Prevention of late effects of irradiation lung damage by ... - CiteSeerX

Gene Therapy (1998) 5, 196–208  1998 Stockton Press All rights reserved 0969-7128/98 $12.00

Prevention of late effects of irradiation lung damage by manganese superoxide dismutase gene therapy M Epperly1, J Bray1, S Kraeger1, R Zwacka2, J Engelhardt2, E Travis3 and J Greenberger1 1

Department of Radiation Oncology, University of Pittsburgh Cancer Institute, Pittsburgh, PA; 2Department of Molecular and Cellular Engineering, University of Pennsylvania, Philadelphia, PA; and 3Department of Radiation Oncology, MD Anderson Cancer Clinic, Houston, TX, USA

Organ and tissue damage caused by ionizing irradiation is directly related to volume irradiated, total dose and dose rate. The acute effects are in part mediated by cellular activation of early response genes, including those for transcriptional activators of genes for humoral cytokines. In the lung, as in other organs, recovery from the acute effects of ionizing irradiation does not always correlate with prevention of the critical late effects, including fibrosis, which contribute to organ failure. An interventional technique by which to protect normal organs from the late effects of

irradiation has remained elusive. We now demonstrate that overexpression of a transgene for human manganese superoxide dismutase (MnSOD) delivered by plasmid– liposome, or adenovirus to the lungs of C57BL/6J or Nu/J mice, respectively, before irradiation exposure, decreases the late effects of whole lung irradiation (organizing alveolitis/fibrosis). These data provide a rational basis for the design of gene therapy approaches to organ protection from irradiation damage.

Keywords: radiation fibrosis; late effects; MnSOD

Introduction While the radiation chemistry of DNA strand breaks and molecular biology of cellular damage caused by ionizing irradiation are becoming more clear (including an understanding of the rapid activation of genes for cell cycle regulation,1–5 apoptosis6–8 and transcriptional and posttranscriptional regulation of cytokine mediators,9–13 the mechanism and therefore the prevention of the late effects of irradiation of tissues and organs have remained elusive. Organ and tissue damage by ionizing irradiation is known to be directly dependent upon the volume exposed, total dose, dose rate (including fraction size in multifraction exposure),14–16 and genetic background of the subject.17–19 In the lung, as in most other organ systems, the acute effects of irradiation exposure have been shown to correlate with elevated levels of cytokines including interleukin-1 (IL-1), tumor necrosis factoralpha (TNF-␣) and transforming growth factor-beta (TGF-␤).20–28 These are humoral regulators common to the response of tissues to other cellular damaging agents including heat, ultraviolet (UV) irradiation, infectious organisms, hyperbaric oxygen and chemical toxins.20–28 Interventional approaches toward modulation of levels of these and other cytokines associated with acute pulmonary irradiation exposure, including the delivery of basic fibroblast growth factor (bFGF)29,30 or corticosteroids,24 have been successful in some cases but do not always correlate with prevention of the late effects, which in the

Correspondence: JS Greenberger, University of Pittsburgh, Department of Radiation Oncology, 200 Lothrop Street, Pittsburgh, PA 15213, USA Received 1 July 1997; accepted 3 October 1997

rodent model include organizing alveolitis/fibrosis. The lack of concordance between prevention of acute irradiation effects (associated with elevated levels of some cytokines) and the late effects of ionizing irradiation may indicate disparate molecular mechanisms for each type of injury.31,32 Alternatively, irradiation damage associated with the late effects of ionizing irradiation may be attributable to molecular mechanisms common to the early effects, but not yet identified. We now demonstrate a decrease in the late effects of ionizing irradiation in the lungs of mice by intrapulmonary delivery and overexpression 1–4 days before irradiation of a transgene for human manganese superoxide dismutase (MnSOD). MnSOD neutralizes irradiationinduced singlet oxygen and superoxides that are created by reaction of intracellular oxygen and water molecules with high energy photons (␥-irradiation) or electrons (␤irradiation).33 Free radicals are known to be associated with the critical single and double strand breaks that cause cell death and activate the cellular repair processes, which are believed to start the process of humoral cytokine production and perhaps also lead to the late effects of organ damage. The bioavailability of MnSOD, the mRNA for which is naturally present at low levels before irradiation,34–39 is rapidly increased within 4 h of exposure to irradiation,33 and the half-life of mRNA is increased in irradiated cells in vitro.33 Since free radicals are produced within 10−10 to 10−9s of irradiation exposure40 and rapidly consume available antioxidant molecules, including MnSOD that are available in the cell, most irradiation exposure overwhelms cellular baseline antioxidant levels. We have tested the strategy of using gene therapy to overexpress MnSOD in the lungs before irradiation

Gene therapy prevention of late irradiation effects M Epperly et al

exposure. Detectable reduction in the late damaging effects of irradiation of the lungs by MnSOD gene therapy would provide a rational basis for developing other gene therapy approaches for radioprotection of other organ systems through pre-emptive intervention in the chemistry of irradiation damage to tip the balance in favor of prevention.

Results Intratracheal MnSOD plasmid–liposome injection decreases lung messenger RNA levels for IL-1, TNF-␣ and TGF-␤ C57BL/6J female (30–33 g) mice were injected intratracheally with human MnSOD plasmid–liposome complex or control lacZ plasmid–liposome complex and irradiated 1 day later to both lungs. Subgroups of mice received 1800, 1900 or 2000 cGy. We confirmed by port film, that the whole lungs of the treated mice were irradiated. The head and abdomen were not irradiated. Although no correlation has been established between increased levels of these cytokines and fibrosis, we sought to obtain evidence that they may correlate with the development of fibrosis. The mice that received plasmid–liposome MnSOD had higher levels of messenger RNA (mRNA) for MnSOD measured by slot blot normalized to actin as a control (on the day of irradiation, but before irradiation) (Figure 1), and these mice showed lower levels of mRNA for inflammatory mediators IL-1, TGF-␤, and TNF-␣ in the lungs at 7 and 14 days after irradiation (Table 1). Lungs from MnSOD plasmid–liposome-injected mice compared with control groups of mice at several timepoints after irradiation, demonstrated a significant decrease in messenger RNA levels by slot blot for TGF␤, IL-1 and TNF-␣ compared with nontransgene-injected irradiated mice, again normalized to actin as a control (Table 1). Levels of plasma TGF-␤ after irradiation were measured at days 1, 4, 14 and 28 and were decreased in the MnSOD gene therapy group, but baseline levels between control groups varied between experiments such that the plasma assay did not detect significant differences. Human-specific MnSOD was detectably transcribed in whole lung explants from the MnSOD plasmid– liposome-injected mice, but not the lacZ plasmid– liposome-injected or uninjected control mice (Figure 2a). Explants of tracheal and alveolar type II cells cultured for 1 day according to published methods41 were also positive for the human MnSOD mRNA as measured by nested reverse transcriptase-polymerase chain reaction (RT-PCR) (Figure 2b). Biochemical activity of MnSOD was significantly increased in the lungs of MnSOD plasmid–liposome-injected mice to 4.98 ± 0.23 U/mg protein (P = 0.038), compared with the MnSOD activity of 4.08 ± 0.19 U/mg and 3.41 ± 0.26 U/mg for control and lacZ plasmid–liposome-injected mice, respectively. Units are as defined in Materials and methods (Figure 2c). Increased expression of MnSOD RNA was detected early after gene therapy in the MnSOD plasmid–liposome groups, but was also induced by irradiation and detected at later time-points in the lacZ plasmid–liposome and control-irradiated groups (Table 2). Thus, the data were consistent with an increase in the whole lung MnSOD RNA activity above background that was attributable to

