar and nonfluores-cent DFH. Intracellular prooxidants oxidize DFH to the fluorescent compoundDF. The intensity of green fluorescence of DF is an indicator of the levels ofprooxidant production. Changes in DFH fluorescence were detected using aFACS (Becton Dickinson, San Jose, CA) equipped with 15-mW and 488-nmargon-ion laser. Excitation was performed at 488 nm and emission at 530 nmand 30-band pass filter. Data from 20,000 cells were collected, and geometricmean fluorescence was calculated using the Cell Quest software (BectonDickinson). MFIs of the unstained cell populations were subtracted from theMFIs of the stained cell populations, and the variations in geometric MFIswere calculated relative to control cells at the time of experimental manipu-lations. The specificity of prooxidant-mediated changes in DFH fluorescencewas determined by staining cells in duplicate dishes with carboxy DCFDA[C-369, 5-(and-6-)-carboxy-2�-7�-dichlorofluorescein diacetate (MolecularProbes, Eugene, OR)]. Carboxy DCFDA is an oxidation-insensitive compoundthat is nearly chemically identical to the oxidation-sensitive compoundDFH-DA except that it lacks the oxidation-sensitive site. Thus, it is anappropriate control for changes in DFH-DA uptake, ester cleavage, or effluxthat might result from various experimental manipulations.

DFH fluorescence was also used to measure prooxidant production duringthe transition of cells from the quiescent phase to the proliferation cycle.Asynchronously growing exponential cultures of MEFs were synchronized byculturing cells in medium containing 0.1% serum for 24 h and serum stimu-lated to reenter the cell cycle. Cells were harvested at representative times afterserum stimulation, stained with DFH-DA, and assayed by flow cytometry asdescribed above. DFH-DA-stained cells were also visualized using a Bio-RadMRC-600 laser-scanning confocal microscope (Central Microscopy Researchand Learning Facility at the University of Iowa).

BrdUrd Pulse-Chase Flow Cytometric Assay for Measurements of CellCycle Phase Transits. BrdUrd pulse-chase flow cytometric assay for cellcycle phase transitions were performed according to previously publishedmethodology (21, 22). Asynchronously growing cell cultures in 100-mm tissueculture dishes were pulse-labeled with 10 �M BrdUrd for 30 min at 37°C in anincubator. At the end of the pulse labeling, monolayer cultures were washedwith sterile, warm PBS to remove any unincorporated BrdUrd from themedium. BrdUrd-labeled cells were either collected immediately at the end ofthe BrdUrd-labeling or continued in culture in regular growth medium sup-plemented with 10 �M thymidine and cytidine (21, 23). Cells were harvestedat regular intervals by trypsinization, washed once with PBS, and fixed in 70%ethanol. Fixed cells were stored at 4°C before analysis by flow cytometry.

Ethanol-fixed cells were washed with PBS and incubated with pepsin (0.4mg/ml in 2 N HCl) for 30 min at room temperature. After neutralization with 0.1M borate buffer, nuclei were isolated by centrifuging the samples at 200–400 � gin a Beckman centrifuge set at 4°C for 5 min. The pellet was washed once withPBS and incubated with anti-BrdUrd antibody (1:20; Becton Dickinson Immuno-cytometry Systems, San Jose, CA) for 1 h at room temperature. At the end of theincubation, samples were diluted with 1 ml of PBS and centrifuged at 200–400 � g. Nuclei were then incubated with FITC-conjugated goat antimouse IgGfor 1 h and were digested with RNase A (0.1 mg/ml) for 30 min at roomtemperature. The nuclei were counterstained with 20 �g/ml PI for 1 h, and stainedcells were analyzed on a FACS 440 flow cytometer. The FACS 440 flowcytometer is equipped with a Coherent I90-5 UV laser operating at 300 mW andat 488 nm excitation wavelengths. Red fluorescence from PI was detected througha 640-nm long pass filter, and green fluorescence from FITC was detected througha 525-nm band pass filter. Data from a minimum of 20,000 nuclei were acquiredin list mode and were processed using Cytomation software (Cytomation, Ft.Collins, CO). The acquired data were displayed as dual-parameter PI versuslog-FITC histograms and three compartments (BrdUrd-positive S-phase cells,BrdUrd-negative G1, and BrdUrd-negative G2 phases) identified using Cytomationsoftware. The fraction of cells in G1 and G2 were used as a measurement of G1 andG2 transit time. Transit through S phase was measured by determining relativemovement as described earlier by Begg et al. (24). Relative movement wascalculated from the mean DNA content of undivided BrdUrd-positive cell popu-lation normalized to mean DNA content of G1 and G2 population. The fraction ofBrdUrd-positive cells that had divided was also calculated for measurements of thetransit of the cells through S phase, G2, and M.

For staining with PI only, cells were harvested by trypsinization and fixedin 70% ethanol. Approximately 1 � 106 cells were incubated with 100 �l ofRNase A for 30 min at room temperature followed by staining with PI (30�g/ml) for 60 min. Samples were analyzed by flow cytometry, and cell cyclephase distribution was determined using CellQuest software (Becton Dickin-son Immunocytometry Systems).

Immunoblotting. Total cellular proteins were separated by SDS-PAGEand transferred to polyvinylidene difluoride membrane (BIORAD Labs.).Blots were incubated with antibodies to cyclin D1 (G124–326; PharMingen),p27 (sc-1641; Santa Cruz Biotechnology), and Rb (G3–245; PharMingen).Immunoreactive polypeptide was visualized using horseradish peroxidase-conjugated secondary antibodies and enhanced-chemiluminescence detectionreagents (Amersham Pharmacia Biotech.) following manufacturer-suppliedprotocols. Blots were reprobed with antibodies to actin (sc-1615; Santa CruzBiotechnology) for comparison of results.

Data Analysis. Experiments were repeated two or three times, and datawere calculated as means, SDs, and SE. Statistical comparisons were carriedout using two-tailed Student’s t test and ANOVA. Ps were calculated with 95%confidence intervals.

