retroviral transfer of a bacterial alkyltransferase gene into … · g418 selection of infected nih...
TRANSCRIPT
Vol. 1, /359-1368, Not’einher 1995 Clinical Cancer Research 1359
Retroviral Transfer of a Bacterial Alkyltransferase Gene
into Murine Bone Marrow Protects against
Chioroethylnitrosourea Cytotoxicity1
Linda C. Harris,2 Upendra K. Marathi,
Carol C. Edwards, Peter J. Houghton,
Deo Kumar Srivastava, Elio F Vanin,
Brian P. Sorrentino, and Thomas P. Brent
Department of Molecular Pharmacology [L. C. H., U. K. M.,
C. C. E., P. J. H., T. P. B.], Department of Biostatistics [D. K. S.],
Division of Experimental Hematology [E. F. V.], and Department of
Biochemistry [B. P. 5.], St. Jude Children’s Research Hospital,
Memphis, Tennessee 38105; Department of Pharmacology, College
of Medicine, University of Tennessee, Memphis. Tennessee 38163
[T. P. B.1; and Genetic Therapy, Inc., Gaithersburg, Maryland20878 [E. F. V.]
ABSTRACT
The chloroethybnitrosoureas (CENUs) are important
antineoplastic drugs for which clinical utility has been re-
stricted by the development of severe delayed myebosuppres-
sion in most patients. To investigate the potential of DNA
repair proteins to reduce bone marrow sensitivity to the
CENUs, we transferred the Escherichia coli ada gene, which
encodes a Mr 39,000 06-alkylguanine-DNA alkyltransferase
(ATase), into murine bone marrow cells by the use of a
high-titer ecotropic retrovirus. The ada-encoded ATase is
resistant to 06-benzylguanine (06-BG), a potent inhibitor of
the mammalian ATases, thus affording the bone marrow an
additional level of protection against CENUs. In methylcel-
luiose cultures, ada-infected hematopoietic progenitor cells
were twice as resistant as uninfected cells to the toxic effects
of l,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) following
treatment with 06-BG. Although showing no obvious pro-
tective effects against leukopenia, overexpression of the bac-
teriab ATase activity reduced the severity of anemia and
thrombocytopenia in mice treated with 06-BG and BCNU.These effects, which were maximal at a BCNU dose of 12.5
mg/kg, were associated with improved survival when BCNU
was given at this dose. At lower BCNU doses cytotoxicity
was limited in both transduced and control mice, and at
higher doses the protective effect was saturated due to cy-
totoxicity. These results suggest that ada gene therapy may be
a feasible approach to amelioration of delayed myelosuppres-
sion following 06-BG plus CENU combination chemotherapy.
Received 3/29/95: revised 6/21/95; accepted 7/13/95.
I Supported in part by NIH Grants CA23099 and CA14799, Cancer
Center Support (CORE) Grant P30 CA21765, and the American Leba-
nese and Syrian Associated Charities.
2 To whom requests for reprints should be addressed, at Department of
Molecular Pharmacology. St. Jude Children’s Research Hospital, P.O.
Box 318, 332 North Lauderdale, Memphis, TN 38105. Phone: (901)
495-3440; Fax: (901) 521-1668.
INTRODUCTION
O”-alkylguanine-DNA ATases3 are conserved DNA repair
proteins found in all living organisms studied to date (1, 2).
They are responsible for the repair of O”-alkybguanine and
04-alkylthymine, which are toxic, mutagenic DNA adducts gen-
erated by the alkylnitrosoureas MNU and N-methyl-N’ -nitro-N-
nitrosoguanidine (1, 2). ATases also repair O”-chloroethylgua-
nine lesions, the DNA cross-link precursors produced by the
CENUs, such as BCNU and l-(2-chloroethyl)-3-(trans-4-meth-
ylcyclohexyl)- 1 -nitrosourea (3). Because DNA cross-links are
extremely cytotoxic, the CENUs possess low mutagenic poten-
tial (4), and, therefore, these drugs are useful antineoplastic
agents (5). The ATase repair reaction is stoichiometnic and
autoinactivating, so that the degree of cytotoxicity produced by
a CENU depends on the amount of repair protein within the cell
(6). Consequently, human tumor xenografts expressing high
levels of this repair enzyme are resistant to CENU therapy (6),
and tissues expressing very low levels (e.g., bone marrow) are
highly sensitive (7, 8). These observations are especially perti-
nent to BCNU, because of its central role in primary therapy for
brain tumors (9); as for all CENUs, its dose-limiting toxicity is
delayed myebosupression (5, 9).
An attractive strategy to overcome bone marrow sensitivity
to the CENUs would rely on overexpression of an ATase gene
and incorporating a mechanism by which endogenous tumor
ATase activity may be selectively inhibited. Several groups
have introduced drug resistance genes into murine bone marrow
with retroviral vectors [e.g. , the multidrug resistance I gene
(MDRJ; Ref. 10) and mutant dihydrofolate reductase cDNAs
(1 1-13)] demonstrating gene expression by selection of taxob-
resistant marrow in s’is’o or improvement in the survival of
methotrexate-treated mice.
In the study reported here, we tested the feasibility of this
approach using the ada ATase gene of Escherichia coli (14),
which encodes a Mr 39,000 protein consisting of two subunits
that repair either O”-alkylguanine and 04-abkylthymine or alky-
lphosphotniester adducts (14, 15). Previously, its expression in
mammalian cells in vitro has been shown to protect against
nitrosourea cytotoxicity (16, 20). The ada protein is similar to
other ATases in that it is autoinactivated following reaction with
its substrate, although in contrast to mammalian ATases, it has
a very low affinity for O”-BG (21, 22), which has been used to
3 The abbreviations used are: ATase, alkyltransferase; CENU, chloro-ethybnitrosourea; MNU, N-methyl-N-nitrosourea; MNNG, N-methyl-N’-
nitro-N-nitrosoguanidine; BCNU, l,3-bis(2-chloroethyb)- 1 -nitrosourea;
MDRI, mubtidrug resistance 1; O”-BG, O”-benzybguanine; PGK, phos-
phoglycerate kinase; IL, interbeukin; LTR, long terminal repeat; RT,
reverse transcription; PEG, polyethylene glycol.
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1360 Alkyltransferase Gene Expression in Munine Bone Marrow
sensitize resistant human tumor xenografts to CENU chemo-
therapy by depleting their ATase (23-26).
Theoretically, 06-BG treatment should sensitize tumors to
the cytotoxic effect of the CENUs, whereas overexpression of
the 06-BG-resistant ATase increases the resistance of normal
bone marrow. We describe successful transfer of the bacterial
ada gene into the hematopoietic cells of CB/CaJ mice, and
protection against 06-BG plus BCNU-induced myelosuppres-
sion and death.
