Aptamers as Ligands for Affinity Chromatography and Capillary Electrophoresis Applications

Eric Peyrin

15.1

Introduction

Since the first reports of the systematic evolution of ligands by exponential enrichment (SELEX) procedure by three independent laboratories (Ellington and Szos-tak, 1990; Tuerk and Gold, 1990; Robertson and Joyce, 1990), the development of in vitro selection has allowed the discovery of aptamers against various targets, such as small molecules, including amino acids and nucleosides, proteins, cells, etc. (Jayasena, 1999). The use of aptamers as tools in analytical chemistry is a very promising and exciting field of research due to their ability to bind specifically the target molecules with an affinity equal or superior to those of antibodies. Aptamers present many advantages over antibodies (O'Sullivan, 2002). They can be regenerated within minutes via a denaturation-renaturation step and are, at least for DNA aptamers, relatively stable over time. The in vitro selection can be manipulated to obtain binding and kinetic properties desirable for specific assays. Aptamers can be produced by chemical synthesis with little or no batch-to-batch variation and reporter molecules can be attached to aptamers at precise locations. Furthermore, they are produced through an in vitro process which does not require animals. Taking all these background features into account, various analytical aptamer-based formats have been exploited, including ELISA-type (ELONA) assays (Ito et al., 1998), biosensors (aptasensors) (Potyrailo et al., 1998), aptazymes (Famulok, 2005), flow cytometry (Davis et al., 1996) or separation techniques. The applications of aptamers as specific ligands in affinity chromatography and capillary electrophoresis are addressed in this chapter.

15.2

Aptamers as Ligands in Affinity Liquid Chromatography (and Electrochromatography)

General Principles of Affinity Chromatography

Affinity chromatography is rapidly becoming the separation method of choice in biochemical, biotechnology, and pharmaceutical sciences. It can be defined as a liquid chromatographic technique that makes use of a biological interaction for the separation, analysis, and purification of specific analytes within a sample. The ligand of interest is classically immobilized on a chromatographic support covalently or via a streptavidin-biotin interaction. The simplest operating scheme (direct approach) for affinity chromatography involves the injection of the sample onto the affinity column under conditions in which the target analyte will bind to the immobilized ligand with high affinity. Because of the specificity of the ana-lyte-ligand interaction, undesirable solutes in the sample tend to have little or no affinity for the immobilized ligand. The elution buffer is then applied to dissociate the retained analyte. Alternatively, chromatographic (or flow-injection assays) based on competitive or sandwich approaches have been developed for the determination of trace analytes that do not produce a detectable signal by themselves in direct affinity chromatography.

When low-to-moderate affinity ligands are immobilized to the chromatographic support, the chromatographic analysis is performed under isocratic operating conditions, allowing the elution of different analytes in relation to their respective affinities for the stationary phase. Although various ligands have been used in affinity chromatography, the most popular format uses the high-affinity and specificity of antibodies to create efficient immunoaffinity columns. However, there are some constraints that reduce the effectiveness of antibodies. The linkage of antibodies to column often results in couplings that are not uniform, leading to reduced binding capacity and can allow leaching of the antibody from the column. Furthermore, antibodies are relatively large, which limits the ligand density at the chromatographic surface. Finally, the elution conditions can be harsh, requiring extremes of pH, detergents, organic solvents, or chaotropic salts, leading to dena-turation of the antibody and possibly the target (protein for example).

Expected advantages of aptamers relative to antibodies for affinity chromatogra-phy include their smaller size, enabling higher density stationary phases, novel approaches for elution, and the possibility of immobilizing the ligand to a chro-matographic surface at a precise location. Moreover, a number of recent reports have shown great interest in the use of immobilized aptamers (DNA or RNA) as affinity ligands in chromatography or electrochromatography. To date, these techniques have been applied to the separation/purification of proteins, separation of small molecules, or chiral resolution (Table 15.1).

