CA2135499A1 - Method and reagent for inhibiting cancer development - Google Patents

Abstract

An enzymatic RNA molecule which specifically cleaves mRNA encoded by an mdr-1 gene, or associated with development or maintenance of chronic myelogenous leukemia, promyelocytic leukemia, Burkitt's lymphoma or acute lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic leukemia, breast cancer, colon carcinoma, neuroblastoma, and lung cancer.

Description

93/230~7 PCT/US93/045i~
1~
,,.

DESCRIPTION ~

METHOD AND REAGENT FOR INHIBITING -CANCER DEVELOPMENT

Backaround of the Invention This invention relates to methods for treating cancer, and in particular, growth of a transformed cell, and inhibition of progression to a transformed phenotype ; 5 in pre-neoplastic cells.
Transformation is a cumulative process whereby ~` normal control of cell growth and differentiation is interrupted, usually through the accumulation of mutations affecting the expression of genes that regulate cell growth and differentiation.
Scanlon WO91/~18625, WO91/18624, and W091~18913 describes~;a~ rlbozyme effe~ctive~to cleave oncogene RNA in ~ the~ H-ras gene.~ This ribozyme is said to inhibit C-fos :~ expressi~on ln response to cis-plantin or other stimuli.
15~ ~Reddy;,~WO 92/OOQ80 and U.S. Serial No. 07/544,199 (filed 3une~26,~19~90~, describes use of ribozymes as therapeutic agents~or 1eukemla~s, such as chronic myelogenous leukemia CML)~ by~targeting specific junction regions of the bcr-abl ~ f~usion transcrlpt.
- SummarY of the Invention This inventlon concerns use of a ribozyme t~argeted~to ~the P-glycoprotein (mdr-1 gene) or other cancer-related genes prior to and/or during administration of~anticancer chemotherap~u!tilc agents. Inclusion of such -~
a ribozyme increases the susceptibility of the transformed cells to such agents.
Applicant notes that relapse of disease caused :`, by cancerous ~ cells after administration of chematherapeutic agents is a major problem in obtaining lasting remissions in a clinic. In some neoplasias, relapse is caused by the expansion of a population of transformed cells resistant to the initial and subsequen~
-,~; ~ . .-.

WO93/230~7 ~ 4 ~ 9 PCT/US93/0~

forms of chemotherapy due to inappropriate expression of the mdr-1 gene, also called P-glycoprotein. Such expression is usually caused by selection of transformed cells that have amplified the mdr-1 gene and thus produce increased amounts of the mdr-1 gene product. Applicant describes treatment of and prevention of this condition by use of ribozymes targeted to the mRNA encoded by this gene.
The mdr-1 gene encodes a 170 kDa integral membrane transport protein that confers resistance to certain chemotherapeutic agents such as colchicine, doxorubicin, actinomycin D and vinblastine ~reviewed ln Gottesman and Pastan, 263 J. Biol. Chem. 12163, 1988).
The gene has been isolated from both human and rodent cells selected in vitro for resistance to such agents (Roninson et al., 309 Nature 626, 1984; and Roninson et al., 83 Proc. Natl. Acad. Sci USA, 4538, 1986), and the entire;4.5-kb MDRl transcript encoding the human MDRl has been sequenced (Chen et al., 47 Cell 381, 1986, EMBL
accession # M14758). The gene is normally expressed in the cells of the colon, small intestine, kidney, liver and adrena~l gland. High levels of MDR1 transcript have been found in adenocarcinomas that are intrinsically resistant to a broad range of chemotherapeutic agent~, such as those derived from adrenal, kidney, liver and bowel.
The in~ention features use of ribozymes to -~ inhibit the development or expression of a transformed , ~ .
phenotype in man and other animals by modulating e~pression of a gene that either contributes to, or inhibits ~he expression of CML, promyelocytic leukemia, Burkitt's lymphoma, acute lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic leukemia, breast cancer, colon carcinoma, neuroblastoma, lung cancer, and other neoplastic conditions. Cleavage of targeted mRNAs expressed in pre-neoplastic and transformed cells elicits -~ inhibition of the transformed state.

.~,;

~ 93/23057 2 1 ~ 5 ~1 ~ 9 PCT/US93/0457~

Riboz~mes are RNA molecules having an enzymatic activity which is able to repeatedly cleave other separate RNA molecules in a nucleotide base sequence specific manner. Such enzymatic RNA molecules can be targeted to virtually any RNA transcript and efficient cleavage has been achieved in vitro. Kim et al., 84 Proc. Natl. Acad.
Sci. USA 8788, 1987; Haseloff and Gerlach, 334 Nature 585, 1988; Cech, 260 JAMA 3030, 1988; and Jefferies et al., 17 Nucleic Acids Research 1371, 1989.
Ribozymes act by first binding to a target RNA.
Such binding occurs through the target RNA binding portion of a ribozyme which is held in close proximity to an enzymatic portion of the RNA which acts to cleave the target RNA. Thus, the ribozyme first recognizes and then binds a target RNA through complementary base-pairing, and once bound to the correct site, acts enzymatically to cut the target RNA. Strat~egic-cleavage of such a target RNA
will destroy its ability to direct synthesis of an encoded protein. After a ribozyme has bound and cleaved its RNA
20~target lt is released from that RNA to search for another ~-~ target~ and can repeatedly bind and cleave new targets.
The enzymatic nature of a ribozyme is advantageous over other technologies, such as antisense ~technology twhere a nucleic acid molecule simply binds to --25 a nucleic acid target to block its translation) since the effective concentration of ribozyme ne`cessary to effect a therapeutic treatment is lower than that of an antisense oligonucleotide. This advantage reflects the ability of the ribozyme to act enzymatically. Thus, a single ~ "
ribozyme molecule is able to cleave many ~molecules of target RNA. In addition, the ribozyme is a highly specific inhibitor, with the specificity of inhibition depending not only on the base pairing mechanism of binding, but also on the mechanism by which the molecule inhibits the expression of the RNA to which it binds.
That is, the inhibition is caused by cleavage of the RNA
target and so specificity is defined as the ratio of the :

.~ 1 3 .i ~ 1t.~
W093/230~7 PCT/US93/045~

i i rate of cleavage of the targeted RNA over the rate of cleavage of non-targeted RNA. This cleavage mechanism is dependent upon factors additional to those involved in base pairing. Thus, it is thought that the specificity of action of a ribozyme is greater than that of antisense oligonucleotide binding the same RNA site. ~-This class of chemicals exhibits a high degree of specificity for clea~age of the intended target mRNA.
Consequently, the ribozyme agent will only affect cells expressing that particular gene, and will not be toxic to normal tissues.
Thus, the invention features novel enzymatic RNA
molecules, or ribozymes, and methods for their use for inhibiting cancer-related mRNA expression. Such ribozymes can be used in a method for treatment of disease caused by expression of the cancer-related genes in man and other animals, including other primates. This conclusion, as noted above,~ is based upon the finding that many forms of cancer~become~ unresponsive to certain chemotherapeutic agents~as a result of overexpression of, e.a., the mdr-1 gene. The advantage of using ribozymes of the present invention is their ability to specifically cleave the :~ ,;
targeted mRNA, ultimately leading to a reduction in target gene ~activity throu~h a decrease in level of the gene

2~5~ product~. Use of mdr-1 specific ribozymes removes the --- mechanism of drug resistance used by transformed cells, and thus enhances drug therapies for tumor cell growth.
These agents can be administered prior to and during chemotherapeutic treatment of those neoplasias known to have a high incidenc'e of drug resistance, or can be used prophvlactically for all neoplasias.
The invention can also be used to treat cancer or pre-neoplastic conditions. Two preferred - administration protocols can be used, either i~ vivo administration to reduce the tumor burden, or ex vivo treatment to eradicate transformed cells from tissues such as bone marrow prior to reimplantation.

~3/23057 ~ t~ ~ 9 PCT/U~93/04573 . .
, Thus, in a first aspec~, the invention features an enzymatic RNA molecule (or ribozyme) which cleaves mdr-1 mRNA (i.e., mRNA expressed from the mdr-1 gene), or its equivalent. In particular, the invention features hammerhead ribozymes designed to cleave accessible areas of the mdr-1 mRNA. Such areas include tr.~se sequences shown in Fig. 2.
In a second aspect, the invention features an enzymatic RNA molecule (or ribozyme) which cleaves mRNA
associated with development or maintenance of CML, promyelocytic leukemia, surkitt~s lymphoma or acute lymphocytic leukemia, follicular lymphoma, s-cell acute lymphocytic leukemia, breast cancer, colon carcinoma, neuroblastoma, and lung cancer, including mRNA targets disclosed in Figs. 3 to 11. Such mRNA is recognized by those in the art to encode an aberrant cel.ular protein which is able to control cellular proliferation, and is directly linked to (correlated with) the presence of the leukemic phenotype.
20 ~ By "enzymatic RNA molecule~ it is meant an RNA
olecule which has complementarity in a substrate binding -, :
region to a specified mRNA target, and also has an ~ enzymatic activity which is active to specifically cleave -~ RNA in that mRNA. That is, the enzymatic RNA molecule is able to intermolecularly cleave mRNA and thereby inactivate a target~mRNA molecule. This complementarity functions to allow sufficient hybridization of the ~, enzymatic RNA molecule to the target RNA to allow the cleavage to occur. For in ~ivo use, such complementarity may be be'tweén `30 and '45 bases. One hundred percent -complementarity is preferred, but complementarity as low as 50-75% may also be useful in this invention.
: By 11 equivalentll RNA to mdr-1 mRNA is meant to include those naturally occurring mRNA molecules associa'ced with neoplastic diseases in various animals, including humans, and other primates, which have similar ~ ~ .
~ - structures and functions to that mdr-1 mRNA in humans.

~ J ~
w0~3/230~7 PCT/US93/0457~: -The deduced sequences of the mouse and human P-glycoproteins are 80~ identical.
In preferred embodiments, the e~zymatic RNA
molecule is formed in a hammerhead motif, but may also be formed in the motif of a hairpin, hepatitis delta virus, group I intron or RNaseP-like RNA (in association with an RNA guide sequence). Examples of such hammerhead motifs are described by Rossi et al., 8 Aids Research and Human Retroviruses 183, 19~2; of hairpin motifs by Hampel et al., ~RNA Catalyst for Cleaving Specific RNA Sequences'~, filed September 20, 1989, which is a continuation-in-part of U.S. Serial No. 07t247,100 filed September 20, 1988, Hampel and Tritz, 28 Biochemistrv 4929, 1989 and Hampel et al., 18 Nucleic Acids Research 299, 1990; an example of the hepatitis delta virus motif is described by Perrotta and Been, 31 B1ochemistrY 16, 1992; of the RNa~seP motif by Guerrier-Takada ~et al.~, 35 Cell 849,~ 1983; and of the group~ ntron;by Cech et al., U.S. Patent 4,987,071.
These~specif1c motlfs are not limiting ln the invention 20~-a~d~those ~skilled~ n the art will recognize that all that is~ lmportant ln an enzymatic RNA molecule of this invention is that it has a specific substrate binding site which is~ complementary to~one or more of the target mRNA
regi~ons,~ and that it have nucleotide sequences within or 25~ surrounding that substrate binding site which impart an mRNA~cleaving activity~to the molecule.
In a related aspect, the invention features a : . ,. - . - , mammalian cell which lncludes an enzymatic RNA molecule as described above. Preferably, the mammalian cell is a human or other primate cell.
In another related aspect, the in~ention features an e~pression vector which includes nucleic acid encoding the enzymatic RMA molecules described above, located~in the vector, ~ , in a manner which allows expression of that enzymatic RNA molecule within a mammalian cell.

