WO1999027133A1 - Improved selex procedure and an anti-cd4 aptamer - Google Patents

Abstract

The invention provides an improved SELEX method for isolating a nucleic acid capable of binding to a target from a library of single stranded nucleic acid members, by selecting library members according to their dissociation kinetics.

Description

IMPROVED SELEX PROCEDURE AND AN ANTI-CD4 APTAMER

The present invention relates to an improved method for the in vitro evolution of nucleic acids in order to isolate nucleic acids which display highly specific binding to any desired molecule or compound. More particularly, the invention relates to an improvement of the Systematic Evolution of Ligands by Exponential enrichment (SELEX) procedure.

SELEX is a method for the in vitro evolution of nucleic acid molecules with highly specific binding to target molecules. It is described, for example, in U.S. patents 5654151, 5503978, 5567588 and 5270163, as well as PCT publication WO 96/38579, each of which is specifically incorporated herein by reference.

The SELEX method involves selection of nucleic acid aptamers, single-stranded nucleic acids capable of binding to a desired target, from a library of oligonucleotides. Starting from a library of nucleic acids, preferably comprising a segment of randomised sequence, the SELEX method includes steps of contacting the library with the target under conditions favourable for binding, partitioning unbound nucleic acids from those nucleic acids which have bound specifically to target molecules, dissociating the nucleic acid-target complexes, amplifying the nucleic acids dissociated from the nucleic acid-target complexes to yield a ligand-enriched library of nucleic acids, then reiterating the steps of binding, partitioning, dissociating and amplifying through as many cycles as desired to yield highly specific, high affinity nucleic acid ligands to the target molecule.

SELEX is based on the principle that within a nucleic acid library containing a large number of possible sequences and structures there is a wide range of binding affinities for a given target. A nucleic acid library comprising, for example a 20 nucleotide randomised segment can have 4 structural possibilities. Those which have the higher affinity constants for the target are considered to be most likely to bind. The process of partitioning, dissociation and amplification generates a second nucleic acid library, enriched for the higher binding affinity candidates. Additional rounds of selection progressively favour the best ligands until the resulting library is predominantly composed of only one or a few sequences. These can then be cloned, sequenced and individually tested for binding affinity as pure ligands.

Cycles of selection and amplification are repeated until a desired goal is achieved. In the most general case, selection/ amplification is continued until no significant improvement in binding strength is achieved on repetition of the cycle. The iterative selection/amplification method is sensitive enough to allow isolation of a single sequence variant in a library containing at least 10¹⁴ sequences. The method could, in principle, be used to sample as many as about 10 18 different nucleic acid species. The nucleic acids of the library preferably include a randomised sequence portion as well as conserved sequences necessary for efficient amplification. Nucleic acid sequence variants can be produced in a number of ways including synthesis of randomised nucleic acid sequences and size selection from randomly cleaved cellular nucleic acids. The variable sequence portion may contain fully or partially random sequence; it may also contain subportions of conserved sequence incorporated with randomised sequence. Sequence variation in test nucleic acids can be introduced or increased by mutagenesis before or during the selection/amplification iterations and by specific modification of cloned aptamers.

A problem which stems from the power of SELEX to enrich for specific binding aptamers amongst up to 10 18 molecules is that it is very often the case that the 10 to 20 cycles of selection needed to achieve this require usually around 8 weeks to complete. Particularly in the medical field, this may be too long; it would be advantageous to be able to isolate therapeutically important aptamers within the lifetime of the patient. It has previously been put forward that the incubation step allowing for aptamer-target binding should be reduced as far as practicable, thus initially selecting those aptamers which have a very high rate of association with (and thus, presumably, affinity for) the target. Summary of the Invention

In a first aspect of the invention, therefore, there is provided a method for isolating a nucleic acid capable of binding to a target from a library of single stranded nucleic acid members by selecting library members according to their dissociation kinetics.

The invention therefore makes use of the rate of dissociation of molecules from the target, rather than their rate of association with it, to select suitable aptamers. It has been found that this permits the more rapid selection of molecules having a higher affinity for the target that the methods of the prior art, which rely on association kinetics. The invention moreover provides aptamers preparable by a method according to the first aspect of the invention.

According to a further aspect of the invention, there is provided an anti-CD4 aptamer preparable by the method of the first aspect of the invention.

Detailed Description of the Invention

SELEX procedures are based on the repeated iterative steps of binding, selecting, eluting and amplifying aptamers. The basic SELEX method has been modified to achieve a number of specific objectives. U.S. 5567588 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. U.S.

5580737 describes a method for identifying highly specific nucleic acid ligands able to discriminate between closely related molecules, termed Counter-SELEX. U.S.

5567588 describes a SELEX-based method which achieves highly efficient partitioning between oligonucleotides having high and low. affinity for a target molecule. U.S.

5,496,938 describes methods for obtaining improved nucleic acid ligands after SELEX has been performed. 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 described, for example, in International patent application WO9507364, that describes oligonucleotides containing nucleotide derivatives chemically modified at the 5- and 2 '-positions of pyrimidines. U.S. 5580737 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). International patent application WO9535102 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 U.S. 5637459 and U.S. 5580737. Each of the above described patent applications which describes modifications of the basic SELEX procedure is specifically incorporated herein by reference. Corresponding published patent documents may be identified in the appropriate databases.

Optionally, further steps of mutation or randomisation may be introduced in between the amplification and binding steps in order to allow for "in vitro evolution" of the aptamers. The particular advantage of SELEX is that the nucleic acid and the biologically effective agent are one and the same, which means that selection according to biological activity allows the isolation of the nucleic acid template which encodes the biological activity.

