| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
NeXstar Pharmaceuticals, Inc., 2860 Wilderness Place, Boulder, CO 80301. Fax 303-444-0672; e-mail sumedha{at}nexstar.com
 |
Abstract |
|---|
 Top Abstract IntroductionAntibodiesAptamersAptamers in Different Diagnostic...ConclusionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionAntibodiesAptamersAptamers in Different Diagnostic...ConclusionReferences |
|---|
|
Antibodies |
|---|
|
TopAbstractIntroductionAntibodiesAptamersAptamers in Different Diagnostic...ConclusionReferences |
|---|
It is important to note, however, that various approaches are being developed to circumvent limitations associated with the current hybridoma-based antibody technology. These approaches, which include humanization of antibodies (9)(10), displaying peptide libraries on phages (11)(12) and ribosomes (13)(14)(15), antibody engineering (16), and in vitro immunization (17), are at various stages of development.
Alternatively, it is possible to consider an entirely different class of molecules, aptamers, to meet the shortcomings of antibodies. In this regard, aptamers have the following advantages:
|
Aptamers |
|---|
|
TopAbstractIntroductionAntibodiesAptamersAptamers in Different Diagnostic...ConclusionReferences |
|---|
The SELEX process begins with a random sequence library obtained from combinatorial chemical synthesis of DNA. Each member in a library is a linear oligomer of a unique sequence. The complexity, or the molecular diversity, of a library is dependent on the number of randomized nucleotide positions. Theoretically, a library containing a 40-nucleotide random region is represented by 1.2 x 1024 individual sequences (420 = 1.2 x 1024). However, in practice, the complexity of a typical combinatorial oligonucleotide library obtained from 1-µmol scale solid-phase DNA synthesis is limited to 1014 to 1015 individual sequences. The success of finding unique and rare molecules that interact with a target parallels the diversity of the libraries used. The degree of molecular diversity present in random sequence oligonucleotide libraries supersedes that of other combinatorial libraries used for screening. These include peptide libraries used for phage display as well as the libraries made up of small organic molecules.
In the screening process, a random sequence oligonucleotide library is
incubated with a target of interest in a buffer of choice at a given
temperature (Fig. 1
). During this step, a very small fraction of individual
sequences tends to interact with the target, and these sequences are
separated from the rest of the library by means of any one of the
physical separation techniques. Typically, nitrocellulose filter
partitioning is used with protein targets that are retained on
nitrocellulose. Small molecular targets are generally immobilized on a
solid support to generate an affinity matrix, in which case sequences
that do not interact with the target on the solid support can be
removed easily by a simple washing step. The population of sequences
bound to the target is isolated and amplified to obtain an enriched
library to be used for the next selection/amplification cycle. The
enrichment efficiency of high-affinity binders is governed by the
stringency of selection at each round. The progress of the enrichment
of high-affinity binders can be determined by carrying out binding
analysis of enriching populations against the target. Once affinity
saturation is achieved after several rounds of selection/amplification,
the enriched library is cloned and sequenced to obtain the sequence
information of each member. Individual sequences are further
characterized on the basis of their ability to bind to the target.
Usually, the majority of individual sequences, >90%, in an enriched
library are "winners", aptamers that bind to the target used for
selection.
|
Aptamers that come out of a SELEX experiment are full-length sequences containing the fixed sequences that were included to aid the amplification process. These full-length aptamers are generally 70–80 nucleotides long and could be truncated to eliminate nucleotide stretches that are not important for direct interaction with the target or for folding into the structure that facilitates target binding. The identification of truncated aptamers restricted to the minimal target-binding domain requires some effort, but it has been successfully carried out to obtain functional aptamers less than 40 nucleotides long (18)(19)(20)(21)(22)(23). In the majority of cases, the fixed sequence regions used for primer binding are unimportant for aptamer function and can be eliminated. Technological advances have already been made to eliminate the requirement for the fixed regions in random sequence libraries used for the SELEX process (Pagratis et al., manuscript in preparation), thereby producing short aptamer sequences.
The number of cycles required for aptamer identification is usually dependent on the degree of stringency imposed at each round as well as on the nature of the target. For most targets, affinity enrichment is reached within 8–15 cycles. In general, a researcher could accomplish one cycle of SELEX every 2 days. Including cloning and sequencing, a typical SELEX experiment may take approximately 2–3 months. Once the sequence is identified, an aptamer is produced by chemical synthesis. Small-scale synthesis and purification of an aptamer do not take more than 3 days and provide ample quantity of aptamer (several nanomoles) for the design and optimization of a diagnostic assay. This whole process is faster than the amount of time typically spent to generate a cell-line to produce a specific monoclonal antibody and purification of the antibody. The SELEX process recently was automated to make aptamer discovery even faster and more economical [Jenison et al., manuscript in preparation and Ref. (24)]. The automated platform carries out the iterative SELEX process around the clock with little or no human intervention and intuition (Jenison et al., manuscript in preparation). It has parallel processing capabilities to handle multiple SELEX experiments run on microtiter plates, allowing fast and high-throughput discovery of aptamers.
Aptamers are known for their remarkable specificity. Aptamers can discriminate targets on the basis of subtle structural differences such as the presence or absence of a methyl (25)(26) or a hydroxyl group (27)(28) and the D- vs L-enantiomer of the target (27)(29). The high degree of specificity often seen in aptamers, sometimes even better than antibodies (25), is a result of the selective demand in the SELEX process that eliminates sequences that bind to closely related analogs of the target. Practically, this is achieved by the process called "counter-SELEX" that effectively discards ligands that have ability to bind the target as well as closely related structural analogs of the target (25). During selection, the population of aptamers bound to the target is subjected to affinity elution with structural analogs and the sequences eluted are discarded. In some diagnostic applications of small molecule targets, it may be important to measure the analyte as well certain structural variants of the analyte. In that case, aptamers retained on the target could be specifically eluted with the structural analogs to select the species that do not discriminate them.