RESULTS

Redox-sensitive Progression from G1 to S Phase in Exponen-tially Growing Asynchronous MEFs. A flow cytometry-basedBrdUrd pulse-chase assay (Fig. 1) was used to monitor the transit of cellsthrough each phase of the cell cycle. Exponentially growing monolayercell cultures were pulse-labeled with BrdUrd, an analogue of thymidine,for 30 min and were chased in regular growth medium containingthymidine and cytidine. Thus, cells synthesizing DNA (S-phase cells,BrdUrd positive; Fig. 1A, box 1) will incorporate BrdUrd during the

2110

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

labeling period, whereas both G1 (Fig. 1A, box 2) and G2 (Fig. 1A, box3) phases will show BrdUrd-negative staining. Fig. 1B shows a repre-sentative histogram 4 h after the BrdUrd-labeling. Two additional groupsof cells were identified: box 4 (Fig. 1B), BrdUrd-positive cells with G1

DNA content that have completed cell division; and box 5 (Fig. 1B),BrdUrd-negative G1 cells entering S phase. Using this assay, we deter-mined three parameters (23, 24) by measuring the progression of cellsthrough different phases of the cell cycle (see Fig. 1B):

Entry into BrdUrd-negative S phase

�Mean fluorescence of cells in box 5

Mean fluorescence of cells in box 2(A)

Transit through S phase4

�

Mean fluorescence of cells in box 1

�mean fluorescence of cells in box 2

Mean fluorescence of cells in box 3

�mean fluorescence of cells in box 2

(B)

Transit through S phase, G2, and M

�1⁄2 � number of cells in box 4

Number of cells in box 1�1⁄2(number of cells in box 4)(C)

The three parameters were calculated relative to the 0-h untreatedcontrol.

Fig. 2 represents HPLC measurements of intracellular low-molec-ular-weight thiol levels in cells cultured in the absence and presenceof NAC. Quantitation of individual thiols (Fig. 2B) was determined byresolving standard solutions of NAC, GSH, �GGC, and cysteinederivatized with NPM and plotting standard curves. Using thismethod, the amounts of intracellular GSH and cysteine in asynchro-nously growing exponential cultures of MEFs were determined to be3.3 � 0.9 (average � SD) nmol/mg protein and 0.67 � 0.16 nmol/mgprotein, respectively (Fig. 2C). Intracellular soluble low-molecular-weight thiol levels increased �6-fold after a 12-h treatment of expo-nentially growing MEF cultures with 20 mM NAC (Fig. 2D). NAC-

induced increases in intracellular soluble low-molecular-weight thiollevels were calculated to be GSH (11.6 � 2.1 nmol/mg protein),cysteine (2.6 � 1.1 nmol/mg protein) and NAC (9.4 � 0.3 nmol/mgprotein). These results show that treatment of MEFs with NAC shiftedthe intracellular redox state toward a more reducing environment.

In addition, intracellular redox state was also determined by flowcytometric measurements of changes in the fluorescence of a prooxidant-sensitive chemical DFH-DA. Asynchronously growing cultures of MEFswere grown in medium containing 20 mM NAC and were harvested at 4-and 12-h intervals for DFH-DA staining and flow cytometric analysis(Fig. 3). The relative fold change in MFI, calculated from the geometric

4 Transit through S phase or Relative Movement [Begg et al. (24)].

Fig. 2. Treatment of MEFs with NAC increases intracellular pools of reduced thiols:HPLC analysis of intracellular thiol levels in exponentially growing cell cultures of MEFs(C) and 12 h after the addition of 20 mM NAC (D). Total cellular protein extracts werederivatized with NPM and resolved using a 15-cm C18 Reliasil column. Quantitation ofindividual thiols was determined by injecting standards containing known amounts ofNAC, GSH, �GGC, and cysteine (Cyst, 50-pmol standards shown in B) and constructingstandard curves of peak area versus concentration of analytes. Peaks attributable to thederivatizing reagent alone in the absence of the added thiols are shown in A.

Fig. 3. A decrease in prooxidant-sensitive DFH fluorescence after the exposure ofMEFs to NAC. Exponential MEF cultures were treated with or without 20 mM NAC (0,4, and 12 h) and stained with a prooxidant-sensitive DFH-DA (A) and insensitive C369 (B)probe. Stained cells were analyzed by flow cytometry, and MFI was calculated usingCellQuest software. Peak 1 in A, autofluorescence in cells without staining, peaks 2 and3, fluorescence of cells cultured in absence (�NAC) and presence (� NAC) of NAC for12 h. Bottom panels, means and SDs of relative MFI calculated from three separateexperiments. Relative fold-change in MFI in each experiment was calculated relative to0 h untreated control.

Fig. 1. Flow cytometric BrdUrd-pulse-chase assay for measurements of progressionthrough the cell cycle: contour plots of DNA content versus log BrdUrd-FITC histogramsof asynchronously growing MEFs and their subdivision into cell cycle compartments. A,exponentially growing, asynchronous population of MEFs analyzed immediately after a30-min pulse labeling with BrdUrd. B, BrdUrd pulse-labeled cells were followed inculture for 4 h in BrdUrd-free regular growth medium before analysis. Five cell cyclecompartments were identified as follows: Box 1, BrdUrd-positive undivided S-phase cells;Box 2, BrdUrd-negative G1 cells; Box 3, BrdUrd-negative G2 cells; Box 4, BrdUrd-positive G1 cells that have completed cell division; Box 5, BrdUrd-negative S-phase cells.

2111

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

mean fluorescence of cells grown in the presence of NAC, showed an�40% decrease after 12 h of NAC treatment compared with cells grownin the absence of NAC (Fig. 3A; P � 0.05). The decrease in DFHfluorescence in NAC-treated cells was specific to NAC-induced changesin intracellular redox state because staining with the oxidation-insensitiveC369 compound did not show any significant difference in MFI betweenthe cells grown in the absence and the cells grown in the presence ofNAC (Fig. 3B). Therefore, the NAC-induced decreases in the oxidationof the DFH was representative of a shift in intracellular redox statetoward a more reducing environment and was not caused by changes inuptake, ester cleavage, or efflux of the compound.

To determine whether changing the intracellular redox state toward amore reducing environment affected progression through G1, S phase,and G2�M, a time course study was performed using the BrdUrdpulse-chase assay. Asynchronously growing exponential cell cultures ofMEFs were pulse-labeled with BrdUrd and were chased in regulargrowth medium for 12 h in the absence and presence of NAC. Cell cyclephase distributions were analyzed using flow cytometric analysis. Resultspresented in Fig. 4 show that increasing thiol pools during the first 9 h ofNAC exposure (5, 10, and 20 mM) accelerated entry into S phase by�1–1.5 h (Fig. 4A; P � 0.05). However, progression through S phase,measured as relative movement (Fig. 4B), was not affected by theincrease in the thiol pools. Similarly, progression through S phase andG2�M, measured as a fraction of BrdUrd-positive cells that have divided(Fig. 4C), also showed no difference.

