The intracellular reduction-oxidation (redox) environment influences cell cycle progression; however, underlying mechanisms are poorly understood. To examine potential mechanisms, the intracellular redox environment was characterized per cell cycle phase in Chinese hamster ovary fibroblasts via flow cytometry by measuring reduced glutathione (GSH), reactive oxygen species (ROS), and DNA content with monochlorobimane, 2′,7′-dichlorohydrofluorescein diacetate (H2DCFDA), and DRAQ5, respectively. GSH content was significantly greater in G2/M compared with G1 phase cells, whereas GSH was intermediate in S phase cells. ROS content was similar among phases. Together, these data demonstrate that G2/M cells are more reduced than G1 cells. Conventional approaches to define regulatory mechanisms are subjective in nature and focus on single proteins/pathways. Proteome databases provide a means to overcome these inherent limitations. Therefore, a novel bioinformatic approach was developed to exhaustively identify putative redox-regulated cell cycle proteins containing redox-sensitive protein motifs. Using the InterPro (http://www.ebi.ac.uk/interpro/) database, we categorized 536 redox-sensitive motifs as: 1) active/functional-site cysteines, 2) electron transport, 3) heme, 4) iron binding, 5) zinc binding, 6) metal binding (non-Fe/Zn), and 7) disulfides. Comparing this list with 1,634 cell cycle-associated proteins from Swiss-Prot and SpTrEMBL (http://us.expasy.org/sprot/) revealed 92 candidate proteins. Three-fourths (69 of 92) of the candidate proteins function in the central cell cycle processes of transcription, nucleotide metabolism, (de)phosphorylation, and (de)ubiquitinylation. The majority of oxidant-sensitive candidate proteins (68.9%) function during G2/M phase. As the G2/M phase is more reduced than the G1 phase, oxidant-sensitive proteins may be temporally regulated by oscillation of the intracellular redox environment. Combined with evidence of intracellular redox compartmentalization, we propose a spatiotemporal mechanism that functionally links an oscillating intracellular redox environment with cell cycle progression.
- redox regulation
the intracellular reduction-oxidation (redox) environment is constituted by reactive oxygen species (ROS) and intracellular antioxidants, primarily the thiol-containing tripeptide glutathione (Glu-Cys-Gly). ROS are highly reactive oxygen derivatives produced as byproducts of aerobic respiration in the electron transport chain of the mitochondrial inner membrane (42). Glutathione reacts with ROS to prevent oxidative damage to intracellular molecules (DNA, proteins, and lipids; 16, 60). Perturbations in the intracellular redox environment are reflected in the relative redox status of the oxidized (GSSG) vs. reduced (GSH) forms of glutathione. Thus cellular redox status is typically assessed by quantifying GSH and GSSG and calculating the half-cell redox potential (Eh, expressed in mV) of the GSSG/2GSH couple with the Nernst equation (4, 40, 45). In this way, intracellular Eh values ranging from −165 (oxidized) to −258 mV (reduced) have been measured in various cell types and cell states (24, 40).
A correlation between intracellular redox status and cell proliferation vs. differentiation has been observed for numerous cell types. Specifically, fibroblasts (20, 35), intestinal epithelial cells (24, 33), cardiomyocytes (38), and osteoclasts (53) are relatively reduced when proliferating and relatively oxidized once differentiated. Intracellular redox status also affects cell cycle progression (20, 31, 35, 48); however, underlying molecular mechanisms remain undefined. The numerous examples of redox modulation of transcription (41, 61), signal transduction (54), biosynthetic pathways (11), and posttranslational modification (37) attest to the potential to which the intracellular redox environment may influence cell cycle progression. Indeed, several redox-sensitive proteins involved in cell cycle progression have been identified, including p53 (9, 15, 56), AP-1 (43, 44), NF-κB (19, 41), PKC (7, 13), and low-molecular-weight protein tyrosine phosphatases (10).
