Ionizing radiation increases adenovirus uptake and improves transgene expression in intrahepatic colon cancer xenografts

ARTICLE

doi:10.1016/S1525-0016(03)00143-6

Ionizing Radiation Increases Adenovirus Uptake and Improves Transgene Expression in Intrahepatic Colon Cancer Xenografts Ming Zhang,1,* Shengping Li,1 Jun Li,1 William D. Ensminger,2 and Theodore S. Lawrence1 1

Department of Radiation Oncology and 2Department of Pharmacology, University of Michigan Medical School, 1331 E. Ann Street, Ann Arbor, Michigan 48109-0582

*To whom correspondence and reprint requests should be addressed. Fax: (734) 763-1581. E-mail: [email protected].

Specific targeting and transgene expression in tumors are critical in adenovirus gene therapy for intrahepatic colon carcinoma metastases. In this study, we investigated if ionizing radiation could increase adenoviral uptake by cells. Various human cell lines and rat hepatocytes were irradiated prior to exposure to a cytomegalovirus (CMV) promoter-driven green fluorescent protein (GFP) marker gene adenoviral vector. We found that ␥-radiation increased the number of GFP-positive cells in a dose- and time-dependent manner for most cells, ranging from 4.6- to 27.1-fold after a 4-Gy treatment. No induction occurred for lentiviral vector, lipofection, or naked plasmid exposure. Preincubation of cells with adenovirus failed to show an increase, suggesting that radiation might mediate adenoviral infection by inducing viral uptake into cells. We demonstrated that radiation induced internalization of a fluorescence-labeled adenovirus (Cy3-Ad) and an increase in intracellular viral DNA content. Rats bearing intrahepatic colon carcinoma xenografts were irradiated in the tumor region followed by portal venous administration of an adenoviral vector containing a CMV–␤-galactosidase (␤-gal) gene. Radiation increased ␤-gal activity in tumors as much as 5.4-fold after a 25-Gy treatment. These data suggest that a combination of regional radiation treatment with adenovirus gene therapy is a rational strategy for improving adenoviral targeting and transgene expression in tumors. Key Words: gene therapy, cytosine deaminase, colorectal neoplasm, 5-fluorouracil, ionizing radiation

INTRODUCTION Only a small fraction of patients with liver-confined colorectal metastases can be cured by resection [1,2]. Although systemic and hepatic arterial chemotherapy can produce objective responses, treatment is limited by systemic and biliary toxicity, respectively. Enzyme prodrug gene therapy has been proposed as a way of achieving high intratumoral concentrations of drug while minimizing systemic and regional toxicity. One approach using an adenovirus vector containing a herpes simplex virus thymidine kinase gene in combination with the prodrug ganciclovir induced tumor necrosis in patients with metastatic colorectal carcinoma in the liver [3]. However, it is difficult to eradicate tumors by this system alone, and there is concern that the adenovirus dose necessary to achieve therapeutic effectiveness may be difficult to attain with acceptable systemic toxicity. Therefore, a multimodality treatment combining other therapeutic approaches

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy 1525-0016/03 $30.00

such as radiation or immunotherapy with gene therapy may be critical in achieving clinical success. We have demonstrated that an adenoviral vector containing cytosine deaminase, a transgene from yeast (yCD) that converts the benign prodrug 5-fluorocytosine (5FC) into the chemotherapeutic agent 5-fluorouracil (5FU), significantly inhibits or eliminates tumor growth in animals [4 –7]. Importantly, 5FU can act as a radiosensitizing agent at approximately 1/10 the concentration of that required to produce direct cytotoxicity [8]. Thus this yCD/prodrug strategy could be useful in combined chemoradiotherapy. We have also obtained selective yCD expression in carcinoembryonic antigen (CEA)-expressing tumors versus normal liver by controlling yCD gene expression with the CEA promoter in adenovirus (CEA-yCD) [9]. However, yCD expression in liver was still detected due to the basal CEA-yCD expression and nonspecific adenoviral targeting in normal liver. These findings prompted us to search for additional

21

ARTICLE

doi:10.1016/S1525-0016(03)00143-6

TABLE 1: Radiation-induced adenovirus infection in other cell lines Cells

Origin

LoVo

Human colon cancer

WB

Rat hepatocytes

Radiation-induced viral infection (fold)

Adenovirus/cell (m.o.i.)

