Co-expression of the herpes simplex virus thymidine kinase ... - Nature

Gene Therapy (1997) 4, 570–576  1997 Stockton Press All rights reserved 0969-7128/97 $12.00

Co-expression of the herpes simplex virus thymidine kinase gene potentiates methotrexate resistance conferred by transfer of a mutated dihydrofolate reductase gene S Mineishi1,2,3, S Nakahara1, N Takebe1,2, D Banerjee1, S-C Zhao1 and JR Bertino1,2 1

Program of Molecular Pharmacology and Experimental Therapeutics, and 2Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA

We have previously shown that transfer of a mutated dihydrofolate reductase (DHFR) confers resistance to methotrexate (MTX) to infected cells. We report herein the construction of a retrovirus vector, DC/SV6S31tk, which carries the herpes simplex virus thymidine kinase gene (HSVtk) as well as the mutated Serine 31 DHFR (S31) cDNA. 3T3 cells infected with DC/SV6S31tk are more resistant to MTX than cells infected with DC/SV6S31, which carries the S31 and Neor gene. In DC/SV6S31tkinfected cells, a fraction of cells (20–40%) were more resistant to MTX compared with DC/SV6S31-infected cells, and these cells survived a 5-day exposure to 200 mM of

MTX. The mechanism of this augmented resistance is attributed to the salvage of thymidine by HSVtk, as the augmentation is reversed when dialyzed serum is used for cytotoxicity assays. The cells that survive high-dose MTX selection have high levels of expression of S31 DHFR and HSVtk, although copy numbers of the proviral sequences do not increase significantly. Transduction of cells with the DC/SV6S31tk vector also sensitizes cells to ganciclovir (GCV). Co-expression of a metabolically related gene in a retroviral vector to potentiate the resistance imparted by a drug resistance gene may be useful for gene therapy for cancer patients.

Keywords: drug resistance; dihydrofolate reductase; thymidine kinase; methotrexate; gene therapy

Introduction One use of gene transfer for cancer treatment is to transfer drug resistance genes into bone marrow cells to protect bone marrow from toxicity of high-dose chemotherapy.1 For this purpose, mutated dihydrofolate reductase (DHFR) 2–4 genes to produce methotrexate (MTX) resistance, and the multidrug resistance gene (Mdr1) 5–7 have received the most attention, although other drug resistance genes such as aldehyde dehydrogenase class I are also being studied.8 As the level of resistance obtained from transfer of one or a few copies of these cDNAs by retroviral vectors into human or large animal marrow cells is modest, efforts have been made to potentiate the resistance conferred by this method. Hematopoietic progenitor cells transduced in vitro can repopulate in bone marrow even without myeloablation, although a large number of hematopoietic cells may be required for engraftment. 9,10 The idea of selection with a drug resistance gene is thus attractive, as drug selection may facilitate engraftment of transduced progenitor cells without myeloablation (Zhao S-C, Banerjee D, Mineishi S and Bertino JR, unpublished observations). An altered Correspondence: S Mineishi, K6/564 Clinical Science Center, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA 3 Current address: Division of Bone Marrow Transplant, University of Wisconsin, Madison, WI 53792, USA Received 24 June 1996; accepted 4 March 1997

DHFR gene is well suited for protection of hematopoietic cells, as the DHFR coding region consists of only 564 bp, is easy to express and several drug resistant mutants are available for study.11–16 In addition, the selection agent, MTX, has been used extensively for treatment of nonmalignant diseases as well as malignant neoplasms, and the spectrum of side-effects is well known.17 In this study, we show that resistance to MTX may be augmented by increasing thymidine salvage by utilizing a retroviral vector containing both a mutated DHFR and the HSVtk cDNA, thus bypassing the block on synthesis of thymidylate. In addition, this construct will also allow eradication of inadvertently transduced tumor cells, a possible drawback of transfer of drug resistance genes into bone marrow cells contaminated by tumor cells. The HSVtk gene sensitizes cells to a prodrug, ganciclovir (GCV), and has been used extensively in gene therapy experiments and clinical applications as a suicide gene.18

Results Retrovirus, viral producer cell lines The retrovirus vector, DC/SV6S31tk used in this study, is shown in Figure 1. DC/SV6S31 has the same structure, except that it contains the neomycin resistance gene (Neor) in place of the HSVtk gene. The amphotropic retroviral producer cell lines, GP-envAM12-DC/SV6S31 and GP-envAM12-DC/SV6S31tk, have approximate titers of 1 × 105 c.f.u./ml.