Table 15.1 Aptamers as ligands in affinity chromatography for analyte capture and separation

Targets/Species Oligo- Separation Appli- References núcleo- systems cations tides

Proteins

L-selectin DNA LC Capture

Thrombin DNA CEC Capture

HCV RNA polymerase RNA LC/chip Capture and replicase

Non-target proteins G-quartet CEC Separation DNA

Romig et al., 1999

Connor and McGown, in press

Chung et al., 2005

Rehder and McGown, 2001;

Rehder-Silinski and McGown,

Small molecules

Non-target analytes

G-quartet CEC DNA

Adenosine and analogs Flavin mononucleotide and other molecules

Nano-LC

Separation Kotia et al., 2000; Charles and McGown, 2002; Vo and McGown, 2004 Separation Deng et al., 2001; Capture Deng et al., 2003 Separation Clark and Remcho, 2003a,b

Enantiomers

Vasopressin

Adenosine Amino acids and derivatives

DNA DNA RNA i-RNA

narrowbore column Micro-LC Micro-LC

Separation Michaud et al., 2003

Separation Michaud et al., 2004 Separation Michaud et al., 2004; Brumbt et al., 2005; Ravelet et al., 2005

LC, liquid chromatography; CEC, capillary electrochromatography.

Separation/Purification of Proteins

Drolet and co-workers (Romig et al., 1999) presented the first work concerning the use of an immobilized DNA aptamer as an affinity stationary phase. An aptamer of 36 nucleotides in length, specific and with high affinity (Kd= 2 nmol/L) for human L-selectin, was applied to the chromatographic purification of recombinant L-selectin-immunoglobulin (Ig) fusion protein from Chinese hamster ovary cell-conditioned medium. The 5'-biotinylated anti-L-selectin DNA aptamer was immobilized on a streptavidin sepharose support which was packed in a short column.

Time (min)

Fig. 15.1 Aptamer affinity chromatography of from the column by a linear EDTA gradient.

partially purified L-selectin receptor globulin fractions. Approximately 220 mg of total protein loaded on a 1 mL aptameric column at a volumetric flow-rate of 0.75mL/min. After washing the column with phosphate-buffered saline (PBS), L-selectin receptor globulin is eluted

Because EDTA absorbs weakly at 280 nm, a slight rise in the baseline is observed for these elution profiles. Arrows indicate the initiation of the gradient. Reprinted from Romig et al. (1999) with permission.

0 10

Time (min)

Fig. 15.1 Aptamer affinity chromatography of from the column by a linear EDTA gradient.

partially purified L-selectin receptor globulin fractions. Approximately 220 mg of total protein loaded on a 1 mL aptameric column at a volumetric flow-rate of 0.75mL/min. After washing the column with phosphate-buffered saline (PBS), L-selectin receptor globulin is eluted

Because EDTA absorbs weakly at 280 nm, a slight rise in the baseline is observed for these elution profiles. Arrows indicate the initiation of the gradient. Reprinted from Romig et al. (1999) with permission.

After loading onto the aptamer affinity column, the L-selectin-Ig fusion protein was eluted under conditions that do not denature proteins: based on the known divalent cation dependence of the active tertiary structure of the aptamer, a linear EDTA gradient appeared to be very efficient to release the captured protein (Fig. 15.1). Application of the aptamer column as the initial purification step resulted in a 1500-fold purification with an 83% single step recovery, demonstrating that aptamers can be effective as affinity purification reagents.

More recently, the use of a thrombin-binding DNA aptamer as a protein capture system in affinity capillary chromatography has been reported (Connor and McGown, in press). The 5'-thiol modified aptamer (15 nucleotides) was covalently attached to the inner surface of a bare fused-silica capillary via an organic linker to serve as stationary phase. After protein capture at 25 °C by incubation overnight, the bound thrombin was released using an elution scheme involving 8mol/L urea and a capillary temperature of 50 °C. The results showed that the aptamer stationary phase was able to bind approximately three times as much thrombin as the control column (scrambled sequence oligonucleotide), in the presence or absence of human serum albumin. However, due to the low binding capacity of the open-tubular capillary, only around 60 pmol of protein can be captured in this system.

Fig. 15.2 Schematic view of micro-affinity purification process using photolytic elution method.

(a) Injection of the protein mixture into the microchip packed with RNA apta-mer-modified microbeads.

(b) Purification of the target protein.

(c) UV irradiation.

(d) Analysis of the photolytically eluted protein. Reprinted from Chung et al. (2005) with permission.

Fig. 15.2 Schematic view of micro-affinity purification process using photolytic elution method.