In contrast, continued treatment with NAC for 12 h resulted in asignificant inhibition of cells entering S phase during both the first andsecond generations (Fig. 4, D–G). In the untreated control group, Sphase cells at 12 h after BrdUrd labeling in generations 1 and 2 were

20 � 0.004% and 28 � 0.015%, respectively. However, the S-phasefraction in NAC-treated cells decreased �2–3-fold to 7 � 0.02% ingeneration 1, and 11 � 0.03% in generation 2 (Fig. 4, F and G). Theinhibition of entry into S phase was also apparent at 24 h of NACtreatment (data not shown). These results show that increasing theintracellular reduced-thiol pools with NAC did not affect progressionfrom late G1 to S phase to G2�M in generation 1. However, progres-sion from early G1 in generations 1 and 2 seemed to be inhibited.These results suggest that the NAC-sensitive redox signaling event(s)initiating progression from G1 to S phase resides early in G1.

To assess whether NAC-induced inhibition in progression from G1

to S phase is associated with cell death in MEFs, a clonogenic cellsurvival assay was performed. Results from the experiment showedthat the concentrations of the NAC used in this study did not affectcell survival (data not shown), suggesting that the NAC-induced cellcycle arrest was not a result of cell death.

To determine whether the NAC-induced G1 arrest could be attrib-utable to its thiol antioxidant/radical scavenger properties or its abilityto increase cellular glutathione content, glutathione synthesis wasinhibited with 1 mM BSO 1 h before and during the 12-h NACtreatment. Cell cycle phase distribution was analyzed by PI stainingand flow cytometry. Results from this experiment showed that morethan 75% of the cells reside in G1 in the NAC-treated cells as well ascells treated with both BSO and NAC (data not shown). These resultsindicate that NAC-induced G1 arrest is independent of the effect ofNAC on cellular glutathione levels. These results suggest that theantioxidant properties of NAC rather than its effect on glutathionelevels were the mechanism regulating NAC-induced G1 arrest. Such

Fig. 4. Treatment with NAC inhibits the progression of cells from G1 to S phase. Exponentially growing asynchronous cultures of MEFs were pulse-labeled with BrdUrd and chased inmedium containing 0, 5, 10, and 20 mM NAC. Cells were harvested at the indicated times and cell cycle phase transit was analyzed by flow cytometric assay. A, transit through G1 (Rel. X S�/G1)was measured by monitoring the mean DNA content of cells in the BrdUrd-negative S phase (Box 5 in Fig. 1B) relative to the mean DNA content of cells in the BrdUrd-negative G1 phase(Box 2 in Fig. 1B). B, relative movement (RM) of BrdUrd-positive undivided cells was used as a measure of transit through S phase. C, fraction (Fr.G1

�) of BrdUrd-positive G1 cells (Box4, Fig. 1B) was calculated to measure transit through S phase and G2 and M phases. D and E, contour plots of exponentially growing MEFs pulse-labeled with BrdUrd and harvested 12 h afterculturing in absence (D) and presence (E) of 20 mM NAC. F and G, S-phase distributions in generations 1 and 2 after 12 h of NAC exposure compared with control. The difference in G1-transitkinetics in cells cultured in absence and presence of NAC (A) was statistically significant as determined by ANOVA (P � 0.05).

2112

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

an antioxidant property of NAC could be directed in the inhibition ofredox reactions occurring on cellular protein thiols.

Redox-sensitive Progression from G1 to S Phase in Exponen-tially Growing MEFs Is Independent of Cellular p53 Status. Todetermine whether the redox-sensitive progression from G1 to S phasedepends on p53 status, asynchronously growing exponential culturesof MEFs [clone 12(1), wild type p53; and clone 10(1) containingmutant p53] were treated with 1–20 mM NAC for 12 h and werepulse-labeled with BrdUrd before harvest. Flow cytometric analysis ofcell cycle phase distribution showed that asynchronous cells in theabsence of NAC had a normal cell cycle distribution with34 � 0.014% and 39 � 0.007% S-phase cells in the wild-type andmutant p53 cells, respectively (Fig. 5A, upper panels, and B, open bargraphs). However, both cell lines showed less than 5% S phase cellsafter 12 h of 20-mM NAC treatment, which indicated a G1 delay inboth cell lines (Fig. 5A, lower panels, and B). The fraction of G1 inboth cell lines increased with 5 mM of NAC and became maximal(�0.7–0.8%) with 20 mM of NAC (Fig. 5B). The fraction of G1 incontrol 12(1) cells was 0.5 � 0.014 and increased to 0.7 � 0.02 and0.79 � 0.007 after exposure to 5 and 20 mM NAC, respectively.Similarly, the fraction of G1 in control 10(1) cells were 0.45 � 0.005and increased to 0.6 � 0.002 and 0.8 � 0.009 after incubation with 5and 20 mM NAC, respectively. These results indicate that the NAC-sensitive progression from G1 to S phase in MEF cell cycle isindependent of cellular p53 status.

Prooxidant Production after Release from NAC-induced G1

Block Is Associated with MEFs Entry into S Phase. To furtherinvestigate the presence of a NAC-sensitive redox event(s) before Sphase entry, exponentially growing asynchronous cultures of MEFswere incubated with 20 mM NAC for 12 h and were released from theNAC-induced G1 block by washing the monolayer with regular cellculture medium without NAC. Cells were continued in culture in theabsence of NAC and were harvested at regular intervals for analysisof glutathione and glutathione disulfide, NAC levels, cysteine levels

and DFH fluorescence measurements. Removal of NAC resulted in adecrease in intracellular glutathione levels from 19.7 � 4 nmol/mgprotein at the time of release to 16.2 � 4 nmol/mg protein at 4 h and8.9 � 3.1 nmol/mg protein at 8 h after release. The intracellularcysteine levels decreased from 1.4 � 0.5 nmol/mg protein at the timeof release to approximately control levels of 0.4 � 0.2 nmol/mgprotein at 8 h after release from the NAC-induced G1 block. Thepercentage of GSSG was 3.4 � 0.5% at the time of release from theNAC treatment, increased to 11.4 � 0.5% at 4 h (P � 0.05) anddecreased to 10.3 � 0.8% at 8 h (P � 0.05) after release from theNAC-induced G1-arrest (Fig. 6B). The transient increase in percent-age GSSG was also associated with an increase in DFH fluorescence.The MFI at the end of the NAC treatment decreased �55 � 0.07%compared with control (Fig. 6A, 0 h, P � 0.05) and subsequentlyincreased to 78 � 0.04% of control at 4 h after release from theNAC-induced G1 block (Fig. 6A, 4 h). These results indicate atransient increase in prooxidant production between 2 and 8 h aftercells were released from the NAC-induced G1 block. After the tran-sient increase in prooxidant production, �49 � 0.06% of the cellsentered S phase at 12 h after removal of NAC (Fig. 6, B and C). Theseresults support the earlier finding of redox-regulated event(s) duringthe entry of G1 cells into S phase (Figs. 4 and 5) and suggest thepresence of a prooxidant-stimulated event during progression from G1

to S phase in MEF cultures after release from NAC-induced G1 arrest.Redox-regulated Event(s) during G1 Correlates with Changes

in G1-Regulatory Protein Levels. To determine whether the redox-regulated event(s) during G1 impacts on the abundance of G1-regulatoryproteins, asynchronously growing exponential cultures of MEFs werecultured in medium containing 20 mM NAC and were harvested atvarious intervals for immunoblotting. In asynchronously growing control