MATERIALS AND METHODS
The plasmid containing the ada gene p062C (27) was
generously provided by Dr. G. P. Margison (Paterson Labora-
tories, Manchester, England), and the retrovinal plasmid pGlNa
(28) was provided by Genetic Therapy, Inc. (Gaithersburg,
MD). The pUCOO7pgk plasmid contained the PGK promoter
(29), and the N2 control retroviral producer clone (30) expressed
the neo gene. NIH 3T3 cells were obtained from the American
Type Culture Collection (Rockville, MD) and grown in DMEM
(GIBCO-BRL, Gaithensburg, MD) supplemented with 10%
newborn calf serum (Sigma Chemical Co., St. Louis, MO), as
were GP + E86 cells (31). PA317 cells (32) were grown in
DMEM containing 10% FCS (Sigma). CBA/CaJ female mice
were obtained from The Jackson Laboratory (Bar Harbor, ME)
and housed in isolation cages in a negative pressure cubical with
access to food and water ad libitum. Human IL-6 was a gift from
Amgen (Thousand Oaks, CA), whereas munine IL-3 was ob-
tamed from GIBCO-BRL. O”-BG was a gift from Dr. R. C.
Moschel (National Cancer Institute, Frederick, MD), and BCNU
(carmustine) was obtained from Bristol-Myers Squibb Co.
(Princeton, NJ). For in vitro studies, O”-BG was dissolved in
DMSO, and BCNU was dissolved in 100% ethanol. For in vivo
studies, the former compound was dissolved in 40% PEG 400 in
saline, with the latter dissolved in 10% ethanol. All radiolabeled
isotopes were from Amersham (Arlington Heights, IL); other
reagents were from Sigma unless otherwise stated.
Generation of Ecotropic Retroviral Producer Clones byTransinfection. Plasmid DNA was initially transfected into
the amphotropic packaging cell line PA317 by a modified
calcium phosphate coprecipitation method (Stratagene, La Jolla
CA; Ref. 33). Forty-eight h later, the medium-containing virus
was removed and diluted 1:5, 1:10, 1:25, 1:50, and 1:100 with
fresh medium. Polybrene (hexadimethrine bromide) at 6 p.g/ml
was added to each of the dilutions, which then were added to
five plates of the ecotnopic packaging cell line GP+E86. After
another 48 h, each of the GP+E86 plates was divided into five
larger 150 x 20-mm plates at the following dilutions: 1:10,
1:20, 1:40, 1:80, and 1:160. One day later, fresh media contain-
ing 500 p.g/ml active G418 (Geneticin; GIBCO-BRL) were
added to the cells and replaced every 3 or 4 days until discrete
colonies were visible. After 10-14 days, 30 colonies were
isolated using glass cloning rings (Bellco, Vineland, NJ) and
placed in the individual wells of 24-well plates. These producer
clones were expanded and characterized for virus production.
Retroviral Titering by PEG Precipitation of RNA andSlot Blot Hybridization. Viral supernatants, 1 ml each, were
transferred from each producer clone to an Eppendorf tube, to
which 0.5 ml PEG solution (30% PEG 8000, 1.5 M NaC1) was
added. The tubes were held on ice for 30 mm before centnifu-
gation at 14,000 rpm for S mm at 4#{176}C.The pellets were resus-
pended in 0.2 ml VTR [0.5 ml Vanadyl Ribonucleoside Com-
plex (GIBCO-BRL), 10 mg/ml tRNA, and S ml TE (10 m�vt
Tnis-HCI, 1 mM EDTA, pH 8.0)], after which 0.2 ml 2X lysis
buffer (1% SDS, 0.6 M NaC1, 20 m�i EDTA, and 20 msi Tnis, pH
7.4) was added. The mixed solution was extracted once with an
equal volume of U-saturated phenol and then with an equal
volume of chloroform:isoamyl alcohol (24:1). The RNA was
then precipitated by the addition of 0.9 ml ethanol and by
holding on dry ice for 15 mm. After centnifugation at 14,000
rpm for 15 mm at 4#{176}C,the pelleted RNA was dissolved in 250
p.1 15% formaldehyde solution (2.1 ml 37.5% formaldehyde, 2.9
ml H2O); 250 p.1 20X SSC were added; and the samples were
heated at 50#{176}Cfor 15 mm. They were then held on ice before
being applied to the slot blot apparatus (Minifold II; Schleicher
& Schuell, Keene, NH), which contained a GeneScreen plus
membrane (New England Nuclear DuPont, Wilmington, DE)
prewetted in lOX SSC. The membrane was hybridized with a
virus-specific probe using standard procedures (34).
Clones that generated the strongest signals, hence the high-
est levels of viral RNA, were subjected to biological titering by
G418 selection of infected NIH 3T3 cells and counting resistant
colonies.
ATase Assays and Fluorography. The O”-ATase assay
1, used to quantitate the activity in the producer clones, relied on
[3HIMNU-treated calf thymus DNA substrate as previously
described (35). Assay 2, originally described by Wu et a!. (36)
and modified by Marathi et a!. (37), relies on PvuII digestion of
a double-stranded oligonucleotide once the O”-methylguanine
adduct within its cleavage recognition sequence has been re-
paired by ATase. The second assay has greater sensitivity than
the first and therefore allowed qualitative assessment of ATase
activity in WBCs isolated from mouse peripheral blood. Blood
(0.7-1.5 ml) was obtained by cardiac puncture from anesthe-
tized mice, which were subsequently killed by cervical disloca-
tion. RBCs were lysed in 10 ml RBC lysis buffer (150 m�i
NH4C1, 10 mM KHCO3, and 0.1 msi EDTA) and centrifuged for
12 mm at 2000 rpm to pellet the WBCs, from which extracts
were prepared as described for assay 1 (35).
Fluorography was performed by incubating producer cell
extracts with [3HJMNU-treated substrate DNA at 37#{176}Cfor 30
mm before PAGE and electroblotting of proteins to Immobil-
lon-P membranes (Millipore, Bedford MA). Because the ATase
proteins covalently bind [3H]methyl groups following their re-
pair, radiolabeled protein was visualized by the addition of
scintillation fluid (Scint-A XF, Meniden, CT) to the membrane
and autoradiography (14).
Genomic DNA Isolation, Labeling of DNA Probes, and
Southern Blot Analysis. Genomic DNA was isolated from
producer cells using a standard protocol, essentially as described
by Sambrook et al. (34), which involves SDS lysis, incubation
in proteinase K (GIBCO-BRL), phenol-chloroform extraction,
and ethanol precipitation. Southern blot analysis was performed
as described by Sambrook et a!. (34); 32P labeling of DNA
probes was carried out with a Random Prime Labeling kit
(Boehninger Mannheim, Indianapolis, IN).
