WO1990004632A1

WO1990004632A1 - Transgenic animals for testing multidrug resistance

Info

Publication number: WO1990004632A1
Application number: PCT/US1989/004686
Authority: WO
Inventors: Ira H. Pastan; Michael M. Gottesman
Original assignee: The United States Of America, Represented By The Secretary, United States Department Of Commerce
Priority date: 1988-10-21
Filing date: 1989-10-20
Publication date: 1990-05-03
Also published as: EP0451157A1; AU637986B2; CA2001128A1; AU4494589A; JPH03504560A; IL92070A0; EP0451157A4

Abstract

Transgenic animals carrying and expressing human MDR1 gene have been produced. These transgenic animals serve as a useful model for testing the efficacy of high dosage chemotherapy and for the development of novel chemotherapeutic agents against cancers.

Description

TRANSGENIC ANIMALS FOR TESTING MULTIDRUG RESISTANCE

The present invention is related generally to genetic manipulation of organisms. More particularly, the present invention is related to the production of transgenic animals suitable for testing _in_ vivo the utility of the expression of the multidrug resistance gene and for the development of novel chemotherapeutic agents against cancers.

BACKGROUND OF THE INVENTION

10 Intrinsic and acquired resistance to multiple chemotherapeutic agents is a major clinical problem in the treatment of cancer. Cell lines resistant to multi¬ ple drugs such as Vinca alkaloids, doxorubicin (Adria- mycin), colchicine and actinomycin D have been studied,

¹⁵ including lines derived from a human KB carcinoma cell line after selection in culture for resistance to a sin¬ gle agent (Akiyama, et al, 1985, So at. Cell Mol. Genet, jy 117-126) . One human gene responsible for multidrug resistance, termed MDR1, encodes a 4.5 kb MRNA which is

20 elevated in highly multidrug-resistant cell lines, in some normal tissues and many tumors (Fojo, et al, 1987, Proc. Natl. Acad. Sci. USA 84.265-269) . These tumors are from intrinsically drug resistant cancers of the colon, adrenal, and kidney, as well as tumors that had acquired

25 drug resistance after chemotherapy. The protein product of the MDR1 gene is a 170 kD membrane glycoprotein (P- glycoprotein), which is overexpressed in multidrug- resistant cell lines and acts as a pump to transport chemotherapeutic drugs out of the cell. As expected for

- - a drug efflux pump, P-glycoprotein is located in the plasma membrane of resistant cells, binds both cytotoxic drugs and ATP and, as indicated by sequence analysis, has 12 membrane-spanning domains (Chen et al, 1986, Cell, 47:381-389) . 5 It has also been reported that full-length cloned human MDR1 or mouse mdr cDNAs can confer multidrug resistance on mouse and human drug sensitive cells after transfection or infection with retroviral vectors (Gros, et al, 1986, Nature _323_:728-731. Ueda, et al, 1987, Proc. Natl. Acad. Sci. USA _84_:3004-3008) . However, it has not heretofore been demonstrated by actual experi¬ mental work that acquired drug resistance can result from 5 somatic expression of the human MDR1 gene in vivo through genetic manipulation of the embryo or the germ plasm.

SUMMARY OF THE INVENTION It is, therefore, an object of the present invention to provide transgenic animals carrying and -^■ - expressing the human MDR1 gene.

It is another object of the present invention to provide an animal model for testing the efficacy of high degree chemotherapy against tumors.

Other objects and advantages will become evi- ¹⁵ dent from the following detailed description of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other objects, features and many of the attendant advantages of the invention will be better 0 understood upon a reading of the following detailed description when considered in connection with the accompanying drawings wherein:

Figures 1A and IB are a schematic representa¬ tion of the construction of expression vectors pHGl and 5 pHG2.

Figure 1A is a construction of pHGl and isola¬ tion of the pronuclear injected fragment.

Figure IB is a construction of pHG2. Letters indicate restriction enzymes used for cloning and orien- 0 tation characterization: B, BamHl; E, EcoRI; N, Ndel; S, Sail; X, Xhol. Restriction enzyme sites within parenthe¬ sis indicate those sites destroyed after ligation.

Figures 2A and 2B show the results of Southern hybridization of tail genomic DNA from mice. 5 Figure 2A is a diagrammatic representation of the B-actin promoter-MDR1 fusion transgene integrated in the mouse genome. A 3.1 kb labeled fragment is expected after EcoRI digestion of a mouse genomic DNA and hybridi¬ zation with the 5A probe, derived from the middle part of MDRl cDNA, if the transgene is integrated. Symbols and letters of restriction sites are the same as described in 5 Figure 1.