As used herein, a "library" is a mixture of nucleic acid molecules, referred to as library "members", which are potentially capable of binding to a target. Preferably, the members of the library are randomised in sequence such that a large number of the possible sequence variations are available within the library. For example, each member may comprise 5' and 3' terminal regions flanking one or more randomised regions. The randomised region(s) may be in essence of any length, but a length of up to 100 nucleotides, which may be interspersed with non-randomised insertion(s), is preferred. Typically, the randomised region will be between 20 and 60, preferably 30 to 50 and most preferably about 36 nucleotides in length. The library according to the present invention thus does not differ in substance to libraries commonly employed in SELEX procedures.

Selected members of libraries according to the invention are also referred to herein as "aptamers" and "ligands". Aptamers are single-stranded nucleic acid molecules capable of binding to a target.

The library members may be DNA or RNA molecules, and libraries may be of homogenous or heterogeneous construction. The DNA and/or RNA, but especially RNA, may be modified in order to improve nuclease resistance of the members. For example, known modifications for ribonucleotides include 2'-O-methyl, 2'-fluoro, 2'- NH₂, and 2'-O-allyl. The modified nucleic acids according to the invention may comprise chemical modifications have been made in order to increase the in vivo stability of the aptamer, enhance or mediate the delivery of the ap tamer, or reduce the clearance rate from the body. Examples of such modifications include chemical substitutions at the ribose and/or phosphate and/or base positions of a given RNA sequence. See, for example, WO 92/03568; U.S. 5,118,672; Hobbs et al., (1973) Biochemistry 12:5138; Guschlbauer et al. , (1977) Nucleic Acids Res. 4: 1933; Schibaharu et al., (1987) Nucleic Acids Res. 15:4403; Pieken et al., (1991) Science 253:314, each of which is specifically incorporated herein by reference.

Targets may be any target which may be bound by an aptamer. For example, suitable targets include any polypeptide molecule, the structures of which lend themselves to binding by aptamers; nucleic acids; lipids; carbohydrates and other biological molecules. The method of the invention allows for the rapid selection of nucleic acid ligands or aptamers which bind to the target with high affinity. Conventional SELEX procedures are based on selection for rapid binding to the target, since the incubation times normally used do not permit equilibrium to be achieved for aptamers having slow binding kinetics. For example, US 5,270,163 allows for an incubation time of 33 minutes, of which 3 minutes is at 0°C; Mannironi et al. , (1997) Biochemistry 36:9726- 9734, recommend an incubation time of only 18 minutes; US 5,580,737 allows only 10 minutes. Such times are insufficient for equilibrium to be established except for the fastest-binding and fastest-dissociating aptamers.

In contrast, it has been determined that high affinity aptamers are more characterised by a slow dissociation rate than by a fast association rate and consequently methods have been developed designed specifically to favour the isolation of such molecules. The invention therefore includes the selection of aptamers having slow dissociation kinetics from their targets. This is preferably achieved by the use of either or both of the following approaches.

In a preferred embodiment of the invention, there is provided a method for isolating a nucleic acid capable of binding to a target from a library of single stranded nucleic acid members, comprising the steps of:

a) incubating the library with the target under conditions suitable for binding such that an equilibrium of dissociation from and association with the target is reached for substantially all library members, including those that have the desired slow rates of dissociation;

b) partitioning the library to isolate the members which are bound to the target;

c) washing the isolated bound target/library member complexes in order to remove weakly bound members; d) eluting the bound library members; and

e) amplifying the selected members.

In a further embodiment, there is provided a method for isolating a nucleic acid from a library of single-stranded nucleic acid members, whose dissociation rate from a target lies within a desired range, comprising the steps of:

a) incubating the library with the target under conditions suitable for binding;

b) eluting the members of the library under conditions that enable the fraction remaining bound to be estimated in real time;

c) discarding those members that elute in less than a time corresponding to a mean desired half life;

d) selecting those members that elute in a time corresponding to a mean desired half life, or longer; and

e) amplifying the selected members.

In one embodiment, therefore, the present invention relies on the establishment of equilibrium for substantially all members of the library, especially those having slow dissociation kinetics. Preferably, the library and the target are incubated together for long periods. The actual period required will depend on the target and library, and also on the round of selection; preferably, for example, the first round of selection may involve an incubation of between about 12 hours and about 48 hours, preferably about 24 hours. Advantageously, the first round of selection is at least about 2 to about 4 hours, preferably 2.5 hours. The remaining rounds involve an incubation of at least about 2 to about 4 hours, preferably 2.5 hours, in order to allow the establishment of a full equilibrium. The mean desired half life of aptamers is determined empirically and depends on the nature of the library and the target, as well as the selection round. Using a biosensor apparatus according to the invention, the dissociation curve obtained will result from the summation of all the component dissociation curves contributed by each aptamer actually present. Its shape is expected to change in successive cycles because both the proportion of those RNAs present that bind to the target will increase and because the combination of kinetic selection and mutagenesis will increase the proportion of these that have off-rates slower than any given value. For this reason, the selection threshold chosen (the mean desired half life) will vary from cycle to cycle, as the characteristics of the RNA population change. In addition, the propensity of the target to interact with RNAs will alter the stringency with which selection is performed at each round.

The use of biosensor apparatus enables the adjustment of the parameters of selection to suit the kinetic behaviour of the nucleic acid pool actually being studied. For example, one would want to retain a greater fraction of the pool during an early round (for example, 50% of the specifically bound total; or those dissociating after 5 min of elution) than a later round (for example, 1 % of the specifically bound total; or those dissociating after 60min of elution), depending on the ease with which high affinity aptamers are evolved to a particular target. Preferably, the aptamers of the final desired affinity will have a mean desired half life of about 30 minutes.

In general, because the observed dissociation curve for the library is the geometric mean of the component dissociations, each selection seeks to produce a next-generation curve whose mean half life more closely resembles that of the desired mean half life. The half life is shifted in the appropriate direction, by taking only those aptamers which dissociate after a desired time.