Genomic DNA was isolated from mouse blood (200 pi) by
lysing RBCs in 10 ml RBC lysis buffer (see above). WBCs were
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Sail
Sail pBRO
Clinical Cancer Research 1361
BamHI Salt BamHl
pGlNa
Fig. I Construction of pGlNada, a retroviral vector derived from the pGlNa plasmid, which contains ada under control of the PGK promoter (pgk
Pr.). Ltr, long terminal repeat; pBRO, pBR322 origin of replication; AMP, ampiciblin resistance gene; neo, neomycin resistance gene.
pelleted by centnifugation at 2000 rpm for 12 mm and resus-
pended in 1 ml cell lysis buffer (0.32 M sucrose, 10 mr�t Tnis-
HC1, pH 7.5, 5 mM MgC12, and 1% Triton X-100). After being
held on ice for S mm, the tubes were centrifuged for 10 mm at
3000 rpm to pellet the nuclei. Genomic DNA was then isolated
as described for the producer cells.
Retrovirab Infection of Murine Bone Marrow Cells.
CBAJCaJ mice were thymectomized at 8 weeks of age so that
future studies may use human tumor xenografts, and, at 10
weeks, received i.v. injections of 150 mg/kg 5-fluorouracil
(Sobopak Laboratories, Inc., Elk Grove Village, IL). Forty-eight
h later, they were sacrificed for isolation of bone marrow from
their tibias and femurs (38). Marrow was flushed from the bones
with a needle and syringe containing PBS and 2% FCS (Hy-
clone, Logan, UT) and plated into DMEM containing 15%
Hyclone FCS, 2x glutamine, IL-3 (3 ng/ml), and IL-6 (100
ng/ml) (39) at a concentration of S x 106 cells/100-mm suspen-
sion dish (Corning Glass Works, Corning, NY). After another
48 h, the marrow cells were harvested and cocultured with
retroviral producer cells at the same density and in the same
media as above with the addition of polybrene (6 p.g/ml).
Following a 48-h infection period, marrow cells were harvested
and used for either transplantation of irradiated recipients or in
vitro methylcellulose colony-forming assays.
Assay for Clonogenic Hematopoietic Progenitor Cells.
Methylcellubose (Stem Cell Technologies, Vancouver, British
Columbia, Canada) was divided into 4-ml aliquots for each drug
dose to be tested. Retrovirally infected bone marrow cells were
then added at a concentration of 1 x 105/ml followed by the
addition of 10 �LM O”-BG. After 2 h, BCNU was added, and
1-ml aliquots were plated into 35-mm dishes and incubated for
1 1 days. The number of colonies was counted by light micros-
copy on day 4. Colonies for RT-PCR analysis were isolated on
day 11.
PCR Analysis. DNA isolated from peripheral mouse
blood was subjected to PCR analysis with either ada and �3-glo-
bin or neo- and �3-globin-specific primers. The sequences of
these oligonucleotides were: ada, 5’-GCTGGCCAATCT-
GTCTFAGC-3’ and 5’-CCGATGTAGATGAAATGGAC-3’,
which generated a 3i9-bp fragment; neo, 5’-TCCACTATG-
GCTGATGCAATGCGGC-3’ and 5’-GATAGAAGGCGAT-
GCGCTGCGAATCG-3’, which generated a 404-bp fragment,
and �3-globin, 5’-GAAG1TGGGTGmGGAGAC-3’ and 5’-
GAGACTGCFCCCFAGAATCG-3’, which generated a 401-bp
fragment. Each PCR reaction contained 200 ng DNA, 1 x Pro-
mega PCR buffer (Promega, Madison, WI), 1 m�i MgCl,, 0.2
mM of each of the four nucleotides, 2 �j.Ci [33P]dCTP, 50 ng of
each oligonucleotide primer, and 2.5 units Taq polymerase
(Promega). Amplifications were performed in a minicycler
(M. J. Research, Watertown, MA) for 25 cycles at 94#{176}Cfor 45
s, 56#{176}Cfor 1 mm 30 s, and 72#{176}Cfor 2 mm. Amplified products
were separated by nondenatuning PAGE (6%) and visualized by
autoradiography of the dried gel.
RT-PCR. RNA was prepared from hematopoietic cob-
nies with RNAzol (Tel-Test, Fniendswood, TX), as described by
the manufacturer; RT was performed with a eDNA cycle kit
(Invitrogen, San Diego, CA). The conditions of RT-PCR were
essentially the same as those described for DNA, except that
reactions were extended to 30 cycles.
Hematobogical Analysis. Blood (20 p.1) was diluted with
9980 p.1 hematobogical diluent (0.1 M NaCI, 16 msi boric acid,
0.1 mM sodium-tetraborate, and 0.5 mts.i EDTA) and analyzed
using a Sysmex automated hematology analyzer.
Statistical Analysis. The measurements for the variables
hemoglobin, hematocnit, RBCs, platelets, and WBCs were re-
corded on days 0, 3, 9, 14, 21, 28, 42, 45, 51, 56, 63, and 70 for
the two groups (ada and neo) of mice. Because the observations
for each mouse on various days are correlated, the data were
analyzed using ANOVA for repeated measures (40). For the
analysis, it was assumed that the underlying covaniance structure
had compound symmetry, and the observations were missing at
random; i.e. , information missing because of death or for any
other reason did not provide any information about the process.
The analysis was performed using the program 5V in the BMDP
statistical software package (41). The significance of the param-
eters in the model was tested using a Wald-type x2 statistic. For
the two groups (ada and neo), Kaplan-Meier plots (42) were
obtained and compared using the exact Wilcoxon-Gehan test as
implemented in StatXact statistical software (43).
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
‘�/ �) �)
49.5 � � .�- ada32.5 -
27.5 �--- � ‘--. .�- human/mouse18.5 -
4 14 18 21 23 29
kb
23.1 -
9.4 -
6.6 -
4.4 -
2.3 -
2.0 -
1.3 -
1.111,1
Cu>
>
(I)0
0
10A A ada-BCNU
.-. neo-BCNU
U #{149}ada-BCNU + O6BG. - . neo-BCNU + 06 BG
1 � � I I I
0 5 10 15 20 25 30
BCNU Dose �M
Fig. 4 Methylcellulose assay for cbonogenic survival of hematopoeitic
progenitor cells among ada- and neo-infected bone marrow cells treated
in vitro with either BCNU alone or BCNU in combination with O”-BG.Points, mean colony number relative to the number in untreated plates;
bars, SDs.