As a first step toward understanding the link between the intracellular redox environment and cell cycle progression, we examined changes in the intracellular redox environment associated with cell cycle progression in untreated, asynchronously growing Chinese hamster ovary (CHO) fibroblasts. Identification of redox-sensitive amino acids or redox-sensitive structures within proteins, such as AP-1 (1, 25) or low-molecular-weight protein tyrosine phosphatases (10), may be used to predict cell cycle proteins that provide a regulatory link between the intracellular redox environment and cell division. Accordingly, a subsequent bioinformatic approach, which takes advantage of extensive proteome and protein motif databases, was implemented to exhaustively identify candidate redox-regulated proteins that may provide a mechanistic link between the intracellular redox environment and control of cell cycle progression.
MATERIALS AND METHODS
CHO fibroblasts were purchased from American Type Culture Collection (Manassas, VA). The cells were cultured in Dulbecco’s modified Eagle’s medium, supplemented with a final glucose concentration of 25 mM, 0.1 mM nonessential amino acids (GIBCO; Invitrogen, Grand Island, NY), 44 mM sodium bicarbonate, 15 mM HEPES, 10% fetal bovine serum (FBS), streptomycin (10,000 U/ml), penicillin (10,000 U/ml), and Fungizone (250 μg/ml) in T-75 cell culture flasks (Costar, Cambridge, MA). The cells were maintained at 37°C in 5% CO2.
Analysis of the intracellular redox environment per cell cycle phase.
The intracellular redox environment was examined per cell cycle phase in live CHO cells using the fluorescent dyes monochlorobimane (mBCl; Molecular Probes, Eugene, OR; 17) for GSH and 2′,7′-dichlorohydrofluorescein diacetate (H2DCFDA; Molecular Probes; 62) for ROS by flow cytometric analysis. Cell cycle phases were distinguished by the addition of the anthraquinone derivative DRAQ5 (Biostatus, Shepshed, UK), which fluoresces in the deep red spectrum and has a high affinity for DNA (49). mBCl is a membrane-permeant probe that fluoresces in the UV spectrum upon reacting with GSH in a reaction catalyzed by the enzyme glutathione-S-transferase (17). Due to the enzymatic catalysis of mBCl-GSH adduct formation, mBCl has greater specificity for GSH compared with other thiol-specific probes such as monobromobimane, which reacts freely with both GSH and intracellular protein thiols (17). Once the membrane-permeant H2DCFDA enters a cell, its acetate moieties are cleaved by intracellular esterases resulting in an impermeable H2DCF form. Subsequent oxidation of H2DCF produces the fluorescent 2′,7′-dichlorofluorescein, which can be detected in the green spectrum.
Cells grown to confluence were subcultured into 6-well plates (Costar) at a density of 4.0 × 105 cells per well and incubated in normal medium overnight at 37°C. Parallel experiments were seeded concurrently from the same culture for ROS and GSH analysis. For GSH measurement, cells were washed with Hanks’ balanced salt solution (HBSS; GIBCO) and treated with 1 ml of warm HBSS containing 80 μM mBCl and 15 μM DRAQ5 for 30 min at 37°C. The cells were subsequently trypsinized with 0.5 ml of 1× trypsin/EDTA in HBSS and resuspended with 0.5 ml HBSS, aliquoted into 1.5-ml microcentrifuge tubes, and stored on ice until further analysis. For ROS measurement, the cells were washed with Ca2+/Mg2+-free phosphate-buffered saline (PBS; GIBCO) and treated with 1 ml of warm PBS containing 10 μM H2DCFDA and 15 μM DRAQ5 for 30 min at 37°C. The cells were then trypsinized, resuspended in 0.5 ml PBS, and stored in 1.5-ml microcentrifuge tubes on ice until further analysis.
Flow cytometric analysis.