Fig. 5. Redox-signaling during progression from G1 to S phase is independent ofcellular p53 status. A, flow cytometric assay of cell cycle phase distribution in asynchro-nously growing MEFs with wild-type [clone 12(1), wtp53, left panels] and mutant p53[clone 10(1), mp53, right panels)] phenotypes cultured in absence (top panels) andpresence (bottom panels) of 20 mM NAC. Cells were pulse-labeled with BrdUrd at the endof a 12-h NAC treatment and were harvested immediately after labeling for flowcytometric analysis. B, distributions of cell cycle phases in the two cell lines aftertreatment with the indicated doses (0–20 mM) of NAC. The fraction of G1 in control and20-mM-NAC-treated cells were as follows: in clone 12(1) wtp53 MEFs, 0.5 � 0.014(mean � SD) and 0.79 � 0.007, respectively; in clone 10(1) mp53 MEFs, 0.45 � 0.005and 0.8 � 0.009, respectively.

Fig. 6. Prooxidant production after release from NAC-induced G1 arrest is associatedwith entry of MEFs into S phase. Exponential MEF cultures were treated with 20 mM

NAC for 12 h and washed with regular growth medium without NAC. Cells werecontinued in culture in medium without NAC and were harvested at regular intervals forDFH staining (A), thiol assays (B), and measurements of S-phase distribution (C). A,relative fold change in MFI of DHF-stained cells at the time of release (0 h) and 4 h afterthe removal of NAC was calculated relative to the untreated control. B, percentage GSSGand fraction of S phase measured in cells at various intervals after the removal of NAC.The changes in percentage of GSSG was statistically significant as determined bytwo-tailed Student’s t test (P � 0.05). C, representative PI versus log BrdUrd-FITChistograms of cells at the end of the NAC treatment (0 h, left histogram), and 12 h afterremoval of NAC (right histogram); arrowhead, position of S-phase cells.

2113

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

cells, cyclin D1 protein levels remained high (0 h, Lane 1, Fig. 7A).However, cells grown in the presence of NAC showed approximately a70% decrease in cyclin D1 protein levels between 6 and 12 h of NACtreatment and remained low at 24 h (Lanes 2–4, Fig. 7A). The decreasein cyclin D1 protein levels in NAC-treated cells at 12 and 24 h wasaccompanied with an increase in the CDK inhibitor p27 protein levels.Similarly, cyclin D1-dependent Rb phosphorylation, a critical regulatorystep in the release of the E2F transcription factor and initiation of S phase,was also differentially regulated in NAC-treated cells. In exponentiallygrowing control cells, hyper- and hypophosphorylated Rb migrated as adiffuse band (0 h, Lane 1, Fig. 7A). In contrast, a faster migratingpolypeptide band representative of the hypophosphorylated form of theRb protein was observed both at 12 and 24 h in cells grown in presenceof NAC (Lanes 3 and 4, Fig. 7A). The alterations in G1-regulatoryproteins in cells cultured in the presence of NAC correlated with theinhibition of the prooxidant event in G1 as well as the inhibition in

progression from G1 to S phase (Figs. 3 and 4). These results furthersupport the idea that early G1-specific redox-regulated event(s) that areinhibited by NAC might stimulate progression from G1 to S phase inMEF cell cycle.

In a separate experiment, NAC-treated cells were washed withregular cell culture medium and continued in culture in the absence ofNAC. Cells were harvested at various intervals after removal of NAC,and G1-regulatory proteins levels were analyzed by immunoblotting.Results presented in Fig. 7B showed that decreases in the thiol poolsand transient increases in the oxidation of DFH and percentage ofGSSG were followed by an increase in cyclin D1 protein levelsbetween 0 and 10 h after release from the NAC treatment (Lanes 5–7,Fig. 7B). In addition, the protein levels of the CDK inhibitor p27decreased between 4 and 10 h after release from the NAC-induced G1

block (Lanes 5–7). Consistent with these changes, a slower migratingpolypeptide band representing the hyperphosphorylated form of theRb protein was observed between 4 and 24 h after release from NACtreatment (Lanes 7–8). The reversal of NAC-induced changes inG1-regulatory protein levels accompanies the cells’ subsequent entryinto S phase (Fig. 6, B and C). These results clearly suggest that theredox-regulated signaling event(s) during G1 precedes entry into Sphase and could impact on the regulation of G1 cell cycle proteins.

The NAC-induced Redox-sensitive Site Responsible for G1 Ar-rest Resides Early in G1. The NAC-induced redox-sensitive site inprogression from G1 to S phase was further mapped in synchronizedMEFs by monitoring the G0 to G1 to S-phase transition after mito-genic stimulation of serum-starved cells. Exponentially growingMEFs were recruited into G0 by serum starvation for 24 h and thenserum stimulated to reenter the cell cycle. Serum-stimulated cellswere cultured in the presence or absence of 20 mM NAC. Resultspresented in Fig. 8 show that �15% of cells in the control group weresynthesizing DNA by 10 h after serum stimulation (Fig. 8A, Control).In contrast, no cells had entered S phase in the NAC-treated group at10 h after serum stimulation (Fig. 8B, (�) panel). In the control group,the fraction of cells entering S phase increased to �38 � 0.02% and84 � 0.03% at 12 and 14 h, respectively (Fig. 8, A and C). In contrast,the majority of the cells (�90%) in the NAC-treated group remainedin G1 during the same time interval (Fig. 8, B and C). These results

Fig. 7. Redox-regulated signaling during G1 impacts on G1-regulatory protein levels.Immunoblot analysis of G1-cell-cycle-regulatory protein levels during the NAC treatment(A) and after release from the NAC-induced G1 arrest (B). Exponentially growing MEFswere cultured in the absence (Lane 1) and presence (Lanes 2–4) of 20 mM NAC, and totalcellular proteins were isolated at the indicated times for immunoblotting. Lanes 5–8,immunoblot analysis of G1-regulatory protein levels at the time of release from theNAC-induced G1 block (Lane 5), and at 4 (Lane 6), 10 (Lane 7), and 24 (Lane 8) h afterthe removal of NAC. Rb and pRb, hypo- and hyperphosphorylated forms, respectively, ofthe Rb protein. Actin protein levels were used for loading control.

