(CANCER RESEARCH 58. 4.191-4401. October 1. 1998]

Retroviral Transfer of Human Cytochrome P450 Genes for Oxazaphosphorine-based Cancer Gene Therapy1

Youssef Jounaidi, Jodi E. D. Hecht, and David J. Waxman2

Division of Ceil and Molecular Biology. Department of Biology, Boston University, Boston. Massachusetts 02215

ABSTRACT

Cyclophosphamide (CPA) and ifosfamide (IFA) are widely used anti-cancer prodrugs that are bioactivated in the liver by specific cytochromeP450 enzymes (CYPs). The therapeutic activity of these antitumor agentscan be compromised by a low therapeutic index that is, in part, due to thesystemic distribution of activated drug metabolites. Here, recombinantretroviruses were used to deliver six different CPA- or IFA-metabolizinghuman CYP genes to 9L gliosarcoma cells: 2B6, 2C8, 2C9, 2CI8 (Met3"5and Thr'"5alíeles),2C19, and 3A4. Intratumoral cytochrome P450 expres

sion conferred substantial sensitivity to CPA cytotoxicity, with the mostdramatic effects seen with CYP2B6. Strong CPA chemosensitivity wasalso seen following transduction of CYP2C18-Met, despite a very low levelof CYP protein expression (>60-fold lower than that of 2B6). In contrastto CPA, the cytotoxicity of IFA was greatest toward tumor cells transduced with CYP3A4, followed by CYPs 2B6 and 2C18-Met. A substantialfurther increase in chemosensitivity was achieved upon transduction of2B6 or 2C18-Met-expressing tumor cells with P450 reducíase,whichprovided for more efficient intratumoral prodrug activation and cytotoxicity at lower drug concentrations. With 2B6- plus P450 reductase-trans-duced tumor cells, CPA but not IFA conferred a strong cell contact-independent bystander cytotoxic effect on non-P450-expressing 9L cells.CPA treatment of tumors that were transduced with 2B6 or 2C18-Mettogether with P450 reducíaseand were grown s.c. in immunodeficientmice resulted in a large enhancement of the liver P450-dependent antitumor effect seen with control 9L tumors, with no apparent increase in hosttoxicity (growth delay of >25-50 days in P450-expressing tumors versus-5-6 days without P450). CYP2B6plus P450 redactase and CYP2d8-Metplus P450 reducíasethus appear to be excellent gene combinations for usewith CPA in P450/prodrug activation-based cancer gene therapy.

INTRODUCTION

Several cancer chemotherapeutic agents, including CPA,3 IFA,procarbazine, and dacarbazine, are prodrugs that undergo liver P450-catalyzed metabolism to yield active, chemotherapeutic metabolites(1,2). The administration of anticancer prodrugs may provide advantages associated with improved chemical stability and slow release ofactivated drug metabolites in situ. However, because the metabolicactivation of classic cancer chemotherapeutic prodrugs largely occursin the liver, it leads to uncontrolled release and systemic distributionof cytotoxic drug metabolites that kill sensitive host cells, in additionto the tumor cell targets. Recently, gene therapy methods have beenintroduced to address this problem by delivering prodrug activationgenes to malignant cells, which thereby acquire the capacity for

Received 5/15/98: accepted 8/3/98.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported in part by NIH Grant CA49248 (to D. J. W.). Y. J. received PostdoctoralFellowship support from L'Association pour la Recherche sur le Cancer.

2 To whom requests for reprints should be addressed, at Department of Biology,Boston University. 5 Cummington Street. Boston. MA 02215. Phone: (617)353-7401:Fax: (617)353-7404; E-mail: djw@bio.bu.edu.

1The abbreviations used are: CPA. Cyclophosphamide; IFA. ifosfamide; CYP, cyto

chrome P450: P¡,inorganic orthophosphate: BROD, P450-catalyzed 7-benzyloxyresorufinO-deethylase activity; RT-PCR. reverse transcription-PCR: 9L/wt. wild-type 9L tumorcells: 9L/pBabe. 9L infected with pBabe-puro control vector retrovirus; 9L/P450. 9Ltransduced with human P450 gene: 9L/2B6/reductase. 9L transduced with 2B6 and P450reducíase; 9L/2C18-Met/reductase. 9L transduced with 2C18-Met and P450 reducíase:

seid, severe combined immunodeficient.

intratumoral prodrug activation (3-5). One widely studied model usesthe antiviral drug ganciclovir (6) in combination with herpes simplexvirus thymidine kinase, which phosphorylates the prodrug and generates nucleoside analogues that induce DNA chain termination andcell death in actively dividing cells (7, 8). Similarly, the bacterial genecytosine deaminase can be used as a "suicide gene" that confers on

tumor cells the ability to activate the antifungal compound 5-fluoro-cytosine to 5-fluorouracil, a potent anticancer agent (9, 10).

Current gene therapy technologies are limited by their inability todeliver prodrug activation or other therapeutic genes to a populationof tumor cells with 100% efficiency. The effectiveness of this cancergene therapy strategy can be greatly enhanced, however, by usingdrugs that exhibit a strong "bystander effect" (5). Bystander cytotox

icity results when active drug metabolites diffuse or are otherwisetransfered from their site of generation within a transduced tumor cellto a neighboring, naive tumor cell. Ideally, the bystander effect leadsto significant tumor regression even when a minority of tumor cells istransduced with the prodrug activation gene (e.g.. Refs. 11 and 12).Bystander cytotoxic responses may also be mediated through theimmune system, following its stimulation by interleukins and othercytokines secreted by tumor cells undergoing apoptosis (13). Although the ganciclovir/herpes simplex virus thymidine kinase and5-fluorocytosine/cytosine deaminase combinations have shown goodactivity in preclinical models, and clinical trials are underway (14-16), their ultimate effectiveness may be limited by the cell cycledependence of their cytotoxicity and by the fact that the bystandercytotoxic effect requires direct cell-cell contact through gap junctionalintercellular communication to transfer the active drug metabolite inthe case of ganciclovir (17). Furthermore, the nonmammalian natureof these two enzymes could elicit immunological responses thatfurther limit their effectiveness in cases in which repeated genedelivery and drug treatments are required. Finally, ganciclovir wasintroduced originally for the treatment of herpes infections (6), andconsequently, its ultimate therapeutic utility for cancer treatment isuncertain.

The oxazaphosphorine prodrugs CPA and IFA, widely used in thetreatment of human cancers (18), are bioactivated in the liver via a4-hydroxylation reaction catalyzed by certain P450 enzymes, withelectron input from the flavoenzyme NADPH P450 reducíase(19).The resultant 4-hydroxy metabolites undergo spontaneous ß-elimina-tion to yield acrolein and a bifunctional DNA alkylator, phosphor-amide mustard from 4-hydroxy-CPA. and isophosphoramide mustardfrom 4-hydroxy-IFA. These electrophilic mustards are active in a cellcycle-independent manner and covalently cross-link genomic DNA,leading to cell death via an apoptotic mechanism (20, 21). In the caseof IFA, an alternate, P450-catalyzed metabolic pathway deactivatesthe drug through side chain oxidation, yielding the inactive metabolites 2-dechloroethyl-IFA and 3-dechloroethyl-IFA and the neurotoxicmetabolite chloroacetaldehyde (22). Although IFA is an isomer ofCPA, it is activated by a distinct subset of P450 enzymes, both inrodent models and in humans (23, 24), and it exhibits a uniquespectrum of antitumor activity, host toxicity, and drug resistance(25-27). Intratumoral expression of the rat CYP gene 2B1 (28), theprotein product of which activates CPA and IFA by 4-hydroxylation(19, 23), has shown promise in preclinical studies as an effective

4391

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6.BASED CANCER GENE THERAPY

strategy to enhance both the selectivity and the cytotoxicity of thesedrugs in a variety of human and rodent tumor models (Refs. 29-32;

for review see Ref. 33). It is anticipated that the clinical developmentof this gene therapy strategy will benefit from the use of human CYPgenes involved in oxazaphosphorine activation (24, 34, 35), whichavoids complications associated with immune responses elicited bythe expression of CYP genes from heterologous species. This studywas undertaken, therefore, to evaluate the efficacy and potential utilityof human P450 genes for CPA- and IFA-based cancer gene therapy.

Recombinant retroviruses based on Moloney murine leukemia virus(36) and packaged in the ecotropic packaging cell line Bosc 23 (37)are presently generated and used to establish 9L gliosarcoma cell linesthat stably express six individual human CYP genes belonging to theCYP 2B, 2C, and 3A families (28). The impact of human CYP genetransfer, both alone and in combination with P450 reducíase genetransfer (38), on the responsiveness to oxazaphosphorine therapy wasevaluated both in vitro and in an in vivo tumor model system.

MATERIALS AND METHODS

Chemicals. CPA and IFA were obtained from the Drug Synthesis andChemistry Branch of the National Cancer Institute (Bethesda, MD). Thechemically activated derivatives of CPA and IFA, 4-hydroxyperoxy-CPA and4-hydroperoxy-IFA, were obtained from Nova Pharmaceutical Corp. (Bethes

da, MD). Puromycin hydrochloride was purchased from Sigma Chemical Co..and hygromycin was from Aldrich.