1362 Alkyltransferase Gene Expression in Munine Bone Marrow
Fig. 2 Fluorography of cell extracts showing sizes of active ATaseproteins. Clones 18 and 29 are two ada-expressing retrovirab producer
clones, GP+E86 is the parental munine line from which they were
derived, and CEM is a human T-lymphoblast cell line that served as a
control. Arrows, ada, human, and munine ATase proteins.
RESULTS
Generation of an ada Ecotropic Retrovirus. The
Moloney leukemia virus-based vector pGlNada was generated
from the pGlNa plasmid by inserting the PGK promoter (29)
and the ada gene (Fig. 1; Ref. 27). The resulting plasmid has the
potential to express the ada gene in mammalian cells under
control of the PGK promoter and the neo gene under control of
the viral promoter contained within the LTR. pGlNada was then
used to generate retroviral producer clones, as described in
‘ ‘Materials and Methods.’ ‘ The clone with the highest titer,
clone 18 (approximately 2 x 106 virus particles/ml,) was used
in all subsequent experiments described in this article. It was
shown to be free of the helper virus, because supernatants
collected from infected NIH 3T3 cells failed to infect a second-
any culture as determined by a marker rescue assay (32).
Characterization of Producer Clones. The O”-ATase
activity of clone 18 (1.5 pmol/mg protein) was 5-fold greater
than the endogenous activity of parental GP+E86 cells (0.3
pmol/mg). Moreover, clone 18 ATase was not inhibited by a 1-h
exposure to 10 p.M O”-BG (data not shown). Incubation of
whole-cell extract with [3H]MNU-tneated substrate DNA, fob-
bowed by PAGE and fluorognaphy, enabled us to visualize the
sizes of radiolabeled ATases, namely, the Mr 39,000 ada and Mr
22,000 mouse proteins (Fig. 2). Although clone 18 expresses
both the neo and ada genes, it was designated the ada-express-
ing virus to distinguish it from N2, which expresses only neo.
Genomic DNA isolated from several producer clones was
digested with the NheI restriction endonuclease (which cuts the
viral DNA once within each of its two LTRs) and subjected to
Southern blot analysis. After hybridization with a neo gene
probe, a single 4-kb band was observed in each ofthe lanes (Fig.
3), demonstrating that full-length proviral DNA had been uni-
fonmly inserted into the host cell genome.
ada Gene Transfer to Murine Hematopoietic Commit-
ted Progenitor Cells in Vitro. CB/CaJ bone marrow cells
were infected with either the clone 18 on N2 netrovirus and
exposed to BCNU, added alone on 1 h after treatment with
O”-BG. After 4 days of growth in methylcellulose, the total
1.1 -
0.9 -
Fig. 3 Southern blot analysis of DNA extracted from six retroviral
producer cell lines probed with the izeo gene. The DNA was digested
with N/tel, which cuts once within each viral LTR to generate a 4063-bp
fragment.
number of hematopoeitic progenitor cell colonies was counted.
The survival responses to BCNU alone of cells infected with
either ada- or neo-expressing viruses were similar; however,
with preexposure to O”-BG the neo-transfected cells became
sensitized to BCNU, whereas those beaning the ada gene did not
(Fig. 4). The D37 for neo-tnansfected cells was 10 p.M BCNU
compared with 20 JiM for the the ada-positive cells, representing
a 2-fold increase in sensitivity following preincubation with
O”-BG.
The presence of ada mRNA was demonstrated by RT-PCR
(Fig. 5) within individual colonies, indicative of ada gene ex-
pression. Only two of the eight colonies exposed to O”-BG but
not to BCNU were positive for ada, compared with eight of
eight exposed to both O”-BG and BCNU (30 p.s�; Fig. 5).
Reconstitution of Lethally Irradiated Mice with Bone
Marrow Infected ex Vivo with the ada or neo Gene. Thirty-
seven female CB/CaJ donor mice were sacrificed 48 h after iv.
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
bp
1 2 3 4 5 6 7 8
+ - + - + - + -+ - + - + - + - RT
0iM BCNU
1 2 3 4 5 6 7 8
+ - + - + -+ - + - #{247}- + -�1- -AT
� � � 30MBCNU310�281- �204-194”
ada_
neo
pglobinlFig. 6 Demonstration of the presence of ada or izeo genes in mouse
genomic DNA. PCR analysis of DNA isolated from the peripheral blood
of eight representative mice is shown. Mice 1-4 (upper panel) werereconstituted with ada virus-infected marrow, and mice 1-4 (lower
panel) were reconstituted with izeo virus-infected marrow. PCRs were
performed with either ada-. iieo-, or 13-gbobin-specific primers.
Clinical Cancer Research 1363
bp
603 -
310281 -204 -
194 “
603-
Fig. 5 Frequency of bone marrow progenitor cell colonies that expressed ada mRNA. ada RT-PCR analysis of individual colonies isolated from
methylceblubose 1 1 days after treatment with O”-BG and a 2-h exposure to either vehicle only (0 p.M BCNU, top) or 30 �J.M BCNU (hotto,n).
administration of 150 mg/kg 5-fluorouracil (38), and hone mar-
row cells were harvested as described in ‘ ‘ Materials and Meth-
ods.’ ‘ Fifty percent of the cells were infected with the clone 18
virus containing the ada gene, and the remainder were infected
with the N2 virus containing neo. Cells (4 x 10”) from either
manipulation were then injected into tail veins of each of 73
CB/CaJ female recipient mice that had been thymectomized and
lethally irradiated (850 cGy). All mice became reconstituted
with donor marrow cells, as determined by 100% survival
beyond 2 weeks after transplantation. After 3 weeks, DNA was
isolated from 200 pi blood isolated from the retroonbital venous
plexus of each mouse. PCR analysis performed on blood sam-
ples from 28 of 38 ada and 12 of 35 iieo recipient mice revealed
the presence of the respective gene in every case examined (Fig.
6). Thus, all mice were likely positive for the virally inserted
genes. At 12 weeks after transplantation, three ada-positive
mice were sacrificed, and their marrow was used to reconstitute
24 additional mice, all of which became PCR positive for the
gene. This finding, together with the observation that three ada
mice remained PCR positive when tested 8 months after trans-
plantation, indicates that long-lived primitive hematopoietic
progenitors had been infected with the netrovirus ex i’is’o.
