Radiation Oncology Investigations 6:1–9 (1998)
DNA Damaging Agents Improve Stable Gene Transfer Efficiency in Mammalian Cells Craig W. Stevens, M.D., Ph.D.,* George J. Cerniglia, Ph.D., Albert R. Giandomenico, and Cameron J. Koch, Ph.D. Department of Radiation Oncology, University of Pennsylvania, Philadelphia, Pennsylvania SUMMARY Gene therapy is an evolving discipline which today relies primarily on viral systems for gene transfer. The primary reason that plasmid vectors have not been widely used for gene therapy trials is their relatively low rate of stable gene transfer. We show here that both ionizing irradiation and hydrogen peroxide can each increase the gene transfer efficiency of plasmids. Hydrogen peroxide improves gene transfer in a linear dose-dependent manner. At equitoxic doses, hydrogen peroxide improves gene transfer by 20-fold over untreated cells and approximately 5 times above that seen for radiation, and this improvement correlates with both the total amount of DNA damage induced and the amount of residual damage after 4 hr of repair. These data suggest that DNA damaging agents may be useful to improve human gene therapy. Radiat. Oncol. Invest. 6:1–9, 1998. © 1998 Wiley-Liss, Inc.
Key words: radiation; gene therapy; hydrogen peroxide; recombination
INTRODUCTION Gene therapy is an evolving discipline with the potential to correct inherited genetic diseases and possibly cure acquired genetic diseases such as cancer. While the genetic basis for many such diseases is defined, the delivery of the appropriate gene to the appropriate site has proved challenging. Gene therapy has been attempted with both DNA and RNA viruses as well as plasmid vectors. The primary difficulty with the use of plasmid-based gene transfer systems has been the relatively low gene transduction efficiency when compared with viral vectors. Plasmid-based gene transfer systems, however, have a number of inherent advantages over viral systems. Plasmids do not induce an immune response, thereby allowing repeat dosing without decreased gene transduction, and they can also carry much larger therapeutic genes than can
viruses [1]. If the stable transfection frequency of plasmids can be increased sufficiently, their advantages in clinical gene therapy may be exploited. The primary effect of ionizing radiation on living cells is to damage macromolecules either directly or indirectly through radicals and other reactive intermediates. One of the most important target molecules is DNA, where single-strand breaks (ssbs), double-strand breaks (dsbs), and other more complicated damage can occur, with approximately 20–40 ssb per dsb. Cells have developed sophisticated methods of repairing radiation-induced DNA damage [2–4] and at least a portion of the DNA dsbs are repaired by recombination [5]. DNA recombination systems have been studied in the past, and evidence has emerged suggesting that ionizing radiation [6,7], ultraviolet (UV) irradiation [8], and Cisplatin [9] can improve stable
Abbreviations: H2O2, hydrogen peroxide; ssb, single-strand DNA break; dsb, double-strand DNA break; TER, transfection enhancement ratio; SF10, dose at which 10% of treated cells survive. Contract grant sponsor: Radiological Society of North America Research and Education Fund Scholars Program Award; Contract grant sponsor: University of Pennsylvania, Cancer Center Pilot Project Award; Contract grant sponsor: NIH; Contract grant number: CA-49498. *Correspondence to: Craig W. Stevens, M.D., Ph.D., Department of Radiation Oncology, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104. E-mail:
[email protected] Received 22 March 1996; Revised 8 July 1997; Accepted 23 July 1997 © 1998 Wiley-Liss, Inc.