The counter-SELEX strategy could be a valuable tool in identifying aptamers aimed at a unique target in a complex mixture, probably even without knowing what the target is, for example, in the search for aptamers that bind to an "epitope" present exclusively on the surface of cancer cells but not in healthy cells, or to find aptamers that interact with molecules present in the serum of patients infected with a pathogen but not present in uninfected individuals. In these applications, it would be possible to use the cells from healthy tissue (or serum from a healthy individual) to remove sequences that bind to the background that does not contain the epitope of interest before the library is challenged with cancer cells (or serum from the patient).
modified libraries for aptamer discovery
Chemically modified oligonucleotide libraries have been introduced
to the SELEX process for several reasons. Unmodified oligonucleotides,
especially RNA, are degraded by nucleases commonly present in
biological fluids. Researchers in the antisense field have made
impressive advances toward making oligonucleotides resistant to
nucleases by introducing various modifications, predominantly to the
oligonucleotide backbone (30)(31). Most of these
backbone modifications are introduced during the chemical synthesis and
are not compatible with the enzymes used in the SELEX process. An
exception to this is phosphorothioate modification. Oligonucleotides
modified with phosphorothioate linkages exert nonspecific interactions
with proteins (32)(33); hence, they may not be
suitable for identifying ligands that confer specificity. Certain
modifications at the 2' position of the sugar (Fig. 2
) make RNA nuclease resistant (34)(35).
Because the nucleases that are most abundant in biological fluids
appear to be pyrimidine-specific endonucleases, substitutions at the 2'
positions of pyrimidine nucleotides alone is sufficient to protect an
RNA sequence from degradation in biological fluids
(35)(36). Most importantly, pyrimidine
nucleotides substituted with amino (NH2) and
fluoro (F) functional groups at the 2' position of the sugar are
substrates for the enzymes used in the SELEX process. As a result,
aptamers with enhanced survival times in biological fluids have been
selected successfully from libraries containing pyrimidines modified
with 2'-NH2 and 2'-F functional groups
(19)(21)(37)(38)(39)(40). These aptamers,
resistant to nucleases, are well suited for both diagnostic and
therapeutic applications.
|
Only four building block monomers are available for making replicable
oligonucleotide libraries. Attachment of different substituents to
nucleotide bases of these four monomers effectively increases the
molecular diversity of oligonucleotide libraries. Such modified
libraries will be useful for aptamer technology provided they can be
substrates for the enzymes used in the SELEX process. There are several
elegant examples in which novel unnatural base pairs have been
specifically incorporated into DNA and RNA (Fig. 3
). Piccirilli et al. (41) and Bain et al.
(42) demonstrated the incorporation of the IsoG/IsoC pair
and the 2,6-diaminopyrimidine/xanthine base pair into DNA and RNA in
experiments aimed at expanding the genetic code. Tor and Dervan
(43) used the methylisoC and (6-aminohexyl)isoG base pair to
introduce aminohexyl modification into unique positions in an RNA
sequence by in vitro transcription. Other appendages, such as biotin
and EDTA, were then introduced to unique alkyl amine functional groups
of RNA posttranscriptionally. Libraries can be tailor-made for certain
classes of targets by decorating nucleotide bases with different
functional groups. For example, modified libraries equipped with
hydrophobic groups on nucleotides may be suitable for finding winning
aptamers to hydrophobic targets. Recently, Schweitzer et al.
(44) and Guckian et al. (45) described a novel
base-pair analog that does not require hydrogen bonding for
recognition. Nonpolar, non-hydrogen-bonding shape mimics (6-methyl
purine and 2,4-difluorotoluene) for an AT base pair had been
enzymatically incorporated into DNA. These base-pair analogs have the
ability to become building blocks of replicable oligonucleotide
libraries, and yet they represent a radical departure from the
classical Watson-Crick base pairing. These modifications will increase
the molecular diversity of oligonucleotide libraries and further
enhance the probability of finding aptamers with unique properties.
|
Similar to the nuclease-resistant aptamers isolated from the libraries containing 2'-modified pyrimidines, aptamers modified at the C-5 position of pyrimidines have also been identified using the corresponding modified libraries. Aptamers consisting of 5-(1-pentynyl)-2-deoxyuridines have been selected to recognize human thrombin using a library containing the modified thymidine nucleotide analog (46). Aptamers isolated from the modified library were strikingly different from those derived from an all-natural DNA library (47) at the level of the primary and the secondary structure. Thus, it is possible that the substitution at the C-5 position of thymidine may have affected the shape repertoire of the library. The attachment of fairly bulky groups, such as a benzoyl group, at the C-5 position of pyrimidines appears to be tolerated by the enzyme used in the SELEX process (48)(49), which expands the possibilities for chemically decorating oligonucleotide libraries. Interestingly, amino acid side chains could also be introduced to create pseudo peptide/nucleic acid libraries.
