NUCLEIC ACID LIGANDS OF TISSUE TARGET FIELD OF THE INVENTION Described herein are methods for identifying and preparing nucleic acid ligands to tissues. Tissues are described herein as a collection of macromolecules in a heterogeneous environment. According to this definition, tissues encompass a single cell type, a collection of cell types, an aggregate of cells or an aggregate of macromolecules. The method utilized herein for identifying such nucleic acid ligands is called SELEX, an acronym for Systematic Evolution of Ligands by EXponential enrichment. Specifically disclosed herein are high- affinity nucleic acid ligands which bind to various tissues.

BACKGROUND OF THE INVENTION A method for the in vitro evolution of nucleic acid molecules with highly specific binding to target molecules has been developed. This method, Systematic Evolution of Ligands by EXponential enrichment, termed SELEX, is described in United States Patent Application Serial No. 07/536,428, entitled"Systematic Evolution of Ligands by Exponential Enrichment,"now abandoned, United States Patent Application Serial No. 07/714,131, filed June 10,1991, entitled"Nucleic Acid Ligands,"now United States Patent No. 5,475,096, United States Patent Application Serial No. 07/931,473, filed August 17,1992, entitled "Methods for Identifying Nucleic Acid Ligands,"now United States Patent No.

(see also PCT Publication WO 91/19813), each of which is herein specifically incorporated by reference. Each of these applications, collectively referred to herein as the SELEX Patent Applications, describes a fundamentally novel method for making a nucleic acid ligand to any desired target molecule.

The SELEX method involves selection from a mixture of candidate oligonucleotides and step-wise iterations of binding, partitioning and amplification, using the same general selection scheme, to achieve virtually any desired criterion of binding affinity and selectivity.

Starting from a mixture of nucleic acids, preferably comprising a segment of randomized sequence, the SELEX method includes steps of contacting the mixture with the target under conditions favorable for binding, partitioning unbound nucleic acids from those nucleic acids which have bound specifically to target molecules, dissociating the nucleic acid-target

partitioning, dissociating and amplifying through as many cycles as desired to yield highly specific, high affinity nucleic acid ligands to the target molecule.

The basic SELEX method has been modified to achieve a number of specific objectives. For example, United States Patent Application Serial No. 07/960,093, filed October 14,1992, entitled"Method for Selecting Nucleic Acids on the Basis of Structure," now abandoned, describes the use of SELEX in conjunction with gel electrophoresis to select nucleic acid molecules with specific structural characteristics, such as bent DNA. United States Patent Application Serial No. 08/123,935, filed September 17,1993, entitled "Photoselection of Nucleic Acid Ligands,"now abandoned, describes a SELEX based method for selecting nucleic acid ligands containing photoreactive groups capable of binding and/or photocrosslinking to and/or photoinactivating a target molecule. United States Patent Application Serial No. 08/134,028, filed October 7,1993, abandoned in favor of United States Patent Application Serial No. 08/443,957, now United States Patent No. 5,580,737, entitled"High-Affinity Nucleic Acid Ligands That Discriminate Between Theophylline and Caffeine,"describes a method for identifying highly specific nucleic acid ligands able to discriminate between closely related molecules, termed Counter-SELEX. United States Patent Application Serial No. 08/143,564, filed October 25,1993, abandoned in favor of United States Patent Application Serial No. 08/461,069, now United States Patent No.

5,567,588, entitled"Systematic Evolution of Ligands by EXponential Enrichment: Solution SELEX,"describes a SELEX-based method which achieves highly efficient partitioning between oligonucleotides having high and low affinity for a target molecule. United States Patent Application Serial No. 07/964,624, filed October 21,1992,"Nucleic Acid Ligands to HIV-RT and HIV-1 Rev,"now United States Patent No. 5,496,938, describes methods for obtaining improved nucleic acid ligands after SELEX has been performed. United States Patent Application Serial No. 08/400,440, filed March 8,1995, now United States Patent No.

5,705,337, entitled"Systematic Evolution of Ligands by EXponential Enrichment: Chemi- SELEX,"describes methods for covalently linking a ligand to its target.

The SELEX method encompasses the identification of high-affinity nucleic acid ligands containing modified nucleotides conferring improved characteristics on the ligand, such as improved in vivo stability or improved delivery characteristics. Examples of such modifications include chemical substitutions at the ribose and/or phosphate and/or base positions. SELEX-identified nucleic acid ligands containing modified nucleotides are

abandoned in favor of United States Patent Application Serial No. 08/430,709, now United States Patent No. 5,660,985, entitled"High Affinity Nucleic Acid Ligands Containing Modified Nucleotides,"that describes oligonucleotides containing nucleotide derivatives chemically modified at the 5-and 2'-positions of pyrimidines. United States Patent Application Serial No. 08/134,028, supra, describes highly specific nucleic acid ligands containing one or more nucleotides modified with 2'-amino (2'-NH2), 2'-fluoro (2'-F), and/or 2'-O-methyl (2'-OMe). United States Patent Application Serial No. 08/264,029, filed June 22,1994, entitled"Novel Method of Preparation of Known and Novel 2'-Modified Nucleosides by Intramolecular Nucleophilic Displacement,"now abandoned, describes oligonucleotides containing various 2'-modified pyrimidines.

The SELEX method encompasses combining selected oligonucleotides with other selected oligonucleotides and non-oligonucleotide functional units as described in United States Patent Application Serial No. 08/284,063, filed August 2,1994, entitled"Systematic Evolution of Ligands by Exponential Enrichment: Chimeric SELEX,"now United States Patent No. 5,637,459, and United States Patent Application Serial No. 08/234,997, filed April 28,1994, entitled"Systematic Evolution of Ligands by Exponential Enrichment: Blended SELEX,"now United States Patent No. 5,683,867, respectively. These applications allow the combination of the broad array of shapes and other properties, and the efficient amplification and replication properties, of oligonucleotides with the desirable properties of other molecules. Each of the above described patent applications which describe modifications of the basic SELEX procedure are specifically incorporated by reference herein in their entirety.

Without question, the SELEX process is very powerful. However, to date the process has been successfully demonstrated primarily with pure, simple targets, such as proteins or small molecules. The present invention provides the first demonstration that complex targets are also compatible with the SELEX process.

It is desirable to be able to obtain nucleic acid ligands to complex tissue targets for various reasons. First, tissue SELEX can be useful to obtain nucleic acid ligands when a distinct target is unknown but a general mode of action of the desired ligand is suggested.

Second, tissue SELEX can be useful when nucleic acid ligands are desired based on functional results. Third, it can be desirable to obtain nucleic acid ligands to a complex tissue target when it is unclear which single target would be effective. It is also useful to obtain nucleic acid ligands to a complex tissue target if the purified target is unavailable or unstable SUBSTITUTE SHEET (RULE 26)

BRIEF SUMMARY OF THE INVENTION The present invention includes methods of identifying and producing nucleic acid ligands to complex targets such as tissues and the nucleic acid ligands so identified and produced. More particularly, nucleic acid ligands are provided that are capable of binding specifically to tissues which are macromolecules in a heterogeneous environment, such as whole cells or substructures thereof, aggregates of cells, collections of cells, aggregates of macromolecules and the like.

Further included in this invention is a method of identifying nucleic acid ligands to tissues comprising the steps of (a) preparing a candidate mixture of nucleic acids, (b) partitioning between members of said candidate mixture on the basis of affinity to tissue. and (c) amplifying the selected molecules to yield a mixture of nucleic acids enriched for nucleic acid sequences with a relatively higher affinity for binding to tissue. Also included are nucleic acid ligands identified according to such method.

Another embodiment of the invention includes methods wherein a negative selection is performed in order to perfect the discrimination between subtle differences of similar tissue types. In this embodiment, the resulting ligands are specific not only for a particular tissue type, but can discriminate between subtly different tissues of the same type. For example, this method can discriminate between normal and abnormal tissue types, between induced and uninduced tissue types, etc.

In another embodiment of the invention, a method is provided for identifying previously unknown or uncharacterized epitopes which are components of a larger unknown macromolecule on the tissue target. The ligands that are evolved by the present invention are capable of binding to previously unknown epitopes and the macromolecule which comprises the unknown epitope can then be identified by standard methods. For example, ligands can be evolved to a previously unknown protein found in the context of a complex tissue target.

The ligand of the invention can be used to purify the protein away from the tissue target by standard protein purification and identification methods. These standard methods include affinity purification, microsequencing and cDNA databank searches. In this aspect, the newly identified epitopes which are components of a larger unknown macromolecule, such as new or previously uncharacterized proteins, are provided by the invention. These new epitopes and the macromolecules of which they are a component will be useful as diagnostic and therapeutic agents as well as the ligands that helped identify them.

More specifically, the present invention includes nucleic acid ligands to peripheral blood mononuclear cells (PBMC), clots, arterial cells and blood vessels, including arteries, including those ligands shown in Tables 2,5,8,10,12 and 13, respectively. Also included are nucleic acid ligands to the above-described tissues that are substantially homologous to any of the given ligands and that have substantially the same ability to bind the above- described tissues. Further included in this invention are nucleic acid ligands to the above- described tissues that have substantially the same structural form as the ligands presented herein.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 shows a schematic representation used in the carotid artery SELEX procedures.

Figure 2 illustrates the progression of the in vivo rat balloon injured carotid SELEX displayed as the accumulation of radioactive RNA bound per wet weight of tissue (CPM/mg) <BR> <BR> <BR> as a function of the number of SELEX rounds. Four to six nmol 32P body-labeled 2'-F RNA (10-15 pM in PBS, 0.1% (w/v) human serum albumin) was injected into a 14-day balloon injured rat via the tail vein. At specific times (between 15 and 60 minutes) the animal was exsanguinated, perfused with normal saline via heart cannulation until fluid from the inferior vena cava was clear, dissected, and tissues were taken for RNA extraction and quantitation.

Blood (B), intercostal muscle (M), normal (N) and balloon denuded carotid artery (I) were harvested for comparison at each round. Shown are the biodistribution profiles of in vivo round 0 (15 minutes), 4 (15 minutes), 6 (30 minutes), 8 (60 minutes), and 10 (60 minutes).

Figure 3 shows the predicted secondary structure of clone 12.2 (SEQ ID NO: 242) identified in Example 4. Lower case letters represent the fixed regions and upper case letters represent the randomized region. Nucleotide position numbers begin at the 5'-end.

Truncation analysis was performed by deleting or substituting regions of clone 12.2 (boxed with a dashed line) and assaying activity in vitro. Alterations that reduce or eliminate activity are bracketed above the structure, and those that have no effect on activity are bracketed below the structure.

Figure 4 shows the predicted secondary structure of truncate 12.2t55L (SEQ ID NO: 259). Residues are numbered to match their positions in Figure 3. Arrows indicate deleted residues #13-41 and #86-87 of clone 12.2. The UCGAC loop of the 3'stem-loop of clone 12.2 has been substituted with the sequence GAAA SUBSTITUTE SHEET (RULE 26)

Figure 5 illustrates the accumulation of radioactive clone 12.2 (SEQ ID NO: 242) and control RNA bound per wet weight of blood (Bld), muscle (Mscl), normal (Nml) and balloon injured (Inj) carotid artery tissue presented as CPM/mg and% injected dose/g (% ID/g).

Figure 6 shows a phosphoimage of a gel obtained by the following procedure: 20 mg of blood, muscle and balloon injured carotid artery tissue and 15 mg of normal carotid artery were extracted and applied to an 8% polyacrylamide, 7M Urea denaturing gel, visualized with a phosphor capture screen (Fuji). Lane 1 is the injected does; lane 2 is blood (Bld), lane 3 is muscle (Mscl), lane 4 is normal artery (Nml) and lane 5 is balloon injured artery (Inj).

Figure 7 illustrates the accumulation of radioactive clones from round 10 of a WHHL in vivo SELEX after in vitro incubation with plaque-containing arterial segments presented as CPM/mm2.

Figure 8 illustrates accumulation of WHHL clone 10.5 (SEQ ID NO: 260) and the negative control 40N8 (unevolved library; SEQ ID NO: 40) in human atherosclerotic artery segments.

Figure 9 illustrates the fraction of 32p ethanol precipitable counts for various tissues after intravenous injection into a WHHL rabbit.

Figure 10 summarizes the data of the concentration of 99mTc-10.31 (SEQ ID NO: 355) in WHHL blood presented as% ID/g as a function of time.

Figure 1 lA illustrates a Region of Interest (ROI) analysis of liver (open circle) and kidney (closed circle) accumulation of 99mTc-10.31 (SEQ ID NO: 355) over time.

Figure 11B illustrates a Region of Interest (ROI) analysis of aortic arch (open circle) and heart (closed circle) accumulation of 99mTc-10. 31 (SEQ ID NO: 355) over time.

Figure 12 summarizes the biodistribution data of 99mTc-10.31 (SEQ ID NO: 355) in various WHHL rabbit tissues sixty minutes after i. v. injection presented as% ID/g.

Figure 13 summarizes the biodistribution of 99m-Tc-10.31 (SEQ ID NO: 355) of selected WHHL (solid bars) and NZW (dashed bars) rabbit tissues sixty minutes after i. v. injection presented as% ID/g.

Figure 14 summarizes the biodistribution of 99m-Tc-10.31 (SEQ ID NO: 355) in various WHHL (solid bars) and NZW (dashed bars) rabbit tissues sixty minutes after i. v. injection presented as% ID/g.

Figure 15 shows the structure of a Technitium-99-complex of the conjugate 5'- [ (mercaptoacetyl-glycyl-glycyl-amidyl)-6-hex-1-yl] phosphoric acid ester of ligand 10.31 (SEQ ID NO: 355). The lower case letters in the ligand sequence represent the fixed sequence

regions.

Figure 16 shows the sequence of clone 10.31 (SEQ ID NO: 268), a truncate of clone 10.31 in which both the 5'and 3'fixed regions have been removed (SEQ ID NO: 356), a truncate of clone 10.31 in which the 5'-fixed region has been removed (SEQ ID NO: 357) and a truncate of clone 10.31 in which the 3'-fixed region has been removed (SEQ ID NO: 358).

Figure 17 shows a proposed secondary structure of a 37-nucleotide truncate (SEQ ID NO: 359) of clone 10.31 (SEQ ID N0: 268).

Figure 18 summarizes the in vitro accumulation assays of analogs of Tr104 (SEQ ID NO: 359), synthesized with increasing numbers of 2'-OH purines substituted with 2'-OMe purines, in WHHL plaque presented as CMP/mm2.

Figure 19 shows the structure of Trl23 having a 5'-pentyl amine and a 3'-3'thymidine (SEQ ID NO: 360).

Figures 20A-H show the structures of 2'-OMe analogs of Trl30, where mG represents 2'-O-methoxy guanosine and mA represents 2'-O-methoxy adenosine.

Figure 21 summarizes the in vitro accumulation assays of various 2'-OMe analogs of Trl30, presented as CPM/mm2.

Figure 22 is a graph summarizing the effect of O-methylation on nuclease resistance for three ligands: Tr104 9 (SEQ ID NO: 359; closed circles, dashed line), Trl28 (SEQ ID NO: 361; closed circles, solid line) and Trl29 (SEQ ID NO: 362; open squares), presented as% full-length transcript versus time.

DETAILED DESCRIPTION OF THE INVENTION This application describes nucleic acid ligands to complex tissue targets identified generally according to the method known as SELEX. As stated earlier, the SELEX technology is described in detail in the SELEX Patent Applications which are incorporated herein by reference. This method, referred to as Tissue SELEX, incorporates complex targets in contrast to the more simple targets previously used in the SELEX process. Certain terms used to describe the invention herein are defined as follows: "SELEX"methodology refers to the combination of selection of nucleic acid ligands which interact with a target in a desirable manner, for example binding to a protein, with amplification of those selected nucleic acids as described in detail above and in the SELEX Patent Applications. Iterative cycling of the selection/amplification steps allows selection of one or a small number of nucleic acids which interact most strongly with the target from a

pool which contains a very large number of nucleic acids. Cycling of the selection/amplification procedure is continued until a selected goal is achieved.

"Tissue SELEX"methodology applies the SELEX methodology to tissue targets.

Tissue SELEX has several advantages. First, using Tissue SELEX one can obtain ligands to specific cell types in the absence of a defined understanding of the involved epitope. The epitope against which a ligand is evolved is usually a substructural component of a larger macromolecule The ligands found by this method could also be useful in identifying new proteins or other new macromolecules on the tissue target. The new proteins or other new macromolecules which comprise a newly identified epitope can be purified and characterized using standard procedures. Second, ligands can be obtained to defined epitopes or macromolecules in the context of their physiologic cellular or membrane environment. Third, it is possible to obtain ligands to tissues in a functionally altered phenotype, e. g., activated, migrating, etc. The ligands and the new macromolecules containing the ligand epitopes identified by this process may be useful as diagnostics or therapeutics.

Tissue SELEX is a powerful methodology which allows one to identify nucleic acid ligands that can mediate many different cell behaviors, such as apoptosis, anergy, differentiation, proliferation, etc., without prior knowledge of the identity of the specific tissue targets that control these changes. The sensitivity of the SELEX process may lead to the generation of oligonucleotides that recognize potentially every different epitope on the complex tissue target. Larger numbers of different sequence motifs are expected using the tissue SELEX process, as compared with simple-target SELEX, since it is believed that different motifs will recognize distinct epitopes on the complex tissue target. Some epitopes may lie within the same protein, but many will be directed to various proteins or other molecules on the tissue. Tissue SELEX can be done in vivo or in vitro.

In one embodiment, a negative selection process (termed counter-SELEX) is employed to enhance the possibility that the ligands derived by tissue SELEX have precise specificity and affinity. In this embodiment, ligands are selected for a specific tissue and then a negative selection is done against a related tissue which does not have certain characteristics for which the ligand is desired. The negative selection can be done against a similar cell line or cell type, different cells, normal tissue, plasma or blood, a non-specific antibody or other available ligand. An example of this negative selection would be to first select using a tumor cell target (such as a malignant melanoma) and then counterselect the resulting nucleic acids against a similar cell type which is not tumorogenic (such as normal human melanocytes).

Ligands that interact with both normal and neoplastic tissue will be removed by this negative selection and only those nucleic acid ligands that specifically bind the tumor cells will be identified (or retained). The resulting nucleic acid ligand would be specific for tumors. This technique will provide the ability to identify nucleic acid ligands that can discriminate between two closely related targets, i. e., between a cancerous cell and an untransformed cell of the same tissue type. The negative selection can also be done in vivo. Using this method one can not only generate ligands to specific targets on complex tissue surfaces, but also be able to recognize the differences between normal and abnormal tissue of a particular type.

"SELEX Target"or"Target"refers to any compound upon which a nucleic acid can act in a predetermined desirable manner. A SELEX target molecule can be a protein, peptide, nucleic acid, carbohydrate, lipid, polysaccharide, glycoprotein, hormone, receptor, antigen, antibody, virus, pathogen, toxic substance, substrate, metabolite, transition state analog, cofactor, inhibitor, drug, dye, nutrient, growth factor, cell, tissue, etc., without limitation.

Virtually any chemical or biological effector would be a suitable SELEX target. Molecules of any size can serve as SELEX targets. A target can also be modified in certain ways to enhance the likelihood of an interaction between the target and the nucleic acid.

"Tissue target"or"Tissue"refers to a certain subset of the SELEX targets described above. According to this definition, tissues are macromolecules in a heterogeneous environment. As used herein, tissue refers to a single cell type, a collection of cell types, an aggregate of cells, or an aggregate of macromolecules. This differs from simpler SELEX targets which are typically isolated soluble molecules, such as proteins. In the preferred embodiment, tissues are insoluble macromolecules which are orders of magnitude larger than simpler SELEX targets. Tissues are complex targets made up of numerous macromolecules, each macromolecule having numerous potential epitopes. The different macromolecules which comprise the numerous epitopes can be proteins, lipids, carbohydrates, etc., or combinations thereof. Tissues are generally a physical array of macromolecules that can be either fluid or rigid, both in terms of structure and composition. Extracellular matrix is an example of a more rigid tissue, both structurally and compositionally, while a membrane bilayer is more fluid in structure and composition. Tissues are generally not soluble and remain in solid phase, and thus partitioning can be accomplished relatively easily. Tissue includes, but is not limited to, an aggregate of cells usually of a particular kind together with their intercellular substance that form one of the structural materials commonly used to denote the general cellular fabric of a given organ, e. g., kidney tissue, brain tissue. The four

general classes of tissues are epithelial tissue, connective tissue, nerve tissue, and muscle tissue.

Examples of tissues which fall within this definition include, but are not limited to, heterogeneous aggregates of macromolecules such as fibrin clots which are acellular; homogeneous or heterogeneous aggregates of cells; higher ordered structures containing cells which have a specific function, such as organs, tumors, lymph nodes, arteries, etc.; and individual cells. Tissues or cells can be in their natural environment, isolated, or in tissue culture. The tissue can be intact or modified. The modification can include numerous changes such as transformation, transfection, activation, and substructure isolation, e. g., cell membranes, cell nuclei, cell organelles, etc.

Sources of the tissue, cell or subcellular structures can be obtained from prokaryotes as well as eukaryotes. This includes human, animal, plant, bacterial, fungal and viral structures.

"Blood vessel"is understood to be any of the vessels through which blood circulates in the body, including the heart, aorta, arteries, veins and capillaries.

"Nucleic acid"means either DNA, RNA, single-stranded or double-stranded and any chemical modifications thereof. Modifications include, but are not limited to, those which provide other chemical groups that incorporate additional charge, polarizability, hydrogen bonding, electrostatic interaction, and fluxionality to the individual nucleic acid bases or to the nucleic acid as a whole. Such modifications include, but are not limited to, modified bases such as 2'-position sugar modifications, 5-position pyrimidine modifications, 8-position purine modifications, modifications at cytosine exocyclic amines, substitution of 5-bromo- uracil; backbone modifications, methylations, unusual base-pairing combinations such as the isobases isocytidine and isoguanidine and the like. Modifications can also include 3'and 5' modifications such as capping. Modifications that occur after each round of amplification are also compatible with this invention. Post-amplification modifications can be reversibly or irreversibly added after each round of amplification. Virtually any modification of the nucleic acid is contemplated by this invention.

"Nucleic acid test mixture"or"nucleic acid candidate mixture"is a mixture of nucleic acids of differing, randomized sequence. The source of a"nucleic acid test mixture"can be from naturally-occurring nucleic acids or fragments thereof, chemically synthesized nucleic acids, enzymatically synthesized nucleic acids or nucleic acids made by a combination of the foregoing techniques. In a preferred embodiment, each nucleic acid has fixed sequences

surrounding a randomized region to facilitate the amplification process. The length of the randomized section of the nucleic acid is generally between 8 and 250 nucleotides, preferably between 8 and 60 nucleotides.

"Nucleic acid ligand"is a nucleic acid which has been isolated from the nucleic acid candidate mixture that acts on a target in a desirable manner. Examples of actions on a target in a desirable manner include, but are not limited to binding of the target, catalytically changing the target, reacting with the target in a way which modifies/alters the target or the functional activity of the target, covalently attaching to the target as in a suicide inhibitor, facilitating the reaction between the target and another molecule. In most, but not all, instances this desirable manner is binding to the target. In the most preferred embodiment, a nucleic acid ligand is a non-natuially occurring nucleic acid ligand having a specific binding affinity for a tissue target molecule, such target molecule being a three dimensional chemical structure other than a polynucleotide that binds to said nucleic acid ligand through a mechanism which predominantly depends on Watson/Crick base pairing or triple helix binding, wherein said nucleic acid ligand is not a nucleic acid having the known physiological function of being bound by the target molecule.

Nucleic acid ligand includes nucleic acid sequences that are substantially homologous to the nucleic acid ligands actually isolated by the Tissue SELEX procedures. By substantially homologous, it is meant a degree of primary sequence homology in excess of 70%, most preferably in excess of 80%, and even more preferably in excess of 90%, 95%, or 99%. The percentage of homology as described herein is calculated as the percentage of nucleotides found in the smaller of the two sequences which align with identical nucleotide residues in the sequence being compared when 1 gap in a length of 10 nucleotides may be introduced to assist in that alignment. In the past it has been shown that the sequence homologies of various nucleic acid ligands to a specific target shows that sequences with little or no primary homology may have substantially the same ability to bind the target. For these reasons, this invention also includes nucleic acid ligands that have substantially the same ability to bind a target as the nucleic acid ligands identified by the Tissue SELEX process. Substantially the same ability to bind a target means that the affinity is within a few orders of magnitude of the affinity of the ligands described herein. It is well within the skill of those of ordinary skill in the art to determine whether a given sequence--substantially homologous to those specifically described herein--has substantially the same ability to bind a tissue target.

