[19]Ribozymes as therapeutic agents and tools for gene analysis

[19]Ribozymes as therapeutic agents and tools for gene analysis

[19] Ribozymes as Therapeutic Agents and Tools for Gene Analysis John J. Rossi, Edovard Bertrand, and Daniela Castanotto Introduction Ribozymes are ...

989KB Sizes 2 Downloads 160 Views

[19]

Ribozymes as Therapeutic Agents and Tools for Gene Analysis John J. Rossi, Edovard Bertrand, and Daniela Castanotto

Introduction Ribozymes are RNA molecules with enzymatic activity. These RNAs can be engineered to base pair with and cleave targeted RNAs, be they of cellular or viral origins. As such, they represent a new approach for specific inactivation of gene expression, and for the therapeutic intervention of disease, both of viral and genetic origin. Ribozymes have been effectively used in cell cultures against viral targets, endogenous transcripts, and transfected genes. They have also been demonstrated to function effectively in vivo in transgenic animals (mouse and Drosophila), allowing the generation of phenotypes that are difficult to achieve by conventional genetic approaches. Ribozymes are receiving considerable attention as therapeutic agents for the treatment of virally initiated as well as endogenous hereditary diseases. Ribozyme engineering involves the design of RNA molecules in which the antisense and enzymatic functions are combined into a single transcript, which can specifically pair and cleave any target RNA. Because of their high target specificity, coupled with the possibility of being delivered either exogenously or endogenously, ribozymes have potential for use in both basic studies as well as therapeutic applications.

Catalytic Motifs There are five catalytic motifs that have been successfully adapted for use in ribozyme applications. These are the group I introns, RNase P, the hammerhead and hairpin motifs, and the self-cleaving domain of the hepatitis delta virus. The catalytic and base pairing domains are depicted in a schematic fashion in Fig. 1. Each of the ribozymes requires a divalent metal cation for activity (usually Mg 2+), which may participate in the chemistry of the cleavage reaction and/or may be important for maintaining the structure of the ribozyme (1). The group I intron of Tetrahymena was the first ribozyme for which the cis-cleaving (on a portion of the same RNA strand) reaction was converted into a trans (on an exogenous RNA molecule) reaction (2). RNase P, which is present in all prokaryotic and eukaryotic cells, has as its primary role in nature the maturation of the 5' ends of precursor tRNA transcripts. In vitro cleavage of non-tRNA molecules has been demonstrated via the use of RNAs designated as external guide sequences (EGS, Fig. 1). These sequences pair with the

348

Methods in Molecular Genetics, Volume 8

Copyright 9 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

[191 RIBOZYMES

349

target RNAs and provide the structural features required for recognition and cleavage by RNase P (3, 4). Small ribozymes have been derived from a motif found in single-stranded plant viroid and virusoid RNAs that replicate via a rolling circle mechanism (5). Based on a shared secondary structure and a conserved set of nucleotides, the term hammerhead has been given to one group of these self-cleavage domains (5). A self-cleaving motif resembling the hammerhead domain has also been found in a newt satellite RNA (6). The simplicity of the hammerhead catalytic domain (Fig. 1) has made it a popular choice in the design of trans-acting ribozymes (7, 8). The requirements at the cleavage site are relatively simple, and virtually any UX (where X is U, C, or A) can be targeted, although the efficiency of the cleavage reaction can vary more than 100-fold between the different cleavage site combinations (9). A second plantderived self-cleavage motif thus far identified only in the negative strand of the tobacco ringspot satellite RNA has been termed the hairpin or paperclip (Fig. 1) (10, 11). An engineered version of this catalytic motif has also been shown to be capable of cleaving and turning over multiple copies of a variety of targets in trans (11). The cleavage site requirements have been discerned and include an obligatory G at the nucleotide 3' to the site of cleavage (12). There is otherwise a great deal of flexibility in the substrate-guide sequence combinations, making this a potentially useful catalyst against a wide array of RNA targets. The other RNA catalytic motif that has been analyzed intensely and has now been engineered for cleavage in trans is the hepatitis delta (6) ribozyme (13-15). The delta virus has two different catalytic domains, one deriving from the plus strand and the other from the minus strand of the virus. Two' different versions of the delta ribozyme, both capable of cleavage in trans, have been described (14, 15). One of these utilizes a 73-nucleotide catalyst that has been demonstrated to turn over multiple target RNAs, and has a guide sequence that can be modified to bind and cleave a variety of RNA targets (15).

