In humans, the oxidatively induced DNA lesion 8-hydroxyguanine (8-oxoG) is removed from DNA by hOgg1, a DNA glycosylase/AP lyase that specifically incises 8-oxoG opposite cytosine. We analysed the expression of hOGG1 mRNA in 18 lung cancer and three normal cell lines. Although hOGG1 was overexpressed in most cell lines, 2/18 (11.1%) showed a lower hOGG1 mRNA and protein expression (∼80% decrease) relative to normal cell lines. Liquid chromatography/mass spectrometry analysis showed increased levels of 8-oxoG in the two cell lines with the lowest hOGG1 mRNA expression. We examined the ability of nuclear and mitochondrial extracts to incise 8-oxoG lesion in cell lines H1650 and H226 expressing lower hOGG1 mRNA and H1915 and H1975 with higher than normal hOGG1 mRNA expression. Both nuclear and mitochondrial extracts from H1915 and H1975 cells were proficient in 8-oxoG removal. However, both cell lines with the lowest hOGG1 mRNA expression exhibited a severe reduction in 8-oxoG incision in both nuclear and mitochondrial extracts. Under-expression of hOGG1 mRNA and hOgg1 protein was associated with a decrease in mitochondrial DNA repair in response to oxidative damaging agents. These results provide evidence for defective incision of 8-oxoG in both nuclear and mitochondria of H1650 and H226 lung cancer cell lines. These results may implicate 8-oxoG repair defects in certain lung cancers.
8-Hydroxyguanine (8-oxoG), also known as 8-oxo-7,8-dihydroguanine, is a premutagenic lesion that pairs with adenine as well as cytosine during DNA replication, potentially resulting in a G:C to T:A transversion. This type of mutation is common in various tumor suppressor genes including p53 gene in human cancers (Hollstein et al., 1996; Gunther et al., 1997). The presence of 8-oxoG in DNA is derived through two pathways: incorporation of the oxidized precursor, 8-oxodGTP into DNA during DNA synthesis, and the direct oxidation of a guanine base in DNA. In yeast, 8-oxoG is a highly mutagenic lesion that if left un-repaired prior to DNA replication can result in the formation of a mutator phenotype (Tajiri et al., 1995; Thomas et al., 1997). Likewise, it has been postulated that lack of adequate 8-oxoG repair in humans could be highly mutagenic and may lead to tumorigenesis. In order to prevent the formation of this potentially harmful lesion, organisms possess three distinct enzymes namely MTH (Escherichia coli MutT), MYH (E. coli MutY) and hOgg1 (E. coli MutM or Fpg, yeast Ogg1) (Michaels and Miller, 1992). MTH is an NTPase that hydrolyses 8-oxodGTP in the nucleotide pool to 8-oxodGMP, thus preventing incorporation of 8-oxodGTP during DNA replication. MYH is a DNA glycosylase that removes adenine opposite 8-oxoG, while Ogg1 is a DNA glycosylase/AP lyase that preferentially removes 8-oxoG opposite cytosine. The existence of these three genes for repair of 8-oxoG supports the fundamental biological importance of this lesion. Structural and functional homologues of these three enzymes are conserved through evolution as supported by their existence in E. coli and yeast. hOgg1 removes 8-oxoG and 2,6-diamino-4-hydroxy-5-formamidopyrimidine (FapyGua) from DNA with similar specificities (Dherin et al., 1999) and that this paper deals with hOgg1 only.
OGG1-deficient mice have been shown to accumulate 8-oxoG in their DNA (Klungland et al., 1999; Arai et al., 2002). One recent knockout study showed that loss of hOgg1 protein function predisposed mice to lung adenocarcinoma development (Sakumi et al., 2003). In humans, low hOgg1 levels in peripheral lymphocytes appeared to convey an increased risk of lung cancer development in smokers (Paz-Elizur et al., 2003). These findings support the notion that defects in hOGG1 and 8-oxoG repair may play an important step in tumorigenesis. The hOGG1 is localized to chromosome 3p25, a region that shows frequent loss of heterozygosity (LOH) in lung and kidney tumors (Audebert et al., 2000; Wikman et al., 2000). These findings suggest that cancer cells may have a reduced capacity to counter the mutagenic effects of reactive oxygen species, a deficiency that could increase genomic instability (decreased genomic integrity). Other studies have suggested that hOGG1 expression can be used as an exposure biomarker to assess cumulative oxidative DNA damage (Cherng et al., 2002).
In this study, we analysed the expression of hOGG1 mRNA by real-time quantitative reverse transcriptase–PCR (RT–PCR) in 18 lung cancer and three normal cell lines. Cellular survival and mitochondrial DNA (mtDNA) repair were analysed after oxidative damage exposure of lung cancer cell lines expressing higher and lower than normal hOGG1 mRNA. Additionally, we analysed 8-oxoG lesions and further examined 8-oxoG incision activity in the cancer cell lines expressing low hOGG1 as well as those with high hOGG1 mRNA and hOgg1 protein expression.