Co-expression of a mutated DHFR and HSVtk S Mineishi et al

571

Figure 1 The structure of the DC/SV6S31tk retrovirus vector. The DC/SV6S31 vector has the same structure as DC/SV6S31tk except that Neor cDNA is in place of HSVtk cDNA.

Cytotoxicity of MTX to 3T3 cells and virus-infected cells The ID50 of MTX in the cells infected with the DC/SV6S31tk vector is increased about 20-fold over noninfected cells and 10-fold compared with DC/SV6S31infected cells, as measured by the XTT colorimetric assay (Figure 2). However, approximately 20% of the cells infected with the DC/SV6S31tk vector, but not with the DC/SV6S31 vector, were very resistant to MTX, and survived a 5-day exposure to .640 mm of MTX. After exposure to 10 mm of MTX for 5 days, most of the cells grew back again when medium was changed to a nondrug-containing one. At a concentration of 200 mm of MTX for 5 days, only a small fraction of cells were viable (data not shown). In contrast, when dialyzed serum was used instead of undialyzed fetal calf serum, a fraction of cells surviving a high dose of MTX was not seen, and essentially all cells died at concentrations greater than 640 nm (Figure 3).

Figure 3 XTT colorimetric assay of the cytotoxicity of MTX, using dialyzed serum. 3T3 cells (– + –), 3T3 cells infected with the DC/SV6S31 vector (– p –), and the DC/SV6S31tk vector (– g –). Except for the use of dialyzed serum, experiments were done in the same way as Figure 2.

Cytotoxicity of ganciclovir to virus-transduced cells DC/SV6S31tk and DC/SV6S31 infected cells were exposed to ganciclovir (GCV) in 96-well plates for 5 days as described (see Materials and methods). As shown in Figure 4, the ID50 of ganciclovir to HSVtk-containing cells was 500-fold less than non-tk containing cells, indicating that cells selected with MTX also expressed HSVtk. Expression of mRNA for mDHFR and HSVtk In order to measure the expression of the transduced cDNAs, Northern analysis was performed (see Materials Figure 4 XTT assay of the cytotoxicity of GCV. 3T3 cells infected with the DC/SV6S31 vector (-– g –-) and the DC/SV6S31tk vector (– p –). After 5 days of incubation with GCV, cell viability was measured by XTT assay.

Figure 2 XTT colorimetric assay of the cytotoxicity of MTX. 3T3 cells (– + –), 3T3 cells infected with the DC/SV6S31 vector (– p –), and the DC/SV6S31tk vector (– g –). Cells were transduced and XTT assay was performed as described in Materials and methods.

and methods). The expression of the mutated human DHFR gene was approximately the same in cells infected with the DC/SV6S31 vector compared to the DC/SV6S31tk vector (Figure 5). The increased protection of the cells from MTX toxicity therefore was not due to increased expression of mutated DHFR in cells infected with the DC/SV6S31tk vector, but secondary to the expression of HSVtk in this construct. Of interest, cells surviving 3 weeks of 1 mm MTX showed an increased level of HSVtk mRNA. In order to determine if increased expression resulted from MTX exposure, we compared the expression of DHFR and HSVtk before and after exposure to various concentrations of MTX for a period of 3 weeks. Expression levels of both DHFR and HSVtk increased with exposure of cells to MTX (Figure 6). It is of note that


572

Figure 5 Northern blot analysis of 3T3 cells transduced with either the DC/SV6S31 or the DC/SV6S31 tk vector. RNA was prepared as described, and Northern analysis performed as described in Materials and methods. (a) RNA was hybridized with a DHFR probe; (b) with a HSVtk probe. Lane 1, 3T3 cells; lanes 2, 3, 3T3 cells infected with the DC/SV6S31 vector; lane 4, 3T3 cells infected with the DC/SV6S31tk vector; lane 5, 3T3 cells infected with the DC/SV6S31tk vector, selected with 1 mm of MTX for 3 weeks. Lanes are numbered the same in both panels. Arrows indicate mRNA of DHFR (0.9 kb) and HSVtk (4 kb), respectively.