2
Stevens et al.: DNA Damage Improves Gene Transfer
gene transfer by inducing recombination. DNA damage repair has been shown to be inducible by irradiation in prokaryotes [10] and lower eukaryotes [11], but the induction of DNA damage repair machinery has been difficult to detect in higher eukaryotes [12]. While the isolation of ionizing radiation-induced DNA repair enzymes in higher eukaryotes has proved difficult, radiation has been shown to induce a wide variety of specific gene transcripts and signal transduction cascades [13]. Several proteins have been shown to be involved in both DNA damage repair and recombination. Extensive data in human cells suggest that both the Ku proteins [14] and the p350 DNA-PK [15] are involved in immunoglobulin gene rearrangement and repair of DNA dsbs induced by ionizing radiation. However, the finding that UV damage can also improve stable gene transfer efficiency suggests that induction of recombination may be quite complex [16]. Non-homologous recombination is an inefficient process, with rates of recombination on the order 10−3 to 10−6 [17], but certain modifications to transfected DNA [18] have been reported to improve recombination efficiency by ∼2-fold. Homologous recombination is even less efficient, with frequencies on the order of 10−7 to 10−8. It has been shown recently that modest doses of ionizing radiation can improve the frequencies of interchromosomal homologous recombination by severalfold [19]. In some viral systems, sequence-specific gene integration has been demonstrated [20,21], although non-homologous recombination is the preferred method of plasmid integration in mammalian cells [22]. Also, fragile sites within chromosomes, known because of hot spots for chromosomal anomalies and gene amplification, have been shown to be associated with exogenous gene integration [reviewed in 23]. Hydrogen peroxide is a DNA damaging compound that induces primarily ssbs, estimated at 400,000 ssb per lethal event [24]. With so many ssbs per lethal event, DNA ssbs are not thought to be the primary cause of cell death for this agent [25]. Thus, hydrogen peroxide is a compound produced by ionizing radiation in small amounts, which produces a large number of DNA ssbs, but is relatively non-toxic. We hypothesized that either the physical changes (e.g., DNA ssb and dsb), one or more of the DNA repair enzymes, or both might enhance the ability of liposome-encapsulated plasmid DNA to integrate into mammalian chromosomes. We show here that the non-homologous gene
integration frequency can be enhanced by treatment of primary, immortalized and fully transformed human and rodent cells with modest doses of radiation. We also show that hydrogen peroxide is even more effective than ionizing radiation at improving gene integration frequencies, and that this improvement correlates with both the total amount of DNA damage induced and the amount of residual damage after 4 hr of repair. These observations may lead to a role for the use of DNA damaging agents to improve human gene therapy with plasmidbased gene transfer systems. MATERIALS AND METHODS Cell Culture A549 and NIH/3T3 cells were obtained from the American Type Culture Collection (ATCC; Rockville, MD) and were routinely passaged twice weekly. A549 and NIH/3T3 were grown as monolayers in Dulbecco’s modified Eagle’s medium (DMEM)/Ham’s F-12 media supplemented with 10% fetal calf serum (FCS) and DMEM supplemented with 10% calf serum, respectively. Human fibroblasts were isolated from foreskins as follows: After circumcision, foreskins were rinsed extensively in Dulbecco’s-phosphate buffered saline and minced. Approximately 20 ml of the enzyme cocktail (0.025% protease type IX, 0.025% collagenase type IA, and 0.025% DNase dissolved in D-PBS) was then added and the mixture incubated at 37°C for 30 min. Any undissociated pieces of tissue were then allowed to settle, and the supernatant was poured off and spun at approximately 500 rpm for 5 min. This cell pellet was then rinsed 3 times with D-PBS before plating into media. The pellet was then seeded into DMEM/Ham’s F-12 supplemented with 10% FCS. The cell strain 39F used in these studies has a population doubling time of approximately 18 hr and will undergo approximately 60 population doublings (15–20 passages) before senescing. Primary cells were used for these experiments within the first 15 population doublings. Cell Irradiation Cells were irradiated as described previously [6]. Briefly, attached cells were irradiated by a 137Cs source (Mark I Irradiation, J.L. Sephard & Association, Glendale, CA) at approximately 1.7 Gy/ min. Lipofection Assay The plasmid used in these studies was pSV2neo. This pBR322-derived vector contains a constitu-
Stevens et al.: DNA Damage Improves Gene Transfer