Fig. 8. NAC-induced redox-sensitive site responsible forG1 arrest resides early in G1. Exponentially growing MEFswere synchronized by culturing the cells in medium contain-ing 0.1% serum and were serum-stimulated to reenter the cellcycle in the absence (A) or presence (B) of NAC. Cells werepulse-labeled with BrdUrd at regular intervals, and cell cyclephase distribution was analyzed by flow cytometric assay. A,representative contour plots of dual parameter PI versus logBrdUrd-FITC histograms of synchronized cell culturesgrown in absence (Control) and presence (B, � NAC) of 20mM NAC; left panel (A), G1, S phase, and G2 populations arelabeled. C. the fraction of cells in S phase at various hoursduring the NAC treatment was calculated from the dualparameter histograms. Each data point, the mean of threeindependent experiments; error bars, �SD.

2114

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

support the previous studies using asynchronous cell populations (Fig.4) and map the NAC-sensitive redox-signaling site to early in G1

phase.The presence of a redox-signaling event early in G1 was further

evaluated by DFH staining. Cells were synchronized by serum starvationand serum stimulated to reenter the cell cycle. Serum-stimulated cells atvarious time intervals were stained with DFH-DA for 15 min and assayedby flow cytometry to determine whether changes in prooxidant produc-tion were occurring before entry into S phase. Results presented in Fig. 9,A and D, show a transient 1.9 � 0.12 (mean and SD)-fold increase inDFH fluorescence as measured by MFI of 20,000 cells at 4 h after serumstimulation (Fig. 9A, peak 3) compared with the MFI of cells at the timeof serum stimulation (Fig. 9A, peak 2). Interestingly, the transient in-crease in DFH fluorescence decreased to 1.4 � 0.12-fold at 6 h afterserum stimulation (Fig. 9A, peak 4). DFH fluorescence in serum-stimu-lated cells was also analyzed by confocal microscopy (Fig. 9C). Thefrequency of DFH-stained cells was low in cultures collected after 5 minof serum stimulation and stained for 15 min (Fig. 9C, left panel). Incontrast, more than 95% of the cells harvested at 4 h after serumstimulation, and stained for 15 min, showed DFH-positive cells (Fig. 9C,right panel). Four to six h after this apparent change in DFH oxidation,�20–40% cells entered S phase 8–10 h after serum stimulation (Fig.9D). In contrast, serum-stimulated cells cultured in the presence of NACdid not show any significant changes in DFH fluorescence at 4–6 h afterserum stimulation and demonstrated no significant increase in the fractionof S-phase cells at 8–10 h poststimulation (Fig. 9, B and D). These resultsshow that the transient increase in prooxidant production (as determinedby MFI) after serum stimulation is an early event in G0 to G1 transitionthat clearly precedes the entry of the cells into S phase. Furthermore,treatment with a thiol antioxidant, NAC, which suppressed the apparentincrease in prooxidant production, inhibited entry into S phase afterserum stimulation. These results suggest a causal relationship betweenthe prooxidant production during G1 and progression of the cells into Sphase.

DISCUSSION

Alterations in intracellular oxidation/reduction (redox) reactions arebelieved to play a regulatory role in various cellular and molecular

events including modulation of protein activities, signal transduction,cell death, and cell proliferation (9, 25–33). Depletion of low-molec-ular-weight thiols such as L-cysteine and glutathione has been shownto suppress thymidine uptake in natural killer cells (30). Lipid perox-ide-induced imbalance of GSH/GSSG in human colon cancer CaCo-2cells is known to arrest cell proliferation in G0-G1 (31). Intracellularglutathione levels have also been reported to vary during the mousefibroblast cell cycle being maximal in mitotic cells (33). Althoughthese previous studies provide some evidence for a mechanistic linkbetween intracellular redox reactions and cell proliferation, the directrole of intracellular redox state during progression through specificphases of the cell cycle as well as the specific cell-cycle-regulatoryprocesses that may be redox regulated are not completely understood.

In the present study, a membrane-permeate aminothiol, NAC, wasused to modulate intracellular redox state and to determine whether sucha manipulation affects progression through specific phases of the cellcycle. NAC is a well-known thiol antioxidant as well as a precursor tointracellular glutathione synthesis via its ability to augment cysteinepools. Glutathione is the major low-molecular-weight intracellular thiolthat acts in a redox-buffering capacity and in general is used as anindicator of changes in intracellular redox state. Treatment of exponentialMEFs with NAC resulted in an increase in intracellular glutathione andcysteine levels, indicating a shift in intracellular redox state toward amore reducing environment. NAC-induced changes in intracellular redoxstate toward a more reducing environment were also supported by adecrease in relative MFI of a prooxidant-sensitive compound, DFH-DA.The shift toward a more reducing environment did not affect the transit ofcells through S phase, G2, and M. In fact, cells from mid- to late G1

appeared to accelerate entry into S phase in the first generation (Fig. 4A).However, cells after the completion of division were unable to progressinto S phase and showed an early G1 arrest in the second generation (Fig.4, D–G). These results clearly indicate that an oxidation event early in G1

may be a critical regulatory step in the progression of cells from G1 to Sphase.

The mapping of the redox-signaling event(s) to early in G1 wasfurther verified in synchronized G0 cells that were serum stimulated toreenter the cell cycle. Detection of a transient increase in DFHfluorescence in serum-stimulated cells at 4 h poststimulation further

Fig. 9. Transient increase in prooxidant production duringG0-G1 precedes entry into S phase. MEFs, synchronized byserum starvation, were serum stimulated to reenter the cellcycle and were collected at various intervals for the meas-urement of prooxidant production by DFH staining and S-phase entry by flow cytometry. DFH fluorescence at 0 h(peak 2), 4 h (peak 3), and 6 h (peak 4) after serum stimu-lation in control (A) and in serum-stimulated cells cultured inthe presence of NAC (B). Peak 1, fluorescence from un-stained cells. C, confocal microscopy images of DFH fluo-rescence of serum-stimulated cells at 5 min and at 4 h afterthe serum stimulation and at 15 min of DFH staining. D,relative fold change in MFI of synchronized cells that wereserum stimulated to reenter the cell cycle in the absence (�)and presence (F) of NAC. Relative fold changes in MFI werecalculated relative to control, and statistical significance wasdetermined using ANOVA (P � 0.05). The P for the 4-h timepoint was calculated to be less than 0.05. Cells in replicatedishes were pulse-labeled with BrdUrd, and cell cycle phasedistribution was assayed by flow cytometry. Fraction of Sphase calculated from the dual parameter PI versus logBrdUrd-FITC histograms: f, cells that were serum stimu-lated in the presence of NAC; Œ, serum-stimulated controlcells. Error bars (D), mean � SD.