In the first round of selection, for example, the mean desired half life may be less than about 20 minutes, preferably between about 2 minutes and about 10 minutes and most preferably is about 5 minutes. In subsequent rounds, the mean desired half life may be increased, in order to select for slower-dissociating aptamers. For example, the mean desired half life may range from between about 10 minutes to about three hours, preferably between about 20 minutes and 1 hour, and most preferably is about 30 minutes.

In a further embodiment, independently of whether or not an equilibrium of binding of aptamer library to target is reached, the method aims to select those library members whose rate of dissociation from the target molecule is slow. Thus, in the initial rounds of selection, most of the eluted library members, which have a fast off-rate, are discarded. Only the very few having a slow off-rate are retained. The extension of initial incubation times to allow full equilibrium to be reached because aptamers with higher binding affinity to be selected, and for aptamers with useful affinity to appear at a relatively early round of selection. These aptamers, it has been determined, are those with the slower off-rates.

Kinetics of dissociation are conveniently assessed by observation of the binding and release of aptamers in real time using a biosensor apparatus. Advantageously, an apparatus such as an instrument as marketed by BIAcore^® may be employed. Such instruments use a flow cell on which target molecules are immobilised. Beams of light are focused on the target molecules, and changes in plasmon resonance caused by binding and release of aptamers detected in real time by an optical detector. The signal is processed by computer and the association and dissociation of aptamers to target molecules displayed, for example as a trace on an X-Y plot.

This permits the visualisation of fast-dissociating aptamers dissociating from the target molecules, followed by the slow-dissociating aptamers. In a preferred embodiment, the biosensor apparatus is used directly to select for the slowly dissociating fraction of a nucleic acid library. The use of a biosensor enables one to adjust the parameters of selection to suit the kinetic behaviour of the nucleic acid pool actually being studied. For example, it is desirable to retain a greater fraction of the pool during an early round (for example, 50% of the specifically bound total; or those dissociating after 5 minutes of elution) than a later round (for example, 1 % of the specifically bound total; or those dissociating after 60min elution), depending on the ease with which high affinity aptamers are evolved to a particular target.

This direct selection on the basis of dissociation rate has been determined to allow the selection of high-affinity aptamers within four cycles of selection that would require 14 rounds of conventional selection, as known to the art, allowing the substantial reduction of total selection time.

Moreover, the small size of the flow cell employed in such an apparatus (approximately 60 nl) allows for extremely small amounts of target to be worked with, which can be extremely advantageous in certain, especially medical, situations. In a preferred embodiment, a plurality of flow cells may be interconnected, in series, in order to screen the library against more than one target. For example, a series of flow cells may be used to remove ligands to undesired targets, before a flow cell comprising the desired target is used to isolate aptamers having affinity for the desired target.

Both the method of the invention and conventional SELEX require the initial preparation of a library composed of candidate nucleic acid members. The individual members contain one or more randomised regions, preferably flanked by sequences conserved in all nucleic acids in the library. The conserved regions are provided to facilitate amplification or selected nucleic acids. Since there are many such sequences known in the art, the choice of sequence is one which those of ordinary skill in the art can make, having in mind the desired method of amplification. The randomised region can have a fully or partially randomised sequence. Alternatively, this portion of the nucleic acid can contain subportions that are randomised, along with subportions which are held constant in all nucleic acid species in the library. For example, sequence regions known to bind, or selected for binding, to the target protein can be integrated with randomised regions to achieve improved binding or improved specificity of binding. Sequence variability in the library can also be introduced or augmented by generating mutations in the nucleic acids in the library during the selection/amplification process.

In principle, the nucleic acids employed in the library can be any length as long as they can be amplified. The method of the present invention is most practically employed for selection from a large number of sequence variants. Thus, it is contemplated that the present method will preferably be employed to assess binding of random nucleic acid sequences ranging in length from about four bases to any attainable size.

Advantageously, the nucleic acids used in selections according to the invention, in addition to a randomisable region, comprise fixed flanking region(s) which for example may be suitable for attachment of extension primers, such as PCR primers. The sequence of such flanking region(s) may be chosen essentially at will according to the requirements of the individual target(s) and aptamer molecules under consideration.

In a preferred aspect of the invention, the library is constructed of oligonucleotides having the structure:

5 ' GCCTGTTGTGAGCCTCCTGTCGAA(36N)TTGAGCGTTTATTCTTGTCTCCC 3 '

Where N stands for any possible base in the random region.

Any suitable chemistry may be employed for the construction of oligonucleotide library members.

The randomised portion of the library members can be derived in a number of ways. For example, full or partial sequence randomisation can be readily achieved by direct chemical synthesis of the members (or portions thereof) or by synthesis of a template from which the members (or portions thereof) can be prepared by use of appropriate enzymes. End addition, catalysed by terminal transferase in the presence of nonlimiting concentrations of all four nucleotide triphosphates can add a randomised sequence to a segment. Sequence variability in the test nucleic acids can also be achieved by employing size-selected fragments of partially digested (or otherwise cleaved) preparations of large, natural nucleic acids, such as genomic DNA preparations or cellular RNA preparations.

A randomised sequence is preferably generated by using a mixture of all four nucleotides (preferably in the ratio 6:5:5:4, A:C:G:T, to allow for differences in coupling efficiency) during the synthesis of each nucleotide in that stretch of the oligonucleotide library.