tively expressed neomycin phosphotransferase gene which confers resistance to the antibiotic G418 in eukaryotic cells. The lipofection protocol was first optimized with respect to the amount of plasmid per 5 × 105 cells that yielded the highest number of drug resistant colonies. This resulted in 5, 0.5, and 50 mg of DNA per 5 × 105 A549, NIH/ 3T3, and 39F cells, respectively (data not shown). Cells were seeded at a density of 5 × 105 cells/25 cm2 of tissue culture surface in 3 ml of the appropriate complete medium. Cells were incubated at 37°C in a 5% CO2 incubator for 24 hr. Cells were then transfected with a quantity of DNA previously determined to optimize the transfection of each cell type, mixed with 15 mg Lipofectamine (Sigma, St. Louis, MO) reagent according to the manufacturer’s recommendations, and applied to monolayers. Sixteen to 17 hr later, cells were treated with radiation. Four hours later, cells were seeded into selective media containing the minimum concentration of Geneticin (G418 sulfate; Gibco, Grand Island, NY) shown to kill 100% of untransfected cells. For expected surviving fractions between 0.1 and 0.01, treatment was performed on 1.5 × 106 cells in 75 cm2 flasks. For lower expected surviving fractions, treatments were performed on 3 × 106 cells in 150 cm2 flasks. Due to the low baseline transfection efficiency of 39F cells, all 39F transfection experiments were performed in 150 cm2 flasks containing 3 × 106 cells. The total volume of complete media was 9 and 18 ml for 75 and 150 cm2 flasks, respectively. Flasks for cytotoxicity determination were seeded at the same time as the treatment flasks. All experiments were performed in triplicate. Radiation survival is not affected by the presence of Lipofectamine, plasmid DNA, or both (data not shown). For assessing the effects of hydrogen peroxide, the above treatment scheme was modified as follows: After 16–17 hr of exposure to plasmid, hydrogen peroxide was added to the media to bring the final concentration to 375 mM. After 1 hr, media were removed and replaced with fresh media. Cells were seeded into selective media 4 hr later. Hydrogen peroxide cytotoxicity was performed with cells at density identical to that in transfected flasks to avoid differences in H2O2 metabolism. Immediately after a 1 hr exposure to H2O2, cells were trypsinized and reseeded at appropriate density for clonogenic survival determination. Colonies were stained with 0.5% crystal violet in 70% methanol and counted 14 days after seeding.
3
Southern Blot Analysis for Plasmid DNA Cellular DNA was extracted as described [26]. DNA was cut with restriction endonucleases according to the manufacturer’s recommendation. Samples containing 15 mg cellular DNA were run on 0.8% agarose gels, blotted, and hybridized with the indicated probes as previously described [6]. DNA Strand Break Determination DNA damage and its repair were assessed by the alkaline elution technique as described by Koch and Giandomenico [27]. Briefly, log-phase cells were incubated with 0.05 mCi/ml [14C]thymidine (∼50 mCi/mmol) for approximately 24 hr on the day before DNA damage determination. Cells were removed from these flasks and seeded, on the morning of the experiment, into 60 mm dishes in 3 ml of complete medium without radiolabel (5 × 105 cells/ dish). Cells were treated with either radiation of hydrogen peroxide at 37°C, allowed to repair for defined time while still at 37°C, then placed at 0°C to halt DNA damage repair until all samples had been loaded onto 1 mm polycarbonate filters. Filters were warmed and eluted as described previously [27]. The DNA elution rate was determined by scintillation counting. The present refinement of the ssb DNA assay provides accurate monitoring of damage at therapeutically relevant, low, radiation doses. The doses used in the present experiments were somewhat higher than the optimal range. In addition, there is much controversy about the mechanism of damage produced by peroxides at low vs. high temperatures. To avoid these problems, cells were treated at 37°C and allowed to repair for a minimum of 30 min at 37°C after hydrogen peroxide or radiation treatment before assaying for ssbs to allow detection of remaining damage in the linear range of the assay. RESULTS Transfection Efficiency in Unirradiated Cells The baseline lipofection efficiency of the three cell lines used in this study is shown in Table 1. The A549 human lung cancer cell line stably integrated DNA with an efficiency of approximately 4 × 10−4, NIH/3T3 immortalized non-tumorigenic mouse fibroblasts with an efficiency of approximately 1 × 10−4, and 39F primary human fibroblasts with an efficiency of approximately 1 × 10−5. Baseline transfection efficiency was not affected by linear-
4
Stevens et al.: DNA Damage Improves Gene Transfer
Table 1. Lipofection Efficiency Without Radiation No. of coloniesa Cell line
EcoR1 linearized
Supercoiled
375 128 12
416 113 9
A549 human lung carcinoma NIH/3T3 immortal mouse fibroblast 39F primary human fibroblast
a Number of G418 resistant colonies per 106 plated cells generated by lipofection with pSV2neo.