Figure 6 Northern blot analysis. Expression of DHFR and HSVtk mRNA was measured with Northern blot analysis. (a) RNA was hybridized with a DHFR probe; (b) with a HSVtk probe. Lane 1, 3T3 cells; lane 2, 3T3 cells transduced with the DC/SV6S31 vector; lane 3, 3T3 cells transduced with the DC/SV6S31tk vector; lanes 4–7, 3T3 cells transduced with the DC/SV6S31tk vector, selected with MTX, 300 nm for 1 week (4), 1 mm for 1 week (5) and 1 mm for 3 weeks (6,7). Lanes are numbered the same in both panels. Arrows indicate mRNA of DHFR (0.9 kb) and HSVtk (4 kb), respectively. The RNA loading was standardized using 36B4 as a control and results are shown as ‘relative mRNA levels’. This experiment was repeated three times with similar results.

HSVtk expression levels were higher after 1 week in the MTX selection, whereas DHFR expression was higher after a 3-week period of selection. These experiments were repeated three times with similar results.

infected with DC/SV6S31tk than the cells infected with DC/SV6S31 or mock-infected cells (Table 1).

Analysis of gene amplification in transduced cells Southern blot analysis was performed to compare the average copy numbers of the proviral sequences in the transduced cell lines. After selection with 1 mm of MTX, despite increased resistance to MTX and increased expression of DHFR and HSVtk by Northern blot, cells did not contain increased copy numbers of DHFR or HSVtk (Figure 7).

Co-expression of HSVtk with Ser31-mutated DHFR augments resistance to MTX. Although the increase in the ID50 was only 10-fold compared to DC/SV6S31, approximately 20% of cells infected with the DC/SV6S31tk vector survived a very high concentration of MTX. The mechanism of this high level of resistance in a subpopulation of cells is probably due to the rescue of MTX toxicity by increased salvage of thymidine as a result of increased levels of HSVtk in these cells (Figure 8). As expected, after dialyzing serum to remove thymidine, the augmented resistance was lost, as shown in Figure 3. HSVtk is commonly used as a dominant selectable marker using HAT (hypoxanthine-aminopterin-thymidine) medium,19 which contains aminopterin, an analog of MTX. 12 Thus the

Transduction of mouse bone marrow in vitro Mouse bone marrow cells were infected with the DC/SV6S31 and DC/SV6S31tk vectors, then cultured 12 days in the presence of MTX at concentrations of 1 and 5 × 10−8 m. More CFU-C colonies were seen in cells

Discussion


573

Figure 8 Enzymatic pathways related to DHFR and HSVtk. MTX or polyglutamates (MTX(GLU)n) inhibits DHFR as well as de novo purine biosynthesis. HSVtk increases the phosphorylation of thymidine to salvage the thymidine pathway, as well as increases the phosphorylation of the ganciclovir to convert the drug into an inhibitor of DNA synthesis.

Figure 7 Southern blot analysis. (a) HSVtk probe (upper panel) and DHFR probe (lower panel). Lane 1, 3T3 cells; lanes 2, 3, 3T3 cells infected with the DC/SV6S31 vector; lanes 4, 5, 3T3 cells infected with the DC/SV6S31tk vector; lanes 6–9, 3T3 cells infected with the DC/SV6S31tk vector and selected with MTX. Cells in lanes 6 and 7 were treated with 1 mm MTX for 1 week; cells in lanes 8 and 9 with 1 mm MTX for 3 weeks, see Materials and methods for detail. (b) Ethidium bromide staining of the gel to show equal loading.