~"

.) 93t230~7 r~ 1 3, ~ q 9 Pcr/ US93/0457 t~

In yet another related aspect, the invention feature~ a method for treatment of an mdr-l gene-related disease, chronic myelogenous leukemia (CML), promyelocytic leukemia, surkitt~s lymphoma or acute lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic leukemia, breast cancer, colon carcinoma, neuroblastoma, or lung cancer, by administering to a patient an enzymatic RNA molecule as described above.
In another related aspect, the invention features a method for treatment of CML by ex vivo treatment of blood or marrow cells with an enzymatic RNA
molecule as described above.
The invention provides a class of chemical cleaving agents which exhibit a high degree of specificity for ~he mRNA causative of CML, promyel~cytic leukemia, Burkitt's ~lymphoma or acute lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic leukemia, ~ breas~ cancer, colon ~carcinoma, and neuroblastoma. If - desired, such~ ribozymes can be designed to target 2~0~ equ~1valent~ slngle-stranded DNAs by methods known in the art.~The~ribozyme molecule is preferably targeted to a highly cQnserved sequence region of the mdr-1 mRNA. Such enzymatic RNA molecules can be delivered exogenously to affect~ed~cells or endogenously to infected cells. In the 25 ~preferred hammerhead~ motif~the small size (less than 40 nucleotides, preferably between 32 and 36 nucleotides in length) of~the molecule allows the cost of treatment to be reduced compared to other ribozyme motifs.
-~ The smallest ribozyme delivered for any type of treatment`reported to date (by Rossi et al.,'1992, supra) -is an in vitro transcript having a length of 142 -~ nucleotides. Synthesis of ribozymes greater than 100 - nucleotides in length is very difficult using automated methods~ and the therapeutic cost of such molecules is prohibitive. Delivery of ribozymes by expression vectors is primarily feasible using only ex vivo treatments. This limits the utility of this approach. In this invention, w093t230~7 PCT/US93/04 small ribozyme motifs ~e.q., of the hammerhead structure, shown generally in Fig. 1) are used for exogenous delivery. The simple structure of these molecules also increases the ability of the ribozyme to invade targeted S regions of the mRNA structure. Thus, unlike the situation when the hammerhead structure i5 included within longer transcripts, there are no non-ribozyme flanking sequences to interfere with correct folding of the ribozyme structure, as well as complementary binding of the ribozyme to the mRNA target.
The enzymatic RNA molecules of this invention can be used to treat human CML, promyelocytic leukemia, Burkitt's lymphoma, acute lymphocytic leukemia, follicular lymphcma, B-cell acute lymphocytic leukemia, breast cancer, or lung cancer. Affected animals can be treated at the time of cancer, or in a prophylactic manner. This -~ timing of treatment wi~l~l reduce the number of affected cells and disable cellular replication. This is possible because the ribozymes are designed to disable those 20~ structures required for successful cellular proliferation.
Ribozymes of this invention may be used as diagnostic tools to examine genetic drift and mutations within diseased cells. The close relationship between rlbozyme activity and the structure of the target RNA
allows the detection of mutations in any region of the molecule which ~alters the base-pairing and three-dimensional ~structure of the target RNA. By using multiple ribozymes described in this invention, one may map nucleotide changes which are important to RNA
structu~re and functioh in vitro, as well as in cells and tissues. Cleavage of target RNAs with ribozymes may be used to inhibit gene expression and define the role (essentially) of specified gene products in the progression of disease. In this manner, other genetic targets may be defined as important mediators of the disease. These experiments will lead to better treatment ~- ~ of the disease progression by affording the possibility of .
, ~93/23057 ~ 3 1 ~ ~ 3 PCT/US93/045 combinational therapies (e.q., multiple ribozymes targeted to different genes, ribozymes coupled with known small molecule inhibitors, or intermittent treatment with combinations of ribozymes and/or other chemical or biological molecules).
Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims.
Descri~tion of the Preferred Embodiments -The drawings will first briefly be described.
Drawinas Fig. 1 is a diagrammatic representation of a hammerhead motif ribozyme showing stems I, II and III
(marked (I), (II) and (III) respectively) interacting with lS a target region. The 5' and 3~ ends of both ribozyme and targ~et are shown. Dashes indicate base-paired nucleot~ides.
Figs. 2 ~- 11 are preferred targets for mdr-l gene, chronic myelogenous leukemia, promyelocytic leukemia, ~Burkitt~s lymphoma or acute lymphocytic - leukemia~,~follicular lymphoma, B-cell acute lymphocytic leukemla~, breast cancer, colon carcinoma, neuroblastoma, and l-ung cancer, respectively.
; Tarqet Sites~
2S ~ ~ Ribozymes targeting selected regions of mRNA
associ~ated with tumor cell growth are preferably chosen to cleave the target RNA in a manner which inhibits ~translation of the mRNA. Genes are selected such that inhibition of translation will preferably inhibit cell ~- 30 replication, e.a., by inhibiting production of a necessary ~- protein. Selection of effective target sites within these critical regions of mRNA entails testing the accessibility , ~, ~., of the target mRNA to hybridization with various oligonucleotide probes. These studies can be performed 35~ using~ RNA or DNA probes and assaying accessibility by cleaving the hybrid molecule with RNaseH (see below).
Alternatively, such a study can use ribozyme probes j 1 3 r~ 3 ~ -W~93/23057 PCT/US93/04 designed from secondary structure predictions of the mRNAs, and assaying cleavage produc~s by polyac~ylamide gel electrophoresis (PAGE), to detect the presence of cleaved and uncleaved molecules.
The following are examples of cancer conditions which can be targetted in this manner.
Chronic MYeloqenous Leukemia Chronic myelogenous leukemia exhibits a characteristic disease course, presenting initially as a chronic granulocytic hyperplasia, and invariably evolving into an acute leukemia which is caused by the clonal expansion of a cell with a less differentiated phenotype (i.e., the blast crisis stage of the disease). CML is an unstable disease which ultimately progresses to a terminal stage which resembles acute leukemia. This lethal disease ~ affects approximately 16,000 patients a year.
;~ Chemotherapeutic a~gents such as hydroxyurea or busulfan ~ can reduce the leukemic burden but do not impact the-life . ~
expectancy~of the patient (e.q., approximateiy 4 years).
Consequently, CML patients are candidates for bone marrow - transplantation (BMT) therapy. However, for those patients which survive BMT, disease recurrence remains a major obstacle. Apperley et al., 69 Br. J. Haematol. 239, 1988.
~25 ~ The Philadelphia (Ph) chromosome which results ; ~ ~ from the translocation of the abl oncogene from chromosome -~9 to the bcr gene on chromosome 22 is found in greater than 95% of CML patients and in 10-25% of all cases of acute lymphoblast1c leukemia (ALL). Fourth International Workshop on Chromosomes in Leukemia. 11 Cancer Genet.
CYtoqenet. 316, 1982. In virtually all Ph-positive CMLs and approximately 50% of the Ph-positive ALLs, the leukemic cells express bcr-abl fusion mRNAs in which exon 2 (b2a2 junction) or exon 3 (b3a2 junction) from the major ` 35 breakpoint cluster region of the bcr gene is spliced to exon 2 of the abl gene. Heisterkamp et al., 315 Nature 758, 1985, Shtivelman et al., 69 Blood 971, 1987. In the . ~
-i~93/23057 ~ PCT/US93/04~7~
~.

remaining cases of Ph-positive ALL, the first exon of the bcr gene is spliced to exon 2 of the abl gene. Hooberman et al., 86 Proc. Natl. Acad. Sci. USA 4259, 1989, HeisterXamp et al., 16 Nucleic Acids Research 10069, 1988.
The b3a2 and b2a2 fusion mRNAs encode 210 kd bcr-abl fusion proteins which exhibit oncogenic activity.
Daley et al., 247 Science 824, 1990, Heisterkamp et al., 344 Nature 251, 1990. The importance of the bcr-abl fusion protein Ip210bCr-abl) in the evolution and maintenance 10 of the leukemic phenotype in human disease has been -demonstrated using antisense oligonucleotide inhibition of p210bC~-abl expression. These inhibitory molecules have been shown to inhibit the in vitro proliferation of leukemic `-cells in bone marrow from CML patients. Szczylik et al., 253 Science 562, 1991. -- c-MYc Gene~
;c-Myc, when activated, can lnduce malignancy in ~ a~;var~i-et`y o~f tissues, most notably hematopoietic tissues ?,`,'~ (Leder~et al.,~222 Sclence 765, 1983). The~most common mechanl~sm of; c-myc activation is translocation to any of - the immunoglobulin (Ig) or T-cell receptor~loci during lymphoid maturation (Croce and Nowe11, 65 Blood 1, 1985;
Kl;ein ~and~ Klein, 6 Immunol. Today 208, 1985). For example,~ ln Burkitt's lymphoma the c-myc locus on chromosome 8 translocates most often to the Ig heavy chain locus on chromosome 14, but also to the lambda or kappa light chain Ig genes on chromosomes 2 and 22 (Magrath, in ~; ~ "Epstein-Barr Virus and Associated Diseases", M. Nijhoff Publishing:631l j1,986).~ , In some instances the c-myc transcription region is altered in the non-coding exon 1 region; in such cases transcription is initiated at a cryptic promoter present in the first intron of the c-myc locus. These rearrangements are thought to lead to - deregulation of c-muc expression.
c-Myc is not normally expressed in quiescent cells, but is temporally expressed in actively-dividing , ~ 1 r;3 5 4 ~
W093/230~7 PCT/US93/04 cells, most prominently during transition from Go to G1 phases of growth induction.
Experiments with transfected cell lines and transgenic animals have shown that c-myc activation plays a critical role, but is not sufficient for transformation (Adams et al., 318 Nature 533, 1985; Lombardi et al., 49 Cell 161, 1987; Schwartz et al., 6 Mol. Cell. Biol. 3221, 1986; Langdon et al., 47 Cell 11, 1986). Targeted inhibition of c-myc expression in tumor cell lines using antisense oligonucleotides has shown that c-myc expression is required for growth in certain lymphomas (McManaway et al., 335 Lancet 808, 1990).
Bcl-2 Gene The bc1-2 gene is abnormally expressed in about 15~ 85% of follicular lymphomas and about 20% of diffuse lymphomas due to a t(14,18)(q32;q21) ~ chromosomal rearrangement~between the bc1-2 locus on chromosome 18 and the immunoglobulin heavy chain locus on chromosome 14 (Yunis et al.,~ 316 N. Ena1. J. Med. 79, 1987). This 20;:chromosoma1~rearrangement represents the most common found in~-ly ~ id malignancies in humans~ A bc7-2/IgH fusion mes~sag~e~is expressed; however, the bc1-2 protein-coding ~-~ region is not interrupted since the major breakpoint region~lies~ ln the 3l nontranslated region of the bc1-2 transcript (Cleary et al.,~47 Cell 19, 1986). The bc1-2 ,,, - , gene repre~sents a new form~of proto-oncogene in`that it encodes a ~mitochondrial protein which inhibits cell senescence (Hockenbery et al., 348 Nature 334, 1990), leading to extended survival of B-cells transfected with this gene (Nunez et al., 86 Proc. Natl. Acad. Sci. USA
4589, 1989).
At least three different forms of bc1-2 mRNAs are found 1n~pre-B-cells and T-cells, which vary due to alternative splicing and;promoter usage. Two different proteins are produced, a 21 kD and a 26 kD peptide which vary at their carboxytermini. Both forms have identical ... .