Construction of Recombinant Retroviruses. cDNAs encoding three human CYP 2C enzymes (39), CYP2C8 (clone 7b), CYP2C18-Met'S5 alíele

(clone 6b: GenBank accession no. HUM2C18), CYP2C18-Thr™5 alíele(clone

29c; GenBank accession no. HUMCYP2CI8). and CYP2C19 (clone Ila:GenBank accession no. HUMCYP2C19), each cloned in the EcoRl site ofpBluescripl-SK+/ . were provided by Dr. J. Goldstein (National Institute of

Environmental Health Sciences, Research Triangle Park, NC). The humanCYP cDNAs 2C9 (plasmid 217; Cys144 variant, cloned in the EcoR\ site of

pUVl; GenBank accession no. HUMCYP2C9A). 2B6 (plasmid 328: cloned inthe £coRI site of pUC9; GenBank accession no. HUMCYP2BB), and 3A4(plasmid 359: cloned in the £cv)RIsite of pGEM7zf+; GenBank accession no.HUMCYPNOA) were obtained from Dr. F. Gonzalez (National Cancer Institute. Bethesda, MD). Rat P450 reducíasecDNA was provided by Dr. G. Gil(University of Massachusetts. Worcester, MA). Retroviral vectors of the pBabeseries (36), which encode genes that confer resistance to either puromycin(pBabe-puro) or hygromycin (pBabe-hygro) and are transcribed from an internal SV40 early promoter, were obtained from Dr. B. Speigelman (Dana-

Farber Cancer Institute, Boston, MA). P450 cDNAs were cloned into thepBabe-puro multiple cloning site, which provides for transcription of the P450

gene from the retroviral long terminal repeat promoter. P450 reducíasecDNAwas subcloned into pBabe-hygro cut with EcoRl-Sall. In the final retroviral

construcls, each of ihe human CYP cDNAs retained the following lengths of5'- and 3'-untranslaled region sequences, respectively: 2C8, 78 and 359 bp;2C9, 11 and 369 bp: 2C18-Met3"5. 43 and 341 bp: 2C18-Thr185, 200 and 333

bp; 2C19, 6 and 268 bp; 3A4, 30 and 457 bp; and 2B6, 7 and 1563 bp.Construction of Stable 9L Gliosarcoma Cell Lines by Retroviral Infec

tion. The ecotropic packaging cell line Bosc 23 (Ref. 37; 2.5 X IO6cells in a60-mm dish), obtained from Dr. J. Aster (Brigham and Women's Hospital.

Boston, MA), was cultured in 3 ml of DMEM containing 10% heat-inactivatedfetal bovine serum. 584 g/liter i.-glutamine. 100 units/ml penicillin, and 100

Hg/ml streptomycin. Prior to transfection, the cells were changed to freshmedium containing 25 JUIMchloroquine. Three h later, 24 ptg of each pBabe-

based plasmid were transfected into the cells using calcium phosphate (37).Plasmid DNA was suspended in 0.5 ml of 0.25 M Cad, and then addeddropwise with shaking in a Falcon tube containing 0.5 ml of 2X HEPES-

buffered saline solution [84 mM HEPES (pH 7.1), 547 mM NaCl, 19.8 mM KC1,0.7 mM Na2HPO40-2H2O, and 5.54 mM glucose). The mixture was added

immediately to the cells and left for 5 h before the medium was changed. Thetotal supernatant (3 ml) containing released viral particles was removed 48 hlater and used to infect 9L rat gliosarcoma cells (0.5 X IO6 cells in a 100-mm

dish) in the presence of 4 jig/ml polybrene. Three h later, the medium was

adjusted Io 10 ml by addition of fresh DMEM. The cells were Irypsinized 48 hlater and split into four 100-mm dishes. Selection for puromycin- or hygro-mycin-resislanl 9L cells was carried out with 2 ng/ml puromycin and/or 250/j.g/ml hygromycin for 2-3 days. Drug-resistanl cells were propagated and then

evaluated for CYP or P450 reducíaseenzyme activities and protein expression(see below). Coexpression of P450 enzymes and P450 reducíasewas achievedby infecting cells with retrovirus encoding a P450 gene, puromycin seleclionof a 9L/P450 pool of cells, clonal seleclion of cells that have elevated P450levels (see below), followed by infection of the clonal 9L/P450 cell line withretrovirus encoding P450 reducíase. Hygromycin selection was then carriedout for 2 days to obtain pools of 9L/P450 cells that overexpress P450 reducíase.

Clonal Selection of 9L/P450 Cell Lines. Cells from each pool of puro-mycin-resistant. P450-expressing 9L cells were trypsinized and diluted lo

aboul one cell per 50 pii and then plated al calculated densities of one. three,and five cells per well in a 96-well tissue culture plate. Wells containing singlecolonies were identified when the colonies were nearly confluent (—15days

later) using a light microscope: the cells were then trypsinized and split intotwo wells of a 48-well tissue culture plate. One well of each clone was

untreated and kept as a control, and the second well was treated with either 2mM CPA for 9L/2B6, 9L/2C8, 9L/2C9. 9L/2C18, and 9L/2CI9 cells or with 2mM IFA for 9L/3A4 cells. Clones that exhibited an enhanced sensitivity toCPA or IFA toxicity were typically detected by day 2 or day 3 of drugtreatment, i.e., ~2 days earlier than seen for the others; these clones were then

propagated and further evaluated for P450 protein levels and enzyme activilies.The yield of clones showing enhanced drug sensitivity was 27 of 36 for9L/2B6, 20 of 34 for 9L/3A4, and 4-5 of 34 for the others. In the case of

CYP3A4, each of the isolated clones grew more slowly than the original9L/3A4 pool or the 9L/wt or 9L/pBabe controls. Slow-growing clones were

occasionally seen for some of the 9L/2C isolates, but these were not selectedor evaluated further.

Western Blot Analysis. Microsomes were prepared from near confluent100-mm dishes of each P450-expressing 9L cell line. Cells were washed withice-cold 50 mM KPrl mM EDTA (pH 7.4) and then scraped into 2 ml of thesame buffer and sonicated for a sufficient time to induce cell lysis (~20 s) in15-ml Corex lubes. The homogenate was centrifuged at 12.000 rpm and 4°C

for 20 min, and the resultant supernatant was then ultracentrifuged for 60 minal 45,000 rpm and 4°C.The microsomal pelici was resuspended by sonicalion

in 50 mM KP,, 1 mM EDTA, and 20% glycerol (pH 7.4). Microsomal protein(60 ¿ig/well) was electrophoresed Ihrough a 10% SDS-polyacrylamide gel.

then transferred to nitrocellulose, and further probed with rabbit polyclonalanti-CYP2B6, anti-CYP3A4. and anli-CYP2C anlibodies prepared againslshort COOH-lerminal synthetic peptides (40), which were kindly provided by

Dr. R. Edwards (Royal Postgraduate Medical School, London, United Kingdom). Lymphoblast-expressed CYP2B6 and CYP3A4 (Genlesl Corp..Woburn, MA) and yeasl-expressed CYPs 2C8, 2C9, 2C18, and 2C19 (oblainedfrom Dr. J. Goldstein: Ref. 35) were electrophoresed at 0.2-1.2 pmol P450/

well and used as standards to quantilale P450 levels in the 9L cells.BROD and P450 ReducíaseAssays. BROD activity was assayed using 70

/ng of microsomal protein in 1 ml containing 0.1 M KP| (pH 7.4), O.I mMEDTA, and 4 ¿tM7-benzyloxyresomfin (Molecular Probes. Inc., Junction Cily.

OR). Reactions were started by adding 1 mM NADPH (final concentration),and Ihe release of the fluorescent metabolite resorufin was measured over an8-min time period at 20-25°C using a Shimadzu RF-1501 fluorescence spec-

trophotometer. Fluorescence was read at 550 nm (excitation) and 586 nm(emission). Activily values were quanlitated using resorufin standard. P450reducíase activity was assayed in 0.3 M KP¡buffer (pH 7.7) using 20 figmicrosomal prolein by monitoring the NADPH-dependent reduction of cyto-chrome C at 550 nm al 30°C (E = 21 mM~'cm~'). Under Ihese assay

conditions, 9L cell microsomes exhibited a P450 reducíaseaclivity of 50-60

nmol cylochrome C reduced/min/mg. corresponding to a P450 reducíasespecific contení of 16-19 pmol of P450 reducíase prolein/mg, based on a

reducíase specific aclivily of 40 f¿mol/min/mg and a P450 reducíase M, of78.000 (41).

RT-PCR Analysis of CYP2C18 Expression. Tolal RNA from 9L/P450 or9L/pBabe cells prepared from confluent 100-mm dishes (42) was trealed with

RQ1 DNase (RNase free: Promega). Five /ig of trealed RNA was then healedat 70°Cfor 10 min in 11 jul of DEPC-trealed H2O containing 10 pmol of a

reverse transcriptase primer. Two fj.\ of 10X RT buffer (Promega). 2 ^.1of 0.1

4392

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

M DTT. 40 units of RNasin, and 100 units of Moloney murine leukemia virusreverse transcriptase were then added in a final volume of 20 fi\. Samples wereincubated at 45°Cfor 1 h, followed by heat inactivation at 60°Cfor 2 min.

RT-PCR was carried out using 1 ^il of the reverse transcription reaction.Negative controls for RT-PCR consisted of mock reverse transcription mix

tures, containing all components except for reverse transcriptase.Cytotoxicity Assays. To evaluate the chemosensitivity of the P450-ex-

pressing 9L tumor cells, cells were plated in triplicate at 500-2000 cells/wellof a 48-well plate 18-24 h prior to drug treatment, unless indicated otherwise.Cells were typically treated with 0-2 mM CPA or IFA, continuously for 4 days.

Cells remaining after this time were quantitated using a crystal violet/alcoholextraction assay. Briefly, culture dishes were washed once with PBS, stainedfor IO min with crystal violet [1.25 g crystal violet (Sigma) dissolved in 500ml containing 50 ml of 37% formaldehyde and 450 ml methanol] and thenwashed with PBS twice for 20 min per wash. Absorbance values weremeasured in an SLT Spectra microtiter plate reader using a 595-nm filter.

Background absorbance determined from wells containing culture mediumalone was subtracted from each value. Data are presented as mean cell numberrelative to drug-free controls ±SD for triplicate samples, unless indicated

otherwise. Error bars not seen in the individual figures are too small to bevisible.