(a) Injection of the protein mixture into the microchip packed with RNA apta-mer-modified microbeads.

(b) Purification of the target protein.

(c) UV irradiation.

(d) Analysis of the photolytically eluted protein. Reprinted from Chung et al. (2005) with permission.

Via an elegant elution strategy based on a photolytic method (Fig. 15.2), detection of hepatitis C virus (HCV) RNA polymerase and replicase from a protein mixture using a microbead affinity chromatography on a chip has been achieved (Cho et al., 2004; Chung et al. 2005). The RNA aptamer directed against the protein was coupled to the beads using a 7-fluorenylmethoxycarbonyl (Fmoc) photo-cleav-able linker and the aptamer-immobilized beads were loaded and packed in a mi-crofluidic chip composed of a transparent microchannel. A protein sample was injected onto the microchamber and incubated for 5-30min. After washing, the microchannel was irradiated by UV light at 360 nm and the captured protein was eluted with the pumped flow of mobile phase. In the first approach for the detection of HCV RNA polymerase (Cho et al., 2004), subsequent trypsin treatment was performed and the peptide mixture was concentrated and applied to matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF) analysis. The detection limit of this system was estimated to be around 10 fmol of HCV RNA polymerase. In the second approach for the analysis of HCV RNA replicase (Chung et al. 2005), the protein was labeled with fluorescein and the eluted target was quantitatively detected via the fluorescence intensity measurement using a confocal microscope. It has been shown that such system was able to detect the HCV RNA replicase from a mixture containing 170 fmol of the target protein.

Alternatively, McGown's group has used the two-plane G-quartet sequence of the antithrombin DNA aptamer and an analogous four-plane G-quartet oligonucleotide of greater stability to separate various non-target proteins in open-tubular capillary electrochromatography (CEC). This approach is quite different from those reviewed previously, as the binding and specificity properties of aptamers are not stricto sensu used to discriminate the proteins. Using the same oligonu-cleotide covalent immobilization as that reported above (Connor and McGown, in press), the four-plane G-quartet stationary phase has been used for the separation of bovine ß-lactoglobulin variants A and B (LgA and LgB), which differ by only two amino acid residues (Rehder and McGown, 2001). In control experiments (i.e. with an oligonucleotide of similar base composition but which did not form the G-quartet structure) no separation was achieved, indicating that the folded structure plays a role in the protein discrimination. Furthermore, this approach has been extended to the separation of various bovine milk proteins (Rehder-Silinski and McGown, 2003). Both the two-plane and four-plane structures have been tested. It has been shown notably that the four-plane G-quartet displayed good resolution for the caseins while the two-plane G-quartet stationary phase did not allow complete separation of these proteins. As the experiments were carried out in the absence of stabilizing K+, the authors attributed this behavior difference to the loss of the G-quartet structure for the less stable oligonucleo-tide.

Finally, these two stationary phases have been also evaluated in presence of 1 mmol/L of K+ to separate albumins of different species (Dick et al., 2004). It appeared that the two-plane phase behaved differently from the four-plane, with a significant resolution between albumins from different species and among the variants within a single albumin variant. Compared with the rigid four-plane structure, the two-plane structure was expected to offer greater flexibility that could facilitate interactions with proteins. All these studies have shown that there are advantages to using G-quartet DNA stationary phases for separation of proteins that exhibit only weak, non-specific interactions with DNA.

Separation of Small Molecules

McGown's group has reported the use of DNA aptamers as stationary phases to separate small solutes. Using two-plane G-quartet sequences and open-tubular CEC systems similar to those described above for the separation of proteins, these authors demonstrated that some non-target small analytes such as amino acids and polycyclic aromatic hydrocarbons can also be resolved (Kotia et al., 2000). The influence of the addition of organic solvents in the mobile phase has been tested. It has been shown notably that the increase of acetonitrile to 60% improved the resolution of naphthalene and benzo-perylene. As shown for the protein separation, the G-quartet conformation was found to play a role in the separation of polycyclic aromatic hydrocarbons. In presence of KCl using a hydro-organic mobile phase (30% of methanol), naphthalene and benzo-perylene were partially resolved while no distinct or resolved peaks were observed in the absence of KCl.

