CA2092571A1 - Ribozyme inhibition of bcr-abl gene expression - Google Patents

Abstract

Ribozymes effective to inhibit bcr-abl expression and the use of ribozymes to block proliferation of leukemia carrying the bcr-abl gene are disclosed. E. Coli transfected with the requisite ribozyme gene provides significant quantities thereof.

Description

~ WO93/03141 2 0 9 2 5 7 ~ PCT/US91/05443 RIBOZYME INHIBITION O~
BCR-ABL GENE EXPRESSION

FIELD OF THE INVENTION
This invention relates to the ribozyme inhibition of bcr-abl gene expression and to the use of such ribozyme inhibition to block proliferation of leukemia cells that carry the bcr-abl gene, to E.
Coli into which the ribozyme gene directed against bcr-abl fusion mRNA has been transfected and from which large quantities of the ribozyme-containing genes can be obtained.
The invention also relates to the purging of leukemia cells from bone marrow of patients Phl+
chronic myelogenous leukemia (CML) or acute lymphoblastic leukemia (ALL) and to the use of the purged bone marrow to reconstitute the patients hematopoietic system.
~ACKGROUND OF THE INVENTION
Chronic myelogenous leukemia (CML) arises from the malignant transformation of a pluipotential stem cell, and represents about 15 to 20 percent of all leukemias. The Philadelphia chromosome (Phl), which results from a reciprocal translocation between chromosomes 9 and 22, is detectable in over 95 percent of patients with CML. This translocation results in transposition of the cellular abl (c-abl) gene from its usual position on chromosome 9 to chromosome 22 (Bartram, C.R., et al. Nature 306:277-280 (1983); Heisterkamp, N., et al. Nature 306:239-242 (1983)). The normal proto-oncogene c-abl contains over 230 kb and 12 exons that encode for a pl45 protein with tyrosine kinase activity. The function of this normal pl45 protein is not known.

,i : - :: -. . ::
:

2 0 9 2 ~ 71 PCT/US91/05443 ~ ¦
_ _ I

The breakpoint on chromosome 9 occurs within a large 200 kb region at the 5' end of the c-abl gene that leaves exons 2 through 11 as an integral part of the fusion gene (Bernards, A., et al., Mol. Cell Biol.
7:3231 (1987); Kurzrock, R. N. Enq. J. Med. 319:990 (1988). In contrast, the breakpoint on chromosome 22 at band qll occurs within a limited 5.8-kb DNA
segment termed the breakpoint cluster region or bcr (Groffen, J., et al. Cell 36:93-99 (1984)), which resides on a gene composed of 140 kb and 23 exons, now designated the bcr gene. The bcr consists of six exons, which correspond to exons XIII-XVIII of the bcr gene. The function of the normal bcr gene is not known, although a pl60 bcr protein has been characterized (Stam, K., et al. Mol. Cell Biol.
7:1955 (1987)). Recent studies have localized the gamma-glutamyl transferase gene within the bcr gene, but the significance of this observation has not yet been determined. The site of the breakpoint in the bcr varies, but involvement of this region appears to be a consistent and necessary aspect of the translocation. Transposition of c-abl into the bcr gene results in the creation of an abnormal fusion gene termed bcr-abl, that occurs either between bcr exons 3 and 4, or between bcr exons 2 and 3, and able exon 2. The mRNA transcript that results from the fusion gene is 8.5 kb (+/- 109 bases depending on whether bcr exon 3 is included or not), and is translated into a p210 protein with augmented tyrosine kinase activity.
~ he bcr-abl gene can be found in virtually all patients with CML, even those in whom the common (Stam, K., et al. Mol. Cell Biol. 7:1955 (1987);
Kurzrock, R., et al., Blood 70:943 (1987)) translocation is replaced by a variant translocation .
.. .. ... . . . . . . . . .
:. .. . ~' .. . . ..