Bystander Cytotoxicity/Coculture Experiments. 9L/pBabe cells wereplated in duplicate at IO5 cells/well in the lower chamber of a six-well Falcon

coculture plate (30-mm diameter wells; Falcon 3046). 9L/2B6/reductase cellswere plated in 25-mm cell culture inserts (0.4-fj.m pore size; Falcon 3090), at

dilutions ranging from 0.1 to 6 times the number of 9L/pBabe cells plated inthe lower chamber. Twenty-four h after plating the cells, culture medium was

aspirated from both compartments and then replaced with 2 ml of fresh DMEMin the lower wells and 1 ml containing 2 or 3 mM CPA or IFA in the upper cellculture insert (final drug concentration, 0.67 or 1 mM. as indicated in eachexperiment). Relative cell numbers were quantitated 4 days later by crystalviolet staining.

Semicarbazide Trapping/Fluorescence Assay for 4-Hydroxy Metabolites. 9L/P450 cells were seeded on a 48-well plate in duplicate at 2 X IO4

cells/well. Twenty-four h later, 2 mM CPA or 2 mM IFA was added to the cells

together with 5 HIMsemicarbazide (final concentration) in 1 ml of DMEM (pH7.4) to trap and stabilize the initial 4-hydroxy metabolite. Cells were counted24 h after drug addition, and the culture medium was analyzed for 4-hydroxy-CPA or 4-hydroxy-IFA as described previously (38), as follows. Culturemedium (400 /t!) was removed and then treated with 160 /j.1of ice-cold 5.5%(w/v) ZnSO4, 160 ¿ilof ice-cold saturated Ba(OH)2, and 80 /u.1of ice-cold 0.01

M HC1. The mixture was vortexed and then centrifuged at 16,000 X g for 15min. Three hundred /j.1of the supernatant were transferred under dim light toa clean tube containing 540 ju.1of water and 160 ^.1 of freshly preparedfluorescent reagent (60 mg of 3-aminophenol and 60 mg of hydroxylamine-HC1 per 10 ml of l M HC1). The tubes were vortexed and heated at 90°Cfor

20 min. Fluorescent readings were obtained at 350 nm (excitation) and 515 nm(emission) using 800 p.\ of each sample. Fluorescent values were converted tonmol 4-hydroxy metabolite/ml of original culture medium based on standardcurves generated using 4-hydroperoxy-CPA and then normalized to units ofnmol 4-hydroxy metabolite formed per 24 h per IO6 cells.

Tumor Growth Delay Assay. Six-week-old male ICR/Fox Chase/outbred

immunodeficient seid mice (Ref. 43; Taconic Farms, Germantown, NY)weighing 26-31 g were given s.c. tumor cell injections to form solid tumors.Tumor cells were grown in culture to —¿�75%confluency, harvested by trypsindigestion, washed in PBS, resuspended in fetal bovine serum-free DMEM at aconcentration of 8 X IO6 cells/ml, and kept on ice. Mice were shaved on bothhind flanks and then were injected s.c. on the left flank with 4 X IO6 9L/pBabetumor cells and on the right flank with 4 X IO6 of either 9L/2B6/reductase or

9L/2C18-Met/reductase tumor cells (eight mice/group). Tumor cell injectionswere in a volume of 0.5 ml of serum-free DMEM using a 0.5-inch 27-gaugeneedle and a 1-ml syringe. Tumor growth was monitored twice a week using

Vernier calipers (Manostat Corp., Switzerland) to measure tumor area. On days26 and 27 after tumor cell inoculation, four animals from each group (four9L/pBabe:9L/2B6/reductase mice and four 9UpBabe:9L/2Cl8-Met/reductase

mice) received daily i.p. injections of CPA at 150 mg/kg dissolved in 0.3 mlof 0.9% NaCl. The remaining animals received saline injections. Averagetumor areas ranged from 80 to 145 mm2 at the time of CPA injection. Mice

were given a second course of CPA treatment (150 mg/kg X two daily

4393

injections) beginning 24-29 days after the first CPA injection, as indicated in

each experiment.

RESULTS

Retroviral Expression of Human P450s in 9L GliosarcomaCells. 9L cells were infected with pBabe-based retroviral particles

(36) produced in the packaging cell line Bosc 23 (37) and engineeredto code for each of six human P450 genes: CYPs 2B6, 2C8, 2C9,2CÌ8,2CI9, and 3A4. The corresponding P450 proteins have eachbeen shown to activate CPA or IFA in a heterologous cDNA expression system (24, 35). Included in these studies are 2C18-Met385 and2C18-Thr385, two allelic variants of CYP2C18 that display differences

in their oxazphosphorine metabolism activities (35). The retroviralapproach to stable transfection used in these studies is rapid and yieldsa pool of cells containing many thousands of independent clones, eachof which expresses the P450 gene at a random integration site. In eachcase, s60-70% of the infected 9L tumor cells acquired resistance to

puromycin, indicating a high efficiency for retroviral gene delivery.P450 protein expression was readily detectable in the pools ofCYP2B6 and 2C9 cells, as shown by Western blot analysis of isolatedcellular microsomes (Fig. \A and data not shown). In the other cases.

A.

9L cell lines

! 2B6

B.

9L cell lines

0.5 1.0 2C18 0.5 0.5

1 8 9 10Fig. 1. Western blot analysis of CYP protein levels in 9L/P450 cell lines. Shown are

Western blots of 9L/wt and 9L/P450 cell microsomes (60 fig/lane) probed with antibodylo P450 2B6 (A) or P450 2C (B). Shown in A, Lane Ì,are 9L microsomes prepared fromthe original pool of 2B6 transduced cells, prior to selection of the 2B6 clone shown inLane 2. cDNA-expressed P450 2B6 (A. Lanes 4 and 5; lymphoblasl expression) and P4502C standards (B, Lanes I. 2, 9, and 10; yeast expression) were analyzed in parallel at theindicated number of pmol P450 protein per well. Very similar band intensities wereobtained for CYP2Cs expressed in lymphoblasts (Gentest). The anti-CYP2C COOH-terminal antibody used in this study (anti-FIPV-COOH) consistently gave strong Westernblot signals with cDNA-expressed 2C8 and somewhat lower signals with 2C9 compared

with 2C18 and 2C19.

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

Fig. 2. BROD activity in 9L/P450 cell lines.Shown are BROD activity values measured in mi-

crosomes prepared from the original retroviralP450-infected 9L cell pools (•)or from individualclonal cell lines (-/. -2. and -3\ C3). which wereidentified on the basis of their enhanced sensitivityto CPA or IFA (see text). Arrows, individual clonesof each 9L/P450 pool selected for subsequent study.

?1000-

e

i

~ 500u

oets

1.n1

só^ó^oÓMXjyovdv^HHœa:ÎNCjOL,tNUuSwwrs e* M es »M n rí 5 S

u S SM o u

é^ÉuéééS^s- —¿�r- " r i r i r i J

H y

9L/P450 cell lines

the level of P450 protein expression in the pool of puromycin-

resistant cells was low (2C8 and 2C19) or not detectable (CYP3A4and both CYP2C18 alíeles).Clonal selection was therefore carriedout to obtain individual 9L/P450 clones with higher levels of P450protein expression. Clones showing enhanced sensitivity to CPA orIFA (see "Materials and Methods") were propagated and analyzed

by Western blotting. In each case, the 9L/P450 clones selected onthe basis of increased sensitivity to CPA or IFA had a higherspecific P450 protein content (Fig. 1 and data not shown) andhigher P450 enzyme activity (Fig. 2). The microsomal P450 content of the selected clones, determined by Western blotting usinglymphoblast-expressed or yeast-expressed human P450s as stand

ards, is presented in Table 1. The highest levels of expression wereobserved with 9L/2B6 (20-25 pmol of P450/mg of microsomalprotein) and 9L/2C9 (10-15 pmol/mg) cells, whereas 2C18-Metwas the lowest (<0.3 pmol/mg). 9L/2C8, 9L/2C18-Thr, 9L/2C19,

and 9L/3A4 cells exhibited intermediate levels of P450 expression(0.5-3 pmol/mg). Expression of both 2C18 transcripts was confirmed by RT-PCR (data not shown). These expressed P450 protein

levels can be compared to an endogenous P450 reducíase proteinlevel of ~ 16-19 pmol of P450 reducíase protein/mg of 9L microsomal protein (see "Materials and Methods").

BROD Activity of the Tumor Cell-expressed P450 Genes.P450-dependent BROD activity, which is catalyzed by many hu

man P450 enzymes at various rates, was used to monitor the

Table 1 P450 protein conteníin 9UP450 clonal cell lines"

9L/P450 clonalcellline2B62C82C92C18-Met2C1

8-Thr2C193A4Clone

no.1213223Specific

P450 contení(pmol of P450/mg

of microsomalprotein)2(1-250.5-110-15£0.30.5-11.5-31-2

" Individual P450 clones were selected from pools of puromycin-resistanl 9L/P450

cells on the basis of their enhanced sensitivity lo CPA (or to IFA, in the case of 9L/3A4),as described in "Materials and Methods." Microsomes prepared from the indicated clones

were analyzed on Western blots (as in Fig. I ) for P450 protein content in comparison loa standard curve based on 0.2-1 pmol of the corresponding cDN A-expressed P450 proteinstandard using either a lymphoblast cDNA-expression system (CYPs 2B6 and 3A4;

Genlesl Corp.) or a yeast expression system (35). Data are shown as a range of values fromtwo or three separate experiments.

enzymatic activity of the transfected P450 genes. Fig. 2 shows thatall of the P450-expressing cell lines exhibited higher BROD activity than parental (wild-type) 9L cells (9L/wt) or 9L/pBabe

controls. Moreover, the isolated 9L/P450 clones (Fig. 2, D) eachexpressed higher BROD activity than the original 9L/P450 retro-

viral pools (Fig. 2, •¿�).9L/P450 clonal cell lines chosen for furtherstudy (Fig. 2, arrows) were selected on the basis of their level ofP450 expression (Western blotting and BROD activity) and theirsensitivity to CPA or IFA.