Oligonucleotides containing certain modifications can be activated by light to generate reactive groups that can form covalent linkages with another molecule in close proximity (50)(51)(52)(53)(54). A special class of aptamers called photo-cross-linkable aptamers can undergo cross-linking between the aptamers and their cognate targets upon brief exposure to ultraviolet light. Modified libraries containing 5-bromouracil and 5-iodouracil residues have been used to generate photo-cross-linkable aptamers [Golden et al., manuscript in preparation, and Ref. (55)]. The photo-cross-linking mediated by a 5-halo-substituted thymidine analog takes place between the C-5 position of the nucleobase and an electron-rich amino acid at close proximity. The potential amino acid candidates for cross-linking are tryptophan, tyrosine, histidine, phenylalanine, and cysteine. A priori, the requirement for a photo-cross-linkable amino acid side chain to reside at close proximity with the correct geometry to a single 5-halo-uracil residue of an aptamer is very stringent. Such a high stringent requirement for photo-cross-linking is expected to bring added specificity to aptamer-protein interactions. Long wavelength monochromatic light sources, 308 nm for 5-bromouracil and 325 nm for 5-iodouracil, used in the photo-cross-linking of aptamer-protein complexes substantially reduce the photodegradation of proteins and aptamers (53)(55)(56). Although identification of photo-cross-linking aptamers directly from 5-halo-substituted libraries is preferred, post-SELEX substitution experiments could also lead to the identification of positions to be substituted by 5-halouracil residues within an aptamer to afford specific photo-cross-linking (20)(57).
The serendipitous discovery of aptamers that undergo covalent cross-linking in the absence of light has been reported (55). Although these aptamers undergo light-independent cross-linking to their cognate protein target, they require 5-iodouracil to do so. The chemistry of light-independent cross-linking has been attributed to the formation of a Michael adduct between the C-6 position on 5-iodouracil and a suitable nucleophile on the protein. These aptamers that undergo light-independent target cross-linking grant all the benefits provided by photo-cross-linking aptamers for diagnostic applications and do not require a light source for cross-linking.
The characteristics that are most important to the success of a diagnostic assay are the affinity and the specificity of the ligand that provides molecular recognition. In heterogeneous assays, nonspecifically retained molecules are washed away under conditions that do not affect the binding equilibrium of the specific target. Ideally, all molecules bound nonspecifically must be eliminated without losing any of the specific target. Chemistries that trigger the formation of a covalent linkage between an aptamer and the specific target will permit the use of extremely harsh conditions to remove molecules that contribute to the background in heterogeneous assays. In homogeneous assays, the generation of such a covalent linkage overcomes any possible dissociation of targets, especially ones with fast off-rates.
properties
Single-stranded oligonucleotide sequences have an exceptional
propensity to assume an array of secondary (and tertiary) structural
motifs with different shapes. The number of possible thermodynamically
stable structural variants of an oligonucleotide sequence is much
higher than the number of variants available for a peptide sequence of
the same length. This is simply based on the ability of nucleotide
bases to interact with each other through canonical Watson-Crick as
well as unusual base pairing. The existence of oligonucleotide
sequences that could assume a myriad of shapes within a random sequence
library is the basis for the remarkable success of generating aptamers
to a wide variety of target molecules (58)(59).
To date, there has not been a restriction on the type of target for which high-affinity aptamers could be identified. Aptamers have been identified that bind to small molecular targets, including metal ions (60)(61), organic dyes (3)(62), drugs (25)(28), amino acids (29)(63)(64)(65), cofactors (66)(67)(68)(69)(70), aminoglycosides (71)(72)(73)(74), antibiotics (75)(76), nucleotide base analogs (77), nucleotides (26)(27)(78), and peptides (79)(80).
There are a growing number of protein targets to which selected aptamers bind. These include enzymes (2)(20)(22)(23)(37)(47)(81)(82)(83)(84)(85), growth factors (19)(39)(86)(87), antibodies (38), gene regulatory factors (88)(89)(90)(91)(92), cell adhesion molecules (21)(40)(93)(94), and lectins (95). Intact viral particles (96) and pathogenic bacteria have also been used to obtain high-affinity aptamers against complex targets whose binding epitopes were not well understood. The equilibrium dissociation constants, Kd, of aptamers range between a few picomoles per liter and a few nanomoles per liter. Therefore, the affinities and specificities of aptamers are comparable to, if not better than, those of antibodies that are used in diagnostic applications. The reader will find several interesting articles reviewing the properties of aptamers, including their affinities, specificities, and molecular structures, that have been identified to recognize a broad spectrum of target molecules (58)(59)(97)(98)(99)(100).
As discussed above, RNA libraries bearing suitable modifications at the 2' position of the sugar have yielded aptamers that are nuclease resistant, thus circumventing the potential limitation of RNA aptamers to measure analytes in biological fluids that contain nucleases. On the other hand, compared with the unmodified RNA sequences, unmodified DNA sequences are generally more nuclease resistant. In fact, DNA aptamers with and without terminal modifications have been effective in biological fluids [Lin and Jayasena, submitted for publication, and Refs. (20)(47)(101)(102)]. Therefore, it is reasonable to assume that DNA aptamers without further modifications could perform in diagnostic assays in which aptamers may come into contact with biological fluids for a brief period of time. If necessary, additional protection from exonucleases can be provided through terminal capping with small molecules such as an amine linker, a phosphate group, or an inverted thymidine residue. For most diagnostic formats, terminal modification of aptamers is expected to be common and provides a route to conjugate aptamers either to reporter molecules or to solid supports.
In the case of the monoclonal antibody technology, a desirable antibody
is selected upon screening of a large number of hybridomas. When the
antibody in hand is not optimal for use, another antibody with improved
characteristics is sought either by the screening of more clones or by
starting the entire process from the beginning. In other words,
improvement of the properties of an existing antibody is not generally
feasible. On the other hand, aptamer technology is gifted by its
capacity to further optimize the characteristics of aptamers obtained
from a SELEX experiment (Fig. 4
). The binding affinity and specificity of an aptamer could be
further improved by subjecting the sequence to a second selection in
which case a biased library is created based on the primary sequence of
an existing aptamer. The biased library increases the complexity of
molecules, mostly represented by the variants of the primary aptamer,
that were not present in the completely randomized library used in the
initial SELEX experiment. As discussed above, the complexity of a
working library is not completely represented by all of the
theoretically possible individual sequences. Thus, the biased library
may present another opportunity to select the best aptamer, which could
have been missed during the initial selection. Conceptually, this
approach is analogous to the "lead optimization" strategy used by
medicinal chemists to identify a potent drug candidate based on the
structure of a lead compound. Biased libraries have been used to
optimize characteristics of aptamers and ribozymes successfully
(64)(68)(79)(103)(104)(105)(106).