hOGG1 expression in lung cancer cell lines
We analysed the expression of hOGG1 mRNA in 18 lung cancer and three normal cell lines by RT quantitative real-time PCR. Figure 1 shows representative RT–PCR amplification curves for β-actin and hOGG1 genes in untreated lung cancer cell lines H1975 and H1650. The primers of hOGG1 used in this assay were chosen from exon 3, which is common to all hOGG1 isoforms, making it possible to quantitate transcripts from all isoforms. These results showed no difference in the expression of β-actin as evidenced by the overlapping amplification curves. However, amplification of hOGG1 showed a three cycle difference corresponding to eightfold (2^3) higher expression in the H1975 cell line. Table 1 shows that 13/18 (72.2%) cancer cell lines overexpressed hOGG1 mRNA, while 2/18 (11.1%) cancer cell lines demonstrated under-expression of hOGG1 mRNA relative to the normal human bronchial epithelial (NHBE), normal retinal epithelial (ARPE-19) cells and NC-37, a lymphoblast cell line. Three of 18 (16.7%) cancer cell lines demonstrated hOGG1 mRNA levels comparable to those of the normal cell lines (0.99–1.32). Four representative lung cancer cell lines were chosen from the above 18 cancer cell lines for further studies. Two of these (H1915 and H1975) displayed higher than normal hOGG1 mRNA expression, while H1650 and H226 exhibited the lowest hOGG1 mRNA expression (Table 1 and Figure 2a). Further analysis of hOGG1 mRNA by RT–PCR showed two major isoforms (α and β) hOGG1 and further that the α isoform (also known as nuclear hOGG1) was the most abundant in all but one cell line, H226, where there was severe reduction in both isoforms (Figure 2b). Figure 2b also shows that H1915 and H1975 expressed both forms of hOGG1 mRNA at levels similar to the normal lung cell line NHBE. However, H1650 cells expressed both α and β-hOGG1 mRNA at levels lower than NHBE and H226 cells had hOGG1 mRNA below detection level by conventional RT–PCR. The expression of G3PDH was the same in all five cell lines (Figure 2b). Additionally, analysis of hOGG1 mRNA expression in response to increasing concentration of hydrogen peroxide (H2O2) for 2 h showed no significant change in expression in H1915, a cell line with high hOGG1 mRNA expression (Figure 2c). Furthermore, mRNA analysis of α and β-hOGG1 showed no significant difference in H1915 in response to increasing concentrations of H2O2 (Figure 2d). Figure 2e shows hOGG1 cDNA expression in response to oxidative damage of H226 cells that exhibited a lower than normal hOGG1 expression. We increased the cycle number to 35 instead of 25 in order to visualize the α-hOGG1 band, however, the β-hOGG1 isoform remained very faint. Similar to the H1915, these results did not reveal any induction of hOGG1 mRNA in H226 cell after H2O2 exposure (Figure 2e). The slight decrease observed in the lanes seems to be due to input differences as evidenced by the different G3PDH levels which were higher in the lanes labeled 0 and 200 compared to lanes labeled 400 and 800 μ M H2O2. Similarly, we did not observe any induction of hOGG1 mRNA after oxidative stress in cell line H1975 and H1650 (data not shown). Western Blot analysis confirmed the RT quantitative real-time PCR results of the above four selected cell lines for both α and β-hOgg1 expression (Figure 3a). This result confirmed that the α-hOgg1 was the most abundant form in all cell lines, and further that cell lines H226 and H1650 had a severe reduction in both hOgg1 isoforms (Figure 3a). Figure 3b shows quantitative analysis for α-hOgg1 of the results in Figure 3a using a densitometry scanner. These results confirmed that low hOGG1 mRNA expression led to low protein levels. The hOGG1 mRNA in cell line H1915 was twofold higher than in H1975 as determined by real-time quantitative RT–PCR. However, conventional RT–PCR (Figure 2b) and Western blot analysis (Figure 3b) shows no difference in the expression of both α and β-hOGG1 in both H1915 and H1975. We attribute this difference to the several isoforms determined by the primers used in the quantitative RT–PCR. It is possible that the other isoforms of hOGG1 mRNA are more abundant in H1915 than in H1975.
Analysis of 8-oxodG by liquid chromatography-mass spectrometry (LC/MS) in genomic DNA
In order to determine whether the low hOGG1 mRNA and protein levels observed in the two lung cancer cell lines H1650 and H226 would compromise the cells' ability to remove 8-oxoG in vivo, we used LC/MS to analyse the nucleoside form of 8-oxoG, that is, 8-oxodG in the genomic DNA before and after a 2 h oxidative damage exposure (400 μ M H2O2). Our results showed that cell lines H1915 and H1975 had no significant difference in 8-oxoG levels before and after H2O2 treatment. However, cell lines H1650 and H226 showed a twofold and 2.5-fold increase in 8-oxoG levels after H2O2 treatment (Figure 4). The results also showed that baseline 8-oxoG levels were similar in all but one cell line (H226). This difference in baseline level of 8-oxoG may be due to differences in other defense mechanisms against 8-oxoG accumulation such as detoxification enzymes, and MTH and MYH enzymes that are involved in 8-oxodGTP hydrolysis prior to incorporation and removal of adenine incorporated opposite an 8-oxoG, respectively. These results suggest that cell lines H1650 and H226 are compromised in their ability to remove 8-oxoG from their DNA.
Incision of 8-oxoG by nuclear extracts of lung cancer cell lines
In order to determine whether low hOgg1 protein levels observed in H1650 and H226 cell lines were associated with reduced 8-oxoG repair, we examined the ability of nuclear extracts prepared from untreated lung cancer cell lines H1975, H1915, H1650 and H226 to incise 8-oxoG-containing oligonucleotide. Cell lines H1975 and H1915 were included as controls expressing higher than normal hOgg1 protein expression. We were unable to include extracts from normal lung NHBE cell line due to the limited passaging of this cell line. The incision activity of 8-oxoG by nuclear extracts from the four lung cancer cell lines is shown in Figure 5. Figure 5a shows a representative incision assay showing dose-dependent incision by nuclear extracts from H1650 and H1975. No incision products were observed when the 8-oxoG-containing oligonucleotide was incubated with buffer without the nuclear extract (lanes labeled 0) or when the control oligonucleotide containing undamaged guanine was incubated with the nuclear extracts (lanes labeled C). Fpg, an enzyme known to incise 8-oxoG opposite cytosine resulted in the expected 10-mer product. Lanes labeled F1 and F2 represent incubations of 8-oxoG oligonucleotide (0.2 pmoles) with 8 U of Fpg for 30 min and 4 h, respectively. The data suggested that the incision activity observed in the lanes with extracts was specific for 8-oxoG. Figure 5b shows a graphical representation of the dose-dependent experiment performed in duplicate. This figure shows that nuclear extracts of H1650 and H226 were severely deficient in 8-oxoG incision compared to cell lines H1915 and H1975. The slight decrease in hOgg1 activity observed in H1915 may be due to the presence of inhibitors or endonucleases that may be absent in H1975 especially at higher protein concentration. Figure 5c is a representation of the kinetic analyses of 8-oxoG incision by nuclear extracts (50 μg) of H1975 and H226. Lanes marked FC and FO represent incubation of control oligonucleotide lacking 8-oxoG and oligonucleotide containing 8-oxoG with 8 units of Fpg, respectively. Figure 5d shows a graphical representation of the kinetic analysis performed in duplicate for all four cell lines. Both the dose and kinetic analyses results showed that nuclear extracts from cell lines H1975 and H1915 were proficient in the removal of 8-oxoG, exhibiting the same initial rate. However, cell lines H1650 and H226 were deficient in 8-oxoG incision, suggesting that low hOGG1 expression in these cells lines led to decreased repair of 8-oxoG.