ization of pSV2neo with EcoRI in any of these three cell lines. Only linearized plasmids were used for the experiments described here, since we have shown that radiation improves the transfer of linearized plasmids much more effectively than supercoiled plasmids [6]. Determination of Radiation Sensitivity and SF10 As shown in Figure 1A, the radiation dose that results in 10% survival is 7, 5, and 3.8 Gy for A549, NIH/3T3, and 39F cells, respectively. The hydrogen peroxide concentration shown to result in 10% survival for A549 cells was 375 mM for a 60 min treatment (Fig. 1B). Both Irradiation and Hydrogen Peroxide Treatment Increase Transfection Efficiency The transfection efficiency of cells treated with SF10 doses of radiation and H2O2 was determined and compared with the transfection efficiency of untreated cells. We previously determined that such enhancement of long-term gene integration by ionizing radiation could only occur with the use of linearized, not supercoiled, plasmids [6]. Therefore, EcoRI linearized pSV2neo was used for all of the following studies. The ratio of the transfection efficiency of treated cells corrected for cytotoxicity to untreated cells was determined. This ratio, the transfection enhancement ratio (TER), is shown in Figure 2 for radiation of A549 cells, NIH/3T3 cells, and primary fibroblasts, and for hydrogen peroxide treatment of A549 cells. The TERs for these cell lines and treatment strategies are 4 ± 0.82, 5 ± 1.97, 6 ± 2.67, and 20 ± 1.73, respectively. Thus, irradiation produces very similar enhancement of stable gene transfer per cell killed for all three cell lines despite an approximately 2-fold variation in
Fig. 1. Radiation and hydrogen peroxide sensitivity. A: Radiation cytotoxicity curves for the following cell lines: (d) A549 human lung carcinoma cells, (j) NIH/3T3 immortalized non-tumorigenic mouse fibroblasts, and (m) 39F primary human fibroblasts. B: Hydrogen peroxide cytotoxicity for A549 cells. The dose at which the surviving fraction is 10% of untreated controls is 7.0, 5.0, and 3.8 Gy, and 375 mM for irradiated A549, NIH/3T3, 39F, and H202 treatment of A549 cells, respectively. Error bars were derived from experiments done in triplicate.
lethality and a 40-fold difference in baseline transfection efficiency. Hydrogen peroxide treatment increases the transfection efficiency of linearized pSV2neo by 20-fold over unirradiated controls, and is approximately 5-fold more efficient than irradiation alone at SF10 in A549 cells. The observed transfection enhancement is stable for at least 90 population doublings, 16 weeks of continuous culture (data not shown).
Stevens et al.: DNA Damage Improves Gene Transfer
Fig. 2. TER determination. The TER is defined as the ratio of the number of drug resistant colonies in treated groups divided by the surviving fraction due to treatment toxicity, to the number of drug resistant colonies in untreated controls. The TER for A549 cells treated with radiation such that 10% of treated cells survive and the TER for A549 cells treated with an equitoxic dose of hydrogen peroxide are shown.
5
Fig. 3. Integration of plasmid into cellular DNA. Approximately 40 population doublings after transfection, DNA was isolated and probed for the presence of plasmidspecific sequences. In intact cellular DNA, only a weak episomal plasmid band is detected at 5.6 kb in this intentionally overexposed blot. The expected 3,017 bp Pst I fragment of pSV2neo can be found in all transfected cell lines and is quite prominent, confirming that plasmid has integrated into the cellular genome.
peroxide improve gene transfer by increasing the number of cells which stably integrate plasmid into their DNA.