The invention also includes nucleic acid ligands that have substantially the same

postulated structure or structural motifs. Substantially the same structure or structural motifs can be postulated by sequence alignment using the Zukerfold program (see Zucker (1989) Science 244: 48-52). As would be known in the art, other computer programs can be used for predicting secondary structure and structural motifs. Substantially the same structure or structural motif of nucleic acid ligands in solution or as a bound structure can also be postulated using NMR or other techniques as would be known in the art.

The term"clone"is used interchangeably with the term"nucleic acid ligand." "Partitioning"means any process for separating nucleic acid ligands from the remainder of the unreacted nucleic acid candidate mixture. Partitioning can be accomplished by various methods known in the art. Filter binding, affinity chromatography, liquid-liquid partitioning, filtration, gel shift, density gradient centrifugation are all examples of suitable partitioning methods. Equilibrium partitioning methods can also be used as described in detail below. Since the tissue targets of the present invention are non-soluble, there are numerous simple partitioning methods which are well suited to this invention. The simple partitioning methods include any method for separating a solid from a liquid, such as, centrifugation with and without oils, membrane separations and simply washing the insoluble tissue target. The ligands can also be specifically eluted from the target with a specific antibody or ligand. The choice of partitioning method will depend on properties of the target and the nucleic acid and can be made according to principles and properties known to those of ordinary skill in the art.

"Amplifying"means any process or combination of process steps that increases the amount or number of copies of a molecule or class of molecules. In preferred embodiments, amplification occurs after members of the test mixture have been partitioned, and it is the facilitating nucleic acid associated with a desirable product that is amplified. For example, amplifying RNA molecules can be carried out by a sequence of three reactions: making cDNA copies of selected RNAs, using the polymerase chain reaction to increase the copy number of each cDNA, and transcribing the cDNA copies to obtain RNA molecules having the same sequences as the selected RNAs. Any reaction or combination of reactions known in the art can be used as appropriate, including direct DNA replication, direct RNA amplification and the like, as will be recognized by those skilled in the art. The amplification method should result in the proportions of the amplified mixture being essentially representative of the proportions of different sequences in the mixture prior to amplification.

It is known that many modifications to nucleic acids are compatible with enzymatic

amplification. Modifications that are not compatible with amplification can be made after each round of amplification, if necessary.

"Randomized"is a term used to describe a segment of a nucleic acid having, in principle, any possible sequence over a given length. Randomized sequences will be of various lengths, as desired, ranging from about eight to more than one hundred nucleotides.

The chemical or enzymatic reactions by which random sequence segments are made may not yield mathematically random sequences due to unknown biases or nucleotide preferences that may exist. The term"randomized"is used instead of"random"to reflect the possibility of such deviations from non-ideality. In the techniques presently known, for example sequential chemical synthesis, large deviations are not known to occur. For short segments of 20 nucleotides or less, any minor bias that might exist would have negligible consequences. The longer the sequences of a single synthesis, the greater the effect of any bias.

A bias may be deliberately introduced into a randomized sequence, for example, by altering the molar ratios of precursor nucleoside (or deoxynucleoside) triphosphates in the synthesis reaction or the ratio of phosphoramidites in the chemical synthesis. A deliberate bias may be desired, for example, to affect secondary structure, to introduce bias toward molecules known to have facilitating activity, to introduce certain structural characteristics, or based on preliminary results.

In its most basic form, the SELEX process may be defined by the following series of steps: 1) A candidate mixture of nucleic acids of differing sequence is prepared. The candidate mixture generally includes regions of fixed sequences (i. e., each of the members of the candidate mixture contains the same sequences in the same location) and regions of randomized sequences. The fixed sequence regions are selected either: (a) to assist in the amplification steps described below, (b) to mimic a sequence known to bind to the target, or (c) to enhance the concentration of a given structural arrangement of the nucleic acids in the candidate mixture. The randomized sequences can be totally randomized (i. e., the probability of finding a base at any position being one in four) or only partially randomized (e. g., the probability of finding a base at any location can be selected at any level between 0 and 100 percent).

2) The candidate mixture is contacted with the selected target under conditions favorable for binding between the target and members of the candidate mixture. Under these circumstances, the interaction between the target and the nucleic acids of the candidate

mixture can be considered as forming nucleic acid-target pairs between the target and those nucleic acids having the strongest affinity for the target.

3) The nucleic acids with the highest affinity for the target are partitioned from those nucleic acids with lesser affinity to the target. Because only an extremely small number of sequences (and possibly only one molecule of nucleic acid) corresponding to the highest affinity nucleic acids exist in the candidate mixture, it is generally desirable to set the partitioning criteria so that a significant amount of the nucleic acids in the candidate mixture (approximately 5-50%) are retained during partitioning.

4) Those nucleic acids selected during partitioning as having the relatively higher affinity to the target are then amplified to create a new candidate mixture that is enriched in nucleic acids having a relatively higher affinity for the target.

5) By repeating the partitioning and amplifying steps above, the newly formed candidate mixture contains fewer and fewer unique sequences, and the average degree of affinity of the nucleic acids to the target will generally increase. Taken to its extreme, the SELEX process will yield a candidate mixture containing one or a small number of unique nucleic acids representing those nucleic acids from the original candidate mixture having the highest affinity to the target molecule.

The SELEX Patent Applications describe and elaborate on this process in great detail.

Included are targets that can be used in the process; methods for partitioning nucleic acids within a candidate mixture; and methods for amplifying partitioned nucleic acids to generate an enriched candidate mixture. The SELEX Patent Applications also describe ligands obtained to a number of target species, including both protein targets where the protein is and is not a nucleic acid binding protein.

The SELEX method further encompasses combining selected nucleic acid ligands with lipophilic compounds or non-immunogenic, high molecular weight compounds in a diagnostic or therapeutic complex as described in United States Patent Application Serial No. 08/434,465, filed May 4,1995, entitled"Nucleic Acid Ligand Complexes."VEGF Nucleic Acid Ligands that are associated with a Lipophilic Compound, such as diacyl glycerol or dialkyl glycerol, in a diagnostic or therapeutic complex are described in United States Patent Application Serial No.

08/739,109, filed October 25,1996, entitled"Vascular Endothelial Growth Factor (VEGF) Nucleic Acid Ligand Complexes."VEGF Nucleic Acid Ligands that are associated with a Lipophilic Compound, such as a glycerol lipid, or a Non-Immunogenic, High Molecular Weight Compound, such as polyalkylene glycol, are further described in United States Patent

Application Serial No. 08/897,351, filed July 21,1997, entitled"Vascular Endothelial Growth Factor (VEGF) Nucleic Acid Ligand Complexes."VEGF Nucleic Acid Ligands that are associated with a non-immunogenic, high molecular weight compound or lipophilic compound are also further described in WO 98/18480, filed October 17,1997, entitled"Vascular Endothelial Growth Factor (VEGF) Nucleic Acid Ligand Complexes."Each of the above described patent applications which describe modifications of the basic SELEX procedure are specifically incorporated by reference herein in their entirety.

Certain embodiments of the present invention provide a complex comprising one or more nucleic acid ligands to a Tissue Target covalently linked with a non-immunogenic, high molecular weight compound or lipophilic compound. A complex as used herein describes the molecular entity formed by the covalent linking of the nucleic acid ligand of a Tissue Target to a non-immunogenic, high molecular weight compound. A non-immunogenic, high molecular weight compound is a compound between approximately 100 Da to 1,000,000 Da, more preferably approximately 1000 Da to 500,000 Da, and most preferably approximately 1000 Da to 200,000 Da, that typically does not generate an immunogenic response. For the purposes of this invention, an immunogenic response is one that causes the organism to make antibody proteins. In one preferred embodiment of the invention, the non-immunogenic, high molecular weight compound is a polyalkylene glycol. In the most preferred embodiment, the polyalkylene glycol is polyethylene glycol (PEG). More preferably, the PEG has a molecular weight of about 10-80K. Most preferably, the PEG has a molecular weight of about 20-45K. In certain embodiments of the invention, the non-immunogenic, high molecular weight compound can also be a nucleic acid ligand.

Another embodiment of the invention is directed to complexes comprised of a nucleic acid ligand to a Tissue Target and a lipophilic compound. Lipophilic compounds are compounds that have the propensity to associate with or partition into lipid and/or other materials or phases with low dielectric constants, including structures that are comprised substantially of lipophilic components. Lipophilic compounds include lipids as well as non- lipid containing compounds that have the propensity to associate with lipid (and/or other materials or phases with low dielectric constants). Cholesterol, phospholipid, and glycerolipids, such as dialkylglycerol, diacylglycerol, and glycerol amide lipids are further examples of lipophilic compounds. In a preferred embodiment, the lipophilic compound is a glycerolipid.

The non-immunogenic, high molecular weight compound or lipophilic compound may be covalently bound to a variety of positions on the nucleic acid ligand to a Tissue Target, such

as to an exocyclic amino group on the base, the 5-position of a pyrimidine nucleotide, the 8- position of a purine nucleotide, the hydroxyl group of the phosphate, or a hydroxyl group or other group at the 5'or 3'terminus of the nucleic acid ligand to a Tissue Target. In embodiments where the lipophilic compound is a glycerolipid, or the non-immunogenic, high molecular weight compound is polyalkylene glycol or polyethylene glycol, preferably it is bonded to the 5'or 3'hydroxyl of the phosphate group thereof. In the most preferred embodiment, the lipophilic compound or non-immunogenic, high molecular weight compound is bonded to the 5'hydroxyl of the phosphate group of the nucleic acid ligand. Attachment of the non-immunogenic, high molecular weight compound or lipophilic compound to the nucleic acid ligand of the Tissue Target can be done directly or with the utilization of linkers or spacers.

A linker is a molecular entity that connects two or more molecular entities through covalent bonds or non-covalent interactions, and can allow spatial separation of the molecular entities in a manner that preserves the functional properties of one or more of the molecular entities. A linker can also be known as a spacer.

The complex comprising a nucleic acid ligand to a Tissue Target and a non- immunogenic, high molecular weight compound or lipophilic compound can be further associated with a lipid construct. Lipid constructs are structures containing lipids, phospholipids, or derivatives thereof comprising a variety of different structural arrangements which lipids are known to adopt in aqueous suspension. These structures include, but are not limited to, lipid bilayer vesicles, micelles, liposomes, emulsions, lipid ribbons or sheets, and may be complexed with a variety of drugs and components which are known to be pharmaceutically acceptable. In the preferred embodiment, the lipid construct is a liposome.

The preferred liposome is unilamellar and has a relative size less than 200 nm. Common additional components in lipid constructs include cholesterol and alpha-tocopherol, among others. The lipid constructs may be used alone or in any combination which one skilled in the art would appreciate to provide the characteristics desired for a particular application. In addition, the technical aspects of lipid constructs and liposome formation are well known in the art and any of the methods commonly practiced in the field may be used for the present invention.

SELEX provides high affinity ligands of a target molecule. This represents a singular achievement that is unprecedented in the field of nucleic acids research. The present invention applies the SELEX procedure to more complicated tissue targets.

Negative selection (Counter-SELEX) is optionally employed before, during or after

the Tissue SELEX process. The negative selection provides the ability to discriminate between closely related but different tissue types. For example, negative selection can be introduced to identify nucleic acid ligands that have a high specificity for a tumor cell but do not recognize the cognate normal tissue. Similarly, nucleic acid ligands can be identified which specifically recognize atherosclerotic arterial tissue but not normal arterial tissue.

Nucleic acid ligands which recognize fibrin, but not fibrinogen can also be identified by this method. Additionally, nucleic acid ligands to a cell type which express a certain receptor can be counter-selected with a cell line engineered not to express the receptor (or other such macromolecule).

One of ordinary skill in the art will readily understand that various mechanisms can be employed to accomplish this negative selection. The following examples are provided mostly for illustrative purposes and are not meant in any way as limiting the procedures of negative selection. Negative selection or Counter-SELEX methods were first described in United States Patent Application Serial No. 08/134,028, filed October 7,1993, abandoned in favor of United States Patent Application Serial No. 08/443,957, entitled"High-Affinity Nucleic Acid Ligands that Discriminate Between Theophylline and Caffeine,"now United States Patent No. 5,580,737, which is herein incorporated by reference. A particular implementation of negative selection is embodied using equilibrium partitioning. In this method, two cell lines or other tissue types are separated by a semi-permeable membrane pm pore size) in an equilibrium dialysis chamber; one cell line is the neoplastic target cell line, the other, the normal tissue used for the negative selection. The choice of cell or tissue type for the negative selection will be determined by the specific end results desired and will sometimes consist of a non-malignant cell line of the same tissue type as the neoplastic target. For other experiments, various normal cell types could be combined to create the negative epitope "sink". The random pool of nucleic acids is placed into the dialysis chamber (on the side of the normal cells; this avoids background from high avidity targets which are common to both the tumor and normal cells) and allowed to equilibrate between the two cell lines. Those nucleic acid sequences that remain bound to the target cell line or tissue at equilibrium are selectively recovered and amplified for the next round of SELEX.

This example of negative selection methodology is quite powerful. First, equilibrium dialysis negative selection allows the positive and negative selection to be carried out simultaneously. Second, the stringency of the negative selection can be varied through the alteration of the relative amounts of"positive"and"negative"cells placed on each side of the

dialysis membrane. These two characteristics of equilibrium dialysis negative selection allow precise control over the evolution of nucleic acid ligands specific for the target cell or tissue type.

This same type of equilibrium partitioning negative selection can be carried out with adherent cell lines. In this embodiment, monolayers of target and negative cells or tissues are plated in different wells of a multi-welled plate. After adherence, media, along with an oligonucleotide pool, is added such that the wells are connected by the volume of cell media.

After equilibration of the oligonucleotide pool, those sequences bound by the target cell line or tissue type would be isolated and amplified for the next round of SELEX.

The equilibrium negative selection strategies above offer a powerful way of generating nucleic acid ligands to tissue targets and especially tumor associated antigens (TAAs).

Additionally, there are several other negative selection methods, which could be classified as"post-SELEX screening procedures."The most simple of these procedures is the testing of individual nucleic acid ligands (those sequences generated by tissue SELEX and demonstrated to be high-affinity ligands for the tissue target) against normal tissue for cross- reactivity. However, this approach is a tedious and time-consuming process.

A more fruitful"post-SELEX"method is to perform a negative selection, for example using a normal tissue as the negative selection target, on a pool that has already been evolved from a SELEX against a desirable complex tissue target, for example a transformed cell line.

This example would suggest the performance of two to three negative selections on a normal tissue using a late-round, highly evolved pool from a SELEX of a transformed cell line. The binding of certain sequences to the normal tissue would be used to subtract these sequences from the evolved pool. This method allows one to quickly eliminate from several hundred to several thousand nucleic acid sequences that show a high affinity for those targets common to both the normal and the transformed cell lines.

Another"post-SELEX"screening method is a variation of a photocrosslinking experiment. As an example, it is possible to synthetically incorporate a highly photoreactive nitrine group (which is also iodinatable) on the 5'end of a PCR primer used in the tissue SELEX protocol. Late-round pools from, for example, a tumor cell line SELEX would be amplified with this photoactivatable (and'25I-labeled) primer, and this sequence pool would then be irradiated in the presence of the tumor cell line, and in the presence of normal tissue.

Membrane proteins would be isolated and solubilized for analysis on an SDS gel. One would

expect to see many different protein epitopes tagged by specific oligonucleotide sequences, for both the tumor and the normal cell lines. A few tagged targets will be unique to the tumor cell line. Because the oligonucleotides have been photochemically linked to the protein targets in a manner which does not destroy the base sequence of the oligonucleotide, it is possible to isolate a tumor-specific band from an SDS gel, and use PCR to recover a specific sequence motif that recognizes a particular tumor antigen. Thus, in one step, it will be possible to remove from a pool oligonucleotide sequences that recognize possibly hundreds of cell surface antigens, leaving one or a few families of sequences that binds specifically to a single tumor-specific antigen.

As described above, the Tissue SELEX methods can include the identification of macromolecules which comprise new epitopes on the tissue target. The nucleic acid ligand to the new epitope component of the macromolecule can be employed to purify, identify and characterize the macromolecule. The new macromolecule can be a previously unknown protein or peptide, lipid, carbohydrate, etc. Virtually any molecule that is part of the molecular make-up of a tissue can be identified by the Tissue SELEX process.

In order to fully exploit this aspect of the invention, it is important to develop strategies for the purification and identification of new macromolecules which comprise the new epitopes and to determine the roles these new macromolecular components of the tissue play in biological systems. The methods for purifying new macromolecules are well-known, especially in the art of protein purification. These standard purification methods include crosslinking, affinity chromatography, peptide microsequencing. Edman sequencing, mass spectrometry, and cDNA library searches.

The following discussion describes this process as it would be applied to the identification of a new tumor-associated antigen (TAA). For the purposes of this discussion, a TAA is a macromolecule that is expressed on a tumor cell, but not on a similar normal cell.

A TAA may or may not be immunogenic. A TAA is merely one example of the kind of macromolecules which can be identified by the Tissue SELEX process and simply used for illustrative purposes. However, it is readily apparent that this process can be extrapolated to any new macromolecule identified by the Tissue SELEX process.

As applied to TAAs, the identification of new TAAs by the Tissue SELEX process is composed of two main parts: one, developing strategies for the purification and identification of new TAAs, and two, the elucidation of the role these tumor antigens play in cancer (i. e., determining the biological significance of each particular TAA in the development and

progression of a particular cancer).

The steps of purification and identification of most of the TAAs should be straightforward and understood by one skilled in the art of protein purification. As with antibodies, SELEX provides a agent-a high-affinity ligand specific for the tumor antigen- -that is incredibly useful for the purification of the antigen from whole cells or other tissues.

As a non-limiting example, most antigens will be amenable to some type of photo-affinity crosslinking or in the negative selection strategies section above. Specific crosslinking of the TAA, using a photoactivatable oligonucleotide with a 3'biotin conjugate will allow one-pass purification of the TAA target using streptavidin coated beads. An alternative method to this purification strategy is to use a column-bound high-affinity nucleic acid ligand to affinity purify the TAA target from solubilized target cell membrane preparations.

There are many compelling reasons to believe that the method provided herein for identifying macromolecules that comprise new epitopes on tissues offers distinct advantages over traditional methods of new macromolecule discovery. Again, the following discussion will be directed to tumor-associated antigen discovery, but one will readily understand that it can be broadly extrapolated to all new macromolecule discovery.

As applied to tumor-associated antigens, one must fully consider that all that is known about tumor antigens has been derived from the immune system's reaction to particular antigens; science has depended on the particular restrictions of the immune system, and the system's repertoires to distinguish antigenic differences between neoplastic and normal tissue.

It is entirely possible that other tumor antigens exist that are not subject to immune response.

Some investigators have hypothesized that there may in fact be many antigenic differences between cancer and normal tissue which are, unfortunately, not immunogenic.

The SELEX methodology provides an improved way to identify TAAs that avoids the restrictions posed by the immune system: a. SELEX can actually provide a deeper search of TAAs than can the entire potential antibody repertoire of an organism--the size of the nucleic acid libraries used in SELEX is unrivaled by any biological system; b. SELEX provides nucleic acid ligands to targets, including those which are not antigenic to the immune system because of tolerance. Many of the TAAs which have been identified are oncofetal--they are antigens expressed at some point during development or cell differentiation. As prior"self'antigens, they elicit no overt immune response because of

earlier immune system tolerization. A SELEX-based search for TAAs avoids the circular nature of using the immune system as a means of identifying tumor antigens; c. SELEX nucleic acid ligands have been shown to be exquisitely sensitive to target conformation. While most antibodies recognize conformational, or discontinuous epitopes, antibody functional epitopes are composed of only a few amino acids. The potential binding surface of an oligonucleotide ligand is much larger than that of an antibody variable region, and may provide greater conformational discrimination of large targets. Additionally, cross- reactivity for SELEX ligands is substantially less of a problem than for monoclonal antibodies. A considerable set of restrictions also controls T-cell mediated tumor responses.

These immune system limitations provide important biological functions; however, they limit the immune system's power for TAA identification. d. SELEX is possibly more sensitive to small quantities of antigen than the immune system. Although the immune system's threshold for reactivity has been estimated to be 200 copies/cell for an antigenic MHC-presented peptide, a B-cell antibody response (necessary for any antigen that is not a peptide-carbohydrates, lipids or conformational antigens) to a monovalent target requires antigen concentrations of about 100 mM. SELEX can generate ligands to TAA targets with a low representation on the cell surface; and e. SELEX provides a rapid and thorough method of TAA discovery. Screening of monoclonal antibodies to tissue sections and purification and identification of MHC peptides are painstaking processes that set practical limits on the depth and completeness of searches for TAAs. Tissue SELEX experiments take a much abbreviated length of time.

Nucleic acid ligands to tissue targets or the tissue epitopes identified by the method of the invention are useful as diagnostic reagents and as pharmaceuticals. The nucleic acid ligands are also useful for the identification of new macromolecules. The nucleic acid ligands are useful in any application that would be suitable for use of an antibody.

One problem encountered in the therapeutic use of nucleic acids is that oligonucleotides in their phosphodiester form may be quickly degraded in body fluids by intracellular and extracellular enzymes such as endonucleases and exonucleases before the desired effect is manifest. Certain chemical modifications of the nucleic acid ligand can be made to increase the in vivo stability of the nucleic acid ligand or to enhance or to mediate the delivery of the nucleic acid ligand. Modifications of the nucleic acid ligands contemplated in this invention include, but are not limited to, those which provide other chemical groups that incorporate additional charge, polarizability, hydrophobicity, hydrogen bonding, electrostatic interaction, and

fluxionality to the nucleic acid ligand bases or to the nucleic acid ligand as a whole. Such modifications include, but are not limited to, 2'-position sugar modifications, 5-position pyrimidine modifications, 8-position purine modifications, modifications at exocyclic amines, substitution of 4-thiouridine, substitution of 5-bromo or 5-iodo-uracil; backbone modifications, phosphorothioate or alkyl phosphate modifications, methylations, unusual base-pairing combinations such as the isobases isocytidine and isoguanidine and the like. Modifications can also include 3'and 5'modifications such as capping.

Where the nucleic acid ligands are derived by the SELEX method, the modifications can be pre-or post-SELEX modifications. Pre-SELEX modifications yield nucleic acid ligands with both specificity for it Tissue Target and improved in vivo stability. Post-SELEX modifications made to 2'-OH nucleic acid ligands can result in improved in vivo stability without adversely affecting the binding capacity of the nucleic acid ligands. The preferred modifications of the tissue nucleic acid ligands of the subject invention are 5'and 3' phosphorothioate capping and/or 3'3'inverted phosphodiester linkage at the 3'end. In one preferred embodiment, the preferred modification of the tissue nucleic acid ligand is 3'3' inverted phosphodiester linkage at the 3'end and 2'fluoro (2'-F) and/or, 2'amino (2'-NH2), and/or 2'O methyl (2'-OMe) modification of some or all of the nucleotides.

As diagnostic reagents, the ligands or tissue epitopes can be used in both in vitro diagnostics and in vivo imaging applications. The SELEX method generally, and the specific adaptations of the SELEX method taught and claimed herein specifically, are particularly suited for diagnostic applications. SELEX identifies nucleic acid ligands that are able to bind targets with high affinity and with surprising specificity. These characteristics are, of course, the desired properties one skilled in the art would seek for a diagnostic ligand. Details regarding use of the ligands in diagnostic applications is well known to one of ordinary skill in the art. Nucleic acid ligands that bind specifically to pathological tissues such as tumors may have a role in imaging pathological conditions such as human tumor imaging and even therapeutic delivery of cytotoxic compounds or immune enhancing substances.

The nucleic acid ligands of the present invention may be routinely adapted for diagnostic purposes according to any number of techniques employed by those skilled in the art. Diagnostic agents need only be able to allow the user to identify the presence of a given target at a particular locale or concentration. Simply the ability to form binding pairs with the target may be sufficient to trigger a positive signal for diagnostic purposes. Those skilled in the art would also be able to adapt any nucleic acid ligand by procedures known in the art to

incorporate a labelling tag in order to track the presence of a ligand. Such a tag could be used in a number of diagnostic procedures. For example, in the field of nuclear medicine, radiodiagnostic agents are useful for imaging sites in the body. United States Patent Application Serial Nos. 08/358,065, filed December 15,1994, and 08/488,290, filed June 7, 1995, both entitled"Conjugates Made of Metal Complexes and Oligonucleotides, Agents Containing the conjugates, their use in Radiodiagnosis as well as Process for Their Production,"which are incorporated herein by reference, describe diagnostic agents comprising oligonucleotides covalently bound to complexing agents which further contain an imaging radioactive isotope such as 99m-Tc.