6.2 ⫾ 0.9*

0.2

27.6 ⫾ 2.3*

40.0

HepG2

Human liver cancer

1.1 ⫾ 0.2

25.0

H1299

Human lung cancer

0.9 ⫾ 0.3

0.5

MCF-10

Human breast epithelium

1.2 ⫾ 0.3

1.5

MCF-7

Human breast cancer

8.4 ⫾ 1.0*

3.0

D54

Human brain cancer

4.6 ⫾ 0.7*

1.5

UMSCC14A

Human head/neck cancer

1.0 ⫾ 0.2

3.0

HF

Human fibroblasts

2.8 ⫾ 0.2*

45.0

Cells in 24-well plates were infected by adenovirus at an optimized amount for 48 h immediately after 4 Gy radiation. GFP-positive cells were counted and radiation induction of viral infection was expressed as fold increase for GFP-positive cells. *Significantly elevated (P ⬍ 0.05) compared to nonirradiated cells of the same type (n ⫽ 4).

approaches that could enhance adenovirus-mediated gene transfer in tumors. Previous studies demonstrated that ionizing radiation greatly improves the stable expression of adenoviral [10] or associated-adenoviral gene transfer into cultured cancer cells [11]. We decided to test whether ionizing radiation might improve adenoviral targeting, and thus transgene expression, in intrahepatic colon carcinoma xenografts as a potential strategy for gene therapy. Importantly, high radiation doses can be safely and effectively administered to intrahepatic cancers either as fractionated radiation [12] or as a single large fraction [13]. Thus, the finding that radiation could increase gene transfer and expression would be highly valuable in designing a future protocol combining gene therapy with focal radiation for patients with intrahepatic colon cancer metastases. In this study, we first demonstrate that adenovirusmediated transgene expression in cultured cells was increased by ␥-radiation. The mechanism of radiation-induced adenoviral gene expression in cells and the applicability of these findings in an intrahepatic model of colon cancer are further explored.

RESULTS Ionizing Radiation Increases Adenovirus-Mediated Gene Transfer in Cells To determine whether radiation could induce adenovirusmediated gene transfer at the cellular level, we used an adenoviral vector containing a CMV promoter-driven green fluorescent protein marker gene (CMV–GFP) to infect various types of cells, including human cancer cells LoVo (colon), HepG2 (liver), H1299 (lung), MCF-7 (breast), D54 (brain), and UMSCC14A (head and neck) as well as normal WB (rat hepatocyte), MCF-10 (human breast epithelium), and HF (human fibroblast) cells. The

22

amounts of virus for infections in these cells were optimized so that transduced cells (GFP-positive) were at ⬃50 for untreated cells. ␥-Radiation treatment at 4 Gy followed by immediate viral infection resulted in significantly increased gene transfer efficiency in LoVo, WB, MCF-7, D54, and HF cells evidenced by increased numbers of GFP-positive cells in irradiated cells compared to nonirradiated cells (6.2 ⫾ 0.9-, 27.6 ⫾ 2.3-, 8.4 ⫾ 1.0-, 4.6 ⫾ 0.7-, and 2.8 ⫾ 0.2-fold, respectively) (Table 1). This demonstrates that radiation can increase transgene expression in many but not all cell types. As our goal is to develop adenovirus gene therapy for intrahepatic colon carcinoma, we decided to explore radiation-induced gene transfer further in rat hepatocyte (WB) and human colon carcinoma cells (LoVo). We demonstrated that the radiation induction was dose-dependent and time-sensitive. GFP-positive cell numbers were dramatically enhanced with an increase in radiation dose in both cell lines compared to nonirradiated cells (representative induction at 4 Gy of radiation treatment is shown in Fig. 1A), leading to a maximum 6.1- and 27.1fold induction for LoVo (190 ⫾ 28 vs 31 ⫾ 10 cells) and WB (298 ⫾ 16 vs 11 ⫾ 2 cells), respectively, when viral vector was given immediately after radiation treatment at 4 Gy (Fig. 1B). Induction was most dramatic when cells were infected immediately after radiation treatment at all doses used (0 h) and decreased significantly with time. Radiation had no effect when infection was delayed for 24 h in LoVo cells and decreased 40% in WB cells (Fig. 1B). To determine whether radiation-induced gene transfer was specific to adenoviral infection, we assessed other gene transfer approaches. In contrast to the marked induction of adenoviral infection, radiation failed to increase GFP-expression cell numbers in either LoVo (Fig. 2A) or WB (Fig. 2B) cells when a lentiviral vector of CMV-