WO 93/03141 2 ~9 2 S ~1 PCl/US91/05443 1-not apparently involving chromosome 9, or in whom no gross cytogenetic evidence of the phl is present.
The p210 tyrosine kinase gene product may represent a cancer specific marker. It has been shown to transform hematopoietic cells in vitro suggesting that it may actually have a causal role in the development of CML. As further evidence for the critical role of the p210 protein, several investigators have reported that expression of this protein in mice using viral vectors leads to the development of CML- or ALL-like disease in many of these animals.
CML patients progress inevitably from a stable chronic phase (CP) to an accelerated phase (AP), and finally to a terminal blast crisis (BC). The death rate for CML patients is 5-10% for the first two years after diagnosis, and increase to 25~ per year thereafter, corresponding to the incidence of progression to AP or BC. The natural course of CML
has not yet been substantially modified by conventional chemotherapeutic agents, by aggressive combination therapy or by the innovative use of biological agents.
Allogeneic bone marrow transplantation is the only proven curative therapy for this disease. Many transplant centers have achieved excellent results with disease free survival (DFS) rats of approximately 50-75% for patients transplanted while in CP. For patients transplanted in AP or BC the DFS
rates are lower, at approximately 40% and 20~ -respectively. since only a minority of patients who might be candidates for BMT have histocompatible sibling donors available, alternative sources of BM
need to be found for BMT. The National Marrow Donor Registry IMMDR) is a source of matched, unrelated , ..
: :~ , ' ~, ' -~ . ' :
.

WO93/03141 2 ~ ~ 2 ~ 7 1 PCT/US91/05443 donors (MUD) whose marrow can be used in the setting of allogeneic BMT to treat many more patients with CML.
However, the chances of finding such a donor through the NMDR vary considerably depending on the ethnic background of the patient. Most minority groups in this country are under-represented in the registry, thus decreasing the odds considerably for a - patient from one of these groups to find a suitable donor. Furthermore, the incidence of complications for MUD BMT, in particular graft rejection, severe graft-vs-host disease, and possibly regimen related toxicities, are higher than those for sibling donor transplants. Many patients may not be candidates for this approach because of age or an unfavorable risk/benefit ratio.
As an alternative, attempts have been made to treat CML patients with autologous BMT based on the following rationale. phl negative myeloid cells can occasionally be found in peripheral blood (PB) or bone marrow (BM) of patients with CML at diagnosis, after intensive chemotherapy (Goto, T., et al., Blood 59:793 (1982)), after interferon therapy or after growth of bone marrow hematopoietic progenitors in semisolid culture or long term liquid c~lture. It has been argued that such Phl negative cells may be malignant; however, these Phl negative cells may represent the progeny of benign, viable stem cells capable of repopulating the hematopoietic system and of providing complete remission from CML after autologous transplantation. If not curative, such an approach may at least prolong CP or establish a second CP after transformation of disease. Such considerations provide the rationale for a number of clinical trials testing the efficacy of autologous .
:, ~

, -. . .

W093/03141 ?~ 7 5 7I PCT/US91/05443 infusion of untreated or treated BM, or PB stem cells after treatment of CML patients with ablative doses of chemotherapy alone or chemoradiotherapy.
Ex vivo treatment of CP marrow with 4-hydroperoxycyclo- phosphamide (4 HC) or with recombinant human gamma interferon (McGlave, P., et al., Blood 74:281A (1989)) followed by autologous BMT
has resulted in complete hematologic remission as well as disappearance of the phl from bone marrow and hematopoietic progenitor metaphases.
The ability to select phl myeloid progenitors from the BM of phl positive CML patients grown in long term liquid culture suggests an alternative method for collection of benign, viable hematopoietic stem cells for autologous transplantation. Initial reports demonstrated successful engraftment and -complete remission in ANLL patients treated with high dose chemoradiotherapy and infusion of BM cells washed from long term liquid cul~ure after ten days of growth. Recently, adaptation of the same marrow treatment method has led to engraftment and short term hematologic remission in CML patients. Longer follow-up and additional patient accrual will be required to determine if this innovative approach to marrow purging with hematopoietic progenitor culture techniques will result in sustained engraftment and remission.
There are potential draw-backs to each of these methods of ex vivo purging. Treatment of the marrow with chemotherapeutic agents can damage normal stem cells as well as the leukemia cells, and thus increase the risk of graft failure. The success rate for culturing-out Phl negative stem cells from CML
marrow is variable. In addition, this population of -. . .. : : ,, ,. , ,: ~ . .