93/23057 ,~ 3~ PCT/US93/04573 ;~

N termini encoded in exon 2 of the gene. Consequently, this region would be suitable for ribozyme targeting. -Breast Cancer ~;
The epidermal growth factor (EGF) receptors have been implicated in human cancer more frequently than any other family of growth factor receptors. The EGF receptor gene is often amplified or overexpressed in squamous cell carcinomas and glioblastomas. Jenkins et al., 39 Cancer `
Genet. C~toaenet. 253, 1989. Similarly, erbB-2 ls often `~
10 overexpressed in adenocarcinomas of the stomach, breast ~-~
and ovary. Turc-Carel et al., 12 ibid. 1, 1984.
Overexpression of either gene under appropriate experimental conditions confers the transformed phenotype.
Heim et al., 32 ibid. 13, 1988. In certain breast carcinomas, the erbB-3 gene is overexpressed. Boehm et al., 7 EMBO J. 385, 1988.
The high incidence of human breast cancer has prompted efforts to model the disease in transgenic mice.
The myc gene is amplified in some human breast cancers, 2~0~Escot ;et ~al.~, 83 Proc. Natl. Acad. Sci. USA 4834, 1986, and ras mutations have been observed. Barbacid, 56 Ann.
Rev. Biochem. 779, 1987.~ Reproduct~ion of disease by ~expresslon of the myc or ras genes in mice have given only -~ ~ sporadlc~ results.
~ ~ Breast cancer progression often correlates with amplification of the tyrosine kinase receptor gene denoted as c-erb-B2 or neu. The ligand for this receptor is unknown. Male and female mice expressing the neu gene both synchronously developed adenocarcinomas encompassing the entire gland.` Muller èt; al., 54 Cell 105, 1988.
Other strains developed tumors stochastically. Bouchard et al., S7 ibid. 931, 1989.
-Cs~sEL5~_ cinoma ,~ .
- The~platelet derived growth factor (PDGF) system has served as a prototype for identification of substrates of the receptor tyrosine kinases. Certain enzymes become ;~ activated by the PDGF receptor kinase, including ':
, phospholipase C and phosphatidylinositol 3' kinase, Ras guanosine triphosphate (GTPase) activating protein (GAP) and src-like tyrosine kinases. GAP regulates the function of the Ras protein. It stimulates the GTPase activity of the 21 kD Ras protein. Barbacid, 56 Ann. Rev. Biochem.
779, 1987. Microinjection of oncogenically activated Ras into NIH 3T3 cells induces DNA synthesis. Mutations that cause oncogenic activation of ras lead to accumulation of Ras bound to GTP, the active form of the molecule. These mutations block the ability of GAP to convert Ras to the inactive form. Mutations that impair the interactions of Ras with GAP also block the biological function of Ras.
While a number of ras alleles exist (N-ras, K-ras, H-ras) which have been implicated in carcinogenesis, the type most often associated with colon and pancreatic carci:nomas is~the K-ras. Ribozymes which are targeted to certain regions of thé K-ras allelic mRNAs may also prove inhibitory t;o the functlon of the other allelic mRNAs of ; the N-ras~and~H-ras genes.
; 20-~Lun~ Cancer/L-mYc Gene Expression of the myc oncogene is known to alter cell growth in a number of tissues. The product of this gene lS~ ~a protein which is known to be a transcriptional activator that can act singly or in combination with other oncogene~ proteins. The L-myc gene is often activated by tra~slo~ations of DNA from other regions of the genome to the~regulatory regions 5' of the m~c gene ORF. After transcription of the L-myc mRNA, alternate splicing of the ., .
transcript is known to occur. Kaye et al., 8 Mol. Cell 30 ~Biol. 196, 1988. !` The ai~ternate mRNAs produced contai~ 'a common 5' exon 1 and portions of a common exon 2. These common regions of m~NA structure allow the use of nucleic acid targeted therapeutics which can inactivate both species of mRNA with one therapeutic molecule.
,,: , .
PromYeloc~tic Leukemia Acute promyelocytic leukemia is characterized by ~- a specific translocation, a t(15;17)(q22;qll.2-12), which ~ :' `

` ~ o g3/23~57 ~ ~ 3 ~J~,~g PCT/US93/0457~

~5 is found in some 90% of the cases. The t(lS;17) is often the only detectable cytogenetic abnormality present in the leukemic cells. This rearrangement results in the fusion of two genes, the promyelocytic leukemia gene t PML) on chromosome 15, and the retinoic acid receptor alpha gene (RARA) on chromosome 17 (J. Borrow et al., 249 Science 1577, 1990; H. de Thé et al., 347 Nature 558, 1990). The RARA is a hormonally-responsive transcriptional regulatory protein, while the function of the PML is as yet unknown.
A fusion message is expressed in the leukemic cells which encodes the N-terminal coding region of the PML gene and the C-terminal coding region of the RARA
gene. Expression of this fusion gene apparently inhibits normal myeloid differentiation. The biological relevance lS of this rearrangement to the etiology of the disease has been exemplified by the discovery that all-trans retinoic acid can~be used to achieve complete clinical remission, presumàbly~by~inducing differentiation of the leukemic cells. ;This suggests that the fusion protein is still hormonally responsive.
, .
The treatment of leukemic cells with retinoic -~ acid is not preferable over the long term because retinoic acld is a generalized inducer of differentiation in all cell ~types, ~not just leukemic cells. Thus, systemic ;25~adminl~stration of these compounds can lead to a number of deleterious side effects by differentiating cells which should not be~ in a differentiated state. A treatment which gi~es suppression of the transformed phenotype in leukemic cells without affecting other cell types is preferable, as deslcribed herein.
B-Cell Acute LYm~hocv_ic Leukemia Leukemia comprises some 3% of the new cancer ;~ - cases~per year, with lymphocytic leukemias accounting for approximately half (National Cancer Institute, 1990 ~; 35 statistics). A subset of lymphocytic leukemias of the ;acute pre-B-cell type are associated with a specific chromosomal translocation, a t(l;l9)(q23;pl3.3) (M.B.

WO93/~30s7 ~l~ 5 4 9 3 PCT/US93/045 ~amps et al., 60 Cell 547, 1990; J. Nourse et al., ibid.
p.535). This rearrangement results in the fusion of two genes, the PBX gene present on chromosome 1, and the E2A
gene present on chromosome 19. While the E2A transcript is found in all B-cell types, the PRL gene is not normally expressed in B-cells. However, an E2A/PRL fusion message is constitutively expressed from this aberrant locus in the leukemic cells. This fusion message encodes the N-terminal region of the E2A, including the transcriptional activating domain of that gene, and the C-terminal region of the PRL gene, which contains a homeodomain DNA binding motif. Thus, a potentially functional chimeric transcriptional regulatory protein is expressed in the leukemic cells.
15The PRL sequences found in the fusion mRNA are : ~ good targets for ribozyme therapy since PRL iS not ~- ~expressed~ in non-leukemic B-cells. Whether the E2A
sequènc~ès~ can ~be targeted by ribozymes is unclear since such~ribozymes~may inhlbit E2A expression in normal B-20~ ~cells.~It~is~not known how normal B-cells are affected by inhibition of E2A.
me following is but one example of a method by which suitable target sltes~can be identified and is not ~ miting ~ln~this invention. Generally, the method s~ 25~ ~involve~s~ identlfylng potential cleavage sites for a hammerheàd~ribozyme, and then testing each of these sites to~determine their suitability as targets by ensuring that secondary structure formation is minimal.
The mRNA sequences are compared in an appropriate target region. Putative ribozyme cleavage sites are identified from weak or non-base paired regions of the mRNA. These sites represent the preferred sites for hammerhead or other ribozyme cleavage within these target mRNAs.
,. . ~
~ ~ Short RNA substrates corresponding to each of the mRNA sites are designed. Each substrate is composed - of two to three nucleotides at the 5~ and 3~ ends that ;. ~93/23057 ~ PCT/US93/0~573 ~s`

will not base pair with a corresponding ribozyme recognition region. The unpaired regions flank a central region of 12-14 nucleotides to which complementary arms in the ribozyme are designed.
The structure of each substrate sequence is predicted using a PC fold computer program. Sequences which give a positive free energy of binding are accepted.
Sequences which give a negative free ener~y are modified by trimming one or two bases from each of the ends. If the modified sequences are still predicted to have a strong secondary structure, they are rejected.
After substrates are chosen, ribozymes are designed to each of the RNA substrates. Ribozyme folding is also analyzed using PC fold.
Ribozyme molecules are sought which form ,~ hammerhead motif stem II (see Fig. 1) regions and contain flanking~arms which are devoid of intramolecular base pairing. Often t-he ribozymes are modified by trimming a base f rom the~ ends of the rlbozyme, or by introducing ~add~ltional~base pairs in stem II to achieve the desired f~old~ Ribozymes with incorrect folding are rejected.
After substrate/ribozyme pairs are found to contain correct~ intramolecular structures, the molecules are foldedl together to predict 1ntermolecular interactions.
-~ ~ 25 A schematic representation of a ribozyme with its coordlnate base pairing to its cognate target sequence is shown in~Fig~ 1. Examples of useful targets are listed in Figs. 2 Those targets thought to be useful as ribozyme targets can be tested to determine accessibility to ~- nucleic acid probes in a ribonuclease H assay (see below).
This assay provides a quick test of the use of the target site without requiring synthesis of a ribozyme. It can be used to~screen~for~sites most suited for ribozyme attack.
35~ SYnthesis of Ribozymes Ribozymes useful in this invention can be ; produced by gene transcription as described by Cech, ., ''' WO93/230s7 ~ g ~CT/US93/045 supra, or by chemical synthesis. Chemical synthesis of RNA is similar to that for DNA synthesis. The additional 2~-OH group in RNA, however, requires a different protecting group strategy to deal with selective 3'-5' internucleotide bond formation, and with RNA
susceptibility to degradation in the presence of bases.
The recently developed method of Z~NA synthesis utilizing the t-butyldimethylsilyl group for the protection of the 2' hydroxyl is the most reliable method for synthesis of ribozymes. The method reproducibly yields RNA with the correct 3'-5' internucleotide linkages, with average coupling yields in excess of 99%, and requires only a two-step deprotection of the polymer.
A method based upon H-phosphonate chemistry of phosphoramidites gives a relatively lower coupling efficiency than a method based upon phosphoroamidite chemis~try.~ Thls is a problem ~for synthesis of DNA as well~ A~ promising approach to scale-up of automatic ol~igonuc-leotide~synt~e~sls has been described recently for 2~0 the~ H~-phosphonates~. A combination of a proper coupling time~and~additional capping of ~Ifailure~ sequences gave high yields in the synthesis of oligodeoxynucleotides in scales ~in~ the range of 14 -~moles with as little as 2 e~uivalents of a~monomer in the coupling step. Another 2S~alternatlve~approach lS to use soluble polymeric supports polyethylene glycolsj, instead of the conventional ;solid ~ ~supports. This method can yield short oligonucleotides in hundred milligram quantities per batch utiliz~ng about 3 equivalents of a monomer in a coupl1ng 30~ step. `
Various modifications to ribozyme structure can be made to enhance the utility of ribozymes. Such --~ modifications~will enhance shelf-life, half-life in vitro, stability, and ease of introduction of such ribozymes to 35 ~the targ~et site, s~g~, to enhance penetration of cellular membranes, and confer the ability to recognize and bind to targeted cells.