transcription of the human MnSOD transgene in cells of the lung before irradiation, including cells explanted to tracheal or alveolar type II cell cultures. Furthermore, MnSOD transgene overexpression before irradiation was associated with a decrease in irradiation-induced MnSOD after day 2 in that group of mice (Table 2). There was also a detectable decrease in transcription of IL-1, TNF-␣ and TGF-␤ in this MnSOD plasmid–liposome-injected group (Table 1).

MnSOD plasmid–liposome gene therapy increases survival and decreases alveolitis in C57BL/6J mice The survival of C57BL/6J female mice that received either 1900 or 2000 cGy to both lungs was significantly improved (P ⬍0.001) in groups receiving MnSOD plasmid–liposomes the day before irradiation (Figure 3a and b). The prolonged survival of control-irradiated mice receiving 2000 cGy compared with 1900 cGy may have been attributable to the use of two different animal care facilities during the 2 years of these experiments. Dose– response survival experiments carried out with irradiated mice in each facility over this time interval support this hypothesis. The late histopathologic effects of irradiation following intratracheal gene therapy in C57BL/6J mice were compared between the MnSOD plasmid–liposome, lacZ plasmid–liposome, and control irradiated groups, but not the uninjected groups. The MnSOD transgenetreated mice developed significantly less alveolitis– fibrosis (P = 0.001) as determined by either two-tailed Wilcoxon test or Kruskal–Wallis test (Figure 3c). The less severe histopathology in an MnSOD plasmid–liposometreated mouse, compared with an irradiated control mouse, is shown by example in the upper panels of Figure 4a. Intratracheal MnSOD adenovirus injection decreases irradiation-induced alveolitis in Nu/J mice In a second model system using a strain with a different genetic background incidence of irradiation-induced late effects,17 and a second independent vector transfer system,42,43 Nu/J mice were studied. We first determined the LD50 lung lethal dose for these mice at 900 cGy and then studied doses of 850, 900 and 950 cGy. Mice were divided into five groups for each study of 10–13 mice per group. Control animals received no irradiation and no gene therapy. The second group of animals received irradiation to both lungs. Three other groups received intratracheal injection of mutant replication-deficient MnSOD adenovirus, copper/zinc superoxide dismutase (Cu/ZnSOD) adenovirus, or lacZ adenovirus, respectively (109 infectious units in 0.05 ml each), 4 days before irradiation. A 4-day interval was chosen to allow viral transgene expression and to minimize the acute cytotoxic effects which are detected even in immune incompetent Nu/J mice.42,43,44–46 Human MnSOD mRNA was detected in the lungs of MnSOD adenovirus-injected mice, but not in the other groups. Biochemical activity in this group at 4 days after injection (on the day of irradiation) was significantly increased. The MnSOD activity increased from 4.17 ± 0.18 U/mg in control mice to 5.35 U/mg (P = 0.036) for mice injected with MnSOD adenovirus. The level was 3.95 ± 0.09 U/mg (P = 0.477) for lacZ adenovirus-injected mice. Cu/ZnSOD biochemical activity increased from 6.34 ± 0.25 U/mg in control mice to 7.40 ± 0.23 (P = 0.037) and 5.92 ± 0.13 (P = 0.217) U/mg in mice injected with

197


198

Figure 1 Effect of MnSOD plasmid–liposome intratracheal injection on pulmonary levels of mRNA MnSOD detected at day 0 by slot blot of whole lung RNA (24 h after injection of plasmid–liposome complex – four mice per group). In separate groups of 12 mice, intratracheal injections of MnSOD plasmid–liposome complexes or lacZ plasmid–liposome complexes were performed, then 24 h later, control mice as well as injected mice were evaluated. The lungs were removed, immediately frozen in liquid nitrogen, and stored at −70°C. RNA was extracted as described above. Slot blots were performed (as described in Methods) to quantify levels of expression of MnSOD (a). The blot was stripped and probed for actin (b) for normalization of RNA loading.

adenovirus for Cu/ZnSOD or lacZ, respectively. We detected a decrease in messenger RNA levels by slot blot for the acute inflammatory proteins in irradiated MnSOD adenovirus-injected mice compared with each of the other groups of mice. There was also a decrease in plasma TGF-␤ levels in MnSOD adenovirus-treated compared with other groups of control mice (data not shown). Evaluation of the histopathology of lung damage was carried out uniformly at 132 days after irradiation of Nu/J mice. In the 850 and 900 cGy dose groups no deaths were detected at this time-point, while in the 950 cGy group some mice in the irradiated control group or lacZ plasmid–liposome group had died. Unirradiated control animals demonstrated no significant alveolitis, while those receiving 850, 900 or 950 cGy showed significant increases in alveolitis (Figure 5). The protective effect of MnSOD adenovirus was detected in the 850, 900 and 950 cGy groups (Figure 5). This protection is shown by example in the lower panels of Figure 4a. LacZ adenovirus intratracheal injection produced an inflammatory response alone (data not shown). There was no detectable decrease (but a potential increase, although the standard error bars were large) in alveolitis in the lacZ adenovirus

plus 850 cGy group compared with the 850 cGy irradiated, noninjected animals (Figure 5). In striking contrast, mice receiving MnSOD adenovirus transgene therapy showed a significant decrease in percentage alveolitis at 132 days in all three dose groups. Mice receiving Cu/ZnSOD adenovirus did not show a significant decrease in alveolitis compared with irradiated control animals. The anatomic localization of adenovirus, compared with the plasmid–liposome-delivered transgene, was tested by immunohistochemistry for the lacZ transgene and showed positivity in the upper as well as lower airway lining cells (Figure 4b). While these immunohistochemistry data with lacZ cannot be reliably compared with the expected anatomic level and percentage of cells positive for the MnSOD transgene, they do provide further support for the positive nested RT-PCR results with the human MnSOD transgene. The lacZ data also confirm the successful injection technique for delivery of transgene to the airway. Our studies show, both with plasmid–liposomes and adenovirus, a pulmonary protective effect of MnSOD transgene therapy before lung irradiation in two mouse strain models.