Incision of 8-oxoG by mitochondrial extracts
We measured the 8-oxoG incision activity in the mitochondrial extracts of the four lung cancer cell lines described above. Figure 6a represents 8-oxoG incision assay gel by mitochondrial extracts of H1915, H1975, H1650 and H226. Lane labeled C represents incubation of H1915 mitochondrial extracts with control oligo lacking 8-oxoG. Figure 6b shows a graphical representation of the 8-oxoG incision results from 50 μg mitochondrial extracts of H1975, H1915, H1650 and H226 lung cancer cell lines performed in triplicate. Owing to limited passaging of the normal lung cell line NHBE, we were unable to obtain mitochondrial extracts for the incision assays. The results indicated that mitochondrial extracts from H1650 and H226 lung cancer cell lines were deficient in the incision of 8-oxoG (Figure 6b), compared with those from the H1975 and H1915. These results suggest a defect in the incision of 8-oxoG by mitochondrial extracts of lung cancer cells expressing lower than normal hOgg1. To ensure that the mitochondrial extracts of cell lines H1650 and H226 that showed a low 8-oxoG incision activity were enzymatically active, we tested the incision of uracil, another lesion repaired by the uracil DNA glycosylase (UDG) through the base excision repair pathway. Only 1/10th (5 μg) of the extract amount used in 8-oxoG incision was used for uracil incision. The results showed that all the mitochondrial extracts were active, with ⩾75% uracil incision in 1 h (Figure 6c). Therefore, the defect observed in cell lines H1650 and H226 was specific for the 8-oxoG lesion and not a general defect of the base excision repair pathway from those cell lines.
hOGG1 expression and mtDNA integrity/repair
Mitochondrial DNA (mtDNA) is believed to be more susceptible to oxidative damage than nuclear DNA (nDNA) (Richter et al., 1988). Several possible factors may account for these differences in susceptibility including exposure to high levels of reactive oxygen species produced during oxidative phosphorylation, lack of protective histones, and limited capacity for repair of DNA damage (Lightowlers et al., 1997). We used quantitative real-time PCR to analyse mtDNA integrity after exposure of the high hOGG1-expressing cell line, H1915 and low hOGG1-expressing cell lines: H226 and H1650 to H2O2. Quantitative real-time PCR assay allows for the measurement of DNA damage in individual amplifiable DNA segments. The fundamental principle of this assay is that DNA damage will impede the progression of the DNA polymerase used in the PCR reactions (Yakes and Van Houten, 1997; Ayala-Torres et al., 2000). Thus, DNA damage is detected as a reduction of the available template for PCR (decreased DNA integrity), resulting in a shift of the amplification curve to the right. Ratios of mtDNA/nDNA are used to obtain the relative extent of damage whereby a lower ratio represents less initial template, denoting a decrease in the integrity of mtDNA. The major advantage of the quantitative real–time PCR assay is that only nanogram quantities of DNA are required, and that DNA damage can be assessed at the individual gene level.
We selected a control region from the nuclear β-actin gene and a specific mtDNA region: D310, encompassed in the D-loop. mtDNA integrity was analysed by calculating the mtDNA/nDNA ratio after normalizing each mitochondrial signal to the corresponding nuclear β-actin signal. Figure 7a shows representative real-time amplification curves for mtDNA-D310 region in untreated H1650 cells (curves marked – D310 control) and at 24 h post H2O2 treatment (curves marked – D310-H2O2), and β-actin before and after H2O2 treatment. The β-actin curves for the control and at 24 h post treatment overlapped, suggesting equal DNA input, and that there was no damage to this area in H1650 cell line. However, upon H2O2 treatment, there was a 2.5 cycle shift to the right from curve D310 control to D310 H2O2 curves (Figure 7a), indicating less available template due to DNA damage, hence a decrease in mtDNA integrity. The fact that DNA damage was detected 24 h post H2O2 treatment suggests that H1650 cells were not able to repair damage induced in the mitochondrial D310 region. In contrast, cell line H1915 showed no difference in D310 amplification between the untreated control and at 24 h post H2O2 treatment (Figure 7b). Likewise, no difference was observed between β-actin amplification in the untreated control and at 24 h post H2O2 treatment as evidenced by the overlapping curves (Figure 7b). Together, these results suggest that the differences observed in D310 amplification between the two cell lines were not due to input differences but rather due to increased damage of mtDNA in H1650. A quantitative analysis of the real-time PCR is shown in Figure 7c, which indicates that cell line H1915 with higher than normal hOgg1 expression was able to recover after 6 h post H2O2 treatment (Figure 7c hatched bars). Conversely, H1650 and H226 with low hOgg1 protein expression showed continued decrease in mtDNA integrity that could not be repaired over 24 h post H2O2 treatment (Figure 7c, black and spotted bars, respectively).
hOGG1 and cellular sensitivity to oxidative damaging agents
We examined whether the differential expression of hOGG1 observed in the above lung cancer cell lines would have any effect on the ability of cells to survive under oxidative stress. Cell lines H1975 and H1915 (both expressing higher than normal hOGG1), and H1650 and H226 (both expressing lower than normal hOgg1) were exposed to H2O2 for 2 h. Figure 8 demonstrates that H1975, H1915 and H1650 were moderately sensitive to high concentrations of H2O2, a model oxidative damaging agent, while H226 was the most sensitive showing only a ∼30% survival after exposure to 400 μ M H2O2. Although the sensitivity of H226 to oxidative damaging agents is not completely understood, low levels of hOgg1 protein likely sensitize these lung cancer cells to oxidative damaging agents.