expression of HSVtk itself may cause partial resistance to MTX when exogenous thymidine is available. Surprisingly, the increased expression noted after MTX exposure for 3 weeks was not secondary to increased cDNA copy numbers, as DHFR mRNA increased with no increase in gene copy number. This increase of mRNA may be due to selection of clones which have higher expression of the mDHFR and the HSVtk genes. More investigation is warranted to clarify this mechanism. We previously reported that partial rescue of the purine biosynthesis pathway by co-expression of xanthine-guanine-phosphoribosyl transferase (XGPRT) can

also augment the resistance to MTX conferred by the S31 mutated DHFR.20 Thus, co-expression of metabolically related genes may potentiate the resistance conferred by a drug resistance gene. The present studies show an advantage of the MTX-DHFR selection system as DHFR is a key enzyme in both thymidilate and purine biosynthesis. Thus, resistance to MTX may be increased by coexpression of genes involved in these pathways (Figure 8). The observation that some fraction of the cells become highly resistant to MTX in the DHFR-HSVtk system was not seen in the DHFR-XGPRT system, and together with the data presented here suggests that the high level of resistance observed with a fraction of cells is due to their ability to salvage thymidine more effectively than cells transduced with mDHFR alone. Another important aspect of the construct which contains HSVtk is that this vector also confers sensitivity to GCV to the infected cells. As bone marrow/peripheral stem cells may be contaminated with tumor cells even after CD34 selection, and inadvertent transduction of tumor cells may result in production of drug-resistant tumor cells, a suicide gene in the same retroviral construct may provide a safeguard against this possibility. Alternatively, constructs containing both these genes may be used to infect tumor cells followed by treatment with MTX to amplify both DHFR and HSVtk followed by treatment with GCV to preferentially eradicate these cells.

Table 1 CFU assays of mouse bone marrow cells infected with retroviral vectors MTX (nm) 0 10 50

AM12

DC/SV6S31

DC/SV6S31tk

100 8.6 ± 2.9 0±0

100 29.5 ± 3.9 18.2 ± 4.0

100 40.0 ± 10.0 36.6 ± 5.8

The experiments were done in triplicate. Results are shown with colony numbers without MTX as 100.


574

HSVtk-containing vectors are also being used for adoptive immunotherapy.21 In certain protocols, T cells transduced with a vector containing HSVtk and a nerve growth factor receptor were selected in vitro after transduction. Transduced T cells can be eliminated with GCV in vivo, but can not be positively selected in vivo. On the other hand, DC/SV6S31tk is a vector which allows both positive selection (with MTX) and negative selection (with GCV) in vivo using clinically approved drugs.

Materials and methods Cells NIH3T3 cells are cultured in Dulbecco’s modified Eaglehigh glucose (with 4.5 g/l of glucose) (DME-HG) medium with 10% fetal bovine serum (FBS). The viral packaging cell lines, GP-E86 and GP-envAM12 line, an ecotropic and an amphotropic line, respectively, have been described previously.22,23 Retrovirus constructs The pSV4S31 mammalian expression vector, which contains the SV40 early promoter and a Serine 31 mutated DHFR cDNA,12 was used as a template to obtain a full length Ser31 mutated DHFR cDNA by PCR. Primers used for this reaction were: GTHindIII-1 TTCAGTATAAGCTTATCCGCGTGCTGTCATGGTTGGTTCGCT and GT564BglII ACAGAACAAGATCTCCTTCATATTAATCATTCTTCTCATA. PCR conditions were 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s, for 35 cycles. The PCR product obtained was digested with HindIII and BglII, and subcloned into a pSV2 vector which was previously digested with HindIII/BglII, to generate the pSV6S31 mammalian expression vector. The pSV6S31 construct contains only the DHFR coding region, and was found to be at least 10-fold more efficient than pSV4S31, which contains approximately 90 bp of the 3′-untranslated region of DHFR cDNA. PvuII/BglII digestion releases the fragment of DNA which contains the SV40 early promoter and S31 DHFR cDNA. This fragment was blunt-ended with Klenow enzyme, and inserted into the SnaBI site of the N2A Moloney murine leukemia virus vector24 to produce DC/SV6S31. This vector contains the SV40 promoter and the S31 DHFR cDNA cloned into the 3′-LTR. Thus upon infection of target cells, this expression unit is duplicated in the 5′-LTR, to form a double copy structure.25,26 This vector was digested with BclI at 65°C overnight, bluntended with Klenow enzyme, and then cut with XhoI. A HSVtk-containing vector, TEL-1 (kind gift of Dr Michel Sadelain, MSKCC)27 was digested with XbaI/XhoI, and the released fragment cloned into pBluescriptSKII (Stratagene, La Jolla, CA, USA) and digested with the same combination of enzymes. The resulting vector, pBHSVtk was digested with ScaI, blunt-ended with T4 DNA polymerase, then digested with XhoI and cloned into the blunt-end/XhoI fragment of DC/SV6S31, resulting in DC/SV6S31tk (Figure 1). Retrovirus producer cell lines Retrovirus vector constructs, DC/SV6S31, and DC/SV6S31tk were either transfected alone or cotransfected with pSV2Neo, using a DOTAP transfection kit (Boehringer Mannheim, Indianapolis, IN, USA) into GP-