Both Ionizing Radiation and Hydrogen Peroxide Induce Plasmid Integration Into Host DNA
Increased Transfection Efficiency Is Correlated With the Induction of DNA Strand Breaks in A549 Cells
In order to confirm plasmid integration, DNA was extracted from transfected cells approximately 40 population doublings after transfection and hybridized with a vector-specific probe. If plasmid DNA remained episomal, a 5.8 kb band would be expected in intact cellular DNA. As can be seen in Figure 3, only a very weak signal is detected in this intentionally overexposed blot. However, if DNA is digested with Pst I prior to electrophoresis, the expected 3,017 bp band is quite prominent in both treated and untreated cells. Since loading is identical (15 mg/lane) for both intact and Pst I-digested DNA, the stronger signal in the Pst I-digested sample demonstrates that pSV2neo was integrated into the genome. No evidence is seen for concatamerization. Similar observations have been made with both NIH/3T3 cells and 39F primary human fibroblasts (data not shown). The plasmid copy number in treated cells is approximately equal to that in untreated cells, but the number of cells stably transfected is much greater in the treated groups. Thus, both ionizing radiation and hydrogen
In an attempt to understand the molecular events important for enhanced gene integration, we hypothesized that gene integration frequencies might correlate with DNA strand breaks. Using the technique of alkaline elution, the elution rate was determined for DNA from A549 cells treated with SF10 doses of either radiation or hydrogen peroxide. Cells were treated and placed at 37°C for various times to allow for repair of DNA damage. The actual elution curves are shown in Figure 4. As the time after treatment is increased, the amount of DNA eluting is reduced dramatically, however, some unrepaired strand breaks remain even after 4 hr of repair. The slopes of the elution curves are plotted vs. the time allowed for repair and are shown in Figure 5. Hydrogen peroxide treatment results in 4.2-fold more residual damage after 4 hr of repair than does treatment with an equitoxic dose of ionizing radiation. Also, the area under the repair curves, i.e., the amount of DNA damage remaining per unit time, is 6.5-fold greater for hydrogen peroxide than for irradiation. Thus, the difference in
6
Stevens et al.: DNA Damage Improves Gene Transfer
Fig. 5. Corrected elution rates. The slopes of the elution rate curves, corrected for elution of DNA from untreated cells, are plotted vs. time allowed for DNA damage repair at 37°C. Hydrogen peroxide treatment results in 4.2-fold more residual damage after 4 hr of repair than does treatment with an equitoxic dose of ionizing radiation. Also, the area under the repair curves is 6.5-fold greater for hydrogen peroxide than for irradiation. Both the residual damage and the residual damage per unit time are similar to the 5-fold better TER seen for hydrogen peroxide over irradiation.
Fig. 4. Elution rates. The fraction of DNA eluted from the filter under alkaline conditions from A549 cells treated with SF10 doses of radiation (A) or hydrogen peroxide (B), with various times for repair of DNA damage. For the 0.5 hr repair elution curve, the initial slope was used to calculate the corrected elution rate for Figure 4.
enhanced gene integration correlates with both the amount of DNA damage and the amount of residual DNA damage after 4 hr of repair. DISCUSSION Clinical trials of gene therapy today are performed primarily with either adenoviral or retroviral vectors [28]. These vector systems have several major limitations. They are limited by the lack of stability of gene transfer in vivo [29] with the half-life of gene expression on the order of 2–4 weeks. These viruses are extremely immunogenic [30], which limits the ability to deliver repeat doses of the gene of interest. The production of viruses must be
closely monitored for the presence of infectious, replication competent particles [31]. The use of liposome-mediated gene transfer has several advantages over virally mediated gene transfer systems. In particular, they are non-immunogenic, can carry large (up to chromosome sized) DNA fragments, and their production does not generate infectious particles [32]. Also, if gene therapy is to be a component of cancer therapy, combined modality treatment with either radiation or chemotherapy would be likely, and the effects of these cytotoxic agents on gene transfer will need to be defined. The observation that DNA damaging agents can improve the frequency of stable gene transfer will be important for gene therapy. If sufficient enhancement of gene transfer can be demonstrated, then gene transfer using DNA damaging agents in conjunction with viruses may be more complete, and gene transfer with non-viral vectors may be possible. Interestingly, we have found radiationinducible recombination to be long lived, such that increased levels of recombination are detectable 2 weeks after a single radiation fraction (unpublished observations). The duration of hyper-recombination after H2O2 treatment is under study. Since it is expected that multiple courses of radiogene therapy
Stevens et al.: DNA Damage Improves Gene Transfer