Cytotoxicity of CPA and IFA toward P450-expressing 9L

Tumor Cells: Growth Inhibition Assays. To evaluate the impactof retroviral P450 transduction on 9L chemosensitivity, cells werecultured in the presence of various concentrations of CPA or IFA,and cytotoxicity was evaluated by a growth inhibition assay scored4 days later. 9L/wt and 9L/pBabe cell lines, used as P450-negative

controls, were insensitive to millimolar concentrations of bothCPA (Fig. 3A) and IFA (Fig. 3C). By contrast, all of the P450-expressing 9L cells showed a concentration-dependent growth

inhibition by CPA. 9L/2B6 cells were the most susceptible to CPAcytotoxicity (s95% growth inhibition at 1 mM CPA), consistentwith the high catalytic activity of CYP2B6 with respect to CPAactivation (24)4 and the comparatively high level of CYP2B6

protein expression achieved using this retroviral expression system(Table 1). Each of the other tumor cell lines, except 9L/2C8, alsoshowed significant acquired drug sensitivity: at 1 mM CPA, ~60%growth inhibition was observed for 9L/2C9 and 9L/2C18-Thr cells,75-85% inhibition was obtained for 9L/2C19 and 9L/2C18-Met(Fig. 3ß),and ~90% growth inhibition was observed for 9L/3A4(Fig. 3A). The cytotoxicity of CPA toward 9L/2C18-Met cells isespecially remarkable, given the very low level of CYP2C18-Met

protein expression in these tumor cells (Table 1). This effect islikely due to the low Km and high catalytic efficiency CYP2C18-

Met for CPA (35). In contrast to the moderate to high sensitivity ofmany of the 9L/P450 cell lines to CPA, only three of the 9L/CYPcell lines showed significant sensitivity toward IFA when tested atconcentrations up to 1 mM. 9L/3A4 cells were, by far, the mostsensitive to IFA, although significant IFA growth inhibition wasalso achieved in 9L/2B6 and 9L/2C18-Met cells (Fig. 3, C and D).

4 P. Roy. L. J. Yu. C. L. Crespi, and D. J. Waxman. unpublished data.

4394

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

CFA Cytotoxicity IFA Cytotoxicity

Fig. 3. Growth inhibition assay to assess chemo-

sensitization to CPA (A and B) and IFA (C and D)in 9L/P450 cells. Cells were seeded at 500 cells/well in 48-well plates and were treated with the

indicated concentrations of CPA or IFA for 4 days.Control cells for each cell line received no drugtreatment. Relative cell number at the end of theexperiment was determined by crystal violet staining as described under "Materials and Methods."

Data points, means (n = 3) based on the crystalviolet absorbance in drug-treated plates as a percentof the corresponding drug-free controls; bars, SD.Data for the P450-defkient control cell lines (9L/WIand 9L/pBabe) shown in A and C are the same asshown in B and D.

0 0.25 0.5 0.75 1

0.25 0.5 0.75 1

CPA (mM)

0.25 0.5 0.75 1

IFA (mM)

The other P450-expressing cell lines were insensitive to IFA under

these conditions (Fig. 3D).P450 Redactase Transduction Further Enhances Cytotoxic Re

sponses. We next investigated whether the endogenous level ofP450 reducíase in 9L tumor cells is sufficient to maximally supportCPA and IFA activation catalyzed by the transduced human P450genes. P450 reducíase activiiies measured in microsomes preparedfrom Ihe 9L/P450 cell lines were, in general, noi subslanliallydifferenl from 9L/wl and 9L/pBabe controls (Fig. 4). An ~5-fold

increase in P450 reducíase aclivily was, however, oblained following relroviral iransduclion of Ihe 9L/2B6 and 9L/2C18-Mel cells

wilh P450 reducíase (Fig. 4). This overexpression of P450 reducíasemarkedly increased the cyloloxicily of CPA toward the 9L/2B6 and 9L/2C18-Mel cells, particularly al lower drug concenlra-lions (e.g., growlh of 2B6-expressing cells decreased from 75% lo

28% of conlrol al 0.2 mM CPA upon coexpression of P450 reducíase;Fig. 5A). Similar effecls of P450 reducíase overexpressionwere observed wilh IFA as subslrale (Fig. 5ß).These effecls were

achieved, allhough endogenous P450 reducíase levels in ihese cells(16-19 pmol of P450 reductase/mg of microsomal protein; see"Materials and Meihods") are comparable lo (9L/2B6 cells) or in

vasi molar excess of (9L/2C18-Mel cells) Ihe expressed P450

prolein levels (compare Table 1). In conlrol experimenls, iransduclion of P450 reducíase Iransduction alone, in Ihe absence ofP450 (Fig. 4; 9L/pBabe/Red), did noi confer CPA or IFA cylotox-icily (Fig. 5). Thus, P450 reducta.se gene iransfer enhances ihera-

peulic aclivily by increasing lumor cell cyloloxicity al a given drugconcenlration. Il is also apparenl from ihese sludies lhal 9L growlhinhibilion required higher concentralions of IFA lhan CPA for bolhCYP2B6 and CYP2C18-Mel (Fig. 5, fi versus A). This is, in part,

a reflection of the greater intrinsic sensitivily of 9L lumor cells loaclivated CPA, as compared lo aclivaled IFA, as delermined incyloloxicily assays using Ihe corresponding chemically activaled4-hydroperoxy compounds (Fig. 6). In ihis lighl. Ihe subsiantially

greaier sensiiiviiy of 9L/3A4 cells to IFA compared to CPA(Fig. 3, C versus A) musi, indeed, reflecl Ihe higher rate of

4395

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

M

> 'S

s E"

u ï» SII

o 100-

I

9L/P450 cell linesFig. 4. P450 reducíaseactivity in 9L/P450 cell lines. Microsomes prepared from each

of the indicated cell lines were assayed for P450 reducíase activity (cytoehrome creduction) as described under "Materials and Methods." Red. cell lines transduced with

P450 reducíase.Columns, means based on n = 3 determinations: bars. SD.

prodrug activation catalyzed by CYP3A4 with IFA compared toCPA (24).4

Metabolic Activation of CPA and IFA Enhanced by P450 Re-ductase Transduction. To investigate whether the P450 reductase-enhanced cytotoxicity of CPA and IFA toward 9L/2B6 and 9L/2C18-

Met tumor cells is due to increased prodrug activation, we measuredthe levels of P450-generated 4-hydroxy metabolites in culture super-

natants from each cell line (Fig. 7). P450 reducíase gene transferstimulated a significant increase in cytotoxic metabolites accumulat

ing in the culture medium, both for CYP2B6 and for CYP2C18-Met

cells. In contrast, transduction of P450 reducíasein 9L parental cellsdid not stimulate prodrug activation. These studies also showed that,in 9L/3A4 cells, IFA 4-hydroxylation was greater than CPA 4-

hydroxylation, consistent with their preferential sensitivity to IFA(Fig. 3). Moreover, 9L/2C8 cells, which are insensitive to IFA (Fig.3D), showed IFA metabolite levels just above the 9L/pBabe background (Fig. 7). Some quantitative discrepancies were observed, however. These include the similar level of activated CPA and IFAmetabolites formed in 9L/2B6/reductase culture medium (Fig. 7)versus the higher level of CPA compared to IFA activation catalyzedby cDNA-expressed CYP2B6 (24), and the similar levels of activatedmetabolites formed by 9L/2B6/reductase compared to 9L/2C18-Met/

reducíase cells (Fig. 7) versus the grealer drug sensilivily of ihe2B6-expressing cells (Figs. 3 and 5). These discrepancies may relate

lo ihe relalive insensilivily of ihe fluorescence meiabolile assay and loihe chemical inslabilily of ihe aclivaied CPA and IFA meiabolilesduring ihe 24-h lissue cullure incubalion in the studies described in

Fig. 7. Finally, the grealer inlrinsic sensilivity of 9L tumor cells loactivated CPA (Fig. 6) is likely lo account for ihe grealer sensilivilyof Ihe 2C18-Mel-expressing cells lo CPA compared lo IFA (Fig. 5),

despite iheir formation of aclivaied CPA and IFA al similar levels(Fig. 7).

Bystander Killing Effect. CPA irealmenl of lumor cells transduced with ral P450 gene CYP2BI is associated wilh a significanlbyslander cyloloxic effecl, whereby neighboring lumor cells lhat donoi express ihe P450 drug susceplibilily gene also become sensilizedlo ihe prodrug (29, 31). Because 9L/2B6/reduclase cells are themselves more readily killed by CPA and IFA lhan 9L/2B6 cells (Fig. 5),we investigated whether bystander tumor cells are also more chemo-sensilive when culiured with P450 reduclase-transduced 9L/2B6 cells.

Byslander cyloloxicity was evaluated by using cell cullure inserts tophysically separate ihe prodrug-activaling 9L/2B6/reduclase (or 9L/

0.5 1 1.5

CPA (mM)0.5 1 1.5

IFA (mM)Fig. 5. Retroviral transduction of P450 reducíasegene enhances P450-mediated CPA and IFA cytotoxicity. 9L/pBabe cells and 9L cells transduced with P450 and/or P450 reducíase

(Red) were seeded at 4000 cells/well in 48-well plates and treated with increasing concentrations of CPA (A) or IFA (fi) for 4 days. Cell growth in comparison to drug-free controlswas determined by crystal violet staining. Data points, means for n = 3 replícales;bars, SD.

4396

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

2B6) tumor cells from the P450-deficient bystander 9L/pBabe cells.

Fig. 8A demonstrates that 9L/2B6/reductase cells exposed to CPAconfer a ~4-fold stronger bystander killing of 9L/pBabe cells than do

9L/2B6 cells, as judged from the 9L/P450:9L/pBabe cell ratio required to effect 50% bystander cytotoxicity. Follow-up experiments

demonstrated that IFA does not display the strong bystander cytotoxicity seen with CPA: CPA treatment of 9L/2B6/reductase cells resultsin —¿�40%bystander toxicity toward non-P450 tumor cells when theP450:non-P450 cell ratio is -0.5:1 and >80% toxicity toward thenon-P450 cells as the cell ratio is increased to 23 (Fig. 8ß).In

contrast, IFA exerted only a modest bystander cytotoxicity to thenon-P450-expressing pBabe controls (up to —¿�20%bystander cytotoxicity at cell ratios up to 3:1 and a maximum of —¿�50%bystander

cytotoxicity at a cell ratio of 6:1: Fig. SB). This observation, togetherwith the greater intrinsic potency of activated CPA compared toactivated IFA in the 9L gliosarcoma model (Fig. 6), demonstrates thatCPA is a superior choice compared to IFA for use in conjunction withCYP2B6 plus P450 reducíasegene therapy.