199

Figure 1 Continued.

Discussion There are gene products known to be associated with rapid cellular responses to ionizing irradiation. For example, exposure of epithelial cells to 100–200 cGy irradiation has been demonstrated to induce egr-1, NF␬␤, c-jun, c-fos and other transcriptional stimulatory elements.6–13 These transcriptional regulators activate genes for many cytokine regulators and cellular protective gene products.9,28 Radioresistance of mammalian cells in culture can be induced by the expression of recombinant oncogenes of several categories47 or by increasing the availability of specific recombinant growth factors.28 The mechanism of these two forms of radiation resistance may have a common molecular pathway, but the molecular mechanism of radiation protection by either an oncogene or a cytokine is unknown. There are data to support a common role of MnSOD and metallothionein in the cellular response to chemical-induced or radiation-induced damage.33,48,49 Recent data from one of these systems support the concept that overexpression before irradiation of an antioxidant gene that is naturally expressed at a lower level may be radioprotective in vivo.49 Protection of the rodent lung from bleomycininduced fibrosis has been reported using a transgene approach.50 For irradiation protection, MnSOD overexpression is theoretically attractive as it does not push cells out of G051 and has an advantage in this regard,

especially for cells in the lung which may not be cycling during irradiation exposure. Our studies with radioprotection of the lung have derived insight from recent research initiatives in cystic fibrosis. The work of several groups demonstrated that adenovirus vectors can transfer the cystic fibrosis transmembrane conductance regulator (CFTR) gene to bronchoalveolar lining cells.42,43 Problems have included shutoff of expression of the transgene, and/or rapid elimination of the vector and transgene43 and inefficient delivery of the gene to less than 50% of airway cells.43 Direct inhalation transfer of a transgene was recently demonstrated by liposome delivery.52 Transfer of transgenes for metabolic or catabolic enzymes such as ␣1 antitrypsin to the tracheobronchial tree has been demonstrated.45 The common factor in these reports was that transgene expression was transient and that ⬍50% of cells was transfected in most cases. The present studies demonstrated clear, significant increases in survival and decreases in irradiation-induced late effects of pulmonary irradiation in both C57BL/6J mice and Nu/J mice by overexpression of MnSOD transgene by intratracheal gene therapy before irradiation. Furthermore, both plasmid–liposome and adenovirus delivery of human MnSOD resulted in a decrease in alveolitis in two model systems. It is possible that some limitations of gene therapy for cystic fibrosis may be of


200 Table 1 Densitometric comparison of levels of mRNA for TGF-␤, IL-1 and TNF-␣ from whole lung RNA slot blots TGF-␤/Actin

IL-1/Actin

TNF-␣/Actin

0 1 4 7 14

0.73 ± 0.44 2.70 ± 1.90 0.213 ± 0.01 0.460 ± 0.19 0.14 ± 0.01

1.26 ± 0.31 1.70 ± 0.87 1.05 ± 0.58 6.50 ± 2.27 0.89 ± 0.05

0.45 ± 0.05 0.45 ± 0.16 0.63 ± 0.20 0.68 ± 0.14 0.55 ± 0.12

1051 ± 132 1052 ± 205 912 ± 143 720 ± 73 1964 ± 284

MnSOD

0 1 4 7 14

0.49 ± 0.18 0.412 ± 0.33 0.16 ± 0.02 * 0.215 ± 0.06 * 0.177 ± 0.05

0.377 ± 0.09 0.67 ± 0.42 * 0.18 ± 0.03 * 2.65 ± 0.58 * 0.48 ± 0.08

0.66 ± 0.11 0.59 ± 0.51 * 0.05 ± 0.01 * 0.32 ± 0.04 * 0.04 ± 0.01

1893 ± 417 1277 ± 258 1593 ± 70 959 ± 99 1943 ± 394

lacZ

0 1 4 7 14

0.285 ± 0.10 0.248 ± 0.15 0.178 ± 0.01 1.475 ± 0.99 0.267 ± 0.09

0.5 ± 0.09 0.67 ± 0.50 0.24 ± 0.05 4.56 ± 3.17 1.1 ± 0.27

0.31 ± 0.09 0.35 ± 0.26 0.65 ± 0.01 0.30 ± 0.13 0.29 ± 0.11

2534 ± 142 1459 ± 201 1604 ± 178 814 ± 82 1943 ± 394

Group

Plasmid

C57BL/6J

Control

C57BL/6J

C57BL/6J

Day after irradiationa

Actin

a

Whole lung RNA was probed for levels of IL-1, TGF-␤ and TNF-␣ as described in Materials and methods. Slot blot analysis was performed on RNA isolated from the lungs of control C57BL/6J mice, or mice which were injected with MnSOD plasmid–liposome complex, or lacZ plasmid–liposome complex on days 0, 1, 4, 7 or 14 after 2000 cGy irradiation. The blots were probed with P32-labeled PCR fragments for either IL-1, TGF-␤ or TNF-␣. The blots were subsequently stripped and probed for actin. Results are presented in densitometric units of IL-1, TGF-␤ or TNF-␣ normalized to actin as the mean ± s.e.m. of three lungs for each of four mice killed at each time-point after irradiation on day 0. Plasmid–liposome was delivered on day −1. Student’s t test was performed comparing control mice and mice injected with lacZ plasmid–liposome with the MnSOD plasmid–liposome complex-injected mice. (*Significant decrease compared with the lacZ plasmid–liposome groups.) Densitometric units for actin, which were used in normalizing results, are shown.

advantage to pulmonary radiotherapy protection. Radioprotective transgene expression was transient in our experiments and this could be an advantage for clinical radiotherapy to avoid any potential deleterious sideeffects of long-term overexpression. The delivery of the MnSOD transgene by plasmid–liposome or adenovirus to less than all airway cells appeared to be adequate for protection from X-ray damage. At present, we do not know whether this effect is mediated by transgene action in epithelial cells of the airway, macrophages or both cell types, nor do we know if the level of expression exerts its protective effect by a ‘bystander effect’ which has been observed in other cancer models.53 The present data support the notion that tracheal epithelial cell and alveolar cell airway damage is in fact critical to the pathophysiology of organizing alveolitis in mice (and its correlates, radiation pneumonitis and fibrosis in humans). Since both plasmid–liposomes and adenovirus were radioprotective in our animal model systems, clinical radiation protection could be feasible using currently available plasmid–liposomes which can reliably reach the tracheobronchial system, perhaps reaching into the terminal bronchioli. Expression of elevated levels of an irradiation protective transgene or combination of genes in bronchoalveolar cells during the 2–3 days preceding and 2–3 days following each fraction of a course of treatment, may prove to provide a significant advantage in the clinic where a protocol of multiple sessions of biweekly inhalation gene therapy would be fashioned. Studies of fractionated irradiation and multiple plasmid– liposome treatments are in progress to define more precisely the dose modifying effects of radioprotective gene therapy.

