The assay system
A standard assay system based on the correction of an integrated mutant eGFP gene was employed to measure gene repair activity. A single clonal isolate from HCT 116 cells (which contain the wild-type complement of p53 genes) containing a single copy of the mutant eGFP gene (which is actively transcribed in these cells) was used for these experiments. A low-copy target gene is advantageous in a model system for oligonucleotide-directed gene repair, conferring only a single site for gene repair activity per cell. Specific oligonucleotides were designed to correct the TA G point mutation at amino acid position 67 which lies in the chromophore region of the eGFP gene and conversion to TA C enables expression of wild-type eGFP, allowing detection of positive conversion events by FACS analysis [11] (see Figure 1A). The standard vector, EGFP3S/47NT, is a 47-mer containing three phosphorothioates at each termini to protect against nuclease activity. This system is well established and has been used in multiple studies to validate gene repair activity [5,6,11,13,23].
Selenomethionine does not induce cell toxicity
We examined cellular toxicity and genotoxicity of selenium in the HCT 116 model system. In general, organic selenium compounds have been found to exhibit low levels of toxicity in most cells. In our hands, selenomethionine was found to be nontoxic in HCT 116 cells, as measured by an MTT reduction assay for viability with concentrations up to 1 mM (see Figure 1B). These results set the viability parameters and indicate that a 24 h pretreatment with 50–200 μM selenomethionine, the experimental conditions used traditionally to assess the modulation of gene repair activity, does not lead to significant cell death.
Selenomethionine does not induce double-stranded breaks in DNA
At the genomic level, no significant DNA breakage in HCT 116 cells is observed by the addition of selenomethionine as evidenced by the pulse field gel (PFGE) presented in Figure 1C. Genomic DNA isolated from selenomethionine-treated cells at concentrations of 100, 200, and 400 μM respectively does not exhibit greater fragmentation compared to an untreated population of cells. This is in sharp contrast to the positive control in which the cells were treated with methyl methanesulfonate (150 μM) [24]. In this case, significant fragmentation is observed, evidenced by the bright smear of DNA containing an extensive number of breaks running into the gel. The bright band near the top of the damaged DNA region represents genomic DNA containing only a few strand breaks. Thus, selenomethionine treatment does not result in extensive DNA damage and therefore does not induce a cellular DNA repair response similar to that seen when etoposide (VP16) or camptothecin is added to a cell culture [25].
Gene repair activity modulated by selenomethionine
Recently, we reported that inducing a low level of DNA damage in cells significantly elevates the level of gene repair activity [5,6]; the introduction of double strand breaks in the target cell led to a cell cycle arrest in response to the creation and/or presence of the DNA damage. During the arrested phase of the reaction, the oligonucleotide is able to more accurately and efficiently gain access to the target site on the DNA, presumably because the cells undergoing gene repair are contained within the population suspended in S phase [see 26 for review]. One response to DNA damage is the elevation of activities which have been shown previously to regulate the gene repair reaction. Thus, agents that increase repair activity are obvious candidates for stimulatory factors of correction, and particularly those that stimulate the repair pathways without causing DNA damage. Along these lines, selenomethionine has been shown to induce DNA repair processes by stabilizing the formation of repair complexes [27]. Thus, we tested for modulation of gene repair by pre-treating the cells with varying doses of selenomethionine. The cells were treated for twenty-four hours, followed first by a wash-out and electroporation of the oligonucleotide, EGFP3S/47NT. Gene repair was evaluated by FACS twenty-four hours later, and the data reveal that correction levels decrease as a function of dose of selenomethionine (50, 100, and 200 μM, respectively) (Figure 2A). These are the same concentrations (described above) that exhibit no significant cytotoxicity (see Figure 1B).