E86 cells and GP-AM12 cells. After a 14-day selection with 750 mg/ml of G418, colonies with high titers were selected, and the supernatants from the ecotropic producer cell lines were used to infect amphotropic producer cell lines two to three times. Titers of these producer cell lines were measured with 3T3 cells as target cells, and the highest titer colonies were used for further experiments. The titers of these colonies were approximately 1 × 105 c.f.u./ml. Producer cells were replated in 60 mm plates, grown up to 70% confluency, then medium was changed to either DME-HG with 10% FBS for 3T3 cell infection or IMDM-10% FBS medium for bone marrow cell infections. The next day the supernatant was collected, centrifuged at 2000 g for 20 min to remove cell debris, and stored frozen at −80°C until use. The titers were measured with NIH3T3 cells as target cells. Briefly, 1 × 105 cells were plated in 60 mm plates and appropriately diluted virus supernatant was added to the plates, in the presence of 8 mg/ml of polybrene. After 3 h of exposure to diluted supernatant, normal medium was added. The next day, medium was changed to the one containing 100 nm of MTX. This concentration of MTX is known to kill all noninfected cells but have a minimal effect on infected cells. The colonies were counted 2 weeks later, and the titer was calculated and is expressed as c.f.u./ml. Helper virus production was monitored with vector rescue assays.28 In brief, 3T3 cells transfected with retrovirus vector containing the Neor gene were infected with viral supernatant. After a week of culture, supernatant from these cells was harvested and wild-type 3T3 cells exposed to these supernatants. The cells were then subjected to G418 selection. No helper virus was detected in viral supernatants.

Cytotoxicity of MTX and GCV to the cells infected with retroviral vectors 3T3 cells were exposed to the supernatant containing retroviral vectors for 3 h, in the presence of 8 mg/ml of polybrene. Multiplicity of infection (MOI) was more than 10 for most of these experiments. After 3 h, fresh medium was added, then at 24 h medium was changed to DMEHG containing 20 nm of trimetrexate (TMTX) for 3-day selection. This concentration of TMTX is 100% lethal to noninfected 3T3 cells, but is not inhibitory to infected cells. Also, this 3-day selection does not change the ID50 of infected cells (data not shown). After the 3-day selection, cells were plated in 96-well plates, and cytotoxicity measured by a colorimetric assay using XTT, sodium 3′(1-((phenylamino)-carbonyl)-3,4-tetrazolium)-bis(4-methoxy-6-nitro)benzene-sulfonic acid hydrate, as described previously.29 Cells were plated at a concentration of 400 cells per well, and 24 h later, medium was changed to a drug-containing one. After 5 days of exposure to drugs, 50 ml of XTT at a concentration of 1 mg/ml, was added together with phenazine methosulfate (PMS). After a 3h incubation at 37°C, the OD at 450 nm was measured using a 96-well plate reader. The absorbance was plotted with wells without drug (with cells) as 100% and wells without cells as 0%. The experiments were done both with pooled cells and selected single colonies. The results shown here were obtained from experiments with pooled cells. ID50 is defined here as the concentration of the drug which inhibits growth of the cells by 50% measured by XTT-colorimetric assay.