:~ , ;

~93/23057 PCT/US93/0457~ ~

1 9 . :
Exogenous delivery of ribozymes benefits from chemical modification of the backbone, e.q., by the overall negative charge of the ribozyme molecule being reduced to facilitate diffusion across the cell membrane.
The present strategies for reducing the oligonucleotide charge include: modification of internucleotide linkages by methylphosphonates, use of phosphoramidites, linking oligonucleotides to positively charged molecules, and creating complex packages composed of oligonucleotides, lipids and specific receptors or effectors for targeted cells. Examples of such modifications include sulfur-containing ribozymes containing phosphorothioates and phosphorodithioates as internucleotide linkages in RNA.
Synthesis of such sulfur-modified ribozymes is achieved by use of the sulfur-transfer reagent, 3H-1,2-benzenedithiol-

3-one 1,1-dioxide. Ribozymes may also contain ribose modified ribonucleotides. Pyrimidine analogues are prepared from uridine using a procedure employing - diethylamino sulphur trifluoride ~DAST) as a starting 20 ~reagent~. ~ Ribaz ~es can also be either electrostatically -~ or covalently attached to polymeric cations for the purpose of reducing charge. The polymer can ~e attached ~-- !~' ~ to the ribozyme by simply converting the 3'-end to a - ~ ribonucleoside dialdehyde which is obtained by a periodate cleavage of the terminal 2~,3~-cis diol system. Depending on the specific requirements for delivery systems, other possible modifications may include different linker arms containing carboxyl, amino or thiol functionalities. Yet further~examples include use of methylphosphonates and 2'-O-methylribose and 5' or 3' capping or blocking with m7GpppG or m32~7GpppG.
For example, a kinased ribozyme is contacted with guanosine triphosphate and guanyltransferase to add a m3G cap to the ribozyme~ After such synthesis, the ribozyme can be gel purified using standard procedure. To ensure that the ribozyme has the desired activity, it may be tested with and without the 5' cap using standard ..

WC) 93/ 230 !7 PCr/ US93/0457 `3 9 procedures to assay both its enzymatic activity and its stability.
Synthetic ribozymes, including those containing various modifiers, can be purified by high pressure liquid chromatography tHPLC). Other liquid chromatography techniques, employing reverse phase columns and anion exchangers on silica and polymeric supports may also be used.
There follows an example of the synthesis of one ribozyme. A ~olid phase phosphoramidite chemistry was employed. Monomers used were 2'-tert-butyl-dimethylsilyl cyanoethylphosphoramidities of uridine, N-benzoyl-cytosine, N-phenoxyacetyl adenosine and guanosine (Glen Research, Sterling, VA). Solid phase synthesis was carried out on either an ABI 394 or 380B DNA/RNA
synthesizér using the standard protocol provided with each machine. ~The only~exception~ was that the coupling step was increased f~rom 10 to 12 minutes. The phosphoramidite concentratlon was 0.1 M. Synthesis was done on z 1 ~mole 20~ sc~alè~uslng~a~l ~ le RNA reaction column (Glen Research).
The~average~coupling efficiencies were between 97% and 98%
for~the 394 model, and between 97% and 99% for the 380B
model, ~as determined by a calorimetric measurement of the released trltyl cation.
25~ Blocked ribozymes were cleaved from the solid support (~ , CPG), and the bases and diphosphoester moiety ~deprotected in a sterile vial by dry ethanolic ammonia (2 mL) at 55C for 16 hours. The reaction mixture was cooled on dry ice. Later, the cold liquid was transferred into a sterile screw cap vial and lyophi`lized.
To remove the 2'-tert-butyl-dimethylsilyl groups from the ribozyme, the residue was suspended in 1 M tetra-n-butylammonium fluoride in dry THF (TBAF), using a 20-" .
fold ex~ess of the reagent for every silyl group, for 16 ~- 35 hours at ambient temperature (about 15-25C). The reaction was quenched by adding an equal volume of sterile .

93/23057 ~ 4 ~ ~ PCT/US93/04~7 1 M triethylamine acetate, pH ~.5. The sample was cooled and concentrated on a SpeedVac to half the initial volume.
The ribozymes were purified in two steps by HPLC
on a C4 300 A 5 mm DeltaPak column in an acetonitrile gradient.
The first step, or ~trityl on" step, was a separation of 5'-DMT-protected ribozyme(s) from failure sequences lacking a 5'-DMT group. Solvents used for this step were: A tO.1 M triethylammonium acetate, pH 6.8~ and -B (acetonitrile). The elution profile was: 20% B for 10 minutes, followed by a linear gradient of 20% B to 50% B
over 50 minutes, 50% B for 10 minutes, a linear gradient of 50% B to 100% B over 10 minutes, and a linear gradient ~ ~ .
of 100% B to 0% B over 10 minutes.
15The ~second step was a purification of ~a completely deblocked ribozyme by a treatment of 2%
trifluoroacetlc acid on a C4 300 A 5 mm DeltaPak column in an ac~etonitrile gradient. Solvents used for this second ---step were:~ A ~0.1 M triethylammonium acetate, pH 6.8) and ',!
20 ~B~(80%~ acetonitrile, 0.1 M triethylammonium ~acetate, pH
6.~8~ The elution profile was: 5% B for 5 minutes, a linear ~gradient of 5% B to 15% B over 60 minutes, 15% B :~"
for 10~mlnutes, and a linear gradient of 15% B to 0% B j-~
over 10 minutes. l~
25~ The fraction containing ribozyme was cooled and r',' lyophillzed on a SpeedVac. Solid residue was dissolved in a minimum amount; of ethanol and sodium perchlorate in acetone. The ribozyme was collected by centrifugation, ~
washed three times with acetone, and lyophilized. ~`
ExPression Vector While synthetic ribozymes are preferred in this -~
invention, those produced by expression vectors can also be used. In designing a suitable ribozyme expression vector the following factors are important to consider. ~-The final ribozyme must be kept as small as possible to minimize unwanted secondary structure within the ribozyme. -~- A promoter (e.q., a T7, human cytomegalovirus immediate ,, W093/23057 PCT/US93/0457'~

early (iel), human beta actin, or U6 snRNA promoters) should be chosen to be a relatively strong promoter, and expressible both in vitro and in vivo (e.q., by co-infection with the T7 RNA polymerase gene, the human cytomegalovirus immediate early (iel) or human beta actin promoters). Such a promoter should express the ribozyme ~-at a level suitable to effect production of enough ribozyme to destroy a target RNA, but not at too high a level to prevent other cellular activities from occurring (unless cell death itself is desired).
A hairpin at the 5' end of the ribozyme is useful to ensure that the required transcription initiation sequence (GG or GGG or GGGAG) does not bind to some other part of the ribozyme and thus affect regulation -of the transcription process. The 5' hairpin is also useful to protect the ribozyme from 5'-3' exonucleases. ~`~
~ A~selected~halrpln at the 3' end of the ribozyme gene is -~ useful since it acts as a transcription termination slgnal, and~ protects the ribozyme from 3'-5' exonuclease ~ 20 actlvity. One example of a known termination signal is r~ that~present on the T7 RNA polymerase system. This signal i;s~about~30~ nucleotides in length. Other 3' hairpins of shorter length can be used to provide good termination and RNA stability. Such hairpins can be inserted within the ~:
vector sequences to allow standard ribozymes to be placed in an ~appropriate orientation and expressed with such sequences attached.
Poly(A) tails are also useful to protect the 3' end of the ribozyme. These can be provided by either including 'a poly(A) signal slte in the expre~ssion vector' (to signal a cell to add the poly(A) tail in vivo), or by ~ -~ introducing a poly(A) sequence directly into the - - expression vector. In the first approach the signal must - be located to prevent unwanted secondary structure formation with other parts of the ribozyme. In the second appr~ach, the poly(A) stretch may reduce in size over time when expressed in vivo, and thus the vector may need to be , .
:: .