Specifically, oligonucleotide ligands with high specificity for particular tumor antigens could become as important as monoclonal antibodies for the detection, imaging, and surveillance of cancer. Modified nucleic acid ligands show nuclease resistance in plasma. and the use of 5'and 3'capping structures will provide stability in animals that rivals that of monoclonal antibodies (and without the immunogenicity of animal-derived MAbs).

Radionuclides, magnetic compounds, and the like can be conjugated to tumor-specific oligonucleotides for cancer imaging. SELEX tumor ligands can also be used to determine if these tumor antigens are sloughed off tumors, and are detectable in the plasma like PSA.

The nucleic acid ligands to tissue targets or newly identified macromolecules components of tissue are also useful as pharmaceuticals. Therapeutic uses include the treatment or prevention of diseases or medical conditions in human patients. Therapeutic uses also include veterinary applications. The ligands can bind to receptors and be useful as receptor antagonists. Conversely, under certain circumstances the ligands can bind to receptors and cause receptor capping and act as receptor agonists.

In order to produce nucleic acids desirable for use as a pharmaceutical, it is preferred that the nucleic acid ligand (1) binds to the target in a manner capable of achieving the desired effect on the target; (2) be as small as possible to obtain the desired effect; (3) be as stable as possible; and (4) be a specific ligand to the chosen target. In most situations, it is preferred that the nucleic acid ligand have the highest possible affinity to the target.

Standard formulations can be used for the nucleic acid ligands of the invention and are known to one of ordinary skill in the art.

The following examples provide a non-limiting description of the present invention.

Example One describes obtaining ssDNA ligands to the complex tissue target peripheral blood mononuclear cells (PBMC). Ligands to PBMC have many uses including imaging

lymph nodes for cancer screening and flow cytometry uses for AIDS monitoring.

Example Two describes the ability to obtain RNA ligands to human fibrin clots. The pyrimidine residues of these RNA ligands have been modified with fluorine at the 2'-position of the sugar. The fibrin ligands are useful as diagnostic agents as described below.

Circulating fibrinogen is converted to insoluble fibrin by the actions of the common product of the intrinsic and extrinsic coagulation cascade, thrombin. Fibrin provides a fibrous network for the clot allowing platelet deposition and later fibroblast invasion. Fibrin is present in large amounts in all thrombi, relatively less in platelet-rich arterial clots than fibrin- rich venous clots. Fibrin also can provide the nidus for atherosclerotic plaques and restenotic lesions by harboring thrombin and other mitogens which can lead to endothelial activation and smooth muscle cell proliferation.

The noninvasive detection and localization of thrombi remains a major challenge in clinical diagnosis. Deep vein thrombosis (DVT) and pulmonary embolism (PE) carry with them a high rate of mortality and morbidity. Deep-vein thrombosis (DVT) is a major complication of hospitalization and is diagnosed by physical exam less than one third of the time. Patients at risk include those with a major medical illness, malignancy, undergoing general abdominal, thoracic surgery or major orthopaedic surgery. High risk patients carry a 40-80% risk of DVT with a 1-2% risk of fatal pulmonary embolism (PE) (Weinmann and Salzman (1994) New Engl. J. Med. 331: 1630-1641). PE accounts for 50,000 deaths/yr. 90% of PEs are non-fatal but carry significant morbidity: dyspnea, pulmonary infarction, abscess, or hypertension. 95% of PEs arise as a complication of DVT. Diagnosis of these conditions is difficult and has not improved, as noted by the high rate of undiagnosed PE on autopsy. which has not improved over time. Freiman et al. found evidence of subclinical PE in 64% <BR> <BR> <BR> <BR> <BR> of consecutive autopsies among persons with various causes of death (Freiman et al. (1965) N. Engl. J. Med. 272: 1278-1280). Arterial thrombus, mostly secondary to atheromatosis. is even more difficult to diagnose non-invasively.

Non-invasive imaging of venous clots has relied on ultrasonic visualization of the deep venous system of the lower extremities. These studies are limited (generally only the thigh region) and are extremely operator dependent. PE diagnosis is generally done by ventilation and perfusion scanning using radioisotopes with the gold-standard being invasive pulmonary angiography. Radiolabeled fibrinogen has been used historically (Lensing and Hirsch (1993) Thromb. Haemost. (Germany) 69: 2-7). It requires either prospective administration or thrombus extension after it becomes clinically apparent. A number of

reports of radiolabeled antibodies to either fibrin or platelets have been reported. These are sensitive but slow, with adequate images appearing 12-48 hours after injection of the tracer.

The need for delayed images is due to clearance of the unbound antibody from the vasculature to allow for adequate signal-to-noise ratio. No significant imaging of coronary artery disease has been reported. The conjecture is that the thickness of the blood pool in the left ventricle of the heart significantly obscures the signal from the small overlying epicardial coronary arteries. Arterial imaging has been performed on the larger vessels of the aorta or femoral arteries using either anti-fibrin or anti-platelet antibodies. Both antibodies have problems: the antifibrin Abs bind to epitopes that are poorly accessible and which are constantly changing through clot stabilization and fibrinolysis; the anti-platelet Abs bind to epitopes which exist in circulating blood, thereby increasing their background. Meaningful high resolution detection of disease in small arteries will require high specificity, rapid clearance of unbound material, and probably 3-dimensional tomographic imaging technologies. In many respects, RNA ligands are suitable agents for these diagnostic approaches. A superior non-invasive diagnostic test for pulmonary embolism would be particularly clinically relevant.

Example Three describes the ability to obtain RNA ligands to rat stenotic carotid arteries. The stenotic carotid arteries ligands are useful as diagnostic and pharmaceutical agents as described below.

Atherosclerosis is one of the major causes of mortality in the world. There has been much effort in identifying and targeting both therapeutics and diagnostic agents to this pathological tissue. Experimentally atherosclerosis suffers from the absence of ideal animal models. Rodent vessels are significantly different from the primate especially with respect to the neointima. Primate models are expensive. The pig or'minipig'provides a model for restenosis but does not provide a good model of primary atherosclerosis. Recently, transgenic mouse models have become available, but they are still poorly defined.

Although mechanisms and components of atherosclerosis are not completely defined, most investigators would agree that smooth muscle cells play an important role. The consensus is that these SMCs proliferate within the intima and are in some form"activated." The rat balloon-injured carotid artery model is one of the best understood models of response to arterial damage. Although there are limits to this model, there is clear evidence that in response to endothelial damage a proliferative response occurs primarily involving the SMCs.

Many unique proteins have been identified from this tissue as well as signals responsible for

SMC activation, migration and proliferation, as well as, extracellular matrix deposition. As such this remains a viable model of restenosis and less directly, primary atherosclerosis.

The rat balloon-injured carotid (RBIC) model provides a unique model for testing the hypothesis that nucleic acid ligands can be evolved by the SELEX methodology which is capable of recognizing pathological tissue to the exclusion of closely related normal tissue.

RBIC are relatively well understood with respect to their composition and structure, are easily and reproducibly produced in a readily available lab animal, and have relevance to human pathologic conditions.

Example 4 describes a second experiment to obtain RNA ligands to rat stenotic carotid arteries involving twelve rounds of ex vivo SELEX followed by twelve rounds of in vivo enrichment, after which a single sequence (clone 12.2; SEQ ID NO: 242) was identified representing 48% of the final enriched library. This sequence accumulated at 1.4 + 0.4% injected dose/gram (% ID/g) in the balloon injured carotid compared to 0.1 + 0.05% ID/g in the contralateral control artery one hour after administration. Blood level at one hour was 0.2 + 0.06% ID/g. Tissue specificity (balloon injured: carotid: normal carotid) was 14, (p<. 05) and the signal to background (balloon injured: carotid: blood) was 7, (p<. 05).

Example 5 describes the ability to obtain RNA ligands to vascular tissue using the Watanabe Heritable Hyperlipidemic (WHHL) Rabbit model of atherosclerosis. The WHHL rabbit plaque tissue ligands are useful as diagnostic and pharmaceutical agents. The WHHL rabbit model provides a novel technique of identifying a nucleic acid ligand which accumulates specifically at a site of induced vascular pathology and for the development of in vivo imaging agents. In this example, a nucleic acid ligand (clone SEQ ID NO: 268) was identified that has a 3.5-10-fold increased affinity for plaqued WHHL vascular tissue over non-plaqued WHHL vascular tissue or vascular tissue.

Example 6 describes the in vitro binding of a nucleic acid ligand to WHHL rabbit atherosclerotic plaque tissue to human atherosclerotic plaque tissue, demonstrating that a WHHL rabbit-derived nucleic acid ligand has cross-reactivity with human atherosclerotic plaque tissue.

Example 7 demonstrates that clone 10.31 (SEQ ID NO: 268), identified in Example 5, accumulates in vascular plaque in rabbits.

Example 8 describes the in vitro binding, in vivo imaging and biodistribution of a 99m-Tc-labelled clone 10.31 identified in Example 5. Phospho-images showed that 15% of the input dose accumulated in the plaqued artery, demonstrating that an in vivo imaging agent

can be developed through this method.

Example 9 describes a comparison of in vivo imaging and biodistribution of a WHHL rabbit-derived nucleic acid ligand in both a WHHL rabbit and a non-atherogenic New Zealand White rabbit. The aortic arch was observed in the WHHL rabbit but could not be visualized in the New Zealand rabbit.

Example 10 describes truncation and post-SELEX modifications of WHHL ligand 10.31 (SEQ ID NO: 268) in which purine positions in the truncate were substituted with 2'- OMe to study the effects of these substitutions on the activity of the truncates.

Example 11 describes a tissue SELEX on human atherosclerotic coronary artery segments.

Example One ssDNA Ligands to Peripheral Blood Mononuclear Cells (PBMC) This example demonstrates the ability to obtain ssDNA ligands to the complex tissue target human peripheral blood mononuclear cells (PBMC). PBMC are isolated from whole blood as described below and contain a complex mixture of cell types including B- lymphocytes, T-lymphocytes and monocytes. Ligands to PBMC have many uses including imaging lymph nodes for cancer screening and flow cytometry uses for AIDS monitoring.

A. MATERIALS and METHODS Isolation of PBMCs Fresh human blood was collected in heparinized vacutainers and up to 35 ml of whole blood was layered atop 10 ml of ficoll (Sigma Histopaque-1077S) in a 50 ml polyethylene conical tube. The samples were centrifuged at 400 x g for 30 minutes at room temperature to separate the blood into three layers: red blood cells (RBCs) below the ficoll, peripheral blood mononuclear cells (PBMCs, including B lymphocytes, T lymphocytes, and monocytes) immediately above the ficoll, and acellular plasma above the PBMCs. Following centrifugation, the plasma was aspirated with a pasteur pipette to within 0.5 cm of the opaque PBMC interface. The PBMC interface, also referred to as the"buffy coat", was transferred to a 15 ml conical tube with a pasteur pipette, 10 ml of phosphate buffered saline solution (PBS, 137 mM NaCI, 2.7 mM KCI, 10.1 mM Na2HPO4,1.8 mM KH2PO4) was added, and the cells were washed by gentle aspiration. The cells were then centrifuged at 250 x g for 10 minutes

at room temperature and the supernatant aspirated and discarded. The cell pellet was resuspended in 5 ml PBS, mixed by gentle aspiration, centrifuged at 250 x g for 10 minutes at room temperature, and the supernatant aspirated and discarded. This washing step was repeated a third time, and the cells were resuspended in a final volume of 0.3 ml PBS, transferred to a 1.7 ml eppendorf tube, and stored on ice. PBMC yield and viability were measured by diluting the cells 1: 50 in PBS, adding an equal volume of 0.4% trypan blue, and counting viable cells with a hemocytometer. Typical yields were 106 cells/ml of whole blood with > 95% viability.

Generation of Degenerate ssDNA Library A library of synthetic DNA oligonucleotides containing 40 random nucleotides flanked by invariant primer annealing sites (oligonucleotide 1,5'-AGGGAGGAC GATGCGG-[N] 40-CAGACGACTCGCCCGA-3')[N] 40-CAGACGACTCGCCCGA-3') (SEQ ID NO: 1) was amplified by the Polymerase Chain Reaction (PCR) for three cycles using oligonucleotides 2 (5'- AGGGAGGACGATGCGG-3') (SEQ ID NO: 2) and 3 (5'- (Biotin) 3- TCGGGCGAGTCGTCTG-3') (SEQ ID NO: 3) as primers. Oligonucleotide 3 had three biotin phosphoramidites conjugated to its 5'terminus. The 72 nucleotide double stranded product was denatured by adding an equal volume of formamide and heating to 95 ° C for 3 minutes, and electrophoresed on an 8% polyacrylamide gel containing 8 M urea. The DNA strand lacking the biotin tag migrates faster than the biotinylated strand, and was isolated by excision from the gel, elution by squashing in 0.4 ml 2 mM EDTA and gentle agitation for 15 minutes, and centrifugation for 5 minutes using a microcentrifuge filter unit (CoStar Spin-X) to partition the ssDNA from the gel slurry. The recovered ssDNA was precipitated with 0.5 M NaCl and 2 volumes of ethanol, pelleted by centrifugation, washed once with 0.4 ml 70% ethanol, dried, and resuspended in deionized, distilled water (ddH, O).

Selection for PBMC Affinitv and Amplification The affinity of the degenerate ssDNA library for PBMCs was determined using a cell- excess nitrocellulose filter binding assay as described in (Carey et al. (1983) Biochemistry 22: 2601-2609. Since the number of possible DNA binding targets on the surface of a PBMC is unknown, affinity values are reported as the concentration of cells (in units of cells/} showing half saturation in this assay. Selections for PBMC affinity were performed under DNA-excess conditions predicted to saturate available target sites, with heparin (Calbiochem, average M. W. 5000) added in excess of DNA to act as a non-amplifiable competitor and to increase stringency. PBMCs, DNA. and heparin were equilibrated for 15 minutes at 37 ° C

and PBMC: DNA complexes were partitioned from free DNA by filtration. PBMC- independent (background) retention of DNA was measured by filtering a similar reaction lacking PBMCs. Filters were pre-wet with 1 ml of wash buffer (50 mM Tris Acetate, pH 7.4) and following application of the sample, washed with 5 ml of wash buffer to remove unbound DNA. For selections 9-21,0.5 M urea was added to the wash buffer to further reduce background retention. To minimize the likelihood of enriching for DNA with an affinity for the filter, we alternated among three different filter types: nitrocellulose (Millipore, Type HA, 0.45 pm), acrylic-coated nylon (Gelman Sciences, Versapor-450,0.45 pm) and glass microfibre (Whatman, GF/C).

For the first selection, 1.4 pM DNA (70 pmoles or about 4 x 10'3 molecules) was equilibrated with 100 pM heparin and PBMCs at final concentrations of 40,000,20,000, 10,000,5,000, and 2,500 cells in 50 pl PBS. The fraction of total DNA complexed to PBMCs and retained by the filters was calculated by measuring Cerenkov radiation in a scintillation counter. A plot of fraction of DNA bound as a function of total DNA gives a linear relationship with a slope equal to the number of DNA molecules bound per cell (an estimate of the number of DNA binding targets per cell). For each subsequent selection, 5-9 PBMC concentrations were tested and plotted in this fashion and the DNA/cell value recorded. In an additional effort to reduce enrichment for filter binders each selection, the filter with the cell concentration retaining between l % and 10% of total DNA, and if possible, at least 10 times more DNA than PBMC-independent (background) retention, was chosen for further amplification and enrichment. The selected DNA was harvested from the filter as described in Tuerk and Gold (1990) Science 249: 505-510, amplified by PCR, and size-purified by electrophoresis on an 8% polyacrylamide, 8 M urea gel as described above.

As enrichment progressed through successive selections, stringency was increased by decreasing the DNA concentration, increasing the heparin concentration, and for selections # 12-21, performing the selections in fresh human plasma instead of PBS. Performing selections in plasma adds an element of specificity, as PBMC-binding DNA molecules with a higher affinity for a plasma component will be depleted from the library. The DNA, PBMC, and heparin concentrations, as well as other relevant selection data, are summarized in Table 1.

Cloning and Sequencing Isolates Following selection #21,2 pmol of the selected library was amplified by PCR using oligonucleotide 4 (5'-CCGAAGCTTAATACGACTCACTATAGGGAGGAC GATGCGG-

3', containing a Hind III restriction endonuclease cleavage site, underlined) (SEQ ID NO: 4) and oligonucleotide 5 (5'-GCCGGATCCTCGGGCGAGTCGTCTG-3', containing a Bam HI site, underlined) (SEQ ID NO: 5) as primers. The double-stranded product was size-purified on an 8% polyacrylamide gel and recovered as described above. Fifteen pmol of the PCR product was digested with Hind III and BamHI, along with 1 pmol pUC19 (all from New England Biolabs) for 3 hours at 37 ° C. Following digestion, the sample was extracted once each with one volume of phenol and chloroform and recovered by precipitation as described above. The selected library was ligated into pUC19 with DNA ligase (New England Biolabs) for 3 hours at 37°C and the ligation product introduced into E. coli DHIa cells by electroporation transformation. Vectors from successful transformations were isolated using a standard plasmid mini-prep protocol and sequenced by dideoxy extension of end-labeled oligonucleotide 6 (5'-TTCACAC AGGAAACAG-3') (SEQ ID NO: 6) with Sequenase T7 DNA Polymerase (United States Biochemical). For a detailed description of these techniques, refer to Schneider et al. (1993) FASEB 7: 201-207. Larger quantities of individual ligands (>20 pmol) were prepared by amplifying the vector inserts by PCR using oligonucleotides 2 and 3 as primers and denaturing and size-purifying the product as described above.

Competition Assay Measuring Disruption of PBMC: DNA Complexes In a 20 pi reaction containing 100 11M heparin in PBS, 10 nM end-labeled DNA was equilibrated with a saturating concentration of PBMCs (10,000 cells/pl) for 10 minutes at 37 ° C. 5 pl of unlabeled competitor DNA was then added to a final concentration ranging from 1.25 nM to 3.2 pM and allowed to equilibrate for 10 minutes at 37°C. Reactions were filtered and the percent of total labeled DNA retained on the filter was recorded.

B. RESULTS Affinitv for PBMCs was Enriched 40-Fold, and is Heparin Dependent Following 21 rounds of enrichment by selection and amplification, the affinity of the DNA library for PBMCs was enriched by a factor of 40. In a cell-excess titration in PBS and 100 pM heparin, the degenerate library (DNA-0) showed half saturation at 43,500 cells/ while the fully enriched library (DNA-21) showed half saturation at 1,000 cells/pl. The difference in affinity between DNA-0 and DNA-21 is heparin dependent and most sensitive in the range of 10-100 pM. Below 10 pM, binding of the random library approaches that of the selected library, while above 100 pM, binding of the selected library begins to decrease

and approach that of the random library. The relationship between heparin concentration and DNA binding demonstrates the ability of heparin to effectively compete for non-specific binding sites on PBMCs.

Enriched Library Consists of Families with Conserved Elements From the enriched library, 34 members were isolated and sequenced as shown in Table 2 (SEQ ID NOs: 7-39). Of these 34 sequences, 33 were unique, and 29 contained the sequence TAGGG (or a variation one base removed) in two locations within the 40 nucleotide random cassette. When aligned by the TAGGG pentamers, additional conserved elements emerged and were used to classify the isolates into families as shown in Table 2.

The sequences of the 34 isolates from the enriched library are aligned by their conserved TAGGG elements (boldface) and classified into families sharing other conserved elements.

Only the sequence of the evolved 40 nucleotide cassette is shown in the alignment. The sequences of the invariant flanking regions are shown in the box and are the same as those from SEQ ID NO: 1. Runs of 2 or more G residues are underlined. The 10 isolates chosen for further characterization are indicated with a bullet. Computer algorithms were unable to identify any stable secondary structures for the selected ligands, possibly due to an overall lack of pyrimidine residues (particularly C residues) in the random cassette. However, conservation of a complex higher-order structure cannot be ruled out, as a large number of GG elements (underlined in Table 2 and consistent with the formation of G-quartet motifs) were selected for in the random region and exist upstream in the invariant flanking region.

Isolates from the Enriched Library Bind PBMCs with High Affinitv To compare the affinity of the selected families for PBMCs, one member of each was chosen for a binding assay (indicated with a bullet in Table 2). The affinities of the chosen ligands in PBS and 100 pM heparin ranged from 400-3,000 cells/pl except ligand L9, which lacked the conserved TAGGG elements and showed half saturation at 15,400 cells/pl as shown in Table 3.

The Enriched Library Binds PBMCs but not RBCs A DNA ligand is most useful if it not only shows high affinity binding to PBMCs, but also shows specific binding to PBMCs. Using the cell-excess binding assay described above, the affinities of DNA-0 and DNA-21 for human PBMCs, rat PBMCs, and human red blood cells (RBCs) were compared. In PBS and 2.5 mM heparin, rat PBMCs mimic human PBMCs in their interaction with each DNA library. In PBS and 100 llM heparin, DNA-21 binds better than DNA-0 to human RBCs, but even at cell concentrations as high as 1 05/ß1

(saturation conditions for PBMC binding to DNA-21), RBCs show only 5% binding to DNA- 21 and less than 1% binding to DNA-0.

DNA: PBMC Complexes are Disrupted bv DNA Competitor A characteristic of dead cells is an inability to pump out internalized DNA. To demonstrate that the DNA binding seen in the binding assays is a measure of complex formation on the surface of viable cells rather than internalization by dead cells, we pre- bound a saturating concentration of PBMCs with radiolabeled DNA-21 and followed with a chase of excess unlabeled DNA-21 at various concentrations. When the data is plotted as the percent of labelled DNA bound as a function of competitor concentration, a sigmoidal relationship is seen showing one-half saturation at approximately 20 nM competitor and approaching zero as the competitor concentration increases. When this data is plotted as a scatchard, two types of interactions are seen: a high affinity interaction with a Kd value of 8 nM and a stoichiometry of 3 x 105 DNA/cell, and a low affinity interaction with a Kd value of 460 nM and a stoichiometry of 3 x 106 DNA/cell. Internalization of DNA by dead PBMCs is inconsistent with these results, as all of the pre-bound DNA-21 is competed off at concentrations of unlabeled DNA-21 above 1000 nM.

Example Two 2'-F RNA Ligands to Human Fibrin Clots This example describes the ability to obtain RNA ligands to human fibrin clots. The pyrimidine residues of these RNA ligands have been modified with fluorines at the 2'- position of the sugar. The fibrin ligands are useful as diagnostic agents as described previously.

A. METHODS Clot formation Human blood was collected in EDTA Vacutainer tubes (Becton-Dickenson), spun at 4 ° C in a clinical centrifuge. Plasma was removed and stored at-70 ° C. Clots were generated in glass tubes by the addition of CaCl2 to a final concentration of 20 mM, incubated for 12-16 hrat37°C.

For the SELEX protocol, the clots were generated in the presence of a glass hanger.

Clots were washed 2 hours at 20°C by continuous exchange of 125 ml 0.01M HEPES, 0.125 M NaCl, 2 mM MgCl2, pH 7.5 (Fibrin buffer).

For the in vitro assays, clots were generated by recalcification of 50 ml plasma in 96-

well microtiter dishes. After 12-16 hours in a humidified chamber at 37 ° C, the clots were washed by 4 x 200 pi buffer changes at 15 minutes each.

For the in vivo pulmonary embolism assay, clots were generated from recalcified plasma as above. Clots from 2 ml plasma were rimmed and centrifuged for 10 min in a clinical centrifuge. They were washed with 2 ml buffer with centrifugation. Clots were then homogenized for 1 min at low speed with a Tissue-Tearor (Biospec Products). Homogenate was washed 3 x 2 ml buffer followed by passage through 18,20,21,22, and 23 Ga needles respectively. Homogenate was resuspended in 0.5 volumes buffer relative to initial plasma volume.