Intracellular fluorescence of CHO cells stained for DNA, GSH, and ROS content was measured by flow cytometry with a MoFlo MLS high-speed flow cytometry instrument (Cytomation, Fort Collins, CO) equipped with a Coherent Innova 90C laser for excitation at 488 nm, a Coherent Innova 70C Spectrum for excitation at 647 nm, and a Coherent I90 for excitation at 351 nm. All three lasers operated at 100 mW. DRAQ5, mBCl, and H2DCFDA fluorescence levels were detected using 675/30, 530/40, and 485/25 band-pass filters, respectively. Flow cytometric data were analyzed using Summit V3.1 software (Cytomation). Initially, cells were gated by scatter properties to exclude cellular debris. Peak forward scatter was plotted against pulse width to exclude doublets. Separate bar regions were used to gate separate cell cycle phases on DNA curves to determine the fluorescence intensity of GSH and ROS probes in separate histograms. A minimum of 10,000 gated events were analyzed per sample.
Results from flow cytometric analyses are the mean values of four separate experiments (n = 3–4 per experiment). One-way ANOVA statistical tests were used to compare phase-specific data. When a significant main effect was observed, differences between groups were determined using t-tests. Values of P < 0.05 were considered statistically significant.
Identification of protein redox motifs.
A list of putative redox-regulated protein motifs (redox motif) was assembled using the InterPro (IPR) protein signature database (http://www.ebi.ac.uk/interpro/), an integrated database of protein domains, families, and functional sites (3). The redox motif list was constructed using the Sequence Retrieval System (SRS) web server at EBI (http://srs.ebi.ac.uk/) as follows. Initially, a broad, text-based search was performed for all entries associated with the key words reduction, oxidation, redox, electron transfer, metal binding, heme, cysteine, and disulfide. Next, motifs identified from the key word search were manually examined to establish their redox sensitivity in relation either to their structure, function, or cofactor association (e.g., metal binding) before being considered redox motifs. The coverage of this approach is limited by the extent to which redox is associated with protein motif annotation. For example, p53 has been demonstrated to be redox regulated (9, 15) and also has an associated IPR accession number (IPR008967). However, this motif does not establish a connection with redox regulation and so was not included in the list.
Query results consisting of IPR accession numbers and corresponding descriptions for each entry were downloaded and formatted using Excel (Microsoft, Redmond, WA). The results (redox motifs) were categorized into the following categories based on specific structural motifs or relevant redox chemistry: active/functional-site cysteines (AC), disulfides (SS), electron transport, heme (H), iron binding (FeB), metal binding (MeB, exclusive of zinc and iron), and zinc binding (ZnB; Fig. 1). These categories were chosen based on prior evidence (redox-sensitive amino acid sequence or redox cofactor binding ability) linking each with redox modulation of protein function (2, 18, 25, 55, 58, 63).
Cell cycle-associated proteins.
Proteins involved in cell cycle progression were queried for putative redox motifs. To limit the query, only eukaryotic proteomes exclusive of the kingdom Viridiplantae were included. The Swiss-Prot and SpTrEMBL protein databases were used as the sources for assembling the cell cycle protein list. The Swiss-Prot database is a curated protein database with a low level of redundancy and extensive integration with other databases including IPR (5). The SpTrEMBL database is also integrated with IPR and serves as a database for proteins that are eligible for addition to the Swiss-Prot database upon further validation. Using the EBI SRS Web server, we performed a text-based search of the Swiss-Prot and SpTrEMBL databases for the key words proliferation, mitosis, cell division, cell cycle, replication, or cytokinesis, to select proteins either directly or purportedly involved functionally in cell cycle progression. Composite query results consisting of protein name, species, Swiss-Prot accession number, and IPR accession number(s) were compiled for each protein.
Identification of putative redox-regulated proteins.
To identify putative redox-regulated cell cycle-associated proteins, the cell cycle protein list was searched for entries annotated with IPR accession numbers included in the redox motif list. Positive hits (proteins with redox motifs) were compiled into candidate protein lists. Each candidate protein was checked to eliminate “false positives,” i.e., to exclude proteins having an IPR accession number included in the redox motif list but lacking amino acids conferring redox sensitivity. For example, the tumor susceptibility protein 101 (Swiss-Prot no. Q99816) is assigned IPR000608, which denotes a ubiquitin conjugating enzyme AC motif; however, this protein lacks the catalytic cysteine conferring redox sensitivity. Candidate proteins were categorized according to cell cycle phase as well as function. Literature queries were performed on all candidate proteins to determine whether they had previously been demonstrated to be redox regulated.