Sequencing of hOGG1 cDNA in lung cancer cell lines
We sequenced hOGG1 in the 18 lung cancer cell lines mentioned in Table 1. We found that, the H1650 cell line with the lowest hOgg1 protein expression but resistant to H2O2-induced oxidative damage, harbored a G to A mutation at codon 308 (data not shown). This mutation results in a Gly to Glu amino-acid change at codon 308. Of the 18 lung cancer cell lines in Table 1, only two (H1650 and H1299) harbored the G-A mutation at codon 308. Thus, 2/18 (11.1%) of the screened lung cancer cell lines harbored this specific codon 308 mutation in hOGG1. The H1299 cell line was also found to be resistant to oxidative damage (data not shown). Table 2 summarizes the major findings in hOGG1 variables studied in this manuscript.
In this study, we have examined the expression profile of hOGG1, which is involved in the repair of 8-oxoG lesions from DNA through the base excision repair pathway. hOGG1 mRNA was overexpressed in most lung cancer cell lines but a few cell lines showed lower expression when compared to normal cell lines. Western blot analysis confirmed that the low hOGG1 mRNA was associated with low hOgg1 protein expression. Low hOgg1 expression was associated with increased 8-oxoG lesions in genomic DNA. Furthermore, we showed that lung cancer cell lines with low hOgg1 expression had reduced 8-oxoG incision activity both in the nuclear and mitochondrial extracts. We also showed that upon oxidative damage exposure, mtDNA integrity was reduced only in the cell lines expressing low hOgg1 protein.
The hOGG1, MYH and MTH1 genes cooperate with each other to suppress the mutagenic events induced by 8-oxoG (Boiteux and Radicella, 1999). There are very limited studies that describe the repair of this potentially harmful lesion in cancer cells. Specifically, defects in 8-oxoG repair have only been reported in one leukemia cell (Hyun et al., 2000), one cervical cell line (Dobson et al., 2000) and two breast cancer cell lines (Mambo et al., 2002). We observed that a decrease in hOgg1 protein expression was associated with increased 8-oxoG lesions in genomic DNA upon oxidative damage exposure, indicating a deficiency in the in vivo removal of 8-oxoG. Consistent with these results, we showed that the removal of 8-oxoG from a synthetic oligonucleotide (8-oxoG incision activity) was compromised in lung cancer cell lines that expressed lower than normal hOGG1 mRNA and protein levels. These results suggest that hOGG1 may be an important base excision repair gene in lung cancer progression. Recent studies (Paz-Elizur et al., 2003) showed that hOgg1 activity was lower in peripheral blood mononuclear cells from patients with non-small-cell lung cancer than in those from control subjects. Paz-Elizur et al. (2003) studies provided evidence for defective repair of 8-oxoG in lung cancer patients suggesting perhaps that the lung cancer cells may be defective in 8-oxoG repair. To date, there are limited studies that have examined the repair of this mutagenic lesion in lung cancer cells. Previous studies have shown that lung cancer tissue harbored higher levels of oxidatively induced DNA lesions than cancer-free surrounding tissue (Olinski et al., 1992; Jaruga et al., 1994), suggesting a possible defect in the repair of these lesions. Recently, low hOgg1 activity was shown to be significantly associated with an increased risk of non-small-cell lung cancer (Paz-Elizur et al., 2003). Other recent studies (Xie et al., 2004) showed that deficiencies in both myh and ogg1 in mice resulted in lung tumor predisposition and G to T mutations in activating hot spot codon 12 of K-ras oncogene in lung tumors. These data support the notion that hOGG1 and 8-oxoG may play a significant role in the initiation of a subset of lung cancers.
There are two major forms of hOGG1, α- and β-hOGG1, that are products of alternative splicing (Kohno et al., 1998; Dherin et al., 1999). Both forms have a putative mitochondria localization signal, but only the α-hOGG1 mRNA has a nuclear localization signal (Nishioka et al., 1999). The β-form has been shown to be targeted to the mitochondria (Takao et al., 1998; Nishioka et al., 1999).
Our analyses of the hOGG1 mRNA by RT–PCR showed that the nuclear form (α-hOGG1) was differentially expressed in the four lung cancer cell lines analysed (Figure 2b), with H1650 and H226 exhibiting the lowest α-hOGG1 mRNA expression. Thus, the reduced 8-oxoG incision activity observed in nuclear extracts of H1650 and H226 can be attributed to a reduction in α-hOgg1 protein expression observed in these cell lines. Our results also showed that the β-hOGG1 mRNA was present in lesser amounts in all but H226 cell line compared to the nuclear α-hOGG1 form. These results are similar to those from human tissues in which the α-hOGG1 was shown to be the most abundant form (Monden et al., 1999; Nishioka et al., 1999). Cell lines H1650 and H226 exhibited a severe reduction in β-hOGG1 (mitochondrial isoform) expression when compared to the H1975 and H1915. These results suggest that the reduced activity in the mitochondrial extracts of H1650 and H226 may be attributed to the low levels of β-hOgg1. Recent findings by Hashiguchi et al. (2004) have provided evidence that suggest that β-hOgg1 may be enzymatically inactive, implying perhaps that deficiencies observed in the mitochondria may be due to other factors. We observed >75% uracil incision by mitochondria extracts of all four lung cancer cell lines, however, only H1650 and H226 showed a defect in 8-oxoG incision in the mitochondria, suggesting that the defect was lesion specific and not a general defect of base excision repair in the mitochondria of these cells. The observed defects of 8-oxoG incision in both H1650 and H226 lung cancer cell lines are likely due to defective expression of both α and β-hOGG1 isoforms.