The effect of selenomethionine treatment on cell cycle progression
As mentioned above, some treatments that increase the efficiency of gene correction do so by inducing DNA damage and cell cycle arrest [26]. Other treatments induce S phase accumulation but without DNA damage and still support high levels of gene repair [13,14]. Since selenomethionine has not been reported to damage DNA directly and we observed no ds breaks (see Figure 1C), we examined the effect of selenomethionine pretreatment on cell cycle progression. As shown in Figure 2B, cells treated with selenomethionine for 24 hrs exhibit a different cell cycle profile compared to untreated cells. FACS analysis reveals that while the overall percentage of cells in S phase is the same for non-treated and selenomethionine-treated cells, selenomethionine-treated cells exhibit some accumulation in G2 with a compensatory reduction in the population of cells in G1 or at the G1/S border. This result is similar to a cell cycle block caused by oxidative damage, where the passage from G2 to mitosis is delayed [28]. Thus, selenomethionine does not arrest or delay cells in S phase, a condition that has been shown to stimulate gene repair activity. Interestingly, the effect on cell cycle is reversible, as the cell cycle profile of treated cells returns to normal 24 hours after the removal of selenomethionine from the culture (Figure 2B).
Selenomethionine does not lead to increased expression or activation of p53 but does increase Ref-1 levels
As shown in Figure 2A, selenium reduces gene repair activity. One may therefore predict that this reduction is transmitted through a Ref-1-mediated redox activation of wild-type p53 [18], which has been shown to suppress gene repair activity [13]. As such, selenomethionine treatment should not increase the overall expression of p53 in HCT116 cells, nor should it induce the activation of p53 through phosphorylation. We tested this prediction by western blot analysis of proteins isolated from HCT116 cells treated with selenomethionine (200 μM) for 24 hours (Figure 3). An immunoblot with the p53-DO antibody, which is specific for total wild-type and mutant p53, exhibited no distinguishable change in p53 expression in non-treated or selenomethionine-treated cells. Likewise, immunoblotting with an antibody specific to an activated form of p53 (phosphorylation of serine-15) [29,30], which occurs in response to DNA damage showed no detectable activated p53 in the non-treated and selenomethionine-treated cells.
Whereas p53 levels are unchanged in the presence of selenomethionine, the Ref-1 protein is present at higher levels in HCT116 after 24 hrs treatment with selenomethionine (200 μM), as determined by immunoblot specific to Ref-1 (Figure 3). This supports our hypothesis that selenomethionine may reduce gene repair activity through p53 modification that is mediated by the Ref-1 protein.
Effect of p53 overexpression (wild-type and mutant) in human cell lines
Wild-type or mutant p53 overexpression plasmids were introduced to cells in the presence of selenomethionine with simultaneous introduction of the targeting oligonucleotide. 5 μg of respective p53 overexpression plasmid and EGFP3S/47NT oligonucleotide were electroporated and evaluated by FACS after 24 hours. As displayed in Figure 4A, cells transfected with the pcDNA empty plasmid and selenomethionine exhibit a reduction in gene repair activity. This result is equivalent to suppression of gene repair activity found in cells transfected with the p72 plasmid, which overexpresses wild-type p53. In combination, selenomethionine treatment in the presence of overexpressed p53 exacerbates this effect further, resulting in a nearly undetectable level of correction. Thus, selenomethionine-dependent redox control of endogenous p53 appears to act synergistically with elevated levels of p53 in the suppression of gene repair.
To explore the role of p53 in this reaction, we utilized a separate cell line that has been used previously as an assay system for gene repair activity. This cell line, DLD-1 [11] contains the same mutated eGFP gene as the HCT116 line used above in low copy number. Using a cell line with a low copy number of integrated targets is important in studies related to gene repair. In addition, this line does not have a full complement of p53 [see 14]. Using the same reaction protocol as carried out for experiments in HCT116 cells, the integrated DLD-1 cell line was electroporated with EGFP3S/47NT and the appearance of corrected cells viewed 24 hours later. Some samples were pre-incubated with the indicated concentration of selenomethionine for 24 hours prior to electroporation. As shown in Figure 4B, selenomethionine does not induce an inhibition of gene repair; in fact a slight but reproducible stimulation is seen in a dose-dependent fashion. Thus, a cell line devoid of the full complement of wild-type p53 responds differently to treatment with selenomethionine than does a cell line containing wild-type p53.
Answers to all your Biology Questions