This approach has also been used to change the specificity of an
existing aptamer. A biased library, based on an aptamer sequence that
recognizes L-citrulline, was subjected to
reselection for binding to L-arginine. This
second selection yielded aptamers that bound to
L-arginine and completely lost their ability to
bind to L-citrulline (67). This is an
interesting example that demonstrates the power of in vitro selection
that allows for changing and further modifying the properties of
aptamers on demand.
|
|
Aptamers in Different Diagnostic Assay Formats |
|---|
|
TopAbstractIntroductionAntibodiesAptamersAptamers in Different Diagnostic...ConclusionReferences |
|---|
The two-site binding assays based on aptamers reported to date involved an antibody as the second ligand. This has been mainly because of the lack of two aptamers that do not share identical or overlapping binding sites on a target of interest. One of the reasons for the lack of aptamers that bind a target at nonoverlapping binding sites could be related to the way in which the selections are generally performed. In a typical SELEX experiment directed to identify the highest affinity aptamers, the affinity enrichment of evolving libraries is pushed to the limit. These efforts generally lead to aptamers with the tightest binding directed to a unique site on the target. It would be reasonable to expect to find aptamers that bind to different binding sites of a target, perhaps with somewhat lower affinity, in the intermediate cycles of the selection. In addition, alternative strategies could also be used to identify aptamers with nonoverlapping target-binding sites. An aptamer already identified for a target could be used in a subsequent SELEX experiment to direct aptamers elsewhere on the same target, or the target could be presented in a different manner in two different SELEX experiments to expose different epitopes of the same target. Aptamers that share nonoverlapping binding sites on a target do exist without the use of any of these techniques. Lochrie et al. (91) had selected aptamers that bound either to nucleocapsid protein or to the matrix protein in the HIV type-1 gag polyprotein target. These two classes of aptamers did not compete for binding to the HIV-1 gag polyprotein. Tasset et al. (20) described the identification of DNA aptamers that bound human thrombin at a site different from the site bound by aptamers isolated by Bock et al. (47). There is a striking similarity between the DNA aptamers identified by the two groups aimed at thrombin binding. Both classes of aptamers have the ability to adapt intramolecular G-quartet structures. However, one class of aptamers bound thrombin at the fibrinogen-recognition exosite (47), whereas the other class bound at the heparin-binding exosite (20). As described previously (20), the identification of two classes of aptamers with nonoverlapping binding sites could have been the result of the two different partitioning methods used in the selections. In one case, thrombin free in solution had been challenged with the single-stranded DNA library, whereas in the other, thrombin immobilized through a glycosylated site was used as the target. Immobilization of the target may have hindered an otherwise preferred binding site, directing aptamers to a different site. Such oligonucleotide ligands with nonoverlapping binding sites on a target protein were used to design a sandwich assay exclusively based on aptamers (Y. Lin and S.D. Jayasena, unpublished results) to detect human thrombin.
Understanding of how aptamers interact with their cognate targets is an intense area of research (108)(109)(110)(111)(112). Molecular details of how aptamers interact with small molecular targets have been obtained by nuclear magnetic resonance spectroscopy (110)(111)(112)(113)(114)(115)(116). These experiments revealed several interesting insights on the subject: (a) Aptamers are folded into unique overall shapes to form intricate binding pockets to accommodate their targets. (b) Functional groups scattered on an aptamer are brought to close proximity to form a cluster of molecular forces that specify target interaction. (c) Aptamers discriminate molecules that are closely related to cognate targets at the atomic level. Nuclear magnetic resonance studies of several aptamer-target complexes indicated that small molecular targets are buried within the binding pockets of aptamers, leaving very little surface to interact with a second molecule. This may limit the possibilities of finding a second aptamer that could interact with a small molecular target that is already bound to the first aptamer.
One experiment that has great value in diagnostics, especially in
detecting small molecular targets, and is still waiting to be carried
out with aptamers is the identification of aptamers that bind to
aptamer-target complexes. One could be optimistic on the success of
isolating aptamers to aptamer-target complexes, based simply on the
fact that nucleic acids interact well with one another. Such aptamers
would be useful in designing two-site binding assays on small molecular
targets and would eliminate the need for antibodies on large protein
targets (Fig. 5
). A diagnostic assay that utilizes an antibody that recognizes
an antibody-target complex is referred to as an "anti-immune complex
immunoassay". An example of this format has been described for
measuring digoxin in serum samples (117). In this assay, a
secondary antibody that specifically recognizes digoxin only when it is
bound to the primary antibody was used. Although the basic idea of this
assay format warrants rapid expansion, the difficulty in generating
secondary antibodies that bind small analyte-primary antibody complexes
may have hampered these efforts. The identification of secondary
aptamers that recognize targets bound to their primary aptamers should
expand this assay format.
|
The exquisite specificity of aptamers should be considered carefully in diagnostic applications. Aptamers have been shown to discriminate closely related molecules from their targets on the basis of small structural changes, such as a methyl group (25)(26), a hydroxyl group (27)(28), and a urea vs a guanidino group (64). They also exhibit a high degree of chiral discrimination of target molecules. This level of specificity in aptamers rivals that of monoclonal antibodies (118)(119)(120). Monoclonal antibodies have been used to develop assays that distinguish minute differences between targets; for example, immunoassays that specifically detect glycated proteins (121) or isoenzymes of different tissue origin (122) have been described. Similar assays that require extreme specificity should be possible to design with aptamers. Assays based on monoclonal antibodies aimed at detecting certain types of targets with molecular heterogeneity, such as pituitary glycoprotein hormones, have been troublesome (123)(124)(125)(126) because of the high specificity of antibodies. These challenges may also exist in aptamer-based assays for detecting protein heterogeneity. However, a cocktail of aptamers, each designed to bind specific variants of the target, could be generated with speed—especially using the automated SELEX process—for assays that are specific for target analytes with molecular heterogeneity.
