|Home | About | Journals | Submit | Contact Us | Français|
Manganese superoxide dismutase plasmid liposomes (MnSOD-PL) confer organ-specific in vivo ionizing irradiation protection. To prepare for potential intravenous clinical trials of systemic MnSOD-PL for radioprotection in humans, plasmid and bacterial sequences were removed and a new minicircle construct was tested. Minicircle MnSOD was purified and then cotransfected into 32D cl 3 murine interleukin-3-dependent hematopoietic progenitor cells along with another plasmid carrying the neo gene. Cells were selected in G418 (50μg/ml) and cloned by limiting dilution. Biochemical analysis of minicircle MnSOD-transfected cells showed an MnSOD biochemical activity level of 5.8±0.5U/mg compared with 2.7±0.1U/mg for control 32D cl 3 cells (p=0.0039). 32D-mc-MnSOD cells were as radioresistant as full-length MnSOD-PL transgene-expressing 2C6 cells, relative to 32D cl 3 parent cells, with an increased shoulder on the radiation survival curve ( and , respectively, compared with 1.5±0.5 for 32D cl 3 cells; p = 0.007). C57BL/6NHsd mice received intraoral mc-MnSOD-PL, mc-DsRed-PL control, full-length MnSOD-PL, or blank-PL and then were irradiated 24hr later with 31 Gy to the esophagus. Mice receiving mc-MnSOD-PL showed increased survival compared with control mice or mice treated with mc-DsRed-PL (p=0.0003 and 0.039, respectively), and comparable to full-length MnSOD-PL. Intravenous, systemic administration of mc-MnSOD-PL protected mice from total body irradiation (9.75 Gy). Therefore, minicircle DNA containing the human MnSOD transgene confers undiminished radioprotection in vitro and in vivo.
Ionizing irradiation induces cellular, tissue, and organ toxicity by mechanisms that include induction of radical oxygen species (ROS) (Hahn et al., 2000; Weiss and Landauer, 2000; Mitchell and Krishna, 2002; Epperly et al., 2004; Spitz et al., 2004). Antioxidant defenses include the bioavailability of ROS-neutralizing enzymes, which modulate ionizing irradiation-induced damage (Spitz et al., 2004). In particular, three forms of superoxide dismutase, two extracellular copper/zinc metalloenzymes, and one mitochondria-targeted manganese metalloenzyme, dismutate superoxide to hydrogen peroxide, which is then further acted on by enzymes including catalase and glutathione peroxidase to form water (Epperly et al., 2004). One approach toward increasing cellular and tissue defenses against ionizing irradiation has been through use of a gene therapy technique to overexpress mitochondrial manganese superoxide dismutase before irradiation exposure or intermittently during fractionated irradiation (Epperly et al., 2001a, 2001b, 2003c; Greenberger et al., 2003; Greenberger and Epperly, 2004, 2007; Carpenter et al., 2005).
For therapeutic application of a gene therapy technique in radiation protection, transgene expression during the time of irradiation exposure is desirable, and should be followed by clearing of the transgene cDNA, as well as RNA and protein, from the system (Epperly et al., 2000, 2003c; Greenberger et al., 2003; Greenberger and Epperly, 2004, 2007). Plasmid liposomes have offered an attractive vehicle by which to achieve such transient overexpression, and this technique of antioxidant gene therapy has been shown to be effective in vitro and in vivo for organ-specific radioprotection of the oral cavity (Guo et al., 2003; Epperly et al., 2007a), esophagus (Epperly et al., 2001a, 2001b), lung (Carpenter et al., 2005), and bladder (Kanai et al., 2002).
Systemic, intravenous administration of plasmid liposomes has been reported to be associated with an inflammatory response that may be attributable to CpG sequences within the bacterial plasmid backbone of the construct (Sellens et al., 2005). A novel approach has been used to create minicircle plasmid constructs in a manner that eliminates the bacterial CpG sequences (Chen et al., 2005). In the present studies, we sought to determine whether minicircle plasmid liposomes containing the human manganese superoxide dismutase (MnSOD) transgene could be used for in vitro and in vivo radioprotection and whether this construct eliminated the inflammatory response through removal of bacterial CpG sequences while retaining therapeutic transfer and transient expression of MnSOD biochemical activity.