Both two-plane and four-plane G-quartet stationary phases have also been investigated to separate the isomeric dipeptides Trp-Arg and Arg-Trp (Charles and McGown, 2002). Although temperature studies suggested that resolution was enhanced when the G-quartet structure was partially destabilized, control experiments in which K+ was not included in the mobile phase confirmed that the folded structure nevertheless played a role in the separation (Fig. 15.3). This work has been further extended using three G-quartet-forming DNA oligonucleotides for the analysis of homodipeptides and alanyl dipeptides in CEC (Vo and McGown, 2004). It has been demonstrated notably that the replacement of G by T in the central loop of the two-plane aptamer increased the affinity of DNA

Fig. 15.3 Separation of non-target analytes Trp-Arg and Arg-Trp (0.5mmol/L each) using 25 mmol/LTris, pH 7.2 as mobile phase, 15 kV voltage, 75 mm internal diameter capillaries, 25 °C.

(a) Electropherogram using a capillary coated with a two-plane G-quartet-forming DNA aptamer (antithrombin aptamer), with 2mmol/L KCl in the mobile phase.

(b) Capillary electrophor-esis using a bare capillary, other conditions same as (a).

(c) Electropherogram without KCl in the mobile phase, other conditions same as (a). Reprinted from Charles and McGown (2002) with permission.

for the dipeptide Ala-Glu, which would indicate that the site of interaction is near the central loop of the G-quartet. The authors have also proposed that the lower stability of this aptamer variant facilitated interactions with the dipeptide by giving greater flexibility to the structure. This was supported by the results for the homodipeptides, in which the resolution of Trp-Trp was better on the two-plane phase than on the more stable four-plane.

Kennedy and co-workers have been the first researchers to report an immobilized aptamer that can selectively retain and separate related target compounds by weak affinity chromatography (Deng et al., 2001). Using the target specificity of the anti-adenosine DNA aptamer, these authors have described the preparation and characterization of an aptamer affinity nano-column for the analysis of ade-nosine and various derivatives such as NAD, AMP, ATP, and ADP. The 3'-bioti-nylated DNA aptamer was immobilized to the streptavidin chromatographic surfaces via a streptavidin-biotin bridge. The DNA-modified beads were packed in fused-silica capillaries of internal diameter of 150 or 50 mm. Using frontal analysis, it was found that (1) the aptamer immobilization did not alter the adenosine-binding properties of the oligonucleotide and (2) a greater surface coverage (about three-fold) was obtained relative to that classically obtained for antibodies. The various species were separated in isocratic conditions in relation to their different dissociation constant values.

The influence of the operating parameters such as particle diameter, capillary internal diameter, buffer composition, ionic strength, and mobile phase pH were evaluated to obtain optimal separation. It appeared notably that when Mg2+ was removed from the column, the affinity of all the analytes was no longer observed (except of some residual retention of adenosine), due to probably the loss of the active Mg2+-dependent tertiary structure.

This aptamer affinity nano-column was further used to develop an efficient adenosine assay in microdialysis samples (Deng et al., 2003). Using an aqueous mobile phase containing 20mmol/L of Mg2+, adenosine was strongly retained on the column.

Various elution schemes have been tested to optimize the target UV detection. Although the elution by chelation of Mg2+ with EDTA would be a suitable approach, a large background disturbance was observed in the chromatograms due to the large change in EDTA concentration. So, a competitive elution with divalent cations such as Ni2+, which is presumed to complex nitrogen atoms in adenosine involved in binding to the aptamer, was preferred (Fig. 15.4). Up to 6 mi-croL of 1.2 micromol/L adenosine could be injected onto the 150 mm (internal diameter) X 7-cm-long nano-column without loss of adenosine. The detection limit was found to be 120fmol. This assay has been successfully applied to the determination of the adenosine concentration in microdialysis samples (without required preparation) collected from the somatosensory cortex of chloral anesthetized rats.