Materials and methods Mice intratracheal liposome injections and irradiation C57BL/6J female mice (30–33 g) (Jackson Laboratories, Bar Harbor, ME, USA) were intratracheally injected with 78 ␮l of either MnSOD plasmid–liposome complex or lacZ plasmid–liposome complex. Nu/J mice (Jackson Laboratories) were injected with 50 ␮l of adenovirus vector (109 p.f.u.) containing transgenes for either lacZ, Cu/ZnSOD, or MnSOD as follows. Mice were anesthetized with 1.25 mg of nembutal anesthesia/20 mg body weight (Abbott Laboratories, North Chicago, IL, USA) as an intraperitoneal injection. The trachea was surgically exposed and a 1 cc syringe with a 28-gauge needle was used to inject into the trachea. The incision was closed with Autoclip 9.0 mm wound clips (Clay Adams, Parsippany, NJ, USA). The pRK5 plasmid containing the human MnSOD cDNA as a transgene39 and pIEP-lacZ plasmid42 were grown in LB (Luria–Bertani) medium and isolated using Qiagen Giga columns (Qiagen, Chatsworth, CA, USA). The plasmid DNA was resuspended in nuclease free water at a concentration of 10 mg/ml. For each mouse to be injected, 500 ␮g of DNA (50 ␮l) and 28 ␮l of Lipofectin (Gibco/BRL, Grand Island, NY, USA) were mixed and allowed to sit at room temperature for 30 min. The complexes were then placed on ice until the time of injection. Groups of at least 13 C57BL/6J mice received 1800, 1900 or 2000 cGy to the pulmonary cavity 1 day after plasmid–liposome injection. All mice received single fraction irradiation at the mid-thoracic plane to both lungs at a dose rate of 200 cGy/min delivered by a 6 MeV linear accelerator (Varian Corporation, Palo Alto, CA,


201

Figure 2 Detection of human MnSOD transcripts in mouse lungs by nested RT-PCR 24 h after MnSOD plasmid–liposome injection in: (a) whole lung lysates; or (b) from 1-day cultures of explanted tracheal cells or alveolar type II (AT-II) cells41 (arrow shows human MnSOD specific transcripts). Day 0 is 24 h after injection, which is the day of irradiation. Human MnSOD mRNA was clearly detectable at day 0 and day 1 in whole lungs (a) of mice Nos 6–8 and Nos 9–11 (No. 5 was negative by both slot blot and RT-PCR). Nested RT-PCR also detected human MnSOD mRNA in both cultured tracheal and alveolar type II cells41 (b) from the treated, but not control groups. (c) Biochemical activity of MnSOD detected in cells in vitro, right panel, compared with known amounts of MnSOD, left panel.

USA). The head and abdomen were shielded by 10 onehalf value layer lead blocks.

Adenovirus vectors For adenovirus injection, the MnSOD transgene was excised from the pRK5-MnSOD plasmid39 by digestion with EcoRI and PuvI. The digests were run on a 1% agarose gel, and a 700 base pair fragment which corresponded to the size of the MnSOD insert was isolated. This fragment contained the entire coding region of the

human MnSOD gene. Cu/ZnSOD constructs were generated from human placental mRNA by reverse transcription followed by polymerase chain reaction (PCR) with primers specific for human Cu/ZnSOD.54 This fragment harbored restriction sites for direct cloning of the Cu/ZnSOD transgene into adenoviral constructs. Recombinant adenoviral constructs were generated by cloning the MnSOD or Cu/ZnSOD transgenes into pAd. CMV.link followed by cotransfection of NheI cut pAd plasmid with ClaI cut Ad5.sub360 (E3-deleted) viral DNA


202

Figure 2 Continued.

Table 2 Effect of MnSOD plasmid–liposome gene therapy before irradiation on MnSOD total RNA expression in C57BL/6J mouse lungs MnSOD expressiona

Group Day 0b

Day 1

Day 4

Day 7

Day 14

MnSOD plasmid–liposome 2000 cGy

10.050 ± 3.252

12.455 ± 2.669

1.618 ± 0.219

2.100 ± 0.632

0.515 ± 0.030

lacZ plasmid–liposome 2000 cGy

0.412 ± 0.025 (P = 0.025)

5.135 ± 0.496 (P = 0.036)

2.247 ± 0.288 (P = 0.132)

11.100 ± 1.283 (P = 0.001)

0.885 ± 0.100 (P = 0.012)

Control 2000 cGy

0.435 ± 0.086 (P = 0.025)

4.970 ± 0.792 (P = 0.036)

2.805 ± 0.264 (P = 0.013)

8.273 ± 1.474 (P = 0.008)

0.707 ± 0.099 (P = 0.112)

a

Expression of MnSOD RNA was measured by slot blot analysis on RNA extracted from the lungs of control mice, lacZ plasmid– liposome, and MnSOD plasmid–liposome-injected mice at 0, 1, 4, 7 and 14 days following 2000 cGy irradiation. MnSOD/actin levels were calculated as described in Table 1, and in Materials and methods. Results are the mean ± s.e.m. of three lungs per group at each time-point. Student’s t test compared significance of the level of MnSOD RNA expression in the MnSOD plasmid–liposome group with the lacZ plasmid–liposome or control irradiation group. b Day 0 denotes 24 h after injection.

according to published methods.42 A shield was designed to facilitate simultaneous irradiation of eight mice, exposing both lungs, but protecting the head and abdomen. Groups of at least 13 Nu/J mice that had been injected with adenovirus 4 days previously, received 850, 900 or 950 cGy to the pulmonary cavity. The surviving mice were killed at 132 days after irradiation.

Culturing of alveolar type II cells Alveolar type II cells were isolated from the lungs of C57BL/6J mice as previously described by Corti et al.41 Briefly, the lungs were perfused by injecting a 0.9% NaCl solution through the right ventricle of the heart, filled with 1 ml of dispase, and then allowed to collapse naturally expelling a portion of the dispase. Low-melt agarose (0.5 ml of 1% stored in a 45°C water bath) was injected into the lungs, covered immediately with crushed ice, incubated for 2 min, removed, placed in 2 ml

of dispase, incubated for 45 min at room temperature, then placed on ice. The lungs were transferred to 7 ml of 0.01% DNase in DMEM medium with the digested lung teased from the airways and swirled for 5–10 min at room temperature. The resulting suspension was filtered through a 40 ␮m cell strainer. The cells were then centrifuged at 25 g for 10 min at 4°C and resuspended in 10 ml of DMEM medium. Biotinylated anti-CD32 (0.65 ␮g per million cells) and biotinylated anti-CD45 (1.5 ␮g per million cells) (Pharmingen, San Diego, CA, USA) were added to the cells and incubated at 37°C for 30 min. The cells were then washed twice, resuspended in 7 ml of DMEM medium, added to the streptavidin-coated beads, then incubated with gentle rocking for 30 min at room temperature. The tube was attached to a magnetic tube separator. The cell suspension was then removed from the bottom of the tube, centrifuged, resuspended in DMEM medium, and incubated overnight in tissue cul-