Intracellular redox environment per cell cycle phase.
G1, S, and G2/M phase CHO cells were identified by DNA content determined by relative fluorescence intensity after staining with the membrane-permeant anthraquinone derivative DRAQ5, which fluoresces maximally in the red spectrum (Fig. 2A). Dual staining of cells with DRAQ5 and the cell-permeant fluorescent dye mBCl allowed analysis of reduced glutathione levels (GSH) per cell cycle phase (Fig. 2B). GSH-mBCl adduct mean fluorescence intensity (MFI) was significantly greater in cells in G2/M phase compared with those in G1 phase. The GSH-mBCl MFI of cells in S phase did not differ significantly from either G1 or G2/M phase cells but was numerically intermediate (Fig. 2B). Oxidative activity per cell cycle phase was determined by staining CHO cells with the ROS-specific probe H2DCFDA. The relative MFI of H2DCFDA did not differ between G1, S, or G2/M phase cells (Fig. 2C). The ratio of the percent change in MFI of GSH (Fig. 2D) and ROS (Fig. 2E) content per phase normalized to G1 phase values (Fig. 2F) indicates the overall reduction of the intracellular redox environment of cells progressing through the cell cycle.
Identification of protein redox motifs.
The final IPR redox motif list included 536 entries that were distributed into functional categories as follows: 3.2% (17 motifs) active/functional-site cysteines (AC); 2.1% (11 motifs) electron transport; 8.2% (44 motifs) heme (H); 12.5% (67 motifs) iron binding (FeB); 23.3% (125 motifs) metal binding (MeB, exclusive of zinc and iron); 23.9% (128 motifs) disulfide (SS); and 26.9% (144 motifs) zinc binding (ZnB). A total of 38 motifs representing 6 categories (AC, FeB, H, MeB, SS, and ZnB) were found in the candidate redox-regulated proteins (Table 1). The entire motif list is in the Data Supplement (available through the Physiological Genomics web site).1
Cell cycle protein lists.
An initial list of 1,634 cell cycle-related proteins was assembled from combined entries from the Swiss-Prot and SpTrEMBL protein databases. Of these, 1,508 proteins had at least one IPR accession number and were subsequently cross-referenced with the IPR redox motif list to identify putative redox-regulated proteins. The remaining 126 cell cycle proteins lacked an IPR accession number and were thus excluded from the analysis. The complete cell cycle protein list is available in the Data Supplement (mentioned above).
Candidate redox-regulated proteins and functional categories.
By cross-referencing the cell cycle protein and redox motif lists, 92 candidate redox-regulated cell cycle proteins were identified. Major functional categories for candidate proteins include transcription (e.g., transcription factors or coactivators), nucleotide metabolism (e.g., methyltransferases and ribonucleotide reductases), (de)ubiquitinylation (ubiquitin pathway), (de)phosphorylation (kinases/phosphatases), and other (e.g., protein inhibitors, assembly proteins, and spindle proteins, and others). The candidate protein list was distributed by functional category as follows: 15.2% transcription (14 proteins), 15.2% nucleotide metabolism (14 proteins), 17.4% ubiquitinylation (16 proteins), 27.2% kinase/phosphatase (25 proteins), and 25.0% other (23 proteins).
Motif frequency and distribution by cell cycle phase.
The frequency and distribution of redox motifs by cell cycle phase within the candidate list of proteins are shown in Table 2. The most prominent motif categories in the candidate proteins were AC, MeB, and ZnB, whereas only a small number of proteins contained FeB, H, or SS motifs. AC-harboring proteins were primarily functional during G2/M phase, with only a few functional during G1 and S phases. The only FeB-containing protein (ribonucleotide reductase small subunit; 12) functions in S phase. H-containing proteins (Speedy/Spy1 and YPL006W) participate exclusively in G2/M. Similar to AC-containing proteins, MeB-containing proteins were functional primarily during G2/M phase. In contrast to AC- and MeB-containing proteins, SS- and ZnB-containing proteins were distributed equally across all cell cycle phases.