Finally, an immobilized specific oligonucleotide ligand based on an RNA aptamer has been designed for CEC applications (Clark and Remcho, 2003a). A 5'-amino-modified antiflavin mononucleotide (FMN) 35-base RNA aptamer was covalently bound to the inner walls of fused capillaries and used in open-tubular CEC and flow-counterbalanced open-tubular CEC. As the three-dimensional structure of this oligonucleotide is known to be stabilized by the presence of divalent cations, the influence of Mg2+ on the retention of FMN and flavin adenine dinucleotide (FAD) was evaluated. The retention increased for FAD between 0 and 0.5 mmol/L, while the affinity of FMN for the immobilized ligand exhibited a maximum at 0.2 mmol/L. The authors have proposed that the increase to higher

Fig. 15.4 Chromatogram illustrating gradient elution of 1.2 mmol/L adenosine (2 mL injected) from an anti-adenosine DNA aptamer nanocolumn 70 X 0.15 (i.d.) mm. Mobile phase A consists of 20 mmol/LTris, 20mmol/L NaCl, 20mmol/L MgCl2 at pH 6.6 and mobile phase B consists of 20 mmol/LTris, 20 mmol/ L NaCl, and 20 mmol/L NiCl2 at pH 3.45. The gradient is 2% B from 0 to 72 s with a linear increase to 90% B from 72 to 180 s. Reprinted from Deng et al. (2003) with permission.

concentration of divalent cations may shield the phosphate group of FMN so that the interaction between this group and a guanine present in the aptamer was lessened. This aptameric column has been further studied to evaluate its ability to discriminate between the target and other molecules which do not contain the flavin moiety recognized by the aptamer (Clark and Remcho, 2003b). It has been shown notably that FMN and anthracene can be separated in this system. However, the well-known inherent instability of RNA, which is significantly higher than that of DNA due to the ability of the 2'-hydroxyl groups to act as intramolecular nucleophiles in both base- and enzyme-catalyzed hydrolysis, is expected to be the major drawback for use of such RNA stationary phase. Unfortunately, the stability of the immobilized ligand over the time was not evaluated in detail in this study.

Target-specific Chiral Separation

For chiral compounds, the efficient monitoring of the selection procedure has allowed in most cases a very high specificity, exemplified by the ability of the apta-mer to bind the target enantioselectively. The chiral discrimination properties of aptamers selected against a target enantiomer have been used by our group to create a new class of target-specific chiral stationary phases.

First, the enantiomers of arginine-vasopressin were separated using an immobilized 55-base DNA aptamer known to bind stereopecifically the d-enantiomer of the oligopeptide (Michaud et al., 2003). Immobilization was achieved using the biotin-streptavidn interaction, as previously described by Kennedy and co-workers (Deng et al., 2001, 2003). The influence of various parameters (such as column temperature, eluent pH, and salt concentration) on the l- and d-peptide retention has been investigated in order to provide information about the binding mechanism and then to define the utilization conditions of the aptamer column. Very important apparent enantioselectivity was observed, the non-target enantiomer not being retained by the column. More, it has been shown by thermodynamic analysis that both dehydration at the binding interface, charge-charge interactions, and adaptive conformational transitions contributed to the specific d-peptide-aptamer complex formation. Furthermore, it was established that the aptamer column was stable over an extended period of time.

In further work, this approach has been extended to the chiral resolution of small molecules of biological interest (Michaud et al., 2004). The DNA aptamers used were selected against the d-adenosine and l-tyrosinamide enantiomers. An apparent enantioseparation factor of around 3.5 (at 20 °C) was observed for the anti-d-adenosine aptamer chiral stationary phase, while very high enantioselectiv-ity was obtained with the immobilized anti-l-tyrosinamide aptamer. This allowed baseline resolution to be attained even at a relatively high column temperature. The anti-d-adenosine aptameric stationary phase can be used for 2 months without loss of selectivity, while some performance degradation was observed for the anti-l-tyrosinamide column over this period.