93/230~7 PCT/US93/0457 checked over time. Care must be taken in addition of a poly(A) tail which binds poly~A) binding proteins which prevent the ribozyme from acting.
RibozYme Tes inq Once the desired ribozymes are selected, synthesized and purified, they are tested in kinetic and other experiments to determine their utility. An example of such a procedure is provided below.
Pre~aration of Ribozvme Crude synthetic ribozyme (typically 350 ~g at a time) was purified by separation on a 15% denaturing polyacrylamide gel (0.75 mm thick, 40 cm long) and visualized by W shadowing. Once excised, gel slices containing full length ribozyme were soaked in 5 ml gel elution buffer (0.5 M NH40Ac, 1 mM EDTA) overnight with shaking at 4C. The eluent was desalted over a C-18 matrix (Sep-Pak cartridges, Millipore, Milford, MA) and vacuum dried. The dried RNA was resuspended in 50-100 ~1 ~ TE (TRIS 10 mM, EDTA 1 mM, pH 7.2). An aliquot of this --f~ 20 ~so~1ution was diluted 100-fold into 1 ml TE, half of which was used to spectrophotometrically quantitate the ribozyme solution. The concentration cf this dilute stock was typically 150~800 nM. Purity of the ribozyme was confirmed by the presence of a single band on a denaturing - .
polyacrylamide gel.
A ribozyme may advantageously be synthesized in two or more portions. Each portion of a ribozyme will generally have only limited or no enzymatic activity, and the actlvity will increase substantially (by at least 5-10 fold) when all portions are ligated (or otherwise juxtaposed) together. A specific example of hammerhead ribozyme synthesis is provided below.
~ .
- The method involves synthesis of two (or more) shorter "half" ribozymes and ligation of them together using T4 RNA ligase. For example, to make a 34 mer ribozyme, two 17 mers are synthesized, one is phosphorylated, and both are gel purified. These purified w093/~3057 ~,~ 3~4~3 PCT/U593/045 17 mers are then annealed to a DNA splint strand complementary to the two 17 mers. (Such a DNA splint is not always necessary.) This DNA splint has a sequence designed to locate the two 17 mer portions with one end of each adjacent each other. The juxtaposed RNA molecules are then treated with T4 RNA ligase in the presence of ATP. The 34 mer RNA so formed is then HPLC purified.
Preparation of Substrates Approximately 10-30 pmoles of unpurified substrate was radioactively 5' end-labeled with T4 polynucleotide kinase using 25 pmoles of [~_32p] ATP. The entire labeling mix was separated on a 20% denaturing polyacrylamide gel and visualized by autoradiography. The full length band was excised and soaked overnight at 4C
in 100 ~l of TE (10 mM Tris-HCl pH 7.6, 0.1 mM ~DTA).
Kinetic Reactions For reactions using short substrates (between 8 and 16 bases) a substrate solution was made lX in assay ; buf~er (75 mM Tris-HCl, pH 7.6; 0.1 mM EDTA, 10 mM MgCl2) such that~ the concentration of substrate was less than 1 nM. A ribozyme solution (typically 20 nM) was made lX
in assay buffer and four dilutions were made using lX
assay buffer. Fifteen ~l of each ribozyme dilution (i.e., 20, 16, 12, 8 and 4 nM) was placed in a separate tube.
These tubes and the substrate tube were pre-incubated at 37C for at least five minutes.
The reaction was started by mixing lS ~l of substrate into each ribozyme tube by rapid pipetting (note that final ribozyme concentrations were 10, 8, 6, 4, 2 nM). Five ~l aliquots were removed at 15 or 30 second intervals and quenched with 5 ~l stop solution (95%
formamide, 20 mM EDTA xylene cyanol, and bromphenol blue dyes~. Following the final ribozyme time point, an aliquot of the remaining substrate was removed as a zero ribozyme control.
The samples were separated on either 15% or 20%
polyacrylamide gels. Each gel was visualized and ~093~3057 PCT/US93/0457 quantitated with an Ambis beta scanner (Ambis Systems, San Diego, CA).
For the most active ribo~ymes, kinetic analyses were performed in substrate excess to determine ~ and Kcat values.
For kinetic reactions with long RNA substrates (greater than 15 bases in length) the substrates were preparéd by transcription using T7 RNA polymerase and defined templates containing a T7 promoter, and DNA
encoding appropriate nucleotides of the target RNA. ~he substrate solution was made lX in assay buffer ~7S mM
Tris-HCl, pH 7.6; 0.1 mM EDTA; 10 mM MgCl2) and contained 58 nanomolar concentration of the long RNA molecules. The reaction was started by addition of gel purified ribozymes 15 to 1 ~M concentration. Aliquots were removed at 20, 40, ; ~ 60, 80 and 100 minutes, then quenched by the addition of 5 ~1 stop solution. Cleavage products were separated using~denaturing PAGE. The bands were visualized and quant~1tat~ed with~an Ambis beta scanner.
Kinetic Analysis ;~ A simple reaction mechanism for ribozyme-mediated cleavage is:
ki k~
- ~ ~
~ R~ S ~ [R:S] ~ [R:P] I ~ R + P
- k l where R = ribozyme, S = substrate, and P = products. The boxed step is important only in substrate excess. Beca~se ribozyme concentration, is~ in excess over substrate concentration, the concentration of the ribozyme-substrate complex (lR:S]) is constant ov-~r time except during the very brief time when the complex is being initially ,.
~; formed, i.e.,:
-~ 35 d~R:Sl = 0 -~ dt where t = time, and thus:
~- (R)(S)~' = (RS~(k~ + k,).
~,'', ' , c~
W(~ 93/230 ~7 ~ I v PCr/US93/0457 r 5 26 `
The rate of the reaction is the rate of disappearance of substrate with time:
Rate = (t) = k2(RS) ., .
Substituting these expressions:
(X)(S)k~ = 1/k2 -d(S) ~k2 + k,) dt or:
-d(S) = _ k,k._ (R) dt S (k2 ~ k,) Integrating this expression wlth respect to time yields:
-ln S = k1k~ ~R) t So ( k2 ~ kl ) where S0 = inltial substrate. Therefore, a plot of the ;~
negative log of fraction substrate uncut versus time (in minutes) yields a straight line with slope:
, slope = k1k. (R) = kObs ~k2 + k,) where kob5 = observed rate constant. A plot of slope (kob5) 20 versus ribozyme concentration yields a straight line with a slope which is:
slope = k.k which is k (k7 ~ k1) ~

Using these equations the data obtained from the , 25 kinetic experiments provides the necessary information to determine which ribozyme tested is most useful, or active.
Such ribozymes can be selected and tested in in vivo or èx vivo systéms. ~ ! `! ' `;. ! ' :
Liposome Preparation Lipid molecules are dissolved in a volatile organic solvent (CHCl3, methanol, diethylether, ethanol, etc.). The organic solvent is removed by evaporation.
The lipid is hydrated into suspension with O.lx phosphate buffered saline (PBS), then freeze-thawed 3x using liquid nitrogen and incubation at room temperature. The ~;~

..~9~t23057 ~ 1~ 3 4 9 9 PC~tUS93tO4~7~
,;.

suspension is extruded sequentially through a 0.4 ~m, O.2 ~m and 0.1 ~m polycarbonate filters at maximum pressure of 800 psi. The ribozyme is mixed with the .-extruded liposome suspension and lyophilized to dry~^ess. ~-The lipid/ribozyme powder is rehydrated with water to one-tenth the original volume. The suspension is diluted to the minimum volume required for extrusion (0.4 ml for 1.5 ml barrel and 1.5 ml for 10 ml barrel) with lxPBS and re-extruded through 0.4 ~m, 0.2 ~m, 0.1 ~m polycarbonate filters. The liposome entrapped ribozyme is separated from untrapped ribozyme by gel filtration chromatography (SEPHAROSE CL-4B, BIOGEL A5M). The liposome extractions are pooled and sterilized by filtration through a 0.2 ~m filter. The free ribozyme is pooled and recovered by ethanol prec}pitation. The liposome concentration is determined by incorporation of a radioactive lipid. The ribozyme concentration is determined by labeling with 32p, Rossi et al., 1992, supra (and references cited therein) ; describe other methods suitable for preparation of liposomes.
Examples of other useful liposome preparations which display similar degrees of upt~ke of both a radioactive lipid marker and an entrapped fluorophore by Vero cells showed different fluorescent staining patterns.
- 25 Specifically, liposomes composed of DPPG/DPPC/Cholesterol ~- (in a ratio of: 50/17/33) gave a punctate pattern of fluorescence, while DOPE/Egg PC/Cholesterol (30/37/33) gave a diffuse, homogeneous pattern of fluorescence in the cytoplasm. Cell fractionation showed that 80% of the entrapped contents from the DPPG/DPPC/Cholesterol formulation was localized in the membrane fraction, whereas the DOPE/Egg PC/Cholesterol formulation was localized in the cytoplasm. Further characterization of the latter formulation showed that after 3 hours, 70% of the fluorescence was cytoplasmic and 30% was in the membrane. After 24 hours, uptake had increased 5-fold and W~l93/23057 ~ 3~4 PCI/IS93/0457'~

the liposome contents were distributed 50/50 between the cytoplasmic and membrane fractions.
Liposomes containing 15 ribozymes (32P-labeled) targeted to the HSV ICP4 mRNA were prepared and incubated with the cells. After 24 hours, 25% of the liposome dose was taken up with approximately 60,000 liposomes per cell.
Thirty percent of the delivered ribozyme was intact after 24 hours. Cell fractionation studies showed 40% of the intact ribozyme to be in the membrane fraction and 52% of the intact ribozyme to be in the cytopIasmic fraction.
In Vi vo Assav The efficacy of action of a chosen ribozyme may be tested in vivo by use of cell cultures sensitive to mdr-1 gene expression, using standard procedures in transformed cells or animals which express the target mRNA
- using standard procedures.
The efficacy of action of a chosen ribozyme may be tested in t~ssue culture by use of transformed cells containing the target mRNA (e.q., K562 cells which express ~;~ 20 the b3;a2~ ~fusion mRNA) using standard procedures.
Alternatively, ribozyme efficacy could be tested with peripheral blood or bone marrow from CML patients using soft-agar colony forming assays. Such methods are known -to those educated in this field.
Ribonuclease Protection AssaY
~ The accumulatlon of target-mRNA in cells or the - cleavage of the mRNA by ribozymes or RNaseH (in vitro or in vivo) can be quantified using an RNase protection assay.
In this method, antisense riboprobes are transcribed from template DNA using T7 RNA polymerase (U.S. Biochemical) in 20 ~l reactions containing lX
transcription buffer (supplied by the manufacturer), 0.2 mM ATP, GTP and UTP, 1 U/~l pancreatic RNase inhibitor 35 (Boehringer Mannheim 8iochemicals) and 200 ~Ci 3~P-labeled CTP (800 Ci/mmol, New England Nuclear) for 1 hour at 37C.
Template DNA is digested with 1 U RNase-free DNaseI (U.S.

.. 093/23057 ~1 3 ~ PCT/US93/0457~ ~
'~:

29 ~`
Biochemical, Cleveland, OH) at 37C for 15 minutes and unincorporated nucleotides removed by G-50 SEPHADEX spin chromatography.
In a manner similar to the transcription of antisense probe, the target mRNA can be transcribed in vi tro using a suitable DNA template. The transcript is purified by standard methods and digested with ribozyme at 37C according to methods described later.
Alternatively, afflicted ~mRNA-expressing~ cells expressing the target mRNA bcr-abl fusion transcript are harvested into 1 ml of PBS, transferred to a 1.5 ml EPPENDORF tube, pelleted for 30 seconds at low speed in a microcentrifuge, and lysed in 70 ~l of hybridization buffer (4 M guanidine isothiocyanate, 0.1% sarcosyl, 25 mM
sodium citrate, pH 7.5). Cell lysate (45 ~l) or defined amounts of in vitro transcript (also in hybridization buffer) is then combined with 5 ~l of hybridization buffer containing 5 x 105 cpm of each antisense riboprobe in 0.5 ml EPPENDORF tubes, overlaid with 25 ~l mineral oil, and hybridization accomplished by heating overnight at 55C.
The hybridization reactions are diluted into 0.5 ml RNase solution (20 U/ml RNaseA, 2 U/ml RNaseTl, 10 ~/ml RNase-free DNaseI in 0.4 M NaCl), heated for 30 minutes at 37C, and 10 ~l of 20% SDS and 10 ~l of Proteinase K (10 mg/ml) added, followed by an additional 30 minutes incubation at : 37C. Hybrids are partially purified by extraction with 0.5 ml of a 1:1 mixture of phenol/chloroform; aqueous ;.
phases are combined with 0.5 ml isopropanol, and RNase- :
resistant hybrids pelleted for 10 minutes at room temperature (about 20C) in a microcentrifuge. Pellets are dissolved in 10 ~l loading buffer (95% formamide, lX :
TBE, 0.1% bromophenol blue, 0.1% xylene cylanol), heated to 95C for five minutes, cooled on ice, and analyzed on

4~ polyacrylamide/7 M urea gels under denaturing :
conditions.