The minicircle MnSOD vector is based on p2ΦC31, the minicircle parent plasmid DNA vector, constructed by Z.-Y. Chen (Chen et al., 2005). The human MnSOD transgene was amplified by polymerase chain reaction (PCR) from plasmid pNGVL3-MnSOD. The 1482-bp human MnSOD gene expression cassette, including a human cytomegalovirus (CMV) promoter, human MnSOD gene, and a poly(A) tail, was inserted into the site between SpeI and XhoI excision points in the p2ΦC31 plasmid, resulting in plasmid p2ΦC31.MnSOD (Fig. 1). A control minicircle construct containing the DsRed (Discosoma sp. red fluorescent protein) transgene was prepared according to the same procedures.
The protocol for minicircle plasmid isolation as previously described by Chen and coworkers (2005) was used to produce minicircle MnSOD. Plasmid p2ΦC31.MnSOD or p2ΦC31.DsRed was transformed into Escherichia coli TOP 10 (Chen et al., 2005). A single colony was grown overnight in fresh Luria–Bertani (LB) broth at 37°C and then pelleted in a centrifuge (Avanti J-20 XP; Beckman Coulter, Fullerton, CA) at 3000rpm for 20min at 20°C. The pellet was resuspended at 4:1 (v/v) in fresh LB broth containing 1% l-arabinose. The bacteria were incubated at 32°C with shaking at 250rpm for 2hr, at which time an additional 1.5 volumes of fresh LB broth (pH 8.0, containing 1% l-arabinose) was added and the incubation was continued at the increased temperature of 37°C for 2hr. Minicircle DNA was prepared according to the instructions included with the plasmid purification kits from Qiagen (Valencia, CA). Agarose gel electrophoresis was used for minicircle DNA purification to obtain 100% pure minicircle DNA.
Minicircle MnSOD DNA (p2ΦC31.MnSOD) was sequenced with an ABI PRISM 3700 DNA sequencer (Applied Biosystems, Foster City, CA) to confirm that the minicircles contained the MnSOD expression cassette.
The 32D cl 3 mouse hematopoietic progenitor cell line, dependent for growth on interleukin (IL)-3, has been described previously (Epperly et al., 2002, 2003b). 32D cl 3 cells were passaged in fresh RMPI 1640 medium containing 10% fetal bovine serum (FBS), 1% l-glutamine, 1% penicillin–streptomycin, and 15% WEHI-3 conditioned medium as a source of IL-3. Twenty-four hours later, 1×107 32D cl 3 cells were washed twice in phosphate-buffered saline (PBS) and resuspended in 0.25ml of PBS. For each electroporation, 5μg (1μg/μl) of p2ΦC31.MnSOD or p2ΦC31.DsRed and 0.5μg of pSV2-neo were added to the 32D cl 3 cells and transferred to an electroporation cuvette (cat. no. 165-2088; Bio-Rad). A Bio-Rad Gene Pulser II set to 280 V and 950μF was used to electroporate the cells. The cells were then placed on ice for 20min and transferred to a flask containing prewarmed medium. Two days after electroporation, transfected cells were selected by addition of G418 (500mg/μl) according to published methods (Epperly et al., 2002). A clonal line was selected by limiting dilution and expanded as previously described (Epperly et al., 2002). Subclones expressing the MnSOD or DsRed transgenes were identified by RT-PCR, using primers specific for MnSOD or DsRed as previously described (Epperly et al., 2002). A subclone of 32D cl 3, over-expressing the human MnSOD transgene (2C6), has been reported previously (Epperly et al., 2002, 2003b). Both cell lines were grown in WEHI-3cell-conditioned medium as a source of IL-3, in a high-humidity incubator, in RMPI 1640 medium supplemented with 10% FBS as described (Epperly et al., 2002).
Cells were irradiated in plastic 10×100mm test tubes at 1×105 cells/ml in tissue culture at a rate of 0.8 Gy/min, with doses ranging from 0 to 8 Gy. The cells were removed from the tubes and plated in semisolid medium according to published methods. The cells were incubated in a 37°C high-humidity incubator with 5% CO2 and colonies consisting of more than 50 cells were scored on day 7. Triplicate cultures for each radiation dose were scored. Each experiment was done three times. Linear quadratic and software programs for measuring radiosensitivity and extrapolation number were used according to published methods (Epperly et al., 2002, 2003b).
The nitroblue tetrazolium (NBT) reduction assay for MnSOD biochemical activity was carried out according to published methods (Epperly et al., 2002).