2115

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

indicated that the redox-sensitive signaling resides early in G1 (Fig.9). Inhibiting this redox signaling with NAC inhibited entry into Sphase. The presence of a redox-signaling event before S phase entrywas also observed in cells released from the NAC-induced G1 arrest.A transient increase in DFH fluorescence and the percentage of GSSGwas observed between 2 and 8 h after the removal of NAC, and,subsequently, more than 50% cells entered S phase at 12 h afterrelease from the NAC-induced G1 arrest (Fig. 6). These results sup-port the hypothesis that a transient increase in intracellular prooxidantlevels early in G1 initiates progression from G1 to S phase and suggestthat such a redox-signaling event is cyclic in nature in the MEF cellcycle (Fig. 10). Although a transient increase in prooxidant levelswould initiate progression from G1 to S phase, antioxidants wouldinhibit such a progression. Although the identity of the prooxidant(s)was not investigated in this study, it is possible that ROS (i.e.,superoxide and hydrogen peroxide), generated during mitochondrialoxidative phosphorylation, and ATP production could act as a signal-ing mechanism early in G1 in preparation for entry of the cells into Sphase. Under normal conditions of cell growth, cellular antioxidantdefense mechanisms, including enzymatic scavenging systems (super-oxide dismutases, catalase, glutathione peroxidases, thioredoxin per-oxidases) and thiol-reducing buffers (small protein- and nonproteinthiols), would neutralize the transient increase in prooxidant(s) levelsproviding a redox balance and continuation of the progression of cellsfrom G1 to S phase. In contrast, failure to maintain such a redoxbalance would halt progression and, depending on the severity of theimbalance, could lead to oxidative stress with deleterious effects. Infact, earlier studies have reported fluctuations in the levels of antiox-idant enzymes in proliferating cells compared with cells in the qui-escent phase (32, 33). Increased levels of superoxide dismutase en-zyme in G1 and variations in intracellular thiol levels during celldivision (32, 33) support the idea of a regulatory role of an intracel-lular redox state during progression through the cell cycle.

In many ways such a redox-sensitive regulation in G1 resembles the“restriction point” (34). Whereas growth factors withdrawal after therestriction point would not affect transit through the remainder of G1, Sphase, G2, and M phases, withdrawal of growth factors before therestriction point halts progression from G1 to S phase. Similarly, manip-ulating the intracellular redox state toward a more reducing environment

in cells in which the prooxidant event had already occurred would not beexpected to affect cells transit through the remainder of G1, S phase, G2,and M phases of the cell cycle (Fig. 10). In contrast, such a manipulationbefore the prooxidant event would inhibit progression into S phase.Interestingly, many of the growth factors (e.g., platelet-derived growthfactor) are known to generate ROS that are believed to be participating incell growth (7). Additional mapping experiments are needed to determinewhether the timing of the redox signaling event and the restriction pointoverlap. Furthermore, the periodicity of the redox signaling during thecell cycle (Fig. 10) also supports the idea of a redox cycle within themammalian cell cycle that could link the activity of crucial metabolicprocesses early in G1 to the activation of cell-cycle-regulatory proteinsnecessary for entry into S phase. Such a redox control in G1 couldfunction as an intracellular redox-status-sensitive cell cycle checkpointpathway. Perhaps loss of such an intracellular redox-status-sensitive cellcycle checkpoint would result in uncontrolled cell growth. In fact a recentstudy reported a preferential sensitivity of thiol antioxidants towardtransformed and tumor-derived human cells (35). Additional study isnecessary to determine whether this differential response could be causedby an alteration in the redox-sensitive cell cycle checkpoint pathway.

The mechanism by which NAC affects progression from G1 to S phaseis unclear. Because simultaneous treatment of MEFs with BSO and NACdid not reverse the NAC-induced G1 arrest both in this study and inhepatic stellate cells (36), these results suggest that the NAC-mediatedincrease in glutathione levels per se does not cause the G1 arrest. Theseresults, however, do not rule out the possible role of other intracellularlow-molecular-weight thiols (i.e., cysteine, thioredoxin, glutaredoxin) inNAC-induced G1 arrest. Indeed cells released from the NAC-induced G1

arrest showed a gradual decrease in cysteine levels, which was accom-panied by a transient increase in percentage of GSSG and MFI beforeentry into S phase. Furthermore, because the DFH fluorescence decreasedafter NAC treatment (Figs. 3 and 6), and DFH has been shown to besensitive to intracellular hydroperoxides, our results indicated that theNAC-induced cell cycle arrest in MEFs was not caused by auto-oxidationof NAC (37) to form greater quantities of hydroperoxides. Therefore, it ispossible that the NAC-induced G1 arrest in MEFs could be attributable tothe antioxidant properties of NAC. Although, such a possibility was ruledout in hepatic stellate cells (36), it was not clear whether transientfluctuations in peroxide level after the release from the NAC-inducedG1-arrest did occur under those experimental systems. Moreover, theaddition of peroxide to cells that are arrested in G1 might complicateinterpretation of results because of peroxide-mediated toxicity. Our re-sults (Figs. 6 and 9) suggest that the redox event early in G1 is transientin nature and is required for entry into S phase.

The antioxidant property of NAC could be directly inhibiting redoxreactions on cellular protein thiols present in the G1-S phase cell-cycle-regulatory proteins (Fig. 10). Thus, target proteins modulated by NACmight contain redox-sensitive cysteine residues that could participate in athiol-disulfide intra- or intermolecular reaction. Although critical cysteineresidues in transcription factors (e.g., activator protein and nuclear factor�B) are known to participate in regulating redox-sensitivity of transcrip-tion factor DNA-binding activity, such a thiol-disulfide reaction is yet tobe clearly identified in G1-cell-cycle-regulatory proteins. However, arecent study has reported a decrease in Cdc25C protein levels afterhydrogen peroxide exposure (38). Cdc25 family of dual specific phos-phatase regulates activation of CDK activity by removing inhibitoryphosphate groups in CDKs. Whereas Cdc25C acts on the activation of themitotic cyclin/CDK kinase complexes, Cdc25A activates the G1-cyclin/CDK complexes. The authors in this study have shown that doublemutations of cysteines at positions 330 and 377 of Cdc25C phosphataseresulted in resistance of the redox modulation as well as inhibited itsdegradation (38). In the present study, we did not observe any significantchanges in Cdc25A protein levels during the NAC-induced G1 arrest

Fig. 10. A schematic illustration of a redox cycle within the mammalian cell cycle.Prooxidants would initiate progression from G1 to S phase, whereas antioxidants wouldinhibit these processes. Once the redox event is initiated, the cells would complete thepresent cell cycle and repeat the process in G1 of the next cell cycle. Target proteins(cyclin D1, p27, Rb, and so forth), modulated by NAC, may react via the cysteine residuesin these proteins that could participate in a thiol-disulfide intra- or intermolecular reaction.CKI, CDK inhibitor; E2F, transcription factor; P, phosphate.