Design of Ribozymes and Antisense Oligonucleotides In this chapter we focus our attention on the hammerhead ribozyme motif because it has received the most attention from those interested in developing ribozymes as surrogate genetic tools. The choice of whether to use a ribozyme or nonenzymatic antisense is the first decision to be made. The advantages of the ribozyme approach are still somewhat theoretical because it is difficult to prove formally that a ribozyme is cleaving the target RNA or turning over multiple substrates in an intracellular environment. However, numerous experiments in which a standard antisense has been used as a control suggest that there is an added effectiveness if there is a catalytic center included within the antisense domain.

C

rn

lilllli

IIIII ~

c

. Z

~~ -zr r

Z Z

Z Z

rr

>

zS,

z~

0

~

lilltll~

~ o<

s

~

,,,

~

z ~ z~z

z~ ~z~z

~

-

z

4-

+

?o

can

= C)..

o

~ aS. o '

-~ -r-

u~

I"

0

IIII

O

l

l

n

l

.

x -r

! ,,0 l

IT"

9.Q

/~

("'

/

m

>

IIIItll

....

~

<<

~i

z

9,--"

0

9

9

-

..

9- . ~ ' . ~

.,~

~

~

~

.I.~

~

~~ ~

~

~ ~

,.~

.,~

.~

~

~

,.0

"

o~,~

~.

~

o

~

",~ ,~

~.~ "~e: ~ ~ ~~"-' ~ o ~ .~oc~ 2 2

[19] RIBOZYMES

351

Choice of Target As a first step in the design of a ribozyme, the target RNA needs to be selected: theoretical considerations suggest that stable rather than unstable RNAs may be the best targets (16). RNAs encoding regulatory proteins are often preferred over the RNAs encoding structural proteins, because small variations in the expression of the former can result in a significant downregulation of the genes with which these proteins interact. In choosing the region(s) to target within the selected RNA molecule, several aspects of mRNA biosynthesis in eukaryotic cells need to be considered: RNAs are always associated with proteins inside the cell; in the nucleus, pre-mRNAs are complexed with heterogeneous nuclear ribonucleoproteins (hnRNPs) (17, 18) and small nuclear ribonucleoproteins (snRNPs) (19, 20), whereas in the cytoplasm mRNAs are associated with the translational machinery. These factors are not distributed equally along the target RNA, so that some regions of the target may be highly accessible to the antisense or the ribozyme, and other regions may be totally inaccessible. The structure of the RNA can affect the availability of a targeted site. If the target sequence is sequestered by a secondary or tertiary structure, the accessibility of this sequence to degradation by a ribozyme or binding by an antisense molecule will probably be impaired. The prediction of RNA secondary structure using computerderived energy minimization programs is the most common method for target site selection (21), although the reliability of such computer-generated structures is not always good. If variations in the sequences are known [such as for variants of human immunodeficiency virus type 1 (HIV-1), a phylogenetic analysis can be a useful adjunct to the computer analysis for the prediction of the RNA structures. It is also useful to probe the structure of RNA in vitro using enzyme or chemical reagents (22). Although it is possible to determine directly the accessibility of a particular sequence in vivo (23) or in cell extracts, it requires intensive labor. Thus, accessible regions of the target should be found empirically by designing separate ribozymes to target several potential sites and testing their relative effectiveness.