Cell lines H1650 and H226 that exhibited a low hOGG1 mRNA expression also showed reduced protein expression of both the α- and β-hOGG1, suggesting that the regulation of hOGG1 gene most likely occurs at a transcriptional level. However, little is known about the factors that regulate hOGG1 expression. Conventional methylation-specific PCR and sequencing of bisulfite-treated DNA revealed no correlation between hOGG1 promoter hypermethylation and low hOGG1 expression (data not shown). The hOGG1 gene is not regulated through the cell cycle and appears to be constitutively expressed (Dhenaut et al., 2000, 2001). Thus, our finding that hOGG1 expression is not controlled through promoter methylation is consistent with its house keeping function. We also showed that in H1915 and H226 cells, hOGG1 mRNA was not induced by H2O2-induced damage, a phenomenon that has been previously observed (Hodges and Chipman, 2002). Further studies of hOGG1 regulation are needed in order to understand the differential expression.
Low expression of hOgg1 is expected to sensitize cells to oxidative damage. Our analysis of cellular survival after exposure to H2O2, an oxidative damaging agent, showed that the cell line with the least hOgg1 protein expression (H226) had decreased cellular survival when compared to the cell lines with high hOgg1 expression such as H1975. Recent studies by Sava et al. (2004) showed that a forced decrease in OGG1 activity was associated with decreased cell viability in PC12 a neuronal cell line. Together, these results suggest that genome integrity is crucial for cellular survival. Our results also showed that cell line H1650 with low hOgg1 protein expression was resistant to H2O2. This observation led us to sequence the open reading frame of hOGG1 and we identified a G → A mutation resulting in glycine to glutamate amino-acid change at codon 308. To date, only a few hOGG1 mutations have been described. Specifically, a G → A mutation resulting in an Arg46Gln amino-acid change has been shown to reduce the enzymatic activity of hOgg1 protein (Audebert et al., 2000), and a G → A mutation resulting in a Gly12Glu amino-acid change blocks the translocation of the protein into the mitochondria (Audebert et al., 2002). The G to A homozygous mutation reported here causes a Gly308Glu amino-acid change and is located in conserved domain V, a position maintained as G in mouse, yeast and the α and β human OGG1. This mutation was previously reported in head and neck (Blons et al., 1999) and in kidney cancer patients (Audebert et al., 2000). To date, there are three hOGG1 mutations reported in primary lung cancer tissue: Arg46Gln, Ala85Ser and Arg131Gln (Chevillard et al., 1998; Wikman et al., 2000). The above substitutions at codons 308, 46 and 85 have been shown to be rare genetic polymorphisms, while the Arg131Glun is a somatic mutation (reviewed in Shinmura and Yokota, 2001). Although the G at codon 308 is highly conserved through many species, and is thus expected to alter the enzyme activity, only one study (Blons et al., 1999) showed no activity difference between the wild-type hOGG1 and the Gly308Glu mutant hOGG1. Other studies have shown that alterations in hOGG1 have been associated with increased risk for lung cancer. Specifically, hOGG1 Ser326Cys polymorphism, commonly observed in lung cancer, has been shown to reduce hOgg1 enzyme activity, suggesting that this polymorphism plays an important role in the risk for lung cancer (Kohno et al., 1998; Wikman et al., 2000). However, the contribution of the Ser326Cys polymorphism to lung cancer remains controversial as other studies suggested that there may be no association of this polymorphism with the risk of lung cancer development, and further that the contribution of Ser326Cys to lung cancer risk may be different among different geographic populations (Wikman et al., 2000; Vogel et al., 2004; Hu and Ahrendt, 2005). Further studies are needed to assess the importance of the interpopulation variation to cancer susceptibility. Therefore, further studies are needed to evaluate the importance of the Ser326Cys polymorphism to cancer susceptibility. The H1650 cells with low hOgg1 expression were resistant to H2O2-induced oxidative damage. The basis for the resistance of the H1650 cells is not yet known.
We analysed mitochondrial DNA integrity in the four lung cancer cell lines discussed above. Our results revealed that cell lines with lower than normal hOgg1 expression (H226 and H1650) had reduced mtDNA integrity after exposure to oxidative damage when compared to the cell line expressing higher than normal hOgg1. Although H2O2 induces a wide range of lesions that include strand breaks, which can block polymerases, the differences in mtDNA integrity between the low and high hOgg1-expressing cell lines could in part be attributed to a deficient repair of 8-oxoG. Some studies have shown that the 8-oxoG lesion caused decreased promoter clearance by E. coli RNA polymerase through stalling, resulting in a reduction in the amount of full-length product formation (Viswanathan and Doetsch, 1998; Kuraoka et al., 2003), indicating that DNA base damage, specifically 8-oxoG could impede RNA polymerases and perhaps DNA polymerases. However, this still remains controversial as other studies have shown that 8-oxoG does not block polymerases (Larsen et al., 2004). Previous studies have suggested that there is transcription-coupled repair (TCR) of 8-oxoG in human cells (Le Page et al., 2000a, 2000b, but this phenomenon is surrounded by some controversy. The findings from those studies revealed that TCR of 8-oxoG was present in ogg1−/− MEF cell line, suggesting that TCR of 8-oxoG is independent of OGG1 (Le Page et al., 2000a), and that several other factors are involved (Le Page et al., 2000b). Furthermore, Le Page's studies showed that Ogg1 protein was required for the repair of 8-oxoG present in the nontranscribed strand. Low hOgg1 protein expression is likely to be associated with a low hOgg1 activity. Consistent with this expectation, our results showed decreased 8-oxoG incision activity and increased 8-oxoG lesion in vivo, in the two cell lines with approximately 80% reduction in hOgg1 protein expression. Low hOgg1 activity can result in more DNA lesions and may give rise to a mutator phenotype in which mutations will arise faster in cells with a low hOgg1 expression. Previous studies by Dobson et al. (2000) have shown that targeting and expressing hOGG1 to the mitochondria enhanced mtDNA repair and increased survival in response to oxidative damage exposure. Conversely, reduction of hOGG1 expression is expected to decrease mtDNA repair and cellular survival in response to oxidative damage. This notion is supported by our mtDNA integrity/repair results, and the survival of H226 cells.