The catalytic power of enzymes is often harnessed to enhance
sensitivity of diagnostic assays. In ELISAs, the secondary antibody is
conjugated to an enzyme such as alkaline phosphatase or horseradish
peroxidase to catalyze a reaction that generates the signal. Although
it is not discussed here, in vitro selection, the same technology that
is used to discover aptamers, is being used to discover
oligonucleotides that catalyze novel reactions. In the future, it
should be possible to use catalytic oligonucleotides (ribozymes or
deoxyribozymes) to carry out signal generation in diagnostic assays
(Fig. 6
), pushing proteins completely out of the picture. Recently,
Wilson and Szostak (127) described the isolation of aptamers
that bind to a fluorophore, a sulfonated rhodamine derivative. Upon
screening of several aptamers, the authors found aptamers (or
deoxyribozymes) that catalyze oxidation of a related molecule,
dihydrotetramethylrosamine, to its oxidized form that emits
fluorescence. Although the activity of the described oligonucleotides
is weak, these deoxyribozymes could be further optimized to enhance
their catalytic power, making them valuable for diagnostic
applications.
|
flow cytometry
Flow cytometry is a powerful analytical tool that allows
multiparameter analysis of cells and microsphere particles. Today, the
technique is being used in basic research as well as in clinical
diagnostics. Diagnostic assays can be constructed on microsphere
particles bearing ligands or analytes. Because particles can be
distinguished on the basis of their color and size, flow cytometry is
an attractive platform for multiplex analysis. Flow cytometers equipped
with state-of-the-art signal processing software programs can measure
these properties on thousands of particles in samples containing very
small amounts of material in a matter of a few seconds. Microspheres
dyed with a spectrum of colors coated with different ligands such as
antibodies and oligonucleotides provide a way to multiplex diagnostic
assays in flow cytometry (128).
Flow cytometry has been used to detect the binding of aptamers to their cognate proteins presented on either microspheres or cell surfaces. A DNA aptamer selected to recognize human neutrophil elastase (HNE) (23) was modified to attach fluorescein at different positions, away from the target binding site, using different linkers, and was used to stain HNE-coated beads for flow cytometry (129). Although every single derivative of the aptamer tested bound the target-coated beads, the signal intensity was dependent on the way in which fluorescein was attached. Fluorescein conjugated through an ethylene glycol linker gave the highest signal intensity, which was even better than that of the fluorescein-conjugated antibody. For cellular staining experiments, aptamers have been selected using recombinant proteins immobilized on beads either by chemical biotinylation (40) or by an expression tag (21)(102)(130). Full-length aptamers containing 2'-F-pyrimidines obtained by in vitro transcription have been conjugated to either fluorescein or phycoerythrin to analyze their binding to human CD4 expressed on cell surfaces (40). Both monovalent fluorescein-conjugated RNA aptamers and multivalent aptamer-phycoerythrin conjugates stained CD4-expressing cells. The results clearly indicated that the aptamers specifically stained subsets of cells, namely T-helper cells and monocytes, expressing CD4 in a heterogeneous cell mixture of human peripheral blood mononuclear cells. Aptamers labeled with one color probe performed well in combination with antibodies labeled with a second color probe in experiments aimed at two-color analysis of two different targets on cell surface.
Truncated DNA aptamers labeled with fluorescein have been chemically synthesized for staining cells that express L-selectin (102)(130). Anti-L-selectin aptamers stained both leukocytes and neutrophils that express L-selectin in human peripheral blood mononuclear cell preparations. This result was very similar to what was obtained with an anti-L-selectin antibody. The addition of the unlabeled aptamer completely inhibited the staining of neutrophils by the labeled aptamer but only partially blocked the binding of the anti-L-selectin antibody. The inability to completely block the antibody binding has been attributed to the undesirable binding of the antibody to the Fc receptors on neutrophils. The staining of P-selectin on human platelets was demonstrated with a 2'-F-pyrimidine-containing RNA aptamer (21). The truncated RNA aptamer specific for P-selectin was chemically synthesized and subsequently derivatized with fluorescein. As expected, this aptamer showed specific binding to activated human platelets that express P-selectin but not to resting platelets with no P-selectin on the cell surface.
Antibodies of the IgG class are bivalent, whereas aptamers that come out of selections are typically monovalent. To understand their behavior in their bivalent forms, aptamers have been dimerized and studied in flow cytometry (12)(9)(130). Dimeric, therefore bivalent, aptamers had been obtained by solid-phase chemical synthesis starting with a commercially available solid support (symmetric branching CPG) that allows symmetric divergent synthesis of two aptamer sequences linked at 3' ends. The divalent anti-HNE aptamer showed a 10-fold higher affinity than the monovalent form in binding to the target either in solution or immobilized on beads (129). The effects of aptamer dimerization on cellular staining were elegantly studied by Ringquist and Parma (130), who used a DNA aptamer isolated to recognize L-selectin. The aptamer binding to L-selectin on cells showed a 20-fold affinity improvement upon dimerization. The rate of dissociation of the divalent aptamer on cells was 10-fold lower than the monomeric form. The half-life of the divalent aptamer on cell surface was superior to that of the divalent antibody (14 min vs 5.5 min) (130), suggesting that aptamer engineering may lead to the discovery of aptamers with further improved characteristics.