W093,0314, 2a92~7l PCT/US91/05443 ~ I

phl negatlve cells may be found to be bcr-abl positive if a sensitive assay such as PCR were used.
This invention utilizes ribozyme technology as a novel molecular biologic approach to this problem.
Ribozymes are a class of RNA molecules that can cleave other RNA sequences enzymatically. These molecules have two domains, the catalytic "hammerhead" portion that cleaves the target by a mechanism dependent on divalent cations, and the flanking oligonucleotides that confer specificity of binding of the ribozyme to the region of the ~NA
molecule that contains the target sequence. Any sequence of nucleotides of G-U-N, where N = A, C, or U, can be targeted by a ribozyme (Ruffner, E.E., et al., Biochemistry 29:10695-10702 (1990)).
Ribozymes act like enzymes in that one molecule can bind to its RNA target, cleave it, then dissociate and bind to a second target, and so on.
Ribozymes have potential to inhibit specifically the expression of a variety of genes, and represent a novel therapeutic approach to controlling viral infections and oncogenesis. Researchers have begun to apply this technology to inhibit expression of HIV
genes (Chang, P.S., et al., Clin. Biotechnoloqy 2:1-9 (1990); Sarver, N., et al., Science 247:1222-1225 (1990)), as well as c-fos (Scanlon, K.J., et al., submitted to Science, 1990) and H-ras (Kashani-Sabet, M., et al., submitted for publication 1990) oncogenes in vitro and in animal models.
SUMMARY OF THE INVENTION
The bcr-abl fusion gene is the molecular counterpart of the Philadelphia chromosome Ph1 which is found in over 95% of patients with CML and about 30~ of adults with ALL. This invention provides ribozyme technology effective to inhibit - : :

, ~VO93~03141 2 ~ 7 1 PCT/US91/05443 expression of the bcr-abl gene in Phl+ leukemia cells. Appropriate ribozyme genes are provided. E.
C _ into which such genes have been transfected replicate the ribozyme gene allowing for large scale isolation of such genes. Ribozyme RNA transcribed from the genes obtained from E. Coli colonies cleaves bcr-abl RNA. Bcr-abl expression in EM-2 cells is inhibited by the liposome vector incorporation of ribozyme mRNA.
A particularly important aspect of the invention entails the purging of leukemia cells from the bone marrow of patients with CML or ALL. The purged bone marrow is then used to reconstitute the patients hematopoietic system.
DESCRIPTION OF THE FIGURES
Figure 1 is a bcr-abl mRNA sequence showing fusion site of splice l. Ribozyme sequence showing "hammerhead" catalytic domain targeted to G-U-U
codon, located just 5' to fusion site, and flanking oligonucleotides complementary to bcr-abl mRNA.
Figure 2 is a Southern blot demonstrating that two E. Coli colonies, #9 and #ll, transfected with pBluescript KS plasmid vector containing the bcr-abl ribozyme gene express the ribozyme gene, as detected by hybridization with P32-labelled oligomer sequence complementary to the ribozyme. Ribozyme gene was cut with Hind III and Sst I endonucleases.
Figure 3 is a ribozyme and pJWp3 (plasmid containing bcr-abl splice l sequence) transcribed to P32-labelled RNA using P32-UTP, then incubated together in cell-free system for 3 hours (Figure 3A) overnight (Figure 3B) at 37C. In Figure 3A, lane l shows that the main transcription product from pJWp3 is partially cleaved to a smaller product in the presence of ribozyme and l0 mM Mg++. Lane 2 shows no - . . , , . - , , . - - - .

., . ,, . , .: . . . . .