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

doi:10.1016/S1525-0016(03)00143-6

ARTICLE

FIG. 1. Effect of radiation on adenovirus infection efficiency. LoVo and WB cultures were irradiated, infected with optimized amounts of adenovirus (CMV-GFP), and assessed for GFP-positive cells 48 h after infection. (A) Representative results of adenovirus-transduced GFP-positive LoVo and WB cells with or without radiation treatment of 4 Gy. Bright and fluorescence fields are shown at the top and bottom, respectively. (B) Quantification of GFP-positive cells for LoVo and WB in response to radiation with increasing delay between irradiation and viral infection (n ⫽ 4).

GFP, a Lipofectamine-conjugated CMV-GFP plasmid, or a CMV-GFP plasmid alone was used as gene transfer method. Thus, radiation appears to induce only adenoviral-mediated gene transfer and not other standard techniques of gene transfer Enhanced Adenoviral Vector Uptake in Response to Radiation To determine whether radiation-induced gene expression was due to an enhanced adenoviral vector uptake or a subsequent intracellular process, we incubated cells with viral vector for 2 h prior to radiation. Radiation treatment (4 Gy) was then given only after removing nonbinding vector from the culture medium. We found that cells preincubated with adenoviral vector failed to respond to radiation treatment, evidenced by unchanged GFP-expressing cell numbers for both LoVo (Fig. 3A) and WB cells (Fig. 3B) compared to cells receiving radiation prior to viral infection. This lack of induction suggests that

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

enhanced adenoviral gene transfer occurs as a result of an increased total viral uptake following radiation treatment. The role of radiation in increasing viral uptake was further demonstrated by assessing adenoviral genome contents in cells. After irradiation with 4 Gy and subsequent adenoviral infection, we extracted total cellular DNA from both LoVo and WB cells and assessed it by real-time PCR quantification of the viral DNA by amplifying the GFP gene sequence from 1 ⫻ 103 infected cells. GFP-containing DNA copies from radiation-treated LoVo and WB cells were much more numerous than those from non-radiation-treated cells (2840 ⫾ 336 vs 1284 ⫾ 142 copies and 2114 ⫾ 263 vs 210 ⫾ 17 copies, respectively) (Fig. 4A), while ␣-actin contents were similar regardless of radiation treatment (data not shown). These changes were further indicated in a representative gel analysis when we used PCR products from 20 cycles (Fig. 4B). Importantly, these increased signals were in agreement

23

ARTICLE

doi:10.1016/S1525-0016(03)00143-6

FIG. 2. Effect of the radiation treatment on gene transfer efficiency by nonadenoviral methods. LoVo and WB cells were prepared and irradiated as described for Fig. 1. Cells were transduced using adenovirus (CMV-GFP), lentivirus (CMV-GFP), Lipofectamine-conjugated plasmid (CMV-GFP), or naked plasmid alone (CMV-GFP) immediately after radiation. GFP-positive cells were counted 48 h later. (A) LoVo cells. (B) WB cells (n ⫽ 4).