Evaluation of Human P450-based Gene Therapy in a seid

Mouse Model. The impact of human P450 gene transfer on 9Lgliosarcoma chemosensitivity was evaluated in vivo using the immu-

nodeficient seid mouse solid tumor model (43). The seid mouse modelis free from the complications associated with immunogenic responses toward 9L tumors expressing human P450 genes, and iteliminates any contributions from immunological components to bystander cytotoxicity (13). seid mice were inoculated s.c. with 9L/pBabe tumors on the left flank and with 9L/2B6/reductase tumors onthe right flank. This experimental design enabled us to control for anysystemic effects that the P450-expressing tumor might have on liver

P450 metabolism. In the absence of drug treatment, 9L/2B6/reductasetumors grew at a somewhat slower rate than 9L/pBabe tumors, as hadbeen observed for the cells in culture (Fig. 9, B versus A). Mice weretreated with CPA given as two daily i.p. injections at 150 mg/kg bodyweight on days 26 and 27 after tumor implantation, at which time thetumors were well established in all of the mice. This initial round ofdrug treatment resulted in no growth delay for the 9L/pBabe tumorscompared to saline-treated controls (Fig. 9A, —¿�CPAversus +CPA). In

contrast, growth of the 9L/2B6/reductase tumors in the same mice wasfully blocked by CPA treatment (Fig. 9B). Administration of a secondcycle of CPA treatment —¿�3.5weeks after the first cycle (Fig. 9,

arrows) effected a modest growth delay effect against 9L/pBabecontrol tumors, whereas it prolonged the strong antitumor effect seenin the case of 9L/2B6/reductase tumors until at least day 75 aftertumor inoculation.

In an experiment designed to evaluate the impact of CYP2C18-Met

gene therapy, CPA treatment effected a modest growth delay effect(—5-6 days) toward the 9L/pBabe control tumors, which likely is aconsequence of liver P450-catalyzed drug activation (Fig. 9Q. In

contrast, a significant, albeit incomplete growth inhibition of the9L/2C18-Met/reductase tumors was observed (growth delay of at least

25 days; Fig. 9D). This effect is especially striking when taken in thecontext of the very low level of P450 protein expression in the9L/2C18-Met tumor cells (Table 1).

DISCUSSION

The primary goals of this study were: (a) to extend earlier preclin-

ical cancer gene therapy studies based on the rat CYP2B1 gene; (b) toidentify human P450 genes that serve as suitable candidates forP450-based prodrug activation/cancer gene therapy: (c) to assess the

extent to which an improvement in the efficiency of intratumoral drugactivation catalyzed by human P450s can be achieved by cotransfer ofthe P450 redactase gene; and (d) to determine whether CPA and IFA

oO

u

I

100-0

75-

50-

25-

4-OOHDrug

Fig. 6. Intrinsic chemosensitivity of 9L/WI cells to chemically activated CPA and IFA.9L/WI cells seeded at 2000 cells/well in 48-well plates were treated for 3-4 h withincreasing concentrations of 4-hydroperoxy-CPA or 4-hydroxyperoxy-IFA (0-100 /IM).Cells were then incubated in fresh drug-free medium and allowed to grow for 4 days. Cellgrowth as a percentage of drug-free controls was then determined by crystal violetstaining. Data points, means range for duplicate samples: bars, ranges.

both exhibit bystander cytotoxic effects when activated within tumorcells by human P450 enzymes. Transduction of the rat gliosarcomacell line 9L with replication-defective retroviral particles encoding

each of six human P450 genes was found to chemosensitize the tumorcells to the cytotoxic effects of CPA, albeit with different apparentefficiencies (2B6 > 2C18-Met - 3A4 > others). In the case of IFA,

CYP3A4 was the most effective in chemosensitizing the target tumorcells, followed by CYPs 2B6 and 2C18-Met. A substantial further

improvement in chemosensitivity toward CPA and IFA was obtainedby transduction of P450-expressing tumor cells with the P450 redué

lase gene. This led to a striking growth delay (>50 days) followingCPA treatment of 2B6- plus P450 reductase-transduced tumors grown

in vivo in seid mice, in which complications associated with immunerejection (44) and immunological contributions to bystander cytotoxicity (13) are avoided. A significant, albeit less dramatic, CPA growthdelay (>25 days) was seen with 9L/2C18-Met/reductase tumors,

despite their very low P450 expression level. The human CYP genes2B6 and 2C18-Met are, thus, strong candidates for further preclinicalevaluation for CPA-based P450 gene therapy.

The six human CYP genes investigated in this study were expressedin the transduced clonal 9L tumor cell lines at widely different proteinlevels (sO.3 to 20-25 pmol of P450 protein/mg of microsomal

protein). Although P450 protein levels and CPA cytotoxicity weresomewhat lower in the corresponding initial pools of transducedtumor cells, the rank order of P450 protein expression level wassimilar (2B6 > 2C9 > 2C8 ~ 2C19 > 2C18 and 3A4; data not

shown). This suggests that the large differences in P450 protein levelsin the isolated clonal cell lines do not primarily reflect clonal differences in the site of retroviral integration. Rather, because all six P450genes were transcribed from the same long terminal repeat retroviralpromoter, the observed differences in P450 protein expression levelsare more likely to be due to differences in the stabilities of theindividual P450 proteins and their mRNAs. Conceivably, some of thedifferences in expression could also reflect differential toxic effectsthat some of the P450 forms may have on the gliosarcoma recipientcell. Of note, clonal 9L cell lines selected for elevated expression ofCYP3A4 consistently exhibited slower growth rates in cell culturecompared to the initial pool of retrovirally infected cells (data notshown). CYP2B6 was expressed at the highest level, and correspond-

4397

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

(Table 1) and given the intrinsic lower cytotoxicity of activated IFAtoward 9L tumor cells compared to activated CPA (Fig. 6). In the caseof CYP3A4, the lower level of tumor cell P450 protein may, in part,reflect toxicity of the CYP3A4 gene product to the 9L cells, assuggested by the slower growth rate that was repeatedly obtainedupon retroviral transduction with CYP3A4, both in retroviral poolsand in individual clones, as noted above. This apparently toxicity ofCYP3A4 might be circumvented by use of an inducible promoter (49,50), thereby augmenting intratumoral expression of CYP3A4 in thecontext of IFA-based gene therapy. However, CYP3A4 is not likely to

exhibit toxicity toward all tumor cells because slower cell growth hasnot been described in other mammalian CYP3A4 expression systems(51).

Although 4-hydroxy-CPA and 4-hydroxy-IFA yield distinct DNA-cross-links (five versus seven atoms in the cross-link, respectively),

the two drugs are presumed to act via similar mechanisms in elicitingapoptosis. Nevertheless, 9L gliosarcoma cells were found to be ~3-

fold more sensitive to activated CPA compared to activated IFA. This

O

•¿�9G

9L Tumor Cell lines

Fig. 7. Scmicarbazide trapping/fluorescence assay for 4-hydroxy metabolites in cellculture. Each of the indicated 9L cell lines was seeded on a 48-well plate in duplicate at2 X 10J cells/well. Twenty-four h later. 2 ITIMCPA (Q) or 2 min 1FA (•)was added to

the colls together with 5 mw semicarbazide (final concentration) in a volume of 1 ml. Theculture medium was analyzed for 4-hydroxy-CPA or 4-hydroxy-IFA 24 h after drug

addition. Data shown is normalized to the total cell number in each well, as described in"Materials and Methods." Background fluorescence observed in culture medium from

9L/Babe cell controls was subtracted from each sample.

ingly, CYP2B6 conferred on the 9L tumor cells the greatest CPAsensitivity. CYP2B6 is thus the gene of choice for CPA-based

cancer gene therapy, both in terms of the apparent greater protein/mRNA stability suggested for CYP2B6 by the current study andfor its high inherent catalytic activity for CPA 4-hydroxylation andits low rate of CPA deactivation by /V-dechloroethylation.4 How

ever, CYP2C18-Met may also be a useful candidate for CPA-based

cancer gene therapy, given the high chemosensitization achieved inCYP2C18-Met-expressing cells in the context of very low ex

pressed P450 protein levels. Higher levels of CYP2C18 proteinexpression may be achievable using improved retroviral or othervectors, including nonviral vectors for gene delivery presentlybeing developed (45-48).

In contrast to the effective intratumoral activation of CPA by abroad range of P450s, intratumoral IFA activation was catalyzed by amore restricted subset of the human P450 genes investigated. Asanticipated from studies using expressed P450 cDNAs and humanliver microsomes (24, 35), CYP3A4 was the most potent activator ofIFA, followed by CYP2B6 and CYP2C18-Met, the only CYP2C

subfamily member able to sensitize 9L tumor cells toward IFA at 1mM. Some IFA chemosensitivity was observed with the other CYP2Cs at higher IFA concentrations (data not shown). The chemosensitization to IFA conferred by CYP3A4 (Fig. 3Q is striking, given themuch lower level of expression of this P450 compared to CYP2B6

4398

25-

12345

Cell Ratio (2B6/Red:pBabe at 2B6:pBabe)

123456

Cell Ratio (2B6/Red:pBabe)

Fig. 8. Bystander cytotoxicity of 9L/2B6/reductase and 9L/2B6 cells. A, bystandercytotoxicity of 9L/2B6/reductase cells is compared to that of 9L/2B6 cells. In both cases,9L/pBabe served as the bystander cells. 9L/pBabe cells were plated in duplicate at IO5cells/well of a 6-well plate (lower culture chamber). 9L/2B6/reductase cells or 9L/2B6cells were plated in 25-mm cell culture inserts (upper chamber) at concentrations rangingfrom IO4 to 6 X IO5 cells, corresponding to ratios of 0.1 to 6 relative to the bystander

9UpBabe cells growing in the lower chamber, as shown on the X axis. Cells wereincubated in a total volume of 3 ml DMEM. 10% FBS. containing 1 mM CPA. final CPAconcentration. B, bystander cylotoxicity of 9L/2B6/reductase cells toward 9L/pBabe cellswith CPA compared to the bystander cytotoxicity with IFA as prodrug. The experimentaldesign was the same as in A, except that the final drug concentration was 0.67 mM. Datashown for both panels correspond to the survival of 9L/pBabe bystander cells in the lowerchamber.