In those cases in which randomised sequence is employed, it is not necessary (or possible from long randomised segments) that the library contains all possible variant sequences. It will generally be preferred that the library contain as large a number of possible sequence variants as is practical for selection, to insure that a maximum number of potential binding sequences are identified. A randomised sequence of 30 nucleotides will contain a calculated 10 18 different candidate sequences. As a practical matter, it is convenient to sample only about 10¹⁴ candidates in a single selection. Practical considerations include the number of templates on the DNA synthesis column, and the solubility of RNA and the target in solution. Therefore, libraries that have randomised segments longer than 30 contain too many possible sequences for all to be conveniently sampled in one selection. However, it is not necessary or even desirable to require that the entire sequence space be sampled. Oligonucleotides of a greater length allow for the formation of certain structures that depend on greater size, thus increasing the range of possible substantially different structures disproportionately. Moreover, if the selected members are amplified by a method such as RT-PCR between each selection cycle, the mutagenesis introduced allows for the effective sampling of more sequence variability than that previously sampled.

It is highly preferred in the invention that the nucleic acids of the library are capable of being amplified. Thus, it is preferred that any conserved regions employed in the test nucleic acids do not contain sequences which interfere with amplification. This invention includes libraries containing all possible variations of a contiguous randomised segment of at least 15 nucleotides. Each individual member in the library may also comprise fixed sequences flanking the randomised segment that aid in the amplification of the selected nucleic acid sequences.

Libraries may also be prepared containing both randomised sequences and fixed sequences wherein the fixed sequences serve a function in addition to the amplification process. In one embodiment of the invention, the fixed sequences in a library may be selected in order to enhance the percentage of members in the library possessing a given nucleic acid motif. For example, the incorporation of the appropriate fixed nucleotides will make it possible to increase the percentage of pseudoknots or hairpin loops in a library.

Libraries containing various fixed sequences or using a partially randomised sequence may be constructed based on information concerning the possible structure of ligands directed at the target, obtained by the method of the invention, conventional SELEX or otherwise. A procedure according to the invention may then be initiated with a library constructed on the basis of such criteria.

Isolation of biologically active aptamers is achieved by partitioning the target/library member complexes from unbound library members. Partitioning includes any process whereby members bound to target molecules can be separated from members not bound to target molecules. This can be accomplished by various methods known in the art. Nucleic acid-protein complexes can be bound to nitrocellulose filters while unbound nucleic acids are not. Columns which specifically retain ligand-target complexes (or specifically retain bound ligand complexed to an attached target) can be used for partitioning. Liquid-liquid partition can also be used as well as filtration gel retardation, electrophoresis gel retardation and density gradient centrifugation. The choice of partitioning method will depend on properties of the target and of the target/library member complexes and can be made according to principles and properties known to those of ordinary skill in the art. In a preferred aspect, however, partitioning is achieved by the use of a biosensor apparatus such as that marketed by BIAcore^® and described above. This permits the visualisation of binding and dissociation between target molecules and library members to be made in real time. Since the target molecules are immobilised in a flow cell of the apparatus, partitioning is achieved by washing unbound library members from the flow cell as desired.

Amplifying means any process or combination of process steps that increases the number of copies of a molecule or class of molecules. Amplifying RNA molecules, for example, may 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 recognised by those skilled in the art.

Preferably, the amplification method should result in the proportions of the amplified library being essentially representative of the proportions of different sequences in the initial library. Alternatively, where desired, amplification itself may be used to selectively amplify desired sequences above undesired sequences.

The Polymerase chain reaction (PCR) is an exemplary method for amplifying of nucleic acids. Descriptions of PCR methods are found, for example in Saiki et al. (1985) Science 230:1350-1354; Saiki et al. (1986) Nature 324: 163-166; Scharf et al. (1986) Science 233: 1076-1078; Innis et al. (1988) Proc. Natl. Acad. Sci. 85:9436-9440; and in U.S. Pat. No. 4,683,195 (Mullis et al.) and U.S. Pat. No. 4,683,202 (Mullis et al.). In its basic form, PCR amplification involves repeated cycles of replication of a desired single-stranded DNA (or cDNA copy of an RNA) employing specific oligonucleotide primers complementary to the 3' and 5' ends of the ssDNA, primer extension with a DNA polymerase, and DNA denaturation. Products generated by extension from one primer serve as templates for extension from the other primer. A related amplification method described in PCT published application WO 89/01050 (Burg et al.) requires the presence or introduction of a promoter sequence upstream of the sequence to be amplified, to give a double-stranded intermediate. Multiple RNA copies of the double-stranded promoter containing intermediate are then produced using RNA polymerase. The resultant RNA copies are treated with reverse transcriptase to produce additional double-stranded promoter containing intermediates which can then be subject to another round of amplification with RNA polymerase. Alternative methods of amplification include among others cloning of selected DNAs or cDNA copies of selected RNAs into an appropriate vector and introduction of that vector into a host organism where the vector and the cloned DNAs are replicated and thus amplified (Guatelli, J. C. et al. (1990) Proc. Natl. Acad. Sci. 87: 1874). In general, any means that will allow faithful, efficient amplification of selected nucleic acid sequences can be employed in the method of the present invention.

In a preferred aspect, the invention relates to chimeric aptamer molecules which are bispecific, that is they comprise two ligands joined together by a suitable linker. Bispecific antibodies have found applications in medical fields. For example, reference is made to international patent application WO88/09344.

The invention moreover provides anti-CD4 aptamers preparable by the method described above. The leukocyte surface antigen CD4 is involved in the recognition of foreign antigens by T-lymphocytes. Binding to CD4 allows T-cells to recognise antigen in association with MHC class II molecules. Consequently aptamers against CD4 may be used diagnostically to measure the number and distribution of CD4+ leukocytes in a variety of disease conditions; therapeutically or experimentally to inhibit the activation or effector function of CD4+ leukocytes (including the induction of tolerance, inhibition of tissue rejection, and suppression of chronic inflammation); and therapeutically to inhibit the infection of CD4+ cells by viruses.

In a preferred aspect, an anti-CD4 aptamer according to the invention has a structure as set forth below: 5 ' GGGAGACAAGAAU AAACGCUCAA-ACUUCAGCUCUAUUAACAGCUC AAG GACUGGCACA-UUCGACAGGAGGCUCACAACAGGC 3'

The dashes (-) delineate the sequence which was randomised in the library.