Research aimed at testing the performance of aptamers in flow cytometry revealed some interesting features. Aptamers conjugated to small fluorophores, such as fluorescein, as well as to structurally large proteins such as phycoerythrin (molecular mass, 240 kDa) retain their binding characteristics, indicating their ability to accommodate a wide variety of the reporter molecules generally used in diagnostic applications. However, the signal intensity may be a function of the chemistry of fluorophore attachment and can easily be optimized. Aptamers can perform equally well in lieu of, or in combination with, antibodies. The generation of divalent, or even multivalent, aptamers with enhanced performance characteristics is relatively easy. The presence of the Fc region on antibodies that may interact with the Fc receptors on cells could complicate the outcome of certain cellular staining experiments (130)(131), making aptamers more attractive for such applications.
sensors
Certain applications demand analyte detection within a very short
period of time. The need for techniques that allow rapid detection and
quantification of analytes cannot be underestimated, especially when
dealing with emergency medical situations or in a battlefield. Portable
equipment that could be used as small hand-held devices would be ideal
for such applications. Sensors based on molecular recognition coupled
to a transducer have been developed to meet the needs of rapid
detection. Ideally, at least three basic criteria should be met in an
affinity sensor: (a) the ability to transduce the binding
event without an extra reagent added; (b) the ability to
detect and quantify the target within the desired concentration range
and the time period; and (c) the ability to turn over the
sensing capacity, i.e., to make repeated measurements on the same
transducer multiple times.
Antibody-based immunosensors have been developed as an alternative to immunoassay techniques. In immunosensors, antibodies have been immobilized on sensor surfaces that come into contact with analytes (132). One of the obvious limitations of immunosensors is their poor capacity to regenerate the antibody surface. Mild conditions are required to preserve the integrity of the antibody function. Although immunosensors have been demonstrated to be reusable (133), the loss of activity of surface-immobilized antibodies is inevitable. On the other hand, several advantages are apparent in aptamer-based sensors. The ability to regenerate the function of immobilized aptamers would be the most attractive feature of aptamers. Being nucleic acids, aptamers could be subjected to repeated cycles of denaturation and renaturation. Several methods are available for aptamer regeneration: heat, salt concentration, pH of the medium, and chelating agents. Except for pH, the other conditions can be varied to any extreme without damaging the aptamer. Extremes of pH should be avoided because they could potentially damage aptamers. As discussed earlier, the ease with which aptamers could be modified for immobilization purposes bestows the second benefit. The attachment of oligonucleotides on different surfaces in a controlled and reproducible manner has been demonstrated (134)(135)(136)(137)(138)(139) and could easily be adapted to aptamers. Homogeneous preparations of chemically synthesized aptamers with appropriate linkers could be deposited precisely on solid surfaces at a desired density. The third benefit of aptamers is the ease of labeling with a wide range of reporters, enabling the design of a variety of detection methods. As discussed below, a fluorophore could be attached to an aptamer such that its characteristics would change upon target binding. One of the attractions of the SELEX technology is that the selections could be carried out under any condition defined by the user to obtain aptamers with the desired characteristics. For example, modified random sequence oligonucleotide libraries containing fluorophores could be subjected to selections to identify aptamers that have the desired binding and fluorescence characteristics for applications in affinity sensors.
Recently, aptamers have been tested in affinity sensors. In one application, a biotinylated RNA aptamer selected to recognize L-adenosine was immobilized on an optical fiber surface derivatized with streptavidin (140). The RNA aptamer on the sensor surface detected L-adenosine conjugated to fluorescein with an affinity similar to that measured in solution. The sensor showed selective binding to L-adenosine with a chiral discrimination of at least 1700-fold. In another application, a DNA aptamer specific for human thrombin was used to detect binding of the target protein by evanescent wave-induced fluorescence anisotropy (141). Here, the 5' end of the aptamer was labeled with fluorescein and its 3' end was modified with an alkyl amine attached to a glass surface. The advantage of this sensor is that it does not require its target to be labeled and, therefore, could be extended to in vivo measurements. The aptamer-sensor was specific for thrombin because there was no change in fluorescence anisotropy when it was challenged with elastase, another basic protein. There was no response when a scrambled DNA sequence labeled with fluorescein was attached, indicating that the sensor was dependent on the specific interaction between the aptamer and target. Repeated measurements of thrombin have been made on the sensor, with the aptamer surface regenerated by the removal of bound thrombin with a denaturant. Although research on aptamer-based sensors has been limited, the promising results obtained thus far indicate their potential utility in developing sensors designed for specific molecules.
An unexplored area of aptamers is in sensors based on electrochemical detection. Aptamers, being polyanionic, may be attractive for sensing the changes in conductance in the presence and absence of target binding. An interesting area of nucleic acid research is the understanding of the principles behind the charge transfer within the DNA helix (142)(143). Although this research is still in infancy, it has great potential in the area of molecular sensing. To date, electron transfer studies have focused purely on DNA strands, but soon they will expand to understand the changes mediated by interacting DNA with other molecules. Applications of aptamers in various diagnostic formats may benefit from these novel discoveries.
fluorescence polarization
Homogeneous assay formats are attractive for their simplicity,
ease of use, speed, and applicability in high-throughput screening. One
homogeneous assay format based on fluorescence polarization (FP) is
being used to measure small target analytes such as steroids in clinics
(144). These are competitive assays based on the interaction
between unlabeled antibodies and their small analytes conjugated to a
fluorophore. The change in FP is a direct reflection of the change in
the tumbling rate of the molecule conjugated to the fluorophore upon
binding to a structurally large antibody (Fig. 7
). The competitive binding of the unlabeled analyte present in
the test sample to the antibody decreases the FP signal in the assay.