Southern blotting Cell lines were cultured to semiconfluency. Genomic DNA was extracted using a buffer containing Tris-EDTA, SDS, and proteinase K, followed by phenol/chloroform extraction. Twenty micrograms of genomic DNA was digested with HindIII/SmaI (for DHFR) or EcoRI (for HSVtk), loaded on to 1.5% agarose gels, and after electrophoresis overnight, DNA was transferred to a nitrocellulose membrane as described previously.30 The membrane was dried, UV-crosslinked, and hybridized with a human DHFR probe or a HSVtk probe labeled with 32P-dCTP. After a vigorous wash at 60°C for 1 h to prevent hybridization of the human probe with mouse endogenous DHFR, the blot was visualized with autoradiography, and quantified with a Betascope 603 blot analyzer (Betagen, Waltham, MA, USA). Northern blotting Total RNA was extracted,31 and 15 mg of RNA electrophoresed on a formamide gel at 90 V for 3 h, transferred to a nitrocellulose filter, dried, UV crosslinked and hybridized to 32P-labeled human DHFR or HSVtk and 36B4 acidic ribosomal phosphoprotein probes,32 the latter as a control for loading. Autoradiography and analysis with the Betascope were performed as described above. Mouse bone marrow CFU-C assays CB57/F male mice were injected with 150 mg/kg of 5FU intraperitoneally 4 days before harvest. Bone marrow cells were harvested from femurs and tibiae, then resuspended in Iscove’s modified Dulbecco’s medium (IMDM) with 10% FBS. Retroviral infection was carried out by the coculture method. In brief, bone marrow cells in IMDM 10% FBS were pipetted into 100-mm plates which were approximately 80% confluent with viral producer cells, which were irradiated at 1500 cGy on the day of coculture. The target to producer cell ratio was approximately 1:1. After 48 h at 37°C, nonadherent cells were harvested, washed with PBS, counted, and 3 × 104 cells in 2 ml of semisolid medium containing 1% methylcellulose, 20% FBS, 10% WEHI-3B conditioning medium, 1% NaHCO3, 1% Na-pyruvate, 1 mm b-mercaptoethanol, 100 U/ml penicillin, 100 mg/ml streptomycin, 1% essential amino acids, 1.5% nonessential amino acids, 0.5% vitamin C were added to 10 × 35 mm plates. The plates are kept at 37°C in a 5% CO2 incubator, and colonies of more than 50 cells were scored after 12 days.33

5

6

7

8

9

10

11 12 13

14

15

16 17

18

19 20

21

Acknowledgements This work was supported by a grant from the USPHS (CA 59350-4).

22

References

23

1 Banerjee D et al. Gene therapy utilizing drug resistance genes: a review. Stem Cells 1994; 12: 378–385. 2 Williams DA, Hsieh K, DeSilva A, Mulligan RC. Protection of bone marrow transplant recipients from lethal doses of methotrexate by the generation of methotrexate-resistant bone marrow. J Exp Med 1987; 166: 210–218. 3 Li M-X et al. Development of a retroviral construct containing a human mutated dihydrofolate reductase cDNA for hematopoietic stem cell transduction. Blood 1994; 83: 3403–3408. 4 Flasshove M et al. Retrovirally mediated gene transfer of a mutant human dihydrofolate reductase (DHFR) gene into pro-