with the induction of GFP expression in cells described above, suggesting that radiation treatment most likely increased viral uptake by cells rather than intracellular DNA replication. To test the role of replication directly, we inhibited cellular DNA replication using hydroxyurea, an inhibitor of ribonucleotide reductase. As expected, 5 mM hydroxyurea failed to affect the radiation induction pattern (data not shown). Moreover, confocal laser scanning of cells infected by a Cy3-labeled adenoviral vector (Cy3-Ad) suggested that radiation treatment substantially increased total viral uptake (Fig. 4C). This evidence strongly supports our hypothesis that radiation treatment of cells facilitates adenoviral uptake. Radiation Enhances Adenovirus-Mediated Gene Expression in Intrahepatic Tumors To determine whether ␥-radiation treatment could increase gene expression in tumors in vivo, we irradiated

nude rats bearing human colon carcinoma xenografts in the liver prior to portal vein infusion of an adenoviral vector. Adenovirus containing CMV-␤-gal was used instead of CMV-GFP for better quantification of transgene expression in tissues. We found that gene expression resulting from adenoviral infusion given at 24 h after radiation treatment was significantly enhanced in tumors as well as the irradiated liver adjacent to the tumor in comparison to gene expression in nonirradiated tissues. This increase was radiation dose-dependent, leading to a maximum of 5.4- or 17.4-fold increase for tumors (0.70 ⫾ 0.21 vs 0.13 ⫾ 0.05 mU/100 ␮g) or adjacent liver (3.13 ⫾ 0.41 vs 0.18 ⫾ 0.05 mU/100 ␮g), respectively (Fig. 5A), after treatment with 25 Gy. Furthermore, radiation-induced gene expression in both tumor and liver appeared to be affected by the timing of the adenoviral infusion with respect to the radiation. We failed to detect the induction

FIG. 3. Effect of sequence of viral infection and radiation in transgene expression. (A) LoVo and (B) WB cells were either incubated with adenovirus (CMV-GFP) for 2 h, washed, and then irradiated (virus–radiation) or irradiated followed immediately by exposure to virus for 2 h (radiation–virus). GFP-positive cells were counted 48 h later (n ⫽ 4).

24

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

ARTICLE

doi:10.1016/S1525-0016(03)00143-6

FIG. 4. PCR detection of adenoviral DNA and viral uptake in cells in response to radiation treatment. LoVo and WB cells were infected by adenovirus (CMV-GFP) immediately after radiation with 4 Gy or no radiation. Total cellular DNA was extracted from infected cells for real-time PCR amplification of adenovirus-derived GFP DNA sequence or ␣-actin DNA sequence. (A) Quantification of intracellular viral DNA copies. *Different from nonradiated cell controls (P ⬍ 0.05). (B) Agarose gel fractionation of products from 20-cycle real-time PCR amplification. M, molecular weight marker. (C) Confocal laser scanning of adenovirus uptake in GFP-expressing stable LoVo and WB cells. Cells in chamber slides were incubated with 1 ⫻ 109 particles of Cy3-Ad immediately after radiation treatment with 4 Gy or no radiation. Two hours later, cells were washed and fixed on slides for dual-color confocal laser scanning (Texas red and FITC, 60⫻ objective on Olympus microscope IX SLA) (n ⫽ 4).

when virus was infused immediately after radiation treatment (0 h) (Fig. 5B). Moreover, delayed viral infusion (0, 12, 24, and 36 h) after a low dose of radiation (5 Gy) induced ␤-gal expression in neither tumor nor liver (Fig. 5C). However, gene expression in both tissues was significantly induced by viral infusion 12 h after treatment with 25 Gy (1.8-, 10.3-, and 1.4-fold increase for remote liver, local liver, and tumor, respectively) and reached a maximum at 24 h followed by a decline at 36 h (Fig. 5D). The gene expression in other tissues including lung, heart, kidney, spleen, colon, and bone marrow was either undetectable or at low background levels and was unaf-

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

fected by liver radiation treatment (data not shown). Thus, induction seemed to be specific to the irradiated region.