Generation of RNA Pool 2'F-pyrimidine, 2'OH-purine RNA was used for this SELEX. The initial DNA template, 40N8, was synthesized on a solid-phase automated DNA synthesizer by standard techniques and had the sequence gggagauaagaauaaacgcucaa-40N-uucgacaggaggcucacaacaggc (SEQ ID NO: 40). All subsequent PCR rounds utilized the primers 5'- taatacgactcactatagggagauaagaauaaacgcucaa (SEQ ID NO: 41) and 5'-gcctgttgtgagcctcctgtcgaa (SEQ ID NO: 42) as the 5'and 3'primers, respectively. PCR, reverse transcription and generation of RNA with T7 RNA polymerase was performed as previously described.

Transcription of 2'F RNA was performed in the presence of 1 mM each ATP and GTP (in the presence or absence of a_3'P-ATP), and 3mM each 2'F UTP and 2'F CTP. Transcription proceeded for 5-14 hours at 37 ° C followed by gel electrophoretic purification in the presence of formamide and 7 M urea.

SELEX Protocol The general protocol used for this SELEX is outlined in Table 4. Clots from 0.5 ml plasma were immersed in a 1-4 mM solution of 2'F RNA pool in fibrin buffer for 1 hr at 20 ° C. Clots were washed by immersion with 4 x 1 ml buffer for 30 min each. The clot was then macerated with a sharp blade and shaken vigorously for 1 hr in 0.6 ml phenol and 0.45 ml 7M urea. 0.4 ml CHCI, is added to elicit a phase separation, followed by centrifugation at 14,000 RPM. The aqueous phase was extracted with equal volumes 1: 1 phenol: CHCI,, then CHCI, and precipitated with 1.5 ml 1: 1 isopropanol: ethanol in the presence of NaOAc and tRNA as a carrier. Generally 0.5-10 pmoles RNA was recovered from a SELEX round.

Stringency and specificity were added to the SELEX after the pool showed signs of increased binding in buffer. Initially, after seven rounds of SELEX, the SELEX binding reaction was done in heparin-anticoagulated plasma. Subsequent washes were done in buffer.

At later rounds, washes were also performed in heparinized plasma. No attempt was made to alter the clot size or RNA concentration. A first SELEX performed in this manner yielded a significant amount of fibrinogen cross-reactivity. A second SELEX was performed which diverged from the first at round six, at which time a fibrinogen'Counter-SELEX'was added.

1-4 nmoles of a 1 mM transcribed pool RNA was premixed with human fibrinogen to a final concentration of 25 pM. After 15 min incubation at 37°C, the solution was filtered two times through three 1 cm diameter,. 45 micron, nitrocellulose filters. This resulted in the removal of 80-90% of the protein. The filtered RNA was requantitated and added to clot SELEX reaction.

Sequence Alignment CLUSTER Algorithm CLUSTER is a program that performs multiple sequence alignment with reoptimization of gap placement within the growing consensus. The algorithm consists of two parts: sequence alignment and clustering. Sequence alignment uses the dynamic programming algorithm of Altschul and Erickson (Altschul and Erickson (1986) Bulletin of Mathematical Biology 48: 603-616) with a weight vector selected on an a priori statistical basis, namely, a match = 1.0, mismatch =-1/3, gap opening =-1.0 and gap extension =-1/3. The total cost of alignment is the sum of each pairwise alignment within the consensus, utilizing the quasi-natural gap costs of Altschul (Altschul (1989) J. Theoretical Biology 138: 297-309). Normalization of alignment costs allows for comparison between alignments that contain different numbers of sequences. The normalization used in CLUSTER compares an alignment to the best possible one in which every position matches.

A normalized score is the cost of alignment divided by the cost of the best possible alignment. The K-Means algorithm clusters sequences into families. Here, the algorithm is modified slightly from the original version (Tou and Gonzales (1974) Pattern Recognition Principes (Addison-Wesley Publishing Company)) to accommodate cost of alignment as the distance measure. Convergence occurs when there is only one family, or the cost to combine any two clusters is beyond a threshold. Optimization (Step 3) pulls out subsets of sequences and realigns them as described by (Subbiah and Harrison (1989) J. Mol. Biol. 209: 539-548).

Fibrinozen binding Fibrinogen binding was determined by standard nitrocellulose filter binding assays as described in the SELEX Patent Applications.

In vitro clot assay To 50 pi clot in microtiter wells was added 5,000 or 25,000 CPM (-. 5 or 2.5 pmoles)

in 100 pl. Clots were incubated for 1 hour followed by 4 x 200 pl buffer washes at 15 minutes each. Microtiter wells were counted directly in the presence of scintillant.

In vivo pulmonary embolism assay Clot homogenate prepared as above from 200 pi plasma was admixed with 100 pmoles (-l x 106 CPM) for 15 min at 22°C just prior to injecting suspension via a 23-ga needle into the tail vein of a 200-250 gm Sprague-Dawley male rat. At predetermined times, the animal was sacrificed by exsanguination followed by removal of the lungs. The left lung, which consists of only one lobe, was pressed onto Whatman one paper and then dried on a gel dryer at 80 ° C for 2 hours and subjected to autoradiography. The multi-lobar right lung was homogenized in 1 ml buffer and quantitated in scintillant.

Histolosic autoradiosraohv To visualize clot-bound RNA, histologic autoradiography was employed. RNA was 5'-end-labeled with y-"P-ATP. Binding was performed as described for the SELEX reactions or for the in vivo pulmonary embolism assay. Tissues were fixed at least 24 hours in 10% neutral buffered formalin, processed to paraffin, and made into 5 pm sections on poly-L- lysine slides. After drying in 60 ° C oven, they were deparaffinized and rehydrated prior to exposure. Lungs were perfused with normal saline via the right atrium and inflated with 10% formalin prior to removal, fixation and imbedding. Slides were dipped in melted nuclear emulsion (Amersham LM-1), allowed to dry, and exposed at 4 ° C. Slides were developed in Dektol developer (Kodak), fixed (Kodak Fixer), and stained in Giemsa (Sigma).

B. RESULTS Two separate SELEX experimentals were performed on fibrin clot as outlined in Table 4. The two SELEX experimentals differed in the degree and method of counter- SELEX. In the first SELEX experiment (termed FC), eleven total rounds were performed.

The binding reaction was performed in buffer for the first seven rounds. The binding was done in human heparinized plasma for rounds eight and nine. The final rounds were done in whole human heparinized blood. In all rounds, the clot was washed with Fibrin buffer. The second SELEX experiment (termed FCN) diverged from the first at round six when there was a first indication of enrichment. A 25 pM fibrinogen counter-SELEX was added to each round beginning at round six. In addition, the binding reactions were done in heparinized human plasma and the clots were washed with plasma instead of physiologic buffer for rounds 7-14. The final round pools bound 2.5% and 6.4%, respectively, in the presence of heparinized plasma. The round twelve and fourteen pools for the first and second SELEX

experiments, respectively, were sequenced. In both cases RNA sequencing indicated considerable nonrandomness. The pools were amplified with new primers containing EcoRl and Hind III sites on the 5'and 3'end, respectively, and cloned into pUC 18.

Visualization of clot binding The round eleven pool from the initial SELEX was 5'-end labeled with 33P. The pool was admixed with clot in an identical manner to a SELEX in Fibrin buffer. After washing the clot was fixed in formalin, imbedded, sectioned and overlaid with autoradiography emulsion.

Development of the sections showed the RNA (visualized as black grains) were coating the outside of the clot with some diffusion into the interstices of the clot. In another experiment, the rat PE model was performed with"P kinased ligand. The pool was pre-bound to the homogenized clot and injected into the tail vein of a rat. At fifteen minutes the rat pulmonary bed was perfused with saline via the right atrium. The lungs were inflated and fixed by injection of 10% formalin into the trachea prior to removal and placement in formalin.

Tissues were processed as above. Tissues showed black grains only in close association with intravascular clots. There was no evidence of RNA pooling downstream of occluded vessels.

Furthermore, when the study was run with a non-evolved round 0 pool no black grains were visualized within the lung.

Sequence analysis and activity screening Seventy-two clones of each were sequenced (SEQ ID NOS: 43-130). Eighty-eight unique clones were seen and 15 clones differed by only one nucleotide. The sequences were combined for analysis and grouped into sequence motifs by the application of CLUSTER and visual inspection as shown in Table 5. Only the sequence of the evolved 40 nucleotide cassette is shown in the alignment. The sequences of the invariant flanking regions are included in each clone and are the same of those in SEQ ID NO: 40. When the unique clones from both SELEXes were combined for CLUSTER analysis they formed 17 separate motifs.

27/88 clones (31%) were grouped into two major motifs. Motif I and II had 15 and 12 members, respectively. A third motif (Motif III) contained 9 members primarily from the first SELEX and had properties similar to Motif I. Four of the motifs had only two members each.

78/88 (89%) clones were screened for binding in the qualitative in vitro microtiter plate assay. These clones were grouped into high, medium and low affinity with 37, 10. and 31 members in each group, respectively. 46/78 clones screened were further screened for fibrinogen-binding activity. The screen was a standard nitrocellulose binding assay

employing a four-point curve from 0.1-10 pM fibrinogen concentration. Sixteen of the clones were further screened for clot binding in the in vivo rat pulmonary embolism assay.

Results for each of these assays are shown in Table 5.

Of the 27 clones in the major two sequence motifs, 24 were evaluated by the initial screen for binding in the microtiter plate assay. Of those, 15/24 (63%) were characterized as high-affinity or moderate-affinity clot binders. The fibrinogen-binding screen was also divided into high, moderate and low affinity groups with 14,6, and 26 in each group, respectively. In the fibrinogen-binding screen, high-affinity binders were included if the Kd<1 mM, while low-affinity binders included those clones with a Kd>l mM. In Motif I, 10/11 (91%) were in the high-or moderate-affinity for fibrinogen binding while, in Motif II, 0/9 (0%) fell in the high-or moderate affinity fibrinogen binding groups. Eleven members of Motifs I, II were tested in the in vivo PE assay. The clones from Motif I had on average 40% increase in clot binding over Motif II when the binding reaction was performed in buffer.

However, when the binding reaction was performed in heparinized plasma, Motif I had a binding decrease by 90% while Motif II had a decrease of only 10%. There was a clear distinction between Motif I and II in the degree of fibrinogen binding. Motif I bound clots with a slightly greater degree than Motif II but had a significant degree of crossreactivity.

More definitive fibrinogen binding curves indicated that Motif I clones had Kd of 200-600 nM. The Kd (fibrinogen) of Motif II is too high to be quantitated accurately. 1-3% binding was seen at the highest fibrinogen concentration of 10 M. One can extrapolate a Kd of greater than 100 p M.

Binding Quantitation The two best binders in the PE model which had the lowest affinity for fibrinogen were pursued. Both of these clones resided in Motif II. Clone 69 (SEQ ID NO: 55) was <BR> <BR> <BR> <BR> analyzed for binding in vitro homogenized clot. By adding a fixed amount of radiolabeled clone 69 (2 nM) to a fixed amount of clot (200 pliters plasma equivalent) with increasing amount of nonradiolabeled ligand, binding could be quantitated. Analyzing data in a Scatchard format yield a two-component curve with high and low affinity binding components. There were 200 nM high affinity sites per 200 pliters plasma equivalent. The ligand bound these sites with a Kd of 10-20 nM. These sites were saturable. Furthermore, if the ligand was pre-bound to the clot homogenate, it could be competed off the clot by the addition of 3 pM unlabeled clone FC 69 with a half-life of 37 minutes. The labeled ligand did not diffuse off the clot homogenate to any significant degree over 4 hours in the presence

of buffer alone or 3 mM of a 2'-F clone which had no measurable affinity for clots. As such it appears that the binding of a specific ligand to clots is specific and stable.

Clone Truncation Boundary experiments were performed in which the ligand was radiolabeled on either the 5'or 3'end. The ligand was subjected to partial cleavage by modest alkaline hydrolysis and bound to fibrin. Binding RNAs were purified and sequenced. The results are shown in Table 6. Typically a ladder was seen until a region critical to binding was lost, at which point there is a step-off on the sequencing gel. Duplicating the reaction with both ends labeled allowed the determination of both the 5'and 3'boundary. Boundary studies were performed on one clone from Motif I and two clones from Motif II. All clones could be folded into a putative secondary structure which was consistent with the boundaries. The Motif I could be folded into a'dumbbell'structure. Motif II used a significant amount of the 3'-fixed region.

It could be folded into a stem-loop/bulge structure. Based on the boundaries and the structure potentials four nested synthetic 2'F oligonucleotides of clone FC 69 (SEQ ID NO: 55) were synthesized by automated solid-phase synthesizer ranging from 25-41 nucleotides in length (SEQ ID NOS: 131-134). These were tested for binding to homogenized clot by competition with full-length material both in vitro and in the rat PE model. In the in vitro assay, qualitatively binding was seen with all four clones, 69.4 (SEQ ID NO: 134) (the longest) being the best. In the rat PE model, again, all four truncates bound clot. The two truncates with four additional nucleotides past the boundary on the 3'-end showed 3-fold increased binding over those whose sequence ended exactly at the 3'-boundary. The binding to clots in the lung as normalized to full-length material was 32,118,36, and 108% for each of the four truncates, respectively. Furthermore, the binding of the best truncate in this assay, 69.2 (SEQ ID NO: 133) (29-nucleotides), was partially inhibited by the addition of 1 pM unlabeled full- length clone FC 69.

Example Three RNA Ligands to Stenotic Carotid Arteries This example describes the ability to obtain RNA ligands to rat stenotic carotid arteries. The stenotic carotid arteries ligands are useful as diagnostic and pharmaceutical agents as described previously.

A. METHODS Generation of RNA Pool 2'F-pyrimidine, 2'OH-purine RNA was used for this SELEX. The initial DNA template, 40N8, was synthesized on a solid-phase automated DNA synthesizer by standard techniques and had the sequence gggagauaagaauaaacgcucaa-40N-uucgacaggaggcucacaacaggc (SEQ ID NO: 40). All subsequent PCR rounds utilized the primers: 5'- taatacgactcactatagggagauaagaauaaacgcucaa (SEQ ID NO: 41) and 5'-gcctgttgtgagcctcctgtcgaa (SEQ ID NO: 42) as the 5'and 3'primers, respectively. PCR, reverse transcription and generation of RNA with T7 RNA polymerase was performed as previously described.

Transcription of 2'F RNA was performed in the presence of 1 mM each ATP and GTP (in the presence or absence of a-32P-ATP), and 3mM each 2'-F UTP and 2'-F CTP. Transcription proceeded for 5-14 hr at 37 ° C followed by gel electrophoretic purification in the presence of formamide and 7 M urea.

SELEX Protocol 250 gm male Sprague-Dawley were subjected to either unilateral or bilateral balloon- injury of the carotids. Rats were anesthetized with isoflorane. The carotids exposed by a 1 cm midline incision. The common, internal, and external carotid were identified. A #2 French Fogarty catheter was inserted into the external carotid just above the bifurcation and advanced to the aortic arch. The balloon was inflated and pulled back to the bifurcation. This was repeated six times. The catheter was removed, and the external carotid was ligated. The skin was closed by cyanoacrylate glue. Injuries were allowed to develop for 10-14 days.

At the time of SELEX, animals were sacrificed under anesthesia by exsanguination.

Both carotid arteries were dissected from the bifurcation to the aortic arch. The arteries were gently stripped of any associated connective tissue. Twelve rounds of SELEX were performed ex vivo as indicated in Table 7. The first three rounds were done by simply immersing two arteries in 0.5 ml of a 2 pM RNA solution. The binding reaction was rotated at 20 ° C. Carotid segments were then washed with four 1 ml buffer washes for 15 min each prior to harvesting the bound RNA. Subsequently, two carotid arteries were ligated together in series with a small length of polyethylene tubing. The distal ends were also canulated with tubing for attachment to a syringe pump and collection of eluent. These procedures were done with minimal disruption to the arterial segments. For the SELEX, in one ml physiologic buffer was passed through the arterial segments at 4 ml/hr. This was

followed by washing the segments with an additional 1 ml of buffer at 4 ml/hr. The segments were taken out of line, counted by Cherenkov radiation, and processed for RNA extraction. Rounds 4-7 were performed in this manner with both artery segments having been balloon-injured. All tissue was processed for RNA extraction. In rounds 8-12, an uninjured artery was ligated upstream from an injured artery as shown in Figure 1. Perfusion proceeded as above. Both artery segments were counted, but only the injured segment was processed for RNA extraction. Rounds 8-12 were done to"counter-SELEX"against the evolved RNA binding normal arterial endothelium. In a subsequent control, it was shown that the uninjured artery had an intact monolayer of intimal endothelium by Factor VIII immunohistochemistry.

Tissue Extraction of 2'-F-RNA Carotid segments were minced with a scalpel and homogenized with 1 ml TRIZOL Reagent (Gibco). Homogenate was clarified by centrifugation, phase-separated with CHCl3, and the aqueous phase precipitated with isopropanol, all according the manufacturers protocol. Purified RNA was resuspended in H20, and digested for 15 minutes at 37 ° C with 0.1 U/pL DNAse I (Pharmacia) and 100 pg/ml RNAse A (Sigma) in reverse transcription buffer. 2'-F-pyrimidine, 2'-OH-purine RNA is stable to RNAse A digestion. The digest was phenol, phenol/CHCl3 extracted and EtOH precipitated out of sodium acetate. The RNA was then subjected to RT/PCR under standard conditions to generate a template for T7 RNA polymerase. After twelve rounds the pool was cloned and sequenced. The sequences identified as C# in Table 8 were obtained by this protocol.

In Vivo SELEX In a subsequent SELEX, 3-5 nmoles of the Round twelve pool was injected directly into the tail vein of a rat with a 14 day unilateral lesion. After 15 minutes, the animal was sacrificed and the carotids processed. RNA was amplified as before. Four Rounds of in vivo SELEX were done as indicated in Table 7. This pool was cloned and sequenced and the sequences from both cloning steps were combined for sequence analysis. The sequences identified as Civ# in Table 8 were obtained by this protocol.

Binding Analysis Binding of RNA either as a pool or individual clones was performed by comparing 32p counts bound to normal versus injured carotid artery segments. Binding was visualized by histologic autoradiography in either the ex vivo perfusion system or by overlaying RNA onto fresh-frozen carotid artery slices.

Histologic autoradiography To visualize carotid-bound RNA, histologic autoradiography was employed. RNA was 5'-end-labeled with y-33P-ATP. Binding was performed as described for the SELEX reactions. Tissues were fixed at least 24 hours in 10% neutral buffered formalin, processed to paraffin, and made into 5 mm sections on poly-L-lysine slides. After drying in 60 ° C oven, they were deparaffinized and rehydrated prior to exposure. Slides were dipped in melted nuclear emulsion (Amersham LM-1), allowed to dry, and exposed at 4 ° C. Slides were developed in Dektol developer (Kodak), fixed (Kodak Fixer), and stained in Giemsa (Sigma).

Fresh-frozen carotid sections were prepared by imbedding normal or injured carotid artery segments in OCT and freezing at-20 ° C. 5 pm sections were cut on a cryostat, placed on a slide (typically a normal and injured section were juxtaposed on a single slide), and stored frozen at-5 ° C. Slides were warmed to room temperature, the paired sections were encircled with a grease pencil, and pre-bound with 30 ml PBS, 0.5% Tween-20,1 mM low molecular weight heparin (Calbiochem). After 15 minutes the solution is removed and 30 pi of the same solution containing 10,000 CPM (~1 pmol) 33P-labeled RNA is added for 30 minutes Slides were washed twice with PBS/Tween-20, twice with PBS. Slides were fixed in 10% neutral buffered formalin, then rinsed in distilled water prior to exposure.

Sequence Alignment CLUSTER Algorithm. CLUSTER is a program that performs multiple sequence alignment with reoptimization of gap placement within the growing consensus. The algorithm consists of two parts: sequence alignment and clustering. Sequence alignment uses the dynamic programming algorithm of Altschul and Erickson (Altschul and Erickson (1986) Bulletin of Mathematical Biology 48: 603-616) with a weight vector selected on an a priori statistical basis, namely, a match = 1.0, mismatch =-1/3, gap opening =-1.0 and gap extension =-1/3. The total cost of alignment is the sum of each pairwise alignment within the consensus, utilizing the quasi-natural gap costs of Altschul (Altschul (1989) J. Theoretical Biology 138: 297-309). Normalization of alignment costs allows for comparison between alignments that contain different numbers of sequences. The normalization used in CLUSTER compares an alignment to the best possible one in which every position matches. A normalized score is the cost of alignment divided by the cost of the best possible alignment. The K-Means algorithm clusters sequences into families. Here, the algorithm is modified slightly from the original version (Tou and Gonzales (1974) Pattern Recognition Principles (Addison-Wesley

Publishing Company)) to accommodate cost of alignment as the distance measure.

Convergence occurs when there is only one family, or the cost to combine any two clusters is beyond a threshold. Optimization (Step 3) pulls out subsets of sequences and realigns them as described by (Subbiah and Harrison (1989) J. Mol. Biol. 209: 539-548).

B. RESULTS SELEX Twelve rounds of ex vivo RBIC SELEX was performed followed by four rounds of in vivo SELEX as indicated in Table 7. Pools were cloned and sequenced after the ex vivo SELEX and the ex vivolin vivo SELEX ; the sequences are provided in Table 8. Only the sequence of the evolved 40 nucleotide cassette is shown in the alignment of Table 8. The sequences of the invariant flanking regions are included in each clone and are the same as those of SEQ ID NO: 40. The last five rounds of the ex vivo SELEX were done with a normal carotid artery as a negative selection (Counter-SELEX). Evaluation of these rounds indicated that over the last five rounds the injured carotid bound between 0.07-0.5% without a trend towards increased binding in the later rounds. The discrimination between normal and injured was 3.2-4.5, again without a trend toward increased discrimination. At round twelve, the RNA pool was sequenced and shown to be significantly non-random.

The pool was then taken forward in the in vivo SELEX. Very little RNA was recovered from the injured carotid arteries (0.2-0.6 pmoles). Comparing CPM in the normal versus injured yielded discrimination values of 2.61-3.54. At the first round of in vivo SELEX, equal amounts of Round 12 RNA and Round 0 RNA were injected into two different animals both with unilateral balloon injuries. There was no discrimination for the Round 0 RNA (i. e. the same number of counts bound the normal as the injured artery), whereas in the round 12 pool, 2.61 times more RNA bound the injured carotid as compared to the uninjured. At Round 15, the evolved pool was injected into the animal or perfused through an ex vivo apparatus exactly as had been done for rounds 8-12. The discrimination of the Round 15 RNA was 4.61, which was higher than had ever been seen during the ex vivo SELEX.

Seventy-two clones from the ex vivo SELEX were sequenced, of which 50 were unique as shown in Table 8. The striking finding was that of the seventy-two clones, two were present in multiple copies. One clone (clone C33; SEQ ID NO: 146) had nine identical or one base difference copies, while another (clone C37; SEQ ID NO: 186) had ten copies.

Thus nineteen of twenty-two copies in the initial sequencing arose from two sequences.

Sequences stemming from those two persisted after the in vivo SELEX with clones related to C33 generating the largest single family in the combined analysis (Motif I).

Analysis of Clone Binding Of the ninety-four unique clones from the two SELEX methods, twenty-eight were screened for binding to fresh-frozen rat carotid artery sections. These were qualitatively graded by intensity of staining (+, ++, +++) and specificity (s, ns) as shown in Table 8.

Clones were seen with a variety of patterns from no visible binding to strong binding of all tissue components. Specificity was graded based on relative intensity of binding to neointimal tissue over normal or injured media or adventia. Early on C33 (SEQ ID NO: 146) was found to have both increased intensity and specificity over both unevolved Round 0 or Round 12 pool RNA. Further screening uncovered three other clones with better binding characteristics: C59 (SEQ ID NO: 150), Civ45 (SEQ ID NO: 202), and Civ37 (SEQ ID NO: 210). Civ41 (SEQ ID NO: 158) was of interest because it was of the same Motif as C33 and C59, but had very intense staining and little specificity: staining neointima, as well as normal an injured media. Civ45 in three independent binding experiments had the most intense staining in a specifically neointimal distribution. This clone also showed an slight increase in binding to injured media over normal media. If smooth muscle cells migrate from the media to the neointima in the injured artery, then it may not be surprising that whatever this clone is binding exists within the injured media. Of the four clones noted with high specificity, two of them are from Motif I and are closely related. Three of the four contain the sequence GLTUUG (underlined in Table 8). Putative secondary structures are shown in Table 9. It is unknown at this time whether these structure correlate to the true structure. In the absence of boundary experiments they provide a basis for truncation studies.