Candidate protein characteristics by cell cycle phase.
G1 phase proteins represented 20.7% (19 proteins) of the redox-regulated proteins identified (Table 3). The distribution of motif types among G1 proteins (i.e., percentage of AC, MeB, etc… per G1 phase) was similar to that of the entire candidate protein population. The distribution of proteins by function was 31.6% transcription (6 proteins), 5.3% nucleotide metabolism (1 protein), 10.5% ubiquitinylation (2 proteins), 26.3% kinase/phosphatase (5 proteins), and 26.3% (5 proteins) other (dehydrogenase, cell adhesion, p53 degradation inhibitor, Cdc28 inhibitor, and unreported). S phase proteins represented only 10.9% (10 proteins) of the total candidate redox-regulated cell cycle proteins (Table 3). Motif category representation within the S phase proteins was similar to the entire candidate protein population. The distribution by function of S phase proteins is 70% nucleotide metabolism (7 proteins), 10% transcription (1 protein), and 20% ubiquitinylation (2 proteins).
Redox motif-harboring proteins functional during G2/M phase represented the largest group, making up 50.0% (46 proteins) of the total list of candidates (Table 4). A greater number of AC- and MeB-containing proteins and fewer ZnB-containing proteins function during G2/M phase compared with the total candidate protein population. The distribution of G2/M phase proteins by function was 4.3% (2 proteins) transcription, 10.9% (5 proteins) nucleotide metabolism, 17.4% (8 proteins) ubiquitinylation, 41.3% (19 proteins) kinase/phosphatase, and 26.1% (12 proteins) other.
Five proteins (5.4% of total proteins) were reported to function in all cell cycle phases (Table 5). These proteins harbored primarily ZnB motifs (3 of the 5 proteins). The proteins were categorized functionally as 40% (2 proteins) transcription, 20% (1 protein) ubiquitinylation, and 40% (2 proteins) other. The cell cycle phase of activity for the remaining 13.0% (12 proteins) of the candidate cell cycle proteins remains unassigned (Table 5). This group was predominated by ZnB-containing proteins. The distribution of proteins by functional category was 25.0% (3 proteins) transcription, 8.5% (1 protein) nucleotide metabolism, 25.0% (3 proteins) ubiquitinylation, 8.5% (1 protein) kinase/phosphatase, and 33.0% (4 proteins) other.
Additional investigation of the literature revealed 22 proteins that have been shown previously to be redox-regulated or are closely related to proteins demonstrated to be redox-regulated (Table 6). These proteins are involved primarily in nucleotide metabolism, ubiquitinylation, and (de)phosphorylation. The proteins active in G1 phase make up 22.72% (5) of the total, whereas 68.18% (15 proteins) are active in G2/M phase, 4.55% (1 protein) are in all phases, and 4.55% (1 protein) are unassigned. Finally, 70% of these proteins contain an AC motif, although proteins containing FeB, MeB, SS, and ZnB motifs were represented. A majority of the proteins are reported to be functional during G2/M phase, with G1 phase having the second largest distribution (Table 7).
The occurrence of redox-sensitive motifs in numerous cell cycle proteins indicates that redox may play a more central role in the regulation of cell division than is currently recognized. This assertion is reinforced by the observed reduction in the intracellular redox environment that parallels cell cycle progression from G1 to G2/M phase. The oscillation in the intracellular redox environment in cells progressing through the cell cycle may represent a fundamental cell cycle mechanism that contributes to the regulation of cell cycle progression through redox-regulated cell cycle proteins.