O”-BG and BCNU Treatment of Mice Transplantedwith Retrovirally Infected Bone Marrow Cells. Five weeks
after transplantation, 35 ada- and 35 iieo-beaning mice, pro-
duced as described above, were divided into five treatment
groups each comprising 7 ada and 7 neo mice. They were
treated as follows: (group 1) vehicle alone; (group 2) 30 mg/kg
O”-BG and 7.5 mg/kg BCNU; (group 3) 30 mg/kg O”-BG and
10 mg/kg BCNU; (group 4) 30 mg/kg O”-BG and 12.5 mg/kg
BCNU; and (group 5) 30 mg/kg O”-BG and 15 mg/kg BCNU.
The drugs were injected i.p., with O”-BG administered 2 h
before BCNU. The mice in group 1 received both vehicles, 40%
PEG in saline and 10% ethanol.
Blood samples (20 �i.l each) were taken from the tail vein
of each mouse on the day of drug treatment (day 0) and subse-
quently on days 3, 9, 14, 21, and 28. On day 42, surviving mice
p globin
1 2 3 4I I I 1 1 1 I I
1 2 3 4I ii 1 1 1 1
received a second dose of O”-BG and BCNU that was identical
to the first treatment, with additional blood collected on days 42,
45, 51, 56, 63, and 70. Measurements of hemoglobin, hemato-
cnit, erythrocytes, WBCs, and platelets were obtained for all
doses at each time point and subjected to statistical analysis.
The two phases of the experiment from day 0 through 28
and from day 42 through 70 were analyzed either separately and
also after combining them together. The most significant differ-
ences in the blood parameters for the two groups were observed
in group 4 following treatment with O”-BG and 12.5 mg/kg
BCNU. Fig. 7 shows the hematopoietic data for this dose. For
group 4, the combined statistical analysis showed there was a
significant difference (at a level of significance a 0.01)
between the two groups (ada and ,zeo) for the variables hemo-
gbobin (P < 0.0001), hematocrit (P 0.0001), and RBCs (P
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Hemoglobin70
Hematocrit
60
6
50
4
40
2
30
16
14
12
10
-... 8
bO
0
10
8
2
0
20
10
0
-u--ad�id��D�.neoI
E
1200
1000
800
600
400
200
0
p = 0.0002
0
Days Days
1364 Alkyltransfenase Gene Expression in Murine Bone Marrow
Red Blood Cells
I I I I
10 20 30 40 50 60 70 80
Fig. 7 Comparison of findings in mouse blood sampled at various times before and after O”-BG and 12.5 mg/kg BCNU treatment. Panels, groupof seven ada mice and seven neo control mice each. Drugs were administered on days 0 and 42. Data points, mean measurements for surviving mice;
bars, SE. Because only one neo mouse survived at day 70, a SE could not be calculated for this point. The statistical significance (a = 0.01) for thedifferences between the two groups of mice analyzed using ANOVA for repeated measures over the entire 70-day time course is shown as the P value.
0.0002). The P value corresponding to the platelet counts for the
two groups was 0.014, which indicates that the two groups differ
significantly at the 0.05 level of significance but not at the 0.01
level of significance. However, there was no difference in the
mean of WBC levels in the two groups (P = 0.136). A similar
trend was also noted when the data in the two phases were
analyzed separately, with the only exception that the difference
between ada and neo in platelet counts was significant at a =
0.01 in the second phase (P = 0.004). There was no difference
in any of the variables in the two groups (ada and neo) when
treated with vehicle alone.
The vast majority of mice survived the first drug treatment;
however, death rates increased substantially after the second
dose (Fig. 8). A clear survival advantage was conferred by the
ada protein in mice treated with BCNU at 12.5 mg/kg (P =
0.021) consistent with the protection of hematobogical parame-
tens shown in Fig. 7. There was no significant difference in the
survival of the two groups for all other dose levels. Necropsies
performed on four mice within 1 h of death revealed severe
myeboid hypoplasia and anemia due to BCNU-induced deficits
in blood elements.
ATase Activity Assay of WBCs Isolated from Trans-
duced Mouse Blood. ATase assay 2 was performed on WBC
extracts isolated from 12 ada mice and 2 neo mice 21 weeks
after reconstitution with transduced marrow (Fig. 9). Extracts
(50 gig) were pretreated for 30 mm at 37#{176}Cwith 50 p.M O”-BG
to inhibit endogenous munine ATase prior to incubation with
0.05 pmol oligonucleotide substrate. A positive control (Fig. 9,
C3) consisted of 10 �i.g ada producer cell extract expressing 3
pmol/mg ada ATase. As indicated in Fig. 9, O”-BG-resistant
activity was evident in at least three of the mice analyzed,
demonstrating that ada ATase was being expressed, albeit at
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
7.5 mg/kg BCNU
30 mg/kg 06 BG
100
10 mg/kg BCNU
30 mg/kg 06 BG
L1�
>
>
(1)
50
0-0 1020304050607080
1’ 1’
12.5 mg/kg BCNU30 mg/kg 06 BG
100
50ada
� neo
0 � � � �
0 1020304050607080
t t15 mg/kg BCNU
30 mg/kg 06 BG
�100
50
‘.
t
1.� I I I I,J �,
0 1020304050607080 0
1’
10 20 30 40
1’
50 60 70 80
Clinical Cancer Research 1365
Time (Days)
Fig. 8 Kaplin-Meien survival curves for mice exposed to 30 mg/kg 06-BG and various doses of BCNU. Arrows, days at which drugs were
administered, days 0 and 42. Each group contained seven mice. Curves for two groups (ada and neo) were compared using the exact Wilcoxon-Gehantest. A clean survival advantage was conferred by ada in the mice treated with 12.5 mg/kg BCNU (P 0.021), but there was no significant differencein the survival of the two groups at all other doses (7.5 mg/kg, P 0.44; 10 mg/kg, P = 0.93; 15 mg/kg, P = 0.92).
low levels, in WBCs after transduction and transplantation of
virally infected marrow cells. ATase cannot be measured in any
other hematopoietic cell compartment.
DISCUSSION
We have generated a high-titer netnoviral producer clone
that expresses high levels of the bacterial ATase gene, desig-
nated ada. This replication-defective retnovirus infects long-
lived munine hematopoetic progenitor cells, as demonstrated by
the persistence of ada gene sequences in peripheral blood and
by the generation of ada-positive secondary bone marrow re-
cipient mice. The functionally active, full-length Mr 39,000 ada
protein expressed by producer clone 18 was easily distinguished
from the smaller Mr 22,000 endogenous munine ATase ex-
pressed by GP+E86 cells (Fig. 2). The identity of the ada
protein was confirmed by its resistance to the mammalian
ATase inhibitor 06-BG (21). Southern blot analysis of genomic
DNA isolated from clone 18 and other producer clones (Fig. 3)
showed that the correct full-length proviral DNA had been
inserted into all clones analyzed, without gross rearrangements
on deletions.