WOg3/230~7 ~13 ~ ~ 9 ~ PCTtUS93/045 Ribozvme Stabilitv The chosen ribozyme can be tested to determine its stability, and thus its potential utility. Sush a test can also be used to determine the effect of various chemical modifications (e.q., addition of a poly(A) tail) on the ribozyme stability and thus aid selection of a more stable ribozyme. For example, a reaction mixture contains 1 to 5 pmoles of 5~ (kinased) and/or 3' labeled ribozyme, 15 ~g of cytosolic extract and 2.5 mM MgCl2 in a total volume of 100 ~1. The reaction is incubated at 37C.
Eight ~1 aliquots are taken at timed intervals and mixed - wlth 8 ~l of a stop mix (20 mM EDTA, 95% formamide).
Samples are separated on a 15% acrylamide sequencing gel, exposed to film, and scanned with an Ambis.
A 3'-labeled ribozyme can be formed by -~ incorporation of the 32P-labeled cordycepin at the 3~ OH
.
- using poly(A) polymerase. For example, the poly(A) polymerase reaction contains 40 mM Tris, pH 8, 10 mM MgCl2, 25~0 mM NaCl, 2.5 mM MnCl2; 3 ~l 32p cordycepin, 500 Ci/mM;
20 ~and 6 unlts poly(A) polymerase in a total volume of 50 ~1.
The reaction mlxture is incubated for 30 minutes at 37C.
Effect of Base Substitution u~on Ribozvme ~ctivity To determine which primary structural characteristics could change ribozyme cleavage of substrate, minor base changes can be made in the substrate cleavage region recognized by a specific ribozyme. For example, the substrate sequences can be changed at the central~nC" nucleotide, changing the cleavage site from a GUC to a GUA motif. The KCat/~ values for cleavage using each substrate are then analyzed to determine if such a `~- change increases ribozyme cleavage rates. Similar M ~ experiments can be performed to address the effects of changing bases complementary to the ribozyme binding arms.
Changes ~predicted to maintain strong binding to the complementary substrate are preferred. Minor changes in nucleotide content can alter ribozyme/substrate ~093/230~7 ` ~ 5 (~ !i 3 PCT/VS93/04573 ~i (.

interactions in ways which are unpredictable based upon binding strength alone. Structures in the catalytic core region of the ribozyme recognize trivial changes in either substrate structure or the three dimensional structure of the ribozyme/substrate complex.
To begin optimizing ribozyme design, the cleavage rates of ribozymes containing varied arm lengths, but targeted to the same length of short RNA substrate can be tested. Minimal arm lengths are required and effective cleavage varies with ribozyme/substrate combinations.
The cleavage activity of selected ribozymes can be assessed using target mRNA-homologous substrates. The assays are performed in ribozyme excess and approximate Kcae/Kmin values obtained. Comparison of values obtained with short and long substrates indicates utility in ~-ivo of a ribozyme.
Intracellular Stabilitv of LiPosome-Delivered ibozvmes , To test the stability of a chosen ribozyme - 20 :in vivo the following test is useful. Ribozymes are 32p_ end-labeled, entrapped in liposomes and delivered to ; target mRNA containing cells for three hours. The cells are ~fractionated and ribozyme is purified by phe~ol~chloroform extraction. Alternatively, cells ( lX107, T-I75 flask) are scraped from the surface of the flask are ~- cultured and washed twice wlth cold PBS. The cells are . .
homogenized by douncing 35 times in 4 ml of TSE (10 mM
Tris, pH 7.4, O.25 M Sucrose, mM EDTA) . Nuclei are pelleted at lOOxglfo~ 10 minutes. Subcellular organelles (the m~ ~ane fraction) are pelleted~at 200,000xg for two - hours ; ng an SW60 rotor. The pellet is resuspended in 1 ml of H buffer (0.25 M Sucrose, 50 mM HEPES, pH 7.4).
The supernatant contains the cytoplasmic fraction (in ; approximately 3 . 7 ml ) . The nuclear pellet is resuspended in 1 ml of 65% sucrose in TM (50 mM Tris, pH 7.4, 2.5 mM
MgCl2) and banded on a sucrose step gradient (1 ml nuclei ~`~ in 65% sucrose TM, 1 ml 60% sucrose TM, 1 ml 55% sucrose W09~/230~7 ' ~ 3 ~ 4 3 ~ PCT/US93/~4~7s. ` ~

.

TM, 50% sucrose TM, 300 ~l 25% sucrose TM) for one hour at 37,000xg with an SW60 rotor. The nuclear band is harvested and diluted to 10% sucrose with TM bu~fer.
Nuclei are pelleted at 37,000xg using an SW60 rotor for lS
minutes and the pellet resuspended in 1 ml of TM buffer.
Aliquots are size fractionated on denaturing polyacrylamide gels and the intracellular localization determined. By comparison to the migration rate of newly syntnesized ribozyme, the various fractions containing 10 intact ribozyme can be determined. ' To investigate modifications which would lengthen the half-life of ribozyme molecules intracellularly, the cells may be fractioned as above and the purity~ of each fraction assessed by assaying enzyme ' lS activity known to exist in that fraction. ' ~;
The various cell fractions are frozen at -70C
and used to determine relative nuclease resistances of modified ribozyme molecules. Ribozyme molecules may be synthesized wlth 5 phosphorothioate (ps), or 2~-O-methyl (2~-OMe~modifications at each end of the molecule. These ' molecules and a phosphodiester version of the ribozyme are end-labeled with 32p and ATP using T4 polynucleotide kinase. Equal concentrations are added to the cell i-cytoplasmlc extracts and aliquots of each taken at 10 25 minute intervals. The samples are size fractionated by `~
denaturing PAGE and relative rates of nuclease resistance -analyzed;by scanning the gel with an Ambis ~-scanner. The results show whether the ribozymes are digested by the 'cytoplasmic extrac,t,land which versions are relativelyi more nuclease resistant. Modified ribozymes generally maintain 80-90% of the catalytic activity of the native ribozyme when short RNA substrates are employed.
Unlabeled, 5' end-labeled or 3' end-labeled ribozymes can be used in the assays. These experiments can also be performed with human cell extracts to verify the observations.
`'~:

.~093/23057 ~ 1 7~ r~ ~ PCT/US93/0457 In one example, Vero or HeLa cells were grown to 90-95% confluency in 175 cm2 tissue culture flasks, scraped into 10 ml of cold phosphate buffered saline tPBS), then washed once in 10 ml of cold PBS and once in 10 ml of cold TSE (10 mM Tris, pH 7.4; 0.25 M sucrose; 1 mM EDTA). The cell pellets were resuspended in 4 ml of TSE, dounced 35x on ice, and the released nuclei pelleted by centrifugation at lOOOg for 10 minutes. The nuclear pellet was resuspended in 1 ml of 65% sucrose TM (50 mM Tris, pH 7.4;
2.5 mM MgCl2) and transferred to seckman ultra-clear tubes.
The following sucrose TM solutions were layered on top of the sample: 1 ml 60%, 1 ml 55~, and 25% sucrose to the top of the tube. Gradients were spun in an SW60 rotor at 37,0nOg for 1 hour. HeLa nuclei banded at the 55-60%
sucrose boundary and Vero nuclei banded at the 50-55%
sucrose boundary. Nuclear bands were harvested, diluted to 10% sucrose with TM buffer, and pelleted by centrifugation at 37,000g for 15 minutes using an SW60 -~ rotor. The nuclear pellet was resuspended in 1 ml of TM
buffer. Subcellular organelles and membrane components in the post nuclear supernatant were separated from the cytoplasmic fraction by centrifugation at 200,000g for 2 hours in an SW60 rotor. The pellet contained the membrane fraction,~ which was resuspended in 1 ml of H buffer (0.25 M sucrose; 50 mM HEPES, pH 7.4), and the supernatant contained the cytoplasmic fraction.
Purity of the various fractions was assessed using enæymatic markers specific for the cytoplasmic and - ~membranous fractions. Three enzyme markers for the membranous fraction were used; hexosaminidase and ~-glucocerebrosidase are localized in lysosomes, while alkaline phosphodiesterase is specific to endosomes.
~- Specifically, the assays were as follows:
~ ,, -; For N-acetyl-beta-hexosaminidase, the reaction mixture contained 0.3 mg/ml 4-methylumbelliferyl-N-acetyl-glucosaminide; 20 mM sodium citrate; pH 4.5; 0.01% Triton -- X-100; and 100 ~l of sample in a final volume of 500 ~l W093/~30~7 ~ 4 ~ ~ PCT/US93/04 (Harding et al., 6~ Cell 393, 1991). The reactions were incubated at 37~C for 1 hour and stopped by the addition of 1.5 ml of stop buffer (0.13 M glycine, 0.07 M NaCl, 0.08 M sodium carbonate, pH 10.6). The reaction product was quantitated in a Hitachi F-4010 fluorescence spectrophotometer by excitation of the fluorophore at 360 nm and analysis of the emission at 448 nm.
For Alkaline Phosphodiesterase, the assay medium contained 25 mM CAPS (3-(Cyclohexylamino)-propanesulfonic 10 acid), pH 10.6; 0.05% Triton X-100; 15 mM MgCl2; 1.25 mg/ml Thymidine-5~-monophosphate-p~nitrophenyl ester; and 100 ~l o, sample in a total reaction volume of 200 ~l. The reactions were incubated at 37C for 2 hours, then diluted to 1 ml with H2O and the absorbance was measured at 400 nm 15 (Razell and Khorana, 234 J. siol. Chem. 739, 1959).
For ~-glucocerebrosidase, the reaction contained 85 nM sodium citrate, pH 5.9; 0.12% Txiton X-100; 0.1%
sodium taurocholate; 5 mM 4-methylumbelliferyl ~-D-glucopyranoside; and 125 ~l of sample in a total volume of - - 20 250 ~l (Kennedy and Cooper, 252 Biochem. J. 739, 198&).
The reaction was incubated at 37C for 1 hour and stopped by the addition of 0.75 ml of stop buffer. Product formation was measured in a fluorescence spectrophotometer ~- ~y using an excitation wavelength of 360 nm and analysis of the emission at 448 nm.
The cytoplasmic enzyme marker, lactate dehydrogenase, was assayed in an assay mixture containing O.2 M Tris; pH 7.4; 0.22 mM NADH; 1 mM sodium pyruvate;
and 50 ~l of sample in a final volume of 1.05 ml. Enzyme "
levels were determined by decreased absorbency at 350 nm resulting from the oxidation of NADH at room temperature ~- (Silverstein and Boyer, 239 J. Biol. Chem. 3901, 1964).
Lactate dehydrogenase was found predominantly in the cytoplasmic fractions of both Vero and HeLa cells, while ~-glucocerebrosidase and alkaline phosphodiesterase were found almost exclusively in the membranous fractions.
The hexosaminidase activity in Vero cell fractions was ~93/230~7 ~; 3'i~ PCT/US93J0457 concentrated in the membranous fraction (70%) with about 20% in the cytoplasmic fraction. The isolation of enzyme markers with the appropriate cellular compartment demonstrated that cytoplasmic, membranous and nuciear fractions can be isolated with minimal intercompartmental contamination using this fractionation scheme.
Nuclease StabilitY of Ribozymes and mRNA
The simplest and most sensitive way to monitor nuclease activity in cell fractions is to use end-labeled oligonucleotides. However, high levels of phosphatase activity in some biological extracts gives ambiguous results in nuclease experiments when 3-P-5~-end-labeled oligonucleotides are used as substrates. To determine the phosphatase activity in the extracts, cellular fractions were incubated with cold ribozymes and trace amounts of

