A mutant eGFP gene, containing a stop codon in the region encoding the chromophore domain of the protein, was integrated in DLD-1 cells and a clonal isolate consisting of 2–3 copies of the mutated gene was isolated and used as the target for oligonucleotide-directed gene repair [10,18-22,25,26]. A single-strand oligonucleotide (ssODN) consisting of 47 bases with three phosphorothioate linkages on each end, designed to hybridize to the nontranscribed strand of the eGFP gene, was used to reverse the mutation (TAG → TAC),. As previously reported, correction resulting in the restoration of green fluorescence expression can be assessed by flow cytometry analyses at various times following ssODN introduction [18].
A variety of reagents have been tested with the aim of elevating the frequency of gene repair in mammalian cells. Among these are drugs that induce double-stranded DNA breaks (ds breaks) and chemicals that form reversible adducts within the helix [27]. In most cases, DNA damage results in a slowing of the cell cycle and an expanded S phase and so, reagents that simply reduce the rate of DNA replication without causing double strand breaks have also been found to improve the efficiency of gene repair [10,19]. These agents are able to increase correction efficiency by expanding the window of time within which the ssODN gains access to the target site [10,19]. Thus, our testable prediction is that a reagent such as thymidine, which increases the cellular levels of dTTP via a depletion of dCTP, resulting in a reduction in the rate of DNA replication [24], will increase gene repair activity.
Pre-incubation with thymidine elevates gene repair by increasing the number of cellular targets amenable for correction
Thymidine was added at various concentrations to the DLD-1 mutant eGFP cell cultures 24 hours prior to the introduction of the ssODN (EGFP3S/47NT) and correction levels were determined 48 hours after electroporation by flow cytometry (Figure 1). The results show a maximal response at concentrations of 1 and 8 mM thymidine, with a significant increase of about 3 fold from the non-treated sample. The decrease in correction efficiency as seen at higher concentrations of thymidine (10 and 12 mM) is most likely due to the enhanced thymidine cytotoxicity reported previously in DLD-1 cells due to their mutation in Msh6 [28]. Although these results differ from the observation of Wu et al. [14], who stated that the incubation of thymidine prior to the introduction of the oligonucleotide actually lowered gene repair in HeLa cells containing >50 copies of the target gene, our results are consistent with other reports that a thymidine pre-treatment is able to elevate gene repair several fold over basal levels.
It is widely known that the addition of thymidine to a mammalian cell culture leads to a synchronization of cells in the G1 phase of the cell cycle or at the G1/S border [29,30]. When cells are released from the thymidine block, a large fraction of the population proceeds into S phase, resulting in an increase of ideal cellular targets for ssODN site association and thus, gene repair activity appears maximized [19,26]. As shown in Figure 2A, the addition of 1 mM thymidine restructures the cell cycle profile such that the majority of cells are primarily stalled in G1, minimally in S phase, and with only a small fraction in G2 (Figure 2A: ii). By 8 hours after the release of the thymidine block, the cells begin to transition into S phase (Figure 2B: ii), while the non-treated samples remain predominantly at the same cell cycle proportions from the time of oligo addition to 8 hours following electroporation (Figure 2A: i and Figure 2B: i). At 8 hours, a majority of the cells released from the thymidine treatment contain actively replicating chromosomes (Figure 2B: iii), as assayed by BrdU incorporation. Under these conditions, the number of actively replicating cells is almost twice as high as the level seen in cells that have not been pretreated with thymidine (Figure 2B: iv vs. 2Biii), an observation that could reflect the increased number of cells in S phase. Thus, the addition of thymidine and its subsequent removal stimulates gene repair activity by enabling more cells to be in S phase when the ssODN initializes the gene repair reaction.