DISCUSSION In this study, we have found that regional ␥-radiation treatment of intrahepatic colon carcinoma xenografts in nude rats significantly improves adenovirus-mediated transgene expression in tumors in a radiation dose-dependent manner. In a cell culture system, ␥-radiation directly induced viral uptake by cells, leading to significantly im-

25

ARTICLE

doi:10.1016/S1525-0016(03)00143-6

FIG. 5. Transgene expression in irradiated intrahepatic tumors and normal liver of nude rats. Intrahepatic colon carcinoma xenografts were established using LoVo cells as described under Materials and Methods. Adenovirus (CMV-␤-gal) at 4 ⫻ 109 pfu was administered through the portal vein after regional radiation treatment. ␤-Gal expression was measured 72 h after infection from nonirradiated remote liver, tumor-surrounding liver, and tumors and is represented as fold increase in comparison to activities measured from nonirradiated rats. (A) Adenovirus was administered at 24 h after radiation treatment. (B) Adenovirus was administered immediately after radiation treatment. (C) Adenovirus was administered at various times after radiation of 5 Gy. (D) Adenovirus was administered at various times after radiation of 25 Gy (n ⫽ 4). *Different from controls (P ⬍ 0.05).

proved gene transfer efficiency in rat hepatocytes and human colon carcinoma cells. A previous study indicated that ionizing radiation induced adenovirus-mediated stable gene expression in cultured cells through a gene integration mechanism [10], and a recent study from this group further demonstrated the roles of activation for DNA-dependent protein kinase and ataxia telangiectasia mutated protein in radiation-enhanced integration [14]. However, it is unlikely that gene integration played a major role in our model system that focused on short- to intermediate-term gene expression. This suggests that radiation has the potential to improve the specificity of gene therapy through more than one mechanism. The mechanism by which viral uptake is increased by radiation is not yet clear. In the case of the cell culture experiments, it seems possible that there is an increase in a cell surface adenoviral receptor. Adenovirus infects cells by binding to the high-affinity coxsackie and adenovirus receptor (CAR) through fiber protein and the low-affinity ␣v integrin receptor through the penton base protein. Our preliminary data from FACS analysis indicated that neither CAR nor integrin ␣v␤3 responded to radiation within 3 h of treatment (data not shown). However, it is not clear whether CAR and/or ␣v␤3 might be playing a role in radiation-induced adenovirus infection observed with a 24-h delay after radiation in our animal model. Another cell surface protein, dynamin, which specifically regulates clathrin-mediated endocytosis, has been demonstrated to

26

play an important role in adenovirus entry via the clathrin-coated pit pathway [15] in a relatively short period of time (half-time ⬍10 min) [16]. The potential regulation of dynamin function by radiation treatment needs to be further investigated. Ionizing radiation has been demonstrated to produce a wide range of other membrane effects such as stimulation of activation of epidermal growth factor receptor within minutes [17] and of ceramide synthase [18]. Furthermore, it has been shown that virus– cell binding can be upregulated by three- to sixfold upon cell cycling, This appears to result from increased receptor expression on the cell surface, suggesting that changes in viral receptor levels are a physiological process [19]. Although it is unlikely that our cell culture observations are a result of cell cycle changes, given the very short time course of the increase, the influence of nuclear and cytoplasmic events on the expression of cell surface viral receptor clearly requires further study. An important question raised in this study is the mechanism underlying the different time courses observed in vitro and in vivo. In our in vitro experiments, we found that the stimulation of viral uptake was greatest when cells were exposed to virus immediately after radiation, whereas in vivo the maximum effect was obtained by administering virus 24 h after radiation. It is possible that radiation could increase vascular permeability [20] and, therefore, viral access, which would not be detected in a cell culture system. Additional studies will be required to