5'-end-labeled ribozyme in the presence of 1 mM Mgt2 (or - Znt2 with HeLa cytoplasmic extracts) to optimize digestion.
After polyacrylamide gel electrophoresis of samples, ~- digestion~of the oligonucleotide was assessed both by staining an~d by autoradiography.
Specifically, the basic oligonucleotide digestion reaction contained substrate nucleic acid (an RNA ollgonucleotide of 36 nucleotides) and cell fraction ~s~ extract in a total volume of 100 ~ liquots ~7 ~l) were taken after various periods of incubation at 37C and added ~to 7 ~l of gel loading buffer (95~ formamide, 0.1 bromophenol blue, 0.1% xylene cyanol, and 20 mM EDTA).
The samples were separated by electrophoresis on a 7 M
urea,~ 20%lj polyaçrylamid~e ,gçl. Intact ribozymes were visualized either by staining with Stains-all (United States Biochemical, Cleveland, OH), or autoradiography of 32P-labeled ribozyme. The stained gels and X-ray films were scanned on a Bio 5000 density scanner (U.S.
Biochemical). Ribozymes were 5' end-labeled with T4 polynucleotide kinase (U.S. Biochemical) using 10 ~Ci of ~- 32p ~-ATP (3,000 Ci/mmole, New England Nuclear, Boston, MA), and 20-25 pmoles of ribozymes. The unincorporated ' W093/230~7 ~ 3~ ~ 9 ~ PCT/US93/0457 .

nucleotides were separated from the product by G-50 spin chromatography. Nuclease assays contained 1-2 pmoles of 32p-labeled ribozyme. All oligonucleotides were synthesized on an Applied Biosystems 394 DNA/RNA
synthesizer (Applied Biosystems Inc., Foster City, CA) according to manufacturer~ 5 protocols. The nuclear fractions were resuspended in a buffer containing 2.5 mM
MgCl2. Experiments involving the nuclear fractions were performed in the presence of 1 mM M~, or in combination with 1 mM Mn'2, Ca~2, or Zn+2.
To measure the stability of mRNA, Vero cells were infected with herpes simplex virus (HSV) at a M.O.I.
of S and total RNA was extracted (Chomczynski and Sacchi, 162 Anal. Biochem. 156, 1987). An RNase protection assay was used to detect mRNA after incubation of total infected cellular RNA in cytoplasmic extracts. RNA probes were produced from PCR-amplified template DNA uslng T7 RNA
polymerase (U.S. Biochemical) in the presence of 32p ~-CTP
(3,000;~ C1~/mmole, New England Nuclear, Boston, MA).
~ ~ 20 TempLate~DNA was~ inactivated with 1 unit of RNase-free ;-~ DNaseI for 15 minutes at 37C. Unincorporated nucleotides were removed by G-50 spin chromatography. Samples (6 ~
were taken from the nuclease assays after various periods ;~ ~of incubation at 37C, added to 40 ~l of 4 M GUSCN buffer 2S (4 M guanidinium thiocyanate; 25 mM sodium citrate, pH 7;
~-; 0.5% sarcosyl; and 0.1 M 2-mercaptoethanol), and 5 ~1 of 32P-labeled RNA probe (5xlOs cpm/5 ~l, specific activity of 1.8x106 cpm/~g) in 4 M GUSCN buffer. Hybridization reactions were~coveredl with~mineral~oil and lncubated at 55C for 12-16 hours, after which the hybridization - reaction was mixed with 500 ~l of RNase buffer (0.4 M
NaCl, 20 ~g/ml RNaseA, 2 units~ml T1 RNase) and incubated for 30 minutes at 37C. RNase activity was quenched by incubation with 10 ~l of 20% SDS and 10 ~l of proteinase K (20 mg/ml), and the RNA was extracted using a phenol/chloroform mixture. The protected RNA fragment was purified by precipitation with an equal volume of -~093/23057 ~ 3; ~ ~ ~ PCT/US93/04573 isopropanol in the presence of 20 ~g of carrier yeast tRNA. The RNA pellets were resuspended in gel loading buffer, heated to 95C for 5 minutes and separated by electrophoresis on a 5% polyacrylamide, 7 M urea gel.
Protected fragments were visualized by autoradiography, and the films were scanned with a sio 5000 density scanner.
In experiments using these methods, the rate of digestion of ribozymes in Vero cell extracts was similar, demonstrating the lack of significant phosphatase activity in Vero cellular fractions. Similar results were observed with HeLa cellular fractions. In most extracts, ladders of digested fragments were observed; such ladders would not be expected if digestion was an artifact of phosphatase action. Thus, digestion using 5~ end-labeled ribozymes is an accurate assessment of nuclease action in ce~llular extracts.
In other experiments, labeled ribozymes were ..~
incubated in various Vero and HeLa cellular fractions.
Incubation of ribozymes in either membranous or nuclear ~ fractions resulted in a linear decrease of intact -~ mo-lecules over time. In contrast, no digestion of ~; rlbozymes occurred durin~ a 24 hour incubation in Vero cytoplasmic extracts, and HeLa cytoplasmic extracts - 2~ exhibited a 20-30 minute delay in the onset of RNA
digestion. After this refractory period, the rate of digestion was linear but not as rapid as the rates observed in any of the nuclear or membranous fractions.
~ ,The effect jofif,our dlvalent cations (Mg~2, Mn~2, Ca~2, and Zn~2) on the nuclease activity of the cellular fractions was assessed. Vero cytoplasmic extracts were ~ stimulated by the addition of 1 mM Mg'2 or Mn~2, while Ca~--~ or Zn~2 had no effect. Nuclease activity in HeLa cytoplasmic extracts was enhanced only by the addition of 1 mM Zn~2. Both Vero and HeLa membranous fractions exhibited maximum nuclease activity with the addition of Mgt~ or Mn~- ions, while the addition of Ca~- significantly ~ .

:' .

WO93/230~;7 lc ~ PCl~/US93/1)4s7 ~`

reduced activity of the HeLa membranous fraction and abolished nuclease activity in the Vero membranous fraction. Addition of Zn~2 to both membranous fractions resulted in a loss of all RNase activity. The Vero nuclear extract demonstrated roughly equivalent nuclease activity in the presence of either Mg~- alone or a Mgt2 and Mn~2 ion combination, less in the presence of Mg2 and Cat2, and no activity in the presence of Mgt2 and Zn~2. The effects of cation addition were not as dramatic with HeLa - 10 nuclear extracts. The nuclease activity of these fractions was greatest in the presence of Mg~2 alone or Mg~
and Ca~- and decreased slightly with the addition of Mn-- or Zn~2 to the Mgt~ present in the extracts.
To verify that nuclease activity was dependent ;
lS upon added divalent cations, nuclease assays were performed uslng 1 mM Mg~2 in the presence and absence of 20 mM EDTA. For the HeLa cytoplasmic fractions,-the Mg~2 was r~eplaced with 1 mM Zn'2. The presence of 20 mM EDTA
; c~omplet~e1y;~abolished nuclease activity in the Vero and 20~HeLa cytoplasmic fractions and Vero nuclear fractions.
~-1 Nuclease acti~ity in the HeLa membranous and nuclear fractions was partially inhibited by the addition of EDTA, ~ ~ while EDTA had no effect on the nuclease activity in the -~ - Vero membranous fraction. For comparative purposes, f~ 25~ reactlons using DNA oligonucleotides were performed using different Vero fractions. All DNase activity in Vero ~;~ cytoplasmic, membranous, and nuclear fractions was inhibited by 20 mM EDTA.
The stability of,RNA oligQnucleotides and HSV-1 mRNA were compared in the presence and absence o activity-enhancing divalent cations (1 mM Mgt2, Vero cells;
-~ ~ 1 mM Zn~2, HeLa cells). Total cellular RNA from HSV-1 , ~1 infected Vero cells (B mg) and tracer amounts of 32p_5'_ I
;~ end-labeled RNA oligonucleotides (1 pmole) were incubated with Vero or HeLa cytoplasmic extracts. In the absence of divalent cations, no substantial decrease of intact ribozymes was detected in assays, although mRNA was ,:-~r093/23057 ~1 3X l.'1.'3 PCrlUS93/04~73 digested in both vero and HeLa cytoplasmic extracts.After addition of divalent cations, digestion of ribozymes occurred in both vero and HeLa cytoplasmic fractions. The rate of ribozyme digestion in HeLa extracts increased to levels similar to those observed with mRNA, while the rate of mRNA digestion remained greater than the rate of ribozyme digestion in Vero cytoplasmic fractions.
Thus, the stability of hammerhead ribozymes were compared in both Vero and HeLa cell cytoplasmic, membranous and nuclear fractions. Vero cytoplasmic and nuclear fractions were found to require Mgt2 for optimal nuclease activity, while the membranous fraction was no, altered by the addition of divalent cations. HeLa membranous and nuclear fractions were also activated by Mg~2, while the cytoplasmic fractions required Znt2 for nuclease activation. Relative stabilities of ribozymes and mRNAs were compared in Vero and HeLa cytoplasmic fractions. In the absence of appropriate divalent cations, little ribozyme digestion was observed in either cytoplasmic preparation while mRNA was rapidly digested.
The addition of Mg~2 to Vero cytoplasmic extracts and Zn~
to the HeLa cytoplasmic extracts stimulated ribozyme - ~ ~ degradation and enhanced mRNA digestion. These data show that the nuclease sensitivity of ribozymes is cell-type specific, varies with the intracellular compartment ; studied and may not be able to be predicted from studies ; with mRNA. Notably, however, ribozymes appear stable in such cellular fractions for a period of time po~entially sufficient to haveja ther~peutically useful activity.
Administration of Rlboz~me Selected ribozymes can be administered prophylactically, or to patients expressing mdr-1 mRNA, or ~ ha~ing CM1, leukemic conditions, Burkitt's lymphoma, -~ follicular lymphoma, breast cancer, colon carcinoma, ~- 35 neuroblastoma, lung cancer, or pretumor cells, e.q., b~-exogenous deli~ery of the ribozyme to an a desired tissue by means of an appropriate delivery vehicle, e.a., a W093/~30s7 ~ PCT/US93/0457, liposome, a controlled release vehicle, by use of iontophoresis, electroporation or ion paired molecules, or covalently attached adducts, and other pharmacologically approved methods of delivery. Routes of administration include intramuscular, aerosol, oral (tablet or pill form), topical, systemic, ocular, intraperitoneal and/or intrathecal. Alternatively, ribozymes may be administered to a tissue or afflicted cell ex vivo to eradicate tumorigenic cells prior to re-implantation (e.a., in the course of autologous bone marrow transplantation therapy).
Expression vectors for immunization with ribozymes and/or delivery of ribozymes are also suitable.
The specific delivery route of any selected ribozymie will depend on the use of the ribozyme.
Generally, a specific delivery program for each ribozyme ; will focus on unmodified ribozyme uptake with regard to intracellular localization, followed by demonstration of efficacy. ~Alternatively, delivery to these same cells in an organ or tissue of an animal can be pursued. Uptake ~stud1es will-include uptake assays to evaluate cellular ribozyme uptake, regardless of the delivery vehicle or strategy. Such assays will also determine the intracellular localization o`f the ribozyme following uptake, ultimately establishing the requirements for malntenance of steady-state concentrations within the cellular ~ompartment containing the target sequence - (nucleus and/or cytoplasm). Efficacy and cytotoxicity can then be tested. Toxicity will not only include cell viability but also cell function.
" ~
Some methods of delivery that may be used include:
a. encapsulation in liposomes, b. transduction by retroviral vectors, c. conjugation with cholesterol, d. localization to nuclear compartment utilizing nuclear targeting site found on most nuclear proteins, v~O 93/23057 ~ l 3 ~ PCT/US93/04573 e. neutralization of charge of ribozyme by using nucleotide derivatives, f. use of blood stem cells to distribute ribozymes throughout the body, and g. electroporation~
At least three types of delivery strategies are useful in the present invention, including- ribozyme modifications, particle carrier drug delivery vehicles, and retroviral expression vectors. Unmodified ribozvmes, 10 like most small molecules, are taken up by cells, albeit ;
slowly. To enhance cellular uptake, the ribozyme may be modified essentially at random, in ways which reduce its charge but maintains specific functional groups. This results in a molecule which is able to diffuse across the cell membrane, thus removing the permeability barrier.
Modification of ribozymes to reduce charge is just one approach to enhance the cellular uptake of these larger~molecules. The random approach, however, is not ~- ~ advisable since ribozymes are structurally and ~functionally more complex than small drug molecules. The structural requirements necessary to maintain ribozyme catalytic activity are well understood by those in the - art. These requirements are taken into consideration when ~;~ designing modifications to enhance celIular delivery. The - ~ 25 modifications are also designed to reduce susceptibility to nuclease degradation. Both of these characteristics should greatly improve the efficacy of the ribozyme.
~ellular uptake can be increased by several orders of ~magnitude without having to alter the phosphodiester linkages necessary for ribozyme cleavage activity.
Chemical modifications of the phosphate b~ckbone ;~ will reduce the negative charge allowing free diffusion across the membrane. This principle has been successfully demonstrated for antisense DNA technology. The similarities in chemical composition between DNA and RNA
make this a feasible approach. In the body, maintenance of an external concentration will be necessary to drive W093/23057 ~1 3~ ~ ~r! ~; PCT/US93/045~
e: cJ ~