Competitive assays are rapid, easy-to-use, and amenable for
high-throughput screening. There are, however, several shortcomings
associated with competitive assays. They are less sensitive, have a
narrow dynamic range, and require precise control of the reagents used.
The inherent nature of antibodies dictates that antibody-based FP
assays be competitive. Structurally bulky antibody molecules tumble
slowly in solution. Therefore, the change in the tumbling rate of
an antibody molecule upon its association with a small analyte is
insignificant. As a result, the change in FP that occurs when a
fluorescently labeled antibody binds to its target could be very small
or even undetectable. In addition, homogeneous preparations of
antibodies labeled with a fluorophore are difficult to obtain. The
scenario with aptamers is, however, expected to be different. In
aptamer-based FP assays, the fluorescence tag could be on the aptamer
itself rather than on the analyte. This arrangement allows these assays
to become noncompetitive. The potential use of fluorescently labeled
aptamers in FP assays is justified in two ways: (a) Aptamers
are relatively small compared with antibodies (one-tenth of the size of
an antibody), and therefore tumble faster than antibodies. The change
in tumbling rate, and thus the change in FP, is expected to be large
when the aptamer interacts with the target. (b) Results
obtained from nuclear magnetic resonance experiments indicate that
aptamers selected to bind small targets undergo target-induced
conformational changes (111)(113). These
conformational transitions induced by target recognition on aptamers
are likely to affect the tumbling rates of aptamers as well.
|
A DNA aptamer identified to recognize HNE was labeled with fluorescein
at the 5' end and used in a FP assay to detect HNE (Fig. 8
). A significant change in FP was observed when the labeled
aptamer interacted with HNE. The addition of a nonspecific target,
human serum albumin, produced a very small change in FP only at high
concentrations. This demonstration suggests that fluorescein-labeled
aptamers could be used to design noncompetitive FP assays for target
detection. Aptamers selected to recognize aminoglycoside antibiotics
have been used in competitive FP assays to understand their binding
characteristics to a family of aminoglycosides
(74)(145). The change in FP, or anisotropy, of a
fluorescein molecule attached to an antithrombin aptamer has been
monitored during its interaction with thrombin in an aptamer-based
biosensor (141). The fluorophore on the aptamer was excited
by an evanescent field to measure the change in polarization as a
function of thrombin concentration at the interface. The thrombin
biosensor detected between 0.7 and 700 amoles of thrombin in a 140-pL
volume. Contrary to the thrombin assay, the fluorescein-labeled aptamer
was free in solution in the HNE assay described in Fig. 8
.
|
Aptamers that change the fluorescence characteristics of a fluorophore attached either to the aptamers themselves (for noncompetitive assays) or to the analyte (for competitive assays) would be useful. In fact, aptamers directed to bind flavins have been shown to quench the fluorescence of the target when it is bound by the aptamer (66). Thus, it is possible that targets that inherently fluoresce could be detected by aptamers by measuring the changes in fluorescence.
communicating with molecular beacons
Recently, a novel class of fluorogenic probes called molecular
beacons was introduced for homogeneous detection of nucleic acid
sequences (146)(147). Molecular beacons are
simple hairpin-loop probes in which a fluorophore is attached to one
terminus and a quencher is attached to the other (Fig. 9
). This mode of attachment brings the fluorophore close to the
quencher when the molecular beacons are folded into hairpins; the
fluorescence is then quenched by the formation of a nonfluorescent
complex between the fluorophore and the quencher (147). The
nucleic acid sequence in the loop of the molecular beacon is designed
to be complementary to the target of interest. The loop of the
molecular beacon interacts with its target sequence to form an
intermolecular hybrid, during which the stem of the beacon unfolds to
move the fluorophore away from the quencher. The end result is the
emission of fluorescence from the previously nonfluorescent beacon.
Homogeneous detection of nucleic acid sequences specific to pathogens
has been demonstrated using molecular beacons (148). Until
recently, molecular beacons were limited to detecting nucleic acid
targets because of their lack of productive interaction with other
classes of targets such as proteins.
|
Aptamers with the ability to interact with nucleic acids as well as other classes of targets have been used to close the communication gap between molecular beacons and targets other than nucleic acids (Lin and Jayasena, submitted for publication). Molecular beacons used in this approach are designed to interact with aptamers and are called ligand beacons. The loop of the ligand beacon is complementary to a nucleotide region, preferably containing unpaired nucleotides, in the aptamer. The nucleotide region of the aptamer targeted by the ligand beacon should allow efficient intermolecular hybridization only in the absence of the target protein. However, when the aptamer is bound to its target, the ligand beacon should be unable to interact productively with the aptamer. This strategy allowed the design of homogeneous and competitive assays to detect proteins for which high-affinity aptamers have been isolated (Lin and Jayasena, submitted for publication). Target proteins have been detected in buffer as well as in plasma, demonstrating the potential application of this approach to clinical practice.
Simultaneous analysis of more than one analyte (multiplexing) provides another dimension into the advantages in an assay. Several molecular beacons, each conjugated to a unique fluorophore that emits at a wavelength different from the others, have been used to analyze the presence of multiple nucleic acid targets (147). The same multiplex approach could be adapted to the ligand beacon assay to measure more than one target in a single tube. A variation of the theme is the use of beads that are easily distinguishable in flow cytometry. The ligand beacon assay has been extended to multiplex analysis to detect more than one protein in a single tube using a flow cytometry platform (Heil et al., manuscript in preparation). This assay, called the solid-phase ligand beacon assay, has aptamers immobilized on beads and ligand beacons in solution. Because the fluorescent measurements take place on beads and not in solution, this format eliminates the need for actual beacons containing a quencher and does away with fluorescein-labeled DNA probes complementary to aptamers.