C57BL/6NHsd female mice (20g, 6–8 weeks old; Harlan Sprague Dawley, Indianapolis, IN) were housed at five per cage according to institutional animal care and use (IACUC) protocols. Animals were irradiated with 31 Gy to the upper body, with shielding of the abdomen and head, using a linear accelerator (Varian, Palo Alto, CA) at 300 monitor units/min according to published methods (Epperly et al., 2003a). In other experiments total body irradiation was delivered to a dose of 9.75 Gy at a rate of 0.8 Gy/min, using a Mark IV cesium irradiator (J.L. Shepherd, Glendale, CA) according to published methods (Epperly et al., 2007b). All animal protocols were approved by the IACUC of the University of Pittsburgh (Pittsburgh, PA). Veterinary care was provided by the Division of Laboratory Animal Research of the University of Pittsburgh.
C57BL/6J female mice received intraoral administration of 100μl of water followed by full-length pNGVL3-MnSOD, blank pNGVL3 plasmid, mc-MnSOD, or control minicircle plasmid by injection of 100-μl volumes of liposomes containing 100μg of pNGVL3-MnSOD or blank pNGVL3 blank plasmid DNA, or 50μg of mc-MnSOD or mc-DsRed plasmid DNA according to published methods (Epperly et al., 2001a, 2001b). The plasmid DNA was mixed with 28μl of Lipofectin (Invitrogen, Carlsbad, CA) and allowed to complex at room temperature for 10min before administration. Mice were not anesthetized, and were allowed to swallow plasmid liposome preparations. Esophageal irradiation was carried out by anesthetizing the mice with Nembutal and irradiating the esophagus to 31 Gy, using a linear accelerator according to previously published procedures (Stickle et al., 1999). Mice were held and monitored for evidence of esophageal toxicity including weight loss, and dehydration. Animals were killed by euthanasia when weight loss was greater than 20%, according to IACUC protocols.
Mice received intravenous administration of 100μl of liposomes containing either 100μg of pNGVL3-MnSOD or pNGVL3 blank plasmid DNA or 50μg of mc-MnSOD or mc-Blank plasmid by tail vein injection, 24hr before irradiation. The DNA was complexed with Lipofectin as described above. A total body dose of 9.75 Gy (the LD50/30, i.e., the dose of radiation required to kill 50% of a test cohort within 30 days) was administered 24hr after gene therapy, using a Shepherd Mark IV cesium irradiator according to published methods (Epperly et al., 2007b). Mice were monitored for the development of hematologic distress and were killed when moribund, or when demonstrating greater than 20% weight loss, according to IACUC protocols.
Minicircle MnSOD or DsRed plasmids were constructed by inserting the MnSOD expression cassette containing the human MnSOD transgene, CMV promoter, and poly(A) tail into the p2ΦC31 plasmid (Fig. 1A). The minicircle MnSOD plasmid was isolated and the gene sequence was obtained, indicating that the minicircle contained 54 bp from the p2ΦC31 plasmid, the CMV promoter, the human MnSOD transgene, and the poly(A) tail (Fig. 1B). A control plasmid containing the CMV promoter, DsRed transgene, and poly(A) tail was also constructed and sequenced (Fig. 1B).
32D cl 3 cells were cotransfected with either minicircle MnSOD plasmid or minicircle DsRed plasmid and pSV2-neo plasmid by electroporation and selected for cells exhibiting Neo resistance. The cells were cloned by limiting dilution and RT-PCR was performed with primers specific for the MnSOD or DsRed transgene to identify clones expressing either the MnSOD or DsRed transgene. Isolation of the mini-circle MnSOD plasmid yielded 0.25 to 0.5mg of 1532-bp minicircle DNA from 1.0 liter of overnight bacterial growth. The DNA purity was about 70 to 80%, compared with the 95% purity reported by Chen and coworkers (2005).