The in vitro methylcellubose assay demonstrated that the
clone 18 retrovirus was able to infect hematopoietic progenitor
cells (Fig. 4). Functional expression of the ATase was indicated
by the 2-fold greaten resistance of clone 18 versus N2-infected
cells to BCNU-induced cytotoxicity. Because these cells were
only partially transduced (-25%), it is likely that the resistance
of the ada transduced component was greaten than 2-fold. It was
surprising that after exposure to BCNU alone, the ada ATase
provided no additional protection to cells beyond that provided
by endogenous ATase levels (7, 8). It is possible that the levels
of ada activity were so low that the total increase in DNA repair
capacity was too small to be significant. However, once the
munine enzyme was inhibited with 10 p.M 06-BG, the neo-
expressing cells became sensitized to BCNU, whereas the ada-
expressing cells retained their resistance, indicating that the
bacterial protein was functionally active in munine hematopoi-
etic cells. The 2-fold increase in resistance generated by clone
18 may have been greater if a higher proportion of cells ex-
pressed the gene. ada mRNA was detected in only two of eight
colonies exposed to the virus in vitro, suggesting that a low
proportion (-25%) of cells expressed the ada gene (Fig. 5).
However, this proportion increased to 100% following treatment
with 10 �LM O”-BG and 30 p.M BCNU, presumably due to
selection for ada-expressing cells containing functionally active
ATase encoded by the bacterial gene.
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
ada neo
12.5mg/kg
BCNUI 10mg/kg
Cl C2 C3 Vehicle BCNU
18 mer 4 -�--� . - - �-
8 mer -�
Vehicle
1366 Alkyltransferase Gene Expression in Munine Bone Marrow
Fig. 9 ATase activity assay of peripheral WBCs isolated from mice 21 weeks after retroviral transduction. Lane Cl, unreacted double-stranded
O”-methybguanine-containing oligonucleotide 18-men substrate; Lane C2, the same oligonucleotide following incubation with P11411, demonstrating
that no cleavage occurs prior to repair of the O”-guanine adduct; Lane C3, incubation with a control producer cell extract-expressing ada. Remaining
lanes, extracts of peripheral blood leukocytes from surviving mice that were transduced with ada or neo and treated with vehicle or different doses
of BCNU as indicated in the legends to Figs. 7 and 8. All of the cell extracts were incubated with 50 p.M O”-BG for I h at 37#{176}Cto inactivate
endogenous mouse ATase prior to addition of the oligonucleotide substrate.
The ability of the ada ATase protein to protect hematopoi-
etic progenitor cells against BCNU in vitro prompted us to use
clone 18 to infect CB/CaJ mouse bone marrow. The modified
marrow was then injected into marrow-depleted recipients for
the study of ada protective effects in vivo. We chose thymec-
tomized CB/CaJ mice for our experiments because they have
been successfully used at this center in therapeutic studies with
human tumor xenografts (6). Our in vivo experiments revealed
that transduction of the ada gene into hematopoietic cells can
protect against nitnosounea-induced myebosupression. After the
first treatment of O”-BG and BCNU, all measured blood ele-
ments (platelets, enythnocytes, hematocnit, hemoglobin, and
WBCs) were significantly suppressed (Fig. 7), but there were
few deaths (Fig. 8). The rationale for giving a second treatment
after the blood values had returned to normal was that ada-
expressing cells might be selected during the first treatment, so
that marrow would become enriched with such cells and be
more resistant when challenged with a second dose. However,
validity of this approach could not be determined because of the
large number of animal deaths that occurred shortly after the
second treatment. Although some blood components were pro-
tected after both 12.5-mg/kg doses of BCNU in ada-positive
compared with neo-positive mice, only one of the seven animals
given the latter gene was alive on day 70, precluding a mean-
ingful statistical test of differences in blood profiles. Nonethe-
less, the fact that six times more ada-positive than neo-positive
mice survived this treatment cleanly indicates a protective effect
from the bacterial enzyme. The low frequency of deaths in both
treatment groups at doses below 12.5 mg/kg and the equally
high frequency at 15 mg/kg suggest a very narrow BCNU dose
range, over which the therapeutic benefits of ATase gene trans-
fen could be realized. This is consistent with the known steep
dose response for BCNU (44).
One might be able to enhance the protective effects of ada
gene transfer by increasing the level of gene expression. Assay
of ATase activity in transduced mouse peripheral blood leuko-
cytes demonstrated a relatively low level ofada expression (Fig.
9). The pattern of ada expression in these cells is probably
different from that in the rapidly proliferating stem cells from
which they were derived and that are the presumed target for
BCNU. The present data suggest that the erythnoid progenitor
cells from which RBCs and platelets are derived express a
higher level of ada ATase, because it was in these lineages that
protection was evident. Because the blood was taken for ATase
assay 21 weeks after transplantation, it is also possible that the
measured levels of ada expressed undernepresent the level of
expression at the time of drug treatment at weeks S and 1 1 after
reconstitution.
The question also arises as to whether the bacterial ada
ATase, with no nuclear localization signal, is able to accumulate
in the cell nucleus (19). Its ability to protect mammalian cells in
vitro against DNA damage (16-20) suggests that at least some
of the protein reaches the nucleus, but it is unknown how much
ATase activity measured in these cells is available for DNA
repair in the nucleus. The human ATase also lacks a recogniz-
able nuclear localization signal, yet it is localized there (45-47),
possibly as a result of its DNA binding properties. To address
these issues in future studies, we plan to use a mutant human
ATase that resists inhibition by O”-BG (48).
We have demonstrated that ada gene transfer to munine
hematopoietic cells confers protection against the toxic effects
of O”-BG plus BCNU, both in vitro and in viva. Such selective
protection of the bone marrow while retaining the capacity for
sensitizing tumor cells to BCNU with O”-BG provides a stnat-
egy for increasing the therapeutic index for BCNU.
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Clinical Cancer Research 1367
ACKNOWLEDGMENTS
We thank Deepthi Jayawardene and the St. Jude Biostatistics
Department for help with statistical analysis. We also thank Pam
Cheshire and Joanna Remack for their excellent technical assistance,
John Gilbert for editing the manuscript, and Dr. Arthur W. Nienhuis for
his helpful comments and discussions.
REFERENCES
1. Pegg, A. E. Mammalian 06-alkylguanine-DNA alkyltransferase: reg-ulation and importance in response to alkylating carcinogenic and ther-
apeutic agents. Cancer Res., 50: 6119-6129, 1990.