W093/03141 2 ~ ~ 2 3 7 1 -8- PCT/US91/05443 ~

cleavage in the absence of Mg++. Lanes 3 and 4 show no brea~down of the bcr-abl RNA in the absence of ribozyme, with or without Mg++. In Figure 3B, lane 3, pJWp3 RNA was completely cleaved after overnight incubation with ribozyme and Mg++/ Lane 1 shows no breakdown of the pJWp3 RNA in the absence of ribozyme, and lane 2 shows no breakdown by ribozyme in the absence of Mg++.
Figure 4 is a P32-labelled ribozyme complexed to lipofectin taken up efficiently by EM-2 cells, but only minimally in the absence of lipofectin.
Figure 5 depicts a gel which demonstrates that intact P32-labelled ribozyme is detected in EM-2 cells after 18 and 42 hours of incubation with ribozyme complexed to lipofectin.
Figure 6 is a Southern blot of bcr-abl DNA
amplified from rNA isolated from EM-2 cells, then subjected to reverse transcriptase followed by PCR.
Figure 7 depicts the design and construction of bcr-abl ribozyme expression plasmid.
DETAILED DESCRIPTION OF THE INVENTION
A procedure for the detection of bcr-abl mRNA by PCR techniques is described in Lange, et al. Blood 73:1735-1741 (1989).
Preparation of Ribozyme:
SYnthesis, Cloninq, Sequencinq A "hammerhead" ribozyme as shown by Figure 1 to -cleave bcr-abl mRNA was prepared. The ribozyme contains the 22 nucleotide catalytic RNA sequence flanked by 15 nucleotide targeting sequences that position the catalytic domain at the target site by Watson-Crick base pairing. Cleavage of the target RNA occurs specifically after the indicated G W
residue, generating a 2'-3' cyclic phosphate and 5'hydroxyl.

. . . .. . .
-.:, : , .
.
:' ' ', ' ':
: -~ ',' .' .` ~ -, , ' , . , :. .
.

~ W093/03141 ~ 9 2 S 7 ~ PCT/USg1/05443 g Oliqonucleotides Synthetic oligodeoxyribonucleotides were prepared on an Applied Biosystems Model 380B DNA synthesizer.
The double-stranded ribozyme gene insert wlth two flanking BamHI restriction sites were prepared from two single-strand oligodeoxyribonucleotides (Oligo 1 primer 5'CCAGATCTGAAGGGCTTTTGCTGATGAGTCCGTGAGG 3', Oligo 2 primer 5'CCAGATCTGGATTTAAGCAGAGTTTCGTCCTCACG-GACT 3', 37 and 39 bases long, with 11 bases complementary at 3' termini.
RibozYme cDNA Synthesis and Subclone to Plasmid To synthesize the double-stranded ribozyme cDNA, the two oligoes described above were used to carry out the in vitro polymerization (PCR) reaction utilizing Taq DNA polymerase. Subsequently, the PCR
product of ribozyme sequence is to be cloned into the BamH1 site of the pBLUESCRIPT II KS vector (Stratagene, LaJolla CA). The fidelity of the synthesis, polymerization, and orientation of the insert was checked by dideoxy sequencing of the cloned ribozyme PCR product (USB sequence 2.0 kit Cleveland, OH).
In Vitro Transcription of RibozYme RNA From Plasmid Template Transcription of RNA from the pBLUESCRIPT-KS(~) plasmid, using T7 promoter was carried out under the conditions adapted from Lowary et al., "Structure and Dynamics of RNA", p. 69, Plenum Press, New York (1986). Briefly, RNA was synthesized by T7 RNA
polymerase transcription of partially duplex synthetic ribozyme DNA templates that will be linearized downstream of the sequence to be transcribed using a five-fold excess of the appropriate restriction endonuclease. The transcription reaction mixtures contain plasmid DNA

. . . - . . . : - -~- . ................. - . - . .

' . ' ' ' - ' : ' . . ~ '' .