Clone C33 and C59 were 33P-labeled and perfused in an ex vivo manner. Although not quantitated, they showed dramatic binding to the lumenal wall of the damage artery but not to the normal vessel.

C59 was used to stain fresh-frozen section of RBIC of different ages. Carotids were harvested at 1,2,4,6,8,16 weeks after balloon injury. The neointimal signal was greatest at 2-6 weeks. It was minimally present at one week and disappeared after six weeks. The pattern of staining neointima in the highly specific clones is diffusely granular. Silver grains are not obviously associated with smooth muscle cell bodies. One hypothesis is that the RNAs are binding to components of the extracellular matrix (ECM). It is known that SMCs require an ECM scaffold to migrate. They have been shown to lay down unique

proteoglycans in the course neointimal proliferation. The presence and disappearance of these unique proteoglycans corresponds temporally to the binding of RNAs to neointima. As such one viable possibility is that the RNAs are binding specifically to these proteoglycans.

EXAMPLE 4 RNA Ligands to Stenotic Carotid Arteries This Example describes the identification of RNA ligands to rat stenotic carotid arteries using a similar method to that described in Example 3 but with a few modifications.

The method used to generate the 2'-F-pyrimidine, 2'-OH-purine RNA pool in this Example was identical to the method described in Example 3.

A. METHODS SELEX Protocol 250 gm Sprague-Dawley were subjected to either unilateral or bilateral balloon-injury of the carotids as described in Example 3.

Initial selections were performed ex vivo. Rat carotid arteries were balloon-denuded with a #2 Fogarty catheter introduced into the external carotid artery and advanced to the aortic arch. After fourteen days the carotid artery was harvested, carefully stripped of excess connective tissue and ligated in line to polyethylene tubing (Figure 1). The arterial segment was placed at 25°C in a phosphate-buffered saline (PBS) bath. 32P body-labeled transcript (0.75 mM) in PBS was passed through the artery once, over a fifteen minute period (4ml/hr).

The arterial segment was then washed in the same manner with PBS. The segment was disconnected, minced, and the RNA extracted as described in Example 3. The RNA was then subjected to RT/PCR under standard conditions to generate a template for T7 RNA polymerase. This set of procedures constituted one round of SELEX. After six rounds, 0.62% of the RNA was bound to the artery. At round seven an uninjured carotid arterial segment from the same rat was ligated in series upstream of the balloon denuded segment to "counter-SELEX"against the evolved RNA binding normal (uninjured) arterial endothelium.

In Vivo SELEX After twelve rounds of ex vivo SELEX, in vivo SELEX was initiated. Four to six nmoles of the ex vivo enriched library was injected into the tail vein of a rat with a unilateral 14-day balloon injured carotid lesion. Animals were sacrificed at 15 min (in vivo round 1-4), 30 min (round 5-6), 40 min (round 7) or 60 min (round 8-12). Tissue was removed and extracted for RT/PCR as described in Example 3. Figure 2 shows the progression of the in

vivo SELEX displayed as the amount of RNA bound per wet weight of tissue (CPM/mg) as a function of the number of SELEX rounds. After twelve rounds of in vivo SELEX the enriched library was cloned.

Binding Analysis Clones were screened for binding neointimal tissue in vitro by histologic autoradiography. Histological autoradiography was performed either by perfusing RNA that was 5-end-labeled with gamma-33P-labeled through the vessel as in Figure 1, or by cutting fresh frozen sections and applying the ligands to the cut surface as described in Example 3.

B. RESULTS Twelve rounds of ex vivo RBIC SELEX were performed followed by twelve rounds of in vivo SELEX. Pools were cloned and sequenced after the in vivo SELEX; the sequences are provided in Table 10 (SEQ ID NOS: 242-258). The sequences shown are the random regions of the sequences; the 3'and 5'sequences of the invariant flanking regions, which are not included in the clones, are same as shown in SEQ ID NOS: 139 and 140, respectively.

The last six rounds of the ex vivo SELEX were done with a normal carotid artery as a negative selection (Counter-SELEX). Evaluation of these rounds indicated that over the last six rounds, the injured carotid bound initially led to a decrease in binding to the balloon injured segment (0.02%), with increased binding in the last five rounds. At the twelfth and final ex vivo round, the binding to the normal and balloon injured segment was 0.1% and 0.34%, respectively.

The pool was then taken forward in the in vivo SELEX for twelve rounds. The round twelve in vivo SELEX was cloned and sequenced. Forty-four clones were sequenced and aligned into two motifs (Table 10). Forty-eight percent of the sequenced library was represented by sequences containing one highly conserved motif, of which clone 12.2 (SEQ ID NO: 242) was the most prevalent member.

Analysis of Clone Binding Autoradiography after perfusion indicated intense uptake on the lumenal surface with little penetration into the neointima. Autoradiography after application of clones to the fresh frozen section showed substantial neointimal binding of the enriched library compared to the random library. The uniform granular silver grain distribution, was not clearly cell- associated, and is suggestive of an extracellular matrix target. Comparison of perfusion and fresh frozen autoradiography led to the conclusion that the clones are binding a neointimal

component which is both displayed on the lumenal surface of the lesion and, in addition, is widely distributed throughout the neointimal matrix.

Representative clones from both sequence families were able to specifically compete with clone 12.2 for binding by histologic autoradiography. Radiolabeled clone 12.2 was competed off the fresh frozen section by unlabeled clone 12.2 (SEQ ID NO: 242), 12.13 (SEQ ID NO: 246), 12.15 (SEQ ID NO: 248) and 12.37 (SEQ ID NO: 250) in a dose dependent manner with complete competition by a 30-fold molar excess of the unlabeled clones.

However, there was no diminution of signal by the addition of a 300-fold molar excess of a sequence scrambled version of clone 12.2. It is interesting to note that the two sequence families which have no discernible sequence similarities are able to compete with each other for target binding.

Secondary structure analysis The secondary structure of clone 12.2 (Figure 3) was predicted using an RNA folding algorithm (Zuker (1989) Science 244: 48-52) and further supported by structure-specific endonuclease and truncation analyses. Computer-assisted structure predictions were supported experimentally by endonuclease probing. Digests were performed under non- denaturing conditions (4°C, 10 minutes in PBS) with single-strand specific endonucleases T, (GpN, Boehringer Mannheim, 1 x 10-3 U/ml), U2 (AptN, USB, 10 U/ml), S, (NpN, Boehringer Mannheim, 4 x 104 U/ml) and P, (NpN, Boehringer Mannheim, 3 U/ml) and double-strand specific endonuclease V, (NpN, Pharmacia, 7 U/ml). Partial cleavage ladders were generated and visualized by denaturing PAGE to identify reactive nucleotides.

Truncation analysis was performed by deleting or substituting regions of clone 12.2 (boxed with a dashed line) and assaying activity in vitro. Alterations that reduce or eliminate activity are bracketed above the structure, and those that have no effect on activity are bracketed below the structure.

The motif consists of two stem loops forming a"dumb-bell"shape with the 5'and 3' termini juxtaposed. Truncated versions of clone 12.2 were transcribed in vitro, each containing the first twelve bases of clone 12.2, which are required for efficient transcription of 2'-F pyrimidine RNAs. Truncates were tested qualitatively in the in vitro histologic autoradiography assay and selected truncates were assayed for balloon injured carotid tissue accumulation in vivo. Results of the truncation analysis are summarized in Figure 3. Most of the 5'stem-loop nucleotides (&num 13-41) can be removed with no loss of activity. The two 3'- terminal nucleotides (#86-87) can also be removed. The sequence of the loop (nucleotides

#65-69) is not important as it can be replaced with a GAAA tetraloop (Uhlenbeck (1990) Nature 346: 613-614) without affecting activity. In contrast, small deletions in the 3'stem critically affect activity. Removal of residue #62 and the two loop-closing base pairs (#63-64 and #70-71) abolishes activity. While loss of residues &num 13-41 is tolerated, removal of residues #13-46 eliminates activity. The most interesting finding is that there appears to be a requirement that the two juxtaposed guanosines (&num 1 and #85) not be linked by a phosphodiester bond. A synthetic oligonucleotide consisting of nucleotides &num 42-85 linked to #1-4 does not have activity, while a transcribed oligonucleotide 12.2t55L (Figure 4; SEQ ID NO: 259) lacking most of the 5'stem-loop (#13-41) but containing the transcription leader sequence (#1-12), has full activity (Figure 4). Flexibility at the"hinge" (residue #46) between the two stem loops appears to be important for high affinity binding. Furthermore, removal of the 5'-terminal GGG (#1-3) in a synthetic truncate of 12.2t55L abolishes activity, consistent with the structure proposed in Figure 4. The 55 nucleotide oligonucleotide 12.2t55L is currently the shortest truncate with identical tissue accumulation as the full length clone 12.2 in vitro.

Biodistribution studies A limited biodistribution study was performed in rats (Figure 5). Sixty minutes after a bolus injection of 2 nmoles (10 mCi) 32P-a-ATP-labeled clone 12.2, radioactivity accumulated in blood to 0.2+/-0.05% ID/g (+/-s. e.), in intercostal muscle to 0.05+/- 0.02% ID/g, in normal carotid to ID/g, and in injured carotid to 1.4+/-0.4% ID/g.

The ratio of balloon injured to normal carotid for clone 12.2 was 14 (p<. 05). An 87- nucleotide 2'-F transcript control with the same terminal invariant regions was used as a non- specific control. The ratio of clone 12.2 to non-specific oligonucleotide binding injured carotid was 7 (p<. 05). These values compare very favorably to those previously reported for vascular imaging agents (Lees et al. (1988) Arteriosclerosis 8: 461-470; Hardoff et al. (1993) J. Clin. Pharmacol. 33: 1039-1047). Clone 12.2 is cleared primarily in the liver, kidney and spleen accumulating to 3-4% ID/g in each of these tissues at 60 minutes Sequence- independent accumulation in these organs is routinely observed. Blood, muscle, normal and balloon injured carotid artery tissue were extracted, normalized to wet tissue weight, and applied to a denaturing gel (Figure 6).

The accumulation of full length transcript into the injured artery was evaluated. RNA extraction with Trizol (Calbiochem) yielded 60-70% of the radiolabel in the aqueous phase after phase separation. If a 2'-F transcript was added to tissue ex vivo just prior to extraction,

>90% of the radiolabel was precipitated as full-length RNA. In the first round (15 min in the rat), 47% of the extracted radiolabel from the injured artery was isopropanol precipitable as full length material. By the twelfth round (60 min in the rat) 85% of the extracted label was precipitable as full length. In contrast, radioactivity extracted from the blood at the twelfth round was only 15% precipitable.

Full length material was easily visualized from the injured carotid tissue but only faintly present in the other tissues. To visualize the binding of clone 12.2, in another animal, the aortic arch and both carotid arteries were dissected in one piece. The vessels were visualized on a phosphor capture screen (Fuji). The image indicated a seven-fold increased uptake per unit area of clone 12.2 in the balloon injured artery compared to the control artery or the aortic arch region.

Binding kinetics Binding kinetics were studied. Binding of clone 12.2 was rapid and approached 3% ID/g at the earliest time point (15 minutes), and was lost from the injured carotid with a tl, 2 = 50 minutes. In this case, the ratio of clone 12.2 present in injured carotid vs. blood peaked at 40 to 60 minutes. It is interesting to note that the signal to noise is the highest at 60 min, the time used for the last rounds of in vivo SELEX.

EXAMPLE 5 RNA Ligands to Watanabe Heritable Hvperlipidemic Rabbit Arteries This example describes the ability to obtain RNA ligands to Watanabe Heritable Hyperlipidemic Rabbit (WHHL) arteries. The Watanabe Heritable Hyperlipidemic Rabbit (WHHL) is an extensively used model of atherosclerosis in which the rabbit has a homozygous deficiency for the low density lipoprotein receptor. These animals maintain very high cholesterol levels and develop atherosclerotic lesions spontaneously.

A. METHODS Generation of RNA Pool 2'-F pyrimidine, 2'-OH purine RNA was used for this SELEX. The initial DNA template, 40N8, was synthesized on a solid-phase automated DNA synthesizer by standard techniques and had the sequence gggagauaagaauaaacgcucaa-40N-uucgacaggaggcucacaacaggc (SEQ ID NO: 40). All subsequent PCR rounds utilized the primers: 5'- taatacgactcactatagggagauaagaauaaacgcucaa (SEQ ID NO: 41) and 5'-gcctgttgtgagcctcctgtcgaa

(SEQ ID NO: 42) as the 5'and 3'primers, respectively. PCR, reverse transcription and generation of RNA with T7 RNA polymerase was performed as previously described.

Transcription of 2'-F RNA was performed in the presence of 1 mM each ATP and GTP (in the presence or absence of a-3zP-ATP), and 3mM each 2'-F UTP and 2'-F CTP.

Transcription proceeded for 5-14 hours at 37°C followed by gel electrophoretic purification in the presence of formamide and 7 M urea.

SELEX Protocol A WHHL or control New Zealand white (NZW) rabbit was anaesthetized with isoflorane. The animal was exsanguinated and the aorta from the arch to the diaphragm was dissected. The arterial segment was placed at 37°C in Hanks solution buffered with 25 mM <BR> <BR> <BR> <BR> HEPES (H/H), pH 7.3. The 32p body-labeled transcript (0.25-1.0 mM) in H/H was incubated with the arterial segment for 45 minutes at 37°C with gentle agitation. The arterial segment was then washed in the same manner four times with increasing wash buffer volumes as the SELEX continued to increase stringency (Table 11). Wash volumes increased from a total of 4 mL in Round 1 to 160 mL in Rounds 9 and 10. The plaque-containing segment was then weighed, minced, and the RNA extracted. For rounds 1-3 (in vitro SELEX), regions containing plaque, which was readily apparent by visual inspection, were isolated as a full thickness including the intima, media, and adventitia. The RNA was reverse transcribed, amplified by the polymerase chain reaction (PCR), and transcribed to RNA. This set of procedures constituted one round of SELEX. Five rounds cf in vitro SELEX were performed, as indicated in Table 11 (Rounds and 10).

In vivo SELEX International Publication No. WO 98/30575, published July 16,1998, entitled "Bioconjugation of Macromolecules,"which is incorporated herein by reference, describes RNA polymerase catalyzed transcriptions of DNA templates initiated with a nucleoside or nucleotide bearing a group other than a triphosphate at the 5'-position. In Rounds 4-8 a SELEX the pool was transcribed in the presence of a five-fold excess guanosine aminohexyl phosphate (guanosine 5'monophosphate with a hexyl amine moiety on the 5'position) over guanosine triphosphate in order to efficiently place a unique primary amine on the 5'-end of the transcript. The transcript pool was conjugated with a tripeptidyl 99m-Tc chelate moiety (mercaptoacetyl-glycyl-glycyl-amidyl)-6-hex-1-yl), as described in United States Patent Application Serial Nos. 08/358,065 and 08/488,290, supra, via N-hydroxy succinamide

chemistry. 5 nmole of the conjugated pool was labeled with 99m-Tc (radiochemical yield after purification 90-95%) to a specific activity of 1,000 Ci/mmole. The labelled pool was injected into the marginal ear vein of an anaesthetized WHHL rabbit. Ten minute images of the rabbit in the left lateral position were obtain with a Siemans LEM DIFITRAC ZLC 10" FOV gamma camera. After 15-65min the animal was exsanguinated, as indicated in Table 11. The thoracic aorta and arch were isolated and the plaque harvested. For rounds 4-6 (in vivo SELEX), regions containing plaque, which was readily apparent by visual inspection, were isolated as a full thickness including the intima, media, and adventitia. For rounds 7-8 (in vivo SELEX, Table 11) the intima and plaque were stripped from the media and adventitia for processing.

Two additional rounds of in vitro SELEX (rounds 9-10) were performed as described above. The pool was incubated with the arterial segment and the arterial segment was then washed as indicated in Table 11. The plaque-containing segment was then weighed, minced, and the RNA extracted. For rounds 9-10 (in vitro SELEX) the intima and plaque were stripped from the media and adventitia for processing.

B. RESULTS The Phospho-Imager images of the aortas from a WHHL rabbit and a New Zealand White (NZW) rabbit after the injection of 5 mCi of Round nine into the marginal ear vein 40 min prior to harvesting, in concert with corroborating in vitro incubation data and RNA pool sequencing data (not shown), supported the conclusion that the initial library had evolved and was ready to sequence. The Round 5 and 10 pools were cloned and sequenced; the sequences from round 10 are provided in Table 12 (SEQ ID NOS: 260-354). From the Round 10 pool, 28% of the sequences were highly related, most likely resulting from a single clone (Family 1; SEQ ID NOS: 260-287). In addition a second family (Family 2; SEQ ID NOS: 288-294) was identified containing 7 highly related sequences. None of the remaining 65% of the sequences (Orphans; SEQ ID NOS: 295-354) have been placed into families or, as yet, screened for activity. Members of the Family 1 were screened by in vitro incubation with plaque-containing arterial segments. The aortic segments were incubated with 50 nM of the clone for 30 min at 37°C, followed by 3 x 40 min washes at 37°C. Figure 7 shows the results of the screen. All but one of the screened clones had high affinity for the plaque. The non- binding clone was determined to be unrelated to the other clones. Clone 10.31 (SEQ ID NO: 268) was chosen as the lead clone for further analysis.

EXAMPLE 6 In vitro bindinz of Clones 10.5 and 10.31 to Human Atherosclerotic Plaque Tissue Clones 10.5 (SEQ ID NO: 260) and 10.31 (SEQ ID NO: 268) from Example 5 were assayed in two separate experiments for in vitro binding to human atherosclerotic plaque tissue. Human tissues were obtained from recipients of heart transplants. The vessels were used within three hours of removal from the recipients and were kept oxygenated in Tyrodes buffer until use. In each experiment the arterial segment was placed at 37°C in Hanks solution buffered with 25 mM HEPES (H/H), pH 7.3. The transcripts were incubated with the arterial segment for 30 minutes, and the arterial segment was washed with 3 x 40 minute washes.

In the first experiment, clone 10.5 was compared to 40N8 (the initial unevolved library; SEQ ID NO: 40). Figure 8 shows binding of family member 10.5 to human plaque- containing coronary artery with 40N8 as the negative control. Clone 10.5 bound-3.5 times better than the unevolved library.

In a separate experiment with an independently harvested artery from a second subject, 10.31 clone was 33P labeled. This allowed for higher spatial resolution on the Phospo-Imager and in addition histologic autoradiographic analysis of the binding. Histologic autoradiography of an ex vivo incubation with clone 10. 31 and a negative control 10.15 (SEQ ID NO: 348) indicated patchy intense binding at the level of the endothelium or subendothelial matrix with very little diffusion into the tissue. No histological structures could be identified to which 10.31 was binding. This experiment demonstrated that a WHHL Rabbit derived nucleic acid ligand has cross-reactivity to human atherosclerotic tissue.

EXAMPLE 7 Analysis of Clone 10.31 for Plaque Accumulation in WHHL Rabbits after Intravenous Iniection Clone 10.31 (SEQ ID NO: 268) from Example 5 was analyzed for plaque accumulation in rabbits. Previous incubation studies indicated that 10-20% of input dose could accumulate in plaque. 3.5 nmole (50 uCi) of clone 10.31 was 32P body-labeled and injected into a WHHL rabbit via the marginal ear vein. The animal was sacrificed at 60 min and the thoracic aorta was harvested, the arch was removed and the aorta was Phospho- Imaged. The images showed that there was clear accumulation of 10.31 in the plaque areas,

which stained red after Oil Red O staining. The ratio of plaqued to non-plaqued vessel accumulation is 5. In one part of the study, tissues were Trizol extracted, isopropanol precipitated, counted and applied to a PAGE gel. Figure 9 shows the fraction of 32P ethanol precipitable counts for various tissues after Trizol extraction. Much of the radioactivity is in the form of non-precipitable counts, indicating degradation of clone 10.31. The amount of degradation of 10. 31 in the plaque is surprising. Of the radioactivity recovered from the tissue after injection, only 20% is precipitable, but all of this is full-length.

EXAMPLE 8 In vitro Binding. In Vivo Imaging and Biodistribution of 99m-Tc-Labelled 10.31 Clone 10.31 (SEQ ID NO: 268) was transcribed in the presence of a five-fold excess guanosine aminohexyl phosphate over guanosine triphosphate and then conjugated with a tripeptidyl 99m-Tc chelate moiety as described in Example 5 via N-hydroxy succinamide chemistry to give transcripts which were >90% conjugated to the chelate moiety. In this example, the conjugated transcripts were then labeled with 99m-Tc to a specific activity of 5000 mCi/mmole transcript.

In one part of the experiment, 50 nM (0.5 mCi) of 99m-Tc labeled 10.31 (Figure 15; SEQ ID NO: 355) was incubated in vitro with a WHHL arterial segment for 30 min at 37°C, followed by 3 x 30 min wash at 37°C. Phospho-Imager images of the arterial segment showed that 15% of the input dose was found in the plaqued artery.

In a second part of the experiment, on the same day, the 99m-Tc labeled 10.31 was injected into another WHHL rabbit for in vivo imaging and biodistribution. The WHHL rabbit was catheterized in the contralateral ear prior to injection of 10.31.7.5 mCi of 99m- Tc-10.31 was injected into the marginal ear vein. At specific times blood was withdrawn from the catheter and 99m-Tc was quantitated. The rapid clearance of radiolabel from the blood can be seen in Figure 10. Radiolabel dropped rapidly by 10 min and then stayed constant, not dropping to zero. From the previous Trizol extraction data (Example 7) this basal level does not represent full-length material For the in vivo imaging, ten minute images were obtained until sacrifice at 60 minutes. The images showed that the uptake of clone 10.31 in the liver is very rapid. The rabbit is different from rats and mice in that over time the liver clears the radiolabel. In the final gamma camera image (50-60 min), the kidney, spleen and bladder had significantly more radiolabel than the liver. In a darker exposure the aortic knob could be visualized.

Region of Interest analysis (Figures 11A and 11 B) indicated that the radiolabel accumulated in the arch very rapidly (within 10 min) and remains for the remainder of the hour. Figure 12 shows the formal biodistribution data. The accumulation of clone 10.31 in the plaque was 0.015% ID/g.

After the in vivo imaging studies were complete, the animal was sacrificed and the aorta was removed for in vitro imaging. The intact aorta was place on the gamma camera. A strong signal was seen with highest intensity in the arch and a second focus of activity distally. These foci of nuclide accumulation were confirmed to be plaque by Oil Red O and delineated more clearly by the Phospho-Imager.

EXAMPLE 9 Comparison of In vivo Imagine and Biodistribution of 10.31 in WHHL and New Zealand Rabbits A second set of imaging experiments were performed in order to obtain biodistribution data and to assess whether the aortic knob could be seen in the NZW rabbit.

This experiment was performed as described in Example 8, except that 5 mCi (1 nmole) 99m- Tc-labelled 10.31 (SEQ ID NO: 355) was injected into both a WHHL and a NZW rabbit.

Again a catheter was placed in the contralateral ear vein of the WHHL rabbit prior to injection for withdrawal of blood samples. The overall biodistributions of the radiolabeled 10.31 in the WHHL and NZW rabbits were similar to the Example 8. In the Phospho-Imager images of the thorax of the NZW rabbit, blood is seen early on in the heart which dissipates.

At no time is the aortic arch in the NZW rabbit visualized. Phospho-Imager images of the WHHL rabbit showed that more label was retained in the heart. The aortic arch and retrocardiac fullness representing the descending aorta were observed throughout the 60 minutes. The results of this experiment showed that the region of the aortic arch was clearly visible in the WHHL rabbit but not in the NZW rabbit.

Sixty minutes after IV injection the animals were sacrificed and the aortas were removed for in vitro imaging. The aortas were placed on the gamma camera. The image of the WHHL rabbit aorta showed intense uptake in the arch of the WHHL rabbit. Figures 13 and 14 show the biodistribution of 99m-Tc labelled 10.31 in the WHHL (solid bars) and NZW (dashed bars) rabbits. Again, the highest accumulation at 60 minutes was in the kidneys. 0.038% ID/g label was found in the WHHL arch as compared to half that in the NZW arch. These numbers are higher than seen in the previous example. The differential

between normal and diseased artery is less than what was seen in the rat restenotic arteries (Example 4, Figure 5). Subsequently, the plaque was dissected out. The plaque did not contain more counts on a per gram basis than normal vessel but did show increased accumulation over normal vessel on a per area basis.