Figure 2B is a Southern blot analysis of genomic DNA (20 ug) digested with EcoRI and hybridized with the 5A probe under conditions of low stringency (Lanes 1-5) and under conditions of high stringency 0 (Lanes 6-10). Genomic DNA isolated from: KB-3-1 cells (Lanes 1,6); Normal mouse mixed with 10 pg of injected DNA fragment (Lanes 2,7); Normal mouse (Lanes 3,8); Negative mouse from a litter produced after pronuclear injection (Lanes 4,9); transgenic mouse from the same 5 litter (Lanes 5,10).

Figures 3A and 3B show blot-hybridization analyses of DNA from founder mice carrying the MDRl transgene.

Figure 3A is a DNA slot blot analysis of 10 ug 0 tail DNA, denatured and hybridized with 5A probe at high stringency condition. NM, normal mouse; M39-M168, trans¬ genic founder mice; 1 C or 10 C, indicate copy number of the transgene as explained in Figure 2B for lanes 2 and 7. 5 Figure 2B is a Southern blot analysis .of tail

DNA from founder mice described in Figure 3A. Hybridiza¬ tion was performed with 5A probe at conditions of high stringency. The arrowhead points to the 3.1 kb fragment; the arrow shows the fragment of a rearranged product of _Q the transgene.

Figures 4A and 4B show the inheritance of the MDRl transgene in mice generated from founder M39 (Figure

3).

Figure 4A is a pedigree of M39 and descendant mice. Squares, males; circles, females; half-filled symbols, carrier mice; open symbols, noncarriers.

Figure 4B is a Southern blot analysis of tail DNA from F2 generation mice. Mouse numbers correspond to numbers shown in the pedigree above. Molecular weight standards (1 kb ladder) (BRL) are indicated.

Figure 4C is a slot blot analysis of tail DNA from mouse 39 and from its Fl and F2 generation progeny. Probe and hybridization conditions are the same as described in Figures 3A and 3B.

Figure 5 shows the pedigree expression analyses of the MDR-39 mouse line. Squares, males; circles, females; open symbols, noncarrier mice; half-filled symbols at bottom, heterozygotes; completely-filled symbols homozygotes, half-strippled symbols at top, mice expressing the transgene as detected by RNA studies; half-hatched symbols at top, mice expressing the trans- gene as detected by protein immunofluorescence localiza¬ tion. *, mice expressing the transgene in spleen but not in bone marrow. Mice marked by letters are mice for which RNA studies are represented in Figure 6.

Figure 6 shows the results of RNA expression analysis of normal and transgenic mice. A slot blot of total RNA samples (10 ug) extracted from bone marrow and spleen of mice A-E (as indicated in Figure 5) was hybri¬ dized with MDRl 5A probe (left) ^■■ and humanV-actin probe (right) at conditions of high stringency and at condi- tions of low stringency, respectively. 3-1, parental drug-sensitive KB cell line; 8-5 and V-l, multidrug resistant sublines; BM, bone marrow; SP, spleen.

Figure 7 demonstrates the tissue specificity of the transgene expression. Figure 7A is a simplified restriction map of

MDRl cDNA and the probes used. E, EcoRI; P, PVUII.

Figure 7B is a blot hybridization of total RNA (10 ug) extracted from: Upper: 1, KB-8-5 drug resistant cell line; 2, KB-3-1 drug sensitive and KB-V-1 drug resistant cell lines; Lower: 1, MDRl non-carrier mouse; 2, MDRl carrier mouse. Li, liver; Ki kidney; Lu, lung; Ov, ovary; Mu, skeletal muscle; BM, bone marrow; Sp, spleen; He, heart; Br, brain. The same blot was hybri¬ dized with different probes, as indicated.

Figures 8A, 8B, 8C and 8D evidences immuno- fluorescence localization of human P170 in bone marrow cells from an MDRl transgenic mouse. Bone marrow samples from either an MDRl transgenic mouse (Figures 8A, 8B and 8C) or a normal sibling (Figure 8D) were smeared on glass slides, air dried, and then fixed in formaldehyde. The smears were labeled using monoclonal antibody MRK16 (anti-human P170) and indirectly labeled with rhodamine. Equal time exposures show bright expression of P170 in all cells from the MDR mouse (Figures 8A' , 8B' and 8C ) but not in cells from the control mouse (Figure 8D' ) . (Figures 8A, 8B, 8C and 8D) represent phase con- trast images of the cells shown in (Figures A', 8B', 8C and 8D'). (Mags = X 630; bar = 6.5 ) .