2. Pegg, A. E., and Byers, T. L. Repair of DNA containing O”-abkyl-guanine. FASEB J., 6: 2302-2310, 1992.
3. Brent, T. P. Suppression of cross-link formation in chloroethylnitro-
sourea-treated DNA by an activity in extracts of human leukemic
bymphoblasts. Cancer Res., 44: 1887-1892, 1984.
4. Erickson, L. C., Bradley, M. 0., and Kohn, K. W. Measurements of
DNA damage in Chinese hamster cells treated with equitoxic andequimutagenic doses of nitrosoureas. Cancer Res., 38: 3379-3384,
1978.
5. Carter, S. K., Schabel, F. M., Bnoder, L. E., and Johnston, T. P.1,3-Bis(2-chboroethyl)-l-nitrosourea (BCNU) and other nitrosoureas incancer treatment: a review. Adv. Cancer Res., /6: 273-332, 1972.
6. Brent, T. P., Houghton, P. J., and Houghton, J. A. 06-alkylguanine-
DNA alkyltransferase activity correlates with the therapeutic response
of human rhabdomyosarcoma xenografts to 1-(2-chboroethyl)-3-(trans-4-methybcyclohexyl)-1 -nitrosourea. Proc. NatI. Acad. Sci. USA, 82:
2985-2989, 1985.
7. Gerson, S. L., Trey, J. E., Miller, K., and Bergen, N. B. Comparisonof 06-alkybguanine-DNA alkyltransferase activity based on cellularDNA content in human, rat and mouse tissues. Carcinogenesis, 7:
745-749, 1986.
8. Gerson, S. L., Trey, J. E., Miller, K., and Benjamin, E. Repair of06-abkylguanine during DNA synthesis in murine bone marrow hema-topoietic precursors. Cancer Res., 47: 89-95, 1987.
9. Philips, G. L., Wolff, S. N., Fay, J. W., Herzig, R. H., Lazarus, H. M.,
Schobd, C., and Herzig, G. P. Intensive 1,3-bis-(2-chloroethyl)-1-nitro-sourea (BCNU) monotherapy and autobogous bone marrow tnansplan-tation for malignant glioma. J. Clin. Oncob., 4: 639-645, 1986.
10. Sorrentino, B. P., Brandt, S. T., Bodine, D., Gottesman, M., Pastan,
I., Cline, A., and Nienhuis, A. W. Selection of drug-resistant bone
marrow cells in vito after retroviral transfer of human MDRI. Science
(Washington DC), 257: 99-103, 1992.
11. Williams, D. A., Hsieh, K., DeSilva, A., and Mulligan, R. C.Protection of bone marrow transplant recipients from lethal doses of
methotrexate by the generation of methotrexate-resistant bone marrow.J. Exp. Med., 166: 210-218, 1987.
12. Corey, C. A., DeSilva, A., Holland, C., and Williams, D. A. Serial
transplantation of methotrexate resistant bone marrow: protection of
murine recipients from drug toxicity by progeny of transduced stem
cells. Blood, 75: 337-343, 1990.
13. Li, M-X., Banerjee, D., Zhao, S-C., Schweitzer, B. I., Mineishi, S.,
Gilboa, E., and Bertino, J. R. Development of a retroviral construct
containing a human mutated dihydrofobate reductase eDNA for hema-
topoietic stem cell transduction. Blood, 83: 3403-3408, 1994.
14. Margison, G. P., Cooper, D. P., and Brennand, J. Cloning of the E.coli 06-methylguanine and methylphosphotniesten gene using a func-
tionab DNA repair assay. Nucleic Acids Res., /3: 1939-1952, 1985.
15. Olsson, M., and Lindahl, T. Repair of alkybated DNA in Esche-
richia coli: methyl group transfer from Ob�methybguanine to a protein
cysteine residue. J. Blob. Chem., 255: 1069-1071, 1980.
16. Brennand, J., and Margison, G. P. Reduction in the toxicity and
mutagenicity of alkylating agents in mammalian cells harbouning the E.coli alkyltransferase. Proc. NatI. Acad. Sci. USA, 83: 6292-6296, 1986.
17. Samson, L., Derfier, B., and Waldstein, E. A. Suppression of human
DNA alkybation repair defects by E. coli DNA repair genes. Proc. Natl.Acad. Sd. USA, 83: 5607-5610, 1986.
18. Kataoka, H., Hall, J., and Karran, P. Complementation of sensitivity
to alkylating agents in E. coli and Chinese hamster ovary cells byexpression of a cloned bacterial DNA repair gene. EMBO J., 5: 3195-
3200, 1986.
19. Dumenco, L. L., Warman, B., Hatzogbou, M., Lim, I. K., Abboud,S. L., and Gerson, S. L. Increase in nitrosourea resistance in mammalian
cells by retroviralby mediated gene transfer of bacterial 06-alkybgua-nine-DNA alkyltransferase. Cancer Res., 49: 6044-6051, 1989.
20. Jelinek, J., Kleibl, K., Dexter, T. M., and Mangison, G. P. Trans-
fection of munine multi-potent haemopoietic stem cells with an E. coliDNA alkyltransferase gene confers resistance to the toxic effects ofalkylating agents. Cancinogenesis, 9: 81-87, 1988.
21. Doban, M. E., Pegg, A. E., Dumenco, L. L., Moschel, R. C., and
Gerson, S. L. Comparison of the inactivation of mammalian and bac-terial Ob�alkylguanine�DNA alkyltransferase by 06-benzybguanine and
06-methylguanine. Carcinogenesis, 12: 2305-2309, 1991.
22. Crone, T. M., and Pegg, A. E. A single amino acid change in human
06-alkybguanine-DNA-alkyltransfenase decreases sensitivity to inactiva-tion by 06-benzylguanine. Cancer Res., 53: 4750-4753, 1993.
23. Dolan, M. E., Stine, L., Mitchell, R. B., Moschel, R. C., and Pegg,
A. E. Modulation of mammalian 06-alkylguanine-DNA alkyltransferasein vivo by 06-benzybguanine and its effects on the sensitivity of a human
glioma tumor to 1-(2-chloroethyl)-3-(4-methylcycbohexyl)-1-nitno-
sourea. Cancer Commun., 2: 371-377, 1990.
24. Mitchell, R. B., Moscheb, R. C., and Dolan, M. E. Effect of
O�’-benzylguanine on the sensitivity of human tumor xenografts to1,3-bis(2-chboroethyl)-1-nitrosourea and on DNA interstrand cross-linkformation. Cancer Res., 52: 1171-1175, 1992.