In summary, our results showed that there was a variable range of hOGG1 mRNA expression in lung cancer cell lines, with most cell lines overexpressing hOGG1 mRNA while only a few (11.1%) exhibited a lower than normal hOGG1 mRNA expression. A lower than normal hOGG1 mRNA expression was associated with increased 8-oxoG lesion in genomic DNA and reduced mtDNA integrity in response to oxidative damage, suggesting that hOGG1 gene plays a role in the maintenance of the mitochondrial genome. Our results showed that a proportion of the lung cancer cell lines had compromised incision of 8-oxoG both in the nucleus and mitochondria, suggesting that abrogation of part of the base excision repair pathway may be important in a subset of lung cancers. Since lung tissue is a target organ for induced DNA damage, it is important to understand DNA repair pathways in this critical tissue. We therefore suggest that examining Ogg1 protein expression in primary lung tumors may lead to an important screening tool in determining individuals at high risk. Additionally, knowledge of hOGG1 expression and mutational status (in addition to other DNA repair genes) may help in determining response to specific chemotherapeutic drugs in lung cancer patients.
Materials and methods
All cell lines except NHBE were obtained from ATCC and grown in their specified media, supplemented with 10% fetal bovine serum (Hyclone), 5% Penicillin–Streptomycin. Total RNA was isolated using TriZol (Invitrogen, Carlsbad, CA, USA) as per the manufacturer's instructions. The normal human bronchial epithelial (NHBE) cell line was purchased from Cambrex (Rockville, MD, USA) and grown in BEGM (Bronchial Epithelial Medium) supplemented with growth factors provided by Cambrex (Cat #CC-3170). All DNA was isolated using the salt extraction method (Miller et al., 1988).
Quantitative RT real-time-PCR analysis of hOGG1 mRNA
The 7900HT sequence-detection system (Applied Biosystems, Foster City, CA, USA) was used to perform real-time quantitative RT-PCR amplification for hOGG1 and β-actin. Table 3 lists the primers and probes used to amplify the respective genes. All primers were obtained from Invitrogen (Carlsbad, CA, USA). All TaqMan probes (Applied Biosystems, Foster City, CA, USA) were labeled with 5′-FAM (6-carboxyfluorescein, fluorescent reporter) and 3′-TAMRA (6-carboxy-tetramethylrhodamine, fluorescence quencher). Quantitative RT–PCR was carried out in 1 × commercial PCR buffer (Invitrogen, Carlsbad, CA, USA), 3 mM MgCl2, 4 ng/μl BSA, 0.2 mM each of dATP, dCTP, dGTP and dTTP, 0.6 mM DTT, 200 nM each of forward and reverse primers, 200 nM TaqMan probe, 90 U of RT, 0.75 U Platinum Taq polymerase and 2% Rox reference dye. RNA (800 ng) was used to amplify both the hOGG1 and the β-actin, a control gene. The real-time quantitative RT–PCR reactions were performed in triplicate. The level of expression in each sample was obtained relative to β-actin.
Western blot analysis
Western blot analysis was performed as described (Mambo et al., 2002) using 30 μg protein lysates. Membranes were hybridized with rabbit polyclonal anti-hOgg1 antibodies (Novus Biologicals, Littleton, CO, USA) and developed with Amersham developer as per the manufacturer's instructions. Loading control for the proteins was performed by reprobing the membrane with nuclear envelope Lamin B and mitochondrial cytochrome c (Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibodies.
Measurement of 8-oxoG in genomic DNA
We used LC/MS as described (Dizdaroglu et al., 2001) to quantify the nucleoside form of 8-oxoG, that is, 8-oxodG in DNA isolated from untreated lung cancer cell lines as well as from lung cancer cell lines treated with 400 μ M of H2O2 for 2 h. All DNA was isolated using the salt extraction method (Miller et al., 1988) immediately after the 2 h treatment.
Oligonucleotides and 5′ end-labeling
Oligonucleotides containing either a single 8-oxoG lesion or uracil were obtained from Midland Certified Reagent Co. (Midland, TX, USA). Oligonucleotide sequences were as described (Bohr et al., 2002). Oligonucleotides (50 pmoles) were 5′ end-labeled using 30 μCi of [γ-32P]ATP 3000 Ci/mmol (Perkin Elmer, Torrance, CA, USA) and T4 polynucleotide kinase (New England Biolabs, Beverly, MA, USA). A fourfold excess of the complementary oligonucleotide was added to each labeled oligonucleotide. The oligonucleotides were denatured at 95°C for 5 min and allowed to anneal by gradual cooling to room temperature.