DETAILED DESCRIPTION OF THE INVENTION The above and various other objects and advan¬ tages of the present invention are achieved by transgenic animals carrying and expressing the human MDRl gene.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned hereunder are incorporated herein by reference. Unless mentioned otherwise, the techniques employed herein are standard methodologies well known to one of ordinary skill in the art.

As a first step toward producing the transgenic animals, a cloned human MDRl cDNA sequence which contains the MDRl translated sequences as well as the endogenous

3' polyadenylation signal under control of the B-actin promoter was constructed. Although there is only one functional _B-actin gene per haploid genome both in human and mouse, _B_-actin is one of the most abundant proteins in eucaryotic cells and is expressed in a variety of tissues _r irrespective of embryonic origin. Since B-actin is abundantly expressed in all cells and evolutionarily conserved, it was rea¬ soned that the ^-actin promoter may satisfy both require¬ mentst high level expression and activity in a wide range of cell types including drug-sensitive cells. The following materials and methods now exemplify the various steps in producing and testing of the MDRl expressing transgenic mice in accordance with the present invention.

MATERIALS AND METHODS Materials Restriction endonucleases and ₄ DNA ligase were purchased from New England Biolabs or Bethesda Research Laboratories, Inc. , and used under conditions recommended by the supplier. Agarose (Seake , GTG and Sea Plaque, LMB) was supplied by FMC (Rockland, ME). Reagents for PAGE were from Bio-Rad. Colchicine was purchased from Sigma chemicals. All other chemicals were of analytical grade. Cell culture media were obtained from Gibco and sera from M.A. Bioproductε, Inc. [p ]- dCTP and nick-translation kit were purchased from Amersham or New England Nuclear.

Bacterial strains, plasmids, cell lines and mouse lines

_E_. coli strain DH5 was used for transforma¬ tion. . pMDR2000XS which carries the full-length MDR-1 cDNA with a unique Xhol site at the 3' end, was con- structed in this laboratory (Pastan, et al, 1988, Proc. Natl. Acad. Sci. USA 85:4486-4490) . The pGEM2 vector was from Promega Biotech (Madison, WI). pBA-CAT which carries the bacterial CAT gene under control of the chicken J-actin promoter in pUClδ was a gift from Dr. B. Paterson. NIH 3T3 and KB-3-1 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and calf serum, respectively. Mouse lines RCNIH and B6SJL F were obtained from Jackson Laboratories (Bar Harbor, ME). Plasmid construction

A plasmid for the expression of the MDRl cDNA under the control of a chicken _B-actin promoter was con¬ structed as shown in Figure 1A. pB_A-CAT which carries a chicken j^-actin promoter was cut with Sail and a 0.33 kb fragment containing the essential J3-actin promoter sequences was eluted after electrophoresis on an agarose gel. The resulting fragment was inserted into the unique Sail site of pMDR2000XS, which carries the full-length MD l cDNA downstream from the T7 promoter in a pGEM2 vector (Pastan, et al, 1988, Proc. Natl. Acad. Sci. USA _8_5_:4486-4490) . The resulting plasmid has a 330 base pair _B_-actin promoter in two orientations at the 5' end of the MDRl cDNA. Restriction analysis with XhoI/BamHI was used to identify a plasmid with the B_-actin promoter in the proper orientation with respect to the MDRl cDNA. The resulting plasmid termed pHG-1 (pBAP-MDR) was used for microinjection. Since pHG-1 was digested with Xhol before microinjection (see below), the effect of truncat¬ ing the actin promoter with Xhol was determined. Accord¬ ingly, pHG-2 (tBAP-MDR) was constructed from pHG-1 as shown in Figure IB, to test the expression level of MDRl under the control of a truncated B_-actin promoter. pHGl was cut with Xhol and the resulting 4.8 kb and 2.9 kb fragments were eluted after separation on an agarose gel. The 2.9 kb fragment containing 60 base pairs of the 5' region of the JB-actin promoter was cleaved with Sail to remove these 60 bp sequences, dephosphorylated with alkaline phosphatase and ligated to the 4.8 kb fragment which contains the MDRl cDNA down¬ stream from the Xhol-SalI (270 bp) truncated promoter. The resulting plasmid had B_-actin promoter-MDRl sequences in two orientations at the 5' end of the multiple cloning sites of the pGEM2 vector sequences. Restriction analysis with Xhol/Ndel was used to identify a plasmid with the truncated JB^actin promoter-MDRl sequence in the proper orientation with respect to the T7 promoter of the pGEM2 vector. The resulting plasmid, pHG2 (tBAP-MDR) as well as pHGl were examined for their size and expression of the MDRl cDNA in human KB-3-1 and mouse NIH 3T3 trans- fectant cells. Cell culture and transfection

