**3.1.1 Nucleus**

276 Selected Topics in DNA Repair

There are evidence that point to a role of redox environment in a short term regulation of the activity of this important enzyme. Minamino et al., 2001, using vascular smooth muscle cells, reported that hypoxia up-regulates telomerase activity. Hypoxia is known to lower oxidative stress and thus to increase levels of glutathione. A specific inhibitor of telomerase, 2-[3-(trifluoromethyl) phenyl]isothiazolin-3-one, reacts with a key cysteine residue, which is essential for telomerase activity and must be kept reduced. Consequently, it has been reported that dithiothreitol reverses this inhibition (Hayakawa et al., 1999). Furthermore, antioxidants have been shown to inhibit nuclear export of telomerase reverse transcriptase and thus delay replicative senescence of endothelial cells (Haendeler et al., 2004). In conclusion, a critical cysteine residue must be kept reduced in order to maintain full telomerase activity. It is likely that the glutathione redox potential may be important in this

Previous findings of our group showed that telomerase is regulated by the shift in glutathione redox potential within values similar to those found *in* vivo and alterations in telomerase activity are coordinated with changes in critical cell cycle proteins, particularly

Fig. 1. Reduced glutathione regulates telomerase activity in 3T3 fibroblasts.

and/or progression of atherothrombosis (Fujii et al., 2006).

Thus, physiological variations in glutathione level induce changes in telomerase activities that are in concordance with changes in cell cycle regulatory proteins. A number of reports have shown similar results. Brown et al., 2007 demonstrated for the first time *in vivo* that high hepatic glutathione levels correlate with increased telomerase activity. Also, the importance of glutathione regulation in telomerase activity has been proved in endothelial progenitor cells (EPC): impairment of antioxidant defences in EPC promoted oxidant mediated apoptosis and telomerase inactivation which subsequently lead to development

Recent data suggest that telomerase activity is regulated and ordered by telomere structure and telomerase assembly. Experimental evidence suggests that the telomere structure may change in a cell cycle-dependent manner to restrict telomerase activity to S phase (Hug & Ligner, 2006). In addition, telomere structure and specially the telomeric G-overhangs generation are strictly regulated during S phase and prolonged to other cell cycle phases depending on whether are telomeres from the lagging or the leading telomere (Dai et al., 2010). For this to happen, the precise control of the changes not only in telomerase conformation, but in chromatin structure (i.e. in its compactation level) as well, is of vital

process.

Id2 and E2F4 (Borras et al., 2004).

Although the role of nuclear GSH in the synthesis of DNA (Thelander & Reichard, 1979) and in protection against oxidative damage or ionizing radiation (Biaglow et al., 1983) is well established, little is known about the concentration of GSH in the nucleus and its regulation. This is due to two main factors. The first is methodological: it is impossible to determine the nuclear concentration of GSH using standard cell fractionation and analytical approaches (for a review see Söderdahl et al., 2003). The second factor is that most, if not all, of the reports share the common view of nuclear GSH distribution in a static situation. Cells are usually studied under steady state conditions *i.e.* when they are confluent (G0/G1 phase of the cell cycle). The nuclear membrane dissolves during mitosis and is formed again around newly replicated DNA packed in chromosomes; this spectacular change involves a variety of regulatory mechanisms. Therefore, if the nuclear GSH distribution is studied, the cell cycle physiology should be carefully considered.

The role of GSH in cell cycle regulation has been addressed mainly from the point of view of its overall cellular content. This is surprising since it is in the nucleus where most cell cycle progression events take place. The nucleus changes dramatically during the different phases of cell cycle, and failing to consider the corresponding changes in its redox environment could confer an important disadvantage in elucidating the actual importance of glutathione in the control of cell proliferation.