203

Figure 3 Effect of intratracheal injection of MnSOD plasmid–liposome complex on survival and development of organizing alveolitis in C57BL/6J mice irradiated to either 2000 cGy or 1900 cGy. Control mice, or mice which had been injected with either MnSOD plasmid–liposome complex, or lacZ plasmid–liposome complex, were irradiated 1 day later to either: (a) 2000 cGy (10 mice per group – experiments carried out during 1996); or (b) 1900 cGy (13 mice per group – experiments carried out during 1997 in a different animal facility thus the different survival curve for controls) to both lungs. Survival was recorded daily and plotted as survival fraction for: control irradiated (쎲); MnSOD plasmid–liposome complex injected (왖); or lacZ plasmid–liposome complex-injected mice (쮿). Results were plotted from the day of irradiation as day 0.63 (c) Mice from the irradiation survival curves shown in (a) and (b) were analyzed for lung damage resulting from either 200 cGy or 1900 cGy using published methods. All survivors in the 1900 cGy group were killed on day 125 to determine the degree of alveolitis at that point. The lungs were excised, expanded in OCT, frozen, sectioned and stained with H&E. For each mouse, 36 sections were visually scored and the percentage of alveolitis determined. Variations between groups of mice in alveolitis rates were estimated as the median, and differences were tested with a Kruskal–Wallis or Wilcoxon test. The panels on the left show the box and whisker plots of distribution of percentage of alveolitis by individual mice, while the panels on the right represent treatment groups. Mouse label prefixes are C for control mice, L for mice receiving lacZ plasmid–liposome complex and M for mice injected with MnSOD plasmid–liposome complex.


204

4a

4b

Figure 4(a) Effect of MnSOD plasmid–liposomes or MnSOD adenovirus on lung histopathology following irradiation and demonstration of localization of transgene to the airway. (a) Upper panels: Lung from a C57BL/6J nonirradiated control mouse, an irradiated control mouse, and an irradiated mouse which were treated with MnSOD plasmid–liposome complex (all three harvested at day 150 after irradiation). (Lower panels): Lung from an Nu/J nonirradiated control mouse, an irradiated control mouse, or a mouse that received MnSOD adenovirus (bottom panels day 120 after irradiation). The section from each irradiated mouse demonstrates the irradiation-induced alveolitis that occurs before death. The section from the MnSOD transgenetreated mouse shows decreased alveolitis similar to the unirradiated control. (b) LacZ-stained airway cells in plasmid–liposome (a) on adenovirus, (c) injected animals, controls are below each as (b) and (d), respectively.


205

Figure 5 Effect of adenoviral MnSOD or Cu/ZnSOD transgene therapy on the development of irradiation damage in Nu/J mice following 850, 900 or 950 cGy irradiation. At 132 days (10 mice per group) unirradiated, control Nu/J mice, irradiated mice or mice injected with adenovirus containing the transgene for either MnSOD, Cu/ZnSOD, or LacZ (4 days before irradiation) were killed. The lungs were then excised, sectioned and H&E stained.17 The percentage of lung containing organizing alveolitis was determined by visually examining (five slides per lobe × five lobes = 25 slides per mouse × 10 = 250 slides per group) the slides under an inverted microscope at ×40 magnification using published methods.17 For the 850 cGy group, there was a difference in MnSOD adenovirus-treated mice, compared with lacZ adenovirus-treated mice (P = 0.065), or control irradiated mice (P = 0.057). For the 900 cGy group, the differences were P ⬍ 0.001 and P ⬍ 0.001, respectively. For the 950 cGy group, the differences were P ⬍ 0.001 and P ⬍ 0.001, respectively. Student’s t-test was used to compare the percentage of alveolitis in MnSOD-treated mice with that of control mice and mice injected with lacZ adenovirus. There was no significant difference in the percentage of alveolitis detected in mice injected with Cu/ZnSOD adenovirus from that detected in control irradiated mice or from mice irradiated then injected with lacZ adenovirus.

ture-treated plates. Nonadherent cells, which include the alveolar type II cells, were centrifuged and resuspended at a concentration of 1 to 1.5 × 106 cells/ml, then plated in six-well tissue culture plates which were coated with fibronectin (10 ␮g/cm2). The cultures were incubated in a humidified, 10% CO2 chamber at 37°C.

Biochemical determination of MnSOD activity C57BL/6J mice were intratracheally injected and irradiated as described above. At various time-points following irradiation, the mice were killed, the lungs excised and snap-frozen in liquid nitrogen. MnSOD activity was determined biochemically as previously described.48,55 Briefly, the lungs were thawed and resuspended in 20 mm Tris (pH 7.6), and homogenized using a Polytron PT2000 (Brinkmann Instruments, Westbury, NY, USA) homogenizer. For total superoxide dismutase (SOD) activity, each assay tube contained 20 mm Tris (pH 7.6), 1 mm DEPATAC, 1 U of catalase, 5.6 × 10−8 m nitroblue tetrazolium (NBT), 0.1 mm xanthine, 0.05 mm bathocuproinedisulfonic acid (BCS), 0.13 mg/ml defatted bovine serum albumin, and 100 ␮g of lung protein. Xanthine oxidase (10−2 m) was added, and the change in absorbance at 560 nm as a function of time was measured. For determination of MnSOD activity, 5 mm of sodium cyanide which inhibits Cu/ZnSOD was added to

each assay tube and incubated for 45 min at room temperature. Xanthine oxidase was added, and the change in absorbance was measured as described.48,55 As a negative control, a cell line (−/−) from MnSOD homologous recombinant knockout mouse fibroblasts was used.56 One unit of activity resulted in a 50% inhibition of the NBT reduction.