W O 93/03141 2 ~ ~ 2 ~ 7 t PC~r/US91/05443 ~ I

templates in the concentration of 0.05 ~g/~l, 0.5 units /~1 T7 RNA polymerase; 40mM Tris-HCL pH 7.9;
20mM MgC12; 10mM NaCl; 10mM DTT; 1.0mM ATP, CTP, and GTP; 0.lmM UTP; 10 ~Ci of (~_p32) UTP; and 1 unit of RNase inhibitor in a 50 ~1 volume. The reaction was carried out at 37C for 1 hour for radiolabeled RNA
and 3 hours for nonradiolabeled RNA. Transcription reactions were fractionated by using acrylamide gels containing 7 M urea. Products were located by autoradiography or W shadowing, eluted in 0.25M
ammonium acetate/10 mM Tris HCl, pH 7.9/ lmM EDTA and then concentrated by ethanol precipitation. See Figure 2.
PreParation of bcr-abl SPlice 1 RNA
The pJWp3 plasmid contains bcr-abl splice 1 cDNA. This clone was provided by Dr. Owen Witte, Molecular Biology Institute, University of California, Los Angeles, Los Angeles, California 90024. This cDNA was incorporated into the pBLUESCRIPT-KS(+) plasmid vector, subcloned, and RNA
transcribed as described above for the bcr-abl ribozyme gene.
Cell-Free Assay of Ribozyme Cleavaqe of bcr-abl mRNA
As depicted by Figures 3A and 3B, P32-labelled ribozyme and bcr-abl RNA molecules were incubated in a cell-free assay using Northern blots to demonstrate the products of the reaction. Titration of the ribozyme:substrate ratio, Mg++ concentration, and determination of temperature and time conditions for this reaction were carried out.
Uptake of Ribozyme by EM-2 Cells Via Liposome Vector LiPosome Vector RNA transfection using cationic liposome is one accepted as a method for introducing genetic information into cultured eukaryotic cells. See, - . :- .
, - - . -~
.. :

~ WO93/03141 2 0 ~ 2 ~ 7 1 PCT/US9"05443 Malone, R.W., et al. PNAS 86:6077-6081 (1989). The liposome interacts spontaneously with DNA or RNA to form a lipid-DNA or lipid-RNA complex with complete entrapment of the DNA or RNA, and the fusion of this complex with cell membranes results in efficient uptake of the DNA or RNA.
RNA Transfection of EM-2 Cell The EM-2 cell line (Keating, A., et al. Normal and Neoplastic HematoPoiesis~ pp. 513-520 ~1983), Alan R. Liss, Inc., New York) (obtained from Dr.
Witte, UCLA) which was derived from a patient with CML in blast crisis, and which contains the bcr-abl fusion gene (splice 1), was maintained in RPMI 1640 with 10% (v/v) FCS in our laboratory. During transfection, the EM-2 cells are washed once with opti-MEM I reduced Serum Medium (Gibco, MD) and incubated with the same medium for 1 hour.
Separately diluted concentrations of ribozyme mRNA
and equal amount of lipofectin (BRL Gathersburg, MD) are then added to the EM-2 cells in an optimal cell concentration, i.e., 5 x 104 cells/ml, and incubated for 5 to 24 hours at 37C in a humidified, 5% C02 environment.
Transfection of Cell Lines by a Plasmid Containinq the Ribozvme Desiqn and Construction of bcr-abl RibozYme Expression Plasmid EM-2 cell lines are developed that express high levels of bcr-abl ribozyme for the purpose of observing the effects of stably maintained expression of ribozyme,. The expression vector which was adapted is that reported by Gunning et al., Proc.
Nat. Acad. Sci. USA _ :4831 (1987), pH~ Apr-l-neo.
This vector contains ~-actin promoter region plus a 5'-untranslated region and an intervening sequence linked to a short DNA polylinker segment containing ': :
.

.

wo 93~03141 2 ~ 9 2 ~ 7 :~ pcr/us91/os443~

unique Sal 1, Hind III, and BamHI restriction endonuclease sites, followed by an SV40 polyadenylation signal, an ampicillin-resistance gene, and a neomycin-resistance gene. Therefore, the expression of bcr-abl ribozyme cDNA is under control of a strong ~-actin promoter. Cells containing the expression plasmid will be selected by their resistance to neomycin. Briefly, as depicted by Figure 7, bcr-abl ribozyme cDNA is prepared by digesting pBLUESCRIPT II KS containing the bcr-abl ribozyme cDNA with restriction enzyme BamHI and purified. This ribozyme cDNA fragment is ligated to a BamHI restricted expression vector, pH ~
Apr-l-neo. Thus, the cDNA insert is located between the human ~-actin promoter and an SV 40 polyadenylation signal.
Bcr-abl RibozYme Expression in CML Cell Lines To transfect the EM-2 CML cell lines by using the expression plasmid pH ~ Apr-1-neo, Lipofectin is utilized following the manufacturer's protocol (BRL, Gaithersburg MD). The cells are incubated at 37 C
with 5% C02 for 24 hours in opti-MEN I Reduced Serum Medium, after which media containing 10% fetal calf serum were added. After an additional 48 hours, the cells were transferred to selective media (opti-MEN
with 10~ FCS and G418). After about two weeks of selection, individual colonies are picked, grown, and screened for expression of P32-labelled bcr-abl ribozyme by the polymerasP chain reaction assay and Northern blot analysis.
Assay of p210bCr-abl Protein Product bY Western Blot AnalYsis To evaluate the effect of anti-bcr-abl ribozyme on downregulating bcr-abl gene expression, the p210bCr-abl protein encoded by the bcr-abl fusion .. . ..
.. . . . , , ~ .
: .
:
,, - . ' . . .
.