In this format, ligand beacons are expected to displace molecules that nonspecifically interact with aptamers, further improving the specificity of assays. In addition, a ligand beacon could also interfere with the specific aptamer-target interaction. This is expected to be especially prominent in cases in which the aptamer-target interaction becomes weak. Thus, aptamers that form tight complexes with slow off-rates are the ideal candidates for the ligand beacon assay. Photo-cross-linkable aptamers that form covalent linkages with their cognate targets are expected to further improve the performance of the ligand beacon assay. Clearly, the ligand beacon assay represents a unique application of aptamers that cannot be replaced by antibodies.
capillary electrophoresis
Capillary electrophoresis (CE) uses the same separation mechanism
used in conventional electrophoresis in a capillary format. The
technique offers the advantages of speed, the use of small sample
volumes, suitability for automation, sensitivity, and possible
multiplex analysis. In CE, component separation occurs on the fly, and
the separated components are identified by online detection. Whereas
conventional gel electrophoresis requires additional detection steps
after electrophoresis, CE combines these two steps. CE is now being
adapted to immunoassays in which antibody-antigen complexes are
separated on a fluidic stream under an applied electric field. Both
competitive (149) and noncompetitive immunoassays
(150) have been performed with CE. In competitive assays,
the unlabeled target in a test sample competes with the same target
labeled with a fluorophore for binding to a limited amount of antibody.
In contrast, in noncompetitive assays, a large excess of fluorescently
labeled antibody interacts with unlabeled target. As described above,
competitive assays have certain limitations. Therefore, noncompetitive
assays are preferred whenever possible. Practical challenges associated
with immunoassays run on a CE format make it difficult to develop
noncompetitive assays. Poor analytical separation is expected between
labeled antibody-target complexes and free labeled-antibodies when the
targets are small and uncharged. Antibodies could exert heterogeneous
electrophoretic patterns, possibly resulting from different degrees of
glycosylation. Moreover, homogeneous preparations of antibodies labeled
with small molecules are difficult to obtain (151).
These limitations of antibodies have prompted German et al. (152) to use an aptamer in CE based on affinity probes, the method referred to as affinity probe capillary electrophoresis (APCE). The authors used a fluorescently labeled DNA aptamer previously isolated to recognize human IgE (38) to detect and quantify human IgE in the presence and absence of serum in APCE. APCE yielded well-separated peaks corresponding to the free aptamer and aptamer-IgE complex as detected by laser-induced fluorescence. The addition of serum-containing IgE samples had little or no effect on the separation and the characteristics of the two peaks, indicating that the binding of the aptamer to the target IgE was not affected in the context of a complex mixture. The apparent Kd value of the fluorescein-conjugated aptamer measured by APCE was approximately sixfold higher than the value obtained for the unlabeled aptamer using the nitrocellulose filter binding technique (38). Although the reduced affinity of the labeled aptamer in APCE could be attributable to the attachment of fluorescein (152), the most likely cause could be the electric field applied to separate the free and bound aptamer. Successful results in APCE have been obtained with an aptamer that interacts weakly (Kd ~ 200 nmol/L) with human thrombin as well.
Aptamers have uniform charge/mass ratios, and therefore have predictable behavior in electrophoresis. They also undergo structural transitions when they interact with small targets. Such structural transitions, accompanied with the change in mass upon binding to the target, could lead to changes in electrophoretic patterns. On the other hand, aptamer-target interactions may be sensitive to an electric potential because of the charged nature of aptamers. The advantage of the aptamer technology is the ability to demand desirable characteristics from aptamers. For example, selection conditions and separation methods could be defined to obtain aptamers that not only form stable complexes with targets but also give a resolvable mobility shift in an electric field when they bind to targets. In other words, CE itself could be used as a separation technique during the SELEX process to isolate aptamers aimed at APCE applications. Interestingly, the aptamers that have been successfully used in CE were not identified in this manner.
aptamers as molecular switches
One of the inherent characteristics of oligonucleotide aptamers is
their robustness under repeated cycles of denaturation and renaturation
conditi
). Each
nucleotide position in the contiguous random region is synthesized upon
delivery of a mixture of phosphoramidites containing all four building
blocks: A, G, C, and T. To obtain an unbiased library with equal
representation of all four nucleotides, the ratios of the four
phosphoramidites in the mixture are adjusted based on the coupling
efficiencies of individual monomers. Fixed sequences are used for
primer binding sites in enzymatic amplification of individual
sequences. Because aptamers are identified from DNA as well as from RNA
libraries, chemically synthesized DNA libraries are enzymatically
converted to RNA libraries. A promoter for an RNA polymerase (for
example, T7 RNA polymerase) is engineered within a fixed sequence
(
) of
the library to accommodate in vitro RNA synthesis (
represents 2,6-diaminopyrimidine and
X represents xanthine (
-electron system
capable of fluorescing. (B), a possible extension of
such catalytic aptamers to in vitro diagnostic applications. This
example shows an ELISA format in which an aptamer is the secondary
ligand in the two-site binding assay. The binding aptamer is conjugated
to a catalytic aptamer that catalyzes the fluorogenic transformation to
generate fluorescence signal.
) in a buffer consisting of 150 mmol/L
NaCl, 100 mmol/L Tris-HCl (pH 7.0), 6 mmol/L KCl, and 2 mmol/L
MgCl2 at 37 °C for 10 min. FP was measured at ambient
temperature using a Beacon Fluorescence Polarization System, model
P2000 (Pan Vera).