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

doi:10.1016/S1525-0016(03)00143-6

ARTICLE

assess the relative importance of cellular and vascular effects on viral uptake. We found that radiation produced a greater induction in the normal liver adjacent to the tumor than in the tumor itself. This observation is consistent with the pattern found in cell culture using WB and LoVo cells. Therefore, it is not clear whether the increased uptake in vivo is a reflection of differences at the cellular level or whether the structure of the liver in the vicinity of the tumor plays a role. One observation that supports the latter concept is the difference in time course between the increase in cellular uptake, which is maximal around the time of radiation, compared to the maximal increase in the normal liver, which occurs 24 h after radiation. Importantly, gene expression in liver and other normal tissues remote from the radiation-treated region remained unchanged. In the clinical treatment of liver metastases, we irradiate a 1-cm region of normal liver around the tumor, based on the surgical finding that patients who undergo resection with a ⬍1-cm margin have a much higher chance of tumor recurrence [21]. The increase in viral uptake, and subsequent 5FU production, anticipated in this narrow zone of irradiated normal liver would be expected to result in an improved therapeutic effect against the microscopic disease that exists in this region, with minimal toxicity produced outside of this region. Our previous studies using an adenoviral vector containing a CEA promoter-driven yCD demonstrated that human colon carcinoma xenografts in nude rats preferentially expressed the yCD gene, which resulted in efficient prodrug 5FC conversion into 5FU in tumor cells [9] for regional chemotherapy and radiosensitization. However, significant yCD expression was still detected from normal liver due to the basal activity of the CEA promoter. Our current finding that regional radiation significantly improved gene expression in tumors suggests the potential to produce a greater regional therapeutic effect while minimizing toxicity to nonirradiated normal liver. This is particularly useful in patients with intrahepatic metastases since a three-dimensional confocal radiation strategy, which has been applied clinically, can be used to deliver high dose radiation either in a fractionated course of treatment [12] or as a single high-dose fraction [13] to a confined tumor-bearing region.

Type 5 adenoviral vectors containing CMV–␤-galactosidase or the CMV–GFP gene were obtained from the Vector Core Facility at University of Michigan. Lentiviral vector of CMV-GFP was prepared in 293T packaging cells using cotransfection of three plasmids coding individually for vesicular stomatitis virus glycoprotein, CMV–GFP transgene, and gag–pol structural gene. Viral titer was determined by infecting HeLa cells.

MATERIALS

Fluorescence labeling of adenoviral vector for cell binding and confocal laser scanning. Adenoviral vectors of CMV-␤-gal at 1 ⫻ 1012 particles were reacted with Cy3 dye (Amersham Pharmacia Biotech, Inc., Piscataway, NJ) in 0.1 M sodium carbonate buffer (pH 9.3) as recommended by the manufacturer. Labeled viral vectors were then dialyzed twice against 500 ml of buffer containing 10% glycerol, 10 mM Hepes, and 1 mM MgCl2 using Slide-A-Lyzer (7000 MWCO; Pierce, Rockford, IL). Aliquots were stored at ⫺80°C for cell binding studies. Stable cell lines of LoVo and WB expressing GFP were established by infecting cells with a lentiviral vector containing the CMV–GFP gene. These cells were first plated on eight-well chamber slides (Nalge Nunc International, Naperville, IL) and were then irradiated with 4 Gy followed by immediate incubation with 1 ⫻ 109 particles of

AND

METHODS

Cell lines and viral vectors. Human tumor cells including LoVo (colon), HepG2 (liver), H1299 (lung), MCF-7 (breast), D54 (brain), and human normal fibroblasts (HF) were purchased from ATCC. Human head/neck cancer cell UMSCC14A and normal breast epithelium cell MCF-10 were kindly provided by Dr. Thomas Carey (Otolaryngology, University of Michigan) and Dr. Stephen Ethier (Radiation Oncology, University of Michigan), respectively. An immortalized rat hepatocyte (WB) line was obtained from Dr. James Trosko (Michigan State University). All cell lines were maintained in RPMI medium containing 10% FBS and antibiotics in 5% CO2 at 37°C.