The Nuclear Compartmentation of Glutathione: Effect on Cell Cycle Progression 279

or inhibitor of differentiation) was demonstrated to be of considerable importance in the regulation of cell growth, differentiation and cancer in many mammalian tissues (Norton, 2000; Yokota & Mori, 2002). Id2, in particular, was shown to disrupt antiproliferative effects of tumour suppressor proteins of the Rb family, thus allowing cell cycle progression (Lasorella et al., 1996). Indeed, the pattern of the Id2 expression detected by Western blotting, confirmed the distribution of the phases of the cell proliferation detected by flow cytometry. This observation was previously published by our group suggesting a redox regulation of this protein (Borras et al., 2004). In addition, the studies of liver regeneration, process that involves DNA synthesis and cell proliferation, gave further support to our

It was demonstrated that when the increase of GSH after partial hepatectomy was prevented, the liver regeneration was delayed and the total liver amount of the DNA was lower than in the control group (Huang et al., 2001). Furthermore, an early increase in Id2 gene has been demonstrated as well as the contribution of Id2 in the control of hepatocyte priming through modulation of c-myc expression (Rodriguez et al., 2006). All these support our notion that Id2 could be an excellent candidate as a protein marker of the redox

PCNA, a proliferating cell nuclear antigen, is a central protein in both DNA replication and repair. It's a "sliding clamp" that localizes proteins, such as DNA polymerase, to DNA and

Replication of mammalian genome starts at thousands of origins, called replication foci, which contain PCNA and are activated at different times during S phase. The dynamics of replication foci is still a matter of debate; there are contradictory reports on the organization of the DNA replication sites in diverse cell types attributable to the differences in the technical approach (Dimitrova & Gilbert, 2000; Kennedy et al., 2000). According to Dimitrova, SD and Berezney, R (Dimitrova & Berezney, 2002) there is no fundamental difference in the spatiotemporal organization of the DNA replication in primary, immortalized and malignant mammalian cells. On the contrary, Kennedy's group (Kennedy et al, 2000), observed different patterns of replication foci in primary versus immortalized cell lines, as well as their perinuclear localization in the contact-inhibited cells prior to cell cycle exit (Barbie et al., 2004). Another fundamental question was weather the replication foci are moving along the DNA in the process of the replication, or the DNA is spooling through fixed replication factories. It seems that the important body of evidence is accumulating supporting the fixed-replication-site model (Dimitrova & Gilbert, 2000; Leonhard et al., 2000). The replication machinary bind to DNA, but they are also tethered to an underlying framework called nuclear matrix or skeleton (Leonhardt et al., 2000). Regardless of the discrepancies in their findings, all the authors call attention to the importance of the preserving nuclear architecture in order to guarantee the correct development of the process of DNA replication (Dimitrova & Gilbert, 2000; Barbie et al., 2004; Leonhardt et al., 2000). In addition, it has been shown that chromosome territory organization depends on association with the nuclear skeleton (Leonhardt et al., 2000). More than 20 years ago, Dijkwell et al. (Dijkwell & Wenink, 1986) postulated that the maintenance of the nuclear matrix, especially nuclear lamina, by preserving disulphide bonds depended on the level of nuclear thiols. In accordance to this work, Oleinick et al. (Oleinik et al., 1987)

**3.2.2 PCNA as a possible redox sensor in the onset of DNA synthesis**

findings.

regulation of cell proliferation in our models.

thus enables the correct DNA replication.

Fig. 2. Modifications of nuclear proteins along the cell cycle.

Our results, in general, are in accordance with the work of Hutter at al, (Hutter et al., 1997), who have studied the redox potential (E) in the normal and malignant cells along the cell cycle. The E of normal cells fluctuates during the cell cycle; for the proliferation to start it has to decrease at least 30mV comparing to its level in the G0 phase (where the cells are 100% confluent). On the contrary, in cancer cells, the E remains low throughout the complete cell cycle, even at a high cell density (Hutter et al., 1997). Our evaluation of the GSH levels along the cell cycle in different models indirectly confirms these findings.

The work of Hutter et al. was further developed and completed by Hoffman et al. (Hoffman et al., 2008), who proposed recently a novel redox model of cell proliferation. They postulate the existence of a redox switch that helps regulate the proliferation within normal cells; its absence in cancer cells enables the bypass of the restriction point and leads to the loss of the control of cell cycle. The authors offer this model as a base to understand the aberrant cellular proliferation that leads to malignant transformation.

According to "redox model of cell proliferation", in normal cells exist a threshold value (θ) of E≤ 207±11mV which initiates the phosphorylation of different regulatory proteins associated with different phases of the cell cycle and, consequently, cell proliferation. When E>θ cell enters G1/G0 phase. However, when cancer cells are concerned, Hoffman does not take into account the fluctuations in the E level. In our hands, the increase in GSH level along the cell cycle (before the onset of the proliferation comparing to the GSH level at the final time point) is, indeed, less striking than in 3T3 fibroblasts; two-fold comparing to fourfold, respectively. The author hypothesizes, though, the existence of a reductive limit of - 260mV which normal cell could not survive, but would not jeopardise cancer cell.