The invention is further described, for the purposes of illustration only, in the following examples.

Example 1

DNA Oligonucleotide Library

DNA oligonucleotides 83 bases in length, having a randomised portion of 36 bases, are used for the development of an aptamer capable of binding CD4. A library of synthetic RNA oligonucleotides having the following structure is prepared constructed (ordered from and provided by Genosys):

5' CCTGTTGTGAGCCTCCTGTCGAA(36N)TTGAGCGTTTATTCTTGTCTCCC 3'

Where N stands for any possible base in the random region. The random region is generated by using a mixture of all four nucleotides (ratio 6:5:5:4, A:C:G:T, to allow for differences in coupling efficiency) during the synthesis of each nucleotide in that stretch of the oligonucleotide library. The resulting complexity is theoretically 4³⁶= 4.7 x 10²¹ molecules, though the scale of synthesis (O. lμmol) followed by gel purification yields 8.8nmol which puts an absolute upper limit of approximately 5xl0¹⁵ on the number of different molecules actually present.

PCR Amplification with a 5' primer that introduces the recognition site for T7 RNA Polymerase (5' TAATACGACTCACTATAGGGAGACAAGAATAAACGCTCAA 3') and 3 'primer (5' GCCTGTTGTGAGCCTCCTGTCGAA 3^*) results in the following template for transcription: O 99/27133 j****. PCT/GB98/03544

5' TAATAGCACTCACTATAGGGAGACAAGAATAAACGCTCAA (36N) TTCGACAGGAGGCTCACAACAGGC 3'

The RNA transcript itself has the following sequence:

5' GGGAGACAAGAAUAAACGCUCAA (36N) - UUCGACAGGAGGCUCACAACAGGC 3'

Example 2

Selection Of Anti-CD 4 Aptamers by BIAcore a) 1st transcription: the template is the "library" as supplied by Genosys, made double stranded by PCR amplification under stringent conditions (as few errors as possible and only five cycles, as described in (Conrad et al., (1996) Meth. Enzymol. 267:336-367). 15 μg template is used in 500 μl volume in an microfuge tube following the method of Fitzwater and Polisky, (1996), Meth. Enzymol. 267:275-301, with a modified transcription buffer as described in Heidenreich et al., (1995) NAR 23:2223- 2228; the DNA template is then removed by DNase treatment, RNA is recovered and adjusted to 100 μg/ml in modified HBS buffer for BIAcore (10 mM HEPES pH 7.4; 150 mM NaCl, 1 mM MgCl₂; ImM CaCl₂; 0.2μm filtered ,and degassed).

b) Injection of 25 μg RNA first onto an Albumin chip (all 4 flow cells on a chip are coated with the protein Albumin) to pre-clear non-specific protein-binding RNA; after passing over all the four flow cells the RNA is recovered automatically in a specialised recovery cell within the machine and delivered to a specified Microfuge tube in a connected rack. This RNA is then injected onto a CD4 chip (all the 4 flow cells are coated with soluble rat CD4 (Davis et al., (1990) J. Mol. Biol. 213:7-10); after incubation for 24 hours, unbound RNA is discarded and bound RNA is eluted by injection of 20μl urea (7M) and 20μl formamide (20% in HBS) (recovery of the eluate as described above). After standard EtOH precipitation of the recovered RNA, this RNA is then reverse transcribed (RT) and amplified by PCR to produce the template for the 2nd transcription (as described in Fitzwater and Polisky, (1996), supra).

c) All the following rounds are done the same way with varying numbers of flow cells used for pre-clearing and/or selection as indicated.

d) Second transcription with 7μg template in a total volume of 250μl in a microfuge tube. Injection of lOOμl at 140μg/ml onto 2 flow cells with albumin (no response, which means that there is no measurable binding of RNA to the Albumin or the chip). Recovered RNA is then injected onto 2 flow cells coated with CD4; it is then incubated for 2.5 hours and eluted with 20 μl formamide, precipitated, and processed by RT and PCR to produce the template for the 3rd transcription.

e) The third transcription is performed with 15 μg template in a total volume of 500 μl. The injection is performed by injecting 100 μl at 200 μg/ml first onto 1 flow cell with Albumin (> > no response). The recovered RNA is then injected separately (50% of the RNA in one and 50% in the other) onto two flow cells coated with CD4, incubated and eluted from with formamide as above; precipitation, RT, and PCR are carried out as before to produce the template for the 4th transcription.

f) The fourth transcription is carried out using 5 μg template in a total volume of 250 μl. The injection is carried out with 100 μl at 130 μg/ml) first onto an albumin flow cell. The recovered RNA is then separately injected onto two CD4 flow cells, incubated and eluted as above, followed by standard precipitation, RT and PCR.

Eluted aptamers from the fourth transcription have a high specific activity for CD4. Several clones are isolated and sequenced. The structure/sequence of a selected CD4- binding aptamer (clone 14) is:

5' GGGAGACAAGAAUAAACGCUCAA-ACUUCAGCUCUAUUAACAGCUCAAG GACUGGCACA-UUCGACAGGAGGCUCACAAC AGGC 3 ' Example 3

Comparison Of The Procedure Of The Invention And Conventional SELEX a) The binding and dissociation kinetics of unselected nucleic acid pools (round 0) and selected pools from rounds 6, 11 and 14 from a SELEX procedure using extended contact times against CD4 according to the first aspect of the present invention. In this experiment, 25 μl of each RNA (100 μg/ml) is applied at 5 μl/min over flow cells coated with soluble rat CD4 (between 3.9ng and 4.2ng per cell). At the end of injection, the following amounts of RNA are bound: round 14, 243pg; round 11, 175pg; round 6, 93pg; unselected, 9pg. The results are shown in Table la.