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy

Radiation treatment. ␥-Radiation was administered using a 60Co unit (Theratron AECL). Radiation doses were calibrated using an electrometer system directly traceable to the National Institute of Standards and Technology standard. Rat tumor model and adenoviral vector infusion. An intrahepatic colon carcinoma model in rat was established as previously described [9]. Briefly, 4 ⫻ 106 LoVo cells were implanted locally under the capsule of the liver. Tumor growth was validated 3 weeks after surgery by laparotomy. After rats were anesthetized, tumors received 5, 15, or 25 Gy of radiation through an anterior field (calculated at 0.5 cm depth) so that most of the normal liver was well out of the radiation field. CMV-␤-gal adenoviral vector (4 ⫻ 109 pfu) was administered by portal vein infusion immediately (0 h) or at various times (12, 24, and 36 h) after radiation treatment. Animals were allowed to recover for 72 h after portal vein infusion. Tissues of tumors, tumor-surrounding liver, remote liver, and other organs, including heart, lung, kidney, spleen, stomach, colon, and bone marrow, were harvested at the end of the experiment for ␤-gal activity measurement in milliunits (mU) from 100 ␮g of total tissue extract as described previously [9]. Animals receiving portal vein infusion of virus without radiation treatment served as controls. Adenoviral infection of cells. Cells were plated on 24-well plates at 2 ⫻ 104 cells/well overnight before being irradiated and infected with adenoviral vector containing a CMV-GFP gene. The multiplicity of infection (m.o.i.) used for each cell line was determined in pilot experiments to achieve countable GFP-positive cell numbers. The m.o.i.’s were 0.2 (LoVo), 40.0 (WB), 25.0 (HepG2), 0.5 (H1299), 1.5 (MCF-10), 3.0 (MCF-7), 1.5 (D54), 3.0 (UMSCC14A), and 45 (HF). In control experiments, lentiviral vector of CMV-GFP at 100 pfu, Lipofectamine (Life Technologies)-conjugated plasmid coding for CMV-GFP, and plasmid alone coding for CMV-GFP at 10 ng were used for non-adenovirus-mediated gene transferring. Transduction efficiencies were quantified by counting GFP-positive cells under a fluorescence microscope. Cellular DNA extraction and PCR. LoVo and WB cells at 2 ⫻ 105 cells/ well on six-well plates were irradiated first at 4 Gy (nonirradiated cells as control), followed by infection using 1 ⫻ 108 particles of CMV-GFP adenovirus vector for 2 h at 37°C. After extensive washing with PBS to remove unbound virus, total cellular DNA from infected cells was prepared using a genomic DNA extraction kit (Promega, Madison, WI). Adenoviral DNA contents were quantified using real-time PCR (Opticon from MJ Research, Waltham, MA) by amplifying the GFP gene sequence from total cellular DNA extracts of 1 ⫻ 103 infected cells using the QuantiTect SYBR Green PCR kit (Qiagen, Santa Clarita, CA) and 5⬘ATGGTGAGCAAGGGCGAGGA and 3⬘ACTTGTACAGCTCGTCCAT as primers. The sequence for the ␣-actin housekeeping gene sequence in cells was also amplified by PCR as an internal control using 5⬘CGAGATCCCTCCAAAATCAA and 3⬘TGTGGTCATGAGTCCTTCCA as primers. PCR products at 20 cycles were fractionated on a 1% agarose gel for visualization.

27

ARTICLE

Cy3-Ad for 2 h at 37°C. Nonirradiated cells were used as controls for basal level of viral binding. After extensive washing in PBS buffer 3⫻ to remove free viral particles, cells were fixed in 4% paraformaldehyde solution for confocal microscopy. Imaging analysis was performed using an Olympus IX SLA confocal laser scanning biological microscope equipped with FITC and Texas red laser guns operated by Fluoview software FV-300 (Olympus USA, Melville, NY). Data analysis and statistics. Data are presented as the means ⫾ the standard errors of four cell culture experiments (each performed in triplicate) or four mice. Values were compared by ANOVA. They were considered significantly different when P ⬍ 0.05.

ACKNOWLEDGMENTS We thank Mary Davis (Radiation Oncology, University of Michigan) for her review of the manuscript. This work was supported by NCI Grants CA80145 and CA84117 and Cancer Center Core Grant C274864 from the University of Michigan. RECEIVED FOR PUBLICATION JANUARY 15, 2003; ACCEPTED APRIL 7, 2003.

REFERENCES 1. Weiss, L, et al. (1986). Haematogenous metastatic patterns in colonic carcinoma. An analysis of 1541 necropsies. J. Pathol. 150: 195–203. 2. Fong, Y, et al. (1997). Liver resection for colorectal metastases. J. Clin. Oncol. 15: 938 –946. 3. Sung, M. W., et al. (2001). Intratumoral adenovirus-mediated suicide gene transfer for hepatic metastases from colorectal adenocarcinoma: results of a phase I clinical trial. Mol. Ther. 4: 182–191. 4. Kievit, E., et al. (1999). Superiority of yeast over bacterial cytosine deaminase for enzyme/prodrug gene therapy in colon cancer xenografts. Cancer Res. 59: 1417–1421. 5. Kievit, E., et al. (2000). Yeast cytosine deaminase improves radiosensitization and bystander effect by 5-fluorocytosine of human colorectal cancer xenografts. Cancer Res. 60: 6649 – 6655. 6. Hamstra, D. A., et al. (1999). Combined radiation and enzyme/prodrug treatment for head and neck cancer in an orthotopic animal model. Radiat. Res. 152: 499 –507. 7. Hamstra, D. A., Rice, D. J., Fahmy, S., Ross, B. D., and Rehemtulla, A. (1999). Enzyme/