MnSOD biochemical activity was determined in 32D cl 3, 2C6, and 32D-mc-SOD cells. Cells were lysed by repeated freeze–thaw cycles with protein concentrations measured. Using an assay in which the reaction between xanthine and xanthine oxide liberates superoxide, which then reduces NBT, resulting in an increase in color change (Epperly et al., 2002). The presence of SOD removes the superoxide, preventing the color change caused by the NBT reduction. Higher concentrations of SOD will result in increased inhibition of the NBT reduction. One unit of SOD activity will result in 50% inhibition of NBT reduction per milligram of protein. Both 32D-mc-MnSOD and 2C6 had increased MnSOD activity, demonstrated by the increased inhibition of NBT color production (Fig. 2A). 32D cl 3 cells had an MnSOD biochemical activity level of 2.7±0.1U/mg protein compared with 5.6±0.3 or 5.6±0.5U/mg protein for 2C6 and 32D mc-SOD cells, respectively (p=0.0008 and 0.0039, respectively) (Fig. 2B).
To demonstrate that 32D mc-MnSOD cells had increased radiation resistance, 32D cl 3, 2C6, and 32D mc-MnSOD cells were irradiated with doses ranging from 0 to 8 Gy, plated in methylcellulose, and incubated for 7 days at 37°C in a 5% CO2 incubator. Seven days later colonies consisting of more than 50 cells were counted and the data were analyzed in a linear quadratic model (Fig. 3). Cell lines 2C6 and 32D mc-MnSOD were more radioresistant as demonstrated by an increase in the shoulder on the survival curve, with and for 2C6 and 32D mc-MnSOD cells, respectively, compared with for 32D cl 3 cells (p=0.0070 and 0.0078, respectively). There was no significant change in D0 between the cell lines.
We had previously demonstrated that intraesophageal administration of MnSOD-PL protects the esophagus from irradiation damage (Stickle et al., 1999; Epperly et al., 2001). To determine whether mc-MnSOD was similarly radioprotective, 100μg of pNGVL3 MnSOD plasmid or blank pNGVL3 DNA was complexed with 28μl of lipofectant and 50μg of mc-MnSOD plasmid or DsRed plasmid was mixed with 28μl of lipofectant and incubated for 10min at room temperature. The plasmid–liposome complexes were each administered by plastic syringe to the top of the esophagus of nonanesthetized C57BL/6NHsd female mice (15 per group) and the mice were allowed to swallow the complexes as described (Stickle et al., 1999). The PL-treated mice as well as untreated control mice were then irradiated 24hr later by administration of 31 Gy to the esophagus and the mice were monitored for the development of esophagitis (Fig. 4). Mice swallowing either mc-MnSOD-PL or MnSOD-PL had improved survival after irradiation compared with the irradiated control mice (p=0.0003 and p<0.0001, respectively). Mice given pNGVL3-PL had some detectable increased survival compared with the control irradiated mice (p=0.0012), but there was further improved survival among those given MnSOD-PL. In contrast, there was no significant difference in survival between control irradiated mice and mice injected with control minicircle DsRed-PL. The mice given mc-Mn-SOD-PL had significantly increased survival compared with control minicircle Blank DsRed-PL-treated mice (p=0.0391).
To demonstrate that overexpression of MnSOD in a pNGVL3 or minicircle plasmid could be safely administered systemically, C57BL/6NHsd female mice (15 mice per group) were injected intravenously via the tail vein with MnSOD-PL, mc-MnSOD-PL, pNGVL3 plasmid, or mc-Blank-PL, and irradiated 24hr later along with control mice to 9.75 Gy whole body irradiation. The mice were monitored for signs of hematologic stress, at which time they were killed (Fig. 5). mc-MnSOD-PL as well as full-length MnSOD-PL provided significant protection from 9.75 Gy total body irradiation (p<0.0001 and p=0.0340, respectively).
The development of systemic protection agents against total body ionizing irradiation is a focus of great scientific interest (Stone et al., 2004). Normal tissue-specific radioprotection is highly desirable in the development of new chemoradiotherapy programs for clinical cancer patient management (Greenberger et al., 2003; Greenberger and Epperly, 2004). Furthermore, normal tissue-specific radioprotection in the total body-irradiated patient has been the subject of particular interest particularly since solid tumors have been demonstrated to have a different redox balance compared with normal tissues (Spitz et al., 2004). Organ-specific systemic radioprotector agents have included strategies to neutralize inflammatory cytokines using small molecule antioxidants, by targeting the apoptotic cellular mechanism of death by administration of agents that upregulate antiapoptotic proteins, or neutralize proapoptotic proteins (Greenberger and Epperly, 2007). Each of these strategies has provided some success. However, management of side effects, and incomplete systemic delivery, have been complicating factors.