EXAMPLE 10 Truncation and Post-SELEX Modifications of WHHL Ligand 10.31 In the progression of development of nucleic acid ligands for clinical applications it is desirable to transform a transcript from an enzymatically generated species to a chemically synthesized one. The first part of this exercise is to reduce the transcript to its minimal binding component, termed truncation. Initially, three truncated analogs of 10.31 (SEQ ID NO: 268) were synthesized by standard techniques on a solid phase automated synthesizer removing the 5'-fixed region (SEQ ID NO: 357), the 3'fixed region (SEQ ID NO: 358), or both the 5'-and the 3'-fixed regions (SEQ ID NO: 356) as shown in Figure 16. In vitro accumulation of 32P radiolabeled polyacrylamide gel electrophoresis purified truncates on WHHL aortic segments by methods described in Example 8 indicated that the 3'-fixed region was not required for binding activity but that some or all of the 5'-fixed region was required.

Secondary structure predictions of the truncate without the 3'-fixed region indicated a strong possible structure of 37-nucleotides (Figure 17). Synthesis and assaying of this molecule, termed alternatively 10.31.37 or Tr104 (SEQ ID NO: 359), confirmed that this was the core binding motif. Figure 17 details putative secondary structure of the core binding Motif. The 8 base pair stem has been confirmed by NMR with NOSY.

In preparation for in vivo studies, Trl23 (Figure 19; SEQ ID NO: 360) was synthesized by adding a 5'-pentyl amine and a 3'-3'thymidine cap to Tr104. The 5'-amine is to allow coupling of the 99m-Tc cage as well as pharmacokinetic modulating molecules, while the 3'cap was added to stabilize the molecule from 3'-exonucleases found in vivo. This ligand also bound WHHL plaque as well or better than Tr 104 (Figure 18).

It has previously been shown that post-SELEX modifications of the 2'positions of purines increases the nuclease resistance of an nucleic acid ligand (Green et al (1995) Current Biology 2: 683-695). A set of Trl 04 analogs was synthesized in which increasing numbers of the 2'-OH purines were substituted with 2'-OMe purines: Trl28 (SEQ ID NO: 361) has 18 out of 18 purines modified with 2'-OMe. Trl29 (SEQ ID NO: 362) has 15 out of 18 purines modified with 2'-OMe. Trl 30 (SEQ ID NO: 363) has 13 out of 18 purines modified with 2'-

OMe. Trl31 (SEQ ID NO: 364) has 8 out of 18 purines modified with 2'-OMe. Based on the results shown in Figure 18 it appeared that at least 13 out of 18 purines (Trl30) could be substituted with 2'-OMe without affecting activity (Figure 18).

Trl30 became the basis for further 2'-O-methylation studies (Figures 20A-H, SEQ ID NOS: 369-376). In vitro accumulation assays indicated that seventeen out of eighteen 2' purine sites could be O-methylated without loss of activity (Figure 21; Trl 59; SEQ ID NO: 375). O-methylating all eighteen sites (Trl28; SEQ ID NO: 360) rendered the aptamer inactive. Further studies indicated that the adenosine at position eighteen required a 2'-OH for the nucleic acid ligand to retain binding activity (Figure 21).

In order to test the effect of O-methylation on nuclease resistance, selected nucleic acid ligands were 32p end-labeled and incubated in the presence of fresh heparinized WHHL rabbit plasma. At specified times over a 24 hr period aliquots were removed, diluted and run on a 8% polyacrylamide, 7 M urea gel. Full-length material was quantitated on a Phospho- imager. Figure 22 shows a plot of three truncates, Tr104, Trl28 and Trl29 (from Figure 18).

After factoring out phosphatase activity (which is not effected by O-methylation) it can be seen that full O-methylation increased the t, by 4.2-fold while the partially O-methylated species has 2.5-fold longer t"2. The O-methylated species in which only A-18 is unsubstituted (Trl59; SEQ ID NO: 375) is expected to have intermediate stability between 2.5 and 4.2 times longer than the parent nucleic acid ligand.

In another experiment, selected uridines in truncate Tr104 were substituted with 5-Br- deoxy uridine (5-BrdU). Trl32 (SEQ ID NO: 365) has a 5-BrdU at position 9, Trl33 (SEQ ID NO: 366) has a BrdU at position 16, Trl34 (SEQ ID NO: 367) has a BrdU at position 17, and Trl35 (SEQ ID NO: 368) has a BrdU at position 22. These truncates were radiolabeled and assayed in a manner analogous to the O-methylated nucleic acid ligand truncates. Again it can be seen that the activity of Trl 04 is sensitive to modifications in the terminal loop region. BrdU substitution at positions 17 and 22 had significant effects on binding while substitutions at positions 9 and 16 did not.

EXAMPLE 11 Tissue SELEX on human atherosclerotic coronary artery segments.

Human coronary arteries were harvested from heart transplant recipients. Heart were transported on ice in oxygenated Tyrodes buffer and used within 3 hr of harvest. 1-3 cm coronary artery segments were canulated with PE tubing in a similar manner to Example 3.

A 0.25 to 2.0 micromolar solution of the RNA library 40 N8 in Hanks/HEPES was perfused through the artery in a recirculating mode at 1 ml/min for 30 minutes, at 25°C. The artery was then perfused with buffer alone without recirculation for an addition 30 minutes. The artery was opened longitudinally. Plaque containing regions were excised and stripped of media and adventitia. The RNA library was extracted from the plaque in a manner analogous to the rabbit plaque (Example 8). The library was cloned and sequenced after Round 5. Of 64 clones sequenced 19% (12/64) represented the WHHL Family I (Table 12), 1.5% (1/64) represented WHHL Family 2,11% (7/64) represented a new family, designated huart. 1,2 sequences were found to exist 3.1% of the time (2/64), and the rest were unique and not grouped into families. Table 13 shows the human artery sequences (SEQ ID NOS: 377-440).