Although many reports have demonstrated that proliferating cells are more reduced and nonproliferating or differentiated cells are more oxidized (20, 23, 29, 33, 35), few have focused on differences in the intracellular redox environment by cell cycle phase. Our observations of a significant increase in GSH combined with static ROS production indicate an overall reduction of the intracellular environment as cells progress from G1 to G2/M phase. These data are in agreement with a recent Jurkat T lymphocyte study that demonstrated, via immunocytochemistry, the greatest GSH content in G2/M cells (51). Reduction of the intracellular environment as cells progress from G1 to G2/M phase may protect genomic DNA from oxidative damage upon breakdown of the nuclear envelope. The induction of the G1 phase checkpoint protein, p53, by ROS agrees with this hypothesis (9). Moreover, in light of the recent report by Menon et al. (31) that demonstrated the requirement for an oxidative event in early G1 phase for cells to proceed to S phase, p53 induction by ROS presupposes a reduction of the intracellular environment prior to S phase entry. Recent evidence that the initiation of DNA replication in yeast is coordinated with the shift from the oxidative to reductive phase of the respiratory oscillation suggests that this phenomenon may be a common mechanism for eukaryotes (26).
The events driving cell cycle progression (e.g., cyclin expression, DNA and protein synthesis, and protein degradation) are mediated by networks of proteins involved in transcription (e.g., AP-1, NF-κB, and E2F), nucleotide metabolism (e.g., ribonucleotide reductases), phosphorylation (MAPK, cyclin D/cyclin-dependent kinase 4), dephosphorylation (Cdc25A, B, and C), and ubiquitinylation (E1-E3 ubiquitin ligases), among others. If one or more proteins required for a central cell cycle process, such as nucleotide metabolism, were redox sensitive, then a significant shift in the intracellular redox environment could, in this case, attenuate DNA synthesis possibly resulting in cell cycle arrest in S phase. Three-fourths (69 of 92) of the candidate proteins identified are directly involved in transcription, nucleotide metabolism, (de)phosphorylation, or (de)ubiquitinylation, which are all essential processes for cell cycle progression. The remaining 23 proteins have diverse functions such as checkpoint proteins (Dma1, CHFR), spindle proteins (Dim1, DIB1), deacetylases (HDAC6), dehydrogenases (DLDH), protein regulators (Cdk-FAR1, Mdm2, Mdm4), nucleoporins (Nup153), kinetochore proteins (CENP-C), and others. The abundance and diversity of function of the candidate proteins suggests that numerous cell cycle processes may be redox sensitive, implying that changes in the intracellular redox environment could affect cell cycle progression in a variety of ways.
Based on our current observations, oxidant-sensitive proteins are less likely to be oxidized (inactivated) during G2/M (reduced) vs. G1 phase (oxidized). Consistent with this hypothesis, 64%, 100%, and 76.5% of AC, H, and MeB proteins, respectively, whose motifs are susceptible to oxidation (27, 30, 55), function during G2/M phase (Table 2). The broad distribution of the few SS proteins identified limits the conclusions that can be drawn from this motif category. However, the distribution of three of the SS proteins, one in G1 phase (interleukin 1 receptor-like 1), one in S phase (ribonucleotide reductase 1), and one in all phases (thioredoxin II), may indicate that the SS proteins identified are less susceptible to oxidation than the AC, H, and MeB proteins.
In contrast to the AC, H, and MeB proteins, the oxidant-sensitive zinc-binding (ZnB) proteins were active in all phases (Table 2; 58, 61). This apparent contradiction might be explained by the spatial distribution of ZnB proteins within the cell. The majority of ZnB proteins functioning in G1 and S phases bind DNA, necessitating their nuclear localization. Cellular compartmentalization contributes to the regulation of transcription (nucleus), protein folding (endoplasmic reticulum), and energy production (mitochondria) by providing biochemically distinct microenvironments. These organelles have dissimilar redox environments compared with the cytosol (6, 21, 40, 47). Relevant to the current study, the nucleus is more reduced than the cytosol, with GSH concentrations of up to 15 mM compared with 11 mM in the cytosol (6, 40, 50). The nuclear GSH pool is also more resistant to oxidation than the cytosolic pool in rat hepatocytes cells and cultured human melanomas treated with buthionine sulfoximine (6, 22). Moreover, the reducing potential of the nucleus is reinforced by the vicinal dithiol-containing protein thioredoxin-1 (Trx1) that shuttles reducing equivalents to nuclear transcription factors via Ref-1 (59). As well, nucleus-specific Trx1 is more reduced (by −20 mV) than its cytosolic counterpart providing additional evidence of a reducing nuclear microenvironment (57).