the diffusion of the modified ribo~yme into the cells of the tissue. Administration routes which allow the diseased tissue to be exposed to a transient high concentration of the drug, which is slowly dissipated by systemic adsorption are preferred. Intravenous administration with a drug carrier designed to increase the circulation half-life of the ribozyme can be used.
The size and composition of the drug carrier restricts rapid clearance from the blood stream. The carrier, made to accumulate at the site of infection, can protect the -ribozyme from degradative processes.
Drug delivery vehicles are effective for both systemic and topical administration. They can be designed to serve as a slow release reservoir, or to deliver their contents directly to the target cell. An advantage of using ~direct delivery drug vehicles is that multiple molecules are delivered per uptake. Such vehicles have beén~shown to increase the circulation half-life of drugs whlch~would~o~therwise~be rapidly cleared from the blood stream~ Some~examples of such specialized drug delivery vehicles ~whlch fall into this category ar~e liposomes, hyd~rogels, cyclodextrins, biodegradable nanocapsules, and bloadhesive microspheres.
From this category of delivery systems, liposomes are preferred. Liposomes increase intracellular ;stabil~ity, increase uptake efficiency and improve biological activity.
Liposomes are hollow spherical vesicles composed of lipids arranged in a similar fashion as those lipids which make up the cell membrane. They have an internal , - :
aqueous space for entrapping water soluble compounds and -~ range in size from 0.05 to several microns in diameter.
~-~ Several studies have shown that liposomes can deliver RNA
to cells and that the RNA remains biologically active.
35For example, a liposome delivery vehicle ;; originally designed as a research tool, Lipofectin, has ~ - 3 ' ~ ~
.j~' ~',' .
~,.''~,:

`~093/23057 ;~ PCT/US93/04573 ! ~

been shown to deliver intact mRNA molecules to cells yielding production of the corresponding protein.
Liposomes offer several advantages: They are non-toxic and biodegradable in composition, they display long circulation hal~-lives; and recognition molecules can be readily attached to their surface for targeting to tissues. Finally, cost effective manufacture of liposome-based pharmaceuticals, either in a liquid suspension or lyophilized product, has demonstrated the viability of this technology as an acceptable drug delivery system.
Other controlled release drug delivery systems, such as nonoparticles and hydrogels may be potential delivery vehicles for a ribozyme. These carriers have been developed for chemotherapeutic agents and protein-based pharmaceuticals, and consequently, can be adaptedfor ribozyme delivery.
;~ ~ Topical adm1nistration of ribozymes is ad~antageous since it allows localized concentration at ` the slte of administration with minimal systemic adsorption.~ This simplifies the delivery strategy of the ribozyme~to the disease site and reduces the extent of toxicological characterization. Furthermore, the amount of mat~erlal to be applied is far less than that required - for other administration routes. Effective delivery 25 ~requires the ribozyme to diffuse into the infected cells -or through the skin to the underlying vasculature.
Chemical modification of the ribozyme to neutralize negative charge may be all that is required for penetration. However, in the event that charge neutralization is insufficient, the modified'ribozyme can be co-formulated with permeability enhancers, such as Azone or oleic acid, in a liposome. The liposomes can .,, ~, ,.
either represent a slow release presentation vehicle in which the modified ribozyme and permeability enhancer transfer from the liposome into the infected cell, or the liposome phospholipids can participate directly with the modified ribozyme and permeability enhancer in ~' ~

, - .

facilitating cellular delivery. In some cases, both the ribozyme and permeability enhancer can be formulated into a suppository formulation for slow release.
Ribozymes may also be systemically administered.
Systemic absorption refers to the accumulation of drugs in the blood stream followed by distribution throughout the entire body. Administration routes which lead to systemic absorption include: intxavenous, subcutaneous, intraperitoneal, intranasal, intrathecal and ophthalmic.
Each of these administration routes expose the ribozyme to an accessible diseased tissue. Subcutaneous administration drains into a localized lymph node which proceeds through the lymphatic network into the circulation. The rate of entry into the circulation has been shown to be a function of molecular weight or size.
The use of a liposome or other drug carrier localizes the ribozyme at the lymph node. The ribozyme can be modified to diffuse into the cell, or the liposome can directly participate in the delivery of either the unmodified or modified ribozyme to the cell.
A liposome formulation containing phosphatidyl-ethanolomidomethylthiosuccinimide which can deliver oligonucleotides to lymphocytes and macrophages is also useful for certain cancerous conditions. Furthermore, a 200 ~m diameter liposome of this composition was internalized as well as 100 nm diameter liposomes. The 200 nm liposomes exhibit a 10-fold greater packaging capacity than the 100 nm liposomes and can accomodate larger molecules such as a ribozyme expression vector.
This oligonucleotide delivery system inhibits viral proliferation in these viruses that infect primary immune cells. This oligonucleotide delivery system prevents mRNA
expression in affected primary immune cells. Whole blood studies show that the formulation is taken up by 90% of 35 the lymphocytes after 8 hours at 37C. Preliminary biodistribution and pharmacokinetic studies yielded 70~ of ~093J23057 PCT/US93/04573 ~

, the injected dose/gm of tissue in the spleen after 1 hour following intravenous administration.
Intraperitoneal administration also leads to entry into the circulation with the molecular weight or size controlling the rate of entry.
Liposomes injected intravenously show accumulation in the liver, lung and spleen. The - composition and size can be adjusted so that this accumulation represents 30% to 40% of the injected dose.
The remaining dose circulates in the blood stream for up t o 24 hours. `
The chosen method of delivery should result in cytoplasmic accumulation and molecules should have some nuclease-resistance for optimal dosing. Nuclear delivery may be used but is less preferable. Most preferred deliver~y~methods include liposomes (10-400 nm), hydrogels, controlled-release polymers, microinjection or electroporation~ (for ex vivo treatments) and other - pharmaceutically applicable vehicles. The dosage will 20' depend upon the disease indication and the route of ; admin1stratlon but should be between 100-200 mg/kg of body weight~/day. The duration of treatmen~ will extend through the course of the disease symptoms, possibly continuously.
-~ The number of~ doses will depend `upon disease delivery -25 vehicle and efficacy data from clinical trials. ;`
Est~ablishment of therapeutic levels of ribozyme within~the cell lS dependent upon the rate of uptake and degradation. Decreasing the dègree of degradation will prolong the intfracellular; half-life of the ribozyme.
Thus, chemically modified ribozymes, e.a., with modification of the phosphate backbone, or capping of the 5' and 3' ends of the rlbozyme with nucleotide analogs may require different dosaging. Descriptions of useful systems are provided in the art cited above, all of which is hereby incorporated by reference herein.
The claimed ribozymes are also useful as diagnostic tools to specifically or non-specifically ~ .

W093/23057 PCT/US93/0457'~-..

detect the presence of a target RNA in a sample. That is, the target RNA, if present in the sample, will be specifically cleaved by the ribozyme, and thus can be readily and specifically detected as smaller RNA species.
S The presence of such smaller RNA species is indicative of the presence of the target RNA in the sample.
Other embodiments are within the following claims.

.
.

Claims (12)

Claims

1. An enzymatic RNA molecule which cleaves mRNA associated with development or maintenance of chronic myelogenous leukemia, promyelocytic leukemia, Burkitt's lymphoma or acute lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic leukemia, breast cancer, colon carcinoma, neuroblastoma, and lung cancer, or which is active to specifically cleave mRNA expressed from a gene encoding multiple drug resistance.

2. The enzymatic RNA molecule of claim 1 which cleaves mRNA produced from the genes PML-RARA, C-myc, bc1-2, E2A-PRL, ErbB2/neu, ras, DCC, N-myc, L-myc or mdr-1.

3. The enzymatic RNA molecule of claim 1, which cleaves target mRNA having a sequence selected from SEQ. ID. NOS. 1-9 in Fig. 2; SEQ. ID. NOS. 1-19 in Fig. 3;
SEQ. ID. NOS. 1-62 in Fig. 4; SEQ. ID. NOS. 1-41 in Fig.
5; SEQ. ID. NOS. 1-22 in Fig. 6; SEQ. ID. NOS. 1-71 in Fig. 7; SEQ. ID. NOS. 1-118 in Fig. 8; SEQ. ID. NOS. 1-26 in Fig. 9; SEQ. ID. NOS. 1-66 in Fig. 10; and SEQ. ID.
NOS. 1-17 in Fig. 11.

4. The enzymatic RNA molecule of claims 1, 2 or 3, wherein said RNA molecule is in a hammerhead motif.

5. The enzymatic RNA molecule of claim 4, wherein said RNA molecule is in a hairpin, hepatitis Delta virus, group 1 intron, or RNaseP RNA motif.

6. The enzymatic RNA molecule of claim 4, wherein said ribozyme comprises between 5 and 23 bases complementary to said mRNA.

7. The enzymatic RNA molecule of claim 6, wherein said ribozyme comprises between 10 and 18 bases complementary to said mRNA.

8. A mammalian cell including an enzymatic RNA
molecule of claims 1, 2 or 3.

9. The cell of claim 8, wherein said cell is a human cell.

10. An expression vector including nucleic acid encoding the enzymatic RNA molecule of claims 1, 2 or 3, in a manner which allows expression of that enzymatic RNA
molecule within a mammalian cell.

11. A method for treatment of a disease caused by expression of an mdr-1 gene, chronic myelogenous leukemia, promyelocytic leukemia, Burkitt's lymphoma or acute: lymphocytic leukemia, follicular lymphoma, B-cell acute lymphocytic, leukemia, breast cancer, colon carcinoma, neuroblastoma, and lung cancer by administering to a patient an enzymatic RNA molecule of claims 1, 2 or 3.

12. The method of claim 11, wherein said patient is a human.