NIH 3T3 mouse cells and human KB-3-1 cells were cultured and transfected by a calcium phosphate precipi- tation method as described by Shen et al, 1986, Mol. Cell Biol. .6_:4039-4044. DNA preparation for microinlection pHGl was cut with Xhol and a 4.7 kb fragment with MDRl downstream from the _B_-actin promoter was separated from the vector fragment by electrophoresis on a low melting point agarose gel (See "Plaque" infra) . The ethidium bromide-stained band containing the BAP-MDRl was excised from the gel and was melted in TE (10 M Tris-HCl ph 8.0, 1 mM EDTA) equilibrated phenol solution at 70°C. Then, 3 volumes of buffer containing 0.25 M NaCl, TE were added and incubated at 70°C for 10 min and at 40°C for 10 min. The aqueous phase was separated, extracted in chloroform (1:1 v/v) and ethanol precipi¬ tated. The DNA pellet was dissolved in 0.2 M NaCl, TE and purified on a NACS column (Bethesda, Research Labs) under conditions recommended by the supplier except that the DNA was eluted with 1.0 M NaCl, TE, 1% caffeine. The eluant was ethanol precipitated to remove excess salt, washed with 75% ethanol and 95% ethanol and the dried DNA pellet was dissolved in sterile H2O. The DNA sample was microdialyzed on a Millipore filter (#VMWP01300) against sterile H2O for 30 min. The sample was removed from the filter, centrifuged in an Eppendorf centrifuge for 15 min and 2/3 volume from the top was collected. The concen- tration of the DNA sample was measured by gel electro¬ phoresis using ethidium bromide stained DNA standards (K DNA/Hind III fragments). Finally, the DNA was diluted with microinjection buffer containing 7.3 mM PIPES pH 7.4, 0.1 mM EDTA, 5 mM NaCl. Microinjection into fertilized mouse eggs

Fertilized eggs were flushed from oviducts of B6SJL _?•__ , females and were microinjected with 1-2 ng of DNA. The microinjected embryos were transferred to CR NIH surrogate females. Manipulation of mice and eggs and the microinjection techniques were done as described by Hogan, et al, 1986, Manipulating the mouse embryo: A laboratory manual. Cold Spring Harbor Laboratory. Cold Spring Harbor, NY. Preparation and analysis of mouse genomic DNA

High molecular weight genomic DNA was isolated from 1-2 cm tail samples by standard procedures. 20 ug of DNA samples digested with EcoRI were electrophoresed on 1% agarose in IX TBE buffer and transferred from the gel to nitrocellulose paper by the method of Southern (Southern, 1975, J. Mol. Biol. _9_8_:503-517) . The RNA was electrophoresed in 1% agarose/6% formaldehyde gels as described by Shen et al, 1986, Science 232:643-645. 10 ug of total RNA was loaded per lane. Only samples in which the ribosomal RNA appeared intact were analyzed. RNA was transferred to nitrocellulose paper or Nytran membrane (Schleicher and Schuell, Inc.) as described by Maniatis, et al, 1982. Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. For semiquantitative analysis, RNA samples were applied to filters by using a slot blot apparatus as described by Fojo, et al, 1987, Proc. Natl. Acad. Sci. USA 84:265-269. Hybridization was done as described for DNA hybridization. Northern and slot blot filters were then washed in 0.2 x SSC 0.1% SDS at 50°C (low strin¬ gency) or at 0.1 x SSC, 0.1% SDS at 70°C (high strin¬ gency) as indicated in text. MDRl hybridization probes

The MDRl probes used were obtained from con¬ tiguous areas of an MDRl cDNA and cover around 85% of the total message. A simplified cDNA map is shown in Figure 7 . Probe 10, a 1.6 kb fragment, was obtained by EcoRI digest of pMDRlO; probe 5A, a 1.4 kb fragment, was obtained by EcoRI digest of pMDR 5A; and probe PVU2 #6,

5 was obtained by purifying the 0.84 kb fragment from a PVU II digest of pMDR2000 (Ueda, et al, 1987, Proc. Natl. Acad. Sci. USA 84:3004-3008) . Immunofluorescence localization

Bone marrow cells were smeared onto glass

10 slides, air-dried, then fixed in 3.7% formaldehyde in phosphate buffered saline for 5 min., at 23°C. The slides from control or transgenic mice were then incu¬ bated with mouse monoclonal antibody MRK16 followed by rhodamine-conjugated affinity-purified goat anti-mouse

I⁵ IgG as described by Willingham et al, 1987, J. Histochem. Cytochem. 35:1451-1456.