Slot blots For slot blots, 10 ␮g of total RNA was dissolved in 100 ␮l of DEPC-treated H2O followed by addition of 300 ␮l of a solution of 6.15 m formaldehyde, and 10 × SSC. The RNA was incubated at 67°C for 15 min and loaded on to a Schleicher & Schuell Minsifold II slot blot system (Schleicher & Schuell, Keene, NH, USA) where the RNA is bound to a BA85 nitrocellulose membrane (Schleicher & Schuell). The membrane was baked for 2 h at 42°C and prehybridized in Denhardt’s buffer (50% formamide, 5 × Denhardt’s reagent, 100 ␮g/ml denatured salmon sperm DNA, 0.1% SDS and 5 × SSPE). The membrane was probed for 18 h at 42°C with P32-labeled PCR fragment of the MnSOD gene, then was washed twice in 1 × SSPE/0.1% SDS for 5 min at room temperature, followed by two washes in 0.1 × SSPE/0.1% SDS at 42°C for 30 min. The blots were placed in an autoradiographic cassette with radiographic film. Densitometric analysis


206

was carried out to determine the binding of the probe to the blot using a Molecular Dynamics Personal Densitometer SI (Molecular Dynamics, Sunnyvale, CA, USA). The membrane was stripped of radioactivity by placing the blot in 0.1 × SSPE/0.1% SDS which had been heated to 100°C. After cooling to 65–70°C, the membrane was prehybridized (as described above) and reprobed for IL1, TNF-␣ or TGF-␤1 with actin as a standard. Densitometry was carried out as published and all results normalized for actin as a standard.57

RT-PCR For RT-PCR, groups of C57BL/6J female mice were injected intratracheally and irradiated as described above. At 0, 1, 4, 7 or 14 days after irradiation, mice were killed by cervical dislocation, lungs were excised and snap-frozen in liquid nitrogen. The lungs were homogenized in the presence of 3 ml of triazol using a Polytron Model PT2000 Homogenizer (Brinkmann). The homogenized samples were incubated for 5 min at room temperature, followed by the addition of 0.6 ml of chloroform, mixed, and incubated at room temperature for 3 min. The samples were then centrifuged at 12 000 g for 15 min at 4°C. The aqueous phase was transferred to a new centrifuge tube, 1.5 ml of isopropyl alcohol was added, incubated at room temperature for 10 min, and centrifuged at 12 000 g for 10 min at 4°C. The pellet was washed with 75% ethanol and centrifuged at 7500 g for 5 min at 4°C, air-dried, and resuspended in 500 ␮l of DEPC-treated water. Two micrograms of each RNA sample were used in the RT-PCR reaction where the mRNA was amplified by mixing the RNA with poly-dT, 10 ␮m mixture of dCTP, dATP, dTTP, dGTP, and Superscript II reverse transcriptase (Gibco/BRL). The tubes were incubated for 50 min at 42°C, 10 min at 95°C, followed by incubation at 4°C. Nested PCR technique58 was used to amplify MnSOD cDNA for the nested RT-PCR reaction or DNA for the nested PCR reaction by mixing 0.01 ␮l of the reverse transcriptase reaction with the first set (broadly detecting both human and endogenous mouse MnSOD) of 5′ and 3′ oligonucleotides, 0.2 mm mixture of dATP, dCTP, dTTP and dGTP, and Taq DNA polymerase (Boehringer Mannheim, Indianapolis, IN, USA). The mixture was subjected to 20 cycles of 94°C (30 s), 60°C (50 s) and 72°C (90 s) in a Perkin Elmer Model 9600 Gene Amp PCR System (Perkin Elmer, Foster City, CA, USA). For the second nested PCR reaction (detecting only the human transgene), 0.1 ␮l of the first reaction was added to the 24 ␮l of PCR reaction mixture in the presence of the second set of 5′ and 3′ primers. This second set of primers which was specific for human MnSOD was internal to the first set of primers and did not overlap in sequences. Thermocycling was identical to that of the first PCR reaction, except that the reaction was 35 cycles. The first set of primers were comprised of a 5′ primer of CGGCGG CATCAGCGGTAAGCCAGCACTA (nucleotides 61–89) and a 3′ primer of TGAGCCTTGGACACCAACAGAT GCA (nucleotides 505–529). The second set of primers consisted of a 5′ primer of GCTGGCTCCGGCTTT GGGGTATCTG (nucleotides 128–152) and a 3′ primer of GCTGAGCTTTGTCCAGAAAATGCTC (nucleotides 388– 412). The expected size for the correct nested MnSOD PCR product was 222 bp. The PCR products (15 ␮l) were electrophoresed on a 1% agarose gel, transferred to a

nylon membrane which was then probed with a P32-PCR fragment of the pRK5-MnSOD plasmid for 20 h in Church Buffer at 65°C. The membrane was washed twice in a solution of 0.1% SDS and 0.1 × SSC for 40 min at 65°C and placed in an autoradiographic cassette with radiographic film.

Histopathology For each treatment group, five to 10 mice were examined histopathologically for irradiation-induced lung damage. The lungs were expanded by intratracheal injection of 0.8 ml of OCT (Miles Laboratories, Elkart, IN, USA) followed by excision. The five lobes of each mouse lung were separated, placed in each of five 15 × 15 mm base molds, and frozen in OCT.59 Nine sections (each 10 microns thick) were prepared from each lobe with 100 microns between each section using a Shandon AS620E Cryotome (Shandon/Lipshaw, Pittsburgh, PA, USA). The sections were hematoxylin and eosin (H&E)-stained by dipping the slides in Harris hematoxylin for 1 min and 45 s, rinsed with tap water until clear, dipped in eosin for 1 min, then rinsed in water. The slides were allowed to air-dry, and then dipped twice in 95% ethanol, twice in 100% ethanol, twice in 50:50 ethanol:xylene, and twice in 100% xylene. The slides were mounted using Permamount then examined under a Zeiss inverted microscope. Fibrosis was confirmed by staining slides with Masson’s Trichrome. Briefly, the slides were placed in saturated picric acid for 30 min, washed in running water until the yellow color disappeared, placed in Harris hematoxylin solution for 10 min, and then rinsed in distilled water. The slides were then placed in Biebrich scarlet-acid fuchsin for 30 min, rinsed in distilled water, placed in 5% phosphomolybdic-phosphotungstic acid solution for 10 min, stained for 3 min with aniline blue, rinsed in distilled water, then placed in 1% acetic acid for 3 min. The slides were next dehydrated (as described above), mounted with Permamount, then visually scored for the percentage of the lung exhibiting fibrosis and/or alveolitis.17 For each mouse, at least seven sections were made from each of five lobes and scored separately. At least 36 total sections were analyzed for each mouse. Immunohistochemistry A rabbit anti-lacZ antibody (5 Prime-3 Prime, Boulder, CO, USA) was used to detect lacZ expression as previously described.60 Briefly, the lungs were excised from C57BL/6J mice which were intratracheally injected with lacZ plasmid–liposome complex 24 h previously, or Nu/J mice which were intratracheally injected with lacZ adenovirus 4 days previously. The lungs were expanded in 1 ml of OCT embedding medium (Baxter Diagnostics, McGaw, MI, USA), frozen in OCT, and sectioned.59 The sections were fixed in methanol for 5 min, and then placed in a solution of 0.5% H2O2 for 30 min at room temperature to block endogenous peroxidase activity. The slides were then washed three times in Tris-buffered saline (TBS), covered with 3% goat serum for 30 min at 24°C, and washed three times in TBS. The slides were covered with 1 ml of a 1:500 dilution of the anti-lacZ antibody, incubated at room temperature for 2 h, washed three times in TBS, and covered with 1 ml of a 1:200 dilution of a biotinylated goat anti-rabbit antibody (Sigma, St Louis, MO, USA). The slides were incubated for 2 h at room temperature, washed three times in TBS,


and covered with 1 ml of a 1:300 dilution of avidin–biotin complex solution for 1 h at room temperature. The slides were washed in TBS and placed in a diaminobenzedine (DAB) solution (5 mg DAB (Sigma), 40 ␮l of 30% H2O2, 40 ml of 0.1 m Tris (pH 7.6)) for 10 min at room temperature. Next, the slides were washed in TBS, followed by successive washes in 95% ethanol, 100% ethanol, 50% ethanol/50% xylene, and xylene. Permamount was used to mount the slides. For scoring, at least five slides from each lobe were examined under an inverted Zeiss optical microscope at ×40 magnification.