b) The binding and dissociation kinetics of unselected nucleic acid pools (round 0) and selected pools from rounds 6, 11, 14 from a conventional SELEX procedure against CD4 (30 min contact time). In this experiment, 25 μl of each RNA (lOOμg/ml) is applied at 5 μl/min over flow cells coated with soluble rat CD4 (between 3.9ng and 4.2ng per cell). At the end of injection, the following amounts of RNA are bound: round 14, 210pg; round 11, 121pg; round 6, 38pg; unselected, 9pg. The results are shown in Table lb.

c) The binding and dissociation kinetics of unselected nucleic acid pools (round 0) and selected pools from round 11 of a conventional SELEX procedure, round 11 of a SELEX procedure using extended contact times and round 4 from a SELEX procedure against CD4 using the BIAcore to select a slow-dissociating fraction according to the second aspect of the present invention. In this experiment, 25 μl of each RNA (100 μg/ml) is applied at 5 μl/min over flow cells coated with soluble rat CD4 (between 3.9ng and 4.2ng per cell). At the end of the injection, the following amounts of RNA are bound: round 11 extended contact time procedure, according to the invention: 175pg; round 11 conventional SELEX procedure: 121pg; round 4 BIAcore-selection procedure, according to the invention: 130pg; unselected: 9pg. The results are shown in Table lc. TABLE 1

Time (sec)

Time (sec)

round 11 : extended contact time round 4: BIAcore selected

Time (sec) d) The binding of CD4 aptamers is highly specific. In this experiment, 25 μl of a cloned CD4 aptamer (clone 14; 100 μg/ml) is applied at 5 μl/min over two flow cells, one coated with soluble rat CD4 and the other with soluble human CD4 (between 3.9ng and 4.2ng per cell). The results are shown in Table 2. Binding to rat CD4 is much stronger than to human CD4.

TABLE 2

D4

Time (sec)

e) Aptamers are moreover shown to recognise native CD4. FACS analysis of rat lymph-node cells either unstained (shaded) or stained with the anti-CD4 monoclonal antibody, W3/25, with or without pre-incubation with clones of anti-rat CD4 aptamer. The results are shown in Table 3. TABLE 3

Inhibition Of Mixed Lymphocyte Reaction By An Anti-CD 4 Aptamer

5xl0⁵ lymph node cells from PVG rats are added to each well of a 96-well microtitre plate. The cells are incubated for 15 mins with the given amounts of RNA or antibody (total volume lOOμl). 5xl0⁵ irradiated (3000 Rad) lymph node cells from DA rats are then added to each well. The mixture is incubated for 3 days in 200μl, pulsed with ³H-Thymidine for 18 hrs, harvested, and counted in a beta-plate scintillation counter. 100% represents the mean value (of 12 wells) for this mixture without added RNA or mAb. Background is less than 1 % .

The results are shown in Table 4. TABLE 4

Example 4

Conventional Aptamer Selection Using Extended Incubation Times The 5'GCCTGTTGTGAGCCTCCTGTCGAA(36N)TTGAGCGTTTATTCTTGTCTCCC 3' oligonucleotide library , the 5' primer (TAATACGACTCACTATAGGGAGACAAGAAT- AAACGCTCAA) and the 3' primer (GCCTGTTGTGAGCCTCCTGTCGAA) for PCR amplification are produced by Genosys (Cambridge, UK). The initial DNA library ( ~ 250 pmol) is transcribed in vitro by T7 RNA polymerase (1500 U) overnight at 37 °C. The 500 μl reaction contains ImM each of 2'F-dUTP and 2'F-dCTP (Amersham Pharmacia Biotech, UK) and ImM each of GTP and ATP in an optimised buffer (17). After transcription, digestion with RQl RNase-free DNase and phenol/chloroform extraction are performed and the product loaded onto a 12% polyacrylamide gel under denaturing conditions to check for full-length product. Transcription products are then dialysed in binding buffer (PBS, ImM CaCl₂, ImM MgCl₂) and then purified by using Sephadex G-50 prepacked NICK Spin Columns.

An initial library of - 10¹⁵ sequence variants is subject to in vitro selection to isolate

2'F-modified RNA aptamers able to specifically bind recombinant rat CD4. Fourteen rounds of SELEX are carried out using rat CD4-coupled Sepharose beads for the affinity selection. Two similar selections are performed where all the protocols and reagents in the procedure are the same with the exception of the contact time in the partitioning step, when RNA molecules and target CD4 are allowed to interact. In one case (limited) the incubation time is limited to 30 minutes, in the other (extended) it is extended to 2 hours and 30 minutes. The aim is to test whether a longer contact time would select for molecules with slower on-rate, possibly associated with slower off-rate and, therefore, lower dissociation rate constants and longer half-lives.

Binding of aptamer pools to the target during both the selections is tested by BIAcore analysis and the results of the two selection methods compared. Briefly, 25 μl of freshly transcribed RNA, either from different rounds of the selection or from single clones, is injected at a concentration of- 50 μg/ml in modified HBBS (10 mM HEPES pH 7.4, 150 mM NaCl, 1 mM MgCl₂, 1 mM CaCl₂) into flow cells previously coupled with rat CD4 or control proteins. For the coupling, either rat CD4 at 70 μg/ml or human CD4 at 20 μg/ml or mouse CD4 at 50 μg/ml in 10 mM NaAc pH 5.0 are directly immobilised via primary amine groups onto pre-activated surfaces in three adjacent flow cells. BIAcore 2000 software outputs are firstly evaluated by the BIAvaluation III program and then exported into Graph Pad Prism and analysed.