Thymidine is able to stimulate and maintain substantial levels of gene repair
After observing the increase in gene repair levels at 48 hours following a 24 hour thymidine pre-treatment, we wanted to examine if this increase remains stable over time, beyond the 48 hour incubation. For this study, we repeated the experiment outlined in Figure 1 using a single concentration of thymidine (1 mM) during the pre-incubation phase and measured correction levels at 48 and 144 hours after ssODN electroporation. Although the initial stimulation is seen at 48 hours post electroporation, Figure 3 illustrates an important phenomenon in which the absolute level of correction decreases with time, corroborating the initial finding reported by Olsen et al. [12,16]. As shown, a statistically significant decrease in repair levels is seen between the 48 and 144 hour time points (approximately 3-fold) for both the pretreated (Figure 3: bar d vs. c) and untreated (Figure 3: bar b vs. a) samples. It has been suggested previously by our group that this lack of maintenance of the correction efficiency is due to the induction of the DNA damage response pathway and its subsequent activation of the checkpoint proteins Chk1 and Chk2. Hence, a "dilution" of the corrected cells is seen as the non-corrected cell population proliferates at normal rates while the corrected cells appear stalled [22].
Based on this observation, we wanted to examine the idea of controlling replication of the entire cell population as a means of preventing the decrease of correction levels over time. Our reasoning behind this was to transiently inhibit the replication and division of non-corrected cells until the corrected cells were able to resume normal replication. Accordingly, we wanted to utilize thymidine and its ability to stall cell replication as a tool to slow down the entire cell culture, in hopes of maintaining the high initial levels of correction for extended periods of time. To address this aim, we treated the cells with 1 mM thymidine for 24 hours prior to the introduction of the 47-mer EGFP3S/47NT and then divided the samples into three distinct groups. The first group of cells was grown in the absence of thymidine for either 48 or 144 hours (Figure 4: a and b) while the second group was grown for 48 or 144 hours in the presence of 2 mM thymidine (Figure 4: c and d). The last group was grown in thymidine for only 48 or 72 hours after electroporation (Figure 4: e and f, respectively), at which point the thymidine was washed out (indicated as "w/o") and the incubation continued for a total of 144 hours in complete medium. The combination of a 1 mM pre- and a 2 mM post-treatment was optimized to produce first an initial increase in repair frequency followed by a sustained level of this with minimal cytotoxicity, data not shown. When no thymidine was present in the post-electroporation cultures, the correction efficiency decreased significantly between the 48 and 144 hour time points, coinciding with the data in Figure 3. In contrast, when 2 mM thymidine was added to the cultures following electroporation, the correction efficiency was largely maintained between 48 and 144 hours at levels significantly above those for the non-treated samples at corresponding time points. In a similar fashion, by removing thymidine after only a 48 or 72 hour incubation, correction efficiencies still remained significantly higher at 144 hours than with out any thymidine post-treatment. Taken together, these data suggest that the presence of thymidine in the post-electroporation reaction results in a maintained level of gene repair for an extended period of time.
Thymidine addition induces cell senescence in DLD-1 cells under certain reaction conditions
After extended culturing in the presence of thymidine, we noticed morphological changes within our cultures that resembled a senescent phenotype, including flattened cell morphology with elongated cellular processes and enlarged nuclei [31]. To confirm these cellular changes, cultures were stained for senescence associated β-galactosidase (SA β-gal) expression [32,33]. In Figure 5A, representative fields of cells treated with thymidine under two different conditions are shown at 48, 72 and 144 hours after introduction of the ssODN. No evidence of senescence is observed in the 48 or 72 hour cultures under either condition (1-0 mM or 1–2 mM). However, cells pretreated with 1 mM thymidine and incubated in 2 mM thymidine after electroporation (1–2 mM) develop a senescent phenotype within 144 hours, as evident by the presence of blue stain in approximately 50% of these cells. In Figure 5B (i, ii, and iii), multiple images obtained at the 144 hours under the 1–2 mM thymidine condition are presented, further highlighting the positive SA β-gal expression. The left-hand panels, showing cell morphology, were photographed under phase contrast while the panels on the right, aimed to enhance the blue color, were photographed with bright field illumination. Here, the evidence for certain reaction conditions inducing cellular senescence is more compelling and this transition to a senescent phenotype could further explain the maintenance of gene correction over 144 hours with the 1–2 mM thymidine treatment combination. In contrast, Figure 6 presents cells incubated for 144 hours with thymidine having been washed out at 48 or 72 hours, post-electroporation. Here, we observe a minimal number of senescent cells at the 48 hour wash-out (less than 10% of the cells) but a significant number is observed when the thymidine is washed out at 72 hours (about 38% of the culture), suggesting that the transition to wide-spread senescence most likely occurs shortly after this point. Furthermore, it is important to note that corrected cells are neither apoptotic nor senescent (Ferrara et al., submitted), suggesting that the transition to the apoptotic state is simply as a result of the thymidine incubation. In addition, prolonged treatments with thymidine did not show evidence of cell death, as determined by a MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) cell viability assay in which the level of production of the purple formazan remained unchanged during the incubation (data not shown).