would be required to achieve significant transduction in vivo, the finding of prolonged hyperrecombination will greatly increase its utility. We have shown that stable gene integration is enhanced by DNA damaging agents in primary and fully transformed human cells, and ionizing radiation, in doses which result in SF10, can improve gene transfection frequency by approximately 5fold. The effects seen with radiation are not due to hydrogen peroxide formation, since the amount of H2O2 generated per gray is on the order of 0.2 mM [33], a dose that does not improve gene integration (unpublished observations). The work summarized by Vos and Hanawalt [34] and Perez et al. [16] extends similar observations to yet other cell types. This broad spectrum of cell types affected suggests that there may be widespread applicability of these techniques. These and other [35] studies using electroporation of restriction endonucleases suggest that DNA dsbs can enhance recombination. Our report is the first to suggest that ssbs can be recombigenic as well. Treatment of cells with hydrogen peroxide at a concentration which induces SF10 toxicity improves gene transfection frequency by 20-fold, an improvement of 5-fold over equally toxic doses of radiation. Hydrogen peroxide is thought to induce DNA ssbs by a free radical-based mechanism. It also induces sister chromatid exchange in a dosedependent manner through a HOz -based mechanism [36]. Interestingly, hydrogen peroxide can cleave DNA in a sequence-specific way assisted by the Flp recombinase in yeast [37]. The mechanism by which this occurs is unknown, but it is hypothesized to occur via a non-radical-based cleavage. Any or all of these types of DNA damage might be responsible for the effects described here, but the presence of a recombinase in the vicinity of hydrogen peroxide-induced strand breaks suggests one possible mechanism for H2O2-mediated recombination. Hydrogen peroxide used at a concentration of 375 mM induces primarily ssbs, with an undetectable number of dsbs [38]. This, in association with our findings that hydrogen peroxide is more efficient than radiation at improving gene integration per cell killed, suggests that repair of ssb may be the mechanism by which gene integration is enhanced. Our data further suggest that recombination may be important in the repair of ssb, as has been shown for the repair of dsbs by Jessberger et al. [5]. One well-studied model for genetic recombination in human cells is that of immunoglobulin gene rearrangement. Several proteins have thus far been identified, most notably the 70 and 80 kDa
7
subunits of the Ku heterodimer, and the p350 gene product, which together make up DNA-dependent protein kinase (DNA-PK). The Ku proteins bind to DNA strand breaks and mediate recombination [39], and the Ku80 has been identified as the XRCC5 DNA repair gene in humans as well as the xrs-6 mutant in Chinese hamster cells [14]. Interestingly, Ku proteins also bind to ssbs, but do not activate DNA-PK [40]. Our observation that only recombination of linearized (not supercoiled) plasmid is enhanced with radiation or H2O2 suggests that Ku proteins could be at each end of recombining DNA. The p350 DNA-PK, which is deficient in scid mice and mediates recombination of both immunoglobulins and T-cell receptors, is also important for both recombination and radiation sensitivity [15]. The presence of functional Ku and DNAPK promotes efficient plasmid transfection [41], providing another link between gene transfer and DNA damage repair. Alkaline elution of cellular DNA after treatment with radiation and H2O2 after various repair times shows that both the residual damage after 4 hr of repair and the total amount of DNA damage are approximately 5–6-fold higher for equitoxic doses of hydrogen peroxide than for radiation. Thus, there is a correlation between the relative number of DNA strand breaks and the enhancement of stable gene transfection. These observations further support the idea that the total amount of DNA damage, not the number of lethal events (i.e., dsbs), could be an important modifier of gene transfer efficiency. We have also noted that recombination enhancement correlates with residual DNA strand breaks after 4 hr of repair. This is consistent with the data of Deng and Nickoloff [42] and Bhattacharyya et al. [43], who showed that unrepaired UV damage correlates with recombination efficiency. It was further shown [43] that at least two components of excision repair, the XP-A gene and an as yet undefined gene, are not involved in the recombination process. The molecular mechanism by which gene transfer is increased with either radiation or H2O2 is unclear, but dsbs may not mediate the effects seen here. Other possible mechanisms by which H2O2 could improve gene integration include 1) membrane damage which could allow increased plasmid uptake, 2) differential protein (DNA binding/repair protein?) induction [13,44], or 3) alterations in chromatin conformation [45]. These are currently under investigation in our laboratory. The ultimate utility of these observations will be in combining gene therapy with cytotoxic therapy for the treatment of cancer. We have shown
8