It was proposed previously that the proliferation occurs within the range of ROS levels, concentrations above or below this range could lead to growth arrest or cell death; hence ROS could act like a dual-edged sword (Davies, 1999).

#### **3.2 Glutathione controls the cell cycle regulatory proteins 3.2.1 Id2 as a redox sensitive protein**

The study of the expression of the Id2 along the cell cycle gave further support to this premise. The family of helix-loop-helix proteins denominated Id (inhibitor of DNA binding

Our results, in general, are in accordance with the work of Hutter at al, (Hutter et al., 1997), who have studied the redox potential (E) in the normal and malignant cells along the cell cycle. The E of normal cells fluctuates during the cell cycle; for the proliferation to start it has to decrease at least 30mV comparing to its level in the G0 phase (where the cells are 100% confluent). On the contrary, in cancer cells, the E remains low throughout the complete cell cycle, even at a high cell density (Hutter et al., 1997). Our evaluation of the GSH levels along

The work of Hutter et al. was further developed and completed by Hoffman et al. (Hoffman et al., 2008), who proposed recently a novel redox model of cell proliferation. They postulate the existence of a redox switch that helps regulate the proliferation within normal cells; its absence in cancer cells enables the bypass of the restriction point and leads to the loss of the control of cell cycle. The authors offer this model as a base to understand the aberrant

According to "redox model of cell proliferation", in normal cells exist a threshold value (θ) of E≤ 207±11mV which initiates the phosphorylation of different regulatory proteins associated with different phases of the cell cycle and, consequently, cell proliferation. When E>θ cell enters G1/G0 phase. However, when cancer cells are concerned, Hoffman does not take into account the fluctuations in the E level. In our hands, the increase in GSH level along the cell cycle (before the onset of the proliferation comparing to the GSH level at the final time point) is, indeed, less striking than in 3T3 fibroblasts; two-fold comparing to fourfold, respectively. The author hypothesizes, though, the existence of a reductive limit of -

It was proposed previously that the proliferation occurs within the range of ROS levels, concentrations above or below this range could lead to growth arrest or cell death; hence

The study of the expression of the Id2 along the cell cycle gave further support to this premise. The family of helix-loop-helix proteins denominated Id (inhibitor of DNA binding

260mV which normal cell could not survive, but would not jeopardise cancer cell.

Fig. 2. Modifications of nuclear proteins along the cell cycle.

the cell cycle in different models indirectly confirms these findings.

cellular proliferation that leads to malignant transformation.

ROS could act like a dual-edged sword (Davies, 1999).

**3.2.1 Id2 as a redox sensitive protein** 

**3.2 Glutathione controls the cell cycle regulatory proteins** 

or inhibitor of differentiation) was demonstrated to be of considerable importance in the regulation of cell growth, differentiation and cancer in many mammalian tissues (Norton, 2000; Yokota & Mori, 2002). Id2, in particular, was shown to disrupt antiproliferative effects of tumour suppressor proteins of the Rb family, thus allowing cell cycle progression (Lasorella et al., 1996). Indeed, the pattern of the Id2 expression detected by Western blotting, confirmed the distribution of the phases of the cell proliferation detected by flow cytometry. This observation was previously published by our group suggesting a redox regulation of this protein (Borras et al., 2004). In addition, the studies of liver regeneration, process that involves DNA synthesis and cell proliferation, gave further support to our findings.

It was demonstrated that when the increase of GSH after partial hepatectomy was prevented, the liver regeneration was delayed and the total liver amount of the DNA was lower than in the control group (Huang et al., 2001). Furthermore, an early increase in Id2 gene has been demonstrated as well as the contribution of Id2 in the control of hepatocyte priming through modulation of c-myc expression (Rodriguez et al., 2006). All these support our notion that Id2 could be an excellent candidate as a protein marker of the redox regulation of cell proliferation in our models.

### **3.2.2 PCNA as a possible redox sensor in the onset of DNA synthesis**

PCNA, a proliferating cell nuclear antigen, is a central protein in both DNA replication and repair. It's a "sliding clamp" that localizes proteins, such as DNA polymerase, to DNA and thus enables the correct DNA replication.