WO93/03141 ~9 ~ j 71 PCT/US~1/05443 gene was detected by Western blot analysis. The ribozyme transfected cells were lysed directly in SDS
gel loading buffer, then analyzed by SDS-polyacryl-amide gel electrophoresis. After transferring the separated components from a gel to a nitrocellulose filter, the target protein was probed by using specific polyclonal antibodies against p2lObCr-ab (oncogene Science, NY).
Inhibition of bcr-abl Expression In EM-2 Cells by Ribozymes RNA isolated from EM-2 cells is subjected to reverse transcriptase and then amplified by PCR.
Southern blot analysis of the amplification product is shown by Figure 6. Lane l is positive control pJWp3 showing splice l as main product of amplification. The next lane is blank. Lane 2 shows complete absence of bcr-abl RNA after 18 hours incubation of EM-2 cells with ribozyme-lipofectin complex. Lane 3 shows intact bcr-abl splice l when cells are incubated with negative control RNA-lipofectin complex. Lane 4 is bcr-abl from EM-2 cells alone. Lane 5 shows weak signal for bcr-abl splice l after 42 hours of incubation with ribozyme-lipofectin. Lanes 6 and 7 are the same as - lanes 3 and 4 except after 42 hours of incubation.
Lane 8 is negative control for the PCR assay with no template added to the reaction mixture.
Cell Bioloqy Effects The impact of ribozyme-mediated inhibition of bcr-abl gene expression on 3H-thymidine incorporation, cell counts, colony formation, and cell phenotype of E~-2 cells was determined by standard techniques including immunophenotyping. --Freshly isolated human leukemia cells derived from -patients with Phl+ leukemias were studied in a . . , ~ . - .

WO93/03141 2 ~ 9 2 ;~ 7 1 PCT/US91/05443 ~ j i similar fashion. One experimental result is reported in Table 1.

3H-Thymidine Incorporation by EM-2 Cells (cpm +/-SD) ~Inhibition 1.Control cellsa 13,441 +/- 1828 -----2."Control" RNAb 13,317 +/- 217 0.9%
3. RibozymeC 9,007 +/- 291 33.0%
a = 5 x 104 EM-2 cells incubated for 18 hours in 1%
FCS, pulsed for 4 hours with 3H-thymidine b = same as a, but 9 ~g total RNA extracted from a normal subjects' peripheral blood mononuclear cells was added to the EM-2 cells via lipofectin c = same as a, but 9 ~g bcr-abl ribozyme added to cells via lipofectin Relevant RibozYmes The ribozyme specifically depicted by Figure 1 is directed against the bcr-abl spice 1 mRNA which fuses bcr exon 3 to abl exon 2. Cells of some patients with Ph+ CML may express either splice 1, splice 2 (bcr exon 2 fused to abl exon 2) or both. Two ribozymes, one targeting splice 1 and one targeting splice 2, may be required to inhibit proliferation of leukemia cells that express both splices. A third ribozyme will be required to inhibit proliferation of the ALL bcr-abl splice. This in~ention accordingly includes all three such ribozymes, combinations of two or more such ribozymes, and use of such ribozymes singularly or in combination to purge bone marrow from patients with Ph~ CML or ALL.

-.. . :
.
v .: . . ' .: , . ' , .~ . , ,: ' , ",' ',' . '.'' '" ., ' " '' ,' ' '''' '''.'' ' ' ''~ '~', '" : ' W O 93/03141 2 ~ 3 ~ PC~r/US91/05443 -15- .