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Fig. 9. Tumor growth delay assay in sciti micebearing 9L/pBabe. 9L/2B6/reductase and 9L/2CI8-Met/reductase solid tumors. Shown are the effectsof CPA treatment on growth of 9L/pBabe tumorsgrowing on the left flank of sciti mice that also beareither a 9L/2B6/reductase tumor (A) or a 9L/2C18-

Met/reductase tumor (O on the right flank. Alsoshown are the effects of CPA on growth of 9L/2B6/reducíase(ß)or 9L/2C18-Met/reductase tumors (D)

in these same mice. Tumor areas were measuredtwice a week with Vernier calipers. Arrows (Xarij), days on which CPA was given by i.p. injection at 150 mg/kg. as described under "Materialsand Methods." Data points, mean tumor areas(mm2: «= 4 tumors per treatment group): bars. SE.

Mice not treated with CPA died earlier than theCPA-treated mice, as indicated by the earlier decrease in number of remaining mice from n = 4(initial number of mice/group) to n = 3, 2, or 1. O,CPA-treated mice; •¿�,saline controls.

CYP2B6-BASED CANCER GENE THERAPY

1200n

N

E

CO<D

oE3

A. 9UpBabe . .CPA o +CPA 1200

1000

800

600

400

200

15 25 35 45 55 65 750

B. 9U2B6/RED

-CPA

_U15 25 35 45 55 65 75

1200

1000

800

600

400-I

200

0

C. 9L/pBabe

n=1-CPA

1200

1000

800

600

400

200

0

D. 9L/2C18-Met/RED-CPA

n=4 +CPA

U_15 25 35 45 55 65 15 25 35 45 55 65

Days After Tumor Cell Injection

differential cytotoxic effect was most evident at lower drug concentrations, which may be particularly relevant to the clinical situation.IFA is known to be activated by hepatic P450s less efficiently thanCPA and is also more extensively deactivated by the W-dechloroethy-lation pathway (22, 25, 52). Although the A/-dechloroethylation by

product chloroacetaldehyde is generally not considered to be anactive, chemotherapeutic metabolite when it is generated via hepaticP450 metabolism, it may nevertheless contribute to IFA cytotoxicitywhen it is generated within the target tumor cell in situ, in the contextof P450-based cancer gene therapy. This possibility is suggested forCYP2B6-expressing tumor cells, which were sensitized to both CPA

and IFA cytotoxicity, despite the fact that CYP2B6 metabolizes IFAprimarily via the A'-dechloroethylation pathway to yield chloroacet

aldehyde rather than by 4-hydroxylation, the major CYP2B6-cata-lyzed metabolic route for CPA.4 Important differences between

CYP2B6-catalyzed CPA and IFA metabolism were also evident from

coculture experiments, in which CPA could mediate a striking bystander cytotoxic effect but IFA could not. This intriguing findingmay relate to the lower intrinsic cytotoxicity of 4-hydroxy-IFA compared to 4-hydroxy-CPA in this tumor cell line, coupled with the factthat CYP2B6 primarily metabolizes IFA via an jV-dechloroethylation

pathway that generates chloroacetaldehyde, which likely contributesto the 9L/2B6 cell's chemosensitivity in these studies but may confer

little or no bystander cytotoxicity. Because the bystander effect provides an important mechanism to enhance the effectiveness of prodrugactivation-based cancer gene therapy by compensating for gene transfer efficiencies that may be well below 100% (5), IFA-based P450

gene therapy is not likely to be as effective using CYP2B6 and will

require a more active catalyst of IFA 4-hydroxylation, such as

CYP3A4.Coexpression of rat P450 reducíase with rat CYP2B1 leads to

enhanced CPA and IFA chemosensitivity (38). This finding wasunexpected, in view of the comparatively high level of endogenousP450 reducíasein 9L gliosarcoma cells and the low absolute level ofP450 protein that is expressed. Large increases in tumor cell sensitivity to CPA and IFA were also seen in this study following trans-duction of P450 reducíase,not only with CYP2B6-expressing tumorcells but also with 9L/CYP2C18-Met cells, which have a very lowexpressed P450 protein level (Table 1), corresponding to a >50-fold

molar excess of P450 reducíase.The increased chemosensitivity of9L/P450 cells transduced with a P450 reducíasegene was paralleledby an enhanced accumulation of active metabolites for both CPA andIFA (Fig. 7) and by a substantial increase in bystander cyloloxicily(Fig. 8/4). P450 reducíase gene Iransfer in the absence of P450expression did not confer any chemosensitivity to CPA or IFA,confirming the P450 dependence of these enhanced cytotoxic responses. Given the clear advantages of transducing lumor cells wilhP450 reducíasein combination with P450, bioreductive prodrugs thatare activated by P450 reducíaseunder hypoxic conditions may readilybe added lo a combinalion chemolherapy/gene iherapy regimen thatincludes a P450-activated prodrug. Examples of such bioreductiveprodrugs include mitomycin C and various other quiñones,nitroimi-dazoles, heterocyclic A'-oxides, and bioreducible DNA alkylators (53,

54). Although this strategy might seem counterintuitive, given therequirement of O2 for P450-catalyzed monoxygenase reactions, receñísludies have shown no significant loss of P450-dependent prodrug

4399

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP2B6-BASED CANCER GENE THERAPY

activation in P450-transduced tumor cells grown under hypoxicconditions.5 This, in turn, suggests that hypoxia response elements

(55) may be useful for transcriptional targeting (56) of P450 incombination with P450 reducíaseto solid tumors, which are characterized by a localized hypoxic environment (57).

Of the four human CYP2C gene products, CYP2C18 is the mostactive catalyst of CPA and IFA 4-hydroxylation (35), and in thisstudy, CYP2C18 was the most active in chemosensitizing 9L gliosar-

coma cells to both CPA and IFA. Of the two CYP2C18 alíelesstudied, CYP2C18-Met was particularly active, despite its very lowlevel of expression, as noted above. Indeed, CYP2C18-Met mediateda 9L cell killing effect that is comparable to that seen in CYP3A4-expressing tumor cells, in which the P450 protein level was several-

fold higher. This observation has important implications both for thepotential utility of CYP2C18-Met for CPA-based cancer gene therapy

and with respect to its role in CPA and IFA activation in human livertissue. Whereas CYP3A4 is the most abundant P450 in a typicalhuman liver sample, CYP2C18 is in much lower abundance (58, 59).Although CYP2C18 is an active CPA and IFA 4-hydroxylase when

assayed in a yeast cDNA expression system (35), the proposed directinvolvement of this or other specific human CYP 2Cs in oxazaphos-

phorine metabolism in human liver (24, 60) cannot be clearly established, owing to the high cross-reactivity among the human P450 2Cproteins and the absence of P450 2C form-specific inhibitory probes.

Nevertheless, the present demonstration that all four CYP 2C members can sensitize 9L tumor cells to CPA provides further support forthe proposal that, in aggregate, these enzymes contribute significantlyto the metabolism of CPA in human liver tissue. Interestingly,CYP2C18-Thr, though just detectable when retrovirally expressed in

9L tumor cells, conferred sensitivity to CPA, but not to IFA up to 1mM. The two CYP2CÌ8allelic variants differ at a single amino acid(Met versus Thr385) and exhibit clear kinetic differences using CPA

and IFA as substrates (35), implicating residue 385 as an importantdeterminant of the specificity of CYP2C18 toward these isomerieprodrug substrates. Clinically, this observation could be importantwhen using IFA, in view of the much more rapid activation of IFA byCYP2C18-Met compared to CYP2C18-Thr, and given that the 2C18-Met alíeleoccurs in ~30% of the Caucasian population, whereas the2C18-Thr alíeleis more abundant, present in —¿�70%of the population

(61).CYP2C9 is the most abundant CYP 2C enzyme expressed in human

liver (58) and can activate CPA, and at a 3-fold lower catalyticefficiency, IFA in a yeast expression system (CYP2C9-Cys144 alíele;

Ref. 35). The CYP2C9-selective inhibitor sulfaphenazole has been

reported to slow the elimination of CPA in a subset of cancer patients(62), suggesting a role for CYP2C9 in hepatic metabolism of thisdrug. Consistent with these observations is our finding that CYP2C9can sensitize 9L tumor cells to CPA, albeit with much lower efficiency than CYP2C18 and CYP2C19, particularly when the several-

fold higher level 9L CYP2C9 protein expression is taken into account.Although CYP2C9 is, thus, not the gene of choice for CPA orIFA-based P450 gene therapy, this and other P450 genes (28) may

prove useful when combined with other cancer chemotherapeuticprodrugs (2).

The use of retroviruses and other viral vectors for delivery oftherapeutic genes to tumor cells has become increasingly feasible withrecent improvements in vector design (45, 63, 64). Nonviral vectors,including cationic liposomes, DNA-protein complexes, nonviral T7

autogene vectors, and other approaches (47, 48, 65) have also undergone significant development and improvement in recent years. Fur-

5 Y. Jounaidi and D. J. Waxman, unpublished data.

ther enhancement of P450/P450 reducíase gene delivery may beachieved by linking these genes using an internal ribosome entry sitesequence (66) to achieve iheir coordinate expression on a bicislronicmessage. Alternatively, construclion of a fusion gene which encodesa calalytically active P450-P450 reducíase fusion protein (67) may

allow for highly efficient expression of P450 activity. Limitationsassociated with comparatively low retroviral liters could, in part, beovercome by ihe use of more powerful promoters lo ensure a highlevel of gene expression, coupled with the use of drugs, such as CPA,that exhibit a sirong bystander cytotoxic effect. Innovative strategiesto achieve tumor-selective gene expression, including transcriptional

targeling, cellular targeting, and selective delivery in situ of both theprodrug and the therapeutic gene have been described (68-70). Further development of CYP2B6, CYP2C18-Met, and other P450 genes

for prodrug activation cancer gene therapy will be directed at theimplementation of these and other approaches to enhance the enzy-

malic activity, tumor cell chemosensitivity, and targeling specificilyof P450/P450-reductase-based gene therapy.