Table 1<BR> Summary of Selection Parameters and Data<BR> for PBMC SELEX Selection [DNA] [PBMC] [heparin] Filter %PBMC %bkgd DNA/cell<BR> nM cells/µl µM Type retention retention<BR> 1 1,400 40,000 100 NC 3.1 0.1 6.0 x 105<BR> 2 900 120,000 100 NC 4.7 0.3 2.0 x 105<BR> 3 800 92,000 100 GF 3.5 0.8 9.0 x 104<BR> 4 480 60,800 100 AN 1.8 0.08 8.7 x 104<BR> 5 500 52,500 500 NC 4.6 0.3 2.2 x 105<BR> 6 990 18,300 500 NC 1.3 0.05 5.0 x 105<BR> 7 750 14,400 2,500 NC 1.6 0.09 2.2 x 105<BR> 8 560 47,800 2,500 AN 4.0 0.2 2.8 x 105<BR> 9* 100 11,200 2,500 NC 7.2 0.5 9.3 x 104<BR> 10* 100 12,600 2,500 AN 3.1 0.6 1.0 x 105<BR> 11* 100 5,100 2,500 NC 5.7 0.7 7.3 x 104<BR> 12* 50 37,500 2,500+ plasma NC 2.2 0.1 2.3 x 104<BR> 13* 50 75,000 2,500+ plasma AN 4.9 0.3 2.2 x 104<BR> 14* 10 32,000 2,500+ plasma NC 0.9 0.2 1.6 x 103<BR> 15* 10 32,000 plasma NC 3.1 0.3 2.3 x 103<BR> 16* 10 112,000 plasma AN 5.6 0.5 3.1 x 103<BR> 17* 10 65,000 plasma NC 3.0 0.6 3.1 x 103<BR> 18* 10 87,300 plasma AN 4.8 0.5 4.2 x 103<BR> 19* 10 118,560 plasma AN 6.0 0.5 3.3 x 103<BR> 20* 10 94,560 plasma NC 4.0 0.3 3.1 x 103<BR> 21* 10 48,000 plasma AN 7.3 0.7 5.8 x 103 Table 2<BR> PBMC SEQUENCES<BR> Degenerate ssDNA Library<BR> 5'- AGGGAGGACGATGCGG - [N] 40 - CAGACGACTCGCCCGA - 3' (SEQ ID NO: 1)<BR> 5' fixed random 3'fixed<BR> region region region<BR> Ligand Random Region (fixed regions are provided as described above) SEQ ID NO:<BR> # L49 GGGGTCGGTTCGGGCATA TAGGG TATTCTTCGTA GAGGG 7<BR> L8 GGGGTCGGTTCGGGCATA TAGGG TATTCTTCGTA AAGGG 8<BR> L35 GGGGTCGGTTCGGGCATA TAGGG TATCCTTCGTA GAGGG 9<BR> # L1,L34 CACGT TAGTAGGAT TAGGA TTATTCAGGTTG TAGGG AACA 10<BR> L29 CACGTTCAGCAGGAT TAGGG TTGTTTNGGTTG TAGGG ACACA 11<BR> L18 CACGG TAGTAAGTAG TAGGG TATTATA AT TAGGG GATCCA 12<BR> L21 CACGG CAGTATTATT CAGGG GTCTTAGATAT TAGGG GGCA 13<BR> # L42 CACGGTAGGTTTTAGA TAGGG ATATTTGGTG TAGGG AGCA 14<BR> L5 CACGGTAGGTTATAGA TAGGG ATATTTNTTG TAGGG AACA 15<BR> L11 CACGGTAGGTTTTAGA TAGGG ATATTTGATG TAGGG AGCA 16<BR> L25 CACGGTAGGTTTTAGA TAGGG ATATTTGGTA TAGGG AGCA 17<BR> L36 CACGGTAGGCTTTAGA TAGGG ATATTTGATG TAGGG AGCA 18<BR> # L26 GGGGAG TAGGG TATTTAAAAATGT TAGGG TAAGTTTCCTC 19<BR> L10 GGGGAG TAGGG TATTTAAAAGTGT CAGGG TAAGTTTCCTC 20<BR> # L7 CGTAGTAAGAAGTATTAT TAGGG ATATTG TAGGG GCGCTA 21<BR> L22 CGTAGTAAGAAGTATTAT TAGGG ATATTG CAGGG GCGCTA 22<BR> # L43 GGCAGCAAGA GTTTGAT TAGGG TATAGT TAGGG GCGCTG 23<BR> L19 GCAGCAAGA GTTTGAT TAGGG TATAGT TAGGG GCGCTGC 24<BR> L41 GCAGTAAGG GTTTGAT TAGGG TATAGT TAGGG GCGCTGC 25<BR> L46 GCAGCAAGAGG TTGAT TAGGG TATAGT TAGGG GCGCTG 26<BR> L20 CGGCAAGATGATTGAA TAGGG GATCTAAAGT TAGGG GCGC 27<BR> L6 GCAGCAGG TG TAGGG GTATAGATGGA TAGGG ATTTCTTCT 28<BR> # L28 CACA TAGGG GAAATGA GAATAG TAGGG TATTAATACAGTG 29<BR> L38 CACA TAGGG GAAATGA GAAGAA TAGGG TATTAATACAGTG 30<BR> L50 CAGGT TAGGG GAAAGGTTTAATAAT TAGGG TATAAT GTG 31<BR> # L14 CAGGTAGAGA TAGGG AAGTTTTATG TAGGG GACAATTCGT 32<BR> L44 CAGGTAGAGA TAGGG AAGTTTTATG TAGGG GACAATTCGT 33 Table 2 (cont'd)<BR> # L31 CACAA TAGGG AAATTT GTTGTTATAGT TAGGG ATACTGGA 34<BR> L40 GGCCGAA TAGGG AAATTTA TTATTACT AACAGTAATCCCC 35<BR> L48 CAGGACT TAGGG ATTTAGTTGTTT TAGGG GTTATGTAGT 36<BR> # L9 GGGGGGATGAGATGTAATCCACATGTCACTTATTAAGTCC 37<BR> L12 CAGGGGATGGGATGTAATCCTCATGTCACTTATTAAGTCC 38<BR> L13 TACGACTACGATTGAGTATCCGGCTATAATATTACCATTG 39 Table 3<BR> Sequences and Affinities of Selected PBMC Ligands<BR> LIGAND SEQUENCE OF RANDOM REGION AFFINITY SEQ ID<BR> (PBMC/µl) NO:<BR> DNA-0 Degenerate DNA Library 43,500 1<BR> DNA-21 Enriched DNA Library 1,000<BR> L1 CACGTTAGTAGGATTAGGATTATTCAGGTTGTAGGGAACA 3,000 10<BR> L7 CGTAGTAAGAAGTATTATTAGGGATATTGTAGGGGCGCTA 700 21<BR> L14 CAGGTAGAGATAGGGAAGTTTTATGTAGGGGACAATTCGT 1,200 32<BR> L26 GGGGAGTAGGGTATTTAAAAATGTTAGGGTAAGTTTCCTC 800 19<BR> L28 CACATAGGGGAAATGAGAATAGTAGGGTATTAATACAGTG 2,400 29<BR> L31 CACAATAGGGAAATTTGTTGTTATAGTTAGGGATACTGGA 1,800 34<BR> L42 CACGGTAGGTTTTAGATAGGGATATTTGGTGTAGGGAGCA 1,100 14<BR> L43 GGCAGCAAGAGTTTGATTAGGGTATAGTTAGGGGCGCTG 700 23<BR> L49 GGGGTCGGTTCGGGCATATAGGGTATTCTTCGTAGAGGG 400 7<BR> L9 GGGGGGATGAGATGTAATCCACATGTCACTTATTAAGTCC 15,400 37 Table 4<BR> Serum Clot SELEX 2'F-40N8-RNA<BR> Round Method [Clot] [RNA] [NaCl] Wash Volume, ml % bound Yield, pm Backgrou@<BR> FC<BR> 1 clot 500 ul 4.00E-06 125 mM Buffer 0.5 1 26<BR> incubation plasma<BR> 2 " " 4.00E-06 125mM " " 0.66 12.3<BR> 3 " " 1.90E-06 125mM " " 0.18 1.6<BR> 4 " " 4.00E-06 125mM " " 0.18 3.5<BR> 5 " " 4.00E-06 125mM " " 0.37 7.8<BR> 6 " " 4.00E-06 125mM " " 2.9 50<BR> 7 " " 4.00E-06 125mM " " 6.2 81<BR> 8 " " 3.00E-06 Hep.Plasma " " 9 67<BR> 9 " " 4.00E-06 Hep.Plasma " " 4 53 long wash<BR> 10 " " 2.90E-06 Hep.Blood " " 7.9 70<BR> 11 " " 4.00E-06 Hep.Blood " " 2.5 46<BR> FCN Counter-Si<BR> 7 clot 500 ul 5.40E-06 Hep.Plasma Hep.Plasma 0.5 0.7 20 0.025 mM fibrino<BR> incubation plasma<BR> 8 " " 1.93E-06 " " " 1.3 8.6 "<BR> 9 " " 2.47E-06 " " " 1.4 15 "<BR> 10 " " 6.00E-07 " " " 2.3 4.7 "<BR> 11 " " 2.00E-06 " " " 2.6 20 "<BR> 12 " " 3.80E-07 " " " 1.8 14 "<BR> 13 " " 1.80E-06 " " " 1.9 12 "<BR> 14 " " 2.10E-06 " " " 5.1 41 " Table 5<BR> Fibrin-Binding Clones<BR> FINAL<BR> CLUSTERS<BR> SEQ In Vitro Fibr<BR> ID Clot gen<BR> NO Binding Affi<BR> FC1 : AGGGCUCGUGUGCCAAAUCGCUAACAAC-AAGCUAGCUGAU 43 O<BR> FCN54 : CUGGGCUCAUCCGGCGAAU-GAUG-CAAGGAAGAUUUCACAU 44 O %<BR> score 0.365079<BR> FC3 : AAGGA-UAG-UGUGCU--CCUGUA--CCAAAUUUCCAAAGCGAUAU 45 O<BR> FC73 : AAAGAGUAA---AGCG--CG-GAA--CAGGAUUCACGUUGCGCUCUU 46 O<BR> FC75 : AAAGAGUAA---AGCG--CG-GAA--CAGGAUUCACGUUGCGCUCA 47 O<BR> FCN 6 : CAACGAUUAUCUUUUCGGCCGUGAAACCCAAACUGACGCC 48 O %<BR> score 0.292824<BR> FC4 : CGCGAGGAUAGGGUG--CAGCUUCUGUUCCAAAUACGUGA-AU 49 *<BR> FC12 : UAAGUCGAAGAGCUCCUGAUCCAAACCAUCGA-AAG-GACGU 50 O<BR> FC28 : GG-UAAGUUG--GAGCUCCUUAUCCAAGCACGCAAUAAGUGAC 51 O<BR> MOTIF II score 0.355556<BR> FC6 : UUUGGCGU-GG-GAU-CCUGGA-CUGAAGG--AUUUUGACGAUGC 52 O<BR> FC45 : AUUCAAGACA-GA-GAC-UUUCCU-U-GAA--U-GCUCUGUCCCAUAA 53<BR> FC54 : ACAAAUGU-GC-GAC-CUUGGA-C-GAAGUUAACUCGGACGGUUC 54 O<BR> FC69 : ACAAAUGU-GC-GAC-CUUGGA-C-GAAGUUAACUCGGACGGUUC 55 * %<BR> FC72 : ACAAAUGU-GC-GAC-CUUGGA-C-GAAGUUAACUCGGACGGUUG 56 * %<BR> FCN16 : ACAAAUGU-GC-GAC-CUUGGA-C-GAAGUUAACUCGGACGGUUC 57 O %<BR> FCN19 : AAGUCU-GA-GACUCCUGGA-CUGAA-UUAGCUAGGACGGCUG 58 # %<BR> FCN30 : UAGGAGCCUAGCAGCC-CCUGCAUC-GA---UCACUAGGAUGGUU 59 * %<BR> FCN38 : AAAGUGUAGC-CUU-CCUGGA-CUGUAGGU-ACUAGGACGGUCC 60 * %<BR> FCN44 : ACAAAUGU-GC-UCC-CUUGGA-C-GAAGUUAACUCGG-CGG 61 * %<BR> FCN55 : AAGAAGCUG-GC-GAC-AGGCGA---AAAGCAGACU-UGAGGGGAA 62 O %<BR> FCN71 : AGUAG-GU-GA-GGCUUCUGGA-CUGAAG-UAACUAGGUCGGUUC 63 * %<BR> score 0.339752<BR> FC8 : CAUGA-GCUGCUGGACCAAA-CAGAUG-GAGG---AACCA-CCGUGU 64 *<BR> FC39 : GA-GCU-CUUGACGAAAACCUAUGCGAGAUGGAUACU-CGGUU 65<BR> FC40 : GA-GCU-CUUGACGAAAACCUAUGCGAGAUGGAUACU-CGGUU 66<BR> FC41 : GA-GCU-CUUGACGAAAACCUAUGCGAGAUGGAUACU-CGGUU 67 O<BR> FC55 : UGA-GCU-CUUGAAGAAGUCC-----GAAC---AUUCUCCUUUCUGCGACU 68 O<BR> FC64 : GA-GCUCCGGGAUCCAAGCG--UGCAACA---ACACU-AUGCCCAC 69 O<BR> FC65 : AAUAC-CCU-CGGGAACCAAUCC-----GACCCU-AUUUUGCAGUUUG 70 *<BR> FCN59 : AUGAUGCU-CCUGAAGUAAUCACCAG-GAC----AUCCU-CGGCAU 71 O %<BR> FC9 : GCAAU--CU-CGGACUAGACCAACGACCUUCGUUUGACGCUC-A 72 O %<BR> FC18 : CCGAUUU--CU-AGGACG-GAUUUACG---GAGAAUUGAGUCGC-AAG 73 O Table 5 (cont'd)<BR> FINAL<BR> CLUSTERS SEQ In Vitro Fibri@<BR> ID Clot gen<BR> NO: Binding Affin@<BR> MOTIF I score 0.262803<BR> FC38 : A-CGG-CGAGAAUGACAAU-GUUAUUCUACGAGCGAAGGAUUA 74<BR> FC50 : GCAAU--CU-CGGACUAGACUAACGACCUUGGUUUGACGCUA-A 75 * &num <BR> FC51 : GCAAU--CU-CGGACUAGACUAACGACCUUCGUUUGACGCUA-A 76 * &num <BR> FC60 : AGAGCAGCCGGAGGUGUGAGCUCUGACUCUG-AACAGCUG 77 O<BR> FC76 : CGGGAUUU--CU-CGGAAAAGACUAACGAC--UAAUUCCAGAACC 78 O<BR> FCN11 : GCAAU--CU-CGGACUAGGCUAACGACCUUUGUUUGACGCUC-A 79 * &num <BR> FCN14 : GCAAU--CU-CGGACUAGACUAACGACCUUCGUUUGACGCUU-A 80 * &num <BR> FCN26 : UC--CA-AGGACCAAACGGGUGUUCGGCAGUGGACU-UU-AGCAA 81 O &num <BR> FCN28 : UGGGCUAC--AU-GUGAGUACACCAGCGUGAGAGUUCUUAGG 82 O &num <BR> FCN29 : CUGUGCAGUAACUGCGGAUGAGACCAACCGG-AUGGCUCAAC 83 # &num <BR> FCN57 : ACAAU--CU-CGGACUAGACUAACGACCUUCGUUUGACGCUU-A 84 * &num <BR> FCN61 : GCAAU-CU-CGGACGAGACUAACGACCUUCGUUUGACGCUU-A 85 * &num <BR> FCN65 : AGCGCUAGAUGGACGAGAGACUUUUAAGUAGC-AAGCGGUA 86 *<BR> score 0.274690<BR> FC10 : CAGACUCAGAGCGCCGUGAGCUUCUGAAG-CAA--UCGCAGGU 87 * &num <BR> FC53 : G-CGGGGAGCUCCUCGAGAAACUGAGUUCAACUUCCCAGGU 88 O &num <BR> MOTIF III score 0.387597<BR> FC14 : GUGUUGGAGCUCUUGAUUGGAAAAG--UAGA--ACAAAUCGAAA 89 * &num <BR> FC30 : GUGUUGGAGCUCUUGAUUGGAAAAU--UAGA--GCAAAUCGAAA 90 * &num <BR> FC49 : CAAUCCGAGCUCUUGAA-GCAAUCC--UUGAUUGCAAGAUGAU 91 O<BR> FC52 : UCGGAUG-AGCUCUUGAA-GCAGUUC-AAGG--ACAGACAUAAAG 92 *<BR> FC59 : UUCCAGGU-UAGCGGCCAAACC--UCGA-CUUGAACAGACUUUA 93 O<BR> FC70 : GUGUUGGAGCUCUUGAUUGGAAAAG--UAGA--GCAAAACGAAA 94 * &num <BR> FC74 : GUGUUGGAGCUCUUGAUUGGAAAAG--UAGA--GCAAAUCGAAA 95 * &num <BR> FCN2 : GUGGUGCUGCAAUUGCUCGGUCGGC--GUGCUCUCUACUUGA 96<BR> FCN3 : GCUCAAGAGACUGAA-GGAAAAGCUUAGAGCUCAAAGC-AUA 97<BR> score 0.306520<BR> FC22 : CGUGUUGGGUUCAAAGACCAGCUUACGGUACACAGUACGA 98 O<BR> FCN31 : UC-UGUUGG-UUCAAAGACUUGCUAAGGGGUCGAAGCACCCU 99 * /<BR> score 0.492063<BR> FC24 : GAC-GACAAAG-AGUCCGUUC-CAAACCUC-UGAGACAGGGU 100 O<BR> FCN8 : U-AGCAAGUCCCACAUCCCAGACGGG-CUAAAAAGAGGUGGA 101 O %<BR> FCN18 : AUGAAC--GACCGCGG--GCAGUCGCGU-UCAAAUGAG-UGGUUUU 102 # %<BR> FCN24 : AUGAGUA-GACCGAGGAAGCACCCGGCUCUCAAAUGAG-UGA 103 * /<BR> FCN64 : AUGAGUA-GACCGAGGAAGCACCCGGCUCUCAAAUGAC-UGA 104 O /<BR> score 0.294667<BR> FC25 : AAGGCC-AU--CAGGGCAAAGACCUCCUAGGUACUGA-CGCUUA 105 O<BR> FC34 : AAGGCCGAA--CAACGAAGUUUGAUUC-AGGUACUCAGCG-UUC 106 O % Table 5 (cont'd)<BR> FINAL<BR> CLUSTERS SEQ In Vitro Fibrin@<BR> ID Clot gen<BR> NO Binding Affini@<BR> FC35 : AAGGCCGAA--CAACGAAGAUUGAUUC-AGGUACUCAGCG-UUC 107 #<BR> PCN34 : AAGGCGGAGGGCAAGCAAGA-ACCU-C-ACGAACAGA-CG-UUAA 108 #<BR> score 0.428395<BR> FC33 : AGCCUGAGGUAUAGUU---ACG-CU-AUAUGGGA--GGUAGGCUUUA 109 *<BR> PCN13 : CGUGAUG-ACAGCUCGGACGGCUCAU-UGCGCGGAGUAG-CUA 110 * %<BR> FCN56 : CGGCUCGAUG-CUAGCUGGGACGGCUCAU-UGAGACUGGUUG 111 # %<BR> score 0.324074<BR> FC42 : AGUGCAACCU-GAACCAAACCAAACUAGCGCGCAGUUGGGU 112 #<BR> FCN36 : AGCAGAUGGUGCUGAGGUAU-CAU-GAAGACGCUGAC-GCUUA 113 # /<BR> FCN49 : GAAUGGAGCCAAGAAAGACAGCGAUGUCUCGGAC-GAUGAG 114 # %<BR> FCN50 : GAAUGGAGCCAGGAAAGACAGCGAUGUCUCGGAC-GAUGAG 115 * /<BR> FCN58 : GCAUGGAGCCAAGAAAGACAGCGAUGUCUCGGAC-GAUGAG 116 # %<BR> PCN67 : CGUG-AGAUUCCCCUGCGUAAGACCA-GAAGACUAUCAG-GCU 117<BR> score 0.335225<BR> FC43 : AGGGUUGAGGCUUAUCCUUCUUUCGUUCGUGACACGAUCG 118 *<BR> FCN69 : GAUUGACACGCA--CUCCAAUGGCUC-UGAAGUGUUCGUGUGC 119 # %<BR> score 0.296296<BR> FC46 : AAAUUCAAUGCUCUGAUGGGUUUAUGAGUUAAUGCGU-GGAC 120 #<BR> FC61 : AAAGGCCCU-UUCAGCAGGGAUC---GAGGUACUGGAU-GGAUA 121 #<BR> FCN27 : AAAGUCG-UGUGC-GAGAGGCUCA-GAUUUAAUGCGGAGGA 122 # %<BR> score 0.343669<BR> FCN9 : CGUU-GAAAUCGCUCCUCAGU-G-UGAGUUGAAUCAGCUGACC 123 # %<BR> PCN21 : AGUUUGGAUUCG-GCAGGUGCUG-AGACUUUGAU-AGCC-ACUA 124 # %<BR> PCN22 : GUGAGAAAU-G-UCGGGGGC-GAUGACUUGGA-CGGUCCACCG 125 # %<BR> FCN60 : CGUU-GAAAUCGCUCCUCAGC-G-UGAGUUGAAUCAGCUGACC 126<BR> score 0.376263<BR> FCN12 : AGAGAGGAACUGCGAUUCAGACCAAAACGGA--AAUGGCUGU 127 # %<BR> FCN25 : GAUAUACUAACUUUCUUUGAAAGCCCAAAAGUAUUAAUG-CG 128 * %<BR> FCN48 : GAUAUACUAACUUUCUUUGAAAGCCCAAAAGUAUUAAUG-CG 129 * /<BR> FCN52 : AGCCGAGCUAAUCCCGAAAGUGACCCGGAACGACG-CGGCA 130 # %<BR> score 0.363636<BR> Table 5 KEY<BR> #-low in vitro clot binding<BR> #-moderate in vitro clot binding<BR> *-high in vitro clot binding<BR> %-low fibrinogen affinity, Kd>1mM<BR> /_moderate fibrinogen affinity<BR> &num -high fibrinogen affinity, Kd<1mM<BR> @-low binding PE assay<BR> $-high binding PE assay Table 6<BR> Truncation Analysis<BR> boundary<BR> FC &num 69<BR> gggagacaagaauaaacgcucaaACAAAUGUGCGCCCUUGGACGAAGUUAACUCGGACGG UUCuucgacaggaggcucacaacaggc 55<BR> 69.1 (25)<BR> CGAAGUUAACUGCGAACGGUUCuucg<BR> 69.2 (29)<BR> CGAAGUUAACUCGGACGGUUCuucgacag<BR> 69.3 (29)<BR> UGGACGAAGUUAACUCGGACGGUUCuucg<BR> 69.4 (41)<BR> CCCUUGGACGAAGUUAACUCGGACGGUUCuucgacaggagg<BR> FCN &num 30 boundary<BR> gggagacaagaauaaacgcucaaUAGGAGCCUAGCAGCCCCUGCAUCGAUCACUAGGAUG GUUuucgacaggaggcucacaacaggc<BR> N30.1 (23)<BR> UCGAUCACUAGGAUGGUUuucga<BR> N30.2 (30)<BR> UCGAUCACUAGGAUGGUUuucgacaggagg<BR> N30.3 (31)<BR> CCCCUGCAUCGAUCACUAGGAUGGUUuucga<BR> N30.4 (38)<BR> CCCCUGCAUCGAUCACUAGGAUGGUUuucgacaggagg<BR> A-C C-U<BR> 5'-C-G-A-A-G--U-U-A U U-C-G-A--U-C-A A<BR> 3'-g-c-u-u-C G-G-C C a-g-c-u G-G-U G<BR> UU A-G G UUU A-G<BR> 69.1 (25) N30.1 (23) Table 7<BR> Rat Carotid Artery SELEX<BR> Dis<BR> Round Method Buffer [2'F-40N8], uM % bound inj<BR> Ex Vivo<BR> 1 minced Ringers lact. 5 6.7<BR> 2 minced Ringers lact. 2 2.7<BR> 3 minced Ringers lact. 3.4 6.9<BR> .<BR> <P>4 perfused.inj. Ringers lact. 2.5 0.2<BR> 5 perfused.inj. Ringers lact. 5.9 0.6<BR> 6 perfused.inj. Ringers lact. 2.3 0.62<BR> 7 perfused.inj. Ringers lact. 2.3 0.7<BR> 8 perf. normal Ringers lact. 0.75 0.54<BR> injured Ringers lact. 0.75 1.04<BR> 9 perf. normal PBS 0.75 0.02<BR> injured PBS 0.75 0.07<BR> 10 perf. normal PBS 0.75 0.1<BR> injured PBS 0.75 0.5<BR> 11 perf. normal PBS 0.75 0.1<BR> injured PBS 0.75 0.35<BR> 12 perf. normal PBS 0.75 0.1<BR> injured PBS 0.75 0.34<BR> In Vivo amount Injected, nmoles<BR> 13 in vivo PBS 3 0.01<BR> 40N8(control) in vivo PBS 3 n.a.<BR> <P>14 in vivo PBS 3 0.02<BR> 15 in vivo PBS 3.8 0.02 0.94<BR> 15 ex vivo PBS 3.8 uM n.a.<BR> <P>16 in vivo PBS 3.9 0.005 Table 8<BR> MOTIF I FINAL CLUSTERS SEQ INTENSITY<BR> ID OF<BR> NO STAINING<BR> p3 : gggagauaaga-auaa-acgcu-caa 139<BR> p5 : uucgacaggaggcucacaacaggc 140<BR> C4 : ACCA-CUGGGC-CCAG-UUUAG-AAA---CU-CAUU-----GCCCAAAUCCGG 141<BR> C13 : AAG--AAGA-AUCG-AAAAAUCUAC--CUUGUUC-GGAGCCUGCUCU 142 +<BR> C15 : U-CUAG-AC-AG-CGAAGGCUGAGCUAUGACACUGAACUUCUUA 143 -<BR> C22 : GCAAUCU-GGA-CUAG-ACUAA-CGAC--CUUCGCU-UGACGCUCA 144<BR> C26 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGC--AUU 145<BR> C33 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUCA 146 ++<BR> C38 : GCAAUCUCGGA-CUCG-ACUAA-CGAC--CUUCGUU-UGACGCUCA 147<BR> C53 : GACAAUAACCGC------ACCAA-CGUU--CU--GUU-CUUCGCUUGCACGU 148 -<BR> C57 : CAAU-UCCCA-CU-G-AUUCG-GGGC--GGUCCUUGCGAUGGCGAGA 149<BR> C59 : CUCAGA-CAACCAACAG-CA-C--GUUC-UC-UGUU-UUCGUCGUUUG 150 +++<BR> C60 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUUA 151<BR> C72 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUCG 152<BR> Civ3 : CGCU-CAUG-ACCAGGCGCUA-CUGACUG-AGAUGUUGAACUUA 153 -<BR> Civ14 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CCUCGUU-UGACGCUCA 154<BR> Civ27 : AUAAGAUCAAC-AUUGG-CG---GUU--UA-UGUUAUUCGUCCGUUUG 155<BR> Civ30 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUUA 156<BR> Civ34 : CACGCGAGAG-CU---UCUAA-AGCU--GCUGAAU-CGA-GCUCCACGA 157<BR> Civ41 : ACAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUUA 158 +++<BR> Civ42 : GCAAUCUCGGA-UUAG-ACUAA-CGAC--CUUUGUU-UGACGCUCA 159<BR> Civ48 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUA-UGACGCUUA 160<BR> Civ50 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGCU-UGACGCUCA 161<BR> Civ53 : GGAGAUCCUCGA-GGAA-ACU---CGAA--CUUCUUCCCGACGUUGA 162<BR> Civ59 : ACAGCUCGGA-UAAG-ACUAA-CGAC--CU-AGUU-UG--GCUAAGCAA 163 ++<BR> Civ65 : GCAAUCUCGGA-CUAG-ACUAA-CGAC--CUUCGUU-UGACGCUCA 164<BR> score 0.268867<BR> C1 : ACCAAGGGAGUCGGUUUAU-UCAGCCUG-UUCGGAACCUGACU 165 +<BR> C44 : AUCCAAGACGCUUAGU---UCUUGCUC-UUCGGGGCUUC-CUA-CG 166<BR> C45 : AAGUAAACUCGAGACCGUUCUGGCUGAUUCGGGGC-ACUCU 167<BR> C55 : ACUUGACAA-UCCCCCUGAUUCGGGGCCUGACUAUCACGA 168 ++<BR> C58 : ACACGACA-UCGAAGUUA-UCCCCCUGAUUCGGAGCCAG-CUG 169 - Table 8 (cont'd)<BR> MOTIF I FINAL CLUSTERS SEQ INTENSITY<BR> ID OF BINDI@<BR> NO STAINING SPECI@<BR> C65 : AGCUGGAAA-UCCAAAUGC-UUUGUCUAGUU-GGGGC--CACUU 170<BR> C67 : AGACUCUUGAUCA-UCCCCCUAGUUCGGGGC-UGACUG-CACU 171<BR> Civ6 : ACUUGACAA-UCCCCCUGAUUCGGGGCCUGACUAUCACGAU 172<BR> Civ24 : GGAGCGAAAUUCUUGAAUA-UCC-ACUGAUUCGGACCGUC-CU 173 -<BR> Civ29 : GCGGGAUUUUCCUGAUCA-UCCCACUGAUUCGGGGCCUUAG 174<BR> Civ31 : AGUUUCUCCUUGGCAA-UCCCCCCUAUUCGGGGCUUCAUUG 175<BR> Civ55 : GAGCGAAAUUCUUGAAUA-UCC-ACUGAUUCGGAGCGUC-CU 176<BR> Civ56 : ACGGCAUUCUAAACAU-UCCCCCUUGUUCGGAGCCACUCU 177<BR> Civ62 : GCGGA-UUUUGAUCA-UCCCCCUGAUUCGGAGAC-CUCUUAC178<BR> score 0.2986720<BR> C3 : GGGAACGAAUCGUCCAAAA--GA-CCUCGCGGAAUCGGC-G-UUA 179 + @<BR> C17 : GCGAGCUCCU-GCACAAAACCGAUCCUCGC--AUACAGCAGGU 180 -<BR> C30 : GGGAACGAAUCGUUCAAAA--GA-CCACGCG-AAUGGC-GCUUA 181<BR> score 0.441975<BR> C10 : GAGCUGUUGACGAAAACUUAUGCGGAGAU--GGAUA-CUCGGU 182 + @<BR> C27 : GAGCUCUUGACGAAAACCUAUUCG-AGAU--GGAUA-CUCGGUU 183<BR> C35 : GGAGCCGAUUG-UACAACCUAGGUG-AGCU---CAAU-CACCUCGC 184<BR> C36 : GGGCCCUCUGCUACAACUUCGGCA-AGGA---CAUU-UUCCGGAC 185<BR> C37 : GAGCUCUUGACGAAAACCUAUGCG-AGAU--GGAUA-CUCGGUU 186 + @<BR> C49 : GAAAGC-CAUGUUGAAAGUUUCACCC-AGAUUCGGA---GUCGUUG 187 ++ @<BR> C62 : ACUGAGCUCGUGU--ACAA--UGUUAG-GGAA--GGACAUCUCGAUA 188<BR> C66 : GAGCUCUUGACGAAAACCUAUGCG-AGAA--GGAUA-CUCGGUU 189<BR> C69 : GAGCUCUUGACGAAAACCUACGCG-AGAU--GGAAA-CUCGGUU 190<BR> Civ32 : CACAGGGGUUUC---AAACCUCCCCC-UGAUUCGGAGG-UUC 191<BR> Civ38 : AACCUCGCCAGGAAUAACU-UGCG-ACUUUCGGAUC-GUCUUA 192 + @<BR> Civ54 : GAGCUCUUGACGAAAACCUAUGCG-AGAU--GGAUA-CUCGGUU 193<BR> Civ64 : CUUUGGAGCUCCUGG---AACGAAAGCG-GAAU---UAAC-UUCCUUA 194<BR> Score 0.303419<BR> C11 : ACAAUUCAGGACGGGG---UUUCUU---GAAUG--GGUUCGACCUU-CA 195<BR> C52 : CCAGUA-GAUCAA-CUCCCUGGCAACU---GGUUCGCCGUUUAUA 196<BR> Civ5 : A-CCUUGAUGUUCA--CUCCCU---AACUCAAGGUUCGACGUC-UA 197<BR> Civ33 : ACAA-CCUGGACAAGGAAUUUUUCU----AGUGUUCGUUGGACGU 198 ++ @<BR> Civ35 : A-CCUUGAUGUUGAA-CUCCCU---AACUCAAGGCUCGACGUC-UA 199 ++ @ Table 8 (cont'd)<BR> FINAL CLUSTERS SEQ INTENSITY<BR> ID OF @<BR> NO STAINING @<BR> score 0.301361<BR> C19 : ACGAAGGCAACUUCA-AACAUUUCCUUACGUUCCG-CGCUCA 200<BR> C51 : ACGGCGCCAACAGCG-AAUGUUCGCCC-CGUUCGGACGCUUA 201<BR> Civ45 : ACCGACACAACCACG--ACGUUCGGUC-GGUUUGUCCGAUUA 202 +++ @<BR> Civ63 : ACGGAGGCAACCAAG--AGAUUUCCAU-CGUUCGUUCGAUUGA 203<BR> Civ70 : UCC-AUCCAACGCGGCAAGAUUUGAUG-GACUUUGACGAUCA 204<BR> score 0.326357<BR> C23 : AA-GCU--CAGC-AGAUCGGGACUUCUGAUCUUCGGGUCGCUUA 205 -<BR> Civ67 : CAACGGUAGCGGCUAGAACGCGCCGACUGAU-UUAGG----CUUA 206<BR> score 0.362963<BR> C28 : UCCUCCUG-UUCG-GAGUCUCAAUGUCGACUCGGCCGGACCU 207<BR> Civ12 : AGA-AAUCCCCUUGAUUCG-GAGUCGUCUUUUCGAG-CGUAGU 208 + @<BR> Civ26 : AGA-AAUCCCCUUGAUUCG-CAGUCGUCUUUUCGAG-CGAAGG 209<BR> Civ37 : GAGAGUCAAC-UG---CGAGAAUGG-CUUUCCCAA-CGGCACCUUU 210 +++ @<BR> Civ40 : AGAUAAUCCCCCGGAUUCG-GAGUCCUCUUGACGAA-CUUCC 211<BR> score 0.356944<BR> C29 : C-GGA-ACAAACGGAAAUGGCACACAGG-AGAAAGACGAGACC 212<BR> C34 : CAGGAGAUUAA-GGAACAGGC-CACAGAUAGAGACACG-GAGC 213<BR> score 0.426357<BR> C31 : AACUGGACGAGAGGAGCUAGCGUCCAAGUUCGGAGCUA 214 + @<BR> Civ44 : ACUGAUUCUCAGCGGCUAGCGCUGAAGUUCG-A-CUAGUUCA 215<BR> score 0.410853<BR> C39 : GGCCACAAGCAG-AGAACAGAACAA-CAGAGCGAUGGAG-AGA 216<BR> C50 : GGAG-CAUCCAGGAUAACAGGCUAAACACCGCAA-GGACCAG 217<BR> score 0.333333<BR> C46 : CGGAGGAAGGAAGAGG--AACCUUCGC-CUCUGAUUUAGCUUA 218<BR> C68 : CGUGGGCAAACUGAGG-CAUUCCCCGCGCUCAGAGAUUCAU 219<BR> C71 : CAAUGGCAACUAGGCCACAAAGUUC-C-CACUGAUUCGACGU 220<BR> Civ19 : GCAAUCGGACCGAAAGG----CCUUACC-GAUUUCUCGACCUUUC 221 +++ @<BR> Civ43 : CGGAGGAAGGAAGAGG--AACCUUUGC-CUCUGAUUUAGCUUA 222<BR> Civ47 : CGGAGGAAGGAAGAGG--AACCUUCGC-CUCUGAUUUAGCUUA 223<BR> Civ57 : AGUCGAGUUUCAAGG--AUCAUCCCC-CUCUUCGGAGCCUUUC 224<BR> Civ58 : CGGAGGAAGGAAGAGG--AGCCUUCGC-CUCUGAUUUAGCUUA 225 Table 8 (cont'd)<BR> SEQ INTENSITY<BR> FINAL CLUSTERS ID OF BINDI@<BR> NO STAINING SPECI@<BR> score 0.347884<BR> C56 : CGAGGCCACCGACAAGGAAGUCG---ACCG-GAGUUGAAGUAAA 226<BR> Civ39 : GGCC-CCUAGCGGGAUGCCGCUAAUCGCGAAUCGAGGUUUA 227<BR> score 0.348485<BR> Civ7 : UCGAUGCU--AUC-GAGUUCU-ACUCGGAAGGUUC-AACGUUUAA 228 ++ @<BR> Civ9 : CCCAUACU-GAGA-AAGAACAGACUUCUCAGGUUCGAACGU 229<BR> Civ13 : CCUGAGACG-GUAC-GAGUUCGGACUC---AGGAUUUAACGCUUU 230<BR> Civ15 : CUUACUCAACCU-GCGA-ACGCACAG-GUU---AG-UUC-ACCGUUUA 231<BR> Civ17 : ACCCACACU-GAGA-AAGAACAGACUCCACAGGUUCGAACGU 232<BR> Civ36 : AAACUCAUUCU-GAGCUAAGCUCA-AGUUC-----UUGCAACGUUUG 233<BR> Civ49 : ACCGAUUCUCGAAG-CAGCACG--CUCC--AGG-UCUGACGUUUU 234<BR> Civ66 : ACCUAUACU-GAGA-AAGAACAGACUUCUCAGGUUCGAACGU 235<BR> score 0.316084<BR> Civ8 : AGGAACUUAUUCGAC---AUCAG-UCGGUUCCCUGGACGGGUUG 236<BR> Civ51 : GAACCUAUUCAACCGGAUUAGGUUGGUUCUC-GGAUGUCUA 237<BR> score 0.439394 Table 9 Carotid Truncates and Putative Structures A G A UC C C u g a C C U G U U U G A G C U u c c g g a c a a a c u c g g a g a c g C12.1 (C33 truncate) (SEQ ID NO: 238) G A U C C A UCC G U U U GCUUuucgaA gga caaacucggaggac c (Civ41 truncate) (SEQ ID NO: 239) GUC C G G G UCG U A <BR> <BR> U U C G U U U G AUUuuc?<BR> <BR> <BR> <BR> g g a c a a a c u g g a g g a c c c (Civ45 truncate) (SEQ ID NO: 240) G G U C U U U G u uc gggacagCg a (C59 truncate) (SEQ ID NO: 241) Table 10<BR> CLONE COPIES SEOUENCE OF RANDOM REGION SEQ<BR> IN N@<BR> 12.2 8 GCGGGAUUUUCCUGAUCAUCCCACUGAUUCGGGGCCUUAC 242<BR> 12.10 8 GGAGCGAUAUUCUUGAACAUCCACUGAUUCGGAGCGUCCU 243<BR> 12.11 4 CGCUGACUUGUUUAUUCCCACUGGUUCGGAGUCUUGCUGU 244<BR> 12.18 1 ACCGUCUUCUUACAUCCCACUGAUUUGGAGUCUCGUUGUA 245<BR> 12.13 4 CACUGAAGCAAAGUUACCUUUCUCAGAUGU UUAGCGCCUA 246<BR> 12.8 3 CGGAGGAAGGAAGAGGAACCUUCGCCUCUGAU UUAGC UUA 267<BR> 12.15 3 CGAAGGACGCGAUGCAAUCGGCCUUUGAAU UUAGCGUUCA 248<BR> 12.9 2 CUGAAGGAUGCAGCCAAGGCCGCCCUUCUAGU UUAGC UUA 249<BR> 12.37 2 CUAAGAACGAAGCCACCGUUUCGCCUUAGUAUUAGCC UUA 250<BR> 12.42 2 CUUUAGUUGUCGCAGGAGCCAAACUAAAGUAUUAGCU UUA 251<BR> 12.17 1 CAACGGUAGCGGCCAGAACGCGCCGACUGAU UUAGGC UUA 252<BR> 12.23 1 UGCGAAGUACAAGACAAACUUGUUUCGACAU UUAGCA UUG 253<BR> 12.5 1 CGCCGUCACGAGGAUGACCUCAGCGGAAGGU UUAACG AUA 254<BR> 12.12 1 CGCAGAUAACGGUCUUUCCGUGUCUGAUGU UUAACGAAGA 255<BR> 12.36 1 CGCGGAACAAUAGAGGAUUGGGUCGGCGAU UUAACGUUCA 256<BR> 12.46 1 CACUAGAUGGGAGAUCCUCAGGCUAGGUGU UUAGG UUCA 257<BR> 12.31 1 AUCCAGAUUGGAAGCAAUCCUGCUGGAUUAUUAGCC UUA 258<BR> 5'-gggagauaagaauaaacgcucaa-[40N]-uucgacaggaggcucacaacaggc-3' 40 Table 11<BR> WHHL SELEX Conditions<BR> In Vitro/ Incubation Wash<BR> Round# In Vivo RNA Time, min Time, mi@<BR> 1 In Vitro 0.54 uM 45 4 x 1ml, 15<BR> 2 In Vitor 1.6 uM 55 2 x 3ml, 15<BR> 1 x 50 ml, 20<BR> 3 In Vitro 1.47 uM 60 3 x 125 ml, 2@<BR> 4 In Vivo 18.3 uM 15<BR> 5 In Vivo 5.0 uM 30<BR> 6 In Vivo 13.3 uM 50<BR> 7 In Vivo 7.0 uM 65<BR> 8 In Vivo N.A. 60<BR> 9 In Vitro 0.50 uM 55 4 x 40 ml, 30<BR> 10 In Vitro 0.27 uM 45 4 x 40 ml, 30 Table 12<BR> Round 10 Clones from Watanabe Heritable Hyperlipidemic Rabbit Atheroscle<BR> plaque SELEX<BR> Family I<BR> 10.5 gggagacaagaauaaacgcucaa GCAACCUCGGACUAGACUAACGACCU GUUUGACACUUA uucgacaggaggcucacaacaggc<BR> 10.1 gggagacaagaauaaacgcucaa UCAAUCUCGGACUAGACUAACGACUUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.12 gggagacaagaauaaacgcucaa GCAAUCUCGGACUAGACUAACGACCUU GUUUGACGCUGA uucgacaggaggcucacaacaggc<BR> 10.14 agacaagaauaaacgcuca ACAAUCUCGGACUAGACUAACGACCUUGGUUUGACGCUCA uucgacaggaggcucacaacaggc<BR> 10.19 gggagacaagaauaaacgcucaa GCAACCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.23 gggagacaagaauaaacgcucaa CCANNCUCGGACUANACUACCCAUCUUCGUUNNACCCUUA uucgacaggaggcucacaacaggc<BR> 10.36 gggagacaagaauaaacgcucaa GCAAUCUCGGACUAGACUAACGA CUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.37 gggacaanaauaaacgcucaa GCAAUCUCGGAUUAGACUAACKA CUUCGUUUGACACUUA uucgacaggaggcucacaacaggc<BR> 10.31 gggagacaagaauaaacgcucaa UCAAUCUCGGACUAGACUAACGACCUUGGUU GACGCUCA uucgacaggaggcucacaacaggc<BR> 10.49 gggagacaagaauaaacgcucaa GCAACCUCGGACUAGACUAACGACCUU GUUUGGCACUUA uucgacaggaggcucacaacaggc<BR> 10.38 gggagacaagaauaacgcucaa GCAAUCUCGGACUAAACUAACGACUUUCGUUCGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.53 gggagacaagaauaacgcucaa GCAACCUCGGAUCAGACUAACGACCUUGGUU GACGCUUA uucgacaggaggcucacaacaggc<BR> 10.55 gggagacaagaauaaacncucaa GCAACCUCGGAUUANACUAACGACCUUCNUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.56 gggagacaagaauaaacgcucaa GCAAUCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.64 gggagacaagaauaaacgcucaa GCAAUCUCGGACUARACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.65 gggagcaagaauaaacgcucaa GCAACCUCGGACUAAACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.68 gggagacaagaauaaacgcucaa GCAAACCUCGGACUAAACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.72 gggagacaakaauaacgcucaa GCKAUCCCGGUACUANACUAACGACCUUGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.73 gggagacaagaauaaacucucaa GCAAUCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.76 gggagacaagaauaaacgcucaa UCAAUCUCGGACUAGACUAACGACUUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.77 gggagacaagaauaaacgcucaa KCAAUCUCGGACUAAACUAACKACUUCBUUUGACGCUUA uucnacaggaggcucacaacaggc<BR> 10.81 gggagauaagaauaaacgcucaa KCAAUCUCGGACUAAACUAACGACCUUGUUUGACSCUUA uucgacaggaggcucacaucaggc<BR> 10.84 gggagacaagaauaaacgcucaa GCAAACCUCGGACUAAACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacangc<BR> 10.95 gggagacaagaauaaacgcucaa GCGAUCUCGGACUAACUAACGACUUCGUUUGACGCUCA uucgacaggaggcucacaacaugc<BR> 10.99 gggagacaagaauaaacgcucaa GCAAUCUCGGACUANACUAACNCYUUCGUUUAACCCCC uuncaanngaaggncccnannnggc<BR> 10.100 gggagacaaaaauaagccucaa GCAACCUCGGACUAAACUAACGACCUUCGUUUCACGCUUA uucgacaggaggcucacaacaggc<BR> 10.101 gggagacaagaauaacgcucaa GCAAUCUCGGACUANACUAACGACCUUCGUUUGACGCUUA uucgacaggaggcucacaacaggc<BR> 10.104 gggagacaagaauaaacgcucaa GCAACCUCGGACUAGACUAACNACCUUGUUUGACGCUA uucgacaggaggcucacaacaggc (cont'd)SEQ.Table12 IDFamilyI NOS.