Activation of the c-abl o~cogene as a result of fusion of bcr sequenced to the 5' end of the abl gene has been postulated to be one of the key initiating steps, if not the key step, in the pathogenesis of Phl+ leukemias. Inhibition of Phl+ cell proliferation and/or differentiation by a ribozyme directed specifically at the bcr-abl mRNA provides strong supporting evidence for this hypothesis.
Clinically, the successful application of ribozyme technology to purge patients' bone marrow in the setting of autologous bone marrow transplantation would have a significant impact on the curability of CML. Most patients with CML in Chronic Phase who are potentially curable by BMT do not have suitable sibling donors. The alternative of performing BMT
using matched unrelated donors is considered to be much more risky than sibling donor transplants, and many patients in this country have little to no chance of finding such a donor.
Purqinq of Autoloqous Bone Marrow Bone marrow cells will be harvested from the posterior iliac crests of patients under general anesthesia following standard procedures. A total of 4 x 108 nucleated cells/kg of body weight of the patient will be collected. This total is twice the usual number so that half of the cells can be preserved as back-up in cells. Peripheral blood stem cells primed by G-CSF will be collected by leukopheresis to supplement the bone marrow harvest if necessary.
Freshly harvested stem cells will then be incubated ln vitro in the presence of the plasmid containing the bcr-abl ribozyme gene, which will be introduced into the bone marrow cells via a liposome vector, or some other method depending on the WO93/0314t 2 0 9 2 ~ 71 PCT/US91/05443 ~ j efficiency of uptake. The cells will be incubated in a sterile co2 incubator for 2-4 days, the exact length of time to be determined by preclinical experiments. The cells will then be recovered, washed, and re-infused into the patient.
During this period of in vitro incubation of the bone marrow cells, the patient will undergo myelo-ablative therapy with a combination of fractionated total body irradiation and high dose chemotherapy consisting of VP-16, +/-cyclophosphamide. The purged bone marrow cells will be re-infused approximately 48 hours after the last chemotherapeutic agent is administered.
The efficiency of the purging technique will be monitored by routine cytogenetic analysis and by PCR
to detect residual Phl+ and bcr-abl+ cells, respectively. Standard bone marrow culture assays for CFU-GM and CFU-GEMM colonies will be carried out to assess the specificity and safety of the purging process.

:' '. , . . . ,- . :, :-, . ': ' ' -' :': '. . :

, - ,. :. ,. :
. - ~ . .
, . ' . .-' .. ', , . . ' ' ~'

Claims (13)

1. CLAIMS:
   l. A method for purging leukemia cells from bone marrow of patients affected with Philadelphia chromosome + chronic myelogenous leukemia and acute lymphoblastic leukemia which comprises treating said bone marrow with a ribozyme effective to inhibit the expression of a bcr-abl gene.
2. 2. A method as defined by claim l in which said bcr-abl gene product is the bcr-abl splice 1 mRNA.
3. 3. A method as defined by claim 1 in which said bcr-abl gene product is the bcr-abl splice 2 mRNA.
4. 4. A method as defined by claim 1 in which said bcr-abl gene product is the "ALL" spliced mRNA.
5. 5. A method which comprises ex vivo purging of leukemia cells from bone marrow by the method defined by claim l or claim 2 followed by autologous transplantation of said purged bone marrow.
6. 6. Bone marrow from which leukemia cells have been purged by the method of claim 1 or claim 2.
7. 7. A Ph+1 cell line into which a ribozyme effective to cleave bcr-abl mRNA has been incorporated.
8. 8. The EM-2 cell line into which the ribozyme shown by Figure 1 has been incorporated.
9. 9. An eukaryotic cell into which a gene which expresses ribozyme effective to cleave a bcr-abl transcript has been transfected.
10. 10. An eukaryotic cell as defined by claim 6 in which said ribozyme is the ribozyme depicted by Figure 1.
11. 11. An eukaryotic cell that expresses a ribozyme gene effective to cleave bcr-abl fusion mRNA.
12. 12. An eukaryotic cell as defined by claim 9, said cell being an E. Coli cell.
13. 13. The plasmid ph .beta.-ribozyme depicted by Figure 7.