Fig. 15.5 (A) Chromatograms for the resolution of arginine using an anti-l-arginine d-RNA chiral stationary phase. Amount of d-, l-argi-nine injected: (a) 10 ng and (b) 100 ng. Column 370 X 0.76 (i.d.) mm; mobile phase phosphate buffer 25 mmol/L, NaCl 25mmol/L, MgCl2 5mmol/L, pH 7.3; column temperature 4 °C; injection volume 100nL; flow rate 50mL/min; detection at208nm. (B) Chromatogram for the

Fig. 15.5 (A) Chromatograms for the resolution of arginine using an anti-l-arginine d-RNA chiral stationary phase. Amount of d-, l-argi-nine injected: (a) 10 ng and (b) 100 ng. Column 370 X 0.76 (i.d.) mm; mobile phase phosphate buffer 25 mmol/L, NaCl 25mmol/L, MgCl2 5mmol/L, pH 7.3; column temperature 4 °C; injection volume 100nL; flow rate 50mL/min; detection at208nm. (B) Chromatogram for the

5 10

Time (min)

resolution of arginine using an anti-d-arginine l-RNA chiral stationary phase. Column 370 X 0.76 (i.d.) mm; mobile phase phosphate buffer 25 mmol/L, NaCl 25 mmol/L, MgCl2 5 mmol/L, pH 7.3; column temperature 4 °C; amount of d-, l-arginine injected 50 ng; injection volume 100nL; flow rate 50mL/min; detection at 208 nm. Reprinted from Brumbt et al. (2005) with permission.

Most of the aptamers reported in the literature are related to RNA sequences (70% of aptamers are RNAs). The ability of RNA aptamers to bind targets with very high stereoselectively has also been observed (Geiger et al., 1996). Thus, the enantioselective properties and the stability of an anti-l-arginine d-RNA apta-mer target-specific chiral stationary phase have been tested (Brumbt et al., 2005). It was found that this immobilized ligand was very quickly degraded by RNases under usual chromatographic utilization and storage. In order to overcome this severe limitation for a practical use, it appeared fundamental to develop a RNA molecule intrinsically resistant to the classical cleaving RNases. A very interesting strategy involving the mirror-image approach has been successfully developed to design biostable l-RNA ligands (Spiegelmers) for potential therapeutic or diagnostic applications (Klussmann et al., 1996). As the structure of nucleases is inherently chiral, the RNases only accept a substrate in the correct chiral configuration (i.e. the "natural" d-oligonucleotide). Thus l-oligonucleotides are expected to be resistant to naturally occurring enzymes. This concept has been successfully applied to create a biostable RNA chiral stationary phase. It was demonstrated that a chiral stationary phase based on l-RNA, that is the mirror-image of the "natural" d-RNA aptamer, was stable for an extended period of time (about 1600 column volumes of mobile phase) under usual chromatographic conditions of storage and experiments. In addition, as expected from the principle of chiral inversion (i.e. the mirror-image of the "natural" aptamer recognizes with the same affinity and specificity the mirror image of the target), d-arginine interacted with the l-RNA stationary phase while l-arginine was not significantly retained by the column. This was responsible for the reversed elution order of enantiomers relative to that obtained using the various d-RNA columns (Fig. 15.5).

Finally, we recently reported for the first time an aptamer-based chiral stationary phase which was able to resolve racemates of not only the target but also various related compounds (Ravelet et al., 2005). The enantiomers of tyrosine and analogs (11 enantiomeric pairs) were separated using an immobilized anti-tyro-sine-specific l-RNA aptamer (Fig. 15.6). It was also found that the immobilized RNA aptamer could be used under hydro-organic mobile phase conditions without alteration of the stationary phase stability (about 3 months of experiments).

Fig. 15.6 Chromatographic resolution of (a) 25mmol/L, MgCl2 5mmol/L, pH 7.4; column tryptophan, (b) 2-quinolyl-alanine, (c) N-acetyl- temperature 10°C; injected concentration tryptophan, and (d) l-methyl-tryptophan using 0.50mmol/L; injection volume 100nL; flow a anti-d-tyrosine l-RNA aptamer chiral sta- rate 15 mL/min; detection at 220 nm. Reprinted tionary phase. Column 350 X 0.76 (i.d.) mm; from Ravelet et al. (2005) with permission. mobile phase Tris-HCl buffer 8mmol/L, NaCl

Fig. 15.6 Chromatographic resolution of (a) 25mmol/L, MgCl2 5mmol/L, pH 7.4; column tryptophan, (b) 2-quinolyl-alanine, (c) N-acetyl- temperature 10°C; injected concentration tryptophan, and (d) l-methyl-tryptophan using 0.50mmol/L; injection volume 100nL; flow a anti-d-tyrosine l-RNA aptamer chiral sta- rate 15 mL/min; detection at 220 nm. Reprinted tionary phase. Column 350 X 0.76 (i.d.) mm; from Ravelet et al. (2005) with permission. mobile phase Tris-HCl buffer 8mmol/L, NaCl