Replication of mammalian genome starts at thousands of origins, called replication foci, which contain PCNA and are activated at different times during S phase. The dynamics of replication foci is still a matter of debate; there are contradictory reports on the organization of the DNA replication sites in diverse cell types attributable to the differences in the technical approach (Dimitrova & Gilbert, 2000; Kennedy et al., 2000). According to Dimitrova, SD and Berezney, R (Dimitrova & Berezney, 2002) there is no fundamental difference in the spatiotemporal organization of the DNA replication in primary, immortalized and malignant mammalian cells. On the contrary, Kennedy's group (Kennedy et al, 2000), observed different patterns of replication foci in primary versus immortalized cell lines, as well as their perinuclear localization in the contact-inhibited cells prior to cell cycle exit (Barbie et al., 2004). Another fundamental question was weather the replication foci are moving along the DNA in the process of the replication, or the DNA is spooling through fixed replication factories. It seems that the important body of evidence is accumulating supporting the fixed-replication-site model (Dimitrova & Gilbert, 2000; Leonhard et al., 2000). The replication machinary bind to DNA, but they are also tethered to an underlying framework called nuclear matrix or skeleton (Leonhardt et al., 2000). Regardless of the discrepancies in their findings, all the authors call attention to the importance of the preserving nuclear architecture in order to guarantee the correct development of the process of DNA replication (Dimitrova & Gilbert, 2000; Barbie et al., 2004; Leonhardt et al., 2000). In addition, it has been shown that chromosome territory organization depends on association with the nuclear skeleton (Leonhardt et al., 2000). More than 20 years ago, Dijkwell et al. (Dijkwell & Wenink, 1986) postulated that the maintenance of the nuclear matrix, especially nuclear lamina, by preserving disulphide bonds depended on the level of nuclear thiols. In accordance to this work, Oleinick et al. (Oleinik et al., 1987)

The Nuclear Compartmentation of Glutathione: Effect on Cell Cycle Progression 281

and Nrf2, cysteine residue within DNA binding domain must be reduced. Both processes are guaranteed by the adequate redox state of the cytosolic and nuclear environment,

Interestingly, the nuclear proteins underwent stronger glutathionylation before and at the onset of cell proliferation than at quiescence. It is not surprising if we bear in mind that high level of GSH in the nucleus could provide protection to the proteins against the oxidative threat coming from the cytoplasm at the early phase of cell proliferation, and that glutathiolation, as it is a reversible modification, could be just the way. On the other hand, based on the simplicity of the redox transition from thiol to disulfide and on the fact that the reversibility was energetically favourable, Cotgrave IA and Gerdes RG (Cortgreave & Gerdes, 1998) more than 10 years ago have proposed glutathionylation as a posttranscriptional modification with the regulative finality. They state that it offers "a strong possibility for transducing "oxidative information" from intracellular oxidants via the GSH redox buffer to individual proteins containing "regulatory thiols". Also, recently, this posttranslational modification was proposed as a likely molecular mechanism for redox dependent signalling mediated by GSH (Fratelli et al., 2005). Thus, high level of GSH in the nucleus, observed before and at the onset of cell proliferation, could provide the "GSH redox buffer" necessary for the progressing of oxidant stimulated mitogenesis. It is encouraging to see that the findings of the present study have provided some support to the assumption that a dynamic intracellular redox environment directs the cell proliferation

**4. The depletion of nuclear glutathione hampers the cell cycle progression**  With the intention of providing further evidence of the importance of nuclear GSH in the initiation of cell proliferation, we have found ourselves in front of a challenge of depleting nuclear glutathione. A number of reports have focused on the consequences of the depletion of cellular glutathione levels on changes in cellular proliferation (Thomas et al., 1995; Hansen et al., 2006). However, all those reports were performed measuring cellular or total glutathione levels, but there is no information relating cellular proliferation with nuclear glutathione levels. A number of studies have indicated the existence of a nuclear GSH pool that resists depletion after exposure of cells to BSO (Thomas et al., 1995). BSO treatment resulted in the concentration dependent depletion of cytoplasmic GSH, while the depletion of mitochondrial and nuclear pool of GSH required concentrations higher than 100 µM, which induced DNA damage (Green et al., 2006). Spyrou and Holmgren (Spyrou & Holmgren, 1996) showed that inhibition of glutathione synthesis by 0.1 mM BSO was able to decrease GSH synthesis after treatment for 12 hours, but GSH-depleted cells grew as well as control 3T6 cells with no decrease in DNA synthesis. Thus, incubation of cells with low concentration of BSO, although decreases glutathione levels, does not change cell proliferation. On the other hand, Thomas *et al.* (Thomas et al., 1995) showed that non toxic concentrations of N-ethyl maleimide or DEM decreased the GSH level in the nucleus and cytoplasm to a similar extent, whereas the nuclear pool of GSH was much more resistant to