28

doi:10.1016/S1525-0016(03)00143-6

8.

9.

10.

11. 12.

13. 14. 15. 16. 17.

18. 19.

20.

21.

prodrug therapy for head and neck cancer using a catalytically superior cytosine deaminase. Hum. Gene Ther. 10: 1993–2003. Lawrence, T. S., Davis, M. A., and Maybaum, J. (1994). Dependence of 5-fluorouracilmediated radiosensitization on DNA-directed effects. Int. J. Radiat. Biol. Phys. 29: 519 –523. Zhang, M., et al. (2002). Regional delivery and selective expression of a high activity yeast cytosine deaminase in an intrahepatic colon cancer model. Cancer Res. 63: 658 – 663. Zeng, M., Cerniglia, G. J., Eck, S. L., and Stevens, C. W. (1997). High-efficiency stable gene transfer of adenovirus into mammalian cells using ionizing radiation. Hum. Gene Ther. 8: 1025–1032. Kanazawa, T., et al. (2001). ␥-Rays enhance rAAV-mediated transgene expression and cytocidal effect of AAV-HSVtk/ganciclovir on cancer cells. Cancer Gene Ther. 8: 99 –106. Dawson, L. A., McGinn, C., Normolle, D., Ten Haken, R. K., Ensminger, W., and Lawrence, T. S. (2000). Escalated focal liver radiation and hepatic artery floxuridine for unresectable liver malignancies. J. Clin. Oncol. 18: 2210 –2218. Herfarth, K. K., et al. (2001). Stereotactic single-dose radiation therapy of liver tumors: results of a phase I/II trial. J. Clin. Oncol. 19: 164 –170. Nimura, Y., Ismail, S. M., Kurimas, A., Chen, D. J., and Stevens, C. W. (2002). DNA-PK and ATM are required for radiation-enhanced integration. Radiat. Res. 157: 562–567. Wang, K., Huang, S., Kapoor-Munshi, A., and Nemerow, G. (1998). Adenovirus internalization and infection require dynamin. J. Virol. 72: 3455–3458. Barlett, J. S., Wilcher, R., and Samulski, R. J. (2000). Infectious entry pathway of adeno-associated virus and adeno-associated virus vectors. J. Virol. 74: 2777–2785. Bowers, G., et al. (2001). The relative role of ErbB1-4 receptor tyrosine kinases in radiation signal transduction responses of human carcinoma cells. Oncogene 20: 1388 – 1397. Lin, X., Fuks, Z., and Kolesnick, R. (2000). Ceramide mediates radiation-induced death of endothelium. Crit. Care Med. 28(4 Suppl.): N87–93. Seidman, M. A., Hogan, S. M., Wendland, R. L., Worgall, S., Crystal, R. G., and Leopold, P. L. (2001). Variation in adenovirus receptor expression and adenovirus vector-mediated transgene expression at defined stages of the cell cycle. Mol. Ther. 4: 13–21. Debbage, P. L., et al. (2000). Vascular permeability and hyperpermeability in a murine adenocarcinoma after fractionated radiotherapy: an ultrastructural tracer study. Histochem. Cell Biol. 114: 259 –275. Poon, R. T., Fan, S. T., Ng, I. O., and Wong, J. (2000). Significance of resection margin in hepatectomy for hepatocellular carcinoma: a critical reappraisal. Ann. Surg. 231: 544 –551.

MOLECULAR THERAPY Vol. 8, No. 1, July 2003 Copyright © The American Society of Gene Therapy