8-oxoG incision assay
Nuclear and mitochondria extract preparation was performed as previously described (Croteau et al., 1997; Mambo et al., 2002). The assay is based on the cleavage of an 8-oxoG-containing oligonucleotide. 8-oxoG incision assay was performed using oligonucleotide sequences as described (Bohr et al., 2002). Dose-dependent experiments were performed for nuclear extracts with 30–150 μg protein and incubated for 4 h at 37°C. Kinetic experiments were carried out with 50 μg nuclear extract and incubated for 2, 4, 6 and 24 h at 37°C. 8-oxoG incision in mitochondria was examined by incubating 50 μg mitochondrial extracts for 6 h at 37°C. Positive control reaction was performed by incubating labeled oligonucleotides (0.2 pmoles) with Fpg, a glycosylase/AP lyase known to cleave 8-oxoG opposite C (Michaels et al., 1991; Tchou et al., 1991). To test for the specificity of the incision activity, two other controls were included: (a) incubation of all reagents in the absence of extracts; and (b) incubation of extracts with a duplex oligonucleotide of the same sequence but without 8-oxoG lesion. Protein extracts (5 μg) were also incubated with a uracil-containing oligonucleotide at 37°C for 1 h. The gels were exposed to PhosphorImager cassettes, and signals were quantified using ImageQuant software (Molecular Dynamics, Sunnyvale, CA, USA).
mtDNA integrity and repair
Cell lines H1915, H1650 and H226 were treated with 400 μ M H2O2 for 2 h and allowed to recover for 4, 6 and 24 h post treatment. The 7900HT sequence detection system was used to perform real-time quantitative PCR amplification for nuclear β-actin and the mtDNA region D-loop region D310. Table 3 lists the primers and probes used to amplify the respective DNA regions. All primers were obtained from Invitrogen (Carlsbad, CA, USA). All TaqMan probes (Applied Biosystems, Foster City, CA, USA) were labeled with 5′-FAM (6-carboxyfluorescein, fluorescent reporter) and 3′-TAMRA (6-carboxy-tetramethylrhodamine, fluorescence quencher). PCR amplifications were carried out as described (Mambo et al., 2003) using 1 ng DNA to amplify both the mitochondrial regions and the β-actin, a nuclear control gene. All reactions were performed in triplicate for each gene and standard curves were obtained by using DNA from untreated cell lines. mtDNA/nuclear DNA ratios were calculated by dividing the mtDNA signal for each gene by the corresponding β-actin signal and expressing the ratio as a percentage of the untreated control.
Cell viability assay
Cell lines H1975, H1650, H1915 and H226 were grown in 35 mm to ∼70% confluence. The cells were washed with Hank's balanced salt solution (HBSS) and treated with 0, 100, 200, 400, 600 and 800 μ M of H2O2 in serum-free medium and incubated at 37°C in 5% CO2 incubator for 2 h. After oxidative damage exposure, the cells were washed with HBSS and then allowed to recover in regular growth media for 16 h. The ATCC-MTT (3-(4,5-dimethylthiazol-2yl)-2,5 diphenyltetrazolium bromide) cell proliferation assay kit was used for the assessment of cellular viability as per the manufacturer's instructions. The percent survival was calculated by assigning the (A570 nm–A650 nm) of the untreated cells to 100% and measuring the treated cells relative to the untreated controls.
Sequencing of lung cancer cell lines
In total, 2 μg of total RNA from four lung cancer cell lines (H1915, H1975, H1650 and H226) were used to prepare cDNA using Superscript II RNase H-Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) as per the manufacturer's instructions. hOGG1 was PCR amplified using forward primer IndexTerm5′ ATGCCTGCCCGCGCGCTTCT 3′ and reverse primer IndexTerm5′ GCCTTCCGGCCCTTTGGAAC 3′. An Expand high fidelity PCR system (Roche, Boulder CO, USA) was used for all PCR reactions. All PCR products were sequenced by automated sequencing. Mutations were further confirmed by sequencing the desired cDNA using internally designed primer IndexTerm5′ GTTCCGTGGACTCCCACT 3′.
Arai T, Kelly VP, Minowa O, Noda T and Nishimura S . (2002). Carcinogenesis, 23, 2005–2010.
Audebert M, Charbonnier JB, Boiteux S and Radicella JP . (2002). DNA Repair (Amst.), 1, 497–505.
Audebert M, Chevillard S, Levalois C, Gyapay G, Vieillefond A, Klijanienko J, Vielh P, El Naggar AK, Oudard S, Boiteux S and Radicella JP . (2000). Cancer Res., 60, 4740–4744.
Ayala-Torres S, Chen Y, Svoboda T, Rosenblatt J and Van Houten B . (2000). Methods, 22, 135–147.
Blons H, Radicella JP, Laccourreye O, Brasnu D, Beaune P, Boiteux S and Laurent-Puig P . (1999). Mol. Carcinog., 26, 254–260.
Bohr VA, Stevnsner T and de Souza-Pinto NC . (2002). Gene, 286, 127–134.
Boiteux S and Radicella JP . (1999). Biochimie, 81, 59–67.
Cherng SH, Huang KH, Yang SC, Wu TC, Yang JL and Lee H . (2002). J. Toxicol. Environ. Health A, 65, 265–278.
Chevillard S, Radicella JP, Levalois C, Lebeau J, Poupon MF, Oudard S, Dutrillaux B and Boiteux S . (1998). Oncogene, 16, 3083–3086.
Croteau DL, ap Rhys CM, Hudson EK, Dianov GL, Hansford RG and Bohr VA . (1997). J. Biol. Chem., 272, 27338–27344.
Dhenaut A, Boiteux S and Radicella JP . (2000). Mutat. Res., 461, 109–118.
Dhenaut A, Hollenbach S, Eckert I, Epe B, Boiteux S and Radicella JP . (2001). Adv. Exp. Med. Biol., 500, 613–616.
Dherin C, Radicella JP, Dizdaroglu M and Boiteux S . (1999). Nucleic Acids Res., 27, 4001–4007.
Dizdaroglu M, Jaruga P and Rodriguez H . (2001). Nucleic Acids Res., 29, E12.
Dobson AW, Xu Y, Kelley MR, LeDoux SP and Wilson GL . (2000). J. Biol. Chem., 275, 37518–37523.
Gunther T, Schneider-Stock R, Rys J, Niezabitowski A and Roessner A . (1997). J. Cancer Res. Clin. Oncol., 123, 388–394.
Hashiguchi K, Stuart JA, de Souza-Pinto NC and Bohr VA . (2004). Nucleic Acids Res., 32, 5596–5608.