Based on this findings, we have designed a model to study the effects on the cell proliferation parameters caused by GSH depletion both in the nucleus and the cytoplasm, using 100µM DEM, comparing to the administration of 10 µM BSO when nuclear GSH level

respectively.

BSO depletion.

is preserved.

through redox sensitive cell cycle proteins.

reported that DNA-protein cross links, present at basal level as normal associations of chromosomal loops with the nuclear matrix proteins, can be increased by ionising radiation and removal of intracellular glutathione, and decreased by hydroxyl radical scavengers. The importance of the GSH in cell proliferation could be extrapolated to the safeguarding of the nuclear architecture, providing in that way a proper milieu for the DNA replication. Our results could provide support to the hypothesis of Bellomo et al (Bellomo et al., 1997) that reduced nuclear glutathione may modulate the structural organization of chromatin. It is tempting to speculate that the high nuclear GSH level we observed before (late G1 phase) and at the onset of cell proliferation (S phase) could provide the redox environment that stimulates chromatin decompaction by reducing disulfide bonds.

#### **3.3 Nuclear compartmentalization of glutathione as an important feature of proliferating cell: reduce to replicate**

The functional compartmentalization is an obvious characteristic of eukaryotic cell. The organelles, visible by light microscopy, are surrounded by membranes, which, although permitting communication, provide unique and defined environment in each one, which guarantee its accurate function. Probably the most remarkable examples of compartmentalization are oxidative phosphorylation in mitochondria, protein folding in endoplasmic reticulum and, for the purpose of this study the most interesting of all, DNA synthesis. It is interesting to note that, when the first two organelles are concerned, the dependence of their function on the correct GSH level has been thoroughly studied. The high intramitochondrial concentration of GSH is maintained by an active multicomponent transport system "pumping" glutathione from the cytosol into the matrix (Matensson et al., 1990). On the contrary, for the correct folding of proteins into a native structure by disulfide bond formation, the GSH level and the ratio GSH/GSSG in the endoplasmic reticulum is maintained at extremely low level by the limited permeability of the vesicle membrane to GSH (Hwang et al., 1992). However, in the case of nuclear compartmentalization of glutathione the reports were scarce and contradictory over the years. This could be attributed to two main factors: methodological difficulties in measuring nuclear glutathione content and the focus of the research generally limited to confluent cells.

#### **3.4 Modifications of nuclear proteins along the cell cycle**

Various studies have demonstrated that the nucleus is more reduced than the cytosol (15mM GSH vs. 11 mM, respectively) (Bellomo et al., 1997; Schafer & Buettner, 2001; Soboll et al., 1995). An important number of nuclear proteins, including transcription factors, require a reduced environment to bind to DNA. More than 62 proteins are involved directly in transcription, nucleotide metabolism, (de)phosphorylation, or (de)ubiquitinylation, which are all essential processes for cell cycle progression (Connour et al., 2004). For instance, it appears that, at the onset of cell proliferation in the early G1 phase, an increase of ROS in the cytoplasm is necessary for the initiation of the phosphorylation cascade mediated by epidermal growth factor (EGF) that, subsequently, activates DNA replication and the cell division (Carpentes & Cohen, 1990). According to Jang and Surh (Jang & Surh, 2003) nuclear GSH may act as a transcriptionalregulator of NF- B, AP-1, and p53 by altering their nuclear redox state. The transcription factor NF-κB is an example of distinct redox-sensitive activation and DNA binding (Hansen et al., 2006); it is activated by various physiological stimuli known to produce ROS; on the contrary, to permit DNA binding, similar to Fos, Jun,

reported that DNA-protein cross links, present at basal level as normal associations of chromosomal loops with the nuclear matrix proteins, can be increased by ionising radiation and removal of intracellular glutathione, and decreased by hydroxyl radical scavengers. The importance of the GSH in cell proliferation could be extrapolated to the safeguarding of the nuclear architecture, providing in that way a proper milieu for the DNA replication. Our results could provide support to the hypothesis of Bellomo et al (Bellomo et al., 1997) that reduced nuclear glutathione may modulate the structural organization of chromatin. It is tempting to speculate that the high nuclear GSH level we observed before (late G1 phase) and at the onset of cell proliferation (S phase) could provide the redox environment that

stimulates chromatin decompaction by reducing disulfide bonds.