2116

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

(data not shown). This difference in results could be because, althoughhydrogen peroxide exposure caused oxidative stress in the previousreport, subtle changes in intracellular redox state after mitogenic stimu-lation (our study) may not result in oxidative stress. In fact, the foldinduction in MFI of oxidation-sensitive probe is much lower (2–4-fold;Figs. 6 and 9) in nonoxidative stress conditions compared with oxidativestress conditions (7–9-fold increase; data not shown). Our results do notrule out the possibility of intra- and/or intermolecular disulfide reactionsin the regulation of Cdc25A phosphatase activity during progression fromG1 to S phase.

NAC-induced inhibition in the redox signaling resulted in de-creased cyclin D1 protein levels, increased p27 protein levels, andhypophosphorylation of Rb. In contrast, reestablishment of the redox-signaling event by release from NAC led to decreases in p27 proteinlevels, increases in cyclin D1 protein levels, and an increase in Rbphosphorylation with subsequent entry into S phase. Whereas addi-tional studies are needed to understand the precise molecular mech-anisms associated with redox signaling early in G1, it is possible thatone such mechanism could include alterations in the redox status ofredox-sensitive thiol-disulfide reactions at critical cysteine residues incyclin D1, p27, and Rb proteins. In fact the B domain of the Rbprotein is a site critical for binding to the LXCXE motif found in theD-type cyclin (39). The Rb protein has 15 cysteine residues, and threeof these cysteines are located within the B domain (GeneBank acces-sion no. M33647 J02994). It is not clear whether cysteine in theLXCXE motif of cyclin D1 forms an intermolecular disulfide linkagewith the cysteines in the Rb B-domain. There are also two cysteineresidues within the cyclin A binding domain of p27 (40). Althoughany possible role of these cysteines in redox regulation of G1 to Sphase is speculative at present, it is reasonable to suggest that thiol-redox reactions could act as “sulfhydryl switches” that reversiblymodulate cell-cycle-regulatory protein activity.

In summary, results from this study indicate the presence of aredox-signaling event early in G1, which is required for entry into Sphase. Although shifting of the intracellular redox state toward a morereducing environment selectively inhibited the entry of early G1 cellsinto S phase, a shift toward an oxidizing environment seemed tostimulate entry into S phase. The periodicity of such a redox cyclewithin the mammalian cell cycle may provide a mechanistic linkbetween metabolic processes early in G1 and the activation of G1-regulatory proteins in preparation for the entry of cells into S phase.

ACKNOWLEDGMENTS

We thank Justin Fishbaugh for assisting with the flow cytometric assays.

REFERENCES

1. Shibanuma, M., Kuroki, T., and Nose, K. Induction of DNA replication and expres-sion of proto-oncogenes c-myc and c-fos in quiescent Balb/3T3 cells by xanthine-xanthine oxidase. Oncogene, 3: 14–21, 1988.

2. DeHaan, J. B., Cristiano, F., Lanello, R., Bladier, C., Kelner, M. J., and Kola, I.Elevation in the ratio of Cu/Zn superoxide dismutase to glutathione peroxidaseactivity induces features of cellular senescence and this effect is mediated by hydro-gen peroxide. Hum. Mol. Genet., 5: 283–292, 1996.

3. Allen, R. G., and Balin, A. K. Oxidative influence in development and differentiation:an overview of a free radical theory of development. Free Radic. Biol. Med., 6:631–661, 1989.

4. Hockenbery, D. M., Oltvai, Z. N., Yin, Y. M., Milliman, C. L., and Korsmeyer, S. J.Bcl-2 function in an antioxidant pathway to prevent apoptosis. Cell, 75: 241–251,1993.

5. Finkel, T. Oxygen radicals and signaling. Curr. Opin. Cell Biol., 10: 248–253, 1998.6. Shackelford, R. E., Kaufmann, W. K., and Paules, R. S. Oxidative stress and cell

cycle checkpoint. Free Radic. Biol. Med., 28: 1387–1404, 2000.7. Sundaresan, M., Yu, Z. X., Ferrans, V. J., Irani, K., and Finkel, T. Requirement for

generation of H2O2 for platelet-derived growth factor signal transduction. Science(Wash. DC), 270: 296–299, 1995.

8. Bae, Y. S., Kang, S. W., Seo, M. S., Baines, I. C., Tekle, E., Chock, P. B., and Rhee,S. G. Epidermal growth factor (EGF)-induced generation of hydrogen peroxide.J. Biol. Chem., 272: 217–221, 1997.

9. Burdon, R. H., and Rice-Evans, C. Free radicals and the regulation of mammalian cellproliferation. Free Radical Res. Commun., 6: 345–358, 1989.

10. Irani, K., Xia, Y., Zweier, J. L., Scollott, S. J., Der, C. J., Fearon, E. R., Sundaresan,M., Finkel, T., and Goldschmidt-Clermont, P. J. Mitogenic signaling mediated byoxidants in Ras-transformed fibroblasts. Science (Wash. DC), 275: 1649–1652, 1997.

11. Muller, R., Mumberg, D., and Lucibello, F. C. Signals and genes in the control ofcell-cycle progression. Biochim. Biophys. Acta, 1155: 151–179, 1993.

12. Grana, X., and Reddy, E. P. Cell cycle control in mammalian cells: role of cyclins,cyclin dependent kinases (CDKs), growth suppressor genes, and cyclin dependentkinase inhibitors (CKIs). Oncogene, 11: 211–219, 1995.

13. Hartwell, L. H., and Weinert, T. A. Checkpoints: controls that ensure the order of cellcycle events. Science (Wash. DC), 246: 629–634, 1989.

14. Sherr, C. J. Mammalian G1 cyclins. Cell, 73: 1059–1065, 1993.15. Musgrove, E. A., Lee, C. S. L., Buckley, M. F., and Sutherland, R. L. Cyclin D1

induction in breast cancer cell lines shortens G1, and is sufficient for cells arrested inG1 to complete the cell cycle. Proc. Natl. Acad. Sci. USA, 91: 8022–8026, 1994.