A reduced nucleus appears to be essential for transcriptional regulation, as cysteine residues critical for protein structure or found within DNA binding domains generally must be reduced to promote transcription in the transcription factors p53, AP-1, Myb, and Sp-1 (1, 14, 15, 41, 61). In addition to promoting the repair of oxidative damage to DNA (47), a reduced nucleus may also shield genomic DNA from oxidation. In the same way, nuclear localization may protect oxidant-sensitive proteins from oxidation, possibly explaining the equal distribution of ZnB proteins between G1 and G2/M phases (Table 2). In support of this hypothesis, 19 of the 22 G1 proteins identified (including AC, H, MeB, and ZnB proteins) function within the nucleus.
Another model for redox regulation of cell cycle progression highlights the potential for oxidation to initiate and terminate the cell cycle. Treatment of cells with serum, PDGF-BB, or thrombin induces second messenger ROS [via NAD(P)H oxidase], resulting in ERK2, JNK1, and p38 MAPK activation, c-Fos, c-Jun, and JunB expression, and ultimately, cell cycle initiation through AP-1 (36, 44). The subsequent reestablishment of an oxidized intracellular environment after cytokinesis could act as a cell cycle brake through the oxidation (inhibition) of cell cycle initiating proteins harboring oxidant-sensitive motifs, such as the M phase-inducing phosphatases Cdc25B and Cdc25C (39, 52). Oxidation of the catalytic cysteine residues (from their AC motifs) of Cdc25B and Cdc25C would inhibit their dephosphorylation (activation) of cyclin/cyclin-dependent kinases and arrest proliferation. Thus a dynamic intracellular redox environment may serve positive and negative regulatory and protective roles during the cell cycle. Specifically, 1) the oxidized G1 phase restricts oxidant-sensitive G2/M cell cycle proteins while favoring mitogen-induced ROS second messenger signal transduction, which 2) induces a reduction in the intracellular environment prior to S phase entry to prevent oxidative DNA damage and 3) enables the function of oxidant-sensitive G2/M cell cycle proteins (Fig. 3).
In summary, evidence of numerous candidate redox-regulated cell cycle proteins contributing to a diverse array of cell cycle processes demonstrates the potential for an oscillating intracellular redox environment to regulate cell cycle progression. Although the approach taken here does not demonstrate empirically how changes in the intracellular redox environment affect candidate cell cycle proteins per se, combining a novel and objective method for the prediction of redox-regulated cell cycle proteins with an in vitro characterization of the intracellular redox environment has provided valuable insight into potential redox regulatory mechanisms influencing cell division and enabled model development to guide further experimentation. Thus far, only 24% (22 of 92) of the candidate proteins identified here have been demonstrated experimentally to be redox-regulated, highlighting the need for future research in this emerging field.
This work was supported by National Institutes of Health Grant RO1-DK-061568 (to H. R. Gaskins) and by a USDA National Needs Graduate Fellowship (to J. E. Conour).
We thank Dr. Barbara Pilas and Ben Montez at the Flow Cytometry Facility for technical assistance with the flow cytometric analysis. We thank Dr. Isaac Cann, Dr. Tarannum Khan, Dr. Dale King, and Dr. Juan Marini for review of the manuscript.
↵1 The Supplementary Material for this article is available online via http://physiolgenomics.physiology.org/cgi/content/full/00058.2004/DC1.
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: H. R. Gaskins, Univ. of Illinois, 1207 W. Gregory Drive, Urbana, IL 61801 (E-mail:).
- Copyright © 2004 the American Physiological Society