CGACCCUGGCUCUAGAGUUCUCGUAUUCGCUGUCGCAUGAuucgacaggaggcucacaac aggc28810.70gggagacaagaauaaacgcucaa CGCACAUGACCAGGCGCUACUGACUGARAUGUUGAACUUAuucgacaggaggcucacaac aggc28910.86gggagacaagaauaaacgcucaa CGACCCUGGCUCUAAARUUCUCGUAUUCGCUGUCGCCUGAuucgacaggaggcucacaac aggc29010.97gggagacaagaauaaacgcucaa CUACCCAMGCCUCAGAGUUCUCGUAUUCGCUGUCGCAUGAuucgacaggaggcucacaac aggc29110.45gggagacaagaauaaacgcucaa CGACCCUGGCCCUARAGUUCUCGUAUUCSCUGUCGCAUGAuucgacaggaggcucacaac aggc29210.78gggagacaagaauaaacgcucaa <BR> <BR> CGACCCUGGCUCUARAGUUCUCGUAUUCGCUGUCGCAUGAuucgacaggaggcucacaac aggc29310.79gggagacaagaauaaacgcucaa CNACCCUGGCUCUAGAGUUCUCGUAUUCGCUGUCGCAUGAuucgacaggaggcucacaac aggc29410.108gggagacaagaauaaacgcucaa Orphans AACUCGUUCAGCAUCUACAAACUACUACAUUGAAGCACCCuucgacaggaggcucacaac aggc29510.41gggagacaagaauaaacgcucaa uucgacaggaggcucacaacaggc29610.46AAGACACCWCWCCAAUUCUAAUAUACAG CACUACAUGGCC AAGUACGGUGUCUACCAAACCGCAUCCAUCACCGAUAuucgacaggaggcucacaacagg c29710.83gggagacaagaauaaacgcucaa AAUUUCACCGAGGCAGACGAACAUCAAKUGCAGCUGCUAAuucgacaggaggcucacaac aggc29810.30gggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> <BR> ACAAAUUCGUACWCACUUCKCGGAGGCMGGGGUAYAUCuucgacaggaggcucacaacag gc29910.90gggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> 10.54 uucgacaggaggcucacaacaggc300ACACUCAUAGAGUCCAAUUCAACGCGCCAUUAU CUCAARA 10.8010.80gggagacaagaauaaacgcucaa 301uucgacaggaggcucacaacaggc <BR> <BR> <BR> AGAGCCAUCCCUUAAACGCAGCUACGGAGGUCGUUCUUAUuucgacaggaggcucacaac aggc30210.57uggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> <BR> AKCAAACAUUCUCAUAUAAAKAGCCAAAAUAGACAACCACuucgacaggaggcucacaac aggc30310.75gggagacaagaauaaacgcucaa <BR> <BR> <BR> AGGCAUGCASCAAAAGCACACCGAAAUAUCUCCAAUCCCUuucgacaggaggcucacaac aggc30410.59uggagacaagaauaaacgcucaa AGGCAUGAGUCCGCACAUAUGUGAACGUGAGCACAUAAAUuucgacaggaggcucacaac aggc30510.47gggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> AGUCGUCGCGCAUAUUGACACAAUAAUCGUGUCAUAACCUuucgacaggaggcucacaac aggc30610.87gggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> AUACUCGCGAUCGCAGCAUCAUCCCGCCUCCUUGAAACACuucgacaggaggcucacaac aggc30710.71gggagacaagaauaaacgcucaa <BR> <BR> <BR> <BR> AUAGGAUGCCGCAUCGGUGUACACUCACACGCUAACCAAUuucgacaggaggcucacaac aggc30810.16gggagacaagaauaaacgcucaa <BR> 10.33uucgacaggaggcucacaacaggc309AUCCUUUAACGAUAGCACAAUAGAAUUA GACUCCAACAAG AUGUCGACAUACAUAAAAUACGUCAAGUCUUGCGAUCGAUuucgacaggaggcucacaac aggc31010.88gggagacaagaauaaacgcucaa CAAGCCAUGCGUUAGUGCCCCUUCCAGUUCCCAuucgacaggaggcucacaacaggc311 10.4gggagacaagaauaaacgcucaa CAACCCGGGCUCNANANUUCUCGUAUCCCCUGUCNCNUUAuucgacaggaggcucacaac aggc31210.96gggagacaagaauaaacgcucaa CAUGAAAAACUUUCUAUUCCGUAAAUCACGUGAGUCGCUAGuucgacaggaggcucacaa caggc31310.22gggagacaagaauaaacgcucaa CAUUGGCAGYWAABCMUCCAYCCAUARACRCUCAACAAGUuucgacaggaggcucacaac aggc31410.62gggagacaagaauaaacgcucaa CCAAGGAGCUUCCUGGCACCGUACCUACAUCGACCUGUUAuucnaacaggaaggcucaca acaggc31510.13gggagacaagaauaaacgcucaa CCAWYCUCUSAYCAGACUAACCACCYYGUAUGAMCCUYAuuccaamggaagsycmcaaca rggv31610.51gggagacaagaauaaacgcucaa CCCAAUAUCAGUUGUCUCAAACUGCGCUACUUCAUGCUUGuucgacaggaggcucacaac aggc31710.25agacaagaauaaacgcucaa CCCCAAACCUACCCCUCUAUAUUCCUUCUCGUAACCCAAuucgacaggaggcucacaaca ggc31810.50agacaagaauaaacgcucaa <BR> <BR> <BR> <BR> <BR> <BR> CCCCUCYYRGACUCCAUCAACAAMUCCCCYCCUGACUCGuucgacaggaggcucacaaca ggc31910.89gggagacaagaauaaacgcucaa <BR> 10.103 gggagacaagaauaaacgcucaa MCCGGUGCAUGUGCUAWAAUAACAUUCGGCCAUUAUCCMA uucuacaggaggcucanaacungc 320 10.27 gggagacaagaauaaacgcucaa CCGUCCUCASAYYCCACUWACGUCCAUKWAUGAAAUWCAC -uccaacaggaggcucacaacaggc 321 10.24 gggagacaagaauaaacgcucaa CCUAUCGUCGUCCUACAUAGAGCUAUCCCUCGCUUCGC uucgacaggaggcucacaacaggc 322 Table 12 (cont'd)<BR> Orphans (Continued)<BR> 10.82 gggagacaagaauaaacgcucaa RCAAUCUCSGAAWMUCACUACCCACUCCCKUCUGACNCUCA uucgacaggaggcucacaacaggc<BR> 10.18 gggagacaagaauaaacgcucaa GCAUCAGAACUCAAUUUCUUCAGCCAGCGUGCUUACGACA uucgacaggaggcucacaacaggc<BR> 10.40 gggagacaagaauaaacgcucaa GCAUUCACAGUCGAUCCAAUUCCGGAACCUUAUCCAACAA uucgacaggaggcucacaacaggc<BR> 10.8 gggagacaagaauaaacgcucaa GCCCCUUCUGGACGCCGCUAA GAUCUCCCCCAACCCGAUA uucgacaggaggcucacaacaggc<BR> 10.107 gggagacaagaauaaacgcucaa GCGGGAUUUUCCNGAUCAUCCCACUGAUUCGGGGCCUUAC uucgacaggaggcucacaacaggc<BR> 10.102 gggagacaagaauaaacgcucaa GCGUAAUGAGCUGUGCCCGCGUUUAUGAUCUAUAUCCUAA uucgacaggaggcucacaacaggc<BR> 10.17 gggagacaagaauaaacgcucaa GCUAGUCAUCUGCCUAACUCCUCACGAGACCCAGCCGUAC uu...<BR> <P>10.35 gggagacaagaauaaacgcucaa GACAUUAUUUGACAUCUAGGUGUAGCAAGUAUAGACCUAA uucgacaggaggcucacaacaggc<BR> 10.92 gggagacaagaauaaacgcucaa GGAUGAACAAUACCCCAUGCAGUCCAAGUCCUGCUUUCAC uuCGACAGGAGGCUCACAACAGGC<BR> 10.39 gggagacaagaauaaacgcucaa GUGUACACRAAGUCAGUUAGCGGACAGUUUGCGCAGCCCGU uucgacaggaggcucacaacaggc<BR> 10.3 gggagacaaaaauaaacgcucaa GUsCaGGUsUGACCACCUGwAsgACCCUGCCCACCCGUCA uucgacaggaggcucacaacaggc<BR> 10.94 gggagacaagaauaaacgcucaa GUUGCCCUCGCGAAUGUACCCAUCGUACAGAACGGUCUAA uucgacaggaggcucacaacaggc<BR> 10.85 gggagacaagaauaaacgcucaa UAACACCCACCACUCRCGCAUGGCAUUGUCVCCAAAUAAC uucgacaggaggcucacaacaggc<BR> 10.67 gggagacaagaauaaacgcucaa UACCGGGCGAKCUACCCUGUACUGCCUCUUCCUUCACAGC uucgacaggaggcucacaacaggc<BR> 10.6 gggagacaagaauaaacgcucaa UACGAACAAAGGAACUCAGUCAAAAACAGCAGUGUACCA uucgacaggaggcucacaacaggc<BR> 10.26 gggagacaagaauaaacgcucaa UCAAC uucgacaggaggcucacaacaggc<BR> 10.11 gggagacaagaauaaacgcucaa UCACGGUAAGGACCCAGAUUCUCCUCUUCCCAACCUCGCA uucgacaggagg<BR> 10.20 gggagacaagaauaaacgcucaa UCAUGCUCACGACUAUUGCCAUAACGCUAUCCACACAACA uucgacaggaggcucacaacaggc<BR> 10.9 gggagacaagaauaaacgcucaa UCCAUGCAUGAUCGUAAGCUAUAAAGGUGCGCAAACCUUG uucgacaggaggcucacaacaggc<BR> 10.52 gggagacaagaauaaacgcucaa UCGGUCUCAACCUACCCACCCCAGUAAGCAUGAGGCUACA uucgacaggaggcucacaacaggc<BR> 10.21 gggagacaagaauaaacgcucaa UCGGCAAAGUCCUCUUAUAUACAGUCUGCAAUCACUCAUU uucgacaggaggcucacaacaggc<BR> 10.106 gggagacaagaauaaacgcucaa UCUAUGCAACAACGANCAUUCACUCCANCUAGGUWCAGG uucgacaggaggcucacaacaggc<BR> 10.10 gggagacaagaauaaacgcucaa UGCCCGGUACAAAUCUUUCUACACCACGAUCUCCCCUAUC uucgacaggaggcucacaacaggc<BR> 10.74 gggagacaagaauaaacgcucaa UGGGAUUGGGUCUUACACGUUCACUCGCUUAUCCUCCCAA uucgacaggaggcucacaacaggc<BR> 10.2 gggagacaagaauaaacgcucaa UGUUACAAAACACAGACGCCGCGCUAAGGUAUAUCCGCUG uucgacaggaggcucacaacaggc<BR> 10.44 gggagacaagaauaaacgcucaa UGUUAAUCAACUACAAUGCACUUGAGCCAAACAACCGACU uucgacaggaggcucacaacaggc<BR> 10.15 gggagacaagaauaaacgcucaa UUCCAUUGGUUUGCCCUAUCGAGACCCGUCCGCUGUUCCU uucgacaggaggcucacaacaggc<BR> 10.48 ggagacaagaauaaacgcucaa UUUCUUGUGAUGUCAGAAUGACCCUAGACUUACGUCCAAA uucgacaggaggcucacaacaggc<BR> 10.7 gggagacaagaauaaacgcucaa UUGACGCACCCAUCGGCCCGUCCACUGCUCCCCACCUAGU uucgacaggaggcucacaacaggc<BR> 10.28 gggagacaagaauaaacgcucaa CRGUAUCAUGAACUCCCACGAACRCCACUGUUUUAAUU uucgacaggaggcucacaacaggc<BR> 10.43 gggagacaagaauaaacgcucaa GCGGGAUUUUCCUGAUCAUCCCACUGAWUCGGGGCCUUAC uucgacaggaggcucacaacaggc<BR> 10.93 gggagacaagaauaaacgcucaa NAGUGGAUAACGUAUAGCCAAUUUUCUCACUCGCCUCGUU uucgacaggaggcucacaacaggc<BR> 10.98 gggagacaaaaauaacgcucaa RAAAACCUACCUUCGUACAUUGGAUARAAAAACGGCUCUU uucgacaggaggcucacaacaggc Table 13<BR> Sequences from Round 5 Human Artery Perfusion Selex<BR> Family I from Watanbe Heritable Hyperlipidemic Rabbit Atherosclerotic Plaque SELEX<BR> 81 GGGAGACAAGAAUAAACGCUCAA AAACAACCUCGGACUAGACUAACGACCUUGUUCGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 5 GGGAGACAAGAAUAAACGCUCAA GCAAUCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 9 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGACUAGACUAACGACCUUAGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 14 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 37 GGGAGACAAGAAUAAACGCUCAA GCAAUCUCGGACUAGACUAACGACCUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 45 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGAUUAGACUAACGACCUUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 54 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGACUAGACUAACGACCUUGUUCGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 68 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGACUAGGCUAACGACCUUGUUUGACACUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 69 GGGAGACAAGAAUAAACGCUCAA GCAACCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 71 GGGAGACAAGAAUAAACGCUCAA GCAAUCUCGGACUAGACUAACGACUUCGCUUGACGCUCA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 78 GGGAGACAAGAAUAAACGCUCAA GCAAUCUCGGACUAGACUAACGACCUUGUUUGACGCCUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 95 GGGAGACAAGAAUAAACGCUCAA GCAAUCUCGGACUAGACUAACGACCUUCGUUUGACGCUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> Family I from Watanbe Heritable Hyperlipidemic Rabbit Atherosclerotic Plaque SELEX<BR> 43 GGGAGACAAGAAUAAACGCUCAA CGACCCUGGCUCUAGAGUUCUCGUAUUCGCUGUCGCAUGA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> Family I from Round 5 Human Perfusion Selex<BR> 1 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUACAGAAGGACUAGAACCCCAAAAGCGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 11 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUACAGACGGACUAGAACCCCAAAAGCGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 13 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUACAGACGGACUAGAACCCCAAAAGCGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 35 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUAUAGACGGACUAGAACCCCAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 41 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUACAGACGGACUAGAACCCCAAAAGCGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 62 GGGAGACAAGAAUAAACGCUCAA GGCUAGAUCUACAGACGGACUAGAACCCCAAAAGCGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 72 GGGAGACAAGAAUAAACGCUCAA GGNUAGAUCUACAGACAGACUAGAACCCCAAAAGUGACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> Sequences Present in Round 5 Human Artery Perfusion Selex in More Than One Copy Numbe<BR> 25 GGGAGACAAGAAUAAACGCUCAA CGUUGAGAGCAACCUGUCGAUCCCGGAGCAGACUAACGAU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 40 GGGAGACAAGAAUAAACGCUCAA CGUUCAGAGCAACCUGUCGAUCCCGGAGCAGACUAACGAU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 23 GGGAGACAAGAAUAAACGCUCAA GUGUUGGAGCUCUUGAUUGGAAAAGUAGAGCAAAUCGAAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 85 GGGAGACAAGAAUAAACGCUCAA GUGUUGGAGCUCUUGAUUGGAAAAGUAGAGCAAAUCGAAA UUCGACAGGAGGCUCACAACAGGCAAGCU Table 13 (cont'd)<BR> Orphans<BR> 38 GGGAGACAAGAAUAAACGCUCAA AAAAUCGUCACCCCUCCGGUCCUCACAUGACAGCAUGAACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 86 GGGAGACAAGAAUAAACGCUCAA AAAGCUAGAUCAGCAGUGAACGACUACAAGUGCAUAGUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 19 GGGAGACAAGAAUAAACGCUCAA AACCGGAGAGCCCGAACCACCGGUAGCAUCCGCAUCAUAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 29 GGGAGACAAGAAUAAACGCUCAA AAUAUAGCCCUGCGAUCUUAGCCCAACUUCCUCAAAGCUU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 82 GGGAGACAAGAAUAAACGCUCAA ACCAGAAAAGAAGCUCAAAACCUUUGCUUGAUCGACACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 91 GGGAGACAAGAAUAAACGCUCAA ACCGAUCGAUAUGACUCGACAUGUCGAUGCACAAAGUAAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 76 GGGAGACAAGAAUAAACGCUCAA ACCUUAUACCAUCCUGUCUCAACCAUACUCUGAUACACAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 77 GGGAGACAAGAAUAAACGCUCAA AGAUGAAUAAUGACCCACAACUGACCCAGCGAUACUAUAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 65 GGGAGACAAGAAUAAACGCUCAA AGCUGCCAGACACAAUCCGGUGGCAGUCCGAUAAAUACAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 12 GGGAGACAAGAAUAAACGCUCAA AGGCCCAAGAUGUACACACGGUCACGUCCUACAUACUACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 79 GGGAGACAAGAAUAAACGCUCAA AGUCCACGCUGCGUGUGCUCCAGCAUACGACUCUUAAGCU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 22 GGGAGACAAGAAUAAACGCUCAA AGUGUGAACUCCUAAACCCCUCCGGACAGAUAACACGGAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 3 GGGAGACAAGAAUAAACGCUCAA AUCUCGACCUCGGUCGGCCCUUCCCGAAGCCGUGUAUAUC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 16 GGGAGACAAGAAUAAACGCUCAA AUGAAGCUGAUGCACCAUUAUCAACACCACCCUACGUUAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 20 GGGAGACAAGAAUAAACGCUCAA AUUAUGGAUAUACGAGACCCACCCUCCUCUCUAGCGUACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 2 GGGAGACAAGAAUAAACGCUCAA CAACUAAUACCGCUAAAAGACUGCAGCCUCAGUACACAAA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 15 GGGAGACAAGAAUAAACGCUCAA CAACAUACUCAACUACCACGAAGCAAACUGCGUAAACCAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 34 GGGAGACAAGAAUAAACGCUCAA CACACCCUAAAUCACCAUCCACUGGCCGUCACCAAUAAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 28 GGGAGACAAGAAUAAACGCUCAA CCCCAAUGUACGAGCCAGUACCAAGCCACCACGAUAUGU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 73 GGGAGACAAGAAUAAACGCUCAA CCCCGUAUCUCUUCGACAGCCCCCUCUUCCUCCAACCACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 80 GGGAGACAAGAAUAAACGCUCAA CUCAACUGGAAACCACGGGAUGACAACCGUCCAUUGCAAU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 6 GGGAGACAAGAAUAAACGCUCAA GACCUAUUUCAACCUGUGCCUGAUCACCUAAAGUUUGCCU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 83 GGGAGACAAGAAUAAACGCUCAA GAGACUCAUUAAGCGCCCGCCGUUGAACGUCACCCCUAUC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 70 GGGAGACAAGAAUAAACGCUCAA GCACCCCAGCAAAAAUCCGAUCCAAACCACACUCCCAAAAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 61 GGGAGACAAGAAUAAACGCUCAA GGAUCUUACUCAGCCCCUGUUUCAACAAUCCAUGCUCCAG UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 89 GGGAGACAAGAAUAAACGCUCAA GGNUNUGCACACAUAGCCAACCNGACCNUUGNUUAAUUCA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 74 GGGAGACAAGAAUAAACGCUCAA GGGAUCAAAAUCCAAACGCGUUAACCGUUAAUACACUUAU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 10 GGGAGACAAGAAUAAACGCUCAA GUCCUGACUACUAACAUUGCCCGUGACCCAUUGCCUUAC UUCGACAGGAGGCUCACAACAGGCAAGCU Table 13 (cont'd)<BR> 92 GGGAGACAAGAAUAAACGCUCAA UAACACAAUGUAAGUCCUUCGAUCACACCUAAUUAGAUCA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 32 GGGAGACAAGAAUAAACGCUCAA UACCCCACACUCCUGAUCACCCCCAUUACUUUCUAUAUAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 55 GGGAGACAAGAAUAAACGCUCAA UAGGGACUAACGCUGUGUGCUACAGGCCCCCCAAACAUCA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 49 GGGAGACAAGAAUAAACGCUCAA UCAAACAGCCUGGAUACCUCUCUCCCUAUCCCCUCACUUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 87 GGGAGACAAGAAUAAACGCUCAA UCGAUACUAGAUCCUAUUGCAGACGUAACGUUGCUUUAAG UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 96 GGGAGACAAGAAUAAACGCUCAA UCGGCCAGCAUCAAGGACAUCACUUACACCUAGUACCUAC UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 90 GGGAGACAAGAAUAAACGCUCAA UCUACCACACGCUUCCCGAACGACCUCCCAAUUAACUCGA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 93 GGGAGACAAGAAUAAACGCUCAA UCUCACAGUUGAAGUAAUCACCAUCGCCAUACAAACUA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 7 GGGAGACAAGAAUAAACGCUCAA UGAUAGAAGCCAAAAGCGCCGUUUGCGACGAUCACCUUAU UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 17 GGGAGACAAGAAUAAACGCUCAA UGAUGUGCCGCGGCCCAACCACAAUAAUCGCACUCUUACA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 94 GGGAGACAAGAAUAAACGCUCAA UUAGCAACCAAGCAUCCGUUAAUAAGCGGAAAAGACACGA UUCGACAGGAGGCUCACAACAGGCAAGCU<BR> 66 GGGAGACAAGAAUAAACGCUCAA NGNUAACGCUUUACCUCUCCCAANCNUCAAACAGGAAUUA UUCGACAGGAGGCUCACAACAGGCAAGCU