16. Ohtsubo, M., and Roberts, J. M. Cyclin-dependent regulation of G1 in mammaliancells. Science (Wash. DC), 259: 1908–1912, 1993.

17. Ridnour, L. A., Winters, R. A., Ercal, N., and Spitz, D. R. Measurement of glutathi-one, glutathione disulfide, and other thiols in mammalian cell and tissue homogenatesusing high-performance liquid chromatography separation of N-(1-Pyrenyl) maleim-ide derivatives. Methods Enzymol., 299: 258–267, 1999.

18. Spitz, D. R., Sim, J. E., Ridnour, L. A., Galoforo, S. S., and Lee, Y. J. Glucosedeprivation-induced oxidative stress in human tumor cells. Ann. N. Y. Acad. Sci.,899: 349–362, 2000.

19. Anderson, M. E. Tissue Glutathione. In: R. A. Greenwald (ed.), Handbook ofMethods for Oxygen Radical Research, pp. 317–323. Boca Raton, FL; CRC Press,1985.

20. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. Protein measurementwith the Folin phenol reagent. J. Biol. Chem., 193: 265–275, 1951.

21. Goswami, P. C., He, W., Higashikubo, R., and Roti Roti, J. L. Accelerated G1 transitfollowing transient inhibition of DNA replication is dependent upon two processes.Exp. Cell Res., 214: 198–208, 1994.

22. Goswami, P. C., Higashikubo, R., and Spitz, D. R. Redox control of cell cycle-coupled Topo II gene expression. Methods Enzymol., 353: 448–459, 2002.

23. Higashikubo, R., Ragouzis, M., and Roti Roti, J. L. Flow cytometric BrdUrd-pulse-chase study of x-ray induced alterations in cell cycle progression. Cell Prolif., 29:43–57, 1996.

24. Begg, A. C., McNally, N. J., Shrieve, D. C., and Karcher, H. A method to measurethe duration of DNA synthesis and the potential doubling time from a single sample.Cytometry, 6: 620–625, 1985.

25. Suzuki, Y. J., Forman, H. J., and Servanian, A. Oxidants as stimulators of signaltransduction. Free Radic. Biol. Med., 22: 269–285, 1997.

26. Hunter, D. E., Till, B. G., and Greene, J. J. Redox state changes in density-dependentregulation of proliferation. Exp. Cell Res., 232: 435–438, 1997.

27. Smith, J., Ladi, E., Mayer-Proschel, M., and Noble, M. Redox state is a centralmodulator of the balance between self-renewal and differentiation in a dividing glialprecursor cell. Proc. Natl. Acad. Sci. USA, 97: 10032–10037, 2000.

28. Sekharam, M., Trotti, A., Cunnick, J. M., and Wu, J. Suppression of fibroblast cellcycle progression in G1 phase by N-acetylcysteine. Toxicol. Appl. Pharmacol., 149:210–216, 1998.

29. Kusano, C., Takao, S., Noma, H., Yoh, H., Aikou, T., Okumura, H., Akiyama, S.,Kawamura, M., Makino, M., and Baba, M. N-acetyl cysteine inhibits cell cycleprogression in pancreatic carcinoma cells. Hum. Cell., 13: 213–220, 2000.

30. Yamauchi, A., and Bloom, E. T. Control of cell cycle progression in human naturalkiller cells through redox regulation of expression and phosphorylation of retinoblas-toma gene product protein. Blood, 89: 4092–4099, 1997.

31. Gotoh, Y., Noda, T., Iwakiri, R., Fujimoto, K., Rhoads, C. A., and Aw, T. Y. Lipidperoxide-induced redox imbalance differentially mediates CaCo-2 cell proliferationand growth arrest. Cell Prolif., 35: 221–235, 2002.

32. Blakely, E., Chang, P., Lommel, L., Bjornstad, K., Dixon, M., Tobias, C., Kumar, K.,and Blakely, W. F. Cell-cycle radiation response: role of intracellular factors. Adv.Space Res., 9: 177–186, 1989.

33. Li, N., and Oberley, T. D. Modulation of antioxidant enzymes, reactive oxygenspecies, and glutathione levels in manganese superoxide dismutase overexpressingNIH/3T3 fibroblasts during the cell cycle. J. Cell. Physiol., 177: 148–160, 1998.

34. Pardee, A. B. A restriction point for control of normal animal cell proliferation. Proc.Natl. Acad. Sci. USA, 71: 1286–1290, 1974.

35. Havre, P. A., O’Reilly, S., McCormick, J. J., and Brash, D. E. Transformed andtumor-derived. human cells exhibit preferential sensitivity to the thiol antioxidants.N-acetyl cysteine and penicillamine. Cancer Res., 62: 1443–1449, 2002.

36. Kim, K. Y., Rhim, T., Choi, I., and Kim, S. S. N-acetylcysteine induces cell cyclearrest in hepatic stellate cells through its reducing activity. J Biol Chem., 276:40591–40598, 2001.

37. Tuttle, S., Horan, A. M., Koch, C. J., Held. K., Manevich, Y., and Biaglow, J.Radiation-sensitive tyrosine phosphorylation of cellular proteins: sensitive to changesin GSH content induced by pretreatment with N-acetyl-L-cysteine or L-buthionine-S,R-sulfoximine. Int. J. Radiat. Oncol. Biol. Phys., 42: 833–838, 1998.

38. Savitsky, P. A., and Finkel, T. Redox regulation of Cdc25C. J. Biol. Chem., 277:20535–20540, 2002.

39. Taya, Y. RB kinases and RB-binding proteins: new points of view. Trends Biochem.Sci., 22: 14–17, 1997.

40. Russo, A. A., Jeffrey, P. D., Patten, A. K., Massague, J., and Pavletich, N. P. Crystalstructure of the p27Kip1 cyclin-dependent-kinase inhibitor bound to the cyclinA-Cdk2 complex. Nature (Lond.), 382: 325–331, 1996.

2117

INTRACELLULAR REDOX AND CELL CYCLE

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

2003;63:2109-2117. Cancer Res   Sarita G. Menon, Ehab H. Sarsour, Douglas R. Spitz, et al.   Mouse Embryo Fibroblast Cell Cycle

to S Phase Transition in the1Redox Regulation of the G

  Updated version

  http://cancerres.aacrjournals.org/content/63/9/2109

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/63/9/2109.full.html#ref-list-1

This article cites 37 articles, 14 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/63/9/2109.full.html#related-urls

This article has been cited by 29 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  SubscriptionsReprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  .permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on August 17, 2015. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from