25. Doban, M. E., Pegg, A. E., Biser, N. D., Moschel, R. C., and
English, H. F. Effect of 06-benzybguanine on the response to 1,3-bis(2-
chboroethyl)-l-nitrosourea in the Dunning R3327G model of prostaticcancer. Cancer Chemothen. Pharmacol., 32: 221-225, 1993.
26. Felker, G. M., Friedman, H. S., Dolan, M. E., Moschel, R. C., and
Schold, C. Treatment of subcutaneous and intracranial brain tumorxenografts with 06-benzylguanine and l,3-bis(2-chloroethyb)-1-nitro-sourea. Cancer Chemothen. Pharmacob., 32: 471-476, 1993.
27. Brennand, J., and Margison, G. P. Expression of the E. coli O�-methybguanine-methylphosphotniester methyltransfenase gene in mam-malian cells. Carcinogenesis, 7: 185-188, 1986.
28. McLachbin, J. R., Mittereder, N., Daucher, M. B., Kadan, M., andEglitis, M. A. Factors affecting retroviral function and structural integ-
nity. Virology, 195: 1-5, 1993.
29. Bodine, D. M., Moritz, T., Donahue, R. E., Luskey, B. D., Kessler,
S. W., Martin, D. I. K., Onkin, S. H., Nienhuis, A. W., and Williams,D. A. Long-term in vivo expression of a munine adenosine deaminasegene in Rhesus monkey hematopoietic cells of multiple lineages afterretroviral mediated gene transfer into CD34� bone marrow cells. Blood,
82: 1975-1980, 1993.
30. Armentano, D., Yu, S-F., Kantoff, P. W., von Ruden, T., Anderson,
W. F., and Gibboa, E. Effect of internal viral sequences on the utility of
retroviral vectors. J. Vinol., 61: 1647-1650, 1987.
31. Markowitz, D, Goff, S., and Bank, A. A safe packaging line forgene transfer separating viral genes on two different plasmids. J. Virob.,
62: 1120-1124, 1988.
32. Miller, A. D., and Buttimore, C. Redesign of retrovirus packaging
cell lines to avoid recombination leading to helper virus production.Mol. Cell. Biol., 6: 2895-2902, 1986.
33. Chen, C., Okayama, H. High efficiency transformation of mamma-
han cells by plasmid DNA. Mob. Cell. Biob., 7: 2745-2752, 1987.
34. Sambrook, J., Fnitsch, E. F., and Maniatis, T. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory, 1989.
35. Brent, T. P. Isolation and purification of 06-alkylguanine-DNAalkyltransferase from human leukemic cells. Prevention of chboroeth-
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1368 Alkyltransfenase Gene Expression in Munine Bone Marrow
ylnitrosoures-induced cross-links by purified enzyme. Phanmacol. &
Ther., 31: 121-140, 1985.
36. Wu, R. S., Hust-Calderone, S., and Kohn, K. W. Measuement of06-alkylguanine-DNA alkyltransferase activity in human cells and tu-
mon tissue by restriction endonuclease inhibition. Cancer Res., 47:
6229-6235, 1987.
37. Marathi, U. K., Dolan, M. E., and Erickson, L. C. Extended deple-tion of O�-methylguanine-DNA methyltransferase activity following06-benzyl-2’-deoxyguanosine or 06-benzylguanine combined with
streptozotocin treatment enhances l,3-bis(2-chloroethyl)-1-nitrosouneacytotoxicity. Cancer Res., 54: 4371-4375, 1994.
38. Bodine, D. M., McDonagh, K. T., Seidel, N. E., and Nienhuis,A. W. Survival and retrovirus infection of murine hematopoietic stemcells in vitro: effects of 5-FU and method of infection. Exp. Hematol.,19: 206-212, 1991.
39. Bodine, D. M., Karlsson, S., and Nienhuis, A. W. Combination ofinterleukins 3 and 6 preserves stem cell function in culture and enhances
netrovirus-mediated gene transfer into hematopoietic stem cells. Proc.
NatI. Acad. Sci. USA, 86: 8897-8901, 1989.
40. Crowder, M. J., and Hand, D. J. Analysis of Repeated Measures.New York: Chapman and Hall, 1991.
41. W. J. Dixon (ed.). BMDP Statistical Software Manual. Los Ange-les: University of California Press, 1992.
42. Kaplan, E. L., and Meien, P. Nonparametnic estimation from incom-plete observations. J. Am. Statist. Assoc., 53: 457-481, 1958.
43. StatXact Statistical Software for Exact Nonparametnic Inference.Cambridge, MA: Cytel Software Corp., 1992.
44. Mitchell, R. B., Moschel, R. C., and Dolan, M. E. Effect of06-benzylguanine on the sensitivity of human tumor xenognafts to
1,3-bis(2-chloroethyl)-1-nitrosourea and on intenstrand cross-link for-mation. Cancer Res., 52: 1171-1175, 1992.
45. Lee, S. M., Rafferty, J. A., Elder, R. H., Fan, C-Y., Bromley, M.,Harris, M., Thatcher, N., Potter, P. M., Altermatt, H. J., Peninat-Frey, T.,Cemy, T., O’Connor, P. J., and Mangison, 0. P. Immunohistologicalexamination of the inter- and intracellular distribution of 06-alkylgua-nine-DNA alkyltnansferase in human liver and melanoma. Br. J. Cancer,66: 355-360, 1992.
46. Aya, T. C., Loh, K. C., Ali, R. B., and Li, B. F. L. Intracellularlocalization of human DNA repair enzyme methylguanine-DNA meth-yltransferase by antibodies and its importance. Cancer Res., 52: 6423-
6430, 1992.
47. Brent, T. P., von Wnonski, M. A., Edwards, C. C., Bromley, M.,Margison, 0. P., Rafferty, J. A., Pegnam, C. N., and Bigner, D. D.Identification of nitrosourea-nesistant human rhabdomyosarcomas by in
situ immunostaining of O�-methylguanine-DNA methyltransfenase. On-cob. Res., 5: 83-86, 1993.
48. Crone, T. M., Goodtzova, K., Edara, S., and Pegg, A. E. Mutationsin human 06-alkylguanine-DNA alkyltransfenase imparting resistance to
O�-benzylguanine. Cancer Res., 54: 6221-6227, 1994.
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1995;1:1359-1368. Clin Cancer Res L C Harris, U K Marathi, C C Edwards, et al. cytotoxicity.murine bone marrow protects against chloroethylnitrosourea Retroviral transfer of a bacterial alkyltransferase gene into
Updated version
http://clincancerres.aacrjournals.org/content/1/11/1359
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
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://clincancerres.aacrjournals.org/content/1/11/1359To request permission to re-use all or part of this article, use this link
Research. on February 2, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from