RNA transcripts from Round 6, Round 11 and Round 14 are injected onto flow cells where rat CD4 has previously been immobilised. BIAcore traces are recorded as response units (RU) vs. time signals and the profiles superimposed for comparison (Table 4, Panel A). Enrichment can be detected in both the extended and the limited selections, when following the process from Round 6 to Round 11 and finally Round 14. Nevertheless, in the case of the extended incubation time method it is possible to reach a good level of enrichment after only six cycles which instead requires 14 cycles of the limited contact time approach.

Moreover, the kinetic properties of the pools selected by the two methodologies are different. Table 5, Panel B shows the half-lives of the aptamer populations at the three different time points for both the methods. Two features can be detected: firstly, when the t y₂ of the two pools are compared within one particular cycle, the extended contact time approach always shows a longer half-life value in respect to the limited one. Secondly, when comparison is done between cycles within each SELEX process, that is comparing round 6, 11 and 14 singularly, the 'extended' selection seems to enrich for aptamers whose average half-life increases during the course of selection. In the case of the limited approach, on the contrary, selection is pushed towards RNA populations with an almost constant or, even, a lower half-life by the time the last cycle of selection is performed. Presumably, the increase in signal seen for the limited SELEX between rounds 6 and 14 resulted more from an increase in the proportion of RNA molecules that bound CD4 rather than an improvement in their half-lives. ( A)

TABLE 5

Table 5

Appearance of CD4-binding molecules within RNA pools during SELEX: BIAcore determination of enrichment and half-lives. RNA transcripts from Round 6, Round 11 and Round 14 of the two selections (extended, solid line; limited, interrupted line) were injected onto a flow cell previously coated with rat CD4 ( - 5000 RU), at a concentration of 50μg/ml. (A) Binding abilities are derived from BIAcore profiles and recorded as response units (RU) at the time point when injection is completed. (B) Half- lives are derived from BIAcore dissociation profiles by non-linear curve-fitting with GraphPad Prism.

Example 5

Aptamer clones: binding properties

The PCR-amplified cDNA from the last round of the selection is cloned in both cases and single clones screened for binding to rat CD4 by BIAcore analysis. Specificity is checked for all the clones by immobilising control proteins on adjacent flow cells. RNA transcripts from each single clone are injected simultaneously onto flow cells where either rat or human or mouse CD4 have previously been immobilised. Again, binding is detected as a RU vs. time signal. Table 6 shows an overlay of BIAcore traces corresponding to the three flow cells: clear specificity for rat CD4 is shown as neither human nor mouse CD4 are recognised by the aptamer.

The half-lives of the single clones are derived from the relative BIAcore dissociation profiles and plotted (Table 7). The majority of the clones in both the selections seemed to cluster on the bottom part of the diagram but two clones selected by the extended contact time method showed substantially higher t Vι values. This is in turn reflected by the mean value for the half-lives within the two groups of clones (represented by a bar) that results to be shifted upwards for the extended selection when compared to the limited one.

time (sec)

Table 6

Aptamer specificity. Typical BIAcore traces of a ratCD4-specific aptamer. Rat CD4 was immobilised on one flow cell (4500 RU) while human CD4 (2000 RU) and rat CD4 (5000 RU) were immobilised on two others as negative controls. RNA from clone 1165 was then simultaneously injected onto the three flow cells and binding recorded as RU vs. time signal: the traces from rat CD4-coated (solid line), human CD4-coated (dotted line) and mouse CD4-coated (interrupted line) flow cells are superimposed.

Table 7

Effect of selection time on dissociation rates of aptamer clones. The PCR products from the last round (Round 14) of the 2h30min/ extended selection (squares) and the 30min/ limited selection (triangles) were cloned. Ten clones from the extended selection and twelve clones from the limited selection were analysed by BIAcore and their half- lives derived from the dissociation profiles. A bar is representative of the mean within the group.

Example 7

Comparison of Aptamers and antibodies

One of the challenges in the application of the SELEX process is to select oligonucleotide species able to compete with or even exceed the high affinity properties of antibodies. In order to compare anti-rat CD4 aptamers with anti-rat CD4 antibodies, RNA transcripts from the best clones of each SELEX method, namely el4 selected by the extended incubation time approach and 1187 selected by the limited incubation time approach, are injected at a concentration of - 50 μg ml onto a BIAcore flow cell that has been previously coated with rat CD4. Furthermore, the very high affinity Fab fragment of the anti-rat CD4 monoclonal antibody W3/25 is injected onto the same flow cell and the traces compared. Table 8 shows the normalised dissociation profiles of el4 aptamer, 1187 aptamer and the Fab fragment. The half-life of each interaction is also reported. Neither aptamer appears to be quite as good a ligand for CD4 as is the W3/25 Fab, to judge from the dissociation curves, although the aptamer from the extended SELEX approaches the Fab closely.

time (sec)

Table 8

Comparison between aptamer clones and Fab fragment. Clone el4 (interrupted line) from the extended selection, clone 1187 (dotted line) from the limited selection and Fab fragment of anti-rat mAb W3/25 (solid line) were injected onto a ratCD4-coupled flow cell. Normalised BIAcore dissociation profiles were superimposed and the relative half- lives recorded.

Example 8 Sequence analysis

The aptamer clones derived from the last round of both the selections are sequenced and the portions corresponding to the variable region of the initial construct are aligned (Table 9). Four groups and one orphan are derived based on sequence similarities. All 12 clones selected by the limited contact time approach that are analysed fell within three of the four groups, namely Group 1, 3 and 4. None of them seemed to have sequence analogies with Group 2 or the orphan clone, represented only by clones selected by the extended selection.

A phylogenetic analysis of these anti-rat CD4 aptamers is also performed and the derived tree represented in Table 10. Similarly, the cloned sequences grouped into four families and one orphan and, again, none of the clones isolated from the limited selection are found in group 2 or are closely related to the orphan sequence ell. Surprisingly, the clones el4 and el 8, selected by the extended approach and belonging to Group 2, or the orphan clone el l, both very distantly related to the limited contact time selected clones, are actually the very ones characterised by the longer half-lives among all the examined clones.