RESULTS Construction of BAP-MDRl plasmids

As described herein above, a chicken _B-actin

²⁰ promoter fragment was inserted 5' to MDRl cDNA in a unique Sail site of a human MDRl ^' cDNA clone (Figure 1A) . The resulting plasmid (BAP-MDR) contains 330 bp of the _B_-actin promoter sequences followed by 4380 bp of the MDRl cDNA, extending from position -140 (Sad site) to

²⁵ position +4240 (EcoRI site) . A 4690 fragment was cleaved with Xhol and used for the microinjection studies. This fragment does not include vector sequences, except for a 24 bp region of the multiple cloning site from pGEM2 in the junction of lϊ-actin promoter, which are followed by

³⁰ 140 bp of the 5' untranslated region of MDRl cDNA, the full-length MDRl translated sequences, and the 3' untranslated MDRl sequences including the endogenous polyA addition signal sequences. However, 60 bp of the 5' end of the _B_-actin promoter do not exist in the micro-

35 injected fragment. Although the remaining 270 bp of B- actin promoter sequences include the consensus CAAT and TATA boxes, it was not known if this truncated fragment would express MDRl. Thus, an expression vector contain¬ ing the MDRl cDNA downstream from this truncated promoter (tBAP-MDR), but identical in all other respects, was constructed (Figure IB), to test its expression of MDRl. Transfection studies

To test the expression level of MDRl under the control of the _B-actin promoter, as well as the function of the truncated _B_-actin promoter, the expression plasmids pHGl and pHG2 were stably transfected into drug- ° sensitive KB-3-1 cells (Table I). pHaMDR, a retroviral expression vector containing two Ha-MSV LTRs was used as a positive control. For a negative control, cells were treated by the same transfection protocol in the absence of DNA. The same negative results were also obtained 5 after transfection using vector DNA without MDRl sequences (data not shown).

Dishes of KB-3-1 cells (1 x 10⁵ cells per 10 cm dish) were transfected with 10 ug of either pHGl, pHG2, or pHaMDR by the calcium-phosphate precipitation method ^ and selected with various concentrations of colchicine (5-8 ng/ml, which is 3-5 times as high as the LD_5Q of colchicine for the parental KB-3-1 cells) for 12 days. The experiments summarized in Table I show comparable results for pjBAP-MDR and pHaMDR in the number of ⁵ resistant colonies at 5 ng/ml of colchicine. However, the, average colony size was smaller for the pJBAP-MDR transfectants than for the pHaMDR transfectants. This size difference as well as the difference in the colony number at high concentrations of colchicine suggests ⁰ higher expression of MDRl under control of the HaMSV promoter than under control of the _B_-actin promoter in KB cells. However, expression of MDRl under control of the truncated JB-actin promoter (tBAP-MDR) is comparable or higher than the expression of MDRl under control of the 5 full-length Jϊ-actin promoter.

To confirm that the B-actin-MDRl expression vectors function in mouse cells, these vectors were also introduced into NIH 3T3 cells, and the colony-forming ability of parental NIH 3T3 cells and transfectants at 60 ng/ml colchicine (which is 3-5 fold as high as the LD-^_Q of colchicine for the parental cells) measured. In this experiment (data not shown), a similar relative colony- forming ability in the presence of colchicine was achieved for NIH 3T3 transfectants with either the full- length or truncated actin promoter construction, there¬ fore, it was concluded that the truncated _B_-actin pro- moter is an adequate promoter to express the MDRl cDNA in transgenic mice. Production of mice carrying the BAP-MDR1

The 4.7 kb fragment obtained from an Xhol digest of JBAP-MDR (Figure 1A) was microinjected into B6SJL (Fl) fertilized eggs to generate MDRl transgenic mice. The mice were screened by Southern blot analysis of tail genomic DNA which was digested with EcoRI. Blots were hybridized with the 5A probe (Figure 2A) . Figure 2B shows a Southern blot analysis of tail DNA of some mice, showing the expected unique 3.1 kb internal fragment from the MDRl cDNA, as well as the mouse endogenous fragments which could be detected under hybridization conditions of low stringency (Lanes 1-5) but not under high stringency conditions (Lanes 6-10). For each blot analysis, the full-length 4.7 kb injected fragment was diluted, mixed with a negative mouse genomic DNA, digested with EcoRI and applied on the gel as an internal control for com¬ plete EcoRI digestion of the genomic DNAs. This sample was also used as a standard to indicate the size as well as to estimate the MDRl copy number in each transgenic mouse (Lanes 2, 7). An EcoRI digest of human KB-3-1 genomic DNA was also used to confirm the estimated MDRl copy number (Lanes 1, 6). The DNA samples from the MDRl positive mice were also analyzed by slot blot analysis to confirm the MDRl copy number in each transgenic mouse (Figure 3A) .