One approach toward radiation protection has focused on the cellular and tissue redox balance mechanisms and their alteration by ionizing irradiation (Epperly et al., 2004; Greenberger et al., 2003). Antioxidant stores within cells include glutathione, and other free radical-scavenging antioxidants. Prominent in this cellular arsenal are the antioxidant enzymes including superoxide dismutases (Hahn et al., 2000; Mitchell and Krishna, 2002; Spitz et al., 2004). Two forms of copper/zinc metalloenzyme, one intracellular and another extracellular, are involved in systemic protection against agents that induce oxidative stress, including tissue inflammation induced by ionizing irradiation (Spitz et al., 2004). Manganese superoxide dismutase, the third form of SOD, targets the mitochondria by a 22-amino acid targeting sequence (Epperly et al., 2003b). Specific localization of SOD to the mitochondria appears necessary for radiation protection of cells in vitro and tissues in vivo because removal of the mitochondrial localization sequence produces MnSOD with cytoplasmic expression and little radioprotective capacity, whereas attaching this same mitochondrial localization signal to copper/zinc SOD results in a mitochondria-targeted and radioprotective copper/zinc metalloenzyme (Epperly et al., 2003b).
Administration of MnSOD protein to cells in culture or tissues in vivo has not shown significant radioprotective capacity, in part because of difficulty in achieving high enough levels of protein at the mitochondrial membrane, where critical antioxidant function is required (Epperly et al., 2001). In contrast, administration of transgene cDNA for MnSOD results in transport to the mitochondria at high efficiency of mitochondria-targeted protein and has been shown to confer radioprotection to cells in vitro and tissues in vivo (Stickle et al., 1999; Epperly et al., 2002, 2003b, 2003c; Carpenter et al., 2005).
For systemic radiation protection, intravenous administration of plasmid liposomes has been shown to induce inflammation (Sellens et al., 2005), and this has been attributed to CpG sequences in the bacterial backbone of the plasmid (Sellens et al., 2005). An important strategy toward eliminating the negative side effects of plasmid liposome gene therapy has been reported by demonstration of a technique by which to construct a minicircle plasmid devoid of bacterial CpG sequences (Chen et al., 2005).
In the present study, we used this published technique of construction of minicircle plasmid to produce minicircle Mn-SOD, demonstrating its effectiveness in transferring human MnSOD transgene-mediated biochemical activity to mice, hematopoietic progenitor cells in vitro, and the esophagus of mice in vivo. Furthermore, intravenous administration of minicircle MnSOD was associated with protection from the LD50/30 total body irradiation dose of 9.75 Gy.
Previous studies demonstrated some radioprotective capacity of empty full-length plasmid, PNGVL3 or PRK5 (Epperly et al., 1999). This protection was associated with an inflammatory response that could be detected in tissues as an increase in mRNA for tumor necrosis factor-α, IL-1, and transforming growth factor-β (Epperly et al., 2001b). Although a nonspecific inflammatory response may be radio-protective in some organ-specific settings, there is concern that systemic introduction of a gene therapy vector that itself is inflammatory might be deleterious in a setting of victims of an irradiation accident or willful irradiation terrorism event, wherein the ionizing irradiation itself and combined injury (heat, physical trauma) would be inflammatory. The present mc-MnSOD gene therapy vehicle should eliminate inflammatory side effects of the vector but maintain sufficient radioprotective capacity to allow its use by intravenous administration in larger mammalian species including humans.
Preparation of mc-MnSOD constructs was not a trivial process. The techniques for bacterial growth, centrifugation, purification, and extraction of minicircle plasmid were complex. Furthermore, previous studies demonstrated yield of minicircle as high as 96% of total recovered plasmid liposome (Chen et al., 2005), whereas in the present study yield rates as low as 68% were noted. These differences were not attributable to obvious changes in laboratory technique, bacterial expression system, centrifugation technique, or extraction methods. It is possible that the MnSOD transgene or other antioxidant transgenes yet to be tested may, as part of the insertion and recombination process, result in a lower yield of minicircle construct. Further studies will be required to increase the percent yield and quantity of mc-MnSOD construct for gene therapy experiments in primates and for clinical trials in patients.
This study was supported by Project I and Pilot Project (Zhang) of NIH/NIAID grant 1U19AI68021-01 (Greenberger) (09/01/05–08/31/10), entitled The Center for Medical Countermeasures against Radiation (CMCR); Mitochondrial Targets against Radiation Damage.
No competing financial interests exist.