**proliferating cell: reduce to replicate** 

**3.3 Nuclear compartmentalization of glutathione as an important feature of** 

content and the focus of the research generally limited to confluent cells.

**3.4 Modifications of nuclear proteins along the cell cycle** 

The functional compartmentalization is an obvious characteristic of eukaryotic cell. The organelles, visible by light microscopy, are surrounded by membranes, which, although permitting communication, provide unique and defined environment in each one, which guarantee its accurate function. Probably the most remarkable examples of compartmentalization are oxidative phosphorylation in mitochondria, protein folding in endoplasmic reticulum and, for the purpose of this study the most interesting of all, DNA synthesis. It is interesting to note that, when the first two organelles are concerned, the dependence of their function on the correct GSH level has been thoroughly studied. The high intramitochondrial concentration of GSH is maintained by an active multicomponent transport system "pumping" glutathione from the cytosol into the matrix (Matensson et al., 1990). On the contrary, for the correct folding of proteins into a native structure by disulfide bond formation, the GSH level and the ratio GSH/GSSG in the endoplasmic reticulum is maintained at extremely low level by the limited permeability of the vesicle membrane to GSH (Hwang et al., 1992). However, in the case of nuclear compartmentalization of glutathione the reports were scarce and contradictory over the years. This could be attributed to two main factors: methodological difficulties in measuring nuclear glutathione

Various studies have demonstrated that the nucleus is more reduced than the cytosol (15mM GSH vs. 11 mM, respectively) (Bellomo et al., 1997; Schafer & Buettner, 2001; Soboll et al., 1995). An important number of nuclear proteins, including transcription factors, require a reduced environment to bind to DNA. More than 62 proteins are involved directly in transcription, nucleotide metabolism, (de)phosphorylation, or (de)ubiquitinylation, which are all essential processes for cell cycle progression (Connour et al., 2004). For instance, it appears that, at the onset of cell proliferation in the early G1 phase, an increase of ROS in the cytoplasm is necessary for the initiation of the phosphorylation cascade mediated by epidermal growth factor (EGF) that, subsequently, activates DNA replication and the cell division (Carpentes & Cohen, 1990). According to Jang and Surh (Jang & Surh, 2003) nuclear GSH may act as a transcriptionalregulator of NF- B, AP-1, and p53 by altering their nuclear redox state. The transcription factor NF-κB is an example of distinct redox-sensitive activation and DNA binding (Hansen et al., 2006); it is activated by various physiological stimuli known to produce ROS; on the contrary, to permit DNA binding, similar to Fos, Jun, and Nrf2, cysteine residue within DNA binding domain must be reduced. Both processes are guaranteed by the adequate redox state of the cytosolic and nuclear environment, respectively.

Interestingly, the nuclear proteins underwent stronger glutathionylation before and at the onset of cell proliferation than at quiescence. It is not surprising if we bear in mind that high level of GSH in the nucleus could provide protection to the proteins against the oxidative threat coming from the cytoplasm at the early phase of cell proliferation, and that glutathiolation, as it is a reversible modification, could be just the way. On the other hand, based on the simplicity of the redox transition from thiol to disulfide and on the fact that the reversibility was energetically favourable, Cotgrave IA and Gerdes RG (Cortgreave & Gerdes, 1998) more than 10 years ago have proposed glutathionylation as a posttranscriptional modification with the regulative finality. They state that it offers "a strong possibility for transducing "oxidative information" from intracellular oxidants via the GSH redox buffer to individual proteins containing "regulatory thiols". Also, recently, this posttranslational modification was proposed as a likely molecular mechanism for redox dependent signalling mediated by GSH (Fratelli et al., 2005). Thus, high level of GSH in the nucleus, observed before and at the onset of cell proliferation, could provide the "GSH redox buffer" necessary for the progressing of oxidant stimulated mitogenesis. It is encouraging to see that the findings of the present study have provided some support to the assumption that a dynamic intracellular redox environment directs the cell proliferation through redox sensitive cell cycle proteins.