15.3

Aptamers as Ligands in Affinity Capillary Electrophoresis

General Principles of Affinity Capillary Electrophoresis

Capillary electrophoresis (CE) may be carried out at high field strengths and provides highly efficient separation of a wide range of analytes. Affinity CE can be defined as electrophoretic separation where the separation patterns are influenced by molecular binding interactions taking place during the separation process. Whereas many of the affinity techniques have been adapted to CE, one area of great importance is the characterization of affinity interactions and the quantification of analyte concentration. Ligands may interact with the analyte before and/or during electrophoresis. The analysis of preincubated target-ligand mixtures is an approach that is especially useful for quantitative measurements of analytes that form high-affinity complexes with the ligand. Such CE assays are based on separating the analyte-ligand complexes from the free analyte and free ligand by CE.

Both competitive and non-competitive (direct) assays can be used. In the competitive assay, the analyte is labeled and competes with the unlabeled solute in the sample for binding to a limited amount of the corresponding ligand. A CE separation of the mixture produces two distinct peaks corresponding to the free labeled analyte and the labeled analyte bound to the ligand, allowing the quantification of the target. In the direct assay, the ligand is labeled and added at a constant concentration to the sample for binding the target. Detection and quantification of the complex peak or free ligand peak is related to the amount of the analyte in solution. Nearly all CE assays rely on laser-induced fluorescence detection because of the sensitivity and selectivity of detection. Various interacting systems have been studied by CE using proteins, antibodies, or double-stranded DNAs as specific ligands. CE assays based on the formation of antigen-antibody complexes are the most frequently reported format.

In principle, direct assays possess several advantages over competitive assays, including a larger dynamic range, detection limits that are less dependent upon binding constant between the analyte and the ligand, and the ability to distinguish between cross-reactive species. However, antibodies present some drawbacks for application in such a non-competitive approach. They are electrophore-tically heterogeneous and do not migrate as a single sharp peak. Moreover, the fluorescent label may interfere with binding if it is too close to the binding site. On the other hand, aptamers possess several advantages over antibodies that render them especially valuable in direct affinity CE. They are simple to label fluorescently, and they possess predictable electrophoresis properties and a relatively low molecular mass, which simplifies the separation of complex from free aptamer. At the present time, aptamers have been used in affinity CE in a direct approach for the quantification of protein and determination of binding parameters (Table 15.2).

Table 15.2 Aptamers as ligands in affinity capillary electrophoresis for protein quantification

Proteins

Oligonucleotides

Particular methodologies

References

IgE

DNA

_

German et al., 1998;

Buchanan et al., 2003

HIV type 1 reverse

DNA

-

Pavski and Le, 2001

transcriptase

Focusing

Wang et al., in press

Thrombin

DNA

-

German et al., 1998;

Buchanan et al., 2003

Stabilizing PEG

Huang et al., 2004

NECEEM

Berezovski et al., 2003

Antithrombin III

DNA

Competitive

Huang et al., 2004

PEG, polyethylene glycol; NECEEM, non-equilibrium capillary electrophoresis of the equilibrium mixture.

PEG, polyethylene glycol; NECEEM, non-equilibrium capillary electrophoresis of the equilibrium mixture.

Affinity Capillary Electrophoresis for Target (Protein) Quantification

Kennedy and co-workers first reported the use of aptamers as ligands in affinity probe CE (German et al., 1998). A DNA aptamer directed against IgE was labeled with fluorescein and used as a specific fluorescent tag for the quantification of IgE using laser-induced fluorescence detection. Aptamer solutions and the target were mixed and incubated for at least 3 min and then injected into the separation capillary hydrodynamically. In order to prevent complex dissociation during the electrophoretic run, the analysis time was reduced to less than 1 min, applying vacuum to the outlet and using a separation distance of just 7 cm. Using fluorescein as internal standard, IgE can be detected with a dynamic linear range of 105 and a detection limit of 46pmol/L. It should be noted that the dissociation constant can also be easily calculated from the free labeled aptamer peak via this methodology. It was found that the assay was highly specific and can be conducted in complex samples such as human serum without significant interferences (Fig. 15.7). This suggests that such assays could have broad utility, including clinical applications.