Statistics Standard survival analysis included Kaplan–Meier plots of survivor function and log rank tests of differences between treatment groups.61,62 Prognostic factors, including radiation dose and radiation protection methods among genetically similar mice, were examined with Cox regression.63 The association between alveolitis rates and survival was examined by adding mouse median alveolitis rates as time-dependent covariates. Treatment group differences in alveolitis were tested by nonparametric methods; either the Wilcoxon test for two samples of the Kruskal–Wallis test, or the Jonckheere–Terpstra method for overall differences or trend.64,65 The group differences were applied to mouse median percentage alveolitis.

9

10

11

12

13

14

15 16

17

18

19

Animal welfare All protocols were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh. Veterinary care was provided by the Central Animal Facility of the University of Pittsburgh in strict accordance with the Institutional Animal Care and Use Committee of the University of Pittsburgh guidelines.

21

Acknowledgements

22

We thank Luann Berry and Anne Fisher for their technical assistance, and Dr William Gooding for the biostatistics. This paper was supported by Research Grants CA39851, CA41068, CA42546, and DE08912 of the National Institutes of Health.

23

20

24 25

References 1 Huang TS et al. Distinct p53-mediated G1/S checkpoint responses in two NIH/3T3 subclone cells following treatment with DNA-damaging agents. Oncogene 1996; 13: 625–632. 2 Maity A et al. Delayed cyclin B1 expression during the G2 arrest following DNA damage. Oncogene 1996; 13: 1647–1657. 3 Bates S, Rowan S, Vousden KH. Characterization of human cyclin G1 and G2: DNA damage inducible genes. Oncogene 1996; 13: 1103–1109. 4 Gadbois DM et al. Alterations in the progession of cells through the cell cycle after exposure to alpha particles or gamma rays. Radiat Res 1996; 146: 414–424. 5 Rudoltz MS et al. Ionizing radiation and the cell cycle: a review. Radiat Oncol Invest 1996; 4: 147–158. 6 Kharbanda S et al. Activation of the c-abl tyrosine kinase in the stress response to DNA-damaging agents. Nature 1995; 376: 85–89. 7 Kharbanda S et al. Nuclear signaling induced by ionizing radiation involves colocalization of the activated p56/p53lyn tyrosine kinase with p34cdc21. Cancer Res 1996; 56: 3617–3621. 8 Kharbanda S et al. The stress response to ionizing radiation

26

27

28

29

30

31

involves c-abl-dependent phosphorylation of SHPTP1. Proc Natl Acad Sci USA 1996; 93: 6989–6901. Weichselbaum RR, Hallahan D, Fuks Z, Kufe D. Radiation induction of immediate–early genes: effectors of the radiationstress response. Int J Radiat Oncol Biol Phys 1994; 30: 229–234. Hallahan DH et al. Protein kinase C mediates X-ray inducibility of nuclear signal transducers EGR and JUN. Proc Natl Acad Sci USA 1991; 88: 2156–2160. Hallahan DE et al. Membrane-derived second messenger regulates X-ray-mediated tumor necrosis factor ␣ gene induction. Proc Natl Acad Sci USA 1994; 91: 4897–4901. Martin M et al. Preferential induction of c-fos, versus c-jun protooncogene during the immediate–early response of pig skin to ␥-rays. Cancer Res 1993; 53: 3246–3249. Brach MA et al. Ionizing radiation induces expression and binding activity of the nuclear factor ␬B. J Clin Invest 1991; 88: 691–695. Rubin P, Cassarett E. Textbook of radiation pathology. In: Hall EJ (ed): Radiobiology For The Radiobiologist, 3rd edn. JB Lippincott Company: Philadelphia, 1988. Puck TT, Markus PI. Action of X-rays on mammalian cells. J Exp Med 1956; 103: 653–666. Elkind MM, Sutton H. Radiation response of mammalian cells grown in culture I. Repair of X-ray damage in surviving Chinese hamster cells. Radiat Res 1960; 13: 556–593. Dileto C, Travis EL. Fibroblast radiosensitivity in vitro and lung fibrosis in vivo: comparison between a fibrosis-prone and fibrosis-resistant mouse strain. Radiat Res 1996; 146: 61–67. Franko AJ, Sharplin J. Development of fibrosis after lung irradiation in relation to inflammation and lung function in a mouse strain prone to fibrosis. Radiat Res 1994; 140: 347–355. Dragani TA, Manenti G, Pierotti MA. Genetics of murine lung tumors. Adv Cancer Res 1995; 67: 83–113. Franko AJ, Sharplin J, Ghahary Z, Barcellos-Hoff MH. Immunohistochemical localization of transforming growth factor beta and tumor necrosis factor ␣ in the lungs of fibrosis-prone and ‘nonfibrosing’ mice during the latent period and early phase after irradiation. Radiat Res 1997; 48: 245–256. Johnston CJ et al. Differences in correlation of mRNA gene expression in mice sensitive and resistant to radiation-induced pulmonary fibrosis. Radiat Res 1995; 142: 197–203. Grande JP. Role of transforming growth factor beta in tissue injury and repair. Proc Soc Exp Biol Med 1997; 214: 27–40. Coker RK et al. Transforming growth factors ␤1, -␤2, and -␤3 stimulate fibroblast procollagen production in vitro but are differentially expressed during bleomycin-induced lung fibrosis. Am J Pathol 1997; 150: 981–991. Ward HE et al. The pulmonary response to sublethal thoracic irradiation in the rat. Radiat Res 1993; 136: 15–21. Fine A, Goldstein RH. The effect of transforming growth factor ␤ on cell proliferation and collagen formation by lung fibroblasts. J Biol Chem 1987; 262: 3897–3902. Finkelstein JN, Johnston CJ, Baggs R, Rubin P. Early alterations in extracellular matrix and transforming growth factor ␤ gene expression in mouse lung indicative of late radiation fibrosis. Int J Radiat Oncol Biol Phys 1994; 28: 621–631. Anscher MS et al. Changes in plasma transforming growth factor ␤ during radiotherapy and the risk of symptomatic radiationinduced pneumonitis. Int J Radiat Oncol Biol Phys 1997; 37: 253–258. Mauch P et al. Hematopoietic stem cell compartment: acute and late effects of radiation therapy and chemotherapy. Int J Radiat Oncol Biol Phys 1995; 31: 1319–1339. Haimovitz-Friedman A et al. Protein kinase C mediates basic fibroblast growth factor protection of endothelial cells against radiation-induced apoptosis. Cancer Res 1994; 54: 2591–2597. Tee PG, Travis EL. Basic fibroblast growth factor does not protect against classical radiation pneumonitis in two strains of mice. Cancer Res 1995; 55: 298–302. Wara WM, Phillips TL, Margolis LW, Smith V. Radiation pneumonitis: a new approach to the derivation of time-dose factors. Cancer 1973; 2: 547–552.