Ribozyme Design In the simplest design of the hammerhead motif, the cleavage site sequences are defined by XUY (where X is U, G, A, or C, and Y is U, C, or A) with cleavage occurring 3' to Y. Different XUY sequences are cleaved with different efficiencies (9). Because the most commonly used site in nature is GUC (5), it is often the site of choice when present in the target RNA. When long and/or highly structured RNAs are targeted, the limiting step in the cleavage reaction is likely to be the binding/ hybridization step (24, 25); thus the choice of the XUY trinucleotide may not be so critical. A potentially useful approach for target site selections is to incubate the RNA

352

IV

RNA EDITING, RIBOZYMES, ANTISENSE RNA

target (end labeled) with a high molar excess of ribozymes with degenerate flanking sequences (26). This ribozyme population can select sites most accessible to binding and cleavage on the RNA substrate. The analysis of the timed appearance of cleavage products on a sequencing gel should result in the generation of a ladder of cleaved RNAs. Sites that are more accessible to ribozyme-mediate cleavage will preferentially show up as the strongest bands of the ladder. The number of base pairs required for optimal ribozyme cleavage at different sites is variable and must be determined empirically for each specific site, because the stability of the rib.ozyme-substrate duplex will be affected by several factors including G-C content, temperature, and RNA structure (27, 28). (a) Twelve to 14 base pairs (bp) is a good starting point, because it has been shown that fewer than 12 bp can result in poor binding and more than 14 bp can reduce turnover by slowing dissociation of ribozyme and cleavage products (25). (b) The inhibition that longer flanking sequences exert on product release may be overcome with the design of asymmetric ribozymes, in which the ribozyme-target pairing is extended on one side of the hybrid and shortened on the other, such that one of the two cleavage products is bound to the ribozyme by only a few bases (four or five; Fig. 2). Ribozymes with long flanking arms might be modified and inactivated by the activity of the dsRNA unwinding/modifying enzyme, which requires only 15-20 bp of double-stranded RNA (29). Increasing the length of the hybrid will decrease ribozyme specificity by stabilizing binding and, thus, cleavage of mismatched targets (30). Because the complexity of human RNA is about 100-fold less than that for human DNA (31), target specificity can be achieved with as few as 12 bp.

Chemical Synthesis of Ribozymes Antisense oligonucleotides and ribozymes can be efficiently synthesized on any DNA/RNA synthesizer. The major advantage that chemical synthesis offers over an endogenous expression system (from a transcriptional unit) is that base or backbone chemical modifications can be readily included in the molecules. Several chemical modifications introduced into ribozymes have been shown to increase stability without impairing catalytic capability. The 2'-ribose hydroxyl group renders RNA more sensitive to nucleases than DNA, and modification of this group can increase ribozyme stability. Only the 2'-OH required for catalysis must be preserved (32). For example, a ribozyme with DNA flanking sequences is catalytically active and two to three times more stable in the cell (33). Ribozymes containing 2'-fluorocytidine and 2'-fluorouridine, or 2'-aminouridines, are considerably more stable in serum and maintain catalytic activity (34). Modification of the 2'-OH in all but the six nucleotides of the hammerhead's conserved catalytic core to 2'-O-allylribonucleotides gave similar results (35). Phosphorothioate backbone substitutions can also render ribozymes more resistant

[191 RIBOZYMES

353 TRANSCRIFTIONALTERMINATOR

PROMOTER

ASYMMETRICRIBOZYMEENCODINGGENE P1

|

P2

__~

- -

NNNNNNNNNNNN~fNNNNN RIBOZYME

PI P2 OH..... NNNN~.r / . _ ~ / ' - NNNNNNNNNNNN~ N N N N N ~ . . ~ . .

FIG. 2 Schematic representation of a ribozyme expression vector and cleavage reaction. An asymmetric ribozyme gene is cloned between a promoter and an appropriate transcriptional terminator. The N's indicate complementary nucleotides on both sides of the cleavage site. The adjacent open boxes represent the non-base-pairing flanking sequences. Below, a diagrammatic representation of the trans-acting ribozyme cleavage reaction. The substrate, cleavage site, and ribozyme generated cleavage products (P1 and P2) are shown. After product dissociation the ribozyme is recycled and can catalyze a new reaction. The cleavage reaction occurs only in the presence of magnesium ions.

to nucleases (25). In addition to increased stability, these types of modifications can be particularly informative for mechanistic, functional, and structural studies (32).

Suppliers The following is a list of suppliers we have used for obtaining research materials: HIV strains, HIV molecular clones, and HIV-specific antibodies for quantitating HIV infection: National Institutes of Health AIDS Research and Reference Reagent Program (Rockville, MD)

354

IV

RNA EDITING, RIBOZYMES, ANTISENSE RNA

Tissue culture supplies: Costar (Cambridge, MA) Tissue culture media and lipofectin reagents: GIBCO-BRL Life Technologies, Inc. (Grand Island, NY) Other sources of cationic liposomes, cloning vectors, and in vitro transcription reagents: Stratagene (La Jolla, CA) and Boehringer-Mannheim (Indianapolis, IN)

P r o t o c o l 1: In Vitro T r a n s c r i p t i o n and T e s t i n g o f R i b o z y m e s 1. RNA molecules may be enzymatically synthesized using RNA polymerases (T7, T3, and SP6 are commonly employed). The DNA bearing the RNA molecule (ribozyme or the target for the in vitro assays) should be cloned in a transcriptional unit. It should be inserted as close to the promoter as possible, and the plasmic should be linearized immediately downstream of the ribozyme gene in order to minimize vector-derived flanking sequences. 2. In vitro transcription can yield RNA up to 50 times the amount of DNA template utilized in the reaction. To produce labeled transcripts use only 0.01-0.5 mM cold UTP and 10/zCi of [T-32p]UTP (3000 Ci/mmol). More than 90% of the radioactivity can be incorporated into the transcripts. The addition of spermidine increases the efficiency of the reaction. This step should be performed at room temperature to avoid template DNA precipitation. 3. Standard cleavage reactions are performed by first heating, to 90~ for 2 min, two separate tubes containing either the ribozyme or substrate (32p-labeled target) in a solution of 20 mM Tris-C1 (pH 7.5) and 0 - 1 4 0 mM KC1 (or NaC1). A noncleaving, mutant version of the ribozyme should be used as a control. This is best accomplished by altering one or more of the conserved nucleotides in the catalytic core (9). 4. Renature the samples for 5 min at the temperature chosen for the cleavage reaction (37-55 ~C) in the presence of 10 mM MgC12. 5. Mix different amounts of ribozyme and target (depending on the purpose of the analysis, at either equimolar, excess of ribozyme, or excess of target) at the desired temperature. Different time points should be taken, from 0 to 3 hr, the reactions can be stopped by adding an equal volume of formamide gel loading dye containing 20 mM EDTA. 6. Denature the samples by heating to 90~ for 2 min and analyze the cleavage products on a 7 M urea polyacrylamide gel. Values for kcat/Km c a n be determined for ribozymes in which product release is not rate limiting by incubating a constant concentration of substrate (around 1 nM) with an increasing excess of ribozyme for a constant time. The k~at/Km value is derived using the equation - ln(Frac S)/t = kcat/Km[ribozyme], where Frac S is the fraction of remaining substrate and t is time (25).

[19] RIBOZYMES

355

In Vivo Analysis o f Ribozyme Activity Exogenous Delivery To examine the effects of a ribozyme in vivo, the first obstacle to overcome is the delivery into the cells of choice. A major drawback of exogenous delivery is that the inhibitory effects of the nucleic acids so delivered are transient and require repeated administrations. Despite this, exogenously delivered molecules can incorporate chemical modifications that increase stability; although these modified bases and sugar-phosphate backbones can contribute to toxicity of the molecules. Further investigation is needed to establish whether the advantages of chemical modifications will overcome the general disadvantages of an exogenous delivery. RNA synthesized in vitro by T7 RNA polymerase can also be exogenously delivered. An advantage of this approach is that the ribozyme can be inserted within transcripts containing structural features (stem-loops) that confer stability. A disadvantage is that these extra sequences can negatively affect the catalytic activity of the ribozyme (36). Many techniques have been developed to introduce functional DNA into cells, and some of these may also be applicable to the delivery of synthetic RNA. DNA is complexed with various compounds (e.g., polylysine) or to lipophilic groups (37-39), which increase cellular uptake or can be complexed with receptor ligands for specific targeting and localization to defined types of cells (40). Conjugates of DNA and lipophilic derivatives can have higher antiviral activity as shown in the case of HIV-1 (39, 41). Antisense oligonucleotides can also be introduced into cells by direct addition to culture medium. RNA molecules, unless extensively modified, are extremely sensitive to degradation in the medium, and require protection from serum ribonucleases. To date, the most commonly utilized technique for delivering presynthesized RNA molecules into cells in culture is via liposome encapsulation. Liposomes are composed of one or more concentric phospholipid bilayers (which can incorporate lipid-soluble substances) surrounding an aqueous compartment that can incorporate water-soluble substances. Size and lipid composition can vary, and different liposomes exhibit different characteristics as in vivo delivery systems. Negatively charged lipids can increase the efficiency of cellular uptake; saturated lipids and the presence of cholesterol can increase liposome stability. Liposomes can be covalently attached to antibody molecules, resulting in specific binding to cellular antigens (42) allowing specific targeting to different types of cells. pH-sensitive liposomes on exposure to the low-pH environment of the endosomes fuse with the endosome membranes. Immunoliposomes (pH-sensitive liposomes conjugated to monoclonal antibodies) were successfully targeted to cell surface receptors in vitro and in vivo (43). The primary mechanism for cellular uptake of liposomes seems to be endocytosis

356

IV RNAEDITING,RIBOZYMES,ANTISENSERNA (44). Once in the cytoplasm, the liposomes are degraded and the nucleic acids contained inside are released.

Protocol 2: Encapsulation of Ribozymes into Cationic Liposomes

Equipment and Reagents Vortex mixer CO: incubator Laminar flow hood Lipofectin (GIBCO-BRL) or other cationic lipid analog reagents: The reagent consists of DOTMA (0.5 mg/ml) and dioleoylphosphatidylethanolamine (0.5 mg/ml) in sterile water. Reagents functionally similar to the lipofectin agent, such as Lipofect ACE (GIBCO-BRL), Transfectase (GIBCO-BRL), Lipofect-AMINE (GIBCO-BRL), Transfectam RM (Promega, Madison, WI), and DOTAP (Boehringer Mannheim Corp.) can all be used in this procedure Opti-MEM I (GIBCO) medium Serum (dependent on cell lines) DMEM high glucose (Irvine, Santa Ana, CA) PBS (1 X) (Irvine) Fungi Bact (Irvine) Pen-Strep (Irvine) Amphotericin B (Fungizone; Irvine) 2-Mercaptoethanol Sodium pyruvate (Irvine) Trypsin (1 X) (Irvine) Sodium bicarbonate (7.5%) solution L-Glutamine (200 mM) Plasmid DNA containing ribozyme gene transcriptional unit RNA produced from in vitro transcription or chemically synthesized Synthetic antisense DNA oligonucleotides Polystyrene tubes, sterile, 17 x 100 mm Culture dish (60 mm; Costar)

Method 1. Plate exponentially growing cells in tissue culture dishes at 5 x 105 cells/well and grow overnight in a CO2 incubator at 37~ to 80% confluency.

[19] RIBOZYMES

357

2. Dilute the presynthesized RNA molecules and the lipofectin reagent (BRL) with Opti-MEM I (GIBCO) medium. The amounts of nucleic acids and liposome suspension need to be optimized for each cell type. 3. Mix the diluted reagent from the previous step, vortex gently for several seconds at room temperature, and incubate for 5 to 10 min at room temperature (prepare this complex in a polystyrene tube, because it can stick to polypropylene). 4. Wash the cells three times with serum-free medium. 5. Add the liposome complex and incubate the cells at 37~ in a CO2 incubator (5-10% CO2) for 3 - 6 hr. In general, transfection efficiency increases with time, although after 8 hr toxic conditions may develop. 6. Add 3 ml of medium with 20% (v/v) serum (the serum is dependent on the cell type; fetal calf serum may be used). 7. Incubate the cells for 2 4 - 4 8 hr at 37~ in a CO2 incubator. 8. Harvest the cells and assay for gene activity.

Endogenous Delivery of Ribozymes Endogenous delivery involves the expression of a ribozyme from a DNA template permanently maintained within the cell. For an example of an expression vector, see Fig. 2. Expression of these molecules can be directed by RNA polymerase II and III (Pol II or Pol III) promoters. Pol II promoters include those promoters of viral origin, the long terminal repeat (LTR) promoter sequences of retroviruses, and strong cellular promoters (e.g., the actin promoter). Tandem repeats of the ribozyme genes, under the control of the same promoter, may help to increase the effective concentration from each transcript. Inducible, repressible, or tissue-specific promoters can be used to confer temporal or cell type specific expression and may temper other problems, such as cellular toxicity generated by high levels of expression within the cells. Endogenous expression from a Pol II promoter, with the exception of a few specialized cases such as the human U1 snRNA Pol II promoter, necessitates a polyadenylation signal, which allows addition of a poly(A) tail. This, along with the 5'-m 7 GpppG cap, common to Pol II transcripts, may prolong the intracellular halflife of the RNA molecules. Expressing these molecules under the control of Pol III promoters affords additional advantages" Pol Ill-driven gene expression seems to occur at high levels in all tissues. The size of Pol Ill-transcribed genes is smaller, presenting a more defined transcript. Other expression strategies are possible. For instance, an snRNA transcription unit (45), which incorporates portions of the snRNA structural sequence and protein-binding sites, could be used. This can facilitate targeting to the nucleus. Another option is to insert a ribozyme into the acceptor arm or the anticodon loop of a tRNA gene, which has resulted in higher levels of expression and stability of the

358

IV

RNA EDITING, RIBOZYMES, ANTISENSE RNA

ribozyme (46). However, this tRNA expression system has been shown to alter posttranscriptional processing and cellular transport (46).

Viral Vectors Although the ribozyme or antisense expression vector can be delivered with the previously described techniques, integration of the foreign DNA into the host genome does not occur with a high frequency. More promising and efficient technologies employ viral vectors. Different viral vectors have the capacity to infect a variety of cell types with high efficiency. Several classes of viral vectors are being exploited for delivery of genes in vivo, including DNA viruses (adenoviruses, herpes virus, and adeno-associated virus) and RNA retroviruses. General concerns persist with the use of viral pathogens such as residual infectivity, toxicity, and rescue of infectivity by recombination. Ad, ditionally, each viral vector has its own set of advantages and disadvantages that ultimately dictate its use in a specific application. To date the most extensively utilized viral vectors have been retroviruses. This class of viruses can infect a wide variety of cell types resulting in long-term persistence as a consequence of integration into the host chromosome. However, the integration process requires cell replication, thereby restricting retroviral use to actively dividing cells. Another promising vector system is adeno-associated virus (AAV). The adenoassociated viruses are nonpathogenic, integrating viruses that require helper viruses for replication of their genome. The AAV vector can exist autonomously at high copy number within a cell, and can integrate into a specific site within chromosome 19 (47). The intracellular transfer of nucleic acids is continuously improving, but many problems remain to be solved. The efficiency of cellular transformation, targeting to specific tissue and organs, subcellular localization of the RNAs, and maintenance of ribozyme activity within the intracellular environment are some of the major issues to be addressed. Other obstacles are the timing of expression and regulation of the level of endogenously synthesized molecules inside the target cells.

In Vivo A s s a y s For in vivo analyses, standard techniques are performed at the levels of RNA [Northern gels, primer extension, polymerase chain reaction (PCR)] and proteins. RNA analysis is not by itself conclusive because it is possible that the cleavage can occur

[191 RIBOZYMES

359

during the RNA extraction. Use a mutant, noncleaving ribozyme as control to establish that any effect seen in vivo is a result of a specific ribozyme activity. The following steps are used for in vivo assays of ribozyme genes. 1. Clone the ribozyme gene in a mammalian expression vector, downstream of a strong promoter (e.g., human/3-actin promoter). The vector must contain a marker for the isolation of transfected clones. 2. Transfect the resulting vector into the appropriate cells and isolate stable clones. It is also possible to work with pools of stable clones, so that clonal variability can be eliminated. 3. Examine these clones for the presence of the ribozyme RNA by Northern gel, RNase protection, or RT-PCR (reverse transcription-polymerase chain reaction) analysis. Assay for ribozyme-mediated inhibition of the targeted function using enzymatic, immunological, or physiological assays for the protein product.

Conclusions The use of ribozymes as surrogate genetic tools and as therapeutic reagents for the treatment of disease is developing rapidly. Ribozymes have the inherent advantage over conventional antisense molecules of not only binding to the target RNA, but also cleaving it, thereby ensuring its permanent inactivation. Because ribozymes base pair with their substrates, antisense effects may contribute to a decrease in steady state levels of the targeted RNA-encoded product. Although this is advantageous from a practical point of view, the optimal design of ribozymes for in vivo use requires the development of assays in which the antisense effects of ribozymes can be quantitated separately from the cleavage effects. To date, despite the potential for catalytic activity, excesses of ribozyme over substrate are often required to achieve an inhibition of expression. We have a great deal to learn about maximizing intracellular targeting and stability of ribozymes, as well as learning more about the mechanisms governing intracellular localization before the maximal effectiveness of ribozymes can be achieved. Despite this, ribozymes have been successfully utilized in cell culture as well as animal models to study the effects of inhibiting specific gene expression, and can therefore be considered as a new surrogate genetic tool.

Acknowledgments This work was supported by American Foundation for AIDS Research (AmFAR) Grant 01917-14-RG, the American Foundation for Pediatric AIDS Research (PAF/AmFAR) Grant 500331-14-PG, and the National Institutes of Health Grants AI25959 and AI29329.

360

IV RNA EDITING, RIBOZYMES, ANTISENSE RNA

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

S. C. Dahm, W. B. Derrick, and O. C. Uhlenbeck, Biochemistry 32, 13040 (1993). A. J. Zaug and T. R. Cech, Science 231, 470 (1986). A. C. Forster and S. Altman, Science 249, 783 (1990). Y. Yuan, E. Hwang, and S. Altman, Proc. Natl. Acad. Sci. U.S.A. 89, 8006 (1992). A. C. Forster and R. H. Symons, Cell 49, 211 (1987). L. M. Epstein and J. G. Gall, Cell 48, 535 (1987). O. C. Uhlenbeck, Nature (London) 328, 596 (1987). J. Haseloff and W. L. Gerlach, Nature (London) 334, 585 (1988). D. E. Ruffner, G. E. Stormo, and O. C. Uhlenbeck, Biochemistry 29, 10695 (1990). J. M. Buzayan, A. Hampel, and G. Bruening, Nucleic Acids Res. 14, 9729 (1986). A. Hampel and R. Tritz, Biochemistry 28, 4929 (1989). S. Joseph et al., Genes Dev. 7, 130 (1993). M. G. Belinsky, G. Dinter-Gotttieb, and J. Taylor, Nucleic Acids Res. 19, 559 (1991). A. D. Branch and H. D. Robertson, Proc. Natl. Acad. Sci. U.S.A. 88, 10163 (1991). A. T. Perotta and M. D. Been, Biochemistry 31, 16 (1992). E. Bertrand, R. Grange, and R. Pictet, in "Gene Regulation Biology of Antisense RNA and DNA" (R. P. Erickson and J. G. Izant, eds.), p. 71. Raven Press, New York, 1992. G. Conway, J. Wooley, T. Bibring, and W: M. Lestourgeon, Mol. Cell. BioL 8, 2884 (1988). G. Dreyfuss, M. S. Swanson, and S. Pinol-Roma, Trends Biochem. Sci. 13, 86 (1988). J. Rinke, A. Bernd, H. Blocker, R. Frank, and R. Lurhmann, Nucleic Acids Res. 12, 4111 (1984). D. L. Black, B. Chabot, and J. A. Steitz, Cell 42, 737 (1985). J. A. Jaeger, D. A. Turner, and M. Zucker, Proc. Natl. Acad. Sci. U.S.A. 86, 7706 (1989). C. Ehresmann, E Baudin, M. Mougel, P. Romby, J. P. Ebel, and B. Ehresmann, Nucleic Acids Res. 15, 9101 (1987). E. Bertrand, M. Fromont-Racine, R. Pictet, and T. Grange, Proc. Natl. Acad. Sci. U.S.A. 90, 3496 (1993). A. Xing and J. L. Whitton, Biochemistry 66, 1361 (1992). O. Heidenreich and E Eckstein, J. Biol. Chem. 267, 1904 (1992). C. Rittner, C. Burmester, and G. Sczakiel, Nucleic Acids Res. 21, 1381 (1993). S. M. Freier, R. Kierzek, J. A. Jaeger, N. Sugimoto, M. H. Caruthers, T. Neilson, and D. H. Turner, Proc. .Natl. Acad. Sci. U.S.A. 83, 9373 (1986). M. J. Fedor and O. Uhlenheck, Proc. Natl. Acad. Sci. U.S.A. 87, 1668 (1990). K. Nishikura, C. Yoo, U. Kim, J. M. Murray, P. A. Estes, E E. Cash, and S. A. Liebhaber, EMBO J. 10, 3523 (1991). D. Hershlag, Proc. Natl. Acad. Sci. U.S.A. 88, 6921 (1991). B. Lewin, "Genes," p. 295. John Wiley & Sons, New York, 1983. O. Heidenreich, W. Pieken, and E Eckstein, FASEB J. 7, 90 (1993). N. R. Taylor, B. E. Kaplan, O. P. Seiderski, H. Li, and J. J. Rossi, Nucleic Acids Res. 20, 4559 (1992). W. A. Pieken, D. B. Olsen, E Benseler, H. Aurup, and E Eckstein, Science 253, 314 (1991).

[191 RIBOZYMES 35. 36. 37. 38. 39. 40. 41.

42. 43. 44. 45. 46. 47.

36 1

G. Paolella, B. S. Sproat, and A. I. Lamond, EMBO J. 11, 1913 (1992). N. Taylor and J. J. Rossi, Antisense Res. Dev. 1, 173 (1991). J. P. Leonetti, G. Degols, and B. Lebleu, Bioconjugate Chem. 1, 149 (1990). A. S. Boutorin, L. V. Guskova, E. M. Inanova, N. D. Kobetz, V. F. Zarytova, A. S. Ryte, L. V. Yurchenko, and V. V. Vlassov, FEBS Lett. 254, 129 (1989). R. L. Letsinger, G. Zhang, D. K. Sun, T. Ikeuchi, and R S. Sarin, Proc. Natl. Acad. Sci. U.S.A. 86, 6553 (1989). G. Y. Wu, J. M. Wilson, E Shalaby, M. Grossman, D. A. Shafritz, and C. H. Wu, J. Biol. Chem. 266, 14338 (1991). T. V. Abromova, V. M. B linov, V. V. Vlassov, V. V. Gorn, V. E Zarytova, E. M. Ivanova, D. A. Konevets, O. A. Plyasunova, A. G. Pokrovsky, L. S. Sandahchiev, E P. Svinarchuk, V. P. Starostin, and S. R. Chaplygina, Nucleotides Nucleosides 10, 419 (1991). S. Wright and L. Huang, Adv. Drug Delivery Rev. 3, 343 (1989). R.J.Y. Ho, B. T. Rouse, and L. Huang, J. Biol. Chem. 262, 13973 (1987). C. R. Alving, Adv. Drug Delivery Rev. 2, 107 (1988). C. Guthrie and B. Patterson, Annu. Rev. Genet. 22, 387 (1988). M. Cotten and M. L. Birnstiel, EMBO J. 12, 3861 (1989). R. M. Kotin, M. Siniscalo, J. Samulski, X. Zhu, L. Hunter, C. Laughlin, S. McLaughlin, N. Muzycka, M. Rocchi, and K. I. Berns, Proc. Natl. Acad. Sci. U.S.A. 87, 2211 (1990).