In ths study, five founder mice containing integrated human MDRl sequences were generated (Figure 3): Mouse 39 (female) had a low copy number (1-3 copies); mouse 132 (female) contained 100-200 copies of MDRl; mouse 168 (male) carried around 50 copies of inte¬ grated DNA; mouse 93 (male, with a high copy number) showed an additional MDRl fragment in a Southern analysis (which probably represented rearranged DNA sequences or junction fragments with host DNA) died soon after wean¬ ing; mouse 104 (female, around 10 copies) killed her

10 first litter and died during her second pregnancy. Transmission of MDRl to progeny

Each of the remaining three transgenic founder mice (M39, M132, and M168) was mated with a normal mouse and tail genomic DNA samples from their progeny were -^■- analyzed by Southern blot hybridization, as well as by slot blot hybridization, for the presence of MDRl sequence. Two of the founder mice (M39 and M132) trans¬ mitted the integrated MDRl sequences through the germ line to about 50% of the progeny, as expected. No major

20 variations in the number or structure of the acquired DNA sequences could be detected in F-_j_ and F2 generations (Figure 4, and data not shown). The DNA analyses of the inheritance of the human MDRl cDNA in the third founder

(M168) revealed that out of the 32 first generation off- ²⁵ spring, only 3 carried the introduced MDRl gene. There¬ fore, it was concluded that this third founder mouse was probably a mosaic. Expression studies in MDRl transgenic mice

The founder mouse M39 (low copy number) and its ⁰ positive progeny were mated with normal mice to generate a MDRl heterozygous mouse line, termed line MDR-39. This line was chosen for RNA expression studies. Total RNA was prepared from 18 different tissues (including brain, liver, kidney, spleen, heart, lung, stomach, small ⁵ intestine, colon, skin, tail, bone, bone-marrow, skeletal muscle, ovary, uterus, oviduct, and testes) of 7 F2 transgenic mice (for pedigree see Figure 5) as well as of 10 negative sibling mice. These RNA samples were analyzed by slot blot hybridization with MDRl probes (Figures 6 and 7). All blots were cross hybridized with a _B_-actin probe as a standard control for accurate RNA loading and filter transfer (Figure 6, and data not shown) . In each experiment, total RNA from multidrug- resistant KB cell lines expressing elevated levels of MDRl mRNA were included so that the unknown samples could be directly compared with samples of known MDRl RNA con- tent and known multidrug resistance. Relative to drug- sensitive KB-3-1, the multidrug-resistant subline KB-8-5 has a 40-fold increase in MDRl mRNA and KB-V-1 (Vbl) has a greater than 500-fold increase. These experiments, summarized in Figure 6, reveal that the human MDRl RNA is expressed mainly in bone-marrow and spleen. Lower expression was also detected in skeletal muscle and ovary (Figure 7B), but not in testes (data not shown). The same blots were cross-hybridized with three different contiguous MDRl probes (10, 5A, and PVUII #6) which together cover around 85% of the MDRl cDNA (Figure 7A) . Except for skeletal muscle which hybridized significantly more to the 5' probe (probe 10) than to the others, all positive tissues showed similar results with all three probes (Figure 7B) . These results indicate that in all the tissues which express human MDRl RNA, except for skeletal muscle, the whole MDRl mRNA was expressed. Northern blot analysis (data not shown) showed an unde¬ fined message size ranging from 4.5 (the expected full- length MDRl message) up to around 11 kb. These results indicate that the endogenous polyadenylation signal at the 3' end of the MDRl cDNA worked weakly in the trans¬ genic mice.

Out of seven MDR-39 mice tested, six showed significant expression in bone marrow, with a somewhat lower expression in spleen. However, one mouse showed expression in spleen but not in bone marrow. One male mouse showed some expression in kidney and liver in addition to skeletal muscle as well as a high signal in bone marrow and spleen. Although some expression was detected in ovary but not in uterus and oviduct of transgenic females, no expression was detected in testes. Expression of P-glycoprotein detected by immunofluores- cence

Immunofluorescence with a mouse monoclonal antibody directed against the human P-glycoprotein (MRK- 16) showed surface expression of human P-glycoprotein in many different bone marrow cells in two MDRl positive mice, when compared with bone marrow derived from a con¬ trol sibling mouse (Figure 8). This could be clearly seen in Figure 8 despite the presence of background mouse plasma IgG detected in the smear using rhodamine- conjugated anti-mouse IgG.