207


208

32 McDonald S, Rubin P, Phillips TL, Marks LB. Injury to the lung from cancer therapy: clinical syndromes, measurable endpoints, and potential scoring systems. Int J Radiat Oncol Biol Phys 1995; 31: 1187–1203. 33 Akashi M et al. Irradiation increases manganese superoxide dismutase mRNA levels in human fibroblasts. J Biol Chem 1995; 270: 15864–15869. 34 Wong GHW, Elwell JH, Oberley JW, Goeddel DV. Manganese superoxide dismutase is essential for cellular resistance to cytotoxicity of tumor necrosis factor. Cell 1989; 8: 923–931. 35 Yen HC et al. The protective role of manganese superoxide dismutase against adriamycin-induced acute cardiac toxicity in transgenic mice. J Clin Invest 1996; 98: 1253–1260. 36 Su WY et al. Extracellular superoxide dismutase mRNA expressions in the human lung by in situ hybridization. Am J Respir Cell Mol Biol 1997; 16: 162–170. 37 Jaeschke H. Mechanisms of oxidant stress-induced acute tissue injury. Proc Soc Exp Biol Med 1995; 209: 104–111. 38 Tsan MF. Superoxide dismutase and pulmonary oxygen toxicity. Proc Soc Exp Biol Med 1993; 203: 286–293. 39 Wispe JR et al. Human Mn-superoxide dismutase in pulmonary epithelial cells of transgenic mice confers protection from oxygen injury. J Biol Chem 1992; 267: 23937–23941. 40 Riley PA. Free radicals in biology: oxidative stress and the effects of ionizing radiation. Int J Radiat Biol 1994; 65: 27–33. 41 Corti M, Brody AR, Harrison JH. Isolation and primary culture of murine alveolar type II cells. Am J Resp Cell Mol Biol 1996; 14: 309–315. 42 Engelhardt JF et al. Adenovirus-mediated transfer of the CFTR gene to lung of nonhuman primates: biological efficacy study. Hum Gene Ther 1993; 4: 759–769. 43 Grubb BR et al. Inefficient gene transfer by adenovirus vector to cystic fibrosis airway epithelia of mice and humans. Nature 1994; 371: 802–806. 44 Flotte TR et al. Gene expression from adeno-associated virus vectors in airway epithelial cells. Am J Respir Cell Mol Biol 1992; 7: 349–356. 45 Rosenfeld MA et al. Adenovirus-mediated transfer of a recombinant alpha 1-antitrypsin gene to the lung epithelium in vivo. Science 1991; 252: 431–434. 46 Kochanek S et al. A new adenoviral vector: replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and ␤-galactosidase. Proc Natl Acad Sci USA 1996; 93: 5731–5736. 47 FitzGerald TJ et al. Resistance of hematopoietic stem cells to Xirradiation at clinical low-dose rate is induced by several classes of oncogenes. Radiat Res 1990; 122: 44–53. 48 Spitz DR, Oberley LW. An assay for superoxide dismutase activity in mammalian tissue homogenates. Anal Biochem 1989; 179: 8–18.

49 Schwarz MA et al. Metallothionein protects against the cytotoxic and DNA-damaging effects of nitric oxide. Proc Natl Acad Sci USA 1995; 92: 4452–4456. 50 Tran PH et al. Prevention of bleomycin-induced pulmonary fibrosis after adenovirus-mediated transfer of the bacterial bleomycin resistance gene. J Clin Invest 1997; 99: 608–617. 51 Epperly MW, Berry LA, Halloran A, Greenberger JS. Inhibition of G1 phase arrest induced by ionizing radiation in hematopoietic cells by overexpression in genes involved in the G1/S phase transition. Radiat Res 1995; 143: 245–254. 52 Stribling R et al. Aerosol gene delivery in vivo. Proc Natl Acad Sci USA 1992; 89: 1364–1370. 53 Wygoda MR et al. Protection of herpes simplex virus thymidine kinase-transduced cells from ganciclovir-mediated cytotoxicity by bystander cells: the good Samaritan effect. Cancer Res 1997; 57: 1699–1703. 54 Peled-Kamar M et al. Oxidative stress mediates impairment of muscle function in transgenic mice with elevated level of wildtype Cu/Zn superoxide dismutase. Proc Natl Acad Sci USA 1997; 94: 3883–3887. 55 Oberley LW, Spitz DR. Nitroblue Tetrazolium. In: Greenberg RA (ed). CRC Handbook of Methods For Oxygen Radical Research. CRC Press: Boca Raton, FL, 1985, pp 217–230. 56 Yibling L et al. Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nat Genet 1995; 11: 376–381. 57 Jahroudi N, Ardekani AM, Greenberger JS. Ionizing irradiation increases transcription of the von Willebrand factor gene in endothelial cells. Blood 1996; 88: 3801–3814. 58 Hurwitz DR et al. Systemic delivery of human growth hormone or human factor IX in dogs by reintroduced genetically modified autologous bone marrow stromal cells. Hum Gene Ther 1997; 8: 137–156. 59 Watkins S. Cryosectioning. In: Ausubel FM et al. (eds). Current Protocols in Molecular Biology. Greene Publishing and WileyInterscience: New York, 1989, pp 14.2.1–14.2.8. 60 Hsu SM, Cossman J, Jaffe ES. Lymphocyte subsets in normal human lymphoid tissues. Am J Clin Pathol 1983; 80: 21–30. 61 Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 23: 457–481. 62 Peto R, Pike M, Armitage P. Design and analysis of randomized clinical trials requiring prolonged observations of each patient: II, analysis and examples. Br J Cancer 1997; 101: 1–39. 63 Cox DR. Regression models and life tables (with discussion). J Roy Stat Soc 1972; 43: 187–220. 64 Hollander M, Wolfe D. Nonparametric Statistical Methods. John Wiley and Sons: New York, 1973. 65 Mack G, Skillings H. A Friedman type rank test for main effects in a two-factor ANOVA. J Am Stat Assoc 1980; 73: 947–951.