ACKNOWLEDGMENTS

The authors thank Dr. Joyce Goldstein (National Institute of EnvironmentalHealth Sciences, Research Triangle Park, NC) and Dr. Frank Gonzalez (National Cancer Institute, Bethesda, MD) for provision of human P450 cDNAsand Dr. Robert Edwards (Royal Postgraduate Medical School, London, UnitedKingdom) for anti-P450 antibodies.

REFERENCES

1. Kivisto. K. T., Kroemer. H. K.. and Eichelbaum. M. The role of human cylochromeP450 enzymes in the metabolism of anticancer agents: implications for drug interactions. Br. J. Clin. Pharmacol., 40: 523-530, 1995.

2. LeBlanc. G. A., and Waxman, D. J. Interaction of anticancer drugs with hepaticmonooxygenase enzymes. Drug Metab. Rev., 20: 395-439. 1989.

3. Moolten, F. L. Drug sensitivity ("suicide") genes for selective cancer chemotherapy.Cancer Gene Ther., /.- 279-287, 1994.

4. Deonarain. M. P.. Spooner. R. A., and Epenetos, A. A. Genetic delivery of enzymesfor cancer therapy. Gene Ther., 2: 235-244, 1995.

5. Pope, I. M.. Poslon. G. J., and Kinsella. A. R. The role of the bystander effect insuicide gene therapy. Eur. J. Cancer. 33: 1005-1016. 1997.

6. Faulds. D.. and Heel. R. C. Ganciclovir. A review of its antiviral activity, pharma-cokinetic properties and therapeutic efficacy in cytomegalovirus infections. Drugs,39: 597-638. 1990.

7. Moolten. F. L.. and Wells. J. M. Curability of tumors bearing herpes thymidine kinasegenes transferred by retroviral vectors. J. Nail. Cancer Inst. (Bethesda). 82: 297-300,1990.

8. Ezzeddine. Z. D.. Marluza. R. L., Platika. D., Short. M. P.. Malick, A.. Choi. B.. andBreakefield, X. O. Selective killing of glioma cells in culture and in vivo by retrovirustransfer of the herpes simplex virus thymidine kinase gene. New Biol.. 3: 608-614.1991.

9. Mullen. C. A.. Kilstrup. M.. and Blaese. R. M. Transfer of the bacterial gene forcytosine deaminase to mammalian cells confers lethal sensitivity to 5-fluorocytosine:a negative selection system. Proc. Nati. Acad. Sci. USA. 89: 33-37, 1992.

10. Huber. B. E., Austin. E. A., Good. S. S.. Knick, V. C., Tibbels. S., and Richards, C. A.In vivo anlitumor activity of 5-fluorocytosine on human coloreclal carcinoma cellsgenetically modified to express cytosine deaminase. Cancer Res., 5.ÃŽ:4619-4626.1993.

11. Freeman. S. M., Abboud, C. N., Whartenby, K. A.. Packman, C. H., Koeplin, D. S.,Moolten. F. L.. and Abraham, G. N. The "bystander effect": tumor regression when

a fraction of the tumor mass is genetically modified. Cancer Res.. 53: 5274-5283.1993.

12. Chen. C. Y.. Chang, Y. N., Ryan, P., Linscotl, M.. McGarrity, G. J., and Chiang. Y. L.Effect of herpes simplex virus thymidine kinase expression levels on ganciclovir-mediated cytotoxicity and the "bystander effect." Hum. Gene Ther.. 6: 1467-1476.

1995.13. Gagandeep, S., Brew, R., Green, B., Christmas, S. E.. Klatzmann, D., Poston. G. J..

and Kinsella. A. R. Prodrug-activated gene therapy: involvement of an immunolog-ical component in the "bystander effect." Cancer Gene Ther.. 3: 83-88. 1996.

14. Link. C. J., Jr., Moorman, D.. Seregina, T., Levy, J. P., and Schabold, K. J. A PhaseI trial of in vivo gene therapy with the herpes simplex thymidine kinase/ganciclovirsystem for the treatment of refractory or recurrent ovarian cancer. Hum. Gene Ther.,7: 1161-1179, 1996.

15. Eck, S. L., Alavi, J. B., Alavi, A., Davis, A., Hackney. D.. Judy, K.. Mollman, J.,Phillips, P. C., Wheeldon, E. B., and Wilson, J. M. Treatment of advanced CNSmalignancies with the recombinant adenovirus H5.010RSVTK: a Phase I trial. Hum.Gene Ther.. 7: 1465-1482, 1996.

4400

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

CYP:BÕ)-BASI-:DCANCERGENETHERAPY

16. Roth, J. A., and Cristiano, R. J. Gene therapy for cancer: what have we done andwhere are we going? J. Nati. Cancer Inst. (Bethesda). S9: 21-39, 1997.

17. Mesnil, M., Piccoli, C., Tiraby, G., Willecke, K., and Yamasaki. H. Bystander killingof cancer cells by herpes simplex virus thymidine kinase gene is mediated byconnexins. Proc. Nati. Acad. Sci. USA, 93: 1831-1835, 1996.

18. Sladek. N. E. Metabolism of oxazaphosphorines. Pharmacol. Ther.. 37: 301-355,

1988.19. Clarke, L., and Waxman, D. J. Oxidative metabolism of cyclophosphamide: identi

fication of the hepatic monooxygenase catalysts of drug activation. Cancer Res.. 49:2344-2350, 1989.

20. Davidoff. A. N., and Mendelow. B. V. Cell-cycle disruptions and apoptosis inducedby the cyclophosphamide derivative mafosfamide. Exp. Hematol., 21: 922-927,

1993.21. Lee, J. H.. Park. J. H.. and Yang. M. H. The effect of cyclophosphamide on

Fas-mediated apoptosis. J. Korean Med. Sci., 12: 185-189, 1997.22. Kaijser. G. P.. Korst, A.. Beijnen. J. H.. Bull. A., and Underberg. W. J. The analysis

of ifosfamide and its metabolites (review). Anticancer Res.. 13: 1311-1324, 1993.23. Weber, G. F.. and Waxman. D. J. Activation of the anti-cancer drug ifosphamide by

rat liver microsomal P450 enzymes. Biochem. Pharmacol.. 45: 1685-1694. 1993.

24. Chang. T. K. H.. Weber, G. F.. Crespi. C. L., and Waxman. D. J. Differentialactivation of cyclophosphamide and ifosphamide by cytochromes P450 2B and 3A inhuman liver microsomes. Cancer Res., 53: 5629-5637, 1993.

25. Fleming, R. A. An overview of cyclophosphamide and ifosfamide pharmacology.Pharmacotherapy. /7(Suppl.).- 146S-I54S. 1997.

26. Demetri. G. D. High-dose ifosfamide in the treatment of sarcomas of soft tissues andbone. Semin. Oncol., 23: 22-26, 1996.

27. Overmoyer. B. A. Ifosfamide in the treatment of breast cancer. Semin. Oncol.. 23:38-41, 1996.

28. Nelson, D. R.. Koymans, L.. Kamataki. T., Stegeman. J. J.. Feyereisen. R.. Waxman,D. J., Waterman. M. R.. Gotoh. O.. Coon. M. J.. Estabrook. R. W.. Gunsalus. 1. C.and Neben. D. W. Cytochrome P450 superfamily: update on new sequences, genemapping, accession numbers, and nomenclature. Pharmacogenetics. 6: 1-42. 1996.

29. Chen. L., and Waxman. D. J. Inlratumoral activation and enhanced chemotherapeuticeffect of oxazaphosphorines following cytochrome P450 gene transfer: developmentof a combined chemotherapy/cancer gene therapy strategy. Cancer Res.. 55: 581-589.

1995.30. Wei. M. X., Tamiya, T., Chase. M.. Boviatsis, E. J.. Chang. T. K. H., Kowall. N. W.,

Hochberg, F. H., Waxman, D. J., Breakefield. X. O., and Chiocca. E. A. Experimentaltumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1gene. Hum. Gene Ther.. 5: 969-978. 1994.

31. Chen. L., Waxman, D. J.. Chen. D., and Kufe. D. W. Sensitization of human breastcancer cells to cyclophosphamide and ifosfamide by transfer of a liver cytochromeP450 gene. Cancer Res., 56: 1331-1340, 1996.

32. Chase, M., Chung, R. Y., and Chiocca. E. A. An oncolytic viral mutant that deliversthe CYP2B1 transgene and augments cyclophosphamide chemotherapy. Nat. Bio-technol.. 16: 444-448, 1998.

33. Waxman, D. J.. Chen, L.. Hecht, J. E. D.. and Jounaidi. Y. Cytochrome P450-basedcancer gene therapy: recent advances and future prospects. Drug Metab. Rev., inpress. 1998.

34. Walker, D.. Flinois. J. P.. Monkman. S. C.. Beloc. C.. Boddy. A. V., Cholerton. S..Daly. A. K.. Lind, M. J.. Pearson. A. D. J.. Beaune. P. H.. and Idle. J. R. Identificationof the major human hepatic cytochrome P450 involved in activation and W-dechlo-roelhylation of ifosfamide. Biochem. Pharmacol.. 47: 1157-1163. 1994.