Another study performed by the same group has evaluated the effects of buffer, electric field, and separation time on the detection of aptamer-ligand complexes (Buchanan et al., 2003). The results showed that the best conditions for the detection of the complexes involved the use of the minimal column length and electric field necessary to achieve separation.

A similar CE assay has been reported for the determination of HIV type 1 reverse transcriptase (HIV-1 RT) using specific DNA aptamers labeled with 5'-car-

Fig. 15.7 Determination of IgE in serum using an anti-IgE DNA aptamer by a non-competitive affinity capillary electrophoresis method. Electrophero-grams obtained for samples prepared in reconstituted human serum. Each sample contains a final concentration of

0 20 40 60 0 20 40 60 0 20 40 60 Time (sec) Time (sec) Time (sec)

300 nmol/L of labeled aptamer (A*) and 0, 100, and 400 nmol/L IgE in (a)-(c), respectively. Reprinted from German et al. (1998) with permission.

boxyfluorescein (Pavski and Le., 2001). The assay was capable of quantifying up to 50 nmol/L HIV-1 RT and was not affected by the presence of other reverse tran-scriptases. Recently, this CE assay has been further improved by this group using a DNA-driven focusing procedure (Wang et al., in press). The key for focusing is to establish electrophoresis conditions under which the mobility of the DNA-protein complex is between those of the electrolytes in the sample and in the running buffer. This has been achieved by using different anions in the sample and in the running buffer, which have different mobilities. The Tris-glycine buffer was used as the running buffer as glycine is characterized by low mobility while acetate of higher electrophoretic mobility was used in the sample. It was shown that the separation efficiency of the aptamer-HIV-1 RT complex reached 5 million theoretical plates/m and the sensitivity for the detection was enhanced by 70-120-fold.

CE assays have been also performed using an antithrombin aptamer (German et al., 1998; Buchanan et al., 2003). However, as this thrombin aptamer is a significantly weaker binder than the IgE and HIV-1 RT aptamers (dissociation constant Kd of ~200 nmol/L versus Kd of ~10 nmol/L or 1-2 nmol/L), a significant loss of the complex during the electrophoretic process was observed, allowing a low detection limit (40 nmol/L). In order to overcome this severe limitation for unstable target-aptamer complexes, Tan and co-workers proposed the use of a relatively long capillary so that no interference occurs between the free labeled aptamer and the aptamer-thrombin peaks (Huang et al., 2004). In this case, the ap-tamer-thrombin complex would almost completely decay during the long analysis time, which was well suited for detection based solely on changes in the free ap-tamer peak. The increase in thrombin in the sample was then reflected by a large decrease in the free labeled apatmer. As this methodology allowed the linear inactive aptamer conformer peak to be differentiated from the folded G-quadruplex

Fig. 15.8 Determination by affinity capillary electrophoresis of thrombin and antithrombin III using an antithrombin DNA aptamer and a PEG-containing sample matrix. Each sample contained a final concentration of 200nmol/L aptamer and (a) 0, (b) 50, and (c) 200nmol/L thrombin. In (e)-(g), each sample contained a final concentration of 200 nmol/L aptamer, 200 nmol/L thrombin, and (e) 50, (f) 100 and (g) 200nmol/L antithrombin III. (d) and (h) are the calibration curves constructed with various concentrations of thrombin and antithrombin III, respectively. The sample matrix consisted of10mmol/L Tris-HCl, 15 mmol/L KCl, and 2% PEG at pH 8.4. The electrophoresis buffer was 10 mmol/LTris-HCl and 15 mmol/L KCl at pH 8.4. Reprinted from Huang et al. (2004) with permission.

active conformer peak, the limit of detection was improved four-fold (10 nmol/L versus 40 nmol/L) relative to the previously reported data (German et al., 1998).

Another approach reported by the same authors was to stabilize the weak apta-mer-thrombin complex via the addition of soluble linear polymer (polyethylene glycol, PEG) to the electrophoretic buffer (Huