AU4107399A - Novel enzymatic RNA molecules - Google Patents

Description

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT *I Applicant: Invention Title: THE SCRIPPS RESEARCH INSTITUTE Novel Enzymatic RNA Molecules The following statement is a full description of this invention, including the best method of performing it known to me/us: -1A- NOVEL ENZYMATIC RNA MOLECULES TECHNICAL FIELD The present invention relates to nucleic acid enzymes or enzymatic RNA molecules for cleaving DNA under physiologic conditions. The present invention also relates to nucleic acid enzymes or enzymatic RNA molecules that are capable of cleaving a variety of bonds, including phosphodiester bonds and amide bonds, in a variety of substrates. Thus, the disclosed enzymatic RNA molecules are capable of functioning as nucleases or as peptidases. The present invention also relates to compositions containing the disclosed S• enzymatic RNA molecules and to methods of making and 15 using such enzymes and compositions.

BACKGROUND

i The need for catalysts that operate outside of their native context or which catalyze reactions that are not represented in nature has resulted in the 20 development of "enzyme engineering" technology. The usual route taken in enzyme engineering has been a "rational design" approach, relying upon the understanding of natural enzymes to aid in the construction of new enzymes. Unfortunately, the state of proficiency in the areas of protein structure and chemistry is insufficient to make the generation of novel biological catalysts routine.

Recently, a different approach for developing novel catalysts has been applied. This method involves the construction of a heterogeneous pool of macromolecules and the application of an in vitro selection procedure to isolate molecules from the pool that catalyze the desired reaction. Selecting catalysts from a pool of macromolecules is not dependent on a comprehensive understanding of their structural and chemical properties. Accordingly, this process has been dubbed "irrational design" (Brenner :f and Lerner, Proc. Natl. Acad. Sci. USA 89: 5381-5383 (1992)).

The process of Darwinian evolution, by which enzymes arise in nature, does not operate by generating a diverse population of variants and harvesting the most advantageous individuals. In biological systems, diversity is maintained by ongoing mutations, and the population is shaped by selection. Novel mutations augment existing variation, so that the evolutionary search is biased, in an appropriate fashion, by selection events that have already occurred (Eigen, et al., J Phys. Chem. 92: 6881 (1988)). The more advantageous mutants, which are relatively abundant in the population, give rise to larger numbers of novel variants when compared to the less advantageous mutants.

15 Most efforts to date involving the rational design of enzymatic RNA molecules or rihnzvmes have not led to molecules with fundamentally new or improved catalytic function. However, the application of irrational design methods via a process we have described as 20 "directed molecular evolution" or "in vitro evolution", which is patterned after Darwinian evolution of organisms in nature, has the potential to lead to the production of RNA molecules that have desirable functional characteristics.

This technique has been applied with varying degrees of success to RNA molecules in solution (see, Mills, et al., PNAS USA 58: 217 (1967); Green, et al., Nature 347: 406 (1990); Chowrira, et al., Nature 354: 320 (1991); Joyce, Gene 82: 83 (1989); Beaudry and Joyce, Science 257: 635-641 (1992); Robertson and Joyce, Nature 344: 467 (1990)), as well as to RNAs bound to a ligand that is attached to a solid support (Tuerk, et al., Science 249: 505 (1990); Ellington, et al., Nature 346: 818 (1990)). It has also been applied to pept.ides attached directly to a solid support (Lam, et al., Nature 354: 82 (1991)); and to peptide epitopes expressed within a viral coat protein (Scott, et al., Science 249: 386 (1990); Devlin, et al., Science 249: 404-406 (1990); Cwirla, et al., PNAS USA 87: 6378 (1990)) However, as disclosed herein; a remarkable degree of success in engineering novel enzymatically active oligonucleotide molecules has now been achieved.

Therefore, the discoveries and inventions disclosed herein are particularly significant, in that they highlight the potential of in vitro evolution as a means of designing increasingly more efficient catalytic molecules.

SUMMARY OF THE INVENTION Site-directed mutagenesis has now been improved by in vitro selected amplification techniques for S"generating large numbers of mutants with subsequent S 15 selection of some desirable property. Individual "macromolecules are selected, and those selected are then amplified to generate a proqeny distribution of favorable mutants. The process is repeated until only those individuals with the most desirable properties 20 remain. For example, by following the within-disclosed guidelines, one may successfully engineer new enzymatically active oligonucleotide molecules. Not only are the within-disclosed techniques useful in the design, identification and use of enzymatically active RNA molecules with improved specificities, reaction rates, and substrate binding capabilities, to name a few examples, success has now been achieved in designing oligonucleotide molecules that cleave bonds other than, or in addition to, phosphodiester bonds generally linking adjacent nucleotides in oligonucleotide molecules.

Therefore, in various disclosed embodiments of the present invention, enzymatic nucleic acids, particularly enzymatic RNA molecules, are prepared, selected, and synthesized in useful quantities for various uses. Selection criteria include, without limitation, the ability of the enzymatic RNA molecule to catalyze a sequence-specific reaction, to cleave nucleic acids, to bind substrate DNA and/or RNA, to display an improved turnover rate, and the like.

Enzymatic RNA molecules having peptidase activity are also disclosed. Enzymatic RNA molecules of the present invention are thus capable of functioning as nucleophiles, cleaving phosphodiester bonds, amide bonds, or both.

Therefore, the present invention contemplates enzymatic RNA molecules capable of specifically cleaving a single-stranded nucleic acid molecule under physiologic conditions, wherein the enzymatic

RNA

molecules include one or more point mutations which improve the enzymatic performance of the enzymatic

RNA

molecules. In various embodiments, the enzymatic

RNA

15 molecule further includes one or more point mutations which affect the substrate specificity of the enzrmatic RNA molecule.

In one variation, the enzymatic performance comprises catalytic efficiency. Alternative 20 embodiments contemplate that the enzymatic RNA molecule has a cleavage rate of about 0.7 min-'.

It is also contemplated that enzymatic performance may comprise substrate binding affinity. In various embodiments, the substrate may comprise DNA, RNA, or composites thereof. In embodiments in which the substrate comprises DNA, an enzymatic RNA molecule may have a substrate binding affinity of at least 10 9

M.

In other variations, an enzymatic RNA molecule of the present invention binds DNA with a K D of less than iM. In alternative embodiments, an enzymatic

RNA

molecule of the present invention binds DNA with a KD of less than 1 jIM; with a KD of less than 50 nM; or with a KD of less than 10 nM.

In embodiments in which the substrate comprises RNA, an enzymatic RNA molecule preferably binds

RNA

with a KD of less than 1.0 nM, and more preferably, with a KD of 0.5 nM or less. In other embodiments, an enzymatic RNA molecule has an RNA substrate cleavage rate up to three times greater than that of wild-type ribozymes. Still other embodiments contemplate enzymatic RNA molecules wherein enzymatic performance comprises substrate specificity. In various embodiments, that specificity is changed via altering the recognition sequence. As noted above, substrates may comprise DNA, RNA, or composites thereof.

In embodiments in which the substrate comprises DNA, an enzymatic RNA molecule invention preferably has a DNA substrate cleavage rate 10-102, 102-103, 103-104, or even 104-105 times greater than that of wild-type ribozymes. A cleavage rate exceeding 105 is also contemplated in various embodiments.

The present invention further contemplates enzymatic RNA molecules derived from group I, II, III, or IV introns. Preferably, an enzymatic RNA molecule of the present invention is derived from a group I intron. In one variation, the group I intron is a Tetrahymena group I intron (shown in SEQ ID NO: In another variation, an enzymatic RNA molecule contemplated herein comprises the portions of a Tetrahymena group I intron having catalytic activity. In yet another embodiment, an enzymatic RNA molecule of the present invention is derived from an L-19 or L-21 RNA molecule and includes the portions of the L-19 or L-21 RNA molecule having the catalytic activity.

Various embodiments of the disclosed invention contemplate that an enzymatic RNA molecule of the 30 present invention includes one or more mutations not typically found in wild-type enzymatic RNA molecules or ribozymes. In various embodiments, mutations are introduced into the group I intron shown in SEQ ID NO:1 and are selected from the group consisting of: 44:C--A; 51/52:insert AGAA; 87:A-, deleted; 94:A-,U; 94:A-C; 115:A-U; 116:G-A; 138:C4A; 166 167:U-G; 170:CU 188:GA; 190:U-*A; 191:G- U; 205:U- C; 215:G-A; 239:U-A; 258:U-C; 312:G-*A; 313:G->U; 313:G- C; 314:A-*G; 317:U-,G; 317:U-C 317:U- A; 333:U-C; 35O:C__>U; and 364:C--.U.

In various alternative embodiments, an enzymatic RNA molecule of the present invention has 1-4 point mutations, 5-8 point mutations, 9-12 point mutations, or 13 or more point mutations.

In various exemplary embodiments, the point mutations of SEQ ID NO: 1 may comprise a 215: mutation, a 25 8: U- C mutation, or both. In another embodiment, an enzymatic RNA molecule includes the following mutations of SEQ ID NO: 1: 94:A- Y, 215:G-A, and 3 13-314:GA ,UG. Still another example includes the mutations 94:A->Y, 215:G-.*A, 313-314:GA-->UG, and 317:U-- R of SEQ ED NO: 1, while yet another example includes the mutations 94:A-+Y, 215:G---A, 313-314:GA- UG and 333:U-4C In another embodiment, an enzymatic RNA molecule includes the following mutations of SEQ ID NO:1I 94:A->Y and 313- 314:GA-4JUG. Still another example includes the following mutations of SEQ ID NO:1: 215:G-4A and 313-314:GA-->UG. In yet another embodiment, an enzymatic RNA molecule of the presene invention includes the following 'mutations of SEQ ID NO: 1: 94:A--3Y, 115:A-.U, 116:G-+A, l90:U-3-A, 191:G-*U, 205:U-3-C, 215:G-,*A, and 313-314:GA--3-UG.

In another variation, the invention contemplates **an enzymatic RNA molecule including the following mutations of SEQ ID NO: 1: 87:A-4del, 94:A-4U, 1 15:A-*U, 1 16:G-*-A, 166:C-3.A, 170:C-->U, 188:G-* A, l90:U-*.A, 191:G-*U, 205:U-3*C, and 215:G-+A.

Other examples of combinations of mutations of SEQ ID NO: 1 which *.:may be present 'in enzymatic RNA molecules of the :present invention include the following: 9 8: C-+U and 3 13-314:GA-*UG; 98:C--3.U, 205:U--3C, and 317:U-+.R; 94:A-.Y and 215:G--3A; 94:A-*Y, 205:U-->C, and 313- 314:GA-*UG; Ce) 94:A-.Y, 98:C->U, and 333:U-i.C; 44:G->A, 191:G-U, 205:U-*C, 215:G-.A, 312:G--.A, and 317:U--.G; (g) 190:U-.-A, 191:G-+U, 205:U--3-C, 215:G-*A, 239:U-.-A, and 312:G--*A; 51/52:insert AGAA, 87:A--.del, 190:U- A, 191:G-*U, 205:U-C, 215:G-*A, 239:U-A, 312:G-*A, 350:C- U, and 364:C-U; or 44:G- A, 51/52:insert AGAA, 87:A-del, 94:A--U, 115:A--U, 116:CG-A, 166:C-A, 170:C-U, 188:G-*A, 190:U-A, 191:G->U, 205:U-C, 215:G-*A, 313:G-*C, and 314:A->G.

In various disclosed embodiments, the mutations are concerted. In one exemplary embodiment, the concerted mutations of SEQ ID NO: 1 comprise a tandem 313-314: GA--UG mutation. In another example, the concerted mutations of SEQ ID NO: 1 comprise a 215:G-A mutation and a 258:U-C mutation.

In yet another variation, the concerted mutations comprise mutations at nucleotide positions 188, 190, and 191 of SEQ ID NO: 1. In still another variation, the concerted mutations comprise mutations at nucleotide positions 115, 116, and 205 of SEQ ID NO:1. Another embodiment includes concerted mutations comprising mutations at nucleotide positions 15, 116, 188, 190, 191, and 205 of SEQ ID NO:1.

The present invention further contemplates an enzymatic RNA molecule capable of specifically cleaving 20 single-stranded DNA under physiologic conditions, wherein the enzymatic RNA molecule includes one or more point mutations which affect the enzymatic performance of the molecule. In various embodiments, an enzymatic RNA molecule of the present invention further includes 25 one or more point mutations which affect the substrate specificity of the molecule.

The present invention also contemplates various methods of making and using enzymatic RNA molecules according to the present invention. For example, a S" 30 method for specifically cleaving a single-stranded DNA molecule under physiologic conditions, is contemplated herein, which comprises the steps of: providing an enzymatic RNA molecule having a deoxyribonuclease activity; contacting the enzymatic RNA molecule with a single-stranded DNA molecule under physiologic conditions; and maintaining the contact for a sufficient time -8to allow the;enzymatic RNA molecule to cause the single-stranded DNA molecule to be cleaved.

In another embodiment, the method further comprises providing the enzymatic RNA molecule in a reaction medium at a concentration sufficient to cause cleavage of about one molecule of DNA per molecule of enzymatic RNA per minute. In yet another embodiment, the method further comprises providing the enzymatic RNA molecule in a reaction medium, wherein the enzymatic RNA molecule is present at a concentration sufficient to cause cleavage of at least 10% of a population of DNA molecules in an hour.

0 0 :0,0 In one variation of the disclosed methods for specifically cleaving a single-stranded DNA molecule S 15 under physiologic conditions, an enzymatic RNA molecule comprises a bindina site fnr nd_ DMA which.. binding site is complementary to nucleotides adjacent to a cleavage site on the single-stranded DNA molecule.

In another variation of the disclosed methods, an 20 enzymatic RNA molecule of the present invention comprises a binding site for single-stranded DNA, which binding site is complementary to nucleotides adjacent to a cleavage site on the single-stranded DNA molecule.

The invention also contemplates a method of producing an enzymatic RNA molecule having a predetermined catalytic activity, comprising: subjecting a population of enzymatic

RNA

molecules to mutagenizing conditions to produce a diverse population of mutant RNA molecules; selecting an enzymatic RNA molecule having a predetermined activity from the diverse population of mutant enzymatic RNA molecules; and separating the RNA molecule from the diverse population of mutant RNA molecules.

In one alternative method, the mutagenizing conditions comprise conditions that introduce defined or random nucleotide substitutions within an enzymatic RNA molecule. In another variation, the mutagenizing conditions comprise chemical modification, incorporation of randomized mutagenic oligodeoxynucleotides, or inaccurate copying by a polymerase. In yet another variation, the mutagenizing conditions comprise use of site-directed mutagenesis, polymerase chain reaction, or self-sustained sequence replication.

In various embodiments, the predetermined activity comprises the ability to cleave DNA under physiologic conditions.

Another variation of the foregoing methods further comprises the step of amplifying the enzymatic RNA molecules selected from the diverse population. In one *9 embodiment, the amplifying is performed using a 15 polymerase chain reaction. In another embodiment, the ego• amplifying is performed using self-sustained sequence r N I~ i r, t-i n n :"As noted hereinabove, the present invention also discloses enzymatic RNA molecules capable of S 20 specifically cleaving amide bonds, wherein the enzymatic RNA molecules include one or more point mutations which improve the enzymatic performance of the enzymatic RNA molecules. In various embodiments, the enzymatic RNA molecule further includes one or more point mutations which affect the substrate specificity of the enzymatic RNA molecule. In one variation, the enzymatic performance comprises catalytic efficiency.

It is also contemplated that enzymatic performance may comprise substrate binding affinity. In various embodiments, the substrate may comprise a polypeptide or protein.

Still other embodiments contemplate enzymatic RNA molecules wherein enzymatic performance comprises substrate specificity. In various embodiments, that specificity is changed via altering the recognition sequence. As noted above, substrates may comprise a polypeptide or protein.

The present invention contemplates enzymatic RNA molecules that cleave amide bonds. In one embodiment the enzymatic RNA molecule is derived from a group

I,

II, III, or IV intron. In one variation, the group

I

intron is derived from a group I intron; in another variation, the group I intron is derived from the group I intron of Tetrahymena thermophila precursor rRNA. In another embodiment, an enzymatic RNA molecule of the present invention is derived from the molecule identified herein as SEQ ID NO 1.

In another variation, an enzymatic RNA molecule contemplated herein comprises the portions of a group I, II, III or IV intron having catalytic activity. In an alternative embodiment, an enzymatic RNA molecule comprises the portions of a Tetrahymena group I intron 15 having catalytic activity. In yet another embodiment, an enzymatic RNA molecule of the present invention is derived from an L-19 or L-21 RNA molecule and includes 9the portions of the L-19 or L-21 RNA molecil having catalytic activity.

The present invention further contemplates amide bond- or peptide bond-cleaving enzymatic RNA molecules including one or more mutations. Various embodiments of the disclosed invention contemplate that an enzymatic RNA molecule of the present invention includes one or more mutations not typically found in wild-type enzymatic RNA molecules or ribozymes. In various embodiments, mutations are introduced into the group I intron shown in SEQ ID NO:1, and include one or more of the following mutations: 44:G-A; 51/52:insert AGAA; 87:A-. deleted; 94:A-U; 94:AC; 115:A U; 116:GA; 138:C-A; 166:C-A; 167:U-.G; 170:CU; 188:G-A; 190:U->A; 191:G-U; 205:U-C; 215:G-.A; 239:U-A; 258:U-C; 312:G-A; 313:G-U; 313:G-*C; 314:A-G; 317:U-G; 317:UC 317:UA; 333:UC; 350:C-U; and 364:CU. In various altetnative embodiments, an enzymatic

RNA

molecule of the present invention has 1-4 point mutations, 5-8 point mutations, 9-12 point mutations, or 13 or more point mutations.

-11- Other examples of combinations of mutations of SEQ ID NO: 1 which may be present in amide bond- or peptide bond-cleaving enzymatic RNA molecules of the present invention include the following: 98:C->U and 313-314:GA>UG; 98:C- U, 205:U C, and 317:U-*R; 94:A-*Y and 215:G- A; 94:A-Y, 205:U C, and 313-314:GA-*UG; (e) 94:A-Y, 98:C-U, and 333:UC; 44:G-A, 94:A-U, 115:A U, 116:G 138:C A, 188:GCA, 190:U A, 191:G-U, 205:U- C, 215:G 312:G-A, and 317:U-*G; 44:G-*A, 94:A-U, 115:AU, 116:G-A, 138:C-*A, 167:U-*G, 188:G-A, 190:U-A, 191:G-*U, 205:U C, 215:GA, 239:U- A, and 312:G+A; 44:G-A, 51/52:insert AGAA, 87:A-*del, 94:A-U, 115:A-U, 116:G-A, 166:C-A, 170:C U, 188:G A, 190:U+A, 191:G 205:U C, 215:G+A, 239:U-)A, 312:G)A, 350:C-*U, and 364:C-*U; or 44:GC-A, 51/52:insert AGAA, 87:A--del, 94:A- U, 115:A--U, 116:G-*A, 166:CA, 170:C-U, 188:G-A, 190:U-A, 191:G)U, 205:U C, 215:G-A, 313:G-C, and 314:A- G.

fees ~The present invention further contemplates an 1:0 20 enzymatic RNA molecule capable of specifically cleaving S0..

amide bonds, wherein the enzymatic RNA molecule includes one or more point mutations which affect the enzymatic performance of the molecule. In other variations, an enzymatic RNA molecule of the present invention includes one or more point mutations which improve the substrate specificity of the molecule. In alternative embodiments, an enzymatic RNA molecule of the present invention includes one or more mutations which improve enzymatic performance and substrate 30 specificity. In an alternative embodiment, an enzymatic RNA molecule capable of specifically cleaving amide bonds is disclosed, wherein the enzymatic RNA molecule includes one or more point mutations which affect the enzymatic performance or substrate specificity of the molecule.

In one variation, the enzymatic performance comprises catalytic efficiency. It is also contemplated that enzymatic performance may comprise -12substrate binding affinity. Still other embodiments contemplate enzymatic RNA molecules wherein enzymatic performance comprises substrate specificity. In various embodiments, that specificity is changed via altering the recognition sequence.

The present invention further contemplates a ribozyme amidase intermediate comprising a ribonucleotide polymer including a 5' terminal nucleotide with a ribone sugar having a 2' hydroxyl, and a piptide having one or more amino acid residues including a carboxy terminal amino acid residue, the carboxy terminal amino acid residue being covalently linked by an ester bond to the 2' hydroxyl of the ribonucleotide polymer. In one alternative embodiment, 15 the ester bond is chemically unstable under physiological conditions. In another the ester bond is acid labile. The invention further contemplates embodiments whereby the ribonucleotide polymer has a catalytic activity for hydrolyzing the ester bond.

In yet another variation, the present invention contemplates a ribozyme amidase intermediate comprising a ribonucleotide polymer; a cofactor including a guanine nucleotide having a ribose sugar with a 2' hydroxyl; and a peptide having one or more amino acid S 25 residues including a carboxy terminal amino acid residue, the carboxyl terminal amino acid residue being covalently linked by an ester bond to the 2' hydroxyl of the guanine nucleotide.

The invention also discloses an enzymatic

RNA

molecule comprising a ribonucleotide polymer having a catalytic activity for hydrolyzing an amide substrate to produce an amino cleavage product and a ribozyme amidase intermediate. In one variation, the ribonucleotide polymer has a 5' terminal nucleotide with a ribose sugar having a nucleophilic 2' hydroxyl, and the ribozyme amidase intermediate includes an ester linkage between the nucleophilic 2' hydroxyl and a carboxy group of the amide substrate. In another -13variation, the 5' terminal nucleotide includes a guanine base.

The present invention also discloses enzymatic RNA molecules wherein the amide substrate includes a peptide having one or more amino acid residues including a carboxy terminal amino acid residue bearing the carboxy group of the amide substrate, the carboxy terminal amino acid residue being covalently linked by the ester linkage to the 2' hydroxyl of the ribonucleotide polymer. In an alternative embodiment, the ribonucleotide polymer has an effective binding affinity for the amide substrate and lacks an effective binding affinity for the amino cleavage product. In another variation, the catalytic activity of the 15 ribonucleotide polymer is dependent upon the presence of divalent ions. An alternative embodiment contemplates that an enzymatic RNA molecule as disclosed herein further comprises a cofactor bound to the ribonucleotide polymer, the cofactor including a 20 guanine nucleotide having a ribose sugar with a nucleophilic 2' hydroxyl capable of forming an acid labile ester intermediate with the carboxy cleavage product.

The present invention also contemplates various S 25 methods of making and using enzymatic RNA molecules according to the present invention. In one embodiment, a method of selecting an enzymatic RNA molecule that cleaves amide bonds, comprising the following consecutive steps: obtaining a population of ribozymes; admixing amide bond-containing substrate molecules with the population of ribozymes to form an admixture; maintaining the admixture for a sufficient period of time and under predetermined reaction conditions to allow the ribozymes and the substrate to interact and form ribozyme-product complexes; isolating any ribozyme-product complexes that form; allowing the ribozyme-product complex to dissociate into separate ribozyme and product; and (f) -14separating the ribozymes from the product.

In other variations of- the aforementioned method, the substrate is tagged with an immobilizing agent. In one embodiment, the agent comprises biotin. In another embodiment, a solid surface incorporated or tagged with avidin is utilized to assist in the process of isolating ribozyme-product complexes. For example, the isolating step may further comprise exposing the ribozyme-product complex to a solid surface having avidin linked thereto, whereby the complex becomes attached to the solid surface.

The present invention further contemplates methods of cleaving an amide bond. In one variation, the method comprises admixing an enzymatic RNA molecule 15 with an amide bond-containing substrate, to form a reaction admixture, and maintaining the admixture Indr predetermined reaction conditions to allow the enzymatic RNA molecule to cleave the amide bond. In an ee. alternative embodiment, the enzymatic RNA molecule is 20 able to cleave an amide bond at a preselected site.

Methods of cleaving amide bonds as disclosed herein may also comprise the steps of separating the products from the enzymatic RNA molecule; and adding additional substrate to the enzymatic RNA molecule to form a new 25 reaction admixture.

Also contemplated herein are methods of engineering enzymatic RNA molecules that cleave amide bonds. In one embodiment, the method comprises the following steps: obtaining a population of ribozymes; introducing genetic variation into the population to produce a variant population; (c) selecting individuals from the variant population that meet predetermined selection criteria; separating the selected individuals from the remainder of the variant population; and amplifying the selected individuals.

SIn another variation, methods of catalytically hydrolyzing an amide substrate are contemplated. In one embodiment, the method comprises the following step A: contacting the amide substrate with a ribozyme comprising a ribonucleotide polymer having a catalytic activity for hydrolyzing the amide substrate and producing an amino cleavage product and a ribozyme amidase intermediate, the ribozyme amidase intermediate including a carboxyl of the amide substrate bonded by an ester bond to a 2' hydroxyl of a ribose sugar on a terminal nucleotide of the ribonucleotide polymer.

In another variation, the method further comprises step B as follows, to be performed after Step A: hydrolyzing the ester bond of the ribozyme amidase intermediate to produce a carboxy cleavage product.

~In another embodiment, the method further 15 comprises providing the enzymatic RNA molecule in a reaction medium at a concentration sufficient to cause cleavage of about one molecule of substrate per molecule of enzymatic RNA per minute. In yet another embodiment, the method further comprises providing the enzymatic RNA molecule in a reaction medium, wherein the enzymatic RNA molecule is present at a concentration sufficient to cause cleavage of at least 10% of a population of substrate molecules in an hour.

The invention also contemplates a method of producing an enzymatic RNA molecule having a S"predetermined catalytic activity, comprising the following steps: subjecting a population of enzymatic RNA molecules to mutagenizing conditions to produce a diverse population of mutant RNA molecules; selecting an enzymatic RNA molecule having a predetermined activity from the diverse population of mutant enzymatic RNA molecules; and separating the RNA molecule from the diverse population of mutant RNA molecules. In various embodiments, the predetermined activity comprises the ability to cleave amide or peptide bonds.

In one alternative method, the mutagenizing conditions comprise conditions that introduce defined -16or random nucleotide substitutions within an enzymatic RNA molecule. In another variation, the mutagenizing conditions comprise chemical modification, incorporation of randomized mutagenic oligodeoxynucleotides, or inaccurate copying by a polymerase. In yet another variation, the mutagenizing conditions comprise use of site-directed mutagenesis, polymerase chain reaction (PCR), mutagenic PCR, or self-sustained sequence replication.

Another variation of the foregoing methods further comprises the step of amplifying the enzymatic RNA molecules selected from the diverse population. In one embodiment, the amplifying is performed using a polymerase chain reaction, preferably a mutagenic 15 polymerase chain reaction. In another embodiment, the amplifying is perfrme using self-sustained seuence replication.

The present invention also discloses various compositions. In one embodiment, a composition including an enzymatic RNA molecule that cleaves amide bonds is disclosed. In another variation, a composition including an enzymatic RNA molecule comprising a ribonucleotide polymer having a catalytic activity for hydrolyzing an amide substrate to produce 25 an amino cleavage product and a ribozyme amidase intermediate is disclosed. In another embodiment, a composition further comprises a cofactor bound to the ribonucleotide polymer, the cofactor including a guanine nucleotide having a ribose sugar with a nucleophilic 2' hydroxyl capable of forming an acid labile ester intermediate with the carboxy cleavage product.

Also contemplated by the within invention are compositions comprising two or more populations of enzymatic RNA molecules having characteristics as disclosed herein and in the claims. In another variation, each population of enzymatic RNA molecules in the composition is capable of recognizing a -17different substrate.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 illustrates the secondary structure of the wild-type Tetrahymena ribozyme (SEQ ID NO 1).

Paired structural elements are designated by the symbol where represents a number or an alphanumeric symbol. Joining regions between paired elements i and j, referred to as J i/j, are not labeled. Nucleotide positions that were partially randomized in the initial population are indicated by shaded regions (also see Fig. The internal guide sequence (IGS) is shown in bold, and the DNA substrate is shown in lowercase letters. Nucleotide positions discussed in the text are labeled.

15 Figures 2A-2C illustrate the general procedure for selective amplification of catalytic RNA. In Figure 2A, the overall procedure for RNA amplification is shown. "RT" reverse transcriptase; "T7 pol" T7 polymerase; "prom" promoter, and "RNA" represents the 20 enzymatic RNA molecule.

In Figure 2B, the procedure for selective amplification based on phosphoester transfer activity of a group I ribozyme is shown. represents the enzymatic RNA molecule; represents substrate; "E-S" S 25 represents enzyme/substrate complex; and "EP" Srepresents enzyme/product complex.

SFigure 2C illustrates the overall in vitro evolution procedure disclosed herein. Step 1 Cleavage of the DNA substrate via phosphoester transfer results in ligation of the 3' portion of the substrate to the 3' end of the ribozyme. Step 2 Selective isothermal amplification of DNA-cleaving ribozymes: first, selective Primer la hybridizes to the extended 3' terminus of active molecules and initiates cDNA synthesis in the presence of reverse transcriptase next, Primer 2, which contains a T7 promoter sequence (T7 Prom), hybridizes to the cDNA and initiates second-strand DNA synthesis; finally, T7 RNA -18polymerase (T7 pol) produces multiple copies of the selected RNA, each of which can enter a new round of amplification. Step 3 Selective cDNA synthesis employing Primer la and reverse transcriptase. Step 4 PCR amplification employing nonselective Primer lb and Primer 2, restores the original terminus of the ribozyme-encoding gene and introduces occasional mutations. Step 5 In vitro transcription to produce the progeny population of ribozymes.

Figures 3A and 3B illustrate the secondary structure of the Tetrahymena ribozyme (L-21 form).

Figure 3A is a diagrammatic representation of the secondary structure of the Tetrahymena ribozyme (L-21 form). Figure 3B is a similar diagram of the L-21 form 15 of Tetrahymena ribozyme; the diagram shows those regions that wre randomly mutagenized (boxed segments), as described herein.

Figure 4 illustrates the course of evolution over 1 0 successive generations, highlighting changes in RNA S 20 population size over time. Closed circles represent RNA population size after transcription, quantitated by 3 Huracil content; open circles represent

RNA

population size at the start of each generation, based on 20-pmol portions; closed squares represent

RNA

population size after reaction with substrate, estimated by the assay described in subsection 4 herein; and open squares represent RNA population size after selective amplification, quantitated by acid precipitation at 4 0 C of [a- 3 2 P]GTP-labeled progeny

RNA.

Figure 5 illustrates the cleavage of [3'-nP]dAlabeled d(GGCCCCTT-A 3

(TA

3 3 [5 -32P]A) (SEQ ID NO 26).

Cleavage of 32 P]dA-labeled d(GGCCCTCT-A 3

(TA

3 3 (SED ID NO 26) was conducted under reaction conditions as described herein prior to autoradiogram.

Substrate enzyme/product and product

(P)

were separated by electrophoresis in a polyacrylamide-8M urea gel. Individual bands were cut -19from the gel and quantitated by Cerenkov counting.

Data points are the average of five replicate experiments performed on three different days with two different preparations of substrate. Error bars correspond to +1 SD.

Figures 6A and 6B illustrate Eadie-Hofstee plots used to determine Km (negative slope) and (yintercept) for cleavage of 32 P)-labeled d(GGCCCTCT-

A

3

(TA

3 3 (SEQ ID NO 17) by wild-type ribozymes and clones 29 and 23 from generation 9. Closed circles represent the wild-type; closed squares represent clone 29; and closed triangles represent clone 23. Each data point is the average of three independent determinations of initial velocity. The extent of the 15 reaction was linear over the chosen time interval (rmi,, 0.94, rav, 0.99).

Fingure 7 illustrates sites at which mutations occurred over the course of evolution, superimposed on the secondary structure of the 20 Tetrahymena ribozyme. Box height corresponds to the frequency of mutations at each nucleotide position, based on 50 subclones sequenced at generations 9 (G9; Fig. 7A), 18 (G18; Fig. 7B), and 27 (G27; Fig. 7C).

Non-mutable primer binding sites are shaded; substrate is shown in black. Commonly-occurring mutations (>30 frequency) are labeled.

"Figure 8 also illustrates sites at which mutations occurred over the course of evolution, superimposed on the secondary structure of the Tetrahymena ribozyme.

Box height corresponds to the frequency of mutations at each nucleotide position, based on 50 subclones sequenced at generation 36. Non-mutable primer binding sites are shaded; substrate is shown in black.

Commonly-occurring mutations (>30 frequency) are labeled (dark bars).

Figures 9A and 9B illustrate the improvement in substrate binding affinity over 27 successive generations of in vitro evolution. Fig. 9A represents a typical binding curve showing data obtained for the G27 population of ribozymes. Data from two different gel-shift experiments are indicated. Data was fit by a least squares method to a theoretical binding curve (indicated by solid line), given by the equation: y KD), where y is the fraction of product (P) bound to ribozyme In this case, KD= 51 2) nM.

Fig. 9B shows the KD for the population of ribozymes at every third generation. Standard errors averaged 11%.

Figure 10 illustrates the cleavage of an amide bond-containing substrate by a ribozyme (with Mg 2 present), showing that it generates a 5' product that carries a terminal amine and a 3' product that carries a terminal carboxyl.

15 Figures 11A-C further illustrate the reaction shown in Figure 10, including the production of intermediates (Fig. 11B) and products (Fig. 11C), as well as the relationship of the substrate to the ribozyme (Fig. 11A). It is also shown in Fig. 11C that the ribozyme-associated product is subsequently hydrolyzed, resulting in generation of a 5' product carrying a terminal amine and a 3' product carrying a terminal carboxyl. Subsequent to hydrolysis of the ribozyme-associated product, the enzyme is free to 25 cycle.

SFigure 12A is a photograph of a gel illustrating the confirmation of successful synthesis of the oligonucleotide-oligopeptide "hybrid". In lane 1, labeled d(GGCCCTCTN) is shown. In lanes 2 and 3, labeled d(GGCCCTCT)-Arg is shown, as measured at 30 and minutes.

Figure 12B is a photograph of a gel illustrating cleavage of a hybrid oligonucleotide-oligopeptide substrate by enzymatic RNA molecules of the present invention. In lane 1, 5'-labeled 8-mer marker is shown. In lane 2, interaction of ribozyme wit- a labeled hybrid substrate generates an 8-mer 5' product -21with a terminal -NH 2 In lane 3, substrate alone in the absence of ribozyme) is shown.

DETAILED DESCRIPTION OF THE INVENTION A. Definitions As used herein, the term "amino acid residue" generally means an amino acid formed upon chemical digestion (hydrolysis) of a polypeptide at its peptide linkages. The amino acid residues described herein are preferably in the isomeric form. However, residues in the isomeric form can be substituted for any L-amino acid residue, as long as the desired functional property is retained by the polypeptide. NH 2 refers to the free amino group present at the amino terminus of a polypeptide. COOH refers to the free carboxy group 15 present at the carboxy terminus of a polypeptide. In keeping with standard polypeptide nomenclature (described in J. Biol. Chem. 243: 3552-59 (1969) and adopted at 37 C.F.R. §1.822(b) abbreviations for amino acid residues are shown in the following Table of Correspondence: oo o go *b 1-Letter

Y

G

F

M

A

S

I

L

T

V

P

K

H

Q

TABLE OF CORRESPONDENCE SYMBOL AP 3-Letter Tyr tyi Gly gly Phe phe Met met Ala alz Ser sei Ile isc Leu le.

Thr th Val va] Pro prc Lys lyI His hiE Gln gl INO ACID osine 'cine nylalanine thionine nine rine oleucine icine reonine Line )line sine stidine utamine -22- E Glu glutamic acid Z Glx Glu and/or Gin W Trp tryptophan R Arg arginine D Asp aspartic acid N Asn asparagine B Asx Asn and/or Asp C Cys cysteine X Xaa Unknown or other It should be noted that all amino acid residue sequences represented herein by formulae have a left to right orientation in the conventional direction of amino-terminus to carboxy-terminus. In addition, the 15 phrase "amino acid residue" is broadly defined to include the amin acids list in the Table of Correspondence and modified and unusual amino acids, such as those listed in 37 C.F.R. §1.822(b) and incorporated herein by reference. Furthermore, it should be noted that a dash at the beginning or end of an amino acid residue sequence indicates a peptide bond to a further sequence of one or more amino acid residues or to an amino-terminal group such as NH 2 or to a carboxy-terminal group such as COOH.

25 The term "conservative substitution" as used S* "herein is meant to denote that one amino acid residue has been replaced by another, biologically similar residue. Examples of conservative substitutions include the substitution of one hydrophobic residue such as Ile, Val, Leu or Met for another, or the substitution of one polar residue for another such as between Arg and Lys, between Glu and Asp or between Gln and Asn, and the like. The term "conservative substitution" also includes the use of a substituted amino acid in place of an unsubstituted parent amino acid provided that such a polypeptide also displays the requisite'binding activity.

The term "correspond" in its various grammatical -23forms is used herein and in the claims in relation to polypeptide sequences to mean the polypeptide sequence described plus -or minus up to three amino acid residues at either or both of the amino- and carboxy-termini and containing only conservative substitutions in particular amino acid residues along the polypeptide sequence.

As used herein, "polypeptide" and "peptide" are terms used interchangeably herein to designate a series of no more than about 50 amino acid residues connected one to the other by peptide bonds between the alphaamino and carboxy groups of adjacent residues.

As used herein, the terms "peptide bond" and "amide bond" may also be used interchangeably, and 15 include amide linkages such as those typically found within polypeptides or proteins. As used herein, it is not necessary that an amide bond or peptide bond link adjacent amino acid residues only; for example, peptide bonds/amide bonds as described herein may be found 20 linking adjacent nucleotides, adjacent amino acids, or linking an amino acid to a nucleotide. The term(s) may also be considered to encompass linkages akin to those including unactivated alkyl amides, as opposed to activated aryl amides.

"Protein" is a term generally used herein to S" designate a series of greater than 50 amino acid residues connected one to the other as in a polypeptide.

The term "sequential subset", as used herein, refers to the fact that a polynucleotide or polypeptide has a nucleotide sequence or an amino acid residue sequence, respectively, corresponding to that of a subset of the sequence of a larger nucleotide or protein (or polypeptide) molecule, respectively. For example, if "ABCDEFGH" represented an amino acid residue sequence of a polypeptide, exemplary sequential subsets thereof would include "ABC", "BCDE", "DEFGH", "ABCDEFG", and so forth.

-24- As used herein, the term "ribozyme" is used to describe an RNA-containing nucleic acid that is capable of functioning as an enzyme. In the present disclosure, the term "ribozyme" includes endoribonucleases and endodeoxyribonucleases of the present invention. The term "ribozyme" also encompasses amide bond- and peptide bond-cleaving nucleic acid enzymes of the present invention. Other terms used interchangeably with ribozyme herein include "enzymatic RNA molecule" and "catalytic RNA molecule", which should be understood to include ribozymes and w* enzymatically active portions thereof, whether derived from Tetrahymena or from other organisms or sources.

The terms "ribozyme", "enzymatic RNA molecule", 15 "enzymatic nucleic acid", and "catalytic RNA molecule" should all be unrood t encompass the enzymatically active molecules of the present invention, whether those molecules are described as endoribonucleases, endodeoxyribonucleases, peptidases, amide-cleaving molecules, or some other equivalent description.

As disclosed herein, the foregoing terms may all be *so used to describe an RNA- and/or DNA-containing nucleic acid that is capable of functioning as an enzyme.

The term "enzymatic RNA molecules" includes

RNA

25 molecules which have complementarity in a substratebinding region to a specified oligonucleotide target or substrate; it also has an enzymatic activity which is active to specifically cleave the oligonucleotide substrate. Stated in another fashion, the enzymatic RNA molecule is capable of cleaving the oligonucleotide substrate intermolecularly. This complementarity functions to allow sufficient hybridization of the enzymatic RNA molecule to the substrate oligonucleotide to allow the intermolecular cleavage of the substrate to occur. While one-hundred percent (100%) complementarity is preferred, complementarity in the range of 75-100%, or 50-100%, is also useful and contemplated by the present invention.

Enzymatic RNA molecules of the present invention may alternatively be described as having amidecleaving, amide bond-cleaving, amidase, peptidase, or protease activity. These terms may be used interchangeably herein.

The term "enzymatic nucleic acid" as used herein encompasses enzymatic RNA or DNA molecules, enzymatic RNA-DNA polymers, and enzymatically active portions or derivatives thereof, although enzymatic RNA molecules are a particularly preferred class of enzymatically Sactive molecules according to the present invention.

The term "endodeoxyribonuclease", as used herein, refers to an enzyme capable of cleaving a substrate comprised predominantly of DNA. The term 15 "endoribonuclease", as used herein, refers to an enzyme capable of cleaving a substrate comprised predominantly of RNA.

As used herein, the term "base pair" (bp) is generally used to describe a partnership of adenine (A) 20 with thymine or uracil or of cytosine with guanine although it should be appreciated that less-common analogs of the bases A, T, C, and G may occasionally participate in base pairings. Nucleotides that normally pair up when DNA or RNA adopts a double- 25 stranded configuration may also be referred to herein as "complementary bases".

"Complementary nucleotide sequence" generally refers to a sequence of nucleotides in a singlestranded molecule of DNA or RNA that is sufficiently complementary to that of another single strand to specifically hybridize to it with consequent hydrogen bonding.

"Nucleotide" generally refers to a monomeric unit of DNA or RNA consisting of a sugar moiety (pentose), a phosphate group, and a nitrogenous heterocyclic base.

The base is linked to the sugar moiety via the glycosidic carbon carbon of the pentose) and that combination of base and sugar is a "nucleoside". When -26the nucleoside contains a phosphate group bonded to the 3' or 5' position of the pentose, it is referred to as a nucleotide. A sequence of operatively linked nucleotides is typically referred to herein as a "base sequence" or "nucleotide sequence" (and their grammatical equivalents) and is represented herein by a formula whose left to right orientation is in the conventional direction of 5'-terminus to 3 '-terminus, urless otherwise specified.

1 0 "Nucleotide analog" generally refers to a purine or pyrimidine nucleotide that differs structurally from A, T, G, C, or U, but is sufficiently similar to substitute for the common or "normal" nucleotide in a nucleic acid molecule. As used herein, the term "nucleotide analog" encompasses altered bases, different sugars, or a combination of the two. A listing of exemplary analogs wherein the base has been altered is provided in Section C hereinbelow; in Example 5, a nucleotide analog including an arabinose S 20 sugar is described.

"Oligonucleotide or polynucleotide" generally refers to a polymer of single- or double-stranded nucleotides. As used herein, "oligonucleotide" and its grammatical equivalents will include the full range of nucleic acids. An oligonucleotide will typically refer to a nucleic acid molecule comprised of a linear strand of ribonucleotides. The exact size will depend on many factors, which in turn depends on the ultimate conditions of use, as is well known in the art.

As used herein, the term "physiologic conditions" is meant to suggest reaction conditions emulating those found in mammalian organisms, particularly humans.

While variables such as temperature, availability of cations, and pH ranges may vary as described in greater detail below, "physiologic conditions" generally comprise a temperature of about 35-40 0 C, with 37 0

C

being particularly preferred, as well as a pH of about 7.0-8.0, with 7.5 being particularly preferred, and -27further comprise the availability of cations, preferably divalent or monovalent cations, in a concentration of about 5-15 mM, with a concentration of about 10 mM being particularly preferred. "Physiologic conditions", as used herein, may optionally include the presence of polyamine or free GoN. As noted previously, preferred conditions are described in greater detail below.

B. Enzymatic Nucleic Acid Molecules Some genes have their coding sequences interrupted by stretches of non-coding DNA. These non-coding sequences are generally termed introns. To produce a mature transcript from these genes, the primary RNA transcript (precursor RNA) must undergo a cleavage- 15 ligation reaction termed RNA splicing. This RNA splicing produces the mature transcript of the polypeptide coding messenger RNA (mRNA), ribosomal RNA, or transfer RNA (tRNA). Introns are grouped into four categories (groups I, II, III, and IV) based on their 20 structure and the type of splicing reaction they undergo.

RNA molecules capable of cleaving other RNA molecules have recently been described. Such RNA- S 2 cleaving RNA molecules, which may also be referred to as ribozymes or enzymatic RNA molecules, may be chosen from group I, II, III, or IV introns, with group I and II introns being of greatest interest. Other enzymatic RNA molecules of interest herein are those formed in ribozyme motifs known in the art as "hammerhead" and "hairpin". Enzymatic RNA molecules of interest herein also include hepatitis delta virus ribozymes and RNaseP or RNaseP-like RNA.

Of particular interest to the present invention are the group I introns. Group I introns undergo an intra-molecular RNA splicing reaction leading to cyclization that does not require protein cofactors, Cech, Science 236: 1532-1539 (1987). (The disclosures of all references cited within this application are -28incorporated by reference herein, where appropriate.) The group I introns, including the intron isolated from the large ribosomal RNA precursor of Tetrahymena thermophila, catalyze a sequence-specific phosphoester transfer reaction involving RNA substrates. (See, Zaug and Cech, Science 229: 1060-1064 (1985); and Kay and Inoue, Nature 327: 343-346 (1987).) This sequence-specific phosphoester transfer reaction leads to the removal of the group I intron from the precursor 10 RNA and to ligation of two exons in a process known as RNA splicing. The splicing reaction catalyzed by group I introns proceeds via a two-step transesterification mechanism. The details of this reaction have been reviewed by Cech, Science 236: 1532-1539 (1987).

The splicing reaction of group I introns is initiated by the binding of gua ine or a guanosine Uilv~ .ox a guanosine nucleotide to a site within the group I intron structure. Attack at the 5' splice site by the 3'hydroxyl group of guanosine results in the covalent 20 linkage of guanosine to the 5' end of the intervening intron sequence. This reaction generates a new 3'hydroxyl group on the uridine at the 3' terminus of the 5' exon. The 5' exon subsequently attacks the 3' splice site, yielding spliced exons and the full-length linear form of the group I intron.

The linear group I intron usually cyclizes following splicing. Cyclization occurs via a third transesterification reaction, involving attack of the 3'-terminal guanosine at an interval site near the end of the intron. The group I introns also undergo a sequence-specific hydrolysis reaction at the splice site sequences as described by Inoue et al., J. Mol.

Biol. 189: 143-165 (1986). This activity has been used to cleave RNA substrates in a sequence-specific manner (Zaug et al., Nature 324: 429-433 (1986)).

The structure of group I introns has also been reviewed by J. Burke, Gene 73: 273-294 (1988). The structure is characterized by nine base paired regions, -29termed P1-P9. (See, Burke et al., Nucleic Acids Res. 15: 7217-7221 (1987).) The folded structure of the intron is clearly important for the catalytic activity of the group I introns, as evidenced by the loss of catalytic activity under conditions where the intron is denatured. In addition, mutations that disrupt essential base-paired regions of the group I introns result in a loss of catalytic activity. (See, Burke, Gene 73: 273-294 (1988).) Compensatory mutations or second-site mutations that restore basepairing in these regions also restore catalytic activity (see Williamson et al., J. Biol. Chem. 262: 14672-14682 (1987); and Burke, Gene 73: 273-294 (1988)) 15 Several different deletions that remove a large nucleotide segment from the group I introns (Figure 1) without destroying its ability to cleave RNA have been reported (Burke, Gene 73: 273-294 (1988)). However, attempts to combine large deletions have resulted in 20 both active and inactive introns (Joyce et al., Nucleic Acid Res. 17: 7879 (1989)).

The Tetrahymena ribozyme is a self-splicing group I intron derived from the large ribosomal RNA (rRNA) S 2 precursor of Tetrahymena thermophila. Its biological 25 function is to catalyze its own excision from precursor rRNA to produce mature rRNA. This function has been expressed in vitro (Kruger, et al., Cell 31: 147 (1982)) and has been generalized to include various phosphoester transfer reactions involving RNA substrates (Zaug, et al., Science 231: 470 (1986); Kay, et al., Nature 327: 343 (1987); Been, et al., Science 239: 1412 (1988); Woodson, et al., Cell 57: 335 (1989); Doudna, et al., Nature 339: 519 (1989)). For example, the ribozyme has been used as a sequence-specific endoribonuclease (Zaug, et al., Id. (1986); Murphy, et al., PNAS USA 86: 9218 (1989)), a reaction that proceeds with high catalytic efficiency (km/Km 10 8

M

min-') (Herschlag, et al., Biochemistry 29: 10159 (1990)) The within-described examples utilize derivatives of the self-splicing group I intron of Tetrahymena thermophila, a ribozyme that is able to catalyze sequence-specific cleavage of single-stranded RNA via a phosphoester transfer mechanism (Zaug and Cech, Science 231: 470-475 (1986); Zaug et al., Nature 324: 429-433 (1986)), although it is expressly to be understood that the present invention is not limited to these 10 embodiments. The Tetrahymena ribozyme consists of 413 nucleotides and assumes a well-defined secondary and tertiary structure that is responsible for its catalytic activity (Burke, et al., Nucleic Acids Res 15: 7217 (1987); Kim, et al., PNAS USA 84: 8788 (1987); Celander, et al., Science 251: 401 (1991); Michel, et Sal., J. Mol. Biol. 16: 8 (1990). (See Fig. for a general diagram.) Phylogenetic analysis, supported by site-directed mutagenesis and deletion studies, points out a distinction between a conserved catalytic core 20 (comprising about one-third of the molecule) and surrounding stem-loop elements that offer structural support but are not essential for catalytic activity.

(See Davies, et al., Nature 300: 719 (1982); Michel, et al., Biochimie 64: 867 (1982); Michel, et al., EMBO J.

2: 33 (1983); Cech, et al., Gene 73: 259 (1988); Price, et al., Nucl. Acids Res. 13: 1871 (1985); Szostak, et al., Nature 322: 83 (1986); Joyce, et al., Nucl. Acids Res. 15: 9825 (1987); Barfod, et al., Genes Dev. 2: 652 (1988); Joyce, et al., Nucleic Acids Res. 17: 7879 (1989); Couture, et al., J. Mol. Biol. 215: 345 (1990); Beaudry and Joyce, Biochemistry 29: 6534 (1990).) The ribozyme contains a template region, referred to as the "internal guide sequence" (IGS), which lies at the 5' end of the molecule and forms Watson-Crick base pairs with the target RNA substrate. The 3'-OH of guanosine (which may be represented by the symbol GH) including a guanosine residue that lies at the 3' end of the ribozyme, is directed to attack a particular -31phosphoester bond within the ribozyme-bound substrate.

A phosphoester transfer reaction ensues, resulting in cleavage of the substrate at a position immediately downstream from the region of base pairing, and concomitant ligation of the 3' portion of the substrate to the 3' oxygen of the attacking guanosine. The wild-type Tetrahymena ribozyme can cleave a single-stranded DNA substrate with low efficiency under conditions of high magnesium concentration (50 mM MgCl 2 and/or high temperature (50 0 C) (Herschlag and Cech, oNature 344: 405-409 (1990a); Robertson and Joyce, Nature 344: 467-468 (1990)). Under more physiologic conditions 37 0 C, 10 mM MgCl.pH however, the DNA-cleavage reaction is almost undetectable.

15 The Tetrahymena ribozyme can also act as a sequence-specific endodeoxyribonuclease (Robertson and Joyce, Id. (1990)), although the efficiency of DNA cleavage is low (km/Km 200 M' min- 1 determined at 0 C, 10mM MgC1 2 (Herschlag, et al., Nature 344: 405 20 (1990)). The efficiency of RNA-catalyzed DNA cleavage under physiologic conditions is even lower (km/K, 36 min-', determined at 37°C, 10mM MgC1 2 Figure 1 illustrates the secondary structure of the Tetrahymena thermophila pre-rRNA intron, with the S 25 recognition sequence and the core structure that is the most conserved region among group I introns shown in bold. The nomenclature used to denote various structural features is the standard nomenclature (see, Burke et al., Nucleic Acids Res. 15: 7217-7221 (1987). The nine conserved pairing regions, P1-P9, and the various loops are shown. The nucleotide sequence is numbered beginning at the 5' terminus of the molecule.

As illustrated in Figure 1, the recognition site generally spans nucleotides 19 to 27. In the illustrated diagram, which represents the L-21 form of the Tetrahymena ribozyme, the recognition site begins -32at nucleotide 22. Similarly, the first spacer region is generally located at nucleotides 27 to 28 and 94 to the P3[5'] region is located at nucleotides 96 to 103, the second spacer region is located at nucleotides 104 to 106, the first stem loop is located at nucleotides 107 to 214, the second stem loop is located at nucleotides 215 to 258, the third spacer region is located at nucleotides 259 to 261 and the third stem loop is located at nucleotides 262 to 314.

10 Tetrahymena group I introns have been shown to cleave ribonucleotide-containing substrates such as RNA molecules. In one instance, cleavage of an RNA-DNA polymer was reported, with cleavage occurring at the RNA-DNA "junction". (See, Zaug et al., Science 231: 470-475 (1986); Sugimoto et al., Nucleic Acids Res. 17: 355-371 (1989); and Cech, See 23: 1532- 1539 (1987).) However, a DNA segment containing deoxycytosines was shown not to be a cleavage substrate for the Tetrahymena IVS, a group I intron, in Zaug et 20 al., Science 231: 470-475 (1986).

Therefore, the identification, enhancement, modification and use of novel enzymatic RNA molecules o with the ability to cleave a variety of substrates as disclosed herein is a significant and useful development.

1. Amide Bond-Cleaving Molecules The utility of molecules with peptidase or protease activity is well-appreciated. Such molecules are used in products as divergent as medical or pharmaceutical agents, food products, personal care products, and cleaning agents, and in various methods and applications industrial, environmental, medical, and numerous others that take advantage of a molecule's ability to cleave bonds between amino acids.

Thus, the within-disclosed methods and compositions useful for cleaving amide bonds in a variety of substrates are of particular significance.

An enzymatic RNA molecule of the present invention -33may be engineered or "evolved" from a wild-type, RNAcleaving ribozyme via methods which tend to generate either "random" or "non-random" mutations. Examples of methods useful in generating enzymatic RNA molecules that include mutations not normally found in wild-type ribozymes include PCR (polymerase chain reaction), 3SR (self-sustained sequence replication), and sitedirected mutagenesis.

Preferably, enzymatic RNA molecules produced as disclosed herein are capable of cleaving an amide bondcontaining substrate. In one preferred embodiment, the substrate is a polypeptide, although enzymatic RNA molecules capable of cleaving "hybrid" oligonucleotideoligopeptide molecules, or oligonucleotides containing 15 one or more amide bonds, are also contemplated. In another preferred variation, an enzymatic RNA molecule of the present invention is able to cleave amide bonds under physiologic conditions. Many enzymatic RNA molecules of the present invention are also capable of 20 cleaving a single-stranded RNA substrate, DNA substrates, or RNA-DNA hybrid substrates.

An enzymatic RNA molecule of the present invention may comprise RNA, modified RNA, RNA-DNA polymer, a 2 modified RNA-DNA polymer, a modified DNA-RNA polymer or a modified RNA-modified DNA polymer. RNA contains nucleotides comprising a ribose sugar and adenine, guanine, uracil or cytosine as the base at the 1' position. Modified RNA contains nucleotides comprising a ribose sugar and adenine, thymine, guanine or cytosine and optionally uracil as the base. An RNA-DNA polymer contains nucleotides containing a ribose sugar and nucleotides containing deoxyribose sugar and adenine, thymine and/or uracil, guanine or cytosine as the base attached to the 1' carbon of the sugar. A modified RNA-DNA polymer is comprised of modified RNA, DNA and optionally RNA (as distinguished from modified RNA). Modified DNA contains nucleotides containing a deoxyribose or arabinose sugar and nucleotides -34containing adenine, uracil, guanine, cytosine and possibly thymine as the base. A modified

DNA-RNA

polymer contains modified DNA, RNA and optionally

DNA.

A modified RNA-modified DNA polymer contains modified RNA-modified DNA, and optionally RNA and DNA.

An enzymatic RNA molecule of the present invention is capable of cleaving an amide bond at a predetermined site. An enzymatic RNA molecule of this invention may also be characterized by a nucleotide sequence defining 10 a recognition site that is contiguous or adjacent to the 5' terminus of the nucleotide sequence, a first "spacer region located 3 '-terminal to the recognition site, a P315'] region located 3 '-terminal to the first S..o0 spacer region, a second spacer region located 3'terminal to the P3[5'] region, a first stem loop located 3-terminal to the second spacer region, a second stem loop located 3'-terminal to the first stem loop, a third spacer region located 3 -'terminal to the second stem loop, and a third stem loop located 3'- 20 terminal to the third spacer region, the third stem loop comprising a 5' stem portion defining a P3[3'] region capable of hybridizing to the P3[5'] region.

It is also to be understood that an enzymatic RNA molecule of the present invention may comprise enzymatically active portions of a ribozyme or may comprise a ribozyme with one or more mutations, e.g., with one or more loops or spacers absent or modified, as long as such deletions, additions or modifications do not adversely impact the molecule's ability to perform as an enzyme.

The recognition site of an enzymatic RNA molecule of the present invention typically contains a sequence of at least 2 to about 12 bases, preferably about 4 to about 8 bases, which are capable of hybridizing to a complementary sequence of bases within a "hybrid" oligonucleotide-oligopeptide substrate or to a specific sequence of amino acids, thus giving the enzymatic

RNA

molecule its high sequence specificity. For example, an enzymatic RNA molecule of the present invention constructed with a recognition site base sequence of would be able to recognize the base sequence 5'-CCCTC-3' present within an oligodeoxynucleotide sequence in a hybrid substrate and to cleave the substrate molecule at a predetermined site (see, Example Similarly, an enzymatic RNA molecule with a recognition sequence of will recognize the target sequence 5'-AGCGG-3' in an oligoribonucleotide sequence in a hybrid substrate.

This same recognition site also allows the enzymatic RNA molecule to cleave hybrid or polypeptide substrates with high sequence specificity.

Modification or mutation of the recognition site via well-known methods allows one to alter the sequence specificity of an enzymatic nucleic acid molecule.

For example, a preferred method of modifying or mutating the recognition site (or other sites of the molecule) is described by Cadwell and Joyce, in PCR 20 Methods and Applications 2: 28-33 (1992). (Also see Cadwell and Joyce, PCR Methods and Applications 3 (Suppl.): S136-S140 (1994).) According to this modified PCR method, random point mutations may be introduced into cloned genes. The method has been 25 used to mutagenize the gene encoding the ribozyme with a mutation rate of 0.66% 0.13% (95% confidence interval) per position per PCR, as determined by sequence analysis, with no strong preferences observed with respect to the type of base substitution. This 30 allows the introduction of random mutations at any position in the molecule. Another available method useful in introducing defined or random mutations is described in Joyce and Inoue, Nucleic Acids Research 17: 711-722 (1989). The modified PCR method of Cadwell and Joyce is, nevertheless, particularly preferred for use as described herein.

Enzymatic nucleic acid molecules of the present invention include those with altered recognition sites.

-36- In various embodiments, these altered recognition sites confer unique sequence specificities on the enzymatic nucleic acid molecule including such recognition sites.

The exact bases present in the recognition site are important in determining the base sequence or amino acid residue sequence that is recognized by the enzymatic RNA molecule, as well as the site at which cleavage will take place. It should be appreciated, however, that other sequences and sites in the enzymatic RNA molecules of the present invention may participate in the recognition-and-cleavage process.

Amino acid residue sequences and conformations of the substrate molecules may also affect this process.

Cleavage of the substrate preferably occurs immediately 3' of the substrate cleavage sequence, the substrate oligomer sequence that associates with the recognition site. For example, if the substrate is an oligonucleotide, this cleavage leaves a 3' hydroxyl group on the substrate cleavage sequence and a 20 phosphate on the nucleotide that was originally immediately 3' of the substrate cleavage sequence in the.original substrate. If, on the other hand, the substrate is (or includes) an amino acid residue sequence, cleavage leaves a 3' amino group on the substrate cleavage sequence and a 5' carboxyl group on the amino acid that was originally immediately 3' of the substrate cleavage sequence in the original substrate. Cleavage can be redirected to a site of 0 choice by changing the bases present in the recognition sequence/internal guide sequence (see Murphy et al., PNAS USA 86: 9218-9222 (1989)) and/or in other sites and sequences of the enzymatic RNA molecule.

The recognition site may also be provided as a separate nucleic acid, an external recognition site not covalently coupled to the rest of the enzymatic

RNA

molecule. External recognition sites may direct cleavage at a specific amino acid or base sequence (see, Doudna et al., Nature 339: 519-522 (1989)).

-37- If an external recognition site is used, the enzymatic RNA molecule used with it would probably not contain a recognition site but would tend to comprise a region, a second spacer region, a first stem loop, a second stem loop, a third spacer region and a third stem loop where the third stem loop comprises a 5' stem portion defining a P3[3'] region capable of hybridizing to said P3[5'] region.

Use of an enzymatic RNA molecule of the present invention with an external recognition site allows the target sequence to be altered by merely changing the external recognition site sequence. Use of a plurality of different external recognition sequences with an enzymatic RNA molecule of the present invention allows the substrate to be cleaved at each of the different residue sequences encoded by the external recognition sequences.

First spacer regions typically contain a sequence of nucleotides of about 3 to 7, preferably about 20 nucleotides in length. In one variation, the nucleotides making up the first spacer have the sequence 5'-NNNNA-3' (SEQ ID NO where N represents the presence of any nucleotide at that position. In i another variation, the first spacer region is defined 25 by the sequence 5'-AACAA-3' (SEQ ID NO 3).

In other embodiments, the first spacer region is comprised of a nucleotide sequence defining two spacer stem loops. In one variation, the first spacer stem loop is 25 nucleotides in length, and the second spacer stem loop is 36 nucleotides in length. In another variation, the first spacer stem loop has the base sequence 5'-AGUUACCAGGCAUGCACCUGGUAGUCA-3' (SEQ ID NO or is as shown in Figure 1. In yet another variation, the second spacer stem loop has the base sequence 5'-GUCUUUAAACCAAUAGAUUGGAUCGGUUUAAAAGGC-3' (SEQ ID NO or is as shown in Figure 1.

As noted previously, the foregoing descriptions of loop and spacer regions are exemplary and are not to be -38construed as limiting the disclosed invention(s) A stem loop is a secondary structure formed by a nucleotide sequence that has "folded over on itself".

A stem loop comprises a 5' nucleotide sequence portion, designated a 5' paring segment that is capable of hybridizing to a nucleotide sequence located 3' of the and is designated the 3' pairing segment In a stem loop, the and are connected by a nucleotide sequence called a loop. The and hybridize and form a nucleic acid duplex. The nucleic acid duplex formed by the and does not have to be a perfect duplex and may contain stretches of nucleotides that are either unpaired or paired to a sequence outside the stem loop.

In various alternative embodiments, an enzymatic RNA molecule of th present invention has an enhanced or optimized ability to cleave amino acid substrates.

As those of skill in the art will appreciate, the rate of an enzyme-catalyzed reaction varies depending upon 20 the substrate and enzyme concentrations and, in general, levels off at high substrate or enzyme concentrations. Taking such effects into account, the kinetics of an enzyme-catalyzed reaction may be described in the following terms, which define the reaction.

The enhanced or optimized ability of an enzymatic RNA molecule of the present invention to cleave a dipeptide or Polypeptide substrate may be determined in a cleavage reaction with varying amounts of labeled 30 peptide substrate in the presence of enzymatic

RNA

molecules as described in Examples 1 and 4. The ability to cleave the substrate is generally defined by the catalytic rate divided by the Michaelis constant The symbol k. represents the maximal velocity of an enzyme reaction when the substrate approaches a saturation value. KM represents the substrate concentration at which the reaction rate is one-half maximal. Values for KM and k. are determined -39in this invention by experiments in which the substrate concentration is in excess over enzymatic RNA molecule concentration Initial rates of reaction (v 0 over a range of substrate concentrations were estimated from the initial linear phase, generally the first 5% or less of the reaction. Typically, eight data points ware fit by a least squares method to a theoretical line given by the equation: v Vm=. Thus, k. and KM are determined by the initial rate of reaction, v 0 and the substrate concentration

IS]

In one embodiment of the present invention, an enzymatic RNA molecule of the present invention exhibits amide bond-cleaving activity not normally found in wild-type ribozymes. In various alternative embodiments, an enzymatic RNA molecule of the present inrntion as an ennced or optimized ability to cleave amino acid substrates, preferably dipeptide or polypeptide substrates. One skilled in the art will 20 appreciate that the enhanced or optimized ability of an enzymatic RNA molecule to cleave amino acid substrates may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention.

25 Enzymatic RNA molecules of the present invention may also be characterized as displaying a KM value that is improved over the wild-type. As noted above, KM represents the substrate concentration at which the reaction rate is one-half maximal; thus, an improved KM 30 indicates an improvement in substrate processing. In various embodiments, enzymatic RNA molecules of the present invention have a KM that is statistically significant when compared with the KM of wild-type ribozymes, the latter of which are apparently unable to cleave amino acid substrates.

One skilled in the art will understand that the enhanced or optimized ability of an enzymatic RNA molecule to process amino acid polypeptide) substrates may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention and may include a reduction of the peptide concentration to favor enzymatic

RNA

molecules with improved substrate processing ability.

In other embodiments, an enzymatic RNA molecule of the present invention has an enhanced or optimized ability to bind an amino acid substrate. The ability of an enzymatic RNA molecule to bind a polypeptide substrate is defined by the dissociation constant

(KD).

The KD is an equilibrium constant describirg the dissociation of the enzymatic RNA molecule:substrate complex into its individual components. The K

D

constant as understood in the context of this invention is determined by a gel-shift analysis to determine the percent enzymatic RNA molecule bound to the amino acid product, as described in the Examples. A binding curve is generated by plotting the percent of product bound 20 to enzymatic RNA molecule over a range of enzymatic

RNA

molecule concentration. The KD is determined by fitting the data to a theoretical binding curve using the least squares method. Typically, the enzymatic RNA molecule concentration vastly exceeds the product; therefore, the theoretical binding curve can be represented by the equation: bound

KD),

where KD when half of the total product is bound to the enzymatic RNA molecule.

An enzymatic RNA molecule of the present invention 30 preferably binds amino acid substrate with a K D which is an improvement over that of wild-type ribozymes.

For example, an enzymatic RNA molecule of the present invention preferably binds peptides with a K D having a value less than 30 AM. In preferred embodiments, enzymatic RNA molecules bind polypeptide with a K

D

having a value less than about 10 AM. In more preferred embodiments, the K

D

of an amino acid-binding -41enzymatic RNA molecule is less than about 1 pM. In an even more preferred embodiment, the KD of an amino acid-binding enzymatic RNA molecule is less than about nM, more preferably less than about 25nM, and even more preferably less than about 10 nM. Especially preferred enzymatic RNA molecules bind peptide substrate with a KD of 5 nM or less, with a K D of about 0.1-5 nM. One skilled in the art will understand that the enhanced or optimized ability of an enzymatic RNA molecule to bind amino acid-containing substrates may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention and may include a reduction of the amino acid concentration to favor enzymatic RNA molecules with improved substrate binding affinity.

In another variation, an enzymatic RNA molecule of the present invention has an enhanced or optimized substrate turnover rate. The enhanced or optimized substrate turnover rate may be determined in single- 20 turnover kinetic experiments with the enzymatic RNA molecule in excess of the substrate as described in the following Examples. Initial rates are obtained using no more than the first 5% of the reaction. Given that km/KM kbS/ each kobSvalue, obtained at 25 different enzymatic RNA molecule concentrations, provides an estimate of km/KM. Generally, eight or more measurements of km/KM are obtained. The value of kc in the presence of limited substrate indicates the substrate turnover number rate and is expressed in the 30 number of catalytic cycles that are completed by the enzyme per unit of time under the assay conditions.

One skilled in the art will appreciate that the enhanced or optimized substrate turnover rate of an enzymatic RNA molecule of the present invention may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention and may include a reduction of the reaction -42time to favor enzymatic RNA molecules with improved substrate turnover rates.

In other embodiments, an enzymatic RNA molecule of the present invention is capable of functioning efficiently over a wide range of temperatures. In yet another variation, an enzymatic RNA molecule of the present invention is capable of functioning efficiently over a wide range of pH.

In various alternative embodiments, an enzymatic RNA molecule of the present invention is capable of functioning efficiently in the presence or absence of Mg 2 Alternatively, an enzymatic RNA molecule of the present invention is capable of functioning efficiently in the presence or absence of divalent cations other than Mg 2 Other suitable divalent cations may be selected f r o m group pised of Mn, Zn, or Ca y %_%.dt1Ljj.LL of Zn" or Ca It is anticipated that cation concentrations similar to those described above for Mg 2 will be useful as disclosed herein.

20 Optionally, monovalent cations may also be present as "alternatives" for the use of divalent cations. For example, monovalent cations such as sodium (Na or potassium may be present, either as dissociated ions or in the form of dissociable compounds such as NaC1 or KC1. In one embodiment, a monovalent cation is present in a concentration ranging from about 0-200 mM.

In other embodiments, monovalent cations are present in a concentration ranging from about 2-100 mM.

Alternatively,- the concentration of monovalent cations 30 ranges from about 2 mM 50 mM. In other embodiments, the concentration ranges from about 2 mM 25 mM, with a concentration of about 2 mM 15 mM being preferred.

In various embodiments, an enzymatic RNA molecule of the present invention optionally includes a 3' hydroxyl of G guanosine, or one of its phosphorylated forms), which functions as a nucleophile it "attacks" substrate molecules, particularly -43hybrid substrates, at a phosphodiester or amide bond.

For example, in the L-21 ribozyme derived from the group I intron of Tetrahymena thermophila, the G264- C311 base pair which is known as the "G-site" binds the G substrate. (See, Wang and Cech, Science 256: 526-529 (1992).) Alternatively, in other embodiments, wherein an enzymatic RNA molecule of the present invention lacks a 3' terminal GOH, the GOH may be supplied as a free unattached) attacking group. In such embodiments, an enzymatic RNA molecule is able to "attack" multiple substrates in sequential fashion. In either case, the term "enzymatic RNA molecules" as used in the present disclosure encompasses enzymatic RNA molecules including, as well as those lacking, a 3' GoH.

In various embodiments, an enzymatic RNA molecule of the present invention may combine one or more modifications or mutations including additions, deletions, and substitutions. In alternative 20 embodiments, such mutations or modifications may be generated using methods which produce random or specific mutations or modifications. These mutations may change the length of, or alter the nucleotide sequence of, a stem loop, the the P3[3'] 25 region, a spacer region or the recognition sequence.

One or more mutations within one catalytically active enzymatic RNA molecule may be combined with the mutation(s) within a second catalytically active enzymatic RNA molecule to produce a new enzymatic RNA molecule containing the mutations of both molecules.

S0. In other preferred embodiments, an enzymatic RNA molecule of the present invention may have random mutations introduced into it using a variety of methods well known to those skilled in the art. For example, the method described by Cadwell and Joyce (PCR Methods and Applications 2: 28-33 (1992)) is particularly preferred for use as disclosed herein, as it efficiently introduces random mutations into -44populations of enzymatic RNA molecules. (Also see Cadwell and Joyce, PCR Methods and Applications 3 E(Supp.i: S136-S140 (1994).) According to this modified PCR method, random point mutations may be introduced into cloned genes. The method has been used to mutagenize the gene encoding the ribozyme with a mutation rate of 0.66% 0.13% (95% confidence interval) per position per PCR, as determined by sequence analysis, with no strong preferences observed with respect to the type of base substitution. This allows the introduction of random mutations at any position in the molecule.

Another method is available which is useful in introducing defined or random mutations (see Joyce and Inoue, Nucleic Acids Research 17: 711-722 (1989)).

This latter method involves exision of a template (coding) strand of a double-stranded

DNA,

*o o reconstruction of the template strand with inclusion of mutagenic oligonucleotides, and subsequent 20 transcription of the partially-mismatched template.

This allows the introduction of defined or random mutations at any position in the molecule by including polynucleotides containing known or random nucleotide sequences at selected positions.

Enzymatic RNA molecules of the present invention may be of varying lengths and folding patterns, as appropriate, depending on the type and function of the molecule. For example, enzymatic RNA molecules derived from group I introns Tetrahymena-derived ribozymes) may be about 413 or more nucleotides in length, although a length not exceeding 413 nucleotides is preferred, to avoid limiting the therapeutic usefulness of molecules by making them too large or unwieldy. In various therapeutic applications, enzymatic RNA molecules of the present invention comprise the enzymatically active portions of ribozymes. In various embodiments, enzymatic

RNA

molecules of the present invention comprise fewer than 400 nucleotides, fewer than 300 nucleotides, fewer than 200 nucleotides, or fewer than 100 nucleotides.

In other therapeutic applications, enzymatic RNA molecules such as "hammerhead" ribozymes are preferably no more than about 50 nucleotides in length, with a length of 30-40 nucleotides being particularly preferred. Even more preferred are hammerhead ribozymes of about 31-36 nucleotides in length.

Moreover, if one intends to synthesize molecules for use as disclosed herein, one should bear in mind that the larger the enzymatic nucleic acid molecule is, the more difficult it is to synthesize. Presumably, those of skill in the art will certainly appreciate these design constraints.

oooo 15 Various preferred methods of modifying ribozymes and other enzymatic RNA molecules, ribonucleases, deoxyribonucleases, and amidases of the present invention are further described in the Examples set forth hereinbelow.

20 2. DNA-Cleaving Molecules In other preferred embodiments, enzymatic RNA molecules produced as disclosed herein are capable of cleaving a DNA substrate. In one preferred embodiment, the DNA substrate is single-stranded, although 25 enzymatic RNA molecules capable of cleaving "loop" RNA and DNA nucleic acids found in stem loops and the like) and double-stranded DNA are also contemplated. In another preferred variation, an enzymatic RNA molecule of the present invention is able to cleave DNA under physiologic conditions. Many enzymatic RNA molecules of the present invention are also capable of cleaving a single-stranded RNA substrate or a modified DNA substrate containing a uracil at the cleavage site rather than a thymine, whereas various enzymatic RNA molecules show a marked preference for DNA as the substrate of choice.

As described above, an enzymatic RNA molecule of the present invention may comprise RNA, modified RNA, -46- RNA-DNA polymer, a modified RNA-DNA polymer, a modified DNA-RNA polymer or a modified RNA-modified DNA polymer RNA generally contains nucleotides comprising a ribose sugar and adenine, guanine, uracil or cytosine as the base at the 1' position. Modified RNA contains nucleotides comprising a ribose sugar and adenine, thymine, guanine or cytosine and optionally uracil as the base. An RNA-DNA polymer contains nucleotides containing a ribose sugar and nucleotides containing deoxyribose sugar and adenine, thymine and/or uracil, guanine or cytosine as the base attached to the 1' carbon of the sugar. A modified RNA-DNA polymer is comprised of modified RNA, DNA and optionally RNA (as distinguished from modified RNA). Modified

DNA

15 contains nucleotides containing a deoxyribose sugar and nucleotides containin nine, uracil, guanine, cytosine and possibly thymine as the base. A modified DNA-RNA polymer contains modified DNA, RNA and optionally DNA. A modified RNA-modified DNA polymer contains modified RNA-modified DNA, and optionally

RNA

and DNA.

An enzymatic RNA molecule of the present invention is capable of cleaving DNA 3' of a predetermined base sequence. An enzymatic RNA molecule of this invention 25 may also be characterized by a nucleotide sequence defining a recognition site that is contiguous or adjacent to the 5' terminus of the nucleotide sequence, a first spacer region located 3 '-terminal to the recognition site, a P3[5'] region located 3 '-terminal to the first spacer region, a second spacer region located 3 '-terminal to the P3[5'] region, a first stem loop located 3 '-terminal to the second spacer region, a second stem loop located 3 '-terminal to the first stem loop, a third spacer region located 3 -'terminal to the second stem loop, and a third stem loop located 3'terminal to the third spacer region, the third stem loop comprising a 5' stem portion defining a P313'] region capable of hybridizing to the P3[5'] region.

-47- It is also to be understood that an enzymatic RNA molecule of the present invention may comprise enzymatically active portions of a ribozyme or may comprise a ribozyme with one or more mutations, e.g., with one or more loops or spacers absent or modified, as long as such deletions, additions or modifications do not adversely impact the molecule's ability to perform as an enzyme.

The recognition site of an enzymatic RNA molecule of the present invention typically contains a sequence of at least 2 to about 8 bases, preferably about 4 to about 7 bases, which are capable of hybridizing to a complementary sequence of bases within the substrate nucleic acid giving the enzymatic RNA molecule its high 15 sequence specificity. For example, an enzymatic RNA molecule of the present invention constructed with a ;rpr~an i t nn Ri t R4-mipnrCI of I I -nnrrr- c; I wA R Ahl ito recognize the base sequence 5'-CCCTCT-3' present within a single-stranded DNA substrate and to cleave 20 same (see, Example Similarly, an enzymatic RNA molecule with a recognition sequence of 3'-UCGCCG- 5' was used to cleave the target sequence 5'-AGCGGT-3' This same recognition site also allows the enzymatic RNA molecule to cleave modified DNA 25 substrates with high sequence specificity.

Modification or mutation of the recognition site via well-known methods allows one to alter the sequence specificity of an enzymatic nucleic acid molecule.

Preferred methods are described in subsection 1 immediately above.

As noted previously, enzymatic nucleic acid molecules of the present invention include those with altered recognition sites. In various embodiments, these altered recognition sites confer unique sequence specificities on the enzymatic nucleic acid molecule including such recognition sites.

The exact bases present in the recognition site determine the base sequence at which cleavage will take -48place. Cleavage of the substrate nucleic acid occurs immediately 3' of the substrate cleavage sequence, the substrate nucleotide sequence that hybridizes to the recognition site. This cleavage leaves a 3' hydroxyl group on the substrate cleavage sequence and a phosphate on the nucleotide that was originally immediately 3' of the substrate cleavage sequence in the original substrate. Cleavage can be redirected to a site of choice by changing the bases present in the recognition sequence (internal guide sequence). See Murphy et al., PNAS USA 86: 9218-9222 (1989).

Moreover, in various embodiments of the present invention, any combination of bases may be present in the recognition site if a polyamine is present. See, 15 for example, Doudna et al., Nature 339: 519-522 (1989).

Examples of useful polyamines include spermidine, putrescine or spermine. A spermidine concentration of about 5 mM was shown to be effective in particular embodiments, as further described hereinbelow, although concentrations ranging from about 0.1 mM to about 10 mM may also be useful.

The recognition site may also be provided as a separate nucleic acid, an external recognition site not covalently coupled to the rest of the 25 endodeoxynuclease. External recognition sites may direct endoribonuclease cleavage at a specific base sequence (see, Doudna et al., Nature 339: 519-522 (1989)). If an external recognition site is used, the enzymatic RNA molecule used with it would probably not contain a recognition site but would tend to comprise a region, a second spacer region, a first stem loop, a second stem loop, a third spacer region and a third stem loop where the third stem loop comprises a stem portion defining a P3[3'] region capable of hybridizing to said P3[5'] region.

Use of an enzymatic RNA molecule of the present invention with an external recognition site allows the target sequence to be altered by merely changing the -49external recognition site sequence. Use of a plurality of different external recognition sequences with an enzymatic RNA molecule of the present invention allows the substrate nucleic acid to be cleaved at each of the different base sequences encoded by the external recognition sequences.

First spacer regions typically contain a sequence of nucleotides of about 3 to 7, preferably about bases in length. In one variation, the nucleotides making up the first spacer have the sequence 3' (SEQ ID NO where N represents the presence of any nucleotide at that position. In another variation, the first spacer region is defined by the sequence AACAA-3' (SEQ ID NO 3).

In other embodiments, the first spacer region is comprised of a nucleotide sequence defining two spacer stem loops. In one variation, the first spacer stem loop is 25 nucleotides in length, and the second spacer stem loop is 36 nucleotides in length. In another S 20 variation, the first spacer stem loop has the base sequence 5'-AGUUACCAGGCAUGCACCUGGUAGUCA-3' (SEQ ID NO 4) or is as shown in Fig.l. In yet another variation, the second spacer stem loop has the base sequence GUCUUUAAACACCAAUAGAUUGGAUCGGUUUAAAAGGC-3' (SEQ ID NO 25 or is as shown in Fig.l.

As noted previously, the foregoing descriptions of loop and spacer regions are exemplary and are not to be construed as limiting the disclosed invention(s).

A stem loop is a secondary structure formed by a nucleotide sequence that has "folded over on itself".

A stem loop comprises a 5' nucleotide sequence portion, designated a 5' paring segment that is capable of hybridizing to a nucleotide sequence located 3' of the and is designated the 3' pairing segment In a stem loop, the and are connected by-a nucleotide sequence called a loop. The and hybridize and form a nucleic acid duplex. The nucleic acid duplex formed by the and does not have to be a perfect duplex and mayi contain stretches of nucleotides that are either unpaired or paired to a sequence outside the stem loop.

In various alternative embodiments, an enzymatic RNA molecule of the present invention has an enhanced or optimized ability to cleave nucleic acid substrates, preferably DNA substrates. As those of skill in the art will appreciate, the rate of an enzyme-catalyzed reaction varies depending upon the substrate and enzyme concentrations and, in general, levels off at high substrate or enzyme concentrations. Taking such effects into account, the kinetics of an enzymecatalyzed reaction may be described in the following terms, which define the reaction.

The enhanced or optimized ability of an enzymatic RNA mr0--11- -f he x su invention to cleave a DNA substrate may be determined in a cleavage reaction with S. varying amounts of labeled DNA substrate in the presence of enzymatic RNA molecules as described in 20 Examples 1 and 2. The ability to cleave the substrate is generally defined by the catalytic rate divided by the Michaelis constant The symbol k.

represents the maximal velocity of an enzyme reaction •when the substrate approaches a saturation value.

KM

25 represents the substrate concentration at which the reaction rate is one-half maximal. Values for KM and k, are determined in this invention by experiments in which the substrate concentration is in excess over enzymatic RNA molecule concentration Initial 30 rates of reaction (v 0 over a range of substrate concentrations were estimated from the initial linear phase, generally the first 5% or less of the reaction.

Typically eight data points were fit by a least squares method to a theoretical line given by the equation: v Thus, k, and KM were determined by the initial rate of reaction, v, and the DNA substrate concentration -51- In various alternative embodiments, an enzymatic RNA molecule of the present invention has an enhanced or optimized ability to cleave nucleic acid substrates, preferably DNA substrates. In preferred embodiments, the enhanced or optimized ability of an enzymatic RNA molecule to cleave DNA substrates shows about a 10- to 9 -fold improvement. In more preferred embodiments, an enzymatic RNA molecule of the present invention is able to cleave DNA substrates at a rate that is about 103- to 10 7 -fold better than wild-type enzymes. In even more preferred embodiments, the enhanced or optimized ability to cleave DNA substrates is expressed as a 104to 10 6 -fold improvement over the wild-type. One skilled in the art will appreciate that the enhanced or optimized ability of an enzymatic RNA molecule to cleave nucleic acid substrates may vary depending upon evolution procedure of the invention.

Enzymatic RNA molecules of the present invention 20 may also be characterized as displaying a KM value that eo is improved at least two-fold over the wild-type. As coo S"noted above, KM represents the substrate concentration e ee at which the reaction rate is one-half maximal; thus, an improved KM indicates an improvement in substrate 25 processing. In various embodiments, enzymatic RNA S"molecules of the present invention have a KM that is to 20-fold better than that of the wild-type. In still other embodiments, enzymatic RNA molecules of the present invention have a KM that is 30- to improved over the wild-type. In various other embodiments, enzymatic RNA molecules of the present invention have a KM that is 40- to 50-fold improved over the wild-type.

One skilled in the art will understand that the enhanced or optimized ability of an enzymatic RNA molecule to process nucleic acid DNA) substrates may vary depending upon the selection constraints -52applied during the in vitro evolution procedure of the invention and may include a reduction of the DNA concentration to favor enzymatic RNA molecules with improved substrate processing ability.

In other embodiments, an enzymatic RNA molecule of the present invention has an enhanced or optimized ability to bind a nucleic acid substrate. The ability of an enzymatic RNA molecule to bind a DNA substrate is defined by the dissociation constant The KD is an equilibrium constant describing the dissociation of the enzymatic RNA molecule:substrate complex into its individual components. The KD constant as understood in the context of this invention is determined by a gel-shift analysis to determine the percent enzymatic RNA molecule bound to the DNA product, as described in A biuing curve is generated by plotting the percent of product bound to enzymatic RNA molecule over a range of enzymatic RNA molecule concentration.

The KD is determined by fitting the data to a 20 theoretical binding curve using the least squares method. Typically, the enzymatic RNA molecule concentration vastly exceeds the product; therefore, the theoretical binding curve can be represented by the equation: bound

KD),

25 where KD when half of the total product is bound to the enzymatic RNA molccule.

An enzymatic RNA molecule of the present invention preferably binds nucleic acid substrate with a K which is an improvement over that of wild-type ribozymes.

For example, an enzymatic RNA molecule of the present rinvention preferably binds DNA with a KD having a value less than 30 MM. In preferred embodiments, enzymatic RNA molecules bind DNA with a KD having a value less than about 10 yM. In more preferred embodiments, the

K

D of a DNA-binding enzymatic RNA molecule is less than about 1 uM. In an even more preferred embodiment, the KD of a DNA-binding enzymatic RNA molecule is less than -53about 50 nM, more preferably less than about 25nM, and even more preferably less than about 10 nM. Especially preferred enzymatic RNA molecules-bind DNA substrate with a KD of 5 nM or less, with a KD of about 0.1- 5 nM.

Alternatively, the enhanced or optimized ability of an enzymatic RNA molecule to bind DNA substrates may be expressed as a five-fold or greater improvement over that of the wild-type. In various embodiments, the enhanced or optimized ability of an enzymatic RNA molecule to bind DNA substrates comprises a 10- to 102fold improvement. In other embodiments, the enhanced or optimized ability of an enzymatic RNA molecule to bind DNA substrates may be expressed as a 102- to 103fold improvement over the wild-type. In still other embodiments, binding K

D

of enzymatic RNA molecules of the present invention is 10 4 -fold improved over the wild-type, or better. One skilled in the art will understand that the enhanced or optimized ability S 20 of an enzymatic RNA molecule to bind DNA substrates may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention and may include a reduction of the DNA concentration to favor enzymatic RNA molecules with 25 improved substrate binding affinity.

In other preferred embodiments, enzymatic RNA molecules of the present invention preferably bind RNA substrate with a KD which is an improvement over wildtype ribozymes. For example, an enzymatic RNA molecule of the present invention preferably binds RNA with a K

D

having a value less than 1.5 nM. In preferred embodiments, enzymatic RNA molecules bind RNA with a KD having a value less than about 1.0 nM. In even more preferred embodiments, enzymatic RNA molecules of the present invention bind RNA with a KD of about 0.5 nM or less.

In another variation, an enzymatic RNA molecule of -54the present invention has an enhanced or optimized substrate turnover rate. The enhanced or optimized substrate turnover rate may be determined in singleturnover kinetic experiments with the enzymatic

RNA

molecule in excess of the substrate as described in Examples 1-3. Initial rates (kobS) were obtained using no more than the first 5% of the reaction. Given that km/KM kob/[E], each kOb, value, obtained at different enzymatic RNA molecule concentrations, provided an estimate of km/KM. Generally, eight or more measurements of km/KM were obtained. The value of k., in the presence of limited substrate indicates the substrate turnover number rate and is expressed in the number of catalytic cycles that are completed by the enzyme per unit of time under the assay conditions.

i eatively, the enhanced or optimized substrate turnover rate of an enzymatic RNA molecule may be described as being improved about 2-fold over the wildtype. In other embodiments, the enhanced or optimized o 20 substrate turnover rate of an enzymatic RNA molecule shows at least a 5- to 25-fold improvement over the wild-type. In still other embodiments, the enhanced or optimized substrate turnover rate of an enzymatic

RNA

molecule of the present invention is about 30-40 times greater than that of the wild-type. In preferred embodiments, the substrate turnover rate is at least ~about 50 times greater than in the wild-type. One skilled in the art will understand that the enhanced or optimized substrate turnover rate of an enzymatic

RNA

30 molecule of the present invention may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention and may include a reduction of the reaction time to favor enzymatic RNA molecules with improved substrate turnover rates.

In other embodiments, an enzymatic RNA molecule of the present invention is capable of functioning efficiently over a wide range of temperatures. In yet another variation, an enzymatic RNA molecule of the present invention is capable of functioning efficiently over a wide range of pH.

In various alternative embodiments, an enzymatic RNA molecule of the present invention is capable of functioning efficiently with or without added polyamine. In another variation, an enzymatic RNA molecule of the present invention is capable of functioning efficiently in the presence or absence of Mg 2 Alternatively, an enzymatic RNA molecule of the present invention is capable of functioning efficiently in the presence or absence of divalent cations other than Mg". Other suitable divalent cations may be selected from the group comprised of Mn 2 Zn 2 or Ca 2 It is anticipated that cation concentrations similar to those described above for M 2 will be useful as disclosed herein.

Optionally, monovalent cations may also be present S 20 as "alternatives" for the use of divalent cations. For o" example, monovalent cations such as sodium or *potassium (K may be present, either as dissociated ions or in the form of dissociable compounds such as NaC1 or KC1. In one embodiment, a monovalent cation is S 25 present in a concentration ranging from about 0-200 mM.

In other embodiments, monovalent cations are present in a concentration ranging from about 2-100 mM.

Alternatively, the concentration of monovalent cations ranges from about 2 mM 50 mM. In other embodiments, the concentration ranges from about 2 mM 25 mM, with a concentration of about 2 mM 15 mM being preferred.

In various embodiments, an enzymatic RNA molecule of the present invention optionally includes a 3' hydroxyl of G guanosine, or one of its phosphorylated forms), which functions as a nucleophile it "attacks" substrate molecules, usually at a phosphodiester bond. For example, in the L-21 ribozyme -56derived from the group I intron of Tetrahymena thermophila, the G264-C311 base pair which is known as the "G-site" binds the G substrate. (See, e.g., Wang and Cech, Science 256: 526-529 (1992).) Alternatively, in other embodiments, wherein an enzymatic RNA molecule of the present invention lacks a 3' terminal GOH, the GOH may be supplied as a free unattached) attacking group. In such embodiments, an enzymatic RNA molecule is able to "attack" multiple substrates in sequential fashion. In either case, the term "enzymatic RNA molecules" as used in the present disclosure encompasses enzymatic

RNA

molecules including, as well as those lacking, a 3' GoH.

In various embodiments, an enzymatic RNA molecule of the present invention may combine one or more modifications or mutations including additions, deletions, and substitutions. In alternative embodiments, such mutations or modifications may be generated using methods which produce random or 20 specific mutations or modifications. These mutations may change th" length of, or alter the nucleotide sequence of, a stem loop, the P315'], the P3[3'] region, a spacer region or the recognition sequence.

One or more mutations within one catalytically active 25 enzymatic RNA molecule may be combined with the mutation(s) within a second catalytically active enzymatic RNA molecule to produce a new enzymatic RNA molecule containing the mutations of both molecules.

In other preferred embodiments, an enzymatic

RNA

30 molecule of the present invention may have random or defined mutations introduced into it using a variety of methods well known to those skilled in the art. For example, the method described by Joyce et al., Nucleic Acids Res. 17: 711-712 (1989), involves excision of a template (coding) strand of a double-stranded

DNA,

reconstruction of the template strand with inclusion of mutagenic oligonucleotides, and subsequent transcription of the partially-mismatched template.

-57- This allows the introduction of defined or random mutations at any position in the molecule by including polynucleotides containing known or random nucleotide sequences at selected positions.

Alternatively, mutations may be introduced into an enzymatic RNA molecule by substituting 5-Br dUTP for TTP in the reverse transcription reaction. 5-Br dU can pair with dG in the "wobble" position as well as dA in the standard Watson-Crick position, leading to A G and G A transitions. Similarly, substituting UTP for UTP in the forward transcription reaction would lead to C U and U C transitions in the subsequent round of RNA synthesis, as described above.

Enzymatic RNA molecules of the present invention may be of varying lengths and folding patterns, as appropriate, depending on the type and function of the molecule. For example, enzymatic RNA molecules derived from group I introns Tetrahymena-derived ribozymes).may be about 413 or more nucleotides in S 20 length, although a length not exceeding 413 nucleotides is preferred, to avoid limiting the therapeutic usefulness of molecules by making them too large or unwieldy. In various therapeutic applications, enzymatic RNA molecules of the present invention 25 comprise the enzymatically active portions of ribozymes. In various embodiments, enzymatic RNA molecules of the present invention comprise fewer than 400 nucleotides, fewer than 300 nucleotides, fewer than *200 nucleotides, or fewer than 100 nucleotides.

In other therapeutic applications, enzymatic RNA molecules such as "hammerhead" ribozymes are preferably no more than about 50 nucleotides in length, with a length of 30-40 nucleotides being particularly preferred. Even more preferred are hammerhead ribozymes of about 31-36 nucleotides in length.

Moreover, if one intends to synthesize molecules for use as disclosed herein, the larger the enzymatic nucleic acid molecule is, the more difficult it is to -58synthesize. Those of skill in the art will certainly appreciate these design constraints.

Various preferred methods of modifying ribozymes and other enzymatic RNA molecules, ribonucleases, and deoxyribonucleases of the present invention are further described in Examples 1-5 hereinbelow.

C. Nucleotide Analogs As noted above, the term "nucleotide analog" as used herein generally refers to a purine or pyrimidine nucleotide that differs structurally from A, T, G, C, or U, but is sufficiently similar to substitute for the normal nucleotide in a nucleic acid molecule. As used herein, the term "nucleotide analog" encompasses altered bases, different sugars, or a combination of the two. Examples of nucleotide analogs useful acording to the present invention include those listed in the following Table, most of which are found in the approved listing of modified bases at 37 CFR §1.822 (which is incorporated herein by reference).

p V0069 0 .0.0 Abbreviation ac4c cm cmnm5s2u d fm galq gm i i6a mla mlf mig Table 1 Nucleotide Analogs Description 4-acetylcytidine 2'-O-methylcytidine 5-carboxymethylaminomethyl-2thioridine dihydrouridine 2'-O-methylpseudouridine I, D-galactosylqueosine 2'-O-methylguanosine inosine N6-isopentenyladenosine 1-methyladenosine 1-methylpseudouridine 1-methylguanosine -59- Abbreviation ml11 m22 g m2a m2g m3 c mE a m7g mam5s2u mang mcm5s2u mo 5u ms2i6a ms2 t6 a (Table 1, cont'd) Descrip~tion 1 -methylinosine 2, 2-dimethylguanosine 2 -methyladenosine 2 -methylguanosine 3 -methylcytidine 5 -methylcytidine NE -methyladenosine 7 -methylguanosine 5 -methylaminomethyluridine 5 -methoxyaminomethyl thiouridine f9, D-mannosylmethyluridine 5 -methoxycarbonylmethyluridine 5-me thoxyuridine 2 -methylthio-NE isopentenyladenosine N- ((9-9-D-ribofuranosyl-2methylthiopurine- 6yl) carbamoyl) threonine N- ((9-9-D-ribofuranosylpurine-6yl) N-methyl-carbanoyl) threonine uridine-5-oxyacetic acid methylester uridine-5-oxyacetic acid (v) wybutoxosime pseudouridine quec sine 2-thiocytidine 5-methyl thiouridine 2-thiouridine 4- thiouridine 5-me thyluridine N- ((9-f9-D-ribofuranosylpurine-6yl) carbamoyl) threonine 2' 2' -0-methyluridine mt Ga osyw p q s2c s2t s2u s4u t t~a (Table 1, cont'd) Abbreviation Description yw wybutosine x 3-( 3 -amino-3-carboxypropyl)uridine, (acp3)u araU S, D-arabinosyl araT S, D-arabinosyl Other useful analogs include those described in published international application no. WO 92/20823 (the disclosures of which are incorporated by reference herein), or analogs made according to the methods disclosed therein. Analogs described in DeMesmaeker, et al., Anqew. Chem. Int. Ed. Enql. 33: 226-229 (1994); DeMesmaeker, et al., Synlett: 733-736 (Oct. 1993); Nielsen, et al., Science 254: 1497-1500 (1991); and Idziak, et al., Tetrahedron Letters 34: 5417-5420 (1993) are also useful according to the withindisclosed invention and said disclosures are 20 incorporated by reference herein.

D. Methods of Cleaving Nucleic Acid Molecules The present invention also describes useful methods for cleaving any single-stranded, looped, partially or fully double-stranded nucleic acid; the 25 majority of these methods employ the novel enzymatically active nucleic acid molecules of the present invention. The methods of this invention may be used to cleave single-stranded nucleic acids or single-stranded portions of double-stranded nucleic acids, whether those nucleic acids are present in an in vitro or ex vivo system, or whether they are present inside a cell (and irrespective of whether those cells are eucaryotic, procaryotic, plant, animal, yeast or bacterial cells).

For example, if the double-stranded nucleic acid is dsDNA, methods of using the enzymatic RNA molecules described herein may be used to cleave single-stranded portions of dsDNA, "looped" DNA or single- -61stranded segments of DNA that are accessible during transcription or translation.

It is also contemplated that enzymatic RNA molecules of the present invention, and the withindisclosed methods of using same, may be used to cleave dsDNA. In one embodiment, cleavage of dsDNA may be accomplished using coupled or paired enzymatic RNA molecules, wherein one member of the pair recognizes and cleaves one target nucleotide sequence, while its "partner" recognizes and cleaves the complementary target nucleotide sequence. In preferred embodiments, however, the enzymatic RNA molecules and methods of this invention are used to cleave single-stranded nucleic acids or single-stranded portions of doublestranded nucleic acids, in vivo, in vitro or ex vivo.

In various embodiments, the single-stranded nucleic acid substrate comprises single-stranded DNA.

modified DNA, RNA and modified RNA. Preferably, an *0 S exemplary nucleic acid substrate is single-stranded at 20 or near the substrate cleavage sequence so that an enzymatic nucleic acid molecule of the present invention can hybridize to the substrate cleavage sequence by virtue of the enzyme's recognition sequence.

25 A nucleic acid substrate that can be cleaved by a method of this invention may be chemically synthesized or enzymatically produced, or it may be isolated from various sources such as phages, viruses, prokaryotic .055. cells, or eukaryotic cells, including animal cells, plant cells, eukaryotic cells, yeast cells and bacterial cells. Chemically synthesized singlestranded nucleic acids are commercially available from many sources including, without limitation, Research Genetics (Huntsville, AL) DNA substrates may also be synthesized, via using an Applied Biosystems (Foster City, CA) oligonucleotide synthesizer according to the manufacturer's instructions. Single-stranded phages -62such as the M13 cloning vectors described by Messing et al., PNAS USA 74: 3642-3646 (1977), and Yanisch-Perron et al., Gene 33: 103-119 (1985) are also sources of DNA substrates. Bacterial cells containing single-stranded phages would also be a ready source of suitable singlestranded DNA. Viruses that are either single-stranded DNA viruses such as the parvoviruses or are partially single-stranded DNA viruses such as the hepatitis virus would provide single-stranded DNA that could be cleaved by a method of the present invention.

Single-stranded RNA cleavable by a method of the present invention could be provided by any of the RNA viruses such as the picornaviruses, togaviruses, orthomyxoviruses, paramyxoviruses, rhabdoviruses, coronaviruses, arenaviruses or retroviruses. As noted previously, a wide variety of prokaryotic and eukaryotic cells may also be excellent sources of @0 suitable nucleic acid substrates.

The methods of this invention may be used on 20 single-stranded nucleic acids or single-stranded o portions of double-stranded DNA (dsDNA) that are present inside a cell, including eucaryotic, procaryotic, plant, animal, yeast or bacterial cells.

Under these conditions an enzymatic nucleic acid 25 molecule an enzymatic RNA molecule or ribozyme) of the present invention could act as an anti-viral agent or a regulator of gene expression. Examples of such uses of enzymatic RNA molecules of the present invention are described in subsection F hereinbelow.

30 In the majority of methods of the present invention, cleavage of single-stranded DNA occurs at the 3'-terminus of a predetermined base sequence. This predetermined base sequence or substrate cleavage sequence typically contains from about 2 to about nucleotides. In a preferred variation, the predetermined base or substrate cleavage sequence comprises about 2 to about 6 nucleotides. In other preferred embodiments, an enzymatic RNA molecule of the -63present invention is able to recognize nucleotides either upstream, or upstream and downstream of the cleavage site. In various embodiments, an enzymatic RNA molecule is able to recognize about 2-10 nucleotides upstream of the cleavage site; in other embodiments, an enzymatic RNA molecule is able to recognize about 2-10 nucleotides upstream and about 2nucleotides downstream of the cleavage site. Other preferred embodiments contemplate an enzymatic RNA molecule that is capable of recognizing a nucleotide sequence up to about 30 nucleotides in length, with a length up to about 20 nucleotides being preferred.

The within-disclosed methods allow cleavage at any nucleotide sequence by altering the nucleotide sequence 15 of the recognition site of the enzymatic RNA molecule.

This allows cleavage of single-stranded DNA in the absence of a restriction endonuclease site at that position.

Cleavage at the 3'-terminus of a predetermined 20 base sequence produces a single-stranded DNA containing the substrate cleavage sequence, with a 3'-terminal hydroxyl group. In addition, the cleavage joins the remainder of the original single-stranded DNA substrate with the enzymatic RNA molecule. This cleavage 25 reaction and products produced from this cleavage :reaction are analogous to the cleavage reaction and cleavage products produced by the Tetrahymena ribozyme described by Zaug and Cech, Science 231: 470-475 (1986) An enzymatic RNA molecule of the present invention may be separated from the remainder of the singlestranded DNA substrate by site-specific hydrolysis at the phosphodiester bond following the 3'-terminal guanosine of the enzymatic RNA molecule, similar to the site-specific cleavage at this position described for the ribozyme acting on RNA by Inoue et al., J. Mol.

Biol. 189: 143-165 (1986). Separation of the enzymatic RNA molecule from the substrate allows the enzymatic -64- RNA molecule to carry out another cleavage reaction.

Generally, the nucleic acid substrate is treated under appropriate nucleic acid cleaving conditions preferably, physiologic conditions with an effective amount of an enzymatic RNA molecule of the present invention. If the nucleic acid substrate comprises DNA, preferably, cleaving conditions include the presence of Mg 2 (usually in the form of MgCl 2 at a concentration of about 2-100 mM. Typically, the DNA cleaving conditions include MgCl 2 at a concentration of about 2 mM to about 50 mM. Preferably, MgCl 2 is present at a concentration of about 5 mM to about 25 mM. More preferably, MgCl 2 is present at a concentration of about 5 mM to about 15 mM, with a concentration of magnesium S" 15 ion of about 10 mM being particularly preferred.

The optimal MgCl 2 concentration to include in the DNA cleaving conditions can be easily determined by determining the amount of single-stranded DNA cleaved at a given MgCl 2 concentration. One skilled in the art will understand that the optimal MgCI 2 concentration may vary depending on the particular enzymatic RNA molecule employed.

An effective amount of an enzymatic RNA molecule is the amount required to cleave a predetermined base 25 sequence present within the single-stranded

DNA.

Preferably, the enzymatic RNA molecule is present at a j molar ratio of RNA molecule to substrate cleavage sites i'.

of 1 to 20. This ratio may vary depending on the length of treatment and efficiency of the particular enzymatic RNA molecule under the particular

DNA

cleavage conditions employed.

"Treating" typically involves admixing, in aqueous solution, the DNA-containing substrate, the enzyme and the MgCl 2 to' form a DNA cleavage admixture, and then maintaining the admixture thus formed under DNA cleaving conditions for a time period sufficient for the enzymatic RNA molecule to cleave the DNA substrate at any of the predetermined nucleotide sequences present in the DNA.

In one embodiment of the present invention, the amount of time necessary for the enzymatic RNA molecule to cleave the single-stranded DNA has been predetermined. The amount of time is from about minutes to about 24 hours and will vary depending upon the concentration of the reactants, and the temperature of the reaction. Usually, this time period is from about 30 minutes to about 4 hours such that the enzymatic RNA molecule cleaves the single-stranded DNA at any of the predetermined nucleotide sequences S present.

The present invention further contemplates that 15 the DNA cleaving conditions include a pH from about pH 6.0 to about pH 9.0. In one preferred embodiment, the pH ranges from about pH 6.5 to pH 8.0. In another preferred embodiment, the pH emulates physiological conditions, the pH is about 7.0-7.8, with a pH of 20 about 7.5 being particularly preferred.

One skilled in the art will appreciate that the methods of the present invention will work over a wide pH range so long as the pH used for DNA cleaving is •go• such that the enzymatic RNA molecule is able to remain 25 in an active conformation. An enzymatic RNA molecule in an active conformation is easily detected by its ability to cleave single-stranded DNA at a predetermined nucleotide sequence.

In various embodiments, the DNA cleaving conditions also include a variety of temperature ranges; as noted previously, temperature ranges consistent with physiological conditions are especially preferred. In one embodiment, the temperature ranges from about 15 0 C to about 60 0 C. In another variation, the DNA cleaving conditions are from about 30 0 C to about 560C. The temperature of the DNA cleaving conditions are constrained only by the desired cleavage rate and the stability of that particular enzymatic RNA -66molecule at that particular temperature. In yet another variation, DNA cleavage conditions include a temperature from about 35 0 C to about 50 0 C. In a preferred embodiment, DNA cleavage conditions comprise a temperature range of about 37 0 C to about 42 0

C.

In various other methods, the present invention contemplates DNA cleaving conditions including the presence of a polyamine. Polyamines useful for practicing the present invention include spermidine, putrescine, spermine and the like. In one preferred variation, the polyamine is spermidine and it is present at a concentration of about .01 mM to about mM. In another variation, the polyamine is present at a concentration of about 1 mM to about 10 mM. DNA cleavage conditions may also include the presence of olyamine at a concentration of about 2 mM to about mM. In various preferred embodiments, the polyamine is spermidine.

A variety of uses for the enzymatic RNA molecules of the present invention are also disclosed herein. For example, in one embodiment, an enzymatic RNA molecule (ribozyme) of the present invention may be useful as an anti-viral agent or a regulator of gene expression.

For example, an enzymatic RNA molecule of the 25 present invention may be used to treat a virally-caused disease by administering to a patient in need of treatment an enzymatic RNA molecule which cleaves viral nucleic acid or virus-encoded nucleic acid. Viruses whose nucleic acids or encoded nucleic acids may be susceptible to cleavage by an enzymatic RNA molecule of the present invention include, for example, papillomavirus, EBV, HSV, HBV, HIV HIV-1, HIV- T-cell leukemia virus HTLV-1, HTLV-2), HCV, CMV, influenza virus, and picornavirus.

In a related aspect of the invention, an enzymatic RNA molecule of the present invention may be used to treat virally-cause diseases in animals, such as feline immunodeficiency virus (FIV), feline leukemia virus -67- (FLV), simian immunodeficiency virus (SIV), bovine leukemia virus (BLV), and simian leukemia virus STLV). Useful enzymatic RNA molecules of the present invention may be selected on the basis of their ability to target a selected region (or regions) of a viral genome. In various embodiments, such enzymatic RNA molecules are able to cleave the target nucleic acid sequence or segment in a manner which inhibits transcription, translation, or expression of the nucleic acid sequence. Target nucleic acid segments are selected so that inhibition of transcription, translation or expression will have maximal effect.

For example, targeting the enzymatic RNA molecules of the present invention to nucleic acid sequences oe 15 involved in protein synthesis, genomic replication, or packaging into virions is expected to inhibit viral replication.

Once an appropriate target is selected, enzymatic RNA molecules capable of cleaving the target nucleotide 20 sequence may be identified from a pool of enzymatic RNA molecules via the use of oligonucleotide probes.

Alternatively, the recognition sequence of an enzymatic RNA molecule of the present invention may be mutated or oeoo modified so that it selectively targets the desired 25 viral sequence. Methods of identifying suitable target S. sequences are available in the art; see, e.g., published PCT application nos. WO 93/23569, WO 91/04319, WO 91/04324, and WO 91/03162; and published European patent application no. EPO 585,549, the disclosures of which are incorporated by reference herein.

Enzymatic RNA molecules of the present invention are preferably targeted to a highly-conserved region of a viral nucleic acid sequence so that all strains and types of such viruses may be treated with a single enzymatic RNA molecule or enzymatic RNA molecule species. Enzymatic RNA molecules of the present invention may be designed to target RNA or DNA -68sequences as appropriate; they may also be designed to target transcripts of viral nucleic acid sequences, whether those transcripts are comprised of RNA or DNA.

Enzymatic RNA molecules according to the present invention may also be designed to target antisense nucleic acid sequences.

Enzymatic RNA molecules of the present invention may further be used to treat transformed eukaryotic cells keratinocytes, hepatocytes or epithelial cells in such a manner that they inhibit the expression of viral genes known or believed to cause cell immortalization or transformation. In another *embodiment, enzymatic RNA molecules may be used to treat latent viral infections, by inhibiting gene 15 expression required for the maintenance of the viral episomal genome.

If desired, a vector such as those described hereinbelow may be used in a therapeutic protocol via use of the methods and systems described in published international application no. WO 92/13070, the disclosures of which are incorporated by reference :9herein. Via one of the disclosed methods, expression of a therapeutic enzymatic RNA molecule of the present invention may be temporally regulated. Thus, a vector comprising sequences encoding therapeutic enzymatic

RNA

molecules of the present invention preferably includes a promoter which expresses enzymatic RNA molecules only in the presence of a nucleic acid molecule which is manifested when an "invading" organism or disease state is present. Such a "timed release" enzymatic

RNA

molecule then functions to impair or destroy the cell in which the unwanted organism or condition is found, to bring about reduced expression of a protein product related to the disease state, or to stimulate production of a "defensive"', protein of the compromised cell.

Enzymatic RNA molecules of the present invention may also be used in vivo or in vitro to make defective -69viral particles which nonetheless retain their immunogenic properties. In this way, an organism's own immune system may be stimulated to combat infection by intact non-defective) viral particles.

In a related aspect, vaccines or other preparations used for immunization may be formed from defective viruses (or portions thereof) created by a method of this invention. Methods for immunizing or vaccinating organisms using defective viral particles with DNA or vectors encoding an enzymatic RNA molecule of this invention under the control of an appropriate promoter) are also contemplated herein.

.r es Enzymatic RNA molecules of the present invention may also be conjugated or otherwise linked to viral 15 particles or viruses, where the latter are used as vectors which may transport enzymatic RNA molecules to cells infected with another virus. Useful vectors of this type may be formed via standard technology; for "example, adenovirus vectors and related methodologies 20 may be effective in this regard.

Diagnostic uses of enzymatic RNA molecules of the present invention are also contemplated. For example, because of the relationship between enzymatic RNA oo molecule activity and target nucleic acid structure, 25 mutations in any region of the target molecule may be :detected, particularly where such mutations alter the base-pairing and three-dimensional structure of the target nucleotide sequence, whether RNA or DNA.

Further, by using multiple enzymatic RNA molecules as described in this invention, one may map nucleotide changes important to DNA or RNA structure and function in vitro, as well as in vivo in cells and tissues. Cleavage of target DNAs or RNAs with enzymatic RNA molecules of the present invention may be used to inhibit gene expression and to assist in defining the role of specified gene products in the progression of disease. In this application, other genetic targets may be identified as important mediators of the disease under investigation. Such experiments may lead to better treatment or modification of disease progression by providing the possibility of combinational therapies. Examples of the latter include application of multiple enzymatic RNA molecules targeted to different genes or nucleotide sequences; enzymatic RNA molecules coupled with known small molecule inhibitors; intermittent treatment with combinations of enzymatic RNA molecules and/or other chemical or biological molecules; and enzymatic RNA molecules administered in conjunction with therapeutic antisense sequences.

The present invention contemplates that enzymatic RNA molecules as described herein may be used as sequence-specific endoribonucleases. In various preferred embodiments, enzymatic RNA molecules cleave S RNA substrates with high catalytic efficiency km/Km 108K' min-') It is also contemplated herein that the enzymatic 20 RNA molecules of the present invention can act as sequence-specific endodeoxyribonucleases. In various preferred embodiments, enzymatic RNA molecules of the present invention cleave DNA with high catalytic efficiency. For example, in one embodiment, an enzymatic RNA molecule of the present invention has the following efficiency: K, 1.9 AM and k. 0.005 min- 1 (km/K 2700 M' minl). In another embodiment, an enzymatic RNA molecule has the following efficiency:

K

m 2.0 AM and k, 0.007 min-' 3600 M' min.') 30 In various embodiments encompassed by the present invention, catalytic efficiencies are greatest under physiologic conditions for example, when the temperature is about 37 0 C and divalent cations are available at a concentration of about 10 mM mM MgCl 2 Thus, the catalytic efficiencies of the enzymatic RNA molecules of the present invention are preferably 10 or more times greater than that of the wild-type, and the molecules also display improvement -71in both Km and k,.

In a related aspect, the enzymatic RNA molecules of this invention may be used to identify the nucleic acid sequence to which a proteinaceous or nonproteinaceous adjunct binds. The proteinaceous or nonproteinaceous adjunct contemplated may bind to specific nucleotide sequences. The term "adjunct" as used herein is meant to include proteinaceous or nonproteinaceous substances which are joined or added to the nucleic acid sequence but are not essentially a part of said sequence. The adjunct may be a protein, metal, inorganic molecule, lipid, or other substance.

For example, multiple enzymatic RNA molecules with known recognition sequences may be used to identify the nucleic acid sequence to which an adjunct is specifically bound. The identification of these nucleic acid sequences may be useful in 4ienUtifyin S* specific nucleic acid target sequences involved in gene S expression.

20 In a related aspect, the enzymatic RNA molecules of this invention may be used to confirm the presence i" or absence of proteins or other adjuncts (proteinaceous or nonproteinaceous) which bind a nucleic acid substrate at a specific recognition site. For example, 25 if a protein or other adjunct specifically binds to a known nucleic acid sequence, an enzymatic RNA molecule which is specific for the same nucleic acid sequences could be used to detect the presence or absence of such a protein or adjunct. The detection of such proteins 30 or adjuncts may be useful for in vivo or in vitro diagnostic assays.

Enzymatic .RNA molecules as described herein may also be used to regulate a variety of reaction systems, whether those reactions are occurring in vitro or in vivo. For example, it is contemplated that-one may take advantage of the endonuclease activities of the molecules of the present invention by using them to regulate or terminate reactions in which transcription -72and/or translation are taking place. For example, enzymatic RNA molecules which recognize a specific nucleic acid residue sequence found in one or more of the PCR primers being used in a particular reaction may be added to such PCR reaction system to modulate or terminate processes dependent upon the primer targeted by the enzymatic RNA molecule. In this way, the entire PCR reaction would not be stopped; rather, only those products dependent upon the targeted primer(s) would be limited or eliminated.

Enzymatic FRNA molecules of the present invention may also be used during the transcription of DNA sequences or oligomers to generate truncated transcripts in the same reaction vessel in which longer transcripts are generated. For example, if one wishes to generate a population of proteins with a membrane- *"*spanning domain attached and a population of proteins with the membrane-spanning domain removed, the transcription/translation reactions may be allowed to 20 run for a predetermined period of time until a predetermined amount of "intact" non-truncated) product is generated. Subsequently, enzymatic RNA molecules of the present invention which have been •engineered to cleave the transcripts at a site-specific location a locus adjacent to the sequence for the membrane-spanning domain) may be added to the 00-. reaction mixture, which will result in the generation of truncated transcripts (and truncated protein products translated therefrom) S 30 Alternatively, the enzymatic RNA molecules may be engineered to cleave the gene from which the messenger RNAs are being transcribed, which will produce a similar result with regard to the protein generated thereby. One of sufficient skill in the art will appreciate, based on the present disclosure, that enzymatic RNA molecules of the present invention may be adapted for a variety of uses; the uses disclosed herein should thus be understood to be exemplary and -73not limiting.

Use of enzymatic RNA molecules of the present invention in a coupled isothermal polynucleotide amplification and translation system could also modulate or "shut down" amplification processes at a predetermined time without simultaneously terminating the linked translation system. Similarly, the enzymatic RNA molecules of the present invention may be used to modulate or terminate therapeutic or diagnostic applications involving the use of "antisense" nucleotide sequences, by cleaving said antisense sequences.

Enzymatic RNA molecules of the present invention may also be used in oligonucleotide DNA) footprinting or footprint analysis of proteinnucleotide binding. For example, enzymatic RNA molecules of the present invention may be prepared as disclosed herein, with a variety of different recognition sequences, and then admixed with a DNA 20 sample to which a particular protein has bound.

ooe• Labeled DNA fragments resulting from cleavage of the DNA sample by enzymatic RNA molecules of the present invention may then be analyzed according to standard protocols to enable one to identify specific protein 25 binding sites on the DNA sample. Additionally, the same types of enzymatic RNA molecules of the present invention may be admixed with a sample of DNA to which the particular protein has not been added. Labeled DNA Do fragments resulting from cleavage of this second DNA sample may then be analyzed and the results compared *with those generated when the protein-bound DNA sample .*was analyzed.

A variety of protocols are available for footprinting and similar analyses which are easily adapted for use with the enzymatic RNA molecules of the present invention, wherein the enzymatic RNA molecules disclosed herein are substituted in place of other nuclease enzymes. For example, see Ausubel, et al.

-74r r r r Current Protocols in Molecular Bioloy, John Wiley Sons, Inc. (1994), for descriptions of a variety of protocols useful in conjunction with the enzymatic RNA molecules of the present invention.

It is anticipated that enzymatic RNA molecules of the present invention may also be engineered and used as highly-specific endonucleases, which makes them uniquely useful in vector construction, particularly when cleavage of single-stranded sequences is desired.

For example, as disclosed above, an enzymatic

RNA

molecule of the present invention may be specifically targeted to a specific nucleotide sequence of from about 2 to 30 nucleotides in length; thus, such a molecule may be used in the design of vectors and cassettes for the delivery and expression of nucleotide sequences, by facilitating the insertion of predetermined nucleotide sequences into a compatible vector.

Other uses of the within-disclosed enzymatic

RNA

molecules will be apparent based on the present disclosure and are thus within the scope of the present invention.

E. Methods of Engineering Enzymatic

RNA

Molecules The present invention also contemplates methods of producing nucleic acid molecules having a predetermined activity. In one preferred embodiment, the nucleic acid molecule is an enzymatic RNA molecule according to the present invention. In another variation, the desired activity is a catalytic activity.

In one embodiment, the present invention contemplates methods of synthesizing enzymatic

RNA

molecules which may then be "engineered" to catalyze a specific or predetermined reaction. Methods of preparing enzymatic RNA molecules are described herein; see, the Examples below. In other embodiments, an enzymatic RNA molecule of the present invention may be engineered to bind small molecules or ligands, such as adenosine triphosphate (ATP). (See, e.g., Sassanfar, et al., Nature 364: 550-553 (1993).) The present invention also discloses thatpopulation of enzymatic RNA molecules may be subjected to mutagenizing conditions to produce a diverse population of mutant enzymatic RNA molecules or ribozymes. Thereafter, enzymatic RNA molecules having desired characteristics are selected and/or separated from the population and are subsequently amplified.

Alternatively, mutations may be introduced in the enzymatic RNA molecule by altering the length of the recognition site (internal guide sequence) of the enzymatic RNA molecule. The recognition site of an enzymatic RNA molecule generally associates with a complementary sequence of bases within a substrate molecule a nucleic acid or a hybrid oligonucleotide-oliqopeptide sequence). Methods of altering the length of the recognition site are known in the art and include PCR, for example. Useful 20 techniques are described further in the Examples below.

oo• Alteration of the length of the recognition site of the enzymatic RNA molecule which retains the ability to hybridize with a complementary sequence of bases within a substrate nucleic acid sequence may have a 25 desirable effect on the binding specificity of the enzymatic RNA molecule. For example, an increase in s the length of the recognition site may increase binding specificity between the enzymatic RNA molecule and the complementary base sequences of the substrate nucleic acid. In addition, an increase in the length of the recognition site may also increase the affinity with which it binds to the nucleic acid substrate. In various embodiments, these altered recognition sites in the enzymatic RNA molecule confer increased binding specificity and affinity between the enzymatic RNA molecule and its nucleic acid substrate.

Similarly, alteration of the length of the recognition site of the enzymatic RNA molecule which -76retains the ability to recognize a sequence of bases within the nucleic acid segment of a hybrid substrate molecule or with an amino acid residue sequence recognized thereby may have a desirable effect on the binding specificity of the enzymatic RNA molecule. For example, an increase in the length of the recognition site may increase binding specificity between the enzymatic RNA molecule and the complementary base sequences of an oligonucleotide in a hybrid substrate, or may enhance recognition of amino acid residue sequences in a hybrid molecule or in a polypepti-e substrate. In addition, an increase in the length of the recognition site may also increase the affinity with which an enzymatic RNA molecule of the present invention binds to a substrate. In various embodimen.ts, these altered regnlition sites in the enzymatic RNA molecule confer increased binding specificity and affinity between the enzymatic RNA molecule and its substrate.

20 It has recently been noted that certain oligonucleotides are able to recognize and bind molecules other than oligonucleotides with complementary sequences. These oligonucleotides are often given the name "aptamers". For example, Ellington and Szostak describe RNA molecules that are able to bind a variety of organic dyes (Nature 346: 818-822 (1990)), while Bock, et al. describe ssDNA o" molecules that bind human thrombin (Nature 355: 564-566 (1992)). Similarly, Jellinek, et al. describe RNA 30 ligands to basic fibroblast growth factor (PNAS USA "11227-11231 (1993)).

Until the advent of the present invention, however, no one has described the existence of catalytically active RNA enzymes with reproducible amide-cleaving capabilities. The art has also been silent with respect to methods of engineering and selecting catalytically active oligonucleotide molecules possessing this ability, until now.

-77- One of skill in the art may realize that the enzymatic RNA molecules of this invention can be altered at any nucleotide sequence, such as the recognition site, by various methods disclosed herein, including PCR and 3SR. Additional nucleotides can be added to the 5' end of the enzymatic RNA molecule by including the additional nucleotides in the primer used to introduce the T7 promoter binding site. The additional nucleotides would be included in the primer between the T7 promoter sequence and the nucleotide sequences which hybridize to the enzymatic RNA molecule at the 5' end.

Enzymatic RNA molecules of the present invention may also be prepared or engineered in a more non-random fashion via use of methods such as site-directed mutagenesis. For example, site-directed mutagenesis may be carried out essentially as described in Morinaga, et al., Biotechnoloqv 2: 636 (1984), which is incorporated herein by reference. Useful site-directed 20 mutagenesis techniques are also described herein; see, Example 1.

In various embodiments, the population of group I nucleic acids is made up of at least 2 group I nucleic acids. In one variation, group I nucleic acids are 25 nucleic acid molecules having a nucleic acid sequence defining a recognition site that is contiguous or adjacent to the 5'-terminus of the nucleotide sequence, a first spacer region located 3'-terminal to the recognition site, a P3[5'] region located 3'-terminal to the first spacer region, a second spacer region located 3'-terminal to the P3[5'] region, a first stem loop located 3'-terminal to the second spacer region, a second stem loop located 3'-terminal to the first stem loop, a third spacer region located 3'-terminal to the second stem loop, and a third stem loop located 3'terminal to the third spacer region, the third stem loop comprising a 5' stem portion defining a P3[3'] region capable of hybridizing to the P3[5'] region.

-78- Other characteristics of enzymatic RIJA molecules produced according to the presently-disclosed methods are described elsewhere herein.

In other embodiments, mutagenizing conditions include conditions that introduce either defined or random nucleotide substitutions within an enzymatic RNA molecule. Examples of typical mutagenizing conditions include conditions disclosed in other parts of this specification and the methods described by Joyce et al., Nucl. Acids Res. 17: 711-722 (1989); Joyce, Gene 82: 83-87 (1989); and Beaudry and Joyce, Science 257: 635-41 (1992) In still other embodiments, a diverse population of mutant enzymatic nucleic acid molecules of the present invention is one that contains at least 2 riu~±x±u u±u mu±euules thnat do not have the exact same 6nucleotide sequence. In other variations, from such a diverse population, an enzymatic RNA molecule or other enzymatic nucleic acid having a predetermined activity 20 is then selected on the basis of its ability to perform the predetermined activity. In various embodiments, the predetermined activity comprises, without limitation, enhanced catalytic activity, decreased KM, enhanced substrate binding ability, altered substrate specificity, and the like.

Parameters which may be considered aspects of enzyme performance include catalytic activity or capacity, substrate binding ability, enzyme turnover S 3 rate, enzyme sensitivity to feedback mechanisms, and 30 the like. In certain aspects, substrate specificity may be considered an aspect of enzyme performance, particularly in situations in which an enzyme is able to recognize and bind two or more competing substrates, each of which affects the enzyme's performance with respect to the other substrate(s) Substrate specificity, as used herein, may refer to the specificity of an enzymatic nucleic acid molecule as described herein for a particular -79substrate, such as one comprising oligonucleotides only, polypeptides only, or a composite of both. In the case of the latter type of substrate, an enzymatic nucleic acid molecule of the present invention may preferentially bind to a particular region of such a hybrid or composite substrate.

Alternatively, substrate specificity, as used herein, may refer to the specificity of an enzymatic nucleic acid molecule as described herein for a particular substrate, such as one comprising RNA only, DNA only, or a composite of both. Substrate specificity also refers to whether an enzymatic nucleic acid molecule of the present invention preferentially eeoc binds a single-stranded nucleic acid substrate, a 15 double-stranded nucleic substrate, or a nucleic acid oee molecule having both single-stranded and doublestranded regions (such as nucleic acid molecules with "loops"). In the case of the latter type of substrate, an enzymatic nucleic acid molecule of the present invention may preferentially bind to a particular region of such a composite substrate.

Substrate specificity may also include sequence specificity; an enzymatic nucleic acid molecule of the present invention may "recognize" and bind to a S. 25 nucleic acid substrate having a particular nucleic acid sequence; to a substrate having a particular amino acid residue sequence; or to a substrate having a particular combination of both. For example, if the substrate recognition site of an enzymatic nucleic acid molecule of the present invention will only bind to substrate molecules having a series of one or two arginine residues in a row, then the enzymatic nucleic acid molecule will tend not to recognize or bind nucleic acid substrate molecules lacking such a sequence.

Similarly, with regard to DNA-cleaving enzymes according to the present invention, substrate specificity may also include sequence specificity; an enzymatic nucleic acid molecule of the present invention may "recognize" and bird to a nucleic acid substrate having a particular nucleic acid sequence.

For example, if the substrate recognition site of an enzymatic nucleic acid molecule of the present invention will only bind to nucleic acid substrate molecules having a series of six adenine residues in a row, then the enzymatic nucleic acid molecule will tend not to recognize or bind nucleic acid substrate molecules lacking such a sequence. In various 10 cumbodiments, selecting includes any means of physically separating the mutant enzymatic nucleic acids having a predetermined activity from the diverse population of mutant enzymatic nucleic acids. Often, selecting comprises separation by size, by the presence of a e catalytic activity, or by hybridizing the mutant nucleic acid to another nucleir acid that is eihe i solution or attached to a solid matrix.

In various embodiments, "selecting" includes any means of physically separating the mutant enzymatic 20 nucleic acids having a predetermined activity from the diverse population of mutant enzymatic nucleic acids.

Often, selecting comprises separation by size, by the e.e presence of a catalytic activity, or by hybridizing the mutant nucleic acid to another nucleic acid or to a peptide that is either in solution or attached to a solid matrix.

In various embodiments, the "predetermined activity" is such that the mutant enzymatic nucleic acid having the predetermined activity becomes labelled in some fashion by virtue of the activity. For example, the predetermined activity may be an enzymatic RNA molecule activity whereby the activity of the mutant enzymatic nucleic acid upon its substrate causes the mutant enzymatic nucleic acid to become covalently linked to it. The mutant enzymatic nucleic acid is then selected by virtue of the covalent linkage. In other embodiments, selecting a mutant enzymatic nucleic acid having a predetermined activity includes -81amplification of the mutant enzymatic nucleic acid (see, Joyce, Gene 82: 83-.87 (1989); Beaudry and Joyce, Science 257: 635-41 (19-92)) F. Comositions i. Compositions IncludinQ DNA-Cleavinq Enzymatic RNA Molecules The invention also contemplates compositions containing one or more types or populations of enzymatic RNA molecules of the present invention; i.e., different types or populations may recognize and cleave 9. different nucleotide sequences. Compositions may further include one or more DNA-containing substrates.

*eoe*.

Compositions according to the present invention may further comprise magnesium ion or other divalent or eee S 15 monovalent cations, as discussed in section B above.

o e Preferably, enzymatic RNA molecules according to the present invention are present at a concentration of about 0.05 AM to about 2 AM. Typically, an enzymatic RNA molecule is present at a concentration ratio of enzymatic RNA molecule to single-stranded DNA substrate ""of from about 1:5 to about 1:50. More preferably, an enzymatic RNA molecule is present in the composition at a concentration of about 0.1 AM to about 1 AM. Even more preferably, compositions contain enzymatic RNA 25 molecules at a concentration of about 0.1 AM to about

AM.

Preferably, single-stranded DNA substrate is present in the composition at a concentration of about AM to about 1000 AM. One skilled in the art will understand that there are many sources of singlestranded DNA including synthetic DNA, phage DNA, "loop" DNA, denatured double-stranded DNA, viral DNA and cellular.

Magnesium ion (or other divalent or monovalent ions, as discussed previously) may also be present in the composition, at a concentration of about 2-100 mM.

More preferably, magnesium ion is present in the composition at a concentration of about 2 mM to about -82mM. Preferably, magnesium ion is present at a concentration of about 5 mM to about 15 mM, with a concentration of about 10 mM being particularly preferred. One skilled in the art will understand that the magnesium ion concentration is only constrained by the limits of solubility of magnesium in aqueous solution and a desire to have the enzymatic RNA molecule present in the same composition in an active conformation.

10 The invention also contemplates compositions containing an enzymatic RNA molecule of the present invention, a substrate containing single-stranded DNA, magnesium ion in concentrations as described hereinabove, and a polyamine. Preferably, the polyamine is spermidine, putrescine, or spermine. More preferably, the polyamine is spermidine and is present at a concentration of about 2 mM to about 10 mM. The invention further contemplates compositions containing an enzymatic RNA molecule of the present invention, 20 single-stranded DNA, magnesium ion at a concentration of greater than 20 millimolar, a second single-stranded DNA molecule ending in a 3'-terminal hydroxyl, and a S" third single-stranded DNA molecule having a guanine nucleotide at its 5'-terminal end.

Also contemplated by the present invention are compositions containing an enzymatic RNA molecule of the present invention, singled-stranded DNA-containing substrate,..and magnesium ion at a concentration of greater than about 2 millimolar, wherein said singlestranded DNA is greater in length than the recognition site present on the enzymatic RNA molecule.

2. Compositions Including Enzymatic RNA Molecules with Amidase/Peotidase Activity The invention also contemplates compositions containing one or more types or populations of amide bond- or peptide bond-cleaving enzymatic RNA molecules of the present invention; different types or -83populations may recognize and cleave different amino acid residue sequences. Compositions may further include a peptide-containing substrate. Compositions according to the present invention may further comprise magnesium ion or other divalent or monovalent cations, as discussed in section B above.

Preferably, enzymatic RNA molecules according to the present invention is present at a concentration of about 0.05 AM to about 2 AM. Typically, an enzymatic RNA molecule is present at a concentration ratio of enzymatic RNA molecule to substrate of from about to about 1:50. More preferably, an enzymatic RNA molecule is present in the composition at a concentration of about 0.1 AM to about 1 AM. Even more 15 preferably, compositions contain enzymatic RNA molecules at a concentration of about 0.1 AM to about 0.5 uM.

Preferably, the substrate is present in the composition at a concentration of about 0.5 AM to about 1000 AM. One skilled in the art will understand that there are many sources of amide bond-containing ':i substrates including naturally-occurring and synthetic amino acids, polypeptides, proteins (including those in denatured form), and molecules containing same.

25 Magnesium ion (or another suitable monovalent or divalent cation, as described previously) may also be present in the composition, at a concentration of about 2-100 mM. More preferably, the magnesium ion is present in the ,composition at a concentration of about 2 mM to about 50 mM. Preferably, magnesium ion is present at a concentration of about 5 mM to about mM, with a concentration of about 10 mM being particularly preferred. One skilled in the art will understand that the magnesium ion concentration is only constrained by the limits of solubility of magnesium in aqueous solution and a desire to have the enzymatic RNA molecule present in the same composition in an active conformation.

-84- The invention also contemplates compositions containing an enzymatic RNA molecule of the present invention, hybrid oligonucleotide-oligopeptide molecules, and magnesium ion in concentrations as described hereinabove. As noted previously, other monovalent or divalent ions Mn 2 may be used in place of magnesium.

Also contemplated by the present invention are compositions containing an enzymatic RNA molecule of 10 the present invention, an appropriate substrate a protein, polypeptide, hybrid oligonucleotideoligopeptide molecules, molecules including amino acids, or other amide or peptide bond-containing molecules capable of being recognized and acted upon by an enzymatic RNA molecule of the present invention), an d m.agnesium ion at a concentration of greater than about 2 millimolar, wherein said substrate is greater in length than the recognition site present on the enzymatic RNA molecule.

20 G. Methods of Using Enzymatic RNA Molecules with Amidase/Pentidase Activity The methods of using enzymatic RNA molecules as disclosed herein are legion. As discussed previously, molecules capable of cleaving the bonds linking neighboring amino acid molecules peptide bonds) have numerous uses encompassing a wide variety of applications. For example, enzymatic RNA molecules having the within-disclosed capabilities, structures, and/or functions are useful in pharmaceutical and medical products for wound debridement, clot dissolution, etc.), as well as in household items detergents, dental hygiene products, meat tenderizers). Industrial utility of the withindisclosed compounds, compositions and methods is also contemplated and well within the scope of the present invention.

H. Vectors The present invention also features expression vectors including a nucleic acid segment encoding an enzymatic RNA molecule of the present invention situated within the vector, preferably in a manner which allows expression of that enzymatic RNA molecule within a target cell a plant or animal cell).

Thus, in general, a vector according to the present invention includes a bacterial, viral or eukaryotic promoter within a plasmid, cosmid, phagemid, virus, viroid, or phage vector. Other suitable vectors include double-stranded DNA (dsDNA), partially doublestranded DNA, dsRNA, partially dsRNA, or singlestranded RNA (ssRNA) or DNA (ssDNA). It should also be appreciated that useful vectors according to the present invention need not be circular.

In one aspect of the present invention, a first enzymatic RNA molecule-encoding nucleotide sequence is transcrintionally linked to a promoter sequence. In another variation, one or more additional enzymatic RNA Smolecule-encoding nucleotide sequences are also 20 included in the vector; said additional enzymatic RNA molecule-encoding sequences may be located on either side, or both sides, of a nucleotide sequence encoding the first enzymatic RNA molecule. Preferably, there are intervening nucleotides or nucleotide sequences S* 25 between successive enzymatic RNA molecule-encoding sequences.

In another variation, nucleotide sequences flanking each of the additional enzymatic RNA moleculeencoding sequences are preferably provided, which sequences may be recognized by the first enzymatic RNA molecule. The intervening or flanking sequences preferably comprise at least 1 nucleotide; more preferably, intervening or flanking sequences are about 2-20 nucleotides in length, with sequences of about 10 nucleotides in length being particularly preferred.

The addition of polyadenine (poly(A)) tails may also be useful to protect the 3' end of an enzymatic RNA molecule according to the present invention. These -86imay be provided by including a poly(A) signal sit.e in the expression vector, which would signal a cell to add the poly(A) tail in vivo. Preferably, the signal is aligned in such a fashion that it prevents unwanted secondary structure formation with other parts of the enzymatic RNA molecule.

Alternatively, a poly(A) tail may be provided by introducing a poly(A) sequence directly into the expression vector. Since the poly(A) sequence may decrease in size over time when expressed in vivo, the vector may need to be monitored over time. Care must be taken, however, in the addition of a poly(A) tail which binds poly(A) binding proteins, which may prevent the enzymatic RNA molecule from acting upon its target S 15 nucleotide sequence. Other vectors and methods of generating same are described in the art; see, e.g., published international application no. WO 93/23569, the disclosures of which are incorporated by reference '.herein.

Thus, in one example, a vector may comprise a 'promoter operatively linked for expression to a nucleotide sequence encoding a first enzymatic RNA molecule followed, in a 3' -3 5' direction, by: a "flanking" nucleotide sequence capable of being 25 recognized and cleaved by said first enzymatic RNA molecule; a nucleotide sequence encoding a second enzymatic RNA molecule; another flanking nucleotide :sequence capable of being recognized and cleaved by said first enzymatic RNA molecule; a nucleotide sequence encoding a third enzymatic RNA molecule; (4) yet another flanking nucleotide sequence capable of being recognized and cleaved by said first enzymatic RNA molecule; and so forth.

Preferably, a vector according to the present invention includes a plurality of nucleic acid sequences encoding the second enzymatic RNA molecule, each flanked by nucleic acid sequences recognized by the first enzymatic RNA molecule. More preferably, -87such a plurality includes at least 5, preferably 7, more preferably 9 or more, nucleic acid sequences. In other embodiments, a vector as disclosed herein includes a promoter which regulates expression of the nucleic acid encoding the enzymatic RNA molecules from the vector.

The invention also contemplates that a promoter sequence is linked to a first or "releasing" enzymatic RNA molecule having an appropriate restriction endonuclease site. A single-stranded oligonucleotide is then provided which encodes the two flanking regions and a second "therapeutic") enzymatic RNA molecule. The oligonucleotides are then allowed to form partial duplexes via hybridization at the flanking regions. The single-stranded sections are then filled in using a DNA polymerase and deoxyribonucleotide triphosphates (dNTPs) to form a dsDNA molecule, which may then be ligated to the restriction endonuclease site to form the desired vector. As noted above, a 20 suitable vector may be chosen from the group comprising plasmids, cosmids, phagemids, virus, viroids, or phage, for example.

Preferably, the plurality of nucleic acid sequences are identical and are arranged in sequential fashion such that each has an identical end nearest to: the promoter. If desired, a poly(A) sequence adjacent to the sequence encoding the first or second enzymatic RNA molecule may be provided to increase stability of the RNA produced by the vector and/or to enhance 30 transport to appropriate cellular compartments.

Further, a restriction endonuclease site adjacent to the nucleic acid encoding the first enzymatic RNA molecule may be provided to allow insertion of nucleic acid encoding the second enzymatic RNA molecule during construction of the vector.

If delivery of a vector construct to a eucaryotic cell is desired, cellular splicing mechanisms within the target cell(s) may be utilized or integrated to -88cleave out the therapeutic second enzymatic RNA molecule(s) by encoding recognition sequences for the second enzymatic RNA molecules within the flanking sequences of the expressed transcript. Multiple copies of the releasing first enzymatic RNA molecule may be provided to enhance release of the second (i.e.

therapeutic) enzymatic RNA molecule if the turnover rate is slower than the degradation rate of the second enzymatic RNA molecule. If the target cell is a bacterial cell, in vitro modifications and certain cell modifications may be enhanced by providing appropriate nucleotide sequences within the vector and are useful in the enhancement of the turnover rate, enzymatic stability, and cleavage activity of the within- 15 disclosed enzymatic RNA molecules.

*method of formi.n.g. enzyma'l- RNP l1c±lU expression vector includes providing a vector comprising nu-leic acid encoding a first enzymatic RNA molecule, as discussed above, and providing a singlestranded DNA molecule encoding a second enzymatic RNA molecule, also as discussed above. The single-stranded DNA is then allowed to anneal to ,orm a partial duplex DNA which can be filled in by treatment with an appropriate enzyme, such as a DNA polymerase in the 25 presence of dNTPs, to form a duplex DNA which can then co be ligated to the vector. Large vectors resulting from *use of this method can then be selected to insure that a high copy number of the single-stranded DNA encoding S"the second enzymatic RNA molecule is incorporated into the vector.

A method for producing enzymatic RNA molecules thus involves providing a vector as described above, expressing a nucleic acid RNA) from that vector, and allowing cleavage by the first enzymatic RNA molecule to release the second (and any subsequent) enzymatic RNA molecule.

Suitable restriction endonuclease sites may also be provided to ease the construction of such a vector -89in DNA vectors or in requisite DNA vectors of an RNA expression system.

The second (and any additional) enzymatic RNA molecule may be any desired type of enzymatic RNA molecule, such as a ribozyme, including group I and group II introns, hammerhead, hairpin, and other types of ribozymes or enzymatically active portions thereof.

The first enzymatic RNA molecule is selected to cleave the encoded cleavage "flanking") sequence, and may also be any desired ribozyme e.g., a ribozyme derived from Tetrahymena which may, for example, include an embedded restriction endonuclease site in the center of a self-recognition sequence to aid in vector construction. This endonuclease site is useful for construction of, and subsequent analysis of, a vector as described herein.

~A vector according to the present invention is preferably operably linked for expression to an appropriate promoter. For example, a vector according 20 to the present invention may comprise an enzymatic RNA molecule under the control of a viral promoter, such as an Epstein-Barr Virus (EBV) promoter. A variety of viral promoters useful for this purpose are known in the art; see, those described in published PCT 25 application no. WO 93/23569, the disclosures of which are incorporated by reference herein.

o In another variation, a vector according to the present invention includes two or more enzymatic RNA molecules. In one embodiment, a first enzymatic RNA S 30 molecule has intramolecular cleaving activity and is able to recognize and cleave nucleotide sequences to release other enzymatic RNA sequences; it is able to function to "release" other enzymatic RNA molecules from the vector. For example, a vector is preferably constructed so that when the first enzymatic RNA molecule is expressed, that first molecule is able to cleave nucleotide sequences flanking additional nucleotide sequences encoding a second enzymatic RNA molecule, a third enzymatic RNA molecule, and so forth.

Presuming said first enzymatic RNA molecule the "releasing" molecule) is able to cleave oligonucleotide sequences intramolecularly, the additional (e.g.

second, third, and so on) enzymatic RNA molecules the "released" molecules) need not possess characteristics identical to the "releasing" molecule.

Alternatively, the first enzymatic RNA molecule may be encoded on a separate vector from the second enzymatic RNA molecule(s) and may have intermolecular cleaving activity. As noted herein, the first enzymatic RNA molecule can be a self-cleaving enzymatic RNA molecule a ribozyme), and the second enzymatic RNA molecule may be any desired type of 15 enzymatic RNA molecule a ribozyme). When a vector is caused to exoress RN from these _nclic -c sequences, that RNA has the ability under appropriate conditions to cleave each of the flanking regions, thereby releasing one or more copies of the second ooo• enzymatic RNA molecule. If desired, several different second enzymatic RNA molecules can be placed in the same cell or carrier to produce different ribozymes.

Methods of isolating and purifying enzymatic

RNA

molecules of the present invention are also e* 25 contemplated. In addition to the methods described e herein, various purification methods those using HPLC) and chromatographic isolation techniques are available in the art. See, the methods described in published international application no. WO 93/23569, the disclosures of which are incorporated herein by reference.

It should also be understood that various combinations of the embodiments described herein are included within the scope of the present invention.

Other features and advantages of the present invention will be apparent from the descriptions hereinabove, from the Examples to follow, and from the claims.

-91-

EXAMPLES

The following examples illustrate, but do not limit, the present invention.

Example 1 In Vitro Evolution of Enzymatic RNA Molecules A. General Principles In vitro selection and in vitro evolution techniques allow new catalysts to be isolated without a priori knowledge of their composition or structure.

Such methods have been used to obtain RNA enzymes with novel catalytic properties. Ribozymes that undergo autolytic cleavage with lead cation have been derived from a randomized pool of tRNA Ph molecules (Pan and Uhlenbeck, Biochemistry 31: 3887-3895 (1992)). Group I ribozyme variants have been isolated that can cleave DNA (Beaidry and Joyce, Science 257: 635-641 (1992)) or that have altered metal dependence (Lehman and Joyce, Nature 361: 182-185 (1993)). Starting with a pool of 20 random RNA sequences, molecules have been obtained that catalyze a polymerase-like reaction (Bartel and Szostak, Science 261: 1411-1418 (1993)). In the present example, refinement of specific catalytic Sproperties of an evolved enzyme via alteration of the selection constraints during an in vitro evolution procedure is described.

~A method of in vitro evolution has now been S. developed for enzyme engineering. For example, the Tetrahymena ribozyme, an RNA enzyme that typically 30 catalyzes sequence-specific phosphoester transfer reactions that result in cleavage or ligation of RNA substrates, is useful in the within-described in vitro evolutionary process. Using the within-disclosed methods, it is now feasible to improve the catalytic efficiency of RNA-catalyzed.DNA cleavage under physiologic conditions, thereby obtaining ribozymes that can cleave DNA in vivo. It is likewise feasible, in view of the present disclosure, to design enzymatic -92- RNA molecules with specific amidase and peptidase activities.

It is not obvious how one should change the Tetrahymena ribozyme to convert it from an RNA-cleaving to a DNA-cleaving enzyme or to an amidase or peptidase.

Thus, directed evolution was selected as a means to acquire the desired phenotype.

Darwinian evolution requires the repeated operation of three processes: introduction of genetic variation; selection of individuals on the basis of some fitness criterion; and amplification of the selected individuals. Each of these processes can be realized in vitro (Joyce, Id. (1989)). A gene can be mutagenized by chemical modification, 15 incorporation of randomized mutagenic oligodeoxynucleotides, or inaccurate copying by a polymerase. (See, Cadwell and Joyce, in PCR Methods and Applications 2: 28-33 (1992); Cadwell and Joyce, PCR Methods and Applications 3 (Suppl.): S136- S140 (1994); Chu, et al., Viroloqv 98: 168 (1979); Shortle, et al., Meth. Enzymol. 100: 457 (1983); Myers, et al., Science 229: 242 (1985); Matteucci, et al., Nucleic Acids Res. 11: 3113 (1983); Wells, et al., Gene 34: 315 (19L5); McNeil, et al., Mol. Cell. Biol. 25 3545 (1985); Hutchison, et al., PNAS USA 83: 710 (1986); Derbyshire, et al., Gene 46: 145 (1986); Zakour, et al., Nature 295: 708 (1982); Lehtovaara, et al., Protein Enq. 2: 63 (1988); Leung, et al., Technique 1: 11 (1989); Zhou, et al., Nucl. Acids Res.

19: 6052 (1991).) The gene product can be selected, for example, by its ability to bind a ligand or to carry out a chemical reaction. (See, Joyce, Id. (1989); Robertson and Joyce, Id. (1990); Tuerk, et al., Id. (1990).) The gene that corresponds to the selected gene product can be amplified by a reciprocal primer method, such as the polymerase chain reaction (PCR). (See, Saiki, et al., Science 230: 1350-54 (1985); Saiki, et al., -93- Science 239: 487-491 (1988).) Alternatively, nucleic acid amplification may be carried out using self-sustained sequence replication (3SR). (See, Guatelli, et al., PNAS USA 87: 1874 (1990), the disclosures of which are incorporated by reference herein.) According to the 3SR method, target nucleic acid sequences may be amplified (replicated) exponentially in vitro under isothermal conditions by using three enzymatic activities essential to retroviral replication: reverse transcriptase, (2) RNase H, and a DNA-dependent RNA polymerase. By mimicking the retroviral strategy of RNA replication by means of cDNA intermediates, this reaction accumulates cDNA and RNA copies of the original target.

In summary, a continuous series of reverse transcription and transcription reactions replicates an RNA target sequence by means of cDNA intermediates.

The crucial elements of this design are the oligonucleotide primers both specify the target and 20 contain 5' extensions encoding the T7 RNA polymerase binding site, so that the resultant cDNAs are competent transcription templates; cDNA synthesis can proceed to completion of both strands due to the degradation of template RNA in the intermediate RNA-DNA hybrid by 25 RNase H; and the reaction products (cDNA and RNA) can function as templates for subsequent steps, enabling exponential replication.

A major obstacle to realizing Darwinian evolution in vitro is the need to integrate mutation and S 30 amplification, both of which are genotype-related, with selection, which is phenotype-related. In the case of RNA enzymes, for which genotype and phenotype are embodied in the same molecule, the task is simplified.

B. Procedures 1. Amplification a. Amplification Method Using a combination of two polymerase enzymes, it is possible to amplify virtually any RNA. (See Kwoh, -94et al., PNAS USA 86: 1173 (1989); Joyce, in Molecular Biology of RNA: UCLA Symposia on Molecular and Cellular Biology, T. R. Cech Liss, NY, 1989, pp.

361-371.) RNA is copied to a complementary DNA (cDNA) with reverse transcriptase and the resulting cDNA is transcribed to RNA with T7 RNA polymerase (T7 Pol).

(See Figs. 2A-C).

Figures 2A and 2B illustrate the general procedure for selective amplification of catalytic RNA enzymatic RNA molecules of the present invention). In Figure 2A, the overall procedure for RNA amplification is shown. "RT" reverse transcriptase; "T7 pol" T7 polymerase; "prom" promoter, and "RNA" represents the enzymatic RNA molecule. In Figure 2B, the procedure 15 for selective amplification based on phosphoester transfer activity of a group I ribozyme is shown. "E" CM4L.LJYkJ 4. £V L.UUZYILl= I .i jUWII.

represents the enzymatic RNA molecule; represents .substrate; represents enzyme/substrate complex; and "EP" represents enzyme/product complex.

Figure 2C illustrates the overall in vitro evolution procedure disclosed herein, in the context of a DNA-cleaving enzyme, which is used as a convenient example. The following steps are illustrated and further described as follows. Step 1 Cleavage of the substrate via phosphoester transfer results in ligation of the 3' portion of the substrate to the 3' end of the ribozyme. Step 2 Selective isothermal amplification of DNA-cleaving ribozymes: first, selective Primer la "hybridizes to the extended 3' terminus of active molecules and initiates cDNA synthesis in the presence of reverse transcriptase next, Primer 2, which contains a T7 promoter sequence, hybridizes to the cDNA and initiates second-strand DNA synthesis; finally, T7 RNA polymerase (T7 pol) produces multiple copies of the selected RNA, each of which can enter a new round of amplification. Step 3 Selective cDNA synthesis employing Primer la and reverse transcriptase. Step 4 PCR amplification employing nonselective Primer lb and Primer 2, restores the original terminus of the ribozyme-encoding gene and introduces occasional mutations. Step 5 In vitro transcription to produce the progeny population of ribozymes.

The foregoing "steps" are further detailed in subsections l.b, 2 and 3 immediately below, where the processes of mutation, selection and amplification are described at length. In general, though, selective amplification of active molecules occurs during transcription as a consequence of the ability of T7 RNA polymerase to generate 200 to 1200 copies of RNA transcript per copy of cDNA template (Chamberlin, et al., in The Enzymes, Vol. 15, P.D. Boyer Academic Press, NY, 1982, pp. 87-108).

The amplification reaction is generally performed in a single test tube at a constant temperature of 37oC res1lting in an increase of 103 to 106 times the original input of RNA after one hour (Guatelli, et al., PNAS USA 87: 1874 (1990); Joyce, in Antisense RNA and 20 DNA, J.A.H. Murray Wiley-Liss, NY, 1992, pp.

353-372). A useful procedure for RNA amplification is described in Beaudry and Joyce, Id. (1992).

b. Example The population of DNA-cleaving ribozymes obtained after 9 generations of in vitro evolution (see Beaudry and Joyce, Id. (1992)) was used as starting material.

It should be understood, however, that ribozymes generated as described in Example 6 below may also be utilized as starting material. The use of ribozymes S 30 derived from group I introns Tetrahymena-derived ribozymes) is described as exemplary.

Ribozymes (0.1 iM) and substrate (0.2 AM) are incubated at 370C for 1 hr in a 100 Ai volume containing 10 mM MgC 2 1 and 30 mM EPPS (pH After ethanol precipitation, a portion of the reaction products (10-50%) was added to a 20 p1 isothermal amplification reaction mixture, containing 10 mM MgCl 2 mM KOAc, 50 mM Tris (pH 5 mM DTT, 2 mM each -96- NTP, 0.2 mM each dNTP, 4 ACi [a- 32 P]GTP, 12.5 U/l MoMLV reverse transcriptase, 50 U/Il T7 RNA polymerase, and pmol each of appropriate primers; the mixture was then incubated at 37 0 C for 2 hours. In experiments designed to optimize DNA cleavage activity, primers 5'-TTTATTTATTTATTT-3' (Primer la, SEQ ID NO 6) and 5'-CTGCAGAATTCTAATACGACTCACTATAGGAGGGAAAAGTTATCAGGC-3' (Primer 2, SEQ ID NO were used. Primer la hybridizes to the 3' portion of the substrate that becomes attached to the 3' end of the ribozyme.

(Primer lb has the sequence 5'-CGAGTACTCCAAAACTAATC-3' (SEQ ID NO Primer Ib hybridizes to the 3' portion of the ribozyme when no substrate or product remains attached. Primers la and Ib, when used, perform 15 similarly.) Primer 2 hybridizes to the 3' end of the resulting cDNA and introduces the T7 promoter sequence.

2. Selection Amplification is performed selectively in that individual RNAs in the population are required to catalyze a particular chemical reaction in order to become eligible for amplification (Joyce, Id. (1989); Robertson and Joyce, Id. (1990); Beaudry and Joyce, Id.

(1992)). One exemplary selection criterion was based on the ability of group I ribozymes to catalyze a S 25 sequence-specific phosphoester transfer reaction involving an oligonucleotide (or oligodeoxynucleotide) substrate. Figure 2B illustrates the procedure for selective amplification based on phosphoester transfer activity of a group I ribozyme.

a. Enhancing DNA Cleaving Activity As described herein, the Tetrahymena ribozyme, an RNA enzyme that typically catalyzes sequence-specific phosphoester transfer reactions that result in cleavage or ligation of RNA substrates, is useful in the withindescribed in vitro evolutionary process. The wild-type enzyme can be used to cleave a single-stranded DNA substrate, albeit only under conditions of high temperature (50 0 C) or high MgC12 concentration -97or both. (See Robertson and Joyce, Id. (1990).)

A

kinetic study showed that, even at 50 0 C, this reaction is inefficient compared to the "native" reaction with an RNA substrate. As noted above, under physiologic conditions 370C, 10mM MgCl 2 the DNA cleavage reaction using wild-type ribozyme is almost undetectable.

To obtain ribozymes that cleave DNA with improved efficiency under physiologic conditions, directed evolution was used to generate and maintain a population of 1013 ribozymes over ten successive generations. Complete access to genotypic and phenotypic parameters for the entire population over the course of its evolutionary history was also maintained.

The amplification was performed selectively in that individual RNAs in the population were required to catalyze a particular chemical reaction in order to become eligible for amplification (Joyce, Id. (1989); 20 Robertson and Joyce, Id. (1990)). The selection was based on the ability of group I ribozymes to catalyze a sequence-specific phosphoester transfer reaction involving an oligonucleotide (or oligodeoxynucleotide) substrate (see Fig. 2B). Figure 2B illustrates the procedure for selective amplification based on phosphoester transfer activity of a group I ribozyme.

The procedure for selective amplification based on phosphoester transfer activity of a group I ribozyme is essentially as follows.

30 The 3' portion of the substrate, d(A 3

(TA

3 3 (SEQ ID NO 21), was transferred to the 3'-terminal guanosine of the ribozyme. Reaction conditions for RNA-catalyzed DNA cleavage were as follows: 1 AM Tetrahymena ribozyme (L-21 form), 10 pM d(GGCCCTCTA 3

(TA

3 3 (SEQ ID NO 17), 10 mM MgCl 2 30 mM N- [2-hydroxymethyl] -piperazine- N-[3-propanesulfonic acid] (EPPS) (pH 37 0 C, 1 hour. Selective amplification occurred as described in subsection 1 above with respect to selective -98amplification of catalytic RNA, with;d((T 3

A)

3

T

3 C) (SEQ ID NO 22) as Primer 1 and with d(ATCGATAATACGACTCACTATAGGAGGGAAAAGTTATCAGGC) (SEQ ID NO 23) as Primer 2. Subsequent selective cDNA synthesis with 0.2 pmol of the selective amplification product under conditions as described in subsection 1 above, but omitting Primer 2 and T7 polymerase.

Subsequent PCR amplification with 0.01 pmol of the selective cDNA synthesis product in a reaction mixture (100 yl volume) containing 0.2 pM d(CGAGTACTCCAAAACTAATC) (Primer Ib; SEQ ID NO 0.2 pM Primer 2 (as above), 0.2 mM dNTPs, 50 mM KC1, 1.5 mM MgC1 2 10 mM tris-HCl (pH 0.01% gelatin, and 2.5 U of Taq DNA polymerase (Perkin-Elmer, Norwalk, CT), 15 cycles of 92 0 C for 1 minute, 45 0 C for 1 minute, and 720C for 1 minute. PCR products were purified by extraction with chloroform and isoamyl alcohol and by precipitation with ethanol, and used to transcribe RNA as described in subsection 3 below.

The product of the reaction was a molecule that contained the 3' portion of the substrate attached to the 3' end of the ribozyme (EP; see Figs. 2A and 2B) Selection occurred when an oligodeoxynucleotide primer was hybridized across the ligation junction and used to initiate cDNA synthesis. The primer did not bind to unreacted starting materials (<10 4 compared to reaction products) and thus led to selective amplification of the catalytically active RNAs.

b. Enhancing Amidase/Peptidase Activity One exemplary procedure for selective amplification based on phosphoester transfer activity of a group I ribozyme is described in Beaudry and Joyce, Id. (1992). Another is described essentially as follows.

Twenty-five percent of the isothermal amplification products (see section B.l.a. above) were used to generate cDNA in a 20 Al reaction mixture containing 10 mM MgC1 2 .50 mM Tris (pH 5 mM DTT, 2 -99mM each NTP, 0.2 mM each dNTP, 0.2 U/pl AMV reverse transcriptase and 20 pmol Primer la, incubated at 37 0

C

for 1 hr. Approximately 5-10% of the resulting cDNA was amplified by the PCR in a 100 gl reaction mixture containing 1.5 mM MgCl 2 50 mM KC1, 10 mM Tris (pH 8.3), 0.1% gelatin, 0.2 mM each dNTP, 20 pmol Primer 1, pmol Primer 2, and 2.5 U Taq DNA polymerase, carried out for 30 cycles of 92 0 C for 1 min, 45°C for 1 min, and 72 0 C for 1 min, and 1 cycle of 720C for 10 min.

Primer lb is complementary to the 3' end of the ribozyme, allowing regeneration of its original, active form. PCR DNA (-250-500 ng, 5-10% of the total) then served as template in an in vitro transcription reaction, carried out in a 25-50 il volume. Errorprone or mutagenic PCR may also be used to generate a higher percentage of variants.

Similarly. when the selection criterion is the ability to bind one or more amino acids, the foregoing procedure is modified to enable the identification and 20 isolation of ribozymes with amino-acid-containing substrate still attached thereto. For example, one or more amino acids in the substrate the terminal amino acids may be "tagged" for identification purposes, via art-recognized procedures. One preferred method of "tagging" or "labeling" substrate amino S' acid(s) involves biotinylation, according to procedures 0 *000 known in the art. (See, Green, et al., Biochem.

J. 125: 781 (1971); Lomant and Fairbanks, J. Mol. Biol.

104: 243-261 (1976); and Mouton, et al., Arch. Biochem.

30 Biophys. 218: 101-108 (1982).) Various reagents and kits for biotinylating amino acids, polypeptides, and proteins are commercially available (see, the biotinylation kits from Pierce Chemicals, Rockford,

IL).

Ribozymes with biotinylated amino acid-containing substrate (or product) attached thereto are then easily identified with the use of a detecting means such as a solid matrix or solid surface containing avidin bound -100thereto or incorporated therein. For example, a sample containing ribozymes admixed with amino acid-containing substrate, wherein said substrate is terminally labeled with biotin, may be run across avidin tips, to "pull out" ribozymes with amino acids or polypeptides attached thereto. Molecules labeled with biotin may easily be detected with indirect immunofluorescence techniques. In addition, a number of fluorochromes, as well as alkaline phosphatase and horseradish peroxidase (which produce colored precipitates) are available directly conjugated to avidin. Streptavidinfluorochrome conjugates are also useful in the S* identification of molecules labeled with biotin and are readily available from various commercial sources 15 Pierce Chem.).

t" Smples coll±ected aftler exposure to avidin may subsequently be subjected to further procedures to separate ribozymes with amino acid-containing product attached thereto from amino acid-containing molecules 20 that are not linked to a ribozyme. Such separations may be done using routine methods, via size separation or via use of a variety of well-known labeling agents and methods. (See, Ausubel, et al. Current Protocols in Molecular Biology, 25 John Wiley Sons, Inc. (1994).) Once the selected enzymatic RNA molecules the ribozymes with attached product) are separated out, they may be prepared for a subsequent amplification step. Preferably, before amplification is initiated, the amino acid-containing product is allowed to dissociate from the ribozyme. As previously noted, adjustment of the pH of the solution may enhance or slow down this dissociation process. Once the ribozymes are "free" 'of attached amino acid-containing product, amplification may be initiated, using appropriate primers, as previously described.

Alternatively, enzymatic RNA molecules capable of binding a particular amino acid (or acids) may be -101identified and removed from a population sample using a column containing one or more amino acids linked to a column matrix,-or via other art-recognized methodologies. For example, if one is seeking to isolate arginine-binding ribozymes, one may take about gg of random "P-labeled ribozymes in water, heated at 65 0 C for 5 minutes, with the salt concentration adjusted to the appropriate level. The RNA is then cooled to 4 0 C over a ten-minute period before the RNA solution in a 25 AL total volume is loaded onto the affinity column. The column, which contains, an L-arginyl-L-cysteine dipeptide linked through the sulfhydryl group to a thiopropyl Sepharose 6B column :matrix (Pharmacia, Piscataway, NJ), is washed at 4°C 15 for approximately 8-12 column volumes, and any remaining RNA sticking to the column is eluted with mM L-arginine in column buffer. cDNA is then synthesized from the arginine-eluted RNA, amplified via PCR, and transcribed into RNA for the next cycle of 20 amplification. (See, Connell et al., Biochemistry 32: 5497-5502 (1993).) It should also be noted that one may adjust the reaction parameters to diminish hydrolysis thus increasing the persistence of the intermediate e.g., by decreasing the pH of the admixture.

Methods of selecting individuals from the population will vary, depending upon the selection criteria applied. For example, if selection of ribozymes that are able to cleave DNA is desired, one may wish to prepare primers that will amplify ribozymes when the DNA-containing product is still attached to the ribozyme. Conversely, when the predetermined selection criterion is the identification of ribozymes with amide-cleaving ability, one may elect to prepare primers that will amplify ribozymes after the amino acid-containing product has dissociated from the ribozyme.

Therefore, if the predetermined selection -102criterion is the identification of amide bond-cleaving enzymatic RNA molecules, the process may be described as follows: obtain a population of ribozymes and place them in an appropriate receptacle; add amide bond-containing substrate molecules to the receptacle, to form an admixture of ribozyme and substrate; (3) maintain the admixture for a sufficient period of time and under predetermined reaction conditions to allow the ribozymes and substrate to interact, to form ribozyme-product complexes; separate the ribozymeproduct complexes from the admixture; allow the ribozyme and product to dissociate from each other; and purify or otherwise separate the ribozymes from the product.

15 As described herein, after step the population may optionally be exposed to mutagenizing :conditions before continuing to step Also as described herein, the selection and separation steps may take advantage of known procedures, the use 20 of biotin labeling of the product and the use of an avidin-coupled support to select out and separate ribozyme-product complexes from the remainder of the admixture. It is also contemplated that the ribozymes identified in step may then be amplified, or may be 25 run through the entire stepwise process one or more subsequent times, with different selection criteria applied each time. It should also be apparent that the foregoing is exemplary and is not intended to limit the scope of the invention.

With regard to amplification, the transcribed RNA is generally isolated by polyacrylamide gel electrophoresis, visualized by UV shadowing, cut and eluted from gel, purified on duPont NENsorb (duPont de Nemours, Wilmington, DE), and quantified spectrophotometrically, as described herein. The entire process is generally repeated 18 times, the first 9 as described in subsection 1 above and the second 9 with the incubation time for the cleavage -103reaction reduced from 1 hr to 5 min. Occasionally, the cDNA was purified to improve the quality of the PCR amplification. To do so, cDNA was synthesiied as above except in the presence of 25-50 ACi 3 2 P]dATP.

Labeled cDNA was isolated by electrophoresis in a polyacrylamide/8 M urea gel, visualized by autoradiography, cut and eluted from gel, and purified on duPont NENsorb.

PCR products are purified by extraction with chloroform and isoamyl alcohol and by precipitation with ethanol, and are used to transcribe RNA as described in subsection 3 below. The product of such a reaction is a molecule that contains the 3' portion of the substrate attached to the 3' end of the ribozyme 15 (EP; see Figs. 2A and 2B). Selection occurs when an oligodeoxynucleotide primer is hybridized across the ligation junction and used to initiate cDNA synthesis.

The primer does not bind to unreacted starting materials (<10- 8 compared to reaction products) and thus 20 leads to selective amplification of the catalytically active RNAs.

3. Introduction of Variation Mutations are introduced in two ways. First, at the outset, a set of mutagenic oligodeoxynucleotides that contain random substitutions at a fixed frequency of occurrence is used. These partially randomized oligodeoxynucleotides are produced on an automated DNA synthesizer with nucleoside 3'-phosphoramidite solutions that have been doped with a small percentage of each of the three incorrect monomers (McNeil, et al., Id. (1985); Hutchison, et al., Id. (1986)).

Second, after each round of selective amplification, random mutations are introduced by performing the PCR under mutagenic conditions (Cadwell and Joyce, PCR Methods and Applications 2: 28-33 (1992); Cadwell and Joyce, PCR Methods and Applications 3 (Suppl.): S136- S140 (1994)).

To generate the initial population of ribozyme -104variants, random mutations are introduced throughout the catalytic core of the molecule. In one example, four synthetic oligodeoxynucleotides are prepared, each of which randomly mutagenizes 35 nucleotide positions at an error rate of 5% per position (not shown). The transcription conditions are essentially as follows: 2 pmol of DNA template (containing mutagenic oligodeoxynucleotides), 2 mM NTP's, 15 mM MgCl 2 2 mM spermidine, 5 mM DTT, 50 mM tris-HCl (pH 1500 U of T7 RNA polymerase are admixed to a volume of 60j2l and held at 37 0 C for 2 hours. RNA is purified by electrophoresis in a 5% polyacrylamide-8M urea gel and subsequent column chromatography on Sephadex The degenerate oligodeoxynucleotides are 15 incorporated into a DNA template that encodes the ribozyme, and the template is transc-ribed directly to, produce the mutant RNAs (Joyce and Inouye, Nucl. Acids Res. 17: 711 (1989)). Twenty pmol (10 1 3 molecules) of material is used at the beginning. Thus, the 20 generation 0 population is expected to contain the wild-type ribozyme, all possible and 4error mutants, and a sampling of the higher-error mutants (see Table 2 in Example 4 below).

In general, when using PCR procedures, each primer works in combination with a second primer to amplify a Starget nucleic acid sequence. The choice of PCR primer pairs for use in PCR is governed by various considerations, as discussed herein. That is, the primers have a nucleotide sequence that is complementary to a sequence conserved in the gene of choice. Useful priming sequences have been disclosed herein Primers 1, Ib, and The strategy used for cloning the selected genes will depend, as is well known in the art, on the type, complexity, and purity of the nucleic acids making up the various genes.

Other factors include whether or not the genes are to be amplified and/or mutagenized.

Typically, the exemplary genes are comprised of -105polynucleotide strands, such as mRNA, cDNA, or the sense strand of genomic DNA, although antisense strands, rRNA, or tRNA may also be used in PCR. If the polynucleotide sequence is in the form of double stranded genomic DNA, it is usually first denatured, typically by melting, into single strands. A gene sequence is subjected to a PCR reaction by treating (contacting) the sequence with a PCR primer pair, each member of the pair having a preselected nucleotide sequence. The PCR primer pair is capable of initiating primer extension reactions by hybridizing to nucleotide sequences, preferably at least about 10 nucleotides in length and more preferably at least about nucleotides in length, conserved within the gene 15 sequence. Primer extension via PCR may be carried out from either end of the molecule, through the amide or through the carboxyester, as desired.

The PCR reaction is performed by mixing the PCR primer pair, preferably a predetermined amount thereof, 20 with the nucleic acids of the selected gene or DNA nucleotide sequence (a predetermined amount thereof, Spreferably) in a PCR buffer to form a PCR reaction admixture. The admixture is maintained under polynucleotide synthesizing conditions for a time 25 period which is typically predetermined sufficient for the formation of a PCR reaction product, 'thereby producing a plurality of different DNA homologs.

The PCR reaction is performed using any suitable method. PCR amplification methods are described in detail in U.S. Patent Nos. 4,683,192, 4,683,202, 4,800,159, and 4,965,188, and at least in several texts including "PCR Technology: Principles and Applications for DNA Amplification", H. Erlich, ed., Stockton Press, New York (1989); and "PCR Protocols: A Guide to Methods and Applications", Innis et al., eds., Academic Press, San Diego, California (1990). Thermus aquaticus DNA polymerase I, which is useful in PCR, is described -106in U.S. Patent No. 4,889,918. (The relevant disclosures of the cited'patents are incorporated by reference herein.) Restriction sites may also be incorporated into the 5' and 3' primers to enable the amplification products to be subcloned into sequencing or expression vectors. It may also be helpful to place a 4-base spacer sequence proximal to the restriction site to improve the efficiency of cutting amplification products with enzymes.

In the presently-described examples, PCR was performed under standard reaction conditions, resulting S" in an error rate of approximately 0.1% per position per generation. A mutagenic, modified PCR procedure that 15 provides an error rate of 0.66 0.13% per position Cadwell and Joyce, PCR Methods and Applications 2: 28- 33 (1992), and Cadwell and Joyce, PCR Methods and Applications 3 (Suppl.): S136-S140 (1994), the 20 disclosures of which are incorporated herein by reference.) The RNAs obtained by selective amplification are subjected to reverse transcription, the resulting cDNAs are PCR amplified, and the PCR products are transcribed to produce a progeny 25 distribution of mutant RNAs.

Integration cf the PCR with the selective RNA amplification procedure is useful in three other ways.

First, it increases the overall amplification by about 103 times. Second, it simplifies the process of subcloning individuals from the evolving population.

Normally, only a small portion of the DNA in the RNA amplification mixture is fully double-stranded, but with the PCR, the amount of double-stranded DNA (dsDNA) is greatly increased. Third, it returns the RNA to a form that can participate in the RNA-catalyzed phosphoester transfer or amide-cleavage reaction.

After phosphoester transfer or amide cleavage, the ribozyme has the 3' portion of the substrate attached -107go a a to its 3' end, and after selective RNA amplification, the substrate sequence remains attached for a time (see Figs. 2A and 2B). However, by subsequent use of PCR, followed by in vitro transcription, the original 3' end of the ribozymes is restored.

Therefore, the entire mutation, selection and amplification process the method of engineering enzymatic RNA molecules capable of cleaving an amide bond may conveniently be described according to the following stepwise procedure.

1. Obtain a population of ribozymes; 2. Introduce genetic variation into the population; 3. Identify individuals from the resulting "mutant" population that are able to meet predetermined selection criteria; 4. Separate the identified (or selected) individuals from the remainder of the population; Prepare appropriate primers; and 20 6. Amplify the selected individuals.

The foregoing steps may be repeated as many times as desired to generate numerous variant populations.

In various embodiments, it is contemplated that the amplified population produced in step six will be used as the "starting population" in the next "generation" beginning with step one.

As those of skill in the art will appreciate, step need not be performed in the time sequence indicated.

That is, primers may be prepared at any time; presumably, preparation of primers is based on understanding of the predetermined selection criteria.

For example, if the predetermined selection criterion is the identification of ribozymes that are able to cleave DNA, one may wish to prepare primers that will amplify ribozymes when the DNA-containing product is still attached to the ribozyme. Conversely, when the selection criterion is the identification of ribozymes with amide-cleaving ability, one may elect to prepare -108primers that will amplify ribozymes after the amino acid-containing product has dissociated from the ribozyme.

4. Substrate Cleavage Activity The entire series of events, beginning with a heterogeneous population of RNAs, proceeding with RNA catalysis in the target reaction, selective amplification of catalytically active RNAs, reverse transcription of the selective amplification products, mutagenic PCR, and in vitro transcription to produce a progeny distribution of RNAs, is referred to as one "generation". Typically, a generation is completed in one to two working days, excluding time for analytic work. The initial population of mutant RNAs is 15 referred to as "generation while subsequent populations Are referred f- as "generatinI, "generation and so forth. In principle, there is no limit to the number of successive generations that can be obtained.

20 Typically, each generation begins with 20 pmol of RNA. The amount of RNA is again quantified after selective amplification and after transcription.

In practice, there is always the danger of odeveloping a "parasite" that circumvents the selection 25 criterion and is amplified more efficiently than the most reactive species. For example, a sequence may arise which allows false hybridization of one of the amplification primers at an internal site, generating a species with a nucleotide deletion that may be amplified more efficiently than the full-length ribozyme. Thus, it is important to monitor the populations generated and remove such "parasites", if and when they appear.

a. DNA-Cleaving Populations To generate the initial population of ribozyme variants, random mutations were introduced throughout the. catalytic core of the molecule. Four synthetic oligodeoxynucleotides were prepared, each of which -109randomly mutagenizes 35 nucleotide positions at an error rate of 5% per position.

Figure 3A is a diagrammatic representation of the secondary structure of the Tetrahymena ribozyme (L-21 form). Figure 3B is a similar diagram of the L-21 form of Tetrahymena ribozyme; the diagram shows those regions that were randomly mutagenized (boxed segments). The transcription conditions were essentially as follows: 2 pmol of DNA template (containing mutagenic oligodeoxynucleotides), 2 mM NTP's, 15 mM MgC1 2 2 mM spermidine, 5 mM DTT, 50 mM tris-HCl (pH 1500 U of T7 RNA polymerase were admixed to a volume of 60yl and held at 37 0 C for 2 hours. RNA was purified by electrophoresis in a 15 polyacrylamide-8M urea gel and subsequent column chromatography on Sephadex The degenerate oligodeoxynucleotides were incorporated into a DNA template that encodes the ribozyme, and the template was transcribed directly to produce the mutant RNAs (Joyce and Inouye, Nucl. Acids "Res. 17: 711 (1989)). Twenty pmol (10" molecules) of material was used at the beginning. Thus, the generation 0 population was expected to contain the wild-type ribozyme, all possible and 4error mutants, and a sampling of the higher-error mutants (see Table 2).

Table 2 illustrates the composition of the initial population (generation The probability of having errors in a doped oligonucleotide of length v and degeneracy d is given by: P(k,v,d) k) !k)]dk(l-d) v k. A total of 140 positions were randomly mutagenized (v 140) at a degeneracy of 5% per position (d 0.05). The number of distinct k-error sequences of length v is given by: Nk !k!]3k.

The expected number of copies per sequence is based on a population size of 20 pmol (1.2 X 10' molecules).

-110- Table 2 Probability Errors Sequences Copies/Sequence 0 (wt) 0.1 1 9 X 109 1 0.6 420 2 X 10 8 2 2.1 9 X 10 4 3 X 10 6 3 5.0 1 X 10 7 5 X 10 4 4 9.0 1 X 10 9 9 X 10 2 12.8 1 X 10"1 6 15.2 7 X 1012 0.3 7+ 55.4 The evolution experiment spanned ten successive generations; each generation began with 20 pmol of RNA.

The amount of RNA was quantified after selective amplification and after trancription (see Fig. 4) Figure 4 illustrates the course of evolution over successive generations, highlighting changes in RNA population size over time. Closed circles represent RNA population size after transcription, quantitated by 3 H]uracil content; open circles represent RNA population size at the start of each generation, based on 20-pmol portions; closed squares represent RNA population size after reaction with substrate, estimated by the assay described herein; and open squares represent RNA population size after selective amplification, quantitated by acid precipitation at of [a- 32 P]GTP-labeled progeny RNA.

DNA cleavage activity for the population as a whole was monitored by a gel electrophoresis assay involving cleavage of 32 P]-labeled d(GGCCCTCT-

A

3

(TA

3 3 (SEQ ID NO 17) to yield d(GGCCCTCT) (SEQ ID NO 16) (data not shown). Cleavage of the substrate d(GGCCCTCT-A 3

(TA

3 3 (SEQ ID NO 17) in the absence of enzyme, in the presence of the wild-type Tetrahymena ribozyme (L-21 form), and in the presence of the population of RNAs obtained at each generation n -111- 0-10) was measured (data not shown).

Reaction conditions were as follows: 0.5 gM ribozyme, 0.1 gM substrate (2.6 iCi/pmol), 30 mM EPPS (pH either 10 mM MgCl2, 37°C, 1 hour (low) or mM MgCl 2 2 mM spermidine, 50 0 C, 1 hour (high) Reaction products were separated by electrophoresis in a 20% polyacrylamide-8M urea gel, of which autoradiograms were made; P represented "P]d(GGCCCTCT) (SEQ ID NO 16), and represented "P]d(GGCCCTCTA) (SEQ ID NO 24) (not shown).

It is generally expected that any given mutation would more likely be detrimental than beneficial, although there may be a substantial number of neutral mutations. Indeed, DNA cleavage activity for the 15 generation 0 population is less efficient than for the wild-type The generation 1 population, having been selected for DNA cleavage activity under physiologic conditions, showed improved catalytic activity compared to generation 0 and was slightly 20 improved over the wild-type. Through successive generations, there was continued improvement of phenotype. By generation 7, the population as a whole cleaved DNA more efficiently at 37 0 C and 10mM MgCl 2 than does the wild-type at the high-temperature, high-MgCl 2 condition. Through generation 10, the rate of improvement had yet to level off.

RNAs from each generation were purified by polyacrylamide gel electrophoresis and Sephadex chromatography. To provide a more formal assay of DNA cleavage activity, d(GGCCCTCTA 3

(TA

3 3 3 P]A) substrate (SEQ ID NO 26) was prepared as follows, and formation of both the ribozyme-d(A 3

(TA

3 3 A) covalent intermediate and the RNA-catalyzed site-specific hydrolysis product d(A 3

(TA

3 3 A) (SEQ ID NO 25) was measured (see Fig. (See also Inoue, et al., J. Mol. Biol. 189: 143 (1986).) Figure 5 illustrates the cleavage of -112labeled d(GGCCCTCT-A 3 (TA3)3[5' 32 pA) (SEQ ID NO 26).

Cleavage of 32 P]dA-labeled d(GGCCCTCT-A 3

(TA

3 3 3 P]A) was conducted under reaction conditions as described hereinabove prior to autoradiogram.

Substrate enzyme/product and product (P) (see Fig. 2) were separated by electrophoresis in a polyacrylamide-8M urea gel. Individual bands were cut from the gel and quantitated by Cerenkov counting. EP of total) is plotted against ribozyme generation and is compared with data using wt ribozymes under "low" and "high" conditions, as discussed above. Data points are the average of five replicate experiments performed on three different days with two different preparations S*of substrate. Error bars correspond to ±1 SD.

15 Substrate was prepared via the following p e Te L Z J'-"-labeled DNA substrate was prepared with terminal deoxynucleotide transferase.

Reaction conditions were as follows: 4 AM d(GGCCCTCT-

A

3

(TA

3 3 (SEQ ID NO 17), 1 iM ([-nP]dATP (3~Ci/pmol), 20 imM CoC12, ImM DTT, 50mM potassium cacodylate (pH 7.2) and terminal transferase (BRL) at 2.7 U/Il, incubated at 370C for 30 minutes.

The product corresponding to addition of a single dA residue was purified by electrophoresis in a 25 polyacrylamide-8M urea gel and subsequent affinity chromatography on Nensorb (duPont, Wilmington, DE).

The hydrolysis product forms either by direct cleavage of the DNA substrate or by cleavage of the ribozymed(A 3

(TA

3 3 A) covalent intermediate. Together, these reactions account for less than 5% of the cleaved substrate.

After ten generations, DNA cleavage activity for the population as a whole was 30 times higher than that of the wild-type. Because selection is based on primer hybridization to the EP covalent intermediate (see Fig.

2B), there is selection pressure against the subsequent site-specific hydrolysis reaction. As a consequence, -113the efficiency of the hydrolysis reaction relative to the initial phosphoester transfer event drops from 4.9% for the wild-type to 1.5% for the generation population. There is selection pressure favoring accurate cleavage of the DNA at the target phosphodiester; inaccurate cleavage would result in partial mismatch of the primer used to initiate selective amplification. The accuracy of cleavage at first declines from 90% for the wild-type to 45% for the generation 8 population, and then rises to 60% for the generation 10 population. There are some individuals in the population that sacrifice accuracy for improved cleavage activity in order to enjoy an .overall selective advantage (see below). Of course, a 15 preferred result is an individual having both high accuracy and high cleavage activity.

b. Amidase/Peptidase Populations Substrate cleavage activity for the population as a whole is generally monitored via gel electrophoresis 20 assay involving cleavage of 3 P)-labeled substrate to yield a specific product. Cleavage of the substrate in the absence of enzyme, in the presence of the wild-type Tetrahymena ribozyme (L-21 form), and in the o presence of the population of RNAs obtained at each 25 generation (Gn, beginning with a value of 0 for n) is measured.

Reaction conditions will vary depending on various S* parameters, substrate recognition, affinity, cleavage, etc. In general, reaction conditions were essentially as follows: 0.5 AM ribozyme, 0.1 AM substrate (2.6 ACi/pmol), 30 mM EPPS (pH and mM MgC 2 1 are admixed and maintained at 37 0 C for about 1 hour. Reaction products were separated by electrophoresis in a 20% polyacrylamide-8M urea gel, of which autoradiograms were made.

One usually expects that any given mutation will more likely be detrimental than beneficial, although there may be a substantial number of neutral mutations.

-114- Through successive generations, however, continued improvement of phenotype was observed to occur, and in further succeeding generations, the rate of improvement is expected to increase. RNAs from each generation are usually purified by polyacrylamide gel electrophoresis and Sephadex chromatography. To provide a more formal assay of cleavage activity, 32 p]-labeled substrate was prepared as follows, and formation of both the ribozyme-coupled covalent intermediate and the RNAcatalyzed site-specific cleavage product is measured.

(See also Inoue, et al., J. Mol. Biol. 189: 143 (1986).) Cleavage of "P-labeled substrate is generally conducted under reaction conditions as described hereinabove prior to autoradiogram. Substrate 15 enzyme/product and product are separated by *I electrophoresis in a 20% polyacrylamide-8M urea gel.

Individual bands are cut from the gel and quantitated by Cerenkov counting. On the average, five replicate experiments are performed on three different days with 20 two different preparations of substrate, before data points are plotted (not shown).

5. Preparation and Sequencing of Subclones Although evolution in natural populations is an accomplished fact, evolution in vitro is a work in 25 progress that allows the experimenter to access any time period in evolutionary history. Subclones are S. S S" obtained from the evolving population at every generation and individual ribozymes are then sequenced.

a. DNA-Cleavin Populations Subclones were obtained from the evolving population at every generation, essentially as follows.

DNAs used to transcribe the population of RNAs at each generation (see subsections 1-3 above) were amplified in a second PCR reaction with primers CCAAGCTTGATCTCGAGTACTCCAAAACTAATC-3' (SEQ ID NO 27) and 5'-CTGCAGAATTCTAATACGACTCACTATAGGAGGGAAAAGTTATCAGGC-3' (SEQ ID NO producing a 435-bp (base pair) fragment with unique restriction sites at its ends. The -115fragment was digested with Eco RI and Hind III and ligated into a pUC18 vector that had been linearized with Eco RI and Hind III and purified in a 1.5% agarose gel. The resulting plasmid DNAs were used to transform competent DH5a-F' cells (see Hanahan, in DNA Cloning: A Practical Approach, D.M. Glover, ed., IRL Press, Oxford, 1985, pp. 109-135), which were then grown on ampicillin-containing plates. Individual colonies were chosen at random and grown overnight in liquid media.

DNA was prepared by the boiling lysis method (Holmes, et al., Anal. Biochem. 114: 193 (1981)) and screened for the insert by restriction digestion.

As noted, subclones were obtained from the evolving population at every generation. Generations 15 3, 6, and 9 were chosen for detailed analysis. DNA was prepared from 25 subclones at generations 3 and 6 and from 50 subclones at generation 9. The nucleotide sequence of the entire ribozyme gene was determined for each of these subclones essentially as follows.

20 Cloned individuals were sequenced by the dideoxy chain-termination method (Sanger, et al., PNAS USA 74: 5463 (1977); Zagursky, et al., Gene Anal. Tech. 2: 89 (1985)) with reciprocal primers 5'-GTAAAACGACGGCCAGT-3' (SEQ ID NO 9) and 5'-CATGATTACGAATTCTA-3' (SEQ ID NO 10), which are compatible with the pUC plasmid.

Sequencing reactions utilized modified T7 DNA polymerase (Sequenase, USB) and 3 5 S] (a-thiophosphate) dATP and were analyzed by electrophoresis in a 6% polyacrylamide-8M urea gel. Nucleotide sequences of individual subclones were also obtained (not shown).

Analysis of the determined sequences indicated how genotype changes over the course of evolutionary history (not shown). From generation 0 to generation 3, variation was discarded throughout much of the catalytic core of the ribozyme. The mean number of mutations per subclone decreased from 7.0 at generation 0 to 2.7 at generation 3. By generation 3, a small number of mutations outside of the original zone of -116random mutation had occurred because of ongoing mutation events (not shown). The consensus sequence was still that of the wild-type, although only one of subclones had the entire wild-type sequence.

From generation 3 to generation 6, the dramatic accumulation of mutations at five positions within the ribozyme coincided with a three-fold improvement in the phenotype of the population as a whole. From generation 6 to generation 9, these positions were further accentuated and aggregate phenotype improved another three-fold. The mean number of mutations per subclone rose to 4.6 at generation 6 and to 5.9 at generation 9 as a larger proportion of subclones *oo. adopted the common mutations and as mutations 15 accumulated outside of the original zone of random mutation.

The most frequent mutation was an A-Y (Y U or C) change at position 94 This mutation was present, as A-U, in only 1 of 25 subclones at generation 3. At generation 6, there were 15 out of occurrences, 12 as A-U and 3 as A-3C; at generation 9, there were 35 out of 50 occurrences, 22 as A-U and 13 as A-C. Position 94 was unpaired in the secondary structure of the wild-type ribozyme (Burke, et al., Nucleic Acids Res. 15: 7217 (1987)). Considering the effect of site-directed mutations made at neighboring positions (Young, et al., Cell 67: 1007 (1991)), the 94:A-Y change may alter the orientation of ribozymebound substrate relative to the catalytic core of the molecule.

Another frequent mutation, occurring in 4 of subclones at generation 3, 6 of 25 subclones at generation 6, and 22 of 50 subclones at generation 9, was a G->A change at position 215. This mutation converts a G-U wobble pair to an A-U Watson-Crick pair within the catalytic core of the ribozyme. Among 87 group I intron sequences that have been analyzed, 39 had a G-U and 28 had a G-C, but only 4 had an A-U at -117this location (Michel, et al., J. Mol. Biol. 216: 585 (1990)).

The most remarkable mutations were. G-U change at position 313 and an A-G change at position 314 that always occur together. These mutations were absent at generation 3, but were present in 5 of 25 subclones at generation 6 and 16 of 50 subclones at generation 9.

The GA sequence normally present at positions 313-314 is thought to form a short duplex structure (the half of the P9.0 pairing) that draws the 3'-terminal guanosine residue of the ribozyme into the catalytic core (Michel, et al., Nature 342: 391 (1989); Michel, et al., Genes Dev. 4: 777 (1990)). The 3'-terminal guanosine was utilized as the nucleophile in the target 15 phosphoester transfer reaction. The 313-314 mutations are expected to destroy the P9.0 pairing, yet confer selective advantage with respect to DNA cleavage (see below).

There was a frequent G-iA change at position 312 20 that occurs only if the 313-314 mutations are not present. The 312:G->A change was present in 4 of. subclones at generation 3 and 8 of 25 subclones at generation 6, but only 5 of 50 subclones at generation S9. In terms of population frequency, the 312:GA mutation declined as the 313-314:GA-UG mutations became more abundant.

b. Amidase/Peptidase Populations Preparation of Subclones One useful method of preparing subclones is described in Beaudry and Joyce, Science 257: 635-641 (1992). For example, DNAs used to transcribe the population of RNAs at each generation were amplified in a second PCR reaction with appropriate primers, producing a 435-bp (base pair) fragment with unique restriction sites at its ends. The fragment was digested with Eco RI and Hind III and ligated into a pUC18 vector that had been linearized with Eco RI and Hind III and purified in a 1.5% agarose gel. (See -118- Beaudry and Joyce, Id. (1992).) The resulting plasmid DNAs were used to transform competent DH5u-F' cells (see Hanahan, in DNA Cloning: A Practical Approach, D.M. Glover, ed., IRL Press, Oxford, 1985, pp. 109- 135), which were then grown on ampicillin-containing plates. Individual colonies were chosen at random and grown overnight in liquid media. DNA was prepared by the boiling lysis method (Holmes, et al., Anal.

Biochem. 114: 193 (1981)) and screened for the insert by restriction digestion.

Another useful method of preparing subclones is as follows. Subclones were obtained using the Invitrogen TA Cloning Kit (Invitrogen, San Diego, CA). The PCR DNA at G27 was ligated into a linearized plasmid, and 15 the resulting DNA was used to transform competent INVaF' cells, which were grownn ampicillin/X-ga plates. Individual colonies containing the insert were identified by their white color, chosen at random, and grown overnight in liquid media. Plasmid DNA was prepared by the boiling, lysis method (Holmes Quigley, Anal. Biochem. 114: 193-197 (1981)) and screened for the presence of insert by restriction digestion.

Sequencing 25 As noted above, subclones may be obtained from the evolving population at every generation. Specific generations may also be chosen for detailed analysis.

The nucleotide sequence of the entire ribozyme gene is determined for each of these subclones essentially as follows.

Cloned individuals are generally sequenced by the dideoxy chain-termination method (Sanger, et al., PNAS USA 74: 5463 (1977); Beaudry and Joyce, Id. (1992); Zagursky, et al., Gene Anal. Tech. 2: 89 (1985)) with reciprocal primers 5'-GTAAAACGACGGCCAGT-3' (SEQ ID NO 9) and 5'-CATGATTACGAATTCTA-3' (SEQ ID NO 10), which are compatible with the pUC plasmid. Sequencing reactions utilized modified T7 DNA polymerase -119- (Sequenase, USB) and 35 S] (a-thiophosphate) dATP and were analyzed by electrophoresis in a 6% polyacrylamide-8M urea gel. Nucleotide sequences of individual subclones were also obtained (not shown).

Individual ribozymes were prepared as follows: the gene encoding the ribozyme was amplified by the PCR using Primer lb and Primer 2; the resulting DNA was used as a template for in vitro transcription; the RNA products were isolated by polyacrylamide gel electrophoresis, and were purified and quantified as described above. (See also Tsang and Joyce, Biochemistry 33: 5966-5973 (1994).) Analysis of the determined sequences indicates how genotype changes over the course of evolutionary 15 history. From generation 0 to generation 3, variation :i.

introduced into the original ribozyme template by the use of mutagenic primers to produce generation 0 was discarded throughout much of the catalytic core of the ribozyme. The mean number of mutations per subclone 20 decreased from 7.0 at generation 0 to 2.7 at generation 3. By generation 3, a small number of mutations outside of the original zone of random mutation in the catalytic core of the ribozyme have occurred because of ongoing mutation events. The consensus sequence still 25 tends to be that of the wild-type. Analysis of subsequent generations suggests that accumulation of 9 mutations coincides with improvement in the phenotype of the population as a whole. The mean number of mutations per subclone was also observed to increase, as a larger proportion of subclones adopt the common mutations and as mutations accumulated outside of the original zone of random mutation.

The relation between genotype and phenotype in the context of an RNA-based evolving system can readily be formalized once catalytic, kinetic, and comparable data are collected and analyzed. Genotype can be represented as a matrix A, the rows corresponding to individuals in the population and the columns -120corresponding to functionally significant positions within the nucleotide sequence. An exemplary analysis is illustrated in Beaudry and Joyce, Science 257: 635- 641 (1992).

The data obtained from a relatively small number of individuals may not be sufficient to provide a meaningful solution to the relation of genotype to phenotype, even for those nucleotide positions that are known to be most significant based on their high frequency of accepted mutation. One may then elect to use an appropriate weighing vector as a guide to help decide which mutations are sufficiently important to Swarrant individual study. (See, Beaudry and Joyce, Id. (1992).) 15 6. Site-Directed Mutaqenesis Individual enzymatic RNA molecules containing single or multiple point mutations may be prepared via ".site-directed mutagenesis for analysis of the relative significance of a particular mutation. Catalytic 20 activity is then studied with an appropriate labeled oligodeoxyribonucleotide substrate. Sitedirected mutagenesis is carried out essentially as described in Morinaga, et al., Biotechnoloqy 2: 636 (1984), which may be described as follows.

25 Plasmid pT7L-21 (Zaug, et al., Biochemistry 27: 8924 (1988)) is digested with either Eco RI and Hind III to remove the ribozyme coding region, or (ii) Bsa I and Xmn I to remove the ampicillin-resistance gene. The resulting fragments are purified in a 1% agarose gel and cross-hybridized in the presence of a synthetic oligodeoxynucleotide that introduces the desired mutation. The annealing mixture typically contains 0.06 pmol of pT7L-21 (AEcoRI- HindIII), 0.06 pmol pT7L-21(ABsaI-XmnI), 15 pmol of mutagenic oligodeoxynucleotide, 40 mM Tris-HCl (pH and 8 mM MgSO 4 in 1 2 -pl volume, which is heated to 100 0 C for three minutes, then incubated at 30 0 C for minutes, and 0 0 C for 10 minutes.

-121- The annealing product is made fully doublestranded with the Klenow fragment of E. coli DNA polymerase I (Boehringer-Mannheim, Indianapolis, IN) and T4 DNA ligase Biochemical, Cleveland, OH) and is then used to transform competent DH5u-F' cells, which are grown on ampicillin-containing plates.

Colonies are screened by the colony hybridization method with [5'-"P]-labeled mutagenic oligodeoxynucleotide as a probe (Grunstein, et al., PNAS USA 72: 3961 (1975)). DNA is prepared from positive colonies and sequenced throughout the ribozyme gene, as described above.

RNA is subsequently prepared from the DNA template by in vitro transcription, which is performed essentially as follows. Transcription conditions: 2 pmol of DNA template (containing mutagenic oligodeoxynucleotides), 2 mM nucleotide triphosphates (NTPs), 15 mM MgClz, 2 mM spermidine, 5 mM dithiothreitol (DTT), 50 mM tris-HCl (ph 1500 U 20 of T7 RNA polymerase; 60 Al volume; 37 0 C, 2 hours. RNA is purified by electrophoresis in a 5% polyacrylamide-8 M urea gel and subsequent column chromatography on Sephadex The foregoing procedures may be repeated as many 25 times as desired to produce enzymatic RNA molecules having one or more point mutations at one or more preselected sites. For example, in addition to use of the within-disclosed in vitro evolution methods to design and identify ribozymes capable of binding amino 30 acids in a polypeptide sequence and cleaving the bond linking adjacent amino acids at a predetermined site, one may use site-directed mutagenesis techniques as disclosed herein to modify the active site on an enzymatic RNA molecule to accomplish the same objective.

For example, one may use the within-disclosed techniques to modify the recognition site on a preselected ribozyme, by altering the nucleotide -122sequence of said site to exactly duplicate, or substantially mimic, a consensus nucleotide sequence which is able to bind one or more particular amino acids. Exemplary consensus sequences which may be incorporated into the recognition site of enzymatic

RNA

molecules according to the within-disclosed methods are available in the art and include those described in Connell, et al., Science 264: 1137-1141 (1994); Connell, et al., Biochemistry 32: 5497-5502 (1993); and Famulok, J. Am. Chem. Soc. 116: 1698-1706 (1994), to name a few examples. Other useful sequences may be identified using the methods described herein; for example, see section B.2 above.

7. Kinetic Analysis Reduction in reaction time tends to favor selection of enzymatic RNA molecules with increased k values. Representative ribozymes may be chosen from the evolving population and analyzed at each generation to determine km and KM values for the individuals 20 selected. It is to be appreciated that the k, and KM values of the selected ribozymes are not necessarily equivalent to the average values for the entire population, however.

Cleavage reactions are generally carried out at 37 0 C in 10 mM MgCl 2 30 mM EPPS (pH and 40 yg/pl BSA, using 32 P)-labeled substrate. BSA is added to prevent oligonucleotides from adhering to the walls of the 500 tl Eppendorf tubes, and does not affect the course of the reaction. Ribozyme and substrate are S 30 preincubated separately for 15 min at 370C, and then mixed to initiate the reaction. Typically, 5 aliquots of 3-10 il each are removed from the reaction mixture at specified times and quenched by addition to 1-2 volumes of an ice-cold mixture containing 8 M urea, 50-100 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue, 10% SDS, 9 mM Tris-borate (pH and sucrose. Substrate and product are separated by electrophoresis in a 20% polyacrylamide/8 M urea gel, -123visualized by autoradiography, excised from gel, and quantified by Cerenkov counting.

KM and k, values are determined in experiments with substrate in excess over ribozyme Initial rates of reaction over a range of substrate concentrations, are estimated from the initial linear phase, generally the first 5% or less of the reaction. Typically 8 data points were fit by a least squares method to a theoretical line given by the equation: v "KM V, Single-turnover experiments are performed with ribozyme in excess of substrate (Herschlag Cech, Biochemistry 29: 10159-10171 (1990b)). Initial rates (kob s are obtained using no more than the first 5% of the reaction. Given that each kob value, obtained at different ribozyme concentrations, provided an estimate of k,/KM. Generally 8 or more measurements of k,/KM are obtained.

Specific catalytic properties of an amide-cleaving o. 20 ribozyme can be optimized by appropriate manipulation *of the selection constraints during an in vitro evolution procedure. For example, beginning with a heterogeneous population of ribozymes, enriched for modest amide bond-cleavage activity, successive S 25 generations are produced to obtain ribozymes with amidase activity that have successively-improved Scatalytic rates and substrate binding affinities.

8. Determination of Binding Constants The equilibrium dissociation constant, KD, of the 30 complex between ribozyme and product is determined by gel-shift analysis in a native polyacrylamide gel (Pyle et al., PNAS USA 87: 8187-8191 (1990)). Ribozyme at twice final concentration is preincubated at 370C for 15 min in 10 mM MgC 2 1 and 30 mM EPPS (pH 7.5) before mixing with an equal volume of 0.05-1 nM DNA product in 10 mM MgCl 2 30 mM EPPS (pH 0.05% xylene cyanol, 3% glycerol, and -124pg/pl BSA. The mixture is allowed to equilibrate at 37 0 C for 15-60 min before loading on a polyacrylamide gel containing 10 mM MgCl 2 and 30 mM EPPS (pH The electrophoresis buffer also contains mM MgCl 2 and 30 mM EPPS (pH The gel is run at 6 milliamps in a 37 0 C room until the sample has entered the gel (-10 min), and is then moved into a 4 0 C cold room where the current is increased to 30 milliamps.

This is done to prevent the temperature of the gel from rising above 37 0 C. The ribozyme-product complex and free product are visualized by autoradiography, cut from the gel, and quantified by Cerenkov counting.

A binding curve is generated by plotting the percentage of product bound to ribozyme bound) over a range of ribozyme concentrations. KDis determined by fitting the data to a theoretical bindina curve using a least squares method. Where ribozyme is in vast excess o ver product, the theoretical binding curve may be ".oe represented by the equation: bound

K)

S 20 where KD when half of the total product is bound to the ribozyme.

The substrate need not be a nucleotide or nucleotide analog. The only requirement is that RNAs that react with the substrate become tagged in some way so that they can be distinguished from nonreactive molecules with respect to the amplification process.

For example, reactive RNAs become joined to a portion of the substrate that is attached to a solid support, while nonreactive RNAs are washed away, leaving the 30 bound RNAs to be selectively amplified. These and other methodologies are further described elsewhere herein.

9. Extension of Directed Evolution to Develop Additional Evolved Species As an in vitro model of Darwinian evolution, a population of macromolecular catalysts was directed toward the expression of novel catalytic function. In the Examples presented herein, the development of -125ribozymes that cleave DNA and those that demonstrate amide bond-cleaving activity with improved efficiency under physiologic conditions has now been demonstrated.

a. Evolution In Vitro Beginning with any generation of a population of ribozymes as described herein, successive generations of in vitro evolution are carried out. Variation in the population is maintained by PCR amplification, which introduces mutations at a rate of per nucleotide position per generation. Because mutation is ongoing, evolution based on Darwinian principles can occur. Progeny ribozymes have the opportunity to acquire new mutations that confer favorable attributes not possessed by the parent molecules. This phenomenon is reflected by the steadily increasing frequency of accepted mutations over subsequent generations (not shown) b. Improvement of Substrate Binding Affinity 20 Beginning with any generation of enzymatic RNA molecules, the concentration of substrate is lowered from 10 AM to 0.2 AM to impose increased selection pressure favoring individuals with enhanced substrate binding affinity. In order to assess the 25 impact of this change, KD values for the complex between ribozyme and product are determined for the population of ribozymes at regular intervals, at every third generation.

It is anticipated that, when the within-disclosed 30 procedures are followed, improvement in substrate binding affinity over successive generations of in vitro evolution may be observed.

The product, rather than substrate, is employed to avoid a cleavage reaction during the gel-shift analysis. The binding affinity for the product is assumed to be similar to that of the substrate, based on previous studies showing that the wild-type ribozyme binds the RNA substrate with the same affinity as it -126binds the product (Pyle et al., PNAS USA 87: 8187-8191 (1990); Herschlag Cech, Biochemistry 29: 10159-10171 (1990)) Example 2 Analysis of Evolving DNA-Cleaving Populations DNA from 14 subclones at generation 9 was transcribed to produce individual RNAs which were purified by polyacrylamide gel electrophoresis and Sephadex chromatography. The catalytic behavior of these RNAs was studied with 32 and 32 p]labeled DNA substrates having the sequence

GGCCCTCTC-

A

3

(TA

3 3 (SEQ ID NO 29) and with 32 and [e- 32 p]_ ATP-labeled RNA substrates having the sequence

GGCCCUCUC-A

3 (UA3)3 (SEQ ID NO 28). The kinetic parameter most relevant to our selection criterion was the proportion of ribozyme molecules that become joined to the 3' portion of the DNA substrate after 1 hour at 37 0 C and 10 mM MgCl 2 These data and comparable data concerning reactions with a DNA substrate at 50 0 C and S 20 50 mM MgCl 2 and with an RNA substrate at 370C and 10 mM MgC 12 were collected and plotted (data not shown).

The catalytic activity of 14 individual ribozymes obtained at generation 9 was determined (not shown) Ribozymes were transcribed and assayed according to the procedures described in Example 1. RNA substrate was prepared by in vitro transcription with a synthetic oligodeoxynucleotide template; reaction conditions were as described previously for the data illustrated in Fig. 3, but included [a- 32 P]ATP at 0.003 pCi/pmol to 30 label the 3' portion of the substrate.

There was considerable heterogeneity among the 14 individual RNAs with respect to DNA cleavage activity in the target reaction. All were more active than the wild-type, with the best (clones 29 and 23) being about 60 times more active. The five most active individuals were more active under physiologic conditions than under the high-temperature and high-MgCl, conditions.

-127- All 14 individuals showed improved activity with the RNA substrate, even though the population had never been challenged with RNA. Improved RNA cleavage activity was largely due to enhanced activity in the site-specific hydrolysis reaction which allowed enhanced turnover (data not shown).

As mentioned previously, there is selection pressure against site-specific hydrolysis of the EP covalent intermediate in the reaction with a DNA substrate. In fact, all but one of the 14 individuals showed decreased hydrolytic cleavage of the attached DNA compared to the wild-type. All but one of the individuals show increased hydrolytic cleavage with the RNA substrate. Furthermore, there was a strong negative correlation between hydrolytic cleavage of DNA and RNA. The population was clearly divided into two groups: those with low DNA and high RNA hydrolysis activity, and those with high DNA and low RNA hydrolysis activity (not shown). All nine 20 members of the former group carry the 313-314:GA-UG mutations, while all five members of the latter group lacked these changes.

Figures 6A and 6B illustrate Eadie-Hofstee plots used to determine Km (negative slope) and V. (y- 25 intercept) for cleavage of 3 2 P)-labeled d(GGCCCTCT-

A

3

(TA

3 3 (SEQ ID NO 17) by wild-type ribozymes and clones 29 and 23 from generation 9. V 0 (nM/min) is plotted against V 0 (103 min') Reaction conditions were as follows: 1 AM ribozyme, 10 mM MgCl 2 30 mM EPPS 30 (pH 370C; for wild-type 10, 20, 40 and 80 IM substrate at 0.25, 30, 60, 120, 180 and 240 minutes; for clone 29, 2.5, 5, 10, and 20 AM substrate at 0.25, 10 and 15 minutes; for clone 23, 2.5, 5, 10 and IM substrate at 0.25, 5, 10, 20 and 30 minutes.

Ribozyme and substrate were first incubated separately in 10 mM MgCl 2 30 mM EPPS (pH 37 0 C, for minutes, then mixed to start the reaction.

-128- Closed circles represent the wild-type; closed squares represent-blone 29; and closed triangles represent clone 23. Each data point is the average of three independent determinations of initial velocity.

The extent of the reaction was linear over the chosen time interval (rmi 0.94, rav, 0.99).

Clones 29 and 23 were chosen for more detailed kinetic analysis, for comparison with the wild-type ribozyme. Initial rates were determined for the reaction with [5'-"P]-labeled d(GGCCCTCT-A 3

(TA

3 3

(SEQ

ID NO 17) substrate at 37°C and 10 mM MgC2,, with 1 AM ribozyme and excess substrate. An Eadie-Hofstee plot of v 0 as a function of v 0 was used to obtain V. and Km (Fig. From this data, k, and km/Km were calculated. For the wild-type ribozyme, Km 6.6 LM and k, 0.0002 min 36 min-') This compares to Km 30 AM and k, 0.006 min-, previously reported for the wild-type ribozyme in a related reaction at 50 0 C and 10 mM MgCl 2 (Herschlag, et al., Id.

20 (1990)). For clone 29, Km 2.0 pM and k, 0.007 min- 1 (k./Km 3600 for clone 23, Km 1.9 pM and km 0.005 min (km/Km 2700 min-') (data obtained at 37 0C and 10 mM MgCl) Thus, the catalytic efficiency of the two evolved RNAs was increased and was about 100 times greater than that of the wild-type, because of improvement in both Km and k,.

A. Correlating Genotype and Phenotype The relation between genotype and phenotype in the context of an RNA-based evolving system can now be 30 formalized. Genotype can be represented as a matrix A, the rows corresponding to individuals in the population and the columns corresponding to functionally significant positions within the nucleotide sequence (Table 3).

Table 3 shows the genotype and phenotype of 14 individuals from generation 9. Genotype is represented as a binary matrix (shown in brackets). Phenotype is -129represented as a column vector with values normalized to wild-type 1.0. DNA cleavage and hydr6lysis activity were determined with

P

labeled DNA substrate under physiologic conditions, as described in Example 1. Accuracy was determined with DNA substrate under physiologic conditions, measuring the fraction of substrate cleavage that occurs at the target phosphodiester bond; reaction conditions were also as described in Example 1.

o e -130- TABLE 3 DNA Hydro- Accu- Clone Errors 94: 98: 205: 215: 313-314: 317: 333: cleavage lysis racy A-Y C-U U-C G-A GA-UG U--R U-C b, b, b, 29 6 1 0 0 1 1 0 0 65 0.1 0.7 23 7 1 0 0 I 1 0 0 57 0.1 0.6 8 1 0 0 1 1 1 0 48 0.1 0.8 43 8 I 0 0 1 0 1 32 0.0 0.4 7 1 0 0 0 1 0 0 22 0.1 0.4 37 4 0 0 0 1 1 0 0 21 0.1 0.6 28 5 0 1 0 0 1 0 0 is 0.0 0.7 2 5 0 0 0 0 1 0 0 II 0.0 0.8 42 6 1 0 0 1 0 0 0 7 0.7 0.8 8 8 1 0 1 0 1 0 0 3 0.2 1.1 40 1 0 0 0 1 0 0 0 3 0.9 0.6 12 6 0 1 1 0 0 1 0 3 0.7 0.8 27 2 1 0 0 0 0 0 0 3 0.8 0.6 I1 6 1 1 0 0 0 0 l 3 1.2 0.8 et wI 0 0 0 0 0 0 0 0 1 1.0 0 s 10 2 -18 13 18 13 -9 o

X

3 0.4 0.7 0.2 0.4 -0.4 -0.2 -0.1 x3: 0.3 0.6 0.4 0.3 0.2 -0.1 -0.3 avg.' 5.6 0.6 0.2 0.1 0.5 0.6 0.1 0.1 21 0.3 0.7 G9 2 5.9 0.7 0.1 0.2 0.4 0.3 0.1 0.2 21 0.3 0.6 1 Avg. the average of the 14 individuals 2 G9 genotype is the average of the 50 subclones obtained from the 9th generation; G9 phenotype is the behavior of the G9 population as a whole -131- As shown in Table 3, phenotype can be represented as a column vector b, whose entries are some measure of fitness (catalytic behavior) of-' the various individuals. One then seeks a row vector x that provides a best fit to the equation: Ax=b, that is, provides a best fit linear estimation of the relation between genotype and phenotype. The solution that minimizes the least-squares error is: x Ab, where A' is the transpose of A. In this way, one obtains a weighing vector x that provides an estimate of phenotype for any given genotype (Table 3).

The data obtained from 14 individuals is not sufficient to provide a meaningful solution to the relation of genotype to phenotype, even for those nucleotide positions that are known to be most significant based on their high frequency of accepted mutation. The weighing vector x is used as a guide to help decide which mutations are sufficiently important to warrant individual study.

20 The following individual mutations were prepared by site-directed mutagenesis: 94:A- U, 94:A- C, 215:G--A, 313:G- U, 314:A-G, and 313-314:GA- UG. Catalytic activity was studied with d(GGCCCTCT-A 3

(TA

3 3 3 2

P]A)

(SEQ ID NO 26) substrate.

S 25 Site-directed mutagenesis was carried out essentially as described in Morinaga, et al., Biotechnoloc 2: 636 (1984), which may be described as follows. Plasmid pT7L-21 (Zaug, et al., Biochemistry 27: 8924 (1988)) was digested with either Eco RI 30 and Hind III to remove the ribozyme coding region, or 6: (ii) Bsa I and Xmn I to remove the ampicillinresistance gene. The resulting fragments were purified in a 1% agarose gel and cross-hybridized in the presence of a 5'-phosphorylated synthetic oligodeoxynucleotide that introduces the desired mutation. The annealing mixture contained 0.06 pmol of pT7L-21(AEco-Hind), 0.06 pmol pT7L-21(ABsa-Xmn), pmol of mutagenic oligodeoxynucleotide, 40 mM Tris-HCl -132- (pH and 8 mM MgSO 4 in 12-yl volume, which was heated to 100 0 C for three minutes, then incubated at 0 C for 30 minutes, and 0 0 C for 10 minutes.

The annealing product was made fully doublestranded with the Klenow fragment of E. coli DNA polymerase I (Boehringer-Mannheim, Indianapolis,

IN)

and T4 DNA ligase Biochemical, Cleveland, OH) and was then used to transform competent DH5a-F' cells, which were grown on ampicillin-containing plates.

Colonies were screened by the colony hybridization method with 32 P]-labeled mutagenic oligodeoxynucleotide as a probe (Grunstein, et al., PNAS USA 72: 3961 (1975)). DNA was prepared from positive colonies and sequenced throughout the ribozyme 15 gene, as described above.

RNA was prepared by in vitro transcription essentially as follows. Transcription conditions: 2 pmol of DNA template (containing mutagenic oligodeoxynucleotides), 2 mM nucleotide triphosphates (NTPs), 15 mM MgC1 2 2 mM spermidine, 5 mM dithiothreitol (DTT), 50 mM tris-HC1 (ph 1500 U of T7 RNA polymerase; 60 pl volume; 37 0 C, 2 hours. RNA was purified by electrophoresis in a 5% polyacrylamide- 8 M urea gel and subsequent column chromatography on 25 Sephadex The individual mutations result in improved activity compared to the wild-type, but they do not result in activity exceeding that of the generation 9 population as a whole. Data were obtained regarding the DNA cleavage activity of individuals obtained by site-directed mutagenesis (not shown). Reaction conditions were as described in Example 1 hereinabove, relating to Fig. 5. The symbol indicates absence of enzyme, while G9 represents generation 9 population as a whole. Reaction products were separated in a polyacrylamide-8M urea gel, an autoradiogram of which is shown.

Activity in the 94:A.U mutant is seven times -133greater and in the 94:A-C mutant it is two times greater than in the wild-type. The 313-314:GA-UG double mutant is more active than either the 313:G->U or 314:A- G single mutant, explaining why the 313-314 mutations occur together among the evolved individuals examined herein. As predicted from the analysis of 14 individuals at generation 9, the 313-314:GA-UG mutations result in diminished site-specific hydrolysis of the DNA substrate compared to the wild-type. These mutations confer both enhanced phosphoester transfer activity and diminished site-specific hydrolysis activity, and thus are well suited to meet the imposed *selection constraint which depends on availability of the EP covalent intermediate.

15 B. Extension of Directed Evolution to Develop Other Evolved Species As an in vitro model of Darwinian evolution, a population of macromolecular catalysts was directed toward the expression of novel catalytic function. In 20 the present Example, the development of ribozymes that cleave DNA with improved efficiency under physiologic conditions has been demonstrated. These evolved RNAs were also used to cleave a target DNA in vivo; ribozymes obtained from generation 9 were expressed in E. coli and shown to prevent infection by M13 singlestranded DNA bacteriophage (not shown) The present successful phylogeny has been continued beyond the tenth generation, after decreasing the concentration of DNA substrate in the target reaction, as further described in Example 2 hereinbelow. Through the first ten generations the substrate concentration was 10/M, roughly matching the Km for the wild-type. Now that the evolved individuals have attained a Km of about 2pM, the substrate concentration has been reduced to subsaturating levels to promote further improvement in substrate binding.

In addition, catalytic turnover in the DNA cleavage reaction is being improved by selecting for both -134phosphoester transfer activity, which generates the EP covalent intermediate, and subsequent RNA-catalyzed site-specific hydrolysis activity, which frees the ribozyme to act on another substrate molecule.

The selection scheme used herein may be applied to various substrates of the form: d(CCCTCNA3(TA 3 3

(SEQ

ID NO 18), where N refers to a nucleotide analog and the ribozyme is selected for its ability to cleave the phosphodiester bond following the sequence CCCTCN (SEQ ID NO 19). Examples of nucleotide analogs useful according to the present invention include those listed in Table 1, most of which are found in the approved listing of modified bases at 37 CFR §1.822 (which is incorporated herein by reference).

Nucleotide analogs that are particularly useful in the enzymatic RNA molecules, nucleotide substrate, and methods disclosed herein include those having the abbreviations cm, d, gm, i, p, s2c, s2u, s4u, t, um, araU, and araT. (The more complete names of these 20 analogs are shown in Table 1, supra.) The substrate need not be a nucleotide or nucleotide analog. The only requirement is that RNAs that react with the substrate become tagged in some way so that they can be distinguished from nonreactive molecules with respect to the amplification process.

For example, reactive RNAs could become joined to a portion of the substrate that is attached to a solid support, while nonreactive RNAs would be washed away, leaving the bound RNAs to be selectively amplified.

These and other methodologies are further described below.

C. Discussion It has now been shown that specific catalytic properties of a DNA-cleaving ribozyme can be optimized by appropriate manipulation of the selection constraints during an in vitro evolution procedure.

Beginning with a heterogeneous population of ribozymes, enriched for modest DNA-cleavage activity, 18 -135additional generations were carried out to obtain DNA-cleaving ribozymes that have a catalytic rate of 0.7 min' 1 and a substrate binding affinity of 10- 9

M.

These catalytic parameters are improved 10 3 -fold and 10 4 -fold, respectively, compared to the wild-type. The greatest improvement in KD and KM, (Fig. 8B; Table 4) occurred between G9 and G18 in response to alteration of the selection constraints to favor ribozymes with enhanced affinity for the DNA substrate. Likewise, based on k values for representative individuals (Table the greatest improvement in k, occurred between G18 and G27, following alteration of the selection constraints to favor a faster rate of catalysis.

The DNA-binding affinity of the G27 #48 and G27 15 #61 ribozymes is comparable to the RNA-binding affinity of the wild-type ribozyme. Previous studies have suggested that the wild-type ribozyme binds RNA more strongly than DNA as a result of interactions between the 2'-OH groups of the RNA substrate and specific 20 nucleotides within the catalytic core of the ribozyme (Pyle Cech, Nature 350: 628-631 (1991); Pyle et al., Nature 358: 123-128 (1992); Herschlag et al., Biochemistry 32: 8299-8311 (1993)). In binding the DNA substrate with nM affinity, the evolved ribozymes must 25 compensate for the lack of substrate 2'-OH groups by forming alternative interactions that provide an additional 5 kcal mol 1 of binding energy (at 37 0

C).

The Tetrahymena ribozyme binds its substrate through a two-step process involving first, Watson-Crick base pairing between the internal guide sequence (IGS) and the substrate, and second, docking of the IGS/substrate duplex (P1 helix) into the catalytic core of the ribozyme via tertiary interactions (Herschlag, Biochemistry 31: 1386-1399 (1992); Bevilacqua et al., Science 258: 1355-1358 (1992)). Because the sequence of both the IGS and substrate is unchanged throughout the in vitro -136evolution procedure, it is unlikely that we have evolved compensatory mutations -Chat operate at the first step of binding. Instead, the 5 kcal mol-' of additional binding energy is likely to result from additional tertiary interactions that affect the second step of binding.

Much more can be learned about such interactions by examining the specific mutations that arose in response to the increased selection pressure aimed to improve substrate binding. For example, mutations at positions 115, 116, and 205 in the J4/5 and J5/4 internal loop of the ribozyme (Fig. 1) became prominent in the population between G9 and G18. Both a tertiary structural model of the wild-type ribozyme (Michel 15 Westhof, J. Mol. Biol. 216: 585-610 (1990)) and experimental evidence suggest that residues in the region may interact with the IGS. On the basis of crosslinking data, Wang et al. (Science 260: 504-508 (1993)) concluded that A114 and A115 lie in close proximity to G22 of the IGS (the 5'-terminal residue of the L-21 form of the ribozyme) when P1 is docked into the ribozyme core. The mutations at positions 115 and 116 may enhance P1 docking by allowing new contacts to be made with the IGS, compensating for the lack of 25 2'-OH groups in the substrate. Such interactions would strengthen binding of both DNA and RNA substrates and might account for the slight improvement in RNA-cleavage activity. This may also explain the observation that the G27 #61 ribozyme efficiently cleaves a modified RNA substrate that has an arabinose sugar at the cleavage site (data not shown). As noted previously, enzymatic RNA molecules according to the present invention are capable of cleaving substrates including nucleotide analogs, irrespective of whether it is the base or the sugar that has been modified, or both.

Co-occurring mutations at positions 188, 190, and 191 in the P5a region also became prominent in the -137population between G9 and G18. The correlation between these mutations and the co-occurring mutations in the and J5/4 iAternal loop (see Results) suggests a possible interaction between these two regions. It has been proposed that the adenosine-rich bulge in (Fig. 1; positions 183-186) interacts with P4 (Flor, et al., EMBO J. 8: 3391-3399 (1989)) by bending at the internal loop (Murphy Cech, Biochem. 32: 5291-5300 (1993)), which would place residues G188, U190, and G191 in close proximity to the J4/5 and J5/4 internal loop. Thus, the mutations in P5a may facilitate the contact between residues in the region and the IGS.

The evolved ribozymes might compensate for the 15 absent substrate 2'-OH groups by forming new tertiary interactions with the bases, phosphates, and/or sugars of the DNA. Studies suggest that residues in J7/8 of the wild-type ribozyme interact with the 2'-OH groups at positions and of the RNA substrate (Pyle 20 Cech, Nature 350: 628-631 (1991); Pyle et al., Nature 358: 123-128 (1992)). Frequent mutations, however, did not occur in the J7/8 region of the evolved ribozymes, suggesting that new contacts are not made in the vicinity of the DNA substrate at positions and In addition, specific base contacts seem unlikely, based on the observation that a DNA substrate S" with a different sequence can be. cleaved efficiently by the ribozyme, provided the IGS has been changed in a complementary manner to maintain Watson-Crick base pairing (Raillard Joyce, unpublished results).

The 10 3 -fold improvement in km over the 27 generations is more difficult to rationalize. k.

reflects all first-order rate constants along the reaction pathway, including those related to P1 docking and substrate cleavage. At least part of the enhancement in k. thus may be attributed to additional tertiary interactions between P1 and the catalytic core, which would favorably affect the docking rate.

-138- The cleavage step of the reaction depends on the appropriate positioning of the 3 '-terminal guanosine of the ribozyme for attack on the target phosphoester bond of the substrate. This is accomplished by the formation of a base triple involving the attacking guanosine and the G264:C311 base pair within the P7 region of the ribozyme (Michel et al., Nature 342: 391-395 (1989)). Mutations at positions 312, 313, and 314 all lie in close proximity to the binding site for the attacking guanosine and may play some role in facilitating the chemical step of the reaction.

However, new mutations that became frequent in the population between G18 and G27 in response to the shorter reaction time occurred in peripheral regions, 15 at positions 51/52 and 170. Such mutations may increase first-order reaction rates indirectly through long-range effects or by facilitating folding of the ribozyme into its active conformation.

It is important to note that some mutations may confer no selective advantage with respect to catalysis, but instead enhance the ability of the polymerase enzymes reverse transcriptase, T7 RNA polymerase, and Taq polymerase) to operate efficiently during the amplification procedure. Future studies, 25 relying on site-directed mutagenesis analysis, will enable us to assess the contribution made by various mutations, in either the conserved core or the peripheral regions, to substrate binding, first-order reaction rates, and ribozyme folding.

Now that it has been demonstrated that substrate binding and first-order rate constants can be specifically enhanced by in vitro evolution, optimization of other catalytic properties of the DNA-cleaving ribozymes, including turnover and substrate specificity, is being attempted. Turnover might be improved by evolving ribozymes that can carry out a site-specific hydrolysis reaction, subsequent to DNA-cleavage, which removes the attached 3' portion of -139the DNA substrate from the 3' end of the ribozyme, returning the molecule to its original form.

Specificity of the ribozymes for DNA versus-RNA substrates might be increased by selecting for DNA-cleavage in the presence of RNA that acts as a competitive inhibitor. One aim is the development of DNA-cleaving ribozymes that have high catalytic efficiency, undergo rapid turnover, and operate in a highly specific manner. Such molecules will contribute to our understanding of the catalytic potential of RNA.

In addition, they may have utility as sequence-specific DNA endonucleases and as therapeutic agents directed against viral pathogens.

Example 3 15 Optimization of a DNA-Cleaving Enzymatic RNA Molecule SA. Optimization and Selection Criteria In previous analyses (see Examples 1 and an in vitro evolution procedure was used to obtain variants S 20 of the Tetrahymena ribozyme with 100-fold improved ability to cleave a target single-stranded DNA under physiologic conditions. Reported herein is the continuation of the in vitro evolution process to achieve 10 5 -fold overall improvement in DNA-cleavage activity. In addition, it is demonstrated herein that, by appropriate manipulation of the selection constraints, one can optimize specific catalytic properties of the evolved ribozymes.

The concentration of the DNA substrate was first reduced 50-fold, to favor ribozymes with improved substrate binding affinity. Next, the reaction time was reduced 12-fold to favor ribozymes with improved catalytic rate. In both cases, the evolving population responded as expected, first improving substrate binding 25-fold, and then improving catalytic rate about 50-fold. The population of ribozymes has undergone 27+ successive generations of in vitro evolution, resulting in, on average, 17 mutations -140relative to the wild-type that are responsible for the improved DNA-cleavage activity.

In vitro selection and in vitro evolution techniques allow new catalysts to be isolated without a priori knowledge of their composition or structure.

Such methods have been used to obtain RNA enzymes with novel catalytic properties. Ribozymes that undergo autolytic cleavage with lead cation have been derived from a randomized pool of tRNAPhC molecules (Pan Uhlenbeck, Biochemistry 31: 3887-3895 (1992)). Group I ribozyme variants have been isolated that can cleave DNA (Beaudry Joyce, Science 257: 635-641 (1992)) or that have altered metal dependence (Lehman Joyce, Nature 361: 182-185 (1993)). Starting with a pool of 15 random RNA sequences, molecules have been obtained that catalyze a polymerase-like reaction (Bartel Szostak, Science 261: 1411-1418 (1993)). In the present example, refinement of specific catalytic properties of an evolved enzyme via alteration of the selection constraints during an in vitro evolution procedure is described.

The within-described examples utilize derivatives 99 of the self-splicing group I intron of Tetrahymena thermophila, a ribozyme that is able to catalyze S. 25 sequence-specific cleavage of single-stranded RNA via a phosphoester transfer mechanism (Zaug Cech, Science 231: 470-475 (1986); Zaug et al., Nature 324: 429-433 (1986)), although it is expressly to be understood that the invention is not limited to these embodiments. The ribozyme contains a template region, referred to as the "internal guide sequence" (IGS), which lies at the end of the molecule and forms Watson-Crick base pairs with the target RNA substrate. The 3'-OH of guanosine, including a guanosine residue that lies at the 3' end of the ribozyme, is directed to attack a particular phosphoester bond within the ribozyme-bound substrate.

A phosphoester transfer reaction ensues, resulting in cleavage of the substrate at a position immediately -141downstream from the region of base pairing, and concomitant ligation of the 3' portion of the substrate to the 3' oxygen of the attacking guanosine. The wild-type Tetrahymena ribozyme can cleave a single-stranded DNA substrate with low efficiency under conditions of high magnesium concentration (50 mM MgCl 2 and/or high temperature (50 0 C) (Herschlag Cech, Nature 344: 405-409 (1990a); Robertson Joyce, Nature 344: 467-468 (1990)). Under more physiologic conditions 37 0 C, 10 mM MgCl 2 pH however, the DNA-cleavage reaction is almost undetectable.

An in vitro evolution procedure that may be used to obtain variants of the Tetrahymena ribozyme that can cleave DNA under physiologic conditions with improved 15 efficiency compared to the wild-type is illustrated in Figures 2A-2C. (See also Beaudry and Joyce (Science 257: 635-641 (1992).) At the beginning of this procedure, a population of ribozyme variants was generated by partially randomizing the phylogenetically 20 conserved portions of the molecule that are known to be essential for catalytic activity. Superior DNA-cleaving ribozymes were distinguished from less active molecules based on the likelihood of attachment of the 3' portion of the substrate to the 3' end of the ribozyme. A DNA primer was hybridized across the ligation junction of successful reaction products, and used to initiate a selective isothermal amplification reaction (see Fig. 2C, bottom). The selectively amplified molecules then served as templates for cDNA synthesis; the resulting cDNA was amplified by the polymerase chain reaction (PCR) (Saiki et al, Science 230: 1350-1354 (1985); Saiki et al, Science 239: 487-491 (1988)); and the PCR products were transcribed to generate a new pool of RNAs. The entire process, beginning with the cleavage reaction and followed by selective isothermal amplification, cDNA synthesis, PCR amplification, and in vitro transcription, constitutes one "generation" of the in vitro evolution procedure.

-142- This in vitro procedure has successfully been used to generate 10 successive generations, starting-with a pool of 101 variants of the Tetrahymena ribozyme (see Example 1 above, and Fig. After the 9th generation individual ribozymes were isolated from the population and shown to catalyze the cleavage of a DNA substrate 100-fold more efficiently compared to the wild-type enzyme. This modest improvement in catalytic efficiency resulted from both an increased catalytic rate and a decreased value for the Michaelis constant The outcome, however, was somewhat dissatisfying because the ribozymes were still inefficient catalysts in an absolute sense, with k,,/KM on the order of 10' For each generation, the 15 evolving population was provided with 10 AM DNA substrate and allowed 1 hr to carry out the DNA-cleavage reaction. By G9, KMhad improved from 6 AM for the wild-type to about 2 AM for the evolved individuals (see Example 1; see also Beaudry Joyce, Id., (1992)). Accordingly, it appeared that the population was no longer under stringent selection pressure to drive further improvement of KM.

Individual cleavage rates, on the other hand, were on :the order of 0.007 min-'by G9, still slow enough to be 25 constrained by the 1 hr incubation period. However, if the reaction rate continued to improve, then the selection constraints would eventually become insufficient to favor further improvement of the catalytic rate. Apparently, additional generations of in vitro evolution, under different selection constraints, would be necessary to obtain substantially greater DNA-cleavage activity.

In the present example, in vitro evolution techniques were applied with a higher level of sophistication and control. Because the outcome of an in vitro evolution experiment depends on the nature of the selection constraints, specific catalytic -143properties of a ribozyme, such as substrate binding affinity, catalytic rate, substrate specificity, and turnover, might be improved by appropriate manipulation of the reaction conditions. With this in mind, optimization of two catalytic properties of the DNA-cleaving ribozymes, namely, substrate binding affinity and catalytic rate was a primary goal. It was hypothesized herein that ribozymes with the greatest affinity for the substrate would enjoy a selective advantage when the substrate is presented at low concentrations. Under saturating conditions, ribozymes with the fastest first-order rate of reaction would be favored when the reaction time is very short.

The previously-characterized G9 population of 15 DNA-cleaving ribozymes (see Example 1) was "resurrected" and 27 additional generations of in vitro evolution were carried out under somewhat different reaction conditions. From generations 10 through 18, the substrate concentration was reduced 50-fold, from 10 AM to 0.2 AM. From generations 19 through 27, the lower substrate concentration was maintained and the reaction time was reduced 12-fold, from 1 hr to 5 min.

On the basis of binding and kinetic studies, the ooo population of ribozymes responded to each alteration of S 25 the selection constraints as predicted, becoming enriched with tighter substrate binders during generations 10-18, and then with faster catalysts during generations 19-27. Generations 28-36 are discussed in subsection 6 hereinbelow.

B. Materials and Methods 1. Materials Unlabeled nucleoside triphosphates (NTPs) and deoxynucleoside triphosphates (dNTPs) were purchased from Pharmacia, and dideoxynucleoside triphosphates (ddNTPs) were from U.S. Biochemical (USB, Cleveland, OH). [a- 2 nPGTP, [g- 32 P]ATP, and 3 HIUTP were from ICN Radiochemicals. Synthetic oligodeoxynucleotides were obtained from Operon Technologies and purified by -144polyacrylamide gel electrophoresis and subsequent chromatography on Se hadex G-25. Restriction enzymes and T4 polynucleotide kinase were from New England Biolabs (Beverly, NA), calf intestine phosphatase from Boehringer (Indianapolis, IN), AMV reverse transcriptase from Life Sciences (St. Petersburg,

FL),

MoMLV reverse transcriptase and Sequenase 2.0 (modified T7 DNA polymerase) from U.S. Biochemical, and Taq DNA polymerase from Cetus (Emeryville, CA). T7 RNA polymerase was prepared as previously described (Davanloo et al., PNAS USA 81: 2035-2039 (1984)) and purified according to a procedure originally developed Sfor SP6 RNA polymerase (Butler Chamberlain, J. Biol.

Chem. 257: 5772-5778 (1982)).

15 2. Preparation of Wild-Type Ribozyme The L-21 form of the Tetrahymena ribozyme was prepared by in vitro transcription of Hind IIl-digested pT7L-21 plasmid DNA (Zaug et al., Biochemistry 27: 8924-8931 (1988)). The transcription reaction mixture S 20 contained 0.1 pg/l of cut plasmid, 15 mM MgCl 2 2 mM spermidine, 50 mM Tris (pH 5 mM DTT, 2 mM each NTP, 0.005 U/ 1 l inorganic pyrophosphatase, and 25 U/l T7 RNA polymerase, incubated at 37 0 C for 2 hr. The 23-nucleotide 3' exon sequence was removed by 25 RNA-catalyzed site-specific hydrolysis (Inoue et al., J. Mol. Biol. 189: 143-165 (1986)): RNA was incubated in the presence of 50 mM CHES (pH 9.0) and 10 mM MgCl 2 at 42 0 C for 1 hr. The resulting RNA was isolated by electrophoresis in a 5% polyacrylamide /8 M urea gel, visualized by UV shadowing, eluted from the gel overnight at room temperature in a buffer containing 200 mM NaC1, 10 mM Tris (pH and 0.5 mM EDTA, and purified by affinity chromatography on DuPont Nensorb (Wilmington, DE). The concentration of ribozyme was determined spectrophotometrically, based on e 2 6 3.2 x 106 M 1 cm"' (Zaug et al., Biochemistry 27: 8924-8931 (1988)) -145- 3. In Vitro Evolution Procedure In vitro evolution was carried out as described previously (see Example above) and as depicted in Fig. 2C. While polymerase chain reaction (PCR) or self-sustained sequence replication (3SR) methods are both useful, the within-described methodology most closely resembles the 3SR method (see, Guatelli et al., PNAS USA 87: 1874-1878 (1990)). The 3SR system is particularly useful in the detection and nucleotide sequence analysis of rare RNAs and DNAs.

The population of DNA-cleaving ribozymes obtained after 9 generations of in vitro evolution in Example 1 above was used as starting material. Ribozymes (0.1 M) and DNA substrate (0.2 AM) were incubated at 37 0

C

S 15 for 1 hr in a 100 Al volume containing 10 mM MgC12 and mM EPPS (pH After ethanol precipitation, a portion of the reaction products (10-50%) was added to a 20 pl isothermal amplification reaction mixture, containing 10 mM MgC1 2 80 mM KOAc, 50 mM Tris (pH 20 5 mM DTT, 2 mM each NTP, 0.2 mM each dNTP, 4 Ci [a- 3 P]GTP, 12.5 U/Ml MoMLV reverse transcriptase,. U/zl T7 RNA polymerase, and 20 pmol each of 5'-TTTATTTATTTATTT-3' (Primer la, SEQ ID NO 6) and 5'-CTGCAGAATTCTAATACGACTCACTATAGGAGGGAAAAGTTATCAGGC-3' (Primer 2, SEQ ID NO which was incubated at 37 0

C

for 2 hr. Primer 1 hybridizes to the 3' portion of the substrate that becomes attached to the 3' end of the ribozyme. (Primers la and Ib, when used, perform similarly.) Primer 2 hybridizes to the 3' end of the resulting cDNA and introduces the T7 promoter sequence.

Twenty-five percent of the isothermal amplification products were used to generate cDNA in a Al reaction mixture containing 10 mM MgC12,50 mM Tris (pH 5 mM DTT, 2 mM each NTP, 0.2 mM each dNTP, 0.2 U/Ml AMV reverse transcriptase and 20 pmol Primer la, incubated at 37 0 C for 1 hr. Approximately 5-10% of the resulting cDNA was amplified by the PCR in a 100 Al reaction mixture containing 1.5 mM MgCl 2 50 mM KC1, -146mM Tris (pH 0.1% gelatin, 0.2 mM each dNTP, pmol 5'-CGAGTACTCCAAAACTAATC-3' (Primer Ib, SEQ ID NO 20 pmol Primer 2, and 2.5 U Taq DNA polymerase, carried out for 30 cycles of 92°C for 1 min, 450C for 1 min, and 72 0 C for 1 min, and 1 cycle of 72°C for min. Primer lb is complementary to the 3' end of the ribozyme, allowing regeneration of its original, active form. PCR DNA (-250-500 ng, 5-10% of the total) then served as template in an in vitro transcription reaction, carried out in a 25-50 il volume.

The transcribed RNA was isolated by polyacrylamide gel electrophoresis, visualized by UV shadowing, cut and eluted from gel, purified on duPont NENsorb (duPont de Nemours, Wilmington, DE), and quantified 15 spectrophotometrically, as described above. The entire process was repeated 18 times, the first 9 as described above and the second 9 with the incubation time for the cleavage reaction reduced from 1 hr to 5 min.

Occasionally, the cDNA was purified to improve the 20 quality of the PCR amplification. To do so, cDNA was synthesized as above except in the presence of 25-50 Ci [a-n 3 P]dATP. Labeled cDNA was isolated by electrophoresis in a 5% polyacrylamide/8 M urea gel, visualized by autoradiography, cut and eluted from gel, 25 and purified on DuPont NENsorb.

4. Shotgun Cloning Sequencing, and Preparation of Individual Enzymatic RNA Molecules The G18 subclones were obtained as previously described (see Example 1 above). The G27 subclones were obtained using the Invitrogen TA Cloning Kit (Invitrogen, San Diego, CA). The PCR DNA at G27 was ligated into a linearized plasmid, and the resulting DNA was used to transform competent INVaF' cells, which were grown on ampicillin/X-gal plates. Individual colonies containing the insert were identified by their white color, chosen at random, and grown overnight in liquid media. Plasmid DNA was prepared by the boiling, lysis method (Holmes Quigley, Anal. Biochem. 114: -147- 193-197 (1981)) and screened for the presence of insert by restriction digestion. Cloned individuals were sequenced by the dideoxy chain-termination method, as previously described (Sanger et al., PNAS USA 74: 5463-5467 (1977); Beaudry Joyce, Id. (1992)).

Complete sequences of individual subclones are available upon request. Individual ribozymes were prepared as follows: the gene encoding the ribozyme was amplified by the PCR using Primer lb and Primer 2; the resulting DNA was used as a template for in vitro transcription; the RNA products were isolated by polyacrylamide gel electrophoresis, and were purified and quantified as described above. i Preparation of Substrate and Product 15 Oligonucleotides The DNA substrate 5'-GGCCCTCTATTTATTTA-3' (SEQ ID i NO 15) and DNA product 5'-GGCCCTCT-3' (SEQ ID NO 16) were 3 )-labeled in a 20 l reaction Lmixture containing 20 pmol oligonucleotide, 10 pmol 20 uCi/pmol) [g- 2 P)ATP, 5 mM MgC12, 25 mM CHES (pH 3 mM DTT, and 1.25 U/pl T4 polynucleotide kinase, incubated at 37 0 C for 1 hr. Labeled oligonucleotide was isolated by electrophoresis in a 20% polyacrylamide/8 M urea gel, visualized by 25 autoradiography, eluted from the gel, and purified on DuPont Nensorb.

The RNA substrate 5'-GGCCCUCUAUUUAUUUA-3' (SEQ ID NO 20) was prepared by in vitro transcription using a partially single-stranded synthetic DNA template (Milligan et al., Nucleic Acids Res. 15: 8783-8798 (1987)), as described previously (Example The RNA transcript was dephosphorylated with calf intestine phosphatase, extracted with phenol and chloroform, and then (5'-"P)-labeled and purified as described above.

6. Kinetics Analysis All cleavage reactions were carried out at 37 0 C in mM MgC12, 30 mM EPPS (pH and 40 ig/pl BSA, using (5'-"P)-labeled substrate. BSA was added to -148prevent oligonucleotides from adhering to the walls of the 500 pl Eppendorf tubes, and did not affect the course of the reaction. Ribozyme and substrate were preincubated separately for 15 min at 37 0 C, and then mixed to initiate the reaction. Typically, 5 aliquots of 3-10 Al each were removed from the reaction mixture at specified times and quenched by addition to 1-2 volumes of an ice-cold mixture containing 8 M urea, 50-100 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue, 10% SDS, 9 mM Tris-borate (pH and sucrose. Substrate and product were separated by electrophoresis in a 20% polyacrylamide/8 M urea gel, Svisualized by auto-radiography, excised from gel, and quantified by Cerenkov counting.

15 KM and k. values were determined in experiments with substrate in excess over ribozvme

(E)

Initial rates of reaction over a range of substrate concentrations, were estimated from the initial linear phase, generally the first 5% or less of 20 the reaction. Typically 8 data points were fit by a least squares method to a theoretical line given by the equation: v "KM V Single-turnover experiments were performed with ribozyme in excess of substrate (Herschlag Cech, 25 Biochemistry 29: 10159-10171 (1990b)). Initial rates (ka, were obtained using no more than the first 5% of the reaction. Given that k./KM= each k,, value, obtained at different ribozyme concentrations, provided an estimate of km/KM. Generally 8 or more measurements of k./KM were obtained.

7. Determination of Binding Constants The equilibrium dissociation constant, KD, of the complex between ribozyme and DNA product was determined by gel-shift analysis in a native polyacrylamide gel (Pyle et al., PNAS USA 87: 8187-8191 (1990)). Ribozyme at twice final concentration was preincubated at 37 0 C for 15 min in 10 mM MgC12 and 30 mM -149- EPPS (pH 7.5) before mixing with an equal volume of 0.05-1 nM 2 P)-labeled DNA product in 10 mM MgC1 2 mM EPPS (pH 0.05% xylene cyanol, 3% glycerol, and pg/pl BSA. The mixture was allowed to equilibrate at 370C for 15-60 min before loading on a polyacrylamide gel containing 10 mM MgC12 and 30 mM EPPS (pH The electrophoresis buffer also contained mM MgCl 2 and 30 mM EPPS (pH The gel was run at 6 milliamps in a 370C room until the sample had entered the gel (-10 min), and then moved into a 4 0 C cold room where the current was increased to 30 milliamps. This was done to prevent the temperature of the gel from rising above 37 0 C. The ribozyme-product complex and free product were visualized by autoradiography, cut S 15 from the gel, and quantified by Cerenkov counting.

A binding curve was generated by plotting the percentage of product bound to ribozyme bound) over ;a range of ribozyme concentrations. KDwas determined by fitting the data to a theoretical binding curve 20 using a least squares method. Because ribozyme was in vast excess over product, the theoretical binding curve could be represented by the equation: bound KD), where KD when half of the total product is bound to the ribozyme.

C. Results 1. Evolution In Vitro *5 Beginning with the 9th generation (G9) population of ribozymes obtained in a previous study (Beaudry Joyce, Id. (1992)), 18 additional generations of in vitro evolution were carried out. Variation in the population was maintained by PCR amplification, which introduces mutations at a rate of per nucleotide position per generation. Because mutation is ongoing, evolution based on Darwinian principles can occur.

Progeny ribozymes have the opportunity to acquire new mutations that confer favorable attributes not possessed by the parent molecules. This phenomenon is reflected by the steadily increasing frequency of -150accepted mutations over the 27 generations.

Sequence data was obtained from 50 randomly-chosen subclones, isolated from the evolving population at G9, G18, and G27 (see Figures 7A-7C). Figures 7A-7C illustrate sites at which mutations occurred over the course of evolution, superimposed on the secondary structure of the Tetrahymena ribozyme. Box height corresponds to the frequency of mutations at each nucleotide position, based on 50 subclones sequenced at G9 (Fig. 7A), G18 (Fig. 7B), and G27 (Fig. 7C).

Nonmutable primer binding sites are shaded; substrate is shown in black. Commonly-occurring mutations (>30 frequency) are labeled.

The mean number of mutations per subclone rose 15 from 5.9 at G9, to 12.7 at G18, and to 16.5 at G27.

Most of the mutations occurred within the phylogenetically conserved portions of the ribozyme that were randomized in the initial population (see Fig. 2C). However, 26% of the total mutations at G18, 20 and 38% at G27, occurred in peripheral regions as a result of ongoing mutagenesis. Most of the commonly-occurring mutations frequency) that occur in the G18 subclones (see Fig. 7B) were not observed at G9 (Fig. 7A), suggesting that these 25 mutations arose in response to the increased selection pressure designed to enhance substrate binding affinity. Between G18 and G27, nearly all of the most commonly-occurring mutations continued to increase in frequency (Fig. 7C). However, two significant mutations, the NGAA insertion between positions 51 and 52 and the C-U change at position 170, first appeared during this interval, suggesting that these mutations arose in response to the increased selection pressure designed to enhance the catalytic rate.

2. Concerted and Mutually-Exclusive Mutations The changes at nucleotide positions 188, 190, and 191 in the P5a region (Fig. 1) co-occur in 90% of subclones, while mutations in the J4/5 and J5/4 -151internal loop at positions 115, 116, and 205 co-occur in 68% of the subclones at G18. Interestingly, the ard J5/4 mutations co-occur only if the set of mutations is also present (c 2 110, p 0.001), suggesting an interaction between these two regions.

The 313:G-*Y and 314:A-G mutations nearly always occur together. These mutations co-occur in 16 of subclones at G9, 11 of 50 subclones at G18, and 44 of subclones at G27. Only two G27 subclones contain the mutation at position 313 but lack the mutation at position 314. At G9 and G18, the 313 mutation always occurs as a G-U change. At G27, however, the 313 mutation occurs primarily as a G-*C change, with the G- U change occurring only once. The GA sequence normally present at positions 313-314 is thought to form a short duplex structure (P9.0) that brings the 3'-terminal guanosine residue of the ribozyme into the catalytic core (Michel et al., Nature 342: 391-395 (1989); Michel et al., Genes Dev. 4: 777-788 (1990); Michel, et al., 20 J. Mol. Biol. 216: 585-610 (1990)). The 3'-OH of this guanosine serves as the nucleophile in the RNA-catalyzed phosphoester reaction. Although the 313-314 mutation would prevent the P9.0 duplex from forming, the 313-314:GAUG change confers selective advantage with respect to the DNA-cleavage reaction, as demonstrated by site-directed mutagenesis studies (Beaudry Joyce, Id. (1992)). The appearance of the i: 313-314:GA-CG change, between G18 and G27, suggests that this altered form of the 313-314 mutation may contribute to the improved catalytic rate of the DNA-cleavage reaction.

The 312:G-*A mutation occurs only if the 313-314:GA- YG mutations are not present. The 312:G-*A change is present in 4 of 25 subclones at G3, 8 of subclones at G6, and 5 of 50 subclones at G9 (Beaudry Joyce, Id. (1992)). There is a dramatic rise in the frequency of the 312:G-A mutation between G9 and G18, followed by an equally dramatic drop between G18 and -152- G27 (see Figs. 7A-C). As the frequency of the 312:G->A mutation declines, the 313-314:GA- YG mutations become more abundant.

The 215:G-+A mutation, present at high frequency in all of the studied populations, putatively allows a WLtson-Crick base pair to form with the U at position 258 (Fig. This change is present in nearly all of the subclones at G18 and G27. Of the 12 individuals that lack this mutation, 11 carry a U->C change at i0 position 258, which would allow a Watson-Crick pair to form with the wild-type G at position 215. Thus, in 99 of 100 subclones from G18 and G27, a Watson-Crick base pair is expected to form between positions 215 and 258.

3. Improvement of DNA Bindin Affinity 15 Beginning with G10, the concentration of DNA *substrate employed during the RNA-catalyzed reaction was lowered from 10 pM to 0.2 pM to impose increased selection pressure favoring individuals with enhanced 2.' substrate binding affinity. In order to assess the 20 impact of this change, KD values for the complex between ribozyme and DNA product (GGCCTCT) were determined for the population of ribozymes at every third generation over the 27 generations (see Figure 9).

25 Figures 9A and 9B illustrate the improvement in substrate binding affinity over 27 successive generations of in vitro evolution. Fig. 9A represents a typical binding curve showing data obtained for the G27 population of ribozymes. A and B indicate data from two different gel-shift experiments. Data was fit by a least squares method to a theoretical binding curve (indicated by solid line), given by the equation: y KD), where y is the fraction of product bound to ribozyme In this case, KD= 51 2) nM. Fig. 9B shows the KD for the population of ribozymes at every third generation. Standard errors averaged 11%.

The DNA product rather than substrate was employed -153to avoid a cleavage reaction during the gel-shift analysis. The binding affinity for the product is assumed to be similar to that of the-substrate, based on previous studies showing that the wild-type ribozyme binds the RNA substrate with the same affinity as it binds the product (Pyle et al., PNAS USA 87: 8187-8191 (1990); Herschlag Cech, Biochemistry 29: 10159-10171 (1990b)).

Binding data for each studied population was fit to a theoretical binding curve, an example of which is shown in Fig. 8A for the G27 population. As expected, the greatest improvement in binding affinity occurred between G9 and G18 (Fig. 8B), subsequent to tightening of the selection constraints. After G18, the S 15 population became saturated with ribozymes having a KD of less than 0.2 pM, accounting for the slow but continued improvement between G18 and G27.

4. Kinetic Analysis Beginning with generation 19, the reaction time 20 was reduced from 1 hr to 5 min to favor selection of ribozymes with increased k, values. To study the effect of this change, two individuals isolated from the population at G9, G18 and G27 were chosen for formal kinetic analysis (Table These ribozymes are representative of the population from which they were isolated because they contain most of the prominent mutations that occur in their respective populations.

In addition, the total number of mutations in each of the studied individuals coincides with the mean number of mutations per subclone in the corresponding population. It is emphasized that the k, and KM values of the studied individuals are not equivalent to the average k, and KM values for the entire population. It is likely that the catalytic efficiencies of the studied ribozymes are somewhat higher than the average because these ribozymes possess a greater fraction of the dominant mutations than a typical individual in the population. Nevertheless, the relative differences in -154k. and KM values between representative pairs of individuals should be comparable. As expected, the improvement in is greatest between the G18 and G27 ribozymes (Table while the improvement in KM is greatest between the G9 and G18 ribozymes.

Table 4, illustrating the catalytic parameters of DNA-cleaving enzymatic RNA molecules, is reproduced hereinbelow.

C*

C

C

C* -155- Table 4 Catalytic Parameters of DNA-Cleaving Ribozymes *9 90 £9 S 9 no.

9**9 9 969099 9 *000 999* *099 C 09 9. 9 9009 0 .9 *b*9 *999 9 9099 *9 9 0S 9.

9* 0 09 kb Kb k ,/KCM Ribozyme Mutations (miw'l) (tiLM) (M-'cmin') Wta0 2.4 x 10-4 6.0 1.7 4.0 x G9 #23 07 5.1 X 1, 1.8 0.3 2.8 G9 #29 6 7.1 X 1, 1.9 0.3 3.8 GIB #13 c 12 f 1.7 x 10-' 0.24 0.04 7.1 G18 #66 13 9 1. 1 x 10.2 0.32 0.08 3. x 104 G27 #48 d 17 A 7.0 x 10-1 0.31 0.05 2.3 x 106 G27 #61 15 3.3 x 10-1 0.11 0.06 2.9 x 106 T) 4t-; prjjreiuslytcmc (see =Tamlelabovetr), modi fid4 slightly as a result of subsequent statistical analysis.

b Measurements were carried out as described in Materials and Methods with: c 0.025 jiM ribozyme and 0.125, 0.25, 0. 5, and 1. 0 AiM DNA substrate d 0.02 AiM ribozyme dnd 0.1, 0.2, 0.4, and 0.8 AiM DNA substrate; or 0.02 jiM ribozyme and 0.05, 0.1, 0.2, and 0.4 AiM DNA substrate. f44:G-*.A, 19 205:U-->C, 215:G-*.A, 312:G-A, and 317:U-3.G. 9 18 8: 190:U-3-A, 19l:G-->U, 205:U-->C, 215:G--*A, 239:U-->A, and 312:Cr--A. h44:G..-A, 51/52:insert AGAA, 87:A-->del, 94:A-->U, 191:G.U, 205:U-3-C, 215:G->A, 239:U-->A, 312:G-+.A, 350:C->.U, and 3 64: C-3.U. '44:G--4A, 51/52:insert AGAA, 87:A->del, 94:A-->U, 19 205:U-->C, 215:G-+A, 313:G-->C, and 314:A-->G.

-156- RNA-Cleavage Activity of G27 Enzymatic RNA Molecules In order to assess the effect of the evolution procedure on RNA-cleavage activity, the efficiency of RNA-cleavage by both the G27 #48 and G27 #61 ribozymes was compared to that of the wild-type. Single-turnover kinetic experiments revealed that the G27 ribozymes have slightly enhanced RNA-cleavage activity: k,/K, values are 2.7 0.2) x 10 7 and 2.3 0.2) x 10 7 M'min-' for clones G27 #48 and G27 #61, respectively, compared to 9.4 3.0) x 106 M- 1 min for the wild-type. Thus, the 27 generations of in vitro evolution resulted in a 10 5 -fold improvement of DNA-cleavage activity and a 2 to 3-fold enhancement of RNA-cleavage activity.

15 Similarly, gel-shift experiments revealed a significantly greater improvement in DNA binding affinity compared to RNA binding affinity. Ribozymes G27 #48 and G27 #61 bind the DNA product with a KD of 4 nM and 1 nM, respectively, compared to 30 AM for the 20 wild-type, and bind the RNA product with a KD of 0.5 nM and 0.4 nM, respectively, compared to 1.5 nM for the wild-type. Thus, the G27 ribozymes exhibit a 10 4 -fold improvement in DNA binding affinity and a 3 to 4-fold improvement in RNA binding affinity.

25 6. Generations 28-36 The aforementioned evolutionary procedures continue to be applied to produce subsequent generations of enzymatic RNA molecules. Data for generations G28-G36 has been gathered and analysis is ongoing. Critical mutation sites identified as described above continue to be of importance, as shown in Figure 9.

Figure 8 also illustrates sites at which mutations occurred over the course of evolution, superimposed on the secondary structure of the Tetrahymena ribozyme.

Box height corresponds to the frequency of mutations at each nucleotide position, based on 50 subclones sequenced at generation 36. Non-mutable primer binding -157sites are shaded; substrate is shown in black.

Commonly-occurring mutations (>30 frequency) are labeled (dark bars).

Example 4 Enzymatic RNA Molecules With Amide-Cleaving Activity Enzymatic RNA molecules (or ribozymes) have now been developed which are capable of cleaving amide bonds inactive alkyl amide bonds via a metal-dependent hydrolytic mechanism. This is comparable to the reaction carried out by protease/peptidase enzymes, which enzymes typically consist of protein themselves.

There have been reports in the literature describing artificial enzymes that promote cleavage of 15 an activated aryl amide; for example, Janda, et al., Science 241: 1188-1192 (1988) describe an antibody with amidase activity. While this is not an insignificant development, it nonetheless involves a protein with enzymatic activity and the bond cleaved is not a peptide bond. There has also been a report showing that a modified Tetrahymena ribozyme has modest ability to accelerate hydrolysis of an aminoacyl ester under certain circumstances (Piccirilli, et al., Science 256: 1420-1424 (1992)). Nevertheless, the reaction discussed by Piccirilli, et al. is easily accomplished by a common hydrolysis reaction in the absence of enzyme. In contrast, amide hydrolysis reactions demand a catalyst.

Thus, the enzymatic RNA molecules disclosed herein are distinguishable from the foregoing, as they catalyze cleavage of an unactivated alkyl amide, which is akin to the amide linkage within a polypeptide.

Furthermore, the within-disclosed molecules, which exhibit amide-cleaving activity, are not themselves proteins.

While the present example employs substrates containing an amide linkage in the context of an oligodeoxynucleotide-polypeptide "hybrid" molecule, -158with 8 nucleotides upstream and one or more amino acids downstream of the target amide, it is expected that any amide-linkage-containing molecule recognized or recognizable by an enzymatic RNA molecule of the present invention may be cleaved as disclosed herein including polypeptides and proteins. In addition, since the ribozyme binds the substrate via Watson-Crick pairing and tertiary contacts involving the upstream nucleotides present in hybrid molecules, thereby drawing the amide into close proximity to a bound Mg 2 cofactor, it is expected that sequential replacement of said upstream nucleotides with amino acids within the framework of the in vitro evolutionary methods disclosed in Example 1 above will produce a ribozyme 15 that binds tightly to polypeptide molecules.

The methods of the present invention are also uniquely useful in facilitating the engineering and selection of catalytically active RNA molecules which are able to cleave a specific amide bond at a desired location. In other words, the present invention permits the construction of a vast array of RNA *o molecules, each having the ability to cleave a specific peptide bond between particular, preselected amino acids. Clearly, the advantages of having the ability S: 25 to efficiently and expeditiously design enzymes of such specificity are inestimable. The present invention is also advantageous in that it obviates the need to screen a significant number of organisms or constructs in an effort to identify a suitable protease; using the methods disclosed herein, one of skill in the art may now design and construct molecules with the desired specificity and activity.

Additionally, as there are no essential contacts with the downstream nucleotides, it is likely that the downstream amino acids can be replaced with other amino acids, peptides, or polypeptides, or with other chemical substituents. Converting an enzymatic RNA molecule to a full-fledged amide bond-cleaving molecule -159that recognizes, binds and cleaves a polypeptide may be accomplished using the within-disclosed in vitro evolution techniques,'selecting for ribozymes that retain amide-cleaving activity and bind a particular protein. (Also see Tuerk and Gold, Science 249: 505- 510 (1990); and Jellinek, et al., PNAS USA 90: 11227- 11231 (1993).) Similarly, the design and construction of an enzymatic RNA molecule with peptidase activity may be accomplished using the within-disclosed guidelines and techniques.

The enzymatic RNA molecules with the ability to cleave target amides are preferably prepared according to in vitro evolution methods such as those described in Example 1 herein. Thus, while the ribozymes 15 disclosed herein may alternatively be described as having the ability to cleave a particular phosphoester bond in the context of a ribonucleotide, deoxyribonucleotide, or some other nucleotidecontaining substrate an arabinonucleotide 20 substrate), it has now been observed that when the evolved ribozymes are presented with a substrate that contains an amide in place of a phosphate, they catalyze cleavage of the amide to generate products with free amine and free carboxyl termini.

In order to stimulate the progressive evolution of enzymatic RNA molecules capable of cleaving amide bonds between neighboring amino acids, various "hybrid" molecules e.g. molecules comprising a series of one or more nucleotides linked to a series of one or more amino acids are first synthesized as described hereinbelow. Such molecules may then be used to identify useful enzymatic RNA molecules according to the present invention.

A. Synthesis of Ribozvmes and Substrates 1. Synthesis of Oligonucleotides The procedure for preparation of the oligonucleotide segment of a hybrid molecule, e.g., d(GGCCCTCTNH) (SEQ ID NO 11), is described essentially -160as follows.

The 7-mer d(GGCCCTC) (SEQ ID NO 12) was prepared on an automated DNA synthesizer, deprotected in the usual way, and purified by polyacrylamide gel electrophoresis and subsequent affinity chromatography on duPont Nensorb (duPont, Wilmington, DE). The TNH2 residue was provided in the form of 3'-

(U.S.

Biochemical, Cleveland, OH) and was coupled enzymatically to the 7-mer using terminal deoxynucleotidyl transferase (TdT; available from U.S.

Biochemical, Cleveland, OH or BRL, Gaithersburg,

MA),

producing the desired 8-mer product.

The 8-mer was purified by polyacrylamide gel 15 electrophoresis and subsequent affinity chromatography.

The 8-mer was found to migrate appreciably slower than the unreacted 7-mer (data not shown). Finally, the purified 8-mer was [5'-"P]-labeled using P]ATP and T4 polynucleotide kinase, according to standard protocols. (In general, the labeling admixture comprised 2 Al 5X buffer, 1 Al 8-mer, 1 il ae-P-ATP, 4 tl H 2 0, and 2 tl T4 kinase, and was maintained at 37 0

C

for 1 hour.) As shown in Fig. 12B, this 8-mer d(GGCCCTCTN 25 (SEQ ID NO 11) marker and the 8-mer 5' product of the enzymatic RNA molecule-catalyzed cleavage of the amidebond-containing substrate have the same mobility. This effectively demonstrates the amide-cleaving activity of the enzymatic RNA molecules of the present invention.

2. Preparation of Ribozymes Enzymatic RNA molecules identified herein as clones 48 and 61 were used in the within-described cleavage experiments, although it is to be appreciated that the present invention is not limited to use of.

said ribozymes. Clones 48 and 61 were optimized for DNA-cleavage ability and were prepared as described in Example 1 (and in Beaudry and Joyce, Science 257: 635- 641 (1992) and Tsang and Joyce, Biochemistry 33: 5966- -161- 5973 (1994)). Ribozymes from clones 48 and 61 were selected from the 27th generation.

Ribozymes 48 and 61 (G27 #48 and G27 #61) ar6 described as follows. Ribozyme G27 #48 includes the following mutations at the sites noted: 44:G- A, 51/52:insert AGAA, 87:A-del, 94:A->U, 115:A-U, 116:G- A, 166:C- A, 170:C-U, 188:G-*A, 190:U->A, 191:G--U, 205:U-*C, 215:G- A, 239:U-A, 312:G-A, 350:C-U, and 364:C-U.

Ribozyme G27 #61 has the following mutations: 44:G-A, 51/52:insert AGAA, 87:A-del, 94:A-U, 115:A-U, 116:G->A, 166:C-A, 170:C-U, 188:G- A, 190:U-A, 191:G--U, 205:U-C, 215:G-*A, 313:G- C, and 314:A- G.

Ribozyme G27 #48 includes the following mutations, which are not present in G27 #61: 239:U-A, 312:G->A, 15 350:CU, and 364:C-U. Similarly, ribozyme G27 #61 includes the following mutations, which are absent in G27 #48: 313:G-C and 314:A-G.

3. Synthesis of Hybrid Molecules As noted above, an oligonucleotide is first S 20 prepared. Next, that nucleotide sequence "head" is linked, via an amide bond, to an amino acid residue sequence "tail" to form a hybrid substrate molecule.

Preferably, an entire "series" of hybrid molecules is prepared for use in a continuing in vitro evolutionary process, whereby the first molecule in an exemplary series may comprise an oligonucleotide sequence S. an 8-mer) linked to a polypeptide a monomer or dimer) by an amide bond. For an example, a first hybrid molecule in such a series may comprise an oligonucleotide 8-mer linked to a polypeptide dimer.

Subsequent hybrid molecules in such a series preferably comprise one fewer nucleotide each time the second molecule in the series comprises an oligonucleotide 7-mer linked to a polypeptide trimer; the third molecule comprises an oligonucleotide 6-mer linked to a polypeptide tetramer; and so on, until only a single nucleotide remains at the "head" of the hybrid molecule. Exemplary hybrid molecules in such a -162- "series" are used in a consecutive manner in conjunction with in vitro evolution methodologies as disclosed herein to identify useful enzymatic RNA molecules in successive rounds of mutation, selection, and amplification.

It is also to be understood that while peptide monomers, dimers, and so forth are described as exemplary, a hybrid molecule according to the present invention may comprise longer and more complex polypeptide sequences. That is, hybrid molecules of the present invention may include as few as one or two amino acid residues, or may include substantially longer polypeptides or proteins, provided that the length of the polypeptide "tail" does not substantially oooo 15 interfere with the ability of enzymatic RNA molecules of the present invention to recognize and bind hybrid molecules, or otherwise interfere with cleavage of amide bonds therein.

S* It should also be appreciated that the sequence of 20 nucleotides and/or amino acids may be varied as desired. For example, the nucleotide sequence at the *.ee "head" of the hybrid may be comprised of common and/or *o unusual or modified nucleotides (as described in 37 CFR §H 1.821 et seq.), in any order. Similarly, while 25 certain exemplary hybrid molecules disclosed herein include pairs of identical amino acids in the "tail" sequence, it is expressly to be understood that the amino acid residue sequence of hybrid molecules according to the present invention may be varied, and may include unusual or modified amino acids, as well.

Hybrid molecules according to the present invention are typically designed and constructed so that the nucleotide and amino acid sequences are linked by an amide bond. In general, methods such as those described by Zieboll and Orgel, J. Mol. Evol 38: 561- 565 (1994) and Ehler, et al., Biochim. et Biophys. Acta 434: 233-243 (1976) the disclosures of which are incorporated by reference herein were used and -163adapted as follows.

Typically, a 0.5 M solution of imidazole is first prepared, into which the amino acid of choice is dissolved. In the present example, arginine

(L-

arginine, 98% purity; Aldrich Chem. Co., Milwaukee,

WI)

was dissolved into a 0.5 M imidazole solution, until a final concentration of arginine of 0.1 M was achieved.

Next, 125 Al of the arginine solution was placed into an Eppendorf tube. Two microliters (2 Al) of oligonucleotide is then placed into a separate, clean Eppendorf tube, dried via spin-vac), and cooled placed on ice). In the present example, 2 pl of radiolabeled d(GGCCCTCTN,) (SEQ ID NO 11) Ssynthesized as described in section 1 above was 15 placed into a separate, clean Eppendorf tube, dried, and placed on ice.

Approximately 0.1 mg 1, 1'-carbonyldiimidazole ST("CDI"; Aldrich, Milwaukee, WT) was measured and added into 0.1 M arginine solution; CDI served to "activate" 20 the amino acids. (See Ehler, et al., Id. (1976).) As soon as the CDI dissolved into the solution, the admixture was placed on ice for about 1 minute. About 20 Al of the above-noted solution was added to the tube containing the d(GGCCCTCTN) (SEQ ID NO 11), on ice at about The tube containing this admixture was then transferred into a cold room and incubated. At various time points 30 minutes, minutes), 10 Al of sample was removed, quenched with 2 X gel loading buffer, and placed on ice. Half of each sample (from each time point) was loaded on an 8M ureapolyacrylamide gel, and run according to standard protocols, as described previously.

Figure 12A illustrates the confirmation of successful synthesis of an exemplary oligonucleotideoligopeptide "hybrid". In lane 1, d(GGCCCTCTNm) is shown. In lanes 2 and 3, d(GGCCCTCT)-Arg is shown, as measured at 30 and minutes.

-164- Two hybrid molecules synthesized as described above were isolated and used in cleavage.reactions conducted essentially as described below. The first molecule, identified herein as "oligo-Arg", had the sequence d(GGCCCTCT)-Arg (SEQ ID NO 13); the second molecule, "oligo-Arg 2 had the sequence d(GGCCCTCT)- ArgArg (SEQ ID NO 14).

It is expressly to be understood, however, that the hybrid and polypeptide substrates cleavable by enzymatic RNA molecules of the present invention are not limited to those containing arginine residues only.

Nor are substrates limited to those containing amide bonds; substrates containing peptide bonds and the like are also contemplated. Substrates lacking 15 arginine, and/or substrates further comprising common or unusual/modified amino acids (preferably in L-form) are also contemplated by the within-disclosed invention. For example, amino acids listed in the Table of Correspondence appearing in Section A of the S 20 Detailed Description are useful in the hybrid and polypeptide substrates of the present invention, as are those described in 37 CFR §1.822.

o Hybrid oligonucleotide-oligopeptide molecules useful in the in vitro evolution procedures disclosed 25 herein may also include uncommon amino acids, variants of "common" amino acids, or amino acid analogs, e.g., S-alanine, S-adenosylmethionine, S-adenosylcysteine,

S-

adenosyl-homocysteine, L(+)-canavanine, hydroxyproline, methioninemethylsulfoniumchloride, and w-nitroarginine. (See, Zieboll and Orgel, J.

Mol. Evol. 38: 561-565 (1994).) 4. Amide. Polvpeptide and Protein Substrates As disclosed herein, the enzymatic RNA molecules of the present invention may be engineered to cleave a bond between adjacent amino acids, with great selectivity and specificity. Any amide, polypeptide or protein substrate, whether naturally-occurring (i.e.

"native"), synthesized, derivatized, or conjugated, is -165an appropriate substrate for the enzymatic RNA molecules disclosed herein.

An amide substrate is a compound which includes a scissile amide bond. Preferred amide substrates include peptides and peptide conjugates having an amide linkage between a secondary amine and a terminal peptide carboxy group. For example, 2-'Amino 3'deoxyribose is a preferred secondary amine. When a ribozyme of the present invention hydrolyzes an amide substrate, it produces an amino cleavage product and a ribozyme-acyl intermediate, a ribozyme amidase intermediate. The amino cleavage product includes the secondary amine, 2'-amino 3'-deoxyribose or a peptide fragment having a free amino terminus. A 15 preferred ribozyme amidase intermediate includes an ester linkage between a ribozyme hydroxyl group and the terminal carboxyl group of a peptide. The peptide may include one or more amino acid residues. Additionally, the peptide may be linear or cyclic and may include 20 non-natural amino acids and amino acids incapable of ribosomal translation. Furthermore, the amino acid residues may be either the D or L isomer.

Alternative amide substrates include amide linked glycopeptides and carbohydrates containing acetylated amino sugars, 2-acetamido-N-(L-aspart-4-oyl)-2- S deoxy-p-D-glucopyranosylamine or asparaginyl-N- "acetylglucosamine. Cleavage of either of these amide substrates yields an amino sugar as the amino cleavage product. Cleavage of N-acetyl amino sugars produces a ribozyme amide intermediate having an acetylated deoxynucleotide. Cleavage of glycopeptides produces a ribozyme amide intermediate having an ester linkage between a ribozyme hydroxyl group and a peptide carboxyl group. The peptide carboxyl group may be a terminal carboxyl group or an aspartic acid or glutamic acid residue. Amide linked glycans, including peptidoglycans, are yet a further class of amide substrate cleavable by amidase active ribozymes.

-166- Although polypeptides of any length are cleavable using RNA enzymes of the present invention, one seeking to design a specific enzymatic RNA molecule according to the present invention may find it convenient to utilize shorter, rather than longer, polypeptides in the initial stages of in vitro evolution. For example, while polypeptides comprising 12 or fewer amino acids dimers, tetramers, 8-mers, etc.) are discussed herein as exemplary, it is expressly to be understood that the invention is not so limited.

A polypeptide used as disclosed herein can be derived from an existing source via proteolysis of a larger polypeptide or protein) or synthesized by any of the peptide synthetic techniques known to those 15 skilled in the art. A summary of some of the '0 techniques available can be found in J.M. Stuard and J.

D. Young, "Solid Phase Peptide Synthesis", W. H.

Freeman Co., San Francisco (1969); J. Meinhofer, "Hormonal Proteins and Peptides" Vol. 2, pp. 46, 20 Academic Press (New York) 1983; E. Schroder and K.

Kubke, "The Peptides", Vol. 1, Academic Press (New York), 1965 for classical solution synthesis, and U.S.

Patent No. 4,631,211, the disclosures of which are incorporated herein by reference.

25 When a polypeptide desired for use according to the present invention is relatively short less than about 25-50 amino acid residues in length) direct peptide synthetic techniques are generally favored, usually by employing a solid phase technique such as that of Merrifield (JACS 85: 2149 (1963)). Appropriate protective groups usable in the aforementioned syntheses are described in the above texts and in J.F.W. McOmie, Protective Groups in Organic Chemistry, Plenum Press, New York, 1973, which is incorporated herein by reference.

A polypeptide useful as disclosed herein can also be synthesized by recombinant DNA techniques. Such recombinant techniques are especially favored when the -167desired polypeptide is relatively long (greater than about 50 amino acids residues in length). When recombinant DNA techniques are employed to prepare an instant polypeptide, a DNA segment encoding the desired polypeptide is incorporated into a preselected vector that is subsequently expressed in a suitable host. The expressed polypeptide is then preferably purified by a routine method such as gel electrophoresis, immunosorbent chromatography, and the like.

Again, while initial rounds of in vitro evolution may conveniently be conducted using small polypeptides during the selection process it should be appreciated that enzymatic RNA molecules of the present invention may be engineered to 15 recognize, bind and cleave polypeptides or proteins of a variety of lengths, conformations and biochemical or physical characteristics, by use of the withindisclosed techniques.

B. Cleavage of Hybrid Molecules 20 Six Al of hybrid molecule prepared as described, 2 pl of 5 X low-Mg" buffer, and 2 il ribozyme were admixed and incubated at 37 0 C for about 8 hours, or overnight. After incubation, a sample comprising approximately one-half of the admixture was labeled, loaded and run on an 8 M urea-20% polyacrylamide gel, as before. A sample of 5'-labeled d(GGCCCTCTNH) (SEQ ID NO 11) was also run as a control.

In Figures 10 and 11A-C, cleavage of a hybrid substrate by a ribozyme of the present invention is illustrated and shown to generate an 8-mer 5' product with a terminal -NH 2 For example, Figure illustrates the cleavage of an amide bond-containing substrate, showing that it generates a 5' product that carries a terminal amine and a 3' product that carries a terminal carboxyl. Figures 11A-C further illustrate the reaction shown in Figure 10, including the production of intermediates (Fig. 11B) and products (Fig. 11C), as well as the relationship of the -168substrate to the ribozyme (Fig. 11A). It also shows that the ribozyme-associated product is subsequently hydrolyzed, resulting in generation of a 5' product carrying a terminal amine and a 3' product carrying a terminal carboxyl (Fig. 11C). Subsequent to hydrolysis of the ribozyme-associated product, the enzyme is free to cycle it is free to cleave another amide bond. (See also Hentzen, et al., Biochimica et Biophysica Acta 281: 228-232 (1972) For purposes of illustration only, Figs. 10 and 11 have been drawn to show the amide bond in the context of an oligonucleotide molecule. It is expressly to be understood that one or both of the motifs identified in Fig. 11A-C as "DNA 1" and "DNA 2" 15 may be replaced by the appropriate amino acid structural formulas and labels. For example, if the motif labeled "DNA 2" were replaced with the label "Arg" and the appropriate chemical drawing, the intermediate shown in Fig. 11B would illustrate that 20 the arginine moiety remains temporarily attached to the ribozyme after the peptide bond is cleaved and is subsequently released via hydrolysis (Fig. 11C).

Figure 12B illustrates the results of an exemplary ribozyme-catalyzed cleavage of a hybrid molecule.

:25 Typical reaction conditions are as follows: 1 AM :ribozyme, 1 !M [5'-"P]-labeled substrate, 10 mM MgC1 2 and 30 mM EPPS, at 37 0 C, pH 7.5, for 8 hours.

Figure 12B is a photograph of a gel illustrating cleavage of a hybrid oligonucleotide-oligopeptide substrate by enzymatic RNA molecules of the present invention. In lane 1, 5'-labeled 8-mer marker is shown. In lane 2, interaction of ribozyme with a labeled hybrid substrate generates an 8-mer 5' product with a terminal -NH 2 In lane 3, substrate alone in the absence of ribozyme) is shown.

As shown in Fig. 12B, the 8-mer d(GGCCCTCTmN)

(SEQ

ID NO 11) marker and the 8-mer 5' product of the enzymatic RNA molecule-catalyzed cleavage of the amide- -169bond-containing substrate have the same mobility. This effectively demonstrates the amide-cleaving activity of the enzymatic RNA molecules of the present invention.

The reaction appears to be dependent upon the presence of Mg 2 although other divalent cations are also expected to be useful; for example, use of Mn 2 instead of Mg 2 also produced satisfactory results. In general, the reactions have been run at 37°C for eight hours or overnight, but it is expected that these parameters will continue to be adjusted as in vitro evolution techniques are applied. For example, selection of enzymatic RNA molecules that carry out the cleavage reaction during shorter time periods will likely be favored. Selection of enzymatic RNA 15 molecules that utilize different monovalent or divalent cations may also be a useful choice.

Thus, as shown in Figure 12B, cleavage of hybrid oligonucleotide-oligopeptide substrate by enzymatic RNA molecules of the present invention has been confirmed.

20 It was observed that ribozyme G27 #48 cleaved the amide bond more rapidly than did G27 #61 (data not shown). It was also noted that the cleavage reaction rate decreases as the temperature is raised to 45 0 C and ceases altogether at 50 0 C (not shown) Unlike experiments involving oligonucleotide cleavage, it was observed that the ribozyme might occasionally cleave one position upstream of the bond "intended" to be cleaved in the hybrid molecule, but it did not cleave downstream of the amide bond. (See, Tsang and Joyce, Biochemistry 33: 5966-5973 (1994).) This "miscleavage" event is illustrated in Figure 12B, in lane 2, as evidenced by the band "beneath" the band representing the product generated when cleavage occurs at the desired (preselected) site. Such "mis-cleavage" events may be selectively eliminated from the evolving population, however; by going back to the previous generation of variants and generating a different subpopulation therefrom, or by applying -170selection criteria designed to eliminate ribozymes that cleave at locations other than the one desired.

C. Cleavage of Polypeptide Molecules The following general procedure is useful for confirming site-specific cleavage between preselected, adjacent amino acids. For example, to confirm that an enzymatic RNA molecule capable of site-specific cleavage of a bond between two arginine molecules has been identified, the following procedure may be used.

A small polypeptide substrate an 8-mer) is prepared as described in section A.4 above and preferably includes a single paired Arg-Arg moiety.

Six yl of the peptide substrate, 2 pl of 5X low-Mg 2 buffer, and 2 pl ribozyme are admixed and incubated at 15 37 0 C for about 8 hours, or overnight. Labeling of the polypeptide substrate will facilitate detection of reaction products and confirm site-specific cleavage, as described previously.

After incubation, a sample comprising 20 approximately one-half of the admixture is loaded and run on an appropriate gel, as described above.

A

sample of substrate polypeptide, in the absence of enzyme, may be run as a control; a sample of expected cleavage product is also preferably run as a control.

Examination of the results will confirm whether sitespecific cleavage has occurred. Subsequent evaluations utilizing larger substrate polypeptides may be performed to further confirm site-specific cleavage between preselected amino acids.

Example Alternative Methods of Preparing Enzymatic RNA Molecules One alternative method of preparing wild-type and mutant ribozymes may be described as follows. Wildtype and mutant ribozymes were produced by first isolating the 443 base-pair Eco RI to Hind III restriction endonuclease fragment from the plasmid PT7- 21 described by Zaug et al., Biochemistry 27: 8924 -171- (1988) using the standard methods described in Current Protocols in Molecular Biology, Ausubel et al., eds.

John Wiley and Sons, New York (1987).

This 443 base-pair fragment contains the T7 promoter described by Dunn et al., J. Mol. Biol. 166: 477-535 (1983) and residues 22-414 of the Tetrahymena IVS and residues 1-25 of the 3' Tetrahymena exon described by Been et al., Cell 47: 207-216 (1986).

This Eco RI and Hind III fragment was inserted into the M13 vector M13mpl8 (which is similar to the vector described by Yanisch-Perron et al., Gene 33: 103-119 (1985)), which vector had been previously cleaved with Eco RI Hind III, according to standard subcloning procedures such as those described in Current Protocols 15 in Molecular Biology, Ausubel et al, eds. John Wiley and Sons, New York (1987). The resulting M13T7L-21 DNA construct was used to transform E. coli host cells according to the transformation procedure described in Molecular Cloning: A Laboratory Manual (Maniatis et 20 al., eds., Cold Spring Harbor Laboratories, Cold Spring Harbor, New York (1989)).

Single-stranded DNA was then prepared from the M13T7L-21-transformed cells according to the procedures described in Current Protocols in Molecular Biology 1987). The accuracy of the above construction was confirmed by DNA sequencing using the klenow fragment of E. coli DNA polymerase I (Boehringer Mannheim Biochemicals, Indianapolis, IN) and the dideoxynucleotide sequencing method (see Sanger et al., PNAS USA 74: 5463-5467 (1977)).

The wild-type and mutant ribozymes were prepared directly from the single-stranded M13T7L-21 DNA using a modification of the technique previously described by Joyce and Inoue, Nucleic Acid Research 17: 711-722 (1989). The technique involves construction of a template strand that optionally includes one or more mutagenic oligodeoxynucleotides. The resulting partially-mismatched double-stranded DNA is transcribed -172directly using T7 RNA polymerase.

Briefly, the procedure-'is as follows. A five-fold molar excess of a terminator polynucleotide and a mutator oligonucleotide were admixed with 5 yg of single-stranded M13T7L-21 DNA and a solution containing mM tris[hydroxy-methyl]aminomethane adjusted to pH with HC1(Tris-HC1), 50 mM NaCI and 2 mM MgCl 2 This solution was maintained at 70 degrees centigrade (70 0

C)

for 5 minutes and then steadily cooled to 30 0 C over minutes. Fifteen units(U) of T4 DNA ligase (U.S.

Biochemicals, Cleveland, OH) and 7.5 U of T4 DNA polymerase Biochemicals) were admixed into the solution, together with sufficient amounts of reagents to make a solution containing a final concentration of 15 20 mM Tris-HC1 at pH 7.5, 50 mM NaC1, 5 mM MgCl 2 2 mM dithiothreitol (DTT), 1 mM adenosine triphosphate (ATP), and 0.5 mM each of dGTP, dTTP, dATP and dCTP (dNTPs). The resulting solution was maintained at 37 0

C

for 60 minutes to complete the synthesis of the mutant 20 strand. The resulting DNA was purified by ethanol precipitation and then used to direct the transcription of mutant RNA.

Transcription took place either in a 10 pl volume containing 1 ig of mutant DNA, 2 iCi [0 2 P] GTP and 50 U 25 of T7 RNA polymerase that was prepared as previously described by Davanloo et al., PNAS USA 81: 2035-2039 (1984), and the resulting product was purified according to a procedure originally developed by Butler Chamberlain, J. Bio. Chem. 257: 5772-5779 (1982), or in a 400 il volume containing 10 pg of mutant DNA, ACi ['H]UTP and 2,400 U of T7 RNA polymerase. In either case, the transcription mixture also contained 40 mM Tris-HCl at pH 7.5, 15 mM MgCl 2 10 mM dithiothreitol, 2 mM spermidine, and 1 mM (each) NTPs, and was incubated at 37 0 C for 90 minutes. The T7 RNA polymerase was extracted with phenol and the transcription products were purified by ethanol precipitation. The mutant RNA was isolated by electrophoresis in a -173polyacrylamide/8 M urea gel, eluted from the gel, and purified by ethanol precipitation and chromatography on Sephadex The 3' exon sequence was removed by RNA-catalyzed site-specific hydrolysis as has been previously, Inoue et al., J. Mol. Biol. 189: 143-165 (1986). Briefly, the RNA was incubated in the presence of 50 mM CHES at pH 9.0 and 10 mM MgCl 2 at 42°C for 1 hour. Wild-type and mutant RNAs were isolated by electrophoresis in a 5% polyacrylamide/8M urea gel, eluted from the gel, and purified by affinity chromatography on du Pont Nensorb (du Pont, Wilmington, DE). RNAs were sequenced by primer extension analysis using AMV reverse Stranscriptase (Life Technologies, Inc., Gaithersburg, 15 MD) in the presence of dideoxynucleotides, using a modification of the methods described by Sanger et al.

(PNAS USA 74: 5463-5467 (1977)), except for those containing the Delta P9 deletion (not shown), which were sequenced from the 3' end by partial RNase 20 digestion, Donis-Keller et al., Nucleic Acids Res. 8783-8798 (1987) Other methods of preparing enzymatic RNA molecules of the present invention are based on chemical synthesis. Methods useful in the chemical synthesis of RNA are similar to those used to synthesize DNA. The additional 2' -OH group in RNA, however, requires a different protecting group strategy to deal with *9 selective internucleotide bond formation, and with RNA susceptibility to degradation in the presence of bases.

The recently-developed method of RNA synthesis utilizing the t-butyldimethylsilyl group for the Sprotection of the 2' hydroxyl seems to be the most reliable method for chemical synthesis of ribozymes.

The method reproducibly yields RNA with the correct 3'internucleotide linkages, with average coupling yields in excess of 99%, and requires only a two-step de-protection of the polymer.

-174- Other useful methods are available. For example, -published PCT application no. WO 93/23569 (the disclosures of which are incorporated by reference herein) describes other useful methods of chemically synthesizing ribozymes.

The foregoing specification, including the specific embodiments and examples, is intended to be illustrative of the present invention and is not to be taken as limiting. Numerous other variations and modifications can be effected without departing from the true spirit and scope of the present invention.

*o 0* 175 SEQUENCE LISTING GENERAL INFORMATION:

APPLICANT:

ADDRESSEE: The Scripps Research Institute, Office of Patent Counsel STREET: 10666 North Torrey Pines Road, TPC-8 CITY: La Jolla STATE: California COUNTRY: USA POSTAL CODE (ZIP): 92037 TELEPHONE: 619-554-2937 TELEFAX: 619-554-6312 (ii) TITLE OF INVENTION: NOVEL ENZYMATIC RNA MOLECULES (iii) NUMBER OF SEQUENCES: 29 (iv) COMPUTER READABLE FORM: MEDIUM TYPE: Floppy disk COMPUTER: IBM PC compatible OPERATING SYSTEM: PC-DOS/MS-DOS SOFTWARE: PatentIn Release Version #1.25 (EPO) CURRENT APPLICATION DATA: APPLICATION NUMBER: PCT/US FILING DATE: 26-APR-1995 (vi) PRIOR APPLICATION DATA: APPLICATION NUMBER: US 08/242,402 FILING DATE: 13-MAY-1994 (vi) PRIOR APPLICATION DATA: APPLICATION NUMBER: US 08/270,180 FILING DATE: 01-JUL-1994 INFORMATION FOR SEQ ID NO:1: SEQUENCE CHARACTERISTICS: LENGTH: 393 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1: a a 176 GGAGGGAAAA GIJUAUCAGGC AUCCACCUCO UAGCUAGUCU GGUUUAAAAG GCAAGACCGU CAAAUUGCGG GAAAGGUC UCUCAGCCA AACUUUGAGA UGGCCUUGCA AACGGUAUCG UCCUAACCAC GCAGCCAAGU CCUAAGUCAA CAGAUCUUCU CAGACUAAAU GUCGGUCGGG GAAGAUGUAU UCUUCUCAUA UUAAUGGGAG CUAGCCGAUG AACUGAUGCA ACACUGGAGC GCGAAAGUAU AUUGAUUAGU UUUGGAGUAC UCG INFORMATION FOR SEQ ID NO:2: SEQUENCE CHARACTERISTICS: LENGTH: 5 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ 1D NO:2:

NNNNA

INFORMATION FOR SEQ ID NO:3: SEQUENCE CHARACTERISTICS: LENGTH: 5 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

AACAA

INFORMATION FOR SEQ ID NO:4: SEQUENCE CHARACTERISTICS: LENGTH: 27 base pairs TYPE: nucleic acid STRANDEDNESS: single

UUAAACCAAU

AACAGCCGUU

UAAUAACCUG

GUUGAUAUGG

AGAUAUACUC

CGCUGGGAAC

AGAUUCCAUC

CACUACCAAG

ACCCACAUGG

AUGCAGUUCA

GCACCUCUCC

UAAUUUGUAU

120 180 240 300 360 393 177 TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: AGUUACCAGG CAUGCACCUG GUAGUCA 27 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 36 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID GUCUUUAAAC CAAUAGAUUG GAUCGGUUUA AAAGGC 36 INFORMATION FOR SEQ ID NO:6: SEQUENCE CHARACTERISTICS: LENGTH: 15 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: TTTATTTATT TATTT INFORMATION FOR SEQ ID NO:7: SEQUENCE CHARACTERISTICS: LENGTH: 48 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 178 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: CTGCAGAATT CTAATACGAC TCACTATAGG AGGGAAAAGT TATCAGGC 48 INFORMATION FOR SEQ ID NO:8: SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: CGAGTACTCC AAAACTAATC INFORMATION FOR SEQ ID NO:9: SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) S (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: GTAAAACGAC GGCCAGT 17 INFORMATION FOR SEQ ID *e SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID CATGATTACG AATTCTA 1- 179 INFORMATION FOR SEQ ID NO:11: SEQUENCE CHARACTERISTICS: LENGTH: 8 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: NAME/KEY: misc_feature LOCATION: 8 OTHER INFORMATION: /label- NH2 /note- "NH2 SIGNIFIES THAT THE T HAS BEEN MODIFIED AND IS 3'-AMINO'3'DEOXYTHYMIDINE-5'-TRIPHOSPHATE" 0 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:11: GGCCCTCT 8 **ooo INFORMATION FOR SEQ ID NO:12: SEQUENCE CHARACTERISTICS: LENGTH: 7 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 0*e* (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12: SGGCCCTC 7 INFORMATION FOR SEQ ID NO:13: SEQUENCE CHARACTERISTICS: LENGTH: 8 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: 180 NAME/KEY: modified_base LOCATION:.8 OTHER INFORMATION: /mod_base- OTHER /label- ARG /note- "ARG SIGNIFIES THAT THE AMINO ACID ARGININE IS COVALENTLY LINKED (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13: GGCCCTCT 8 INFORMATION FOR SEQ ID NO:14: SEQUENCE CHARACTERISTICS: LENGTH: 8 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURKE: NAME/KEY: modifiedbase LOCATION: 8 OTHER INFORMATION: /modbase- OTHER /label- ARGARG /note- "ARGARG SIGNIFIES THAT THE T HAS THE AMINO ACID ARGININE COVALENTLY LINKED TO IT AND THAT A SECOND ARGININE IS COVALENTLY LINKED TO THE FIRST (xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:

GGCCCTCT

INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID GGCCCTCTAT TTATTTA 181 INFORMATION FOR SEQ ID NO:16: SEQUENCE CHARACTERISTICS: LENGTH: 8 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:16: GGCCCTCT 8 INFORMATION FOR SEQ ID NO:17: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17: GGCCCTCTAA ATAAATAAAT AAA 23 INFORMATION FOR SEQ ID NO:18: SEQUENCE CHARACTERISTICS: LENGTH: 21 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: NAME/KEY: misc_feature LOCATION: 6 OTHER INFORMATION: /label- N /note- "N SIGNIFIES A NUCLEOTIDE ANALOG" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18: 182 CCCTCNAAAT AAATAAATAA A 21 INFORMATION FOR SEQ ID NO:19: SEQUENCE CHARACTERISTICS: LENGTH: 6 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: NAME/KEY: miscfeature LOCATION: 6 OTHER INFORMATION: /label- N /note- "N SIGNIFIES A NUCLEOTIDE ANALOG" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:19: CCCTCN 6 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID GGCCCUCUAU UUAUUUA 17 INFORMATION FOR SEQ ID NO:21: SEQUENCE CHARACTERISTICS: LENGTH: 15 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 183 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:21: AAATAAATAA ATAAA INFORMATION FOR SEQ ID NO:22: SEQUENCE CHARACTERISTICS: LENGTH: 16 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22: TTTATTTATT TATTTC 16 INFORMATION FOR SEQ ID NO:23: e SEQUENCE CHARACTERISTICS: LENGTH: 42 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:23: ATCGATAATA CGACTCACTA TAGGAGGGAA AAGTTATCAG GC 42 INFORMATION FOR SEQ ID NO:24: SEQUENCE CHARACTERISTICS: LENGTH: 9 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24:

GGCCCTCTA

184 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 16 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID AAATAAATA AATAAAA 16 INFORMATION FOR SEQ ID NO:26: SEQUENCE CHARACTERISTICS: S(A) LENGTH: 24 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ij) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:26: S GGCCCTCTAA ATAAATAAAT AAAA 24 INFORMATION FOR SEQ ID NO:27: SEQUENCE CHARACTERISTICS: LENGTH: 33 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27: CCAAGCTTGA TCTCGAGTAC TCCAAAACTA ATC 33 INFORMATION FOR SEQ ID NO:28: SEQUENCE CHARACTERISTICS: LENGTH: 24 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear 185 (ii) MOLECULE TYPE: RN~A (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:28: GGCCCUCUCA AAUAAAUAAA UAAA 24 INFORMATION FOR SEQ ID NO:29: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29: GGCCCTCTAA ATAAATAAAT AAA 23

Claims (1)

186- THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS: 1. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains a mutation selected from the group consisting of: 44: 51/52: insert AGAA; 87: A deleted; 115: 116: 138: 166: C-A; 167: 170: 188: G-+A; 190: U-A; 191: G-U; 239: 258: U-C; 313: 350: C-U and 364: C-U. 2. An enzymatic RNA molecule according to claim 1 provided that the sequence contains one mutation selected from the group consisting of: 44: 51/52: insert AGAA; 87: A deleted; 115: 116: G-A; 138: C-A; 166: C-+A; 167: U- G; 170: 188: 190: 191: 239: 258: U-C; 313: G-C; 350: C-U and 364: C-+U 15 3. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 215: G-A and 258: U-+C. 4. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence 20 contains the mutations: 188: 190: U-A and 191: G-U. 5. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 115: 116: G-A and 205: U-+C. 6. An enzymatic RNA molecule which has a nucleotide 25 sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 115: 116: G-A; 188: G-+A; 190: 191: G-+U and 205: U-+C. 7. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 94: A-Y; 115: 116: 188: G-A; 190: 191: 205: 215: and 313-314: GA-UG. 8. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 44: G- A; 87: A del; 94: A-U; 115: \\GHSYDNTusrs\Spec300 399300 349\34938.doc 21/07/99 -187- A-U; 116: 166: C-A; 170: 188: 190: U-A; 191: 205: and 215: G-+A. 9. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 44: G-A; 94: 115: A-U; 116: G- A; 138: 188: G-A; 190: 191: 205: U-+C; 215: G-A; 312: G-A and 317: U-G. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 44: 94: A-U; 115: 116: 138: C-A; 167: U-G; 188: G-A; 190: 191: G-U; 205: 215: 239: U-+A and 312: G-A. 11. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 provided that the sequence 5 contains the mutations: 44: 51/52: insert AGAA; 87: A del; 94: 115: 116: G-A; 166: 170: C-U; 188: 190: 191: 205: U-C; 215: G-A; 239: U-A; 312: 350: C-U and 364: C-U. 12. An enzymatic RNA molecule which has a nucleotide 20 sequence shown in SEQ ID NO: 1 provided that the sequence contains the mutations: 44: 51/52: insert AGAA; 87: A del; 94: A-U; 115: 116: G-A; 166: C-A; 170: C-U; 188: G-A; 190: 191: 205: 215: G-A; 313: G-C; and 314: A-G. 25 13. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO:1 wherein the molecule is a modified RNA. 14. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO:1 wherein the molecule is an RNA-DNA polymer. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO:1 wherein the molecule is a modified RNA-DNA polymer. 16. A enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO: 1 wherein the molecule is a modified DNA-RNA polymer. \\GHSYDNTI\users\Speci300 39930OO 349\34938.doc 21/07/99 -188- 17. An enzymatic RNA molecule which has a nucleotide sequence shown in SEQ ID NO:1 wherein the molecule is a modified RNA- modified DNA polymer. 18. A composition comprising an enzymatic molecule according to any one of claims 1 to 17 and a carrier. 19. A kit comprising an enzymatic molecule according to any one of claims 1 to 17. A method for cleaving a nucleic acid molecule, the method comprising contacting the nucleic acid molecule with an enzymatic RNA molecule according to any one of claims 1 to 17 under conditions suitable for cleaving the nucleic acid molecule so that the nucleic acid molecule is cleaved. 21. A method according to claim 20 wherein the 15 conditions suitable for cleaving the nucleic acid molecule S" with the enzymatic RNA molecule include a Mg 2 concentration of 2-100 mM, a pH of 6.0 to 9.0, and a temperature of 15 0 C to 60 0 C. 22. A method according to claim 21 wherein the S 20 conditions suitable for cleaving the nucleic acid molecule with the enzymatic RNA molecule further include a polyamine concentration of 0.01 mM to 15 mM. oeo• ~23. A method according to claim 20 wherein the conditions suitable for cleaving the nucleic acid molecule 25 with the enzymatic RNA molecule are physiologic conditions. "24. A method according to claim 20 wherein the nucleic acid molecule is single stranded, looped, or partially or fully double stranded. A method according to claim 20 wherein the nucleic acid molecule is cleaved in vivo or in vitro. 26. A method according to claim 20 wherein the nucleic acid molecule comprises DNA, RNA or modified RNA. 27. A method according to claim 26 wherein the nucleic acid molecule is chemically synthesised, enzymatically produced or isolated from a source selected \\GlSYDNTlXusersSpecA3OO 399X300- 349\3493.doc 21107i99 -189- from the group consisting of phage, virus,prokaryotic cells or eukaryotic cells. 28. A method of treating viral disease in a patient, the method comprising administering a therapeutically effective amount of an enzymatic RNA molecule according to any one of claims 1 to 17, or a composition according to claim 18 to the patient. 29. A method according to claim 28 wherein the viral disease is caused by virus selected from the group consisting of EBV, HSV, HBV, HIV, T-cell leukemia virus, HCV, CMV, influenza, picornavirus, FIV, FLV, SIV, BLV and simian leukemia virus. A method of treating a transformed eukaryotic cell, the method comprising administering an amount of an 15 enzymatic RNA molecule according to any one of claims 1 to 18 to the cell which is sufficient to inhibit the expression of viral genes that cause transformation of the cell. 31. A method according to claim 30 wherein the S 20 transformed eukaryotic cell is a keratinocyte, hepatocyte or epithelial cell. 32. A method of detecting the presence of a nucleotide sequence mutation in a nucleic acid molecule, the method comprising: 25 a)contacting an enzymatic RNA molecule according to any one of claims 1 to 17 with the nucleotide sequence of the nucleic acid molecule containing the mutation under conditions suitable for cleavage of the nucleic acid molecule by the enzymatic RNA molecule and b)detecting the presence of cleavage products of the nucleic acid molecule, wherein the presence of cleavage products indicates the presence of the nucleotide sequence mutation in the nucleic acid molecule. 33. A method of detecting the binding of an adjunct to a nucleic acid molecule, the method comprising: \\GHSYNTIuserSpcA300 3993OO 349\34938.doc 21/07199 -190- a) contacting an enzymatic RNA molecule according to any one of claims 1 to 17 with the nucleotide sequence of the nucleic acid molecule to which the adjunct is capable of binding under conditions suitable for cleavage of the nucleic acid molecule by the enzymatic RNA molecule and b) detecting the presence of cleavage products of the nucleic acid molecule, wherein the absence of cleavage products indicates the binding of the adjunct to the nucleic acid molecule. 34. A method of inhibiting transcription from a template nucleic acid molecule, the method comprising contacting an enzymatic RNA molecule according to any one of claims 1 to 17 under conditions suitable for the cleavage of the template nucleic acid molecule by the !iii 15 enzymatic RNA molecule so that the template nucleic acid molecule is cleaved by the enzymatic RNA molecule. 35. A method of inhibiting transcription from a template nucleic acid molecule, the method comprising contacting an enzymatic RNA molecule according to any one of claims 1 to 17 with an oligonucleotide which is hybridised to the template nucleic acid molecule under conditions suitable for cleavage of the oligonucleotide by o.e. the enzymatic RNA molecule, so that the oligonucleotide is ooo. cleaved by the enzymatic RNA molecule. 25 36. A method according to claim 34 or 35 wherein the transcription is reverse transcription. 37. A method of isolating an enzymatic RNA molecule which has a predetermined activity, the method comprising: a) mutating an enzymatic RNA molecule according to any one of claims 1 to 12 so as to produce a population of mutated molecules; b) selecting an enzymatic RNA molecule having the predetermined activity from the population of mutated molecules; and c) isolating the selected enzymatic RNA molecule from the population of mutated molecules. \\GHSYDNTrIusers\Spec300 399N300 349\34938.doc 21107/99 -191- 38. A method according to claim 37 wherein a population of enzymatic RNA molecules containing molecules according to any of claims 1 to 12 is mutated so as to produce the population of mutated molecules. 39. A method according to claim 37 wherein the enzymatic RNA molecule is mutated by random mutagenesis. A method according to claim 37 wherein the enzymatic RNA molecule is mutated by site directed mutagenesis. 41. A method according to claim 37 wherein the internal guide sequence (IGS) of the enzymatic RNA molecule is mutated. 42. A method according to claim 37 wherein the predetermined activity includes increased catalytic activity, decreased KM, enhanced substrate binding ability, or altered substrate specificity. 43. A method according to claim 34 wherein the enzymatic RNA molecule having predetermined activity is selected according to the following steps: 20 cleaving a substrate molecule for which the enzymatic RNA molecule has predetermined activity so as to form a ligation product in which the 3' portion of the substrate molecule is ligated to the 3' end of the enzymatic RNA molecule; 25 (ii) producing a complementary copy of the nucleotide sequence of the ligation product by reverse transcription from the 3' portion of the substrate molecule; and (iii) producing a copy of the enzymatic RNA molecule by transcription from the complementary copy of the nucleotide sequence of the ligation product. Dated this 21st day of July 1999 THE SCRIPPS RESEARCH INSTITUTE By their Patent Attorneys GRIFFITH HACK \\GHSYDNTusersSpcc300 399\3OO 349\34938.doc 21107/99