Hodges NJ and Chipman JK . (2002). Carcinogenesis, 23, 55–60.
Hollstein M, Shomer B, Greenblatt M, Soussi T, Hovig E, Montesano R and Harris CC . (1996). Nucleic Acids Res., 24, 141–146.
Hu YC and Ahrendt SA . (2005). Int. J. Cancer, 114, 387–393.
Hyun JW, Choi JY, Zeng HH, Lee YS, Kim HS, Yoon SH and Chung MH . (2000). Oncogene, 19, 4476–4479.
Jaruga P, Zastawny TH, Skokowski J, Dizdaroglu M and Olinski R . (1994). FEBS Lett., 341, 59–64.
Klungland A, Rosewell I, Hollenbach S, Larsen E, Daly G, Epe B, Seeberg E, Lindahl T and Barnes DE . (1999). Proc. Natl. Acad. Sci. USA, 96, 13300–13305.
Kohno T, Shinmura K, Tosaka M, Tani M, Kim SR, Sugimura H, Nohmi T, Kasai H and Yokota J . (1998). Oncogene, 16, 3219–3225.
Kuraoka I, Endou M, Yamaguchi Y, Wada T, Handa H and Tanaka K . (2003). J. Biol. Chem., 278, 7294–7299.
Larsen E, Kwon K, Coin F, Egly JM and Klungland A . (2004). DNA Repair (Amst.), 3, 1457–1468.
Le Page F, Klungland A, Barnes DE, Sarasin A and Boiteux S . (2000a). Proc. Natl. Acad. Sci. USA, 97, 8397–8402.
Le Page F, Kwoh EE, Avrutskaya A, Gentil A, Leadon SA, Sarasin A and Cooper PK . (2000b). Cell, 101, 159–171.
Lightowlers RN, Chinnery PF, Turnbull DM and Howell N . (1997). Trends Genet., 13, 450–455.
Mambo E, Gao X, Cohen Y, Guo Z, Talalay P and Sidransky D . (2003). Proc. Natl. Acad. Sci. USA, 100, 1838–1843.
Mambo E, Nyaga SG, Bohr VA and Evans MK . (2002). Cancer Res., 62, 1349–1355.
Michaels ML and Miller JH . (1992). J. Bacteriol., 174, 6321–6325.
Michaels ML, Pham L, Cruz C and Miller JH . (1991). Nucleic Acids Res., 19, 3629–3632.
Miller SA, Dykes DD and Polesky HF . (1988). Nucleic Acids Res., 16, 1215.
Monden Y, Arai T, Asano M, Ohtsuka E, Aburatani H and Nishimura S . (1999). Biochem. Biophys. Res. Commun., 258, 605–610.
Nishioka K, Ohtsubo T, Oda H, Fujiwara T, Kang D, Sugimachi K and Nakabeppu Y . (1999). Mol. Biol. Cell, 10, 1637–1652.
Olinski R, Zastawny T, Budzbon J, Skokowski J, Zegarski W and Dizdaroglu M . (1992). FEBS Lett., 309, 193–198.
Paz-Elizur T, Krupsky M, Blumenstein S, Elinger D, Schechtman E and Livneh Z . (2003). J. Natl. Cancer Inst., 95, 1312–1319.
Richter C, Park JW and Ames BN . (1988). Proc. Natl. Acad. Sci. USA, 85, 6465–6467.
Sakumi K, Tominaga Y, Furuichi M, Xu P, Tsuzuki T, Sekiguchi M and Nakabeppu Y . (2003). Cancer Res., 63, 902–905.
Sava V, Mosquera D, Song S, Cardozo-Pelaez F and Sanchez-Ramos JR . (2004). Free Radic. Biol. Med., 36, 1144–1154.
Shinmura K and Yokota J . (2001). Antioxid. Redox Signal, 3, 597–609.
Tajiri T, Maki H and Sekiguchi M . (1995). Mutat. Res., 336, 257–267.
Takao M, Aburatani H, Kobayashi K and Yasui A . (1998). Nucleic Acids Res., 26, 2917–2922.
Tchou J, Kasai H, Shibutani S, Chung MH, Laval J, Grollman AP and Nishimura S . (1991). Proc. Natl. Acad. Sci. USA, 88, 4690–4694.
Thomas D, Scot AD, Barbey R, Padula M and Boiteux S . (1997). Mol. Gen. Genet., 254, 171–178.
Viswanathan A and Doetsch PW . (1998). J. Biol. Chem., 273, 21276–21281.
Vogel U, Nexo BA, Wallin H, Overvad K, Tjonneland A and Raaschou-Nielsen O . (2004). Biochem. Genet., 42, 453–460.
Wikman H, Risch A, Klimek F, Schmezer P, Spiegelhalder B, Dienemann H, Kayser K, Schulz V, Drings P and Bartsch H . (2000). Int. J. Cancer., 88, 932–937.
Xie Y, Yang H, Cunanan C, Okamoto K, Shibata D, Pan J, Barnes DE, Lindahl T, McIlhatton M, Fishel R and Miller JH . (2004). Cancer Res., 64, 3096–3102.
Yakes FM and Van Houten B . (1997). Proc. Natl. Acad. Sci. USA, 94, 514–519.
This work was supported by National Institutes of Health Grant 5 U01 CA084986-04 titled Early Detection Research Network-Integrated Development of Novel Molecular Markers, and NIH Grant 5P50CA096784-03 entitled spore in lung cancer. Certain commercial equipment or materials are identified in this paper in order to specify adequately the experimental procedure. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified are necessarily the best available for the purpose.
About this article
Cite this article
Mambo, E., Chatterjee, A., de Souza-Pinto, N. et al. Oxidized guanine lesions and hOgg1 activity in lung cancer. Oncogene 24, 4496–4508 (2005). https://doi.org/10.1038/sj.onc.1208669
- lung cancer
- DNA repair