In summary, the results presented herein clear¬ ly demonstrate successful generation of transgenic mice carrying and expressing the human MDRl cDNA under the control of a heterologous promoter. As shown herein above, the strategy was to construct an expression vector carrying the MDRl cDNA sequence 3' to a chicken B-actin promoter. The _B_-actin promoter per se seems to be a constitutive promoter and is only down-regulated by a region 3' to the promoter sequences and by a small region which is located 5' to the polyadenylation signal of the B_-actin gene itself. Neither of these inhibitory regions were contained in the constructions of the present inven¬ tion. A comparative study of DNA-mediated transfer of expression vectors containing a cloned bacterial chloram- phenicol acetyltransferase (CAT) gene into a wide range of cell types has shown that the jB_-actin promoter activ¬ ity is equal to or stronger than the SV40 early promoter (Gunning et al, 1987, Proc. Natl. Acad. Sci. USA 84:4831- 4835). Since B_-actin is abundantly expressed in all cells in culture, it was reasoned that a transgene carry¬ ing MDRl sequences under control of _B-actin promoter may express at a high level in a wide range of cell types, and therefore may confer drug resistance. To test this assumption, first the ability of the _B_-actin MDRl expres¬ sion vector (pHGl) to confer a drug resistance phenotype in cell culture was studied and indeed significant acquired drug-resistance was found.

It has been shown that the presence of pro- caryotic vector sequences is highly inhibitory to appro¬ priate expression of certain genes in transgenic mice. Moreover, in mice carrying a transgene adjacent to pro- caryotic vector sequences, aberrations such as local instability at the site of integration and insertional mutagenesis are observed. In order to eliminate vector sequences in the pronuclear injected DNA in accordance with the present invention, an Xhol fragment of pHGl was isolated. This strategy also deleted 60 bp in the 5' region of the B_-actin promoter. To test the expression ability of this truncated _B-actin promoter, an identical plasmid to pHGl, but lacking the upstream 60 bp of the _B_- actin promoter (pHG2), was constructed and transfected into human and mouse cells in culture. These studies established that the truncated _B-actin promoter activity is equal to or greater than the full-length promoter in cultured cells. Hence, in accordance with the present invention, the Xhol ϊ-actin MDRl fragment was found ade- quate to introduce the human MDRl cDNA into mice.

From the generated MDRl founder mice and their progeny lines, one line (MDR-39) was chosen for expres¬ sion studies. Restriction fragment pattern analyses of genomic DNAs from this founder and over 100 descendant mice (Figure 5, data not shown) indicate that 1-3 copies of the injected fragment were integrated into the mouse DNA. Also, neither internal rearrangement, deletion nor duplication could be detected. Pedigree analyses indi¬ cated that the transgene is: (a) transmitted through the germ line; (b) integrated into a single chromosome, since the transgene is stably inherited by about 50% of the progeny of matings of heterozygous and normal mice; and (c) inherited in an autosomal fashion and is neither x- or y- linked, since male-to-male and female-to-female transmission of the transgene occurred.

It is known that the insertion of foreign DNA sequence into the cellular genome can cause mutational changes by disrupting the function of endogenous genes or of control elements. Most insertional mutations in transgenic mice are recessive, but in most of the matings between heterozygous mice their embryonic lethal pheno- type is demonstrated by significantly reduced litter size and by the inability to produce homozygous mice. Since homozygous matings of MDR-39's progeny revealed normal litter size, and phenotypically normal homozygotes could be generated (data not shown), it was concluded that no insertional mutation of endogenous housekeeping or con¬ trol genes occurred during the integration of the trans¬ gene in this mouse line.

RNA studies of the MDR-39 transgenic mice showed that the transgene is expressed mainly in hemo- poietic tissues such as bone marrow and spleen. Lower expression is detected in skeletal muscle and ovary. However, no phenotypic and morphologic changes nor func¬ tional or behavioral disturbances could be detected in the transgenic mice studied. Primer extension studies of RNA extracted from tissues expressing the transgene with a human specific 35 base synthetic oligonucleotide (which does not cross- hybridize to mouse mdr sequences at conditions of high stringency) (Ueda et al, 1987, J. Biol. Chem. 262:505- 508), resulted in a single extension product of around 192 bases, as expected if the MDRl cDNA is expressed under the control of the _B_-actin promoter. These results suggest a single major transcription initiation site of the transgene under control of the J3-actin promoter, rather than under a mouse endogenous promoter adjacent to the integration site.

Quantitative measurements of RNA expression levels in bone marrow of MDR-39 transgenic mice showed relatively high expression of MDRl. The expression level is comparable to, or in some mice up to 3-fold greater than expression levels of MDRl mRNA in KB-8-5 drug resistant cells. These levels (which are 40- to 120-fold higher than the basal expression level in drug sensitive cells) are at least as high as MDRl RNA levels in multi- drug resistant human tumors.

The present invention now makes it possible for MDRl transgenic mice which express the human MDRl gene in bone marrow to be resistant to doses of chemotherapy which would be toxic to the bone marrow of the normal mice. For example, an intraperitoneal dose of the chemo¬ therapeutic agent daunomycin produces bone marrow suppression and death 2-3 weeks after treatment in the majority of the mice so treated. In contrast, the trans¬ genic mice of the present invention are resistant to that dosage of a chemotherapeutic agent which would be lethal to normal mice. Hence, by establishing tumors in MDRl transgenic mice (either by transplantation or by intro¬ duction of genetic alterations which predispose the mice to such tumors, or by other suitable methodology), the tolerance limits of potent chemotherapeutic preparations, containing either a single or a combination^"of drugs, can be determined.

Transgenic mice of the present invention also allow genetic manipulations to improve suitability of the MDR-39 and other MDRl transgene mice for various pur¬ poses. These manipulations include the matings of MDRl mice to produce mice homozygous at the MDRl locus and to produce fully inbred strains of mice for tumor xenograft studies. Furthermore, by crossing MDRl mice with nude immunodeficient mice, animals are obtained that accept human tumors while yet being multidrug resistant. This allows the testing of tolerable limits and efficacy of anti-tumor or anti-cancer agents.

An additional utility of the cloned MDRl is as a dominant selectable marker to introduce linked genes into the bone marrow and other tissues of humans and animals. The MDRl mice allow the determination of the dosages of drugs (such as daunomycin, vincristine, vinblastine, VP-16, VM-26, actinomycin D, adriamycin and the like) needed to select cells or tissues expressing the human MDRl gene and introduced into recipient host. This is accomplished by using MDRl mice to determine the tolerable dosages of these and other chemotherapeutic agents for bone marrow and other tissues expressing MDRl gene. Of course, such studies are needed prior to devel¬ opment of effective gene therapy regimens using the MDRl gene as a selectable marker.

The MDRl transgenic mice also allow the study of the role of multidrug transporter encoded by this gene. Expression of the MDRl gene at relatively high levels in tissues in which this gene is not normally expressed, allows the determination of the physiological effects of the transporter on various tissues. Further- more, the MDRl transgenic animals can serve as donors for various cells and tissues expressing the human MDRl gene. These cells can be established in culture or can be used as the sources of multidrug resistant tissues in animal transplant experiments. it is understood that the examples -and embodi¬ ments described herein are for illustrative purposes only and that various modifications in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this applica¬ tion and scope of the appended claims. Table 1. Frequency of drug-resistant colonies

KB-3-1 (1 x 10~⁶ cells per 10-cm dish) were trans- fected with 10 ug of DNA; 48 hr after transfection, cells were split into 4 dishes and cultured in the presence of colchicine at the indicated concentration. On day 12, cells were stained and colonies counted.

Claims

WHAT IS CLAIMED IS:

1. Transgenic animals carrying and expressing human MDRl gene.

2. The animals of claim 1 being mice.

3. A method for testing high dosage chemo¬ therapeutic modalities against tumors, comprising:

(a) implanting or inducing tumors in transgenic animals of claim 1;

(b) then administering a chemotherapeutic agent, either alone or in combination, to a plurality of groups of animals of step (a) in increasing amounts to each group of animals; and

(c) determining that amount of the chemo- therapeutic agent which prevents the growth or proliferation of the tumor without intolerable side effects.

4. A method for testing efficacy of gene transfer protocols in which the MDRl gene is used as a selectable marker, comprising:

(a) determining maximum tolerable dose of a chemotherapeutic agent in the presence of which MDRl expressing cells grow; (b) then selecting, by employing the maximum tolerable dose, multidrug- resistant somatic cells into which the MDRl gene has been integrated after said cells are introduced into host animals.

5. Cells or tissues expressing the human MDRl gene, obtained from the animals of claim 1.