Prolonged incubation with thymidine leads to a slow-down in DNA replication and cell division
As shown in Figure 2A, the exogenous addition of thymidine to a DLD-1 cell culture results in an accumulation of cells in G1 or at the G1/S border. In effect, thymidine reduces the overall rate of DNA replication in a population of cells by synchronizing them into G1 and preventing their entry into S phase. Thus, the presence of thymidine in a gene repair reaction could simply be maintaining the relative percentages of corrected and non-corrected cells between 48 and 144 hours by inhibiting the replication of the entire cell population and there by, preventing the dilution of the corrected cells. To examine this possibility, we measured the capacity of cells to incorporate BrdU, which would indicate the percent of cells undergoing active replication at 48, 72, and 144 hours. When the population was pretreated with thymidine for 24 hours followed by a removal of the thymidine at the time of electroporation (Figure 7: 1-0 mM), replication activity remained fairly constant overtime. In contrast, when the pretreatment was followed by sustained incubation in thymidine following electroporation (Figure 7: 1–2 mM), replication activity dropped dramatically and little incorporation of BrdU was evident. Thus, it seems likely that the maintained correction efficiency seen between 48 and 144 hours under the latter condition (1–2 mM) is due to an overall inhibition of DNA replication and in part by the entry into cellular senescence, as seen in Figure 6. However, when thymidine was washed out at 48 or 72 hours (Figure 7: 1–2 mM w/o 48 hrs and 1–2 mM w/o 72 hrs, respectively), and cells allowed to resume growth in complete medium, DNA replication resumed, reaching approximately half the normal level of activity by 144 hours of incubation. More importantly, this washout protocol preserved high levels of correction efficiency between 48 and 144 hours (see Figure 4). Therefore, the presence of thymidine during the early times of the gene repair reaction appears to help maintain correction levels without irreversibly disabling the capacity of the cell to resume DNA replication.
In order to confirm these results and to evaluate the potential residual effects if thymidine on the cell cycle following the wash-out protocol, we utilized an assay to determine division rates of cells. PKH26, a fluorescent cell label that binds irreversibly within the lipid region of cell membranes, was used to determine rate and extent of cell proliferation following the removal of the thymidine post electroporation. As cells divide following intercalation of the dye, each resulting daughter cell receives half of the fluorescent dye during subsequent rounds of cell division [34,35]. Thus, by using flow cytometric analysis to measure the change in fluorescence intensity of PKH26, we are able to estimate the number of divisions a cell population has undergone within a given period. At time zero (T0), or at the time of electroporation following a 1 mM thymidine pre-treatment, cells were labeled with PKH26 and the starting mean fluorescence intensity (m.f.i.) was found to be 254.8 (Figure 8: i). At 48 hours following staining, cells that received no thymidine in the post treatment (Figure 8, ii) displayed a shift in their mean fluorescence intensity to 62.6, corresponding to two cellular divisions, and again at the 144 hour point (Figure 8: iii), indicating an additional three cell divisions (total of 5 within the 144 hours). In contrast, with thymidine in the post treatment for 48 hours (Figure 8: iv), there was an insignificant change in the mean fluorescent intensity, indicating that no cell division had occurred. Following the removal of thymidine at 48 hours, the cells underwent 2.5 cell divisions by 144 hours (Figure 8: v), as determined by the change in the mean fluorescence intensity from 48 hours. Thus, it seems as if washing out thymidine at 48 hours can enable cell division to resume at a rate nearly that of the 1-0 mM counterpart. Combined with the BrdU data, a case can be made that the maintenance of correction efficiency is accompanied by DNA replication and normal cell division processes following the removal of thymidine.
Answers to all your Biology Questions