24 25

26

27

genitors from human peripheral blood (PB). Blood 1995; 85: 566–574. Guild BC, Mulligan RC, Gros P, Housman DE. Retroviral transfer of a murine cDNA for multidrug resistance confers pleiotropic drug resistance to cells without prior drug selection. Proc Natl Acad Sci USA 1988; 85: 1595–1599. Ueda K, Cardarelli C, Gottesman MM, Pastan I. Expression of full-length cDNA for the human ‘Mdr1’ gene confers resistance to colchicine, doxorubicin, and vinblastine. Proc Natl Acad Sci USA 1987; 84: 3004–3008. Sorrentino BP et al. Selection of drug-resistant bone marrow cells in vivo after retroviral transfer of human Mdr1. Science 1992; 257: 99–103. Magni M et al. Induction of cyclophosphamide-resistance by aldehyde dehydrogenase gene transfer. Blood 1996; 87: 1097– 1103. Stewart FM et al. Long-term engraftment of normal and post-5uracil murine marrow into normal nonmyeloablated mice. Blood 1993; 81: 2566–2571. Schiffman R et al. Transfer of the human glucocerebrosidase gene into hematopoietic stem cells of nonablated recipients: successful engraftment and long-term expression of the transgene. Blood 1995; 86: 1218–1227. Bertino J. Ode to methotrexate. J Clin Oncol 1993; 11: 5–14. Schweitzer BI, Dicker AP, Bertino JR. Dihydrofolate reductase as a therapeutic target. FASEB J 1990; 4: 2441–2452. Banerjee D et al. Transfection with a cDNA encoding a Ser31 or Ser34 mutant human dihydrofolate reductase into Chinese hamster ovary and mouse marrow progenitor cells confers methotrexate resistance. Gene 1994; 139: 269–274. Dicker AP et al. Methotrexate resistance in an in vivo mouse tumor due to a nonactive-site dihydrofolate reductase mutation. Proc Natl Acad Sci USA 1993; 90: 11797–11801. Banerjee D et al. Transfection of a nonactive site mutant murine DHFR cDNA (the tryptophan 15 mutant) into Chinese hamster ovary and mouse marrow progenitor cells imparts MTX resistance in vitro. Cancer Gene Ther 1994; 1: 181–184. Spencer HT et al. A gene transfer strategy for making bone marrow cells resistant to trimetrexate. Blood 1996; 87: 2579–2587. Mineishi S, Bertino JR. Methotrexate. In: McDonald C (ed). Immunomodulatory and Cytotoxic Agents in Dermatology. MarcelDekker, New York, 1997, pp 9–20. Culver KW et al. In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science 1992; 256: 1550–1552. Ausubel FM et al (eds). Current Protocols in Molecular Biology. Wiley: New York, 1987, section 9.5.4. Mineishi S, Banerjee D, Zhao SC, Bertino JR. Construction of a retrovirus vector which confers both augmented resistance to methotrexate and sensitivity to 6-thioxanthine. Proc Am Soc Clin Oncol 1995; 14: A587 (Abstr.). Bonini C et al. Transfer of the HSVtk gene into donor peripheral blood lymphocytes for in vivo immunomodulation of donor antitumor immunity after allo-BMT. Blood 1994; 84 (Suppl. 1): 110a (Abstr.). Markowitz D, Goff S, Bank A. A safe packaging line for gene transfer: separating viral genes on two different plasmids. J Virol 1988; 62: 1120–1124. Markowitz D, Goff S, Bank A. Construction and use of a safe and efficient amphotropic packaging cell line. Virology 1988; 167: 400–406. Armentano D et al. Effect of internal sequences on the utility of retroviral vectors. J Virol 1987; 61: 1647–1650. Hantzopoulos PA, Sullenger BA, Ungers G, Gilboa E. Improved gene expression upon transfer of the adenosine deaminase minigene outside the transcriptional unit of a retroviral vector. Proc Natl Acad Sci USA 1989; 86: 3519–3523. Li M et al. Comparison of the expression of a mutant dihydrofolate reductase under control of different internal promoters in retroviral vectors. Hum Gene Ther 1992; 3: 381–390. Babra D, Hardin J, Sadelain M, Gage FH. Development of anti-

575


576

tumor immunity following thymidine kinase-mediated killing of experimental brain tumors. Proc Natl Acad Sci USA 1994; 91: 4348–4352. 28 Kaleko M, Garcia JV, Osborne WRA, Miller AD. Expression of human adenosine deaminase in mice after transplantation of genetically modified bone marrow. Blood 1990; 75: 1733–1741. 29 Rodgers GH, Hatfield SM, Glasebrook AL. An improved colorimetric assay for cell proliferation and viability utilizing the tetrazolium salt XTT. J Immunol Meth 1991; 142: 257–265. 30 Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975; 98: 503–517.

31 Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156–159. 32 Laborda J. 36B4 cDNA used as an estradiol-independent mRNA control is the cDNA for human acidic ribosomal phosphoprotein PO. Nucleic Acids Res 1991; 19: 3998. 33 Zhao SC et al. Long term protection of recipient mice from lethal doses of methotrexate by marrow transduced with a double copy retrovirus vector containing a mutated DHFR. Cancer Gene Ther 1994; 1: 27–33.