Stevens et al.: DNA Damage Improves Gene Transfer
that DNA damaging agents can improve gene transfer into those cells which survive the initial cytotoxic insult. Since a much higher percentage of surviving cells stably express a transgene, combined modality therapy has the potential to be more efficacious than either alone. In summary, the findings presented here suggest: 1) DNA damaging agents can improve gene integration frequency, 2) this improvement is not dependent on events which are necessarily cytotoxic, 3) the improvement in integration can be correlated with the total number of DNA strand breaks induced by the damaging agent, and 4) ssbs may be repaired by a recombination-dependent pathway. REFERENCES 1. Singhal A, Huang L: Direct gene transfer by liposomes. J Liposome Res 4:289–299, 1994. 2. Weichselbaum RR, Beckett MA, Hallahan DE, Kufe DW, Vokes EE: Molecular targets to overcome radioresistance. Seminars in Oncology 99:14–20, 1992. 3. Arrand JE, Michael BD: Recent advances in the study of ionizing radiation damage and repair. Int J Radiat Biol 61:717–720, 1992. 4. Rojas A, Denekamp J: Modifiers of radiosensitivity. Experimentia 45:41–52, 1989. 5. Jessberger R, Podust V, Hubscher U, Berg P: A mammalian protein complex that repairs double-strand breaks and deletions by recombination. J Biol Chem 268:15070–15079, 1993. 6. Stevens CW, Zeng M, Cerniglia GJ: Ionizing radiation greatly improves gene transfer efficiency in mammalian cells. Hum Gene Ther 7:1727–1734, 1996. 7. Zeng M, Cerniglia GJ, Eck SL, Stevens CW: Highefficiency stable gene transfer of adenovirus into mammalian cells using ionizing radiation. Hum Gene Ther 8:1025–1032, 1997. 8. Van Duin M, Westerveld A, Hoeijmakers JH: UV stimulation of DNA-mediated transformation of human cells. Mol Cell Biol 5:734–741, 1985. 9. Son K, Huang L: Exposure of human ovarian carcinoma to Cisplatin transiently sensitizes the tumor cells for liposome-mediated gene transfer. Proc Natl Acad Sci USA 91:12669–12672, 1994. 10. Ewing D, Franklin RB, Damsker KE: The X-ray induction of SOS repair activity in E. coli: Euoxic vs. anoxic DNA damage. Biochem Mol Biol Int 38:267–273, 1996. 11. Fortunato EA, Osman F, Subramani S: Analysis of spontaneous and double-strand break-induced recombination in rad mutants of S. pombe. Mutat Res 364:14– 60, 1996. 12. Caldecott K, Banks G, Jeggo P: DNA double-strand break repair pathways and cellular tolerance to inhibitors of topoisomerase II. Cancer Res 50:5778–5783, 1990. 13. Boothman DA, Meyers M, Fukunaga N, Lee SW: Iso-
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
lation of X-ray-inducible transcripts from radioresistant human melanoma cell lines. Proc Natl Acad Sci USA 90:7200–7204, 1993. Taccioli GE, Gottlieb TM, Blunt T, Priestly A, Demengeot J, Mizuta R, Lehmann AR, Alt FW, Jackson SP, Jeggo PA: Ku80: Product of the XRCC5 gene and its role in DNA repair and V(D)J recombination. Science 265:1442–1445, 1995. Lees-Miller SP, Godbout R, Chan DW, Weinfeld M, Day RS, Barron GM, Allanunis-Turner J: Absence of p350 subunit of DNA-activated protein kinase from a radiosensitive human cell line. Science 267:1183–1185, 1995. Perez CF, Botchan MR, Tobias CA: DNA-mediated gene transfer is efficiently enhanced by ionizing radiation and ultraviolet irradiation of rodent cells in vitro. Radiat Res 104:200–213, 1985. Fell HP, Yarnold S, Hellstrom I, Hellstrom KE, Folger KR: Homologous recombination in hybridoma cells: Heavy chain chimeric antibody produced by gene targeting. Proc Natl Acad Sci USA 86:8507–8511, 1989. Chang XB, Wilson JH: Modification of DNA ends can decrease end joining relative to homologous recombination in mammalian cells. Proc Natl Acad Sci USA 84:4959–4963, 1987. Benjamin MB, Little JB: X-rays induce interallelic homologous recombination at the human thymidine kinase gene. Mol Cell Biol 12:2730–2738, 1992. Hacker D, Fluck MM: High-level recombination specific to polyomavirus genomes targeted to the integration-transformation pathway. Mol Cell Biol 9:995– 1004, 1989. Roth DB, Wilson JH: Nonhomologous recombination in mammalian cells: Role for short sequence homologies in the joining reaction. Mol Cell Biol 6:4295– 4304, 1986. Hasson JF, Mougneau E, Cuzin F, Yaniv M: Simian virus 40 illegitimate recombination occurs near short direct repeats. J Mol Biol 177:53–68, 1984. Popescu NC, DiPaolo JA: Preferential sites for viral integration on mammalian genome. Cancer Genet Cytogenet 42:157–171, 1989. Elkind MM, Redpath JL: Molecular and cell biology of radiation lethality. In Becker RR (ed): Cancer: A Comprehensive Treatise. Vol. 6. pp 51–70, 1977. Ward JF, Blakely WF, Joner EI: Mammalian cells are not killed by DNA single-strand breaks caused by hydroxyl radicals from hydrogen peroxide. Radiat Res 103:383–392, 1985. Stevens CW, Manoharan H, Fahl WE: Characterization of mutagen-activated cellular oncogenes that confer anchorage independence to human fibroblasts and tumorigenicity to NIH 3T3 cells: Sequence analysis of an enzymatically amplified mutant HRAS allele. Proc Natl Acad Sci USA 85:3875–3879, 1988. Koch CJ, Giandomenico AR: The alkaline elution technique for measuring DNA single strand breaks: Increased reliability and sensitivity. Ann Biochem 220: 58–65, 1994.