35. Chang. T. K.. Yu, L.. Goldstein. J. A., and Waxman. D. J. Identification of thepolymorphically expressed CYP2C19 and the wild-type CYP2C9-ILE359 alíeleaslow-íímcatalysts of cyclophosphamide and ¡fosfamideactivation. Pharmacogenetics,7:211-221,1997.

36. Morgenstern, J. P., and Land, H. Advanced mammalian gene transfer: high litreretroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res., 18: 3587-3596. 1990.

37. Pear, W. S., Nolan, G. P., Scott, M. L.. and Baltimore. D. Production of high-tilerhelper-free retroviruses by transient transfection. Proc. Nati. Acad. Sci. USA. 90:8392-8396. 1993.

38. Chen, L.. Yu, L. J., and Waxman. D. J. Potentiation of cytochrome P450/cyclophos-phamide-based cancer gene therapy by coexpression of the P450 reducíase gene.Cancer Res., 57: 4830-4837. 1997.

39. Goldstein. J. A., and de Moráis. S. M. Biochemistry and molecular biology of thehuman CYP2C subfamily. Pharmacogenetics, 4: 285-299. 1994.

40. Edwards. R. J.. Singleton, A. M.. Murray. B. P.. Davies, D. S., and Boobis. A. R.Short synthetic peptides exploited for reliable and specific targeting of antibodies tothe C-termini of cytochrome P450 enzymes. Biochem. Pharmacol.. 49: 39-47. 1995.

41. Yasukochi, Y., and Masters. B. S. S. Some properties of a delergent-solubilizedNADPH-cytochrome c (cytochrome P450) reducíasepurified by biospecific affinitychromatography. J. Biol. Chem.. 251: 5337-5344. 1976.

42. Chirgwin. J. M.. Przybyla, A. E.. MacDonald. R. J.. and Rutter. W. J. Isolation ofbiologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18: 5294-5299. 1979.

43. Paine-Murrieta. G. D., Taylor. C. W.. Curtis. R. A.. Lopez. M. H.. Dorr. R. T..Johnson, C. S., Funk, C. Y., Thompson, F., and Hersh. E. M. Human tumor modelsin the severe combined immune deficient (seid) mouse. Cancer Chemother. Pharmacol., 40: 209-214, 1997.

44. Tapscott. S. J.. Miller, A. D., Olson. J. M.. Berger, M. S., Groudine, M., and Spence,A. M. Gene therapy of rat 9L gliosarcoma tumors by transduction with selectablegenes does not require drug selection. Proc. Nati. Acad. Sci. USA, 91: 8185-8189.

1994.

45. Robbins. P. D.. Tahara. H.. and Ghivizzani. S. C. Viral vectors for gene therapy.Trends Biotechnol., 16: 35-40, 1998.

46. Kim. V. N.. Mitrophanous, K.. Kingsman. S. M.. and Kingsman. A. J. Minimalrequirement for a lentivirus vector based on human immunodeficiency virus type I.J. Virol.. 72: 811-816. 1998.

47. Nakanishi. M. Gene introduction into animal tissues. Crit. Rev. Ther. Drug CarrierSystems, 12: 263-310, 1995.

48. Pawelek, J. M.. Low, K. B.. and Bermudes, D. Tumor-targeted Salmonella as a novelanticancer vector. Cancer Res.. 57: 4537-4544. 1997.

49. Hofmann. A.. Nolan. G. P.. and Blau. H. M. Rapid retroviral delivery of tetracycline-

inducible genes in a single autoregulalory cassette. Proc. Nati. Acad. Sci. USA, 93:5185-5190, 1996.

50. Massie, B., Couture, F.. Lamoureux, L., Mosser, D. D., Guilbault. C.. Jolicoeur, P..Belanger. F.. and Langelier. Y. Inducible overexpression of a toxic protein by anadenovirus vector with a tetracycline-regulatable expression cassette. J. Virol., 72:2289-2296, 1998.

51. Schneider. A.. Schmalix. W. A., Siruguri. V., de Groene. E. M., Horbach. G. J..Kleingeist. B., Lang. D.. Bocker, R.. Belloc, C.. Beaune, P.. Greim. H.. and Doehmer,J. Stable expression of human cytochrome P450 3A4 in conjunction with humanNADPH-cytochrome P450 oxidoreductase in V79 Chinese hamster cells. Arch.Biochem. Biophys.. 332: 295-304. 1996.

52. Yu. L.. and Waxman. D. J. Role of cytochrome P450 in oxazaphosphorine metabolism. Deactivation via A'-dechloroethylation and activation via 4-hydroxylation cal-

alyzed by distinct subsets of rat liver cytochromes P450. Drug Metab. Dispos.. 24:1254-1262. 1996.

53. Patterson. A. V.. Barham. H. M.. Chinje. E. C.. Adams, G. E., Harris, A. L.. andStratford. 1. J. Importance of P450 reducíase activity in determining sensitivity ofbreast tumour cells to the bioreductive drug, tirapazamine (SR 4233). Br. J. Cancer,72: 1144-1150, 1995.

54. Workman, P., and Stratford. I. J. The experimental development of bioreductive drugsand their role in cancer Iherapy. Cancer Mela.sta.sis Rev.. 12: 73-82, 1993.

55. O'Rourke. J. F.. Dachs, G. U.. Oleadle. J. M.. Maxwell, P. H.. Pugh. C. W.. Stratford.

I. J., Wood, S. M.. and Ratcliffe. P. J. Hypoxia response elements. Oncol. Res.. 9:327-332, 1997.

56. Dachs. G. U.. Patlerson, A. V., Firth, J. D.. Ratcliffe. P. J.. Townsend. K. M..Stralford, I. J., and Harris, A. L. Targeling gene expression lo hypoxic tumor cells.Nat. Med.. 3: 515-520, 1997.

57. Brown, J. M., and Giaccia. A. J. The unique physiology of solid tumors: opportunities(and problems) for cancer Iherapy. Cancer Res.. 58: 1408-1416. 1998.

58. Shimada. T.. Yamazaki. H.. Mimura. M.. Inui, Y.. and Guengerich, F. P. Interindi-vidual variations in human liver cytochrome P-450 enzymes involved in the oxidation

of drugs, carcinogens and toxic chemicals: studies with liver microsomes of 30Japanese and 30 Caucasians. J. Pharmacol. Exp. Ther., 270: 414-423, 1994.

59. Furuya. H.. Meyer, U. A., Gelboin, H. V.. and Gonzalez. F. J. Polymerase chainreaction-directed identification, cloning, and quantification of human CYP2C18mRNA. Mol. Pharmacol.. 40: 375-382. 1991.

60. Ren, S.. Yang, J. S., Kalhorn. T. F.. and Slaltery. J. T. Oxidation of cyclophosphamideto 4-hydroxycyclophosphamide and deschloroethylcyclophosphamide in human livermicrosomes. Cancer Res.. 57: 4229-4235. 1997.

61. Goldstein. J. A.. Falcilo. M. B.. Romkes-Sparks. M., Sullivan. T.. Kitareewan. S.,

Raucy. J. L., Lasker, J. M., and Ghanayem, B. I. Evidence that CYP2C19 is the major(S)-mephenytoin 4'-hydroxylase in humans. Biochemistry. 33: 1743-1752, 1994.

62. Faber. O. K.. Mouridsen, H. T., and Skovsted. L. The effect of chloramphenicol andsulphaphenazole on the biotransformation of cyclophosphamide in man. Br. J. Clin.Pharmacol.. 2: 281-285, 1975.

63. Heise. C., Sampson-Johannes, A.. Williams. A.. McCormick. F., Von Hoff. D. D.. andKim. D. H. ONYX-015. an E1B gene-atlenualed adenovirus. causes lumor-specific

cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents. Nat. Med.. 3: 639-645. 1997.

64. Feng. M.. Jackson. W. H.. Jr., Goldman. C. K.. Rancourt. C., Wang. M., Dusing.S. K., Siegal. G.. and Curici. D. T. Stable in vivo gene Iransduction via a noveladenoviral/retroviral chimeric vector. Nat. Biotechnol.. 15: 866-870. 1997.

65. Chen. X.. Li. Y.. Xiong, K.. Aizicovici, S., Xie, Y., Zhu, Q., Sturtz. F.. Shulok, J..Snodgrass. R.. Wagner, T. E.. and Platika. D. Cancer gene therapy by direct tumorinjections of a nonviral T7 vector encoding a Ihymidine kinase gene. Hum. GeneTher., 9: 729-736. 1998.

66. Morgan. R. A.. Couture. L.. Elroy-Stein. O.. Ragheb, J., Moss. B.. and Anderson.W. F. Retroviral vectors containing putative internal ribosome entry sites: development of a polycistronic gene transfer system and applications to human gene therapy.Nucleic Acids Res., 20: 1293-1299, 1992.

67. Fisher. C. W.. She!. M. S.. and Estabrook. R. W. Construction of plasmids andexpression in Escherichia cali of enzymatically active fusion proteins containing theheme-domain of a P450 linked to NADPH-P450 reducíase.Methods Enzymol., 272:15-25. 1996.

68. Gunzburg, W. H., Karle. P.. Mrochen. S.. Sparmann. G.. Sailer. R., Klein, D.. Uckert,W., and Salmons, B. Regulated gene expression after relroviral vector-mediateddelivery of cancer-relevant therapeutic genes. Ree. Results Cancer Res., 144: 116-

126, 1998.69. Walther. W., and Stein, U. Cell type specific and inducible promoters for vectors in

gene therapy as an approach for cell targeting. J. Mol. Med., 74: 379-392, 1996.70. Schnierle, B. S.. and Groner. B. Retroviral targeled delivery. Gene Ther., 3: 1069-

1073, 1996.

4401

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1998;58:4391-4401. Cancer Res   Youssef Jounaidi, Jodi E. D. Hecht and David J. Waxman  Oxazaphosphorine-based Cancer Gene TherapyRetroviral Transfer of Human Cytochrome P450 Genes for

  Updated version

  http://cancerres.aacrjournals.org/content/58/19/4391

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/19/4391To request permission to re-use all or part of this article, use this link

on April 22, 2020. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from