In conclusion, the extended contact time method allows enrichment at an earlier stage, when compared to the 'limited' one; this permits reduction of the number of SELEX cycles and, thus, shortens the whole process. Moreover, the pool of aptamers selected by this approach displays different kinetic properties with, on average, longer half-lives. In particular, selected clones present a significantly lower dissociation rate constant as estimated from the half-lives, almost comparable to those typical of antibodies. For this reason, the hypothesis - that longer contact times might allow the more rapid selection of aptamers with improved dissociation kinetics - is confirmed by these results.

GROUP 1

E10 CUCAGAG.AC AGAGCAGAAA GGACAGUUCA AGC . GACA

E8 CUCAGAG.AC AGAGCAGAAA CGACAGUUCA AGCCGA.A

LI87 UUCAGAG.AC AAAGCAGGAA CGACAAUUCA AGC . GACA

LI80 CUCAGAGUAC AGAGCAGGAA CGACAGUUCA AGC . GACA

GROUP 2

El8 AGCACUUCAG ..AUAUGAUA ACAGGUUCAA GGA.UGUGCA CA

E14 ...ACUUCAG CUCUAU..UA ACAGC.UCAA GGACUG.GCA CA

GROUP 3

E6 AAGACCACAC AUA.AGA..A ACAGGGAACA GCGUUCAA

E33 AAGACCACAC ACACAGA..A ACAGGGAACA GCGUUCAA

LI78 AAGACCACAC ACACUG..AA ACAGGGAACA GCGUCCAC

LI47 AAGACCACAA ACACAG ..AA ACGGGGAACA GCGUCUAA

L45 AAGACCACAC ACACAG..AA ACAGGGAACA GCGUCCA.

El6 GAGCAUUAAC AAAAAGAUCA CUAGG.AACA GCGG.UA.

LI04 AAGCACGAAC AUA.AGC.AA CUCGG.AUCA GCGACGUAU

L76 GA.AGAAGAA ACG.AGCUGA CGGAGGCGUG AGAGGAUGA

L74 ...CCUGUGC CCAUGCACAC UCAGUGUACA GUGUUCAG

GROUP 4

E29 GUAGUCAAAA GUCAUACAGC UCUACCAACA GCUC .... GA .

L90 CCAGCUGAAA G....ACAGC UCAAUUAACA GCUC....AA C

LI65 .CAGCUAGAA UA...ACAGC UCAGGAGCAC CUUCCCGAGA C

L235 .CAACUCAAU UA...ACAGU UCAAGAGCCG GAGCUAAACC A

E2 . CCACCAGCC CAGAGAUAGC CCAGACACCA C .... CAAGA C

ORPHAN

Ell CCCAAUUCCG ACCUGCGUCU ACGUAACCUG CCAUCG

Table 9

Sequence alignment of ratCD4-specific aptamers. The clones analysed by BIAcore were sequenced and the portions corresponding to the variable region of the initial construct were aligned using the Pileup program of the GCG package. Groups were derived based on sequence similarity. Clones denoted with the letter "e" belong to the extended selection; those denoted with the letter "1" belong to the limited selection.

Table 10

Phylogenetic analysis of ratCD4-specific aptamers. The cloned sequences corresponding to the variable region of the initial construct were analysed by the GCG program 'distances' and represented using the program 'growtree'. Clones denoted with the letter "e" belong to the extended selection; those denoted with the letter "1" belong to the limited selection.

Claims

Claims

1. A method for isolating a nucleic acid capable of binding to a target from a library of single stranded nucleic acid members by selecting library members according to their dissociation kinetics.

2. A method according to claim 1, comprising the steps of:

a) incubating the library with the target under conditions suitable for binding such that an equilibrium of dissociation from and association with the target is reached for substantially all library members, including those that have slow rates of dissociation;

b) partitioning the library to isolate the members which are bound to the target;

c) washing the isolated bound target/library member complexes in order to remove weakly bound members;

d) eluting the bound library members; and

e) amplifying the selected members.

3. A method according to claim 1 for isolating a nucleic acid from a library of single-stranded nucleic acid members, whose dissociation rate from a target lies within a desired range, comprising the steps of:

a) incubating the library with the target under conditions suitable for binding;

b) eluting the members of the library under conditions that enable the fraction remaining bound to be estimated in real time; c) discarding those members that elute in less than a time corresponding to a mean desired half life;

d) selecting those members that elute in a time corresponding to a mean desired half life, or longer; and

e) amplifying the selected members.

4. A method according to any preceding claim, wherein steps a) to e) are repeated iteratively.

5. A method according to any preceding claim, further comprising the step of incubating the library, before the selection step d), with a further target, thus removing those library members capable of binding the further target from the library.

6. A method according to any preceding claim, wherein the library members are constructed from RNA.

7. A method according to claim 6, wherein the library members comprise modified ribonucleotides or ribonucleotide analogues.

8. A method according to any preceding claim, wherein steps a) to d) at least are performed on a biosensor apparatus capable of monitoring the association and dissociation of the target and the library members in real time.

9. A method according to each preceding claim, wherein each library member is composed of 5' and 3' constant regions flanking a region which is at least partly randomised.

10. A method according to claim 9 wherein the randomised region comprises between 30 and 50 bases.

11. A method according to any preceding claim wherein step e) is performed by PCR.

12. An anti-CD4 aptamer preparable by a method according to any one of claims 1 to 11.

13. An anti-CD4 aptamer having the structure:

5 ' GGGAGACAAGAAUAAACGCUCAA-ACUUCAGCUCUAUUAACAGCUC AAG GACUGGCACA-UUCGACAGGAGGCUCAC AACAGGC 3 '