Stevens et al.: DNA Damage Improves Gene Transfer 28. Morgan RA, Anderson WF: Human gene therapy. Annu Rev Biochem 62:191–217, 1993. 29. Robbins PD, Tahara H, Mueller G, Hung G, Bahnson A, Zitvogel L, Galea-Lauri J, et al.: Retroviral vectors for use in human gene therapy for cancer, Gaucher disease, and arthritis. Ann NY Acad Sci 716:72–88, 1994. 30. Zabner J, Couture LA, Gregory RJ, Grahm SM, Smith AE, Welsh MJ: Adenovirus-mediated gene transfer transiently corrects the chloride transport defect in nasal epithelia of patients with cystic fibrosis. Cell 75: 207–216, 1993. 31. Kotani H, Newton PB, Zhang S, Chiang YL, Otto E, Weaver L, Blaese RM, Anderson WF, McGarrity GJ: Improved methods of retroviral vector transduction and production for gene therapy. Hum Gene Ther 5:19–28, 1994. 32. Farhood H, Gao X, Son K, Yang YY, Lazo JS, Huang L, Barsoum J, Bottega R, Epand RM: Cationic liposomes for direct gene transfer in therapy of cancer and other diseases. Ann NY Acad Sci 716:23–34, 1994. 33. Biaglow JE, Mitchell JB, Held K: The importance of peroxide and superoxide in the X-ray response. Int J Radiat Oncol Biol Phys 22:665–669, 1992. 34. Vos JMH, Hanawalt PC: Effect of DNA damage on stable transformation of mammalian cells with integrative and episomal plasmids. Mutat Res 220:205–220, 1989. 35. Yorifuji T, Mikawa H: Co-transfer of restriction endonucleases and plasmid DNA into mammalian cells by electroporation: Effects on stable transformation. Mutat Res 243:121–126, 1990. 36. Tachon P: Intracellular iron mediates the enhancing effect of histidine on the cellular killing and clastogenicity induced by H2O2. Mutat Res 4:343–348, 1990.
9
37. Kimball AS, Lee J, Jayaram M, Tullius TD: Sequencespecific cleavage of DNA via nucleophilic attack of hydrogen peroxide, assisted by Flp recombinase. Biochemistry 32:4698–4701, 1993. 38. Baker MA, He S: Elaboration of cellular DNA breaks by hydroperoxides. Free Radical Biol Med 11:563–572, 1991. 39. Rathmel WK, Chu G: Involvement of the Ku autoantigen in the cellular response to DNA double strand breaks. Proc Natl Acad Sci USA 91:7623–7627, 1994. 40. Turchi JJ, Henkels K: Human Ku autoantigen binds Cisplatin-damaged DNA but fails to stimulate human DNA-activated protein kinase. J Biol Chem 271: 13861–13867, 1996. 41. Jeggo PA, Smith-Ravin J: Decreased stable transfection frequencies of six X-ray-sensitive CHO strains, all members of the xrs complementation group. Mutat Res 218:75–86, 1989. 42. Deng WP, Nickoloff JA: Preferential repair of UV damage in highly transcribed DNA diminishes UVinduced intrachromosomal recombination in mammalian cells. Mol Cell Biol 14:391–399, 1994. 43. Bhattacharyya NP, Mayer VM, McCormick JJ: Effect of nucleotide excision repair in human cells on intrachromosomal homologous recombination induced by UV and 1-nitropyrene. Mol Cell Biol 10:3945–3951, 1990. 44. Ward JF: DNA damage as the cause of ionizing radiation-induced gene activation. Radiat Res 138:s85–s88, 1994. 45. Vaughan ATM, Milner AM, Gordon DG, Schwartz JL: Interaction between ionizing radiation and supercoiled DNA within human tumor cells. Cancer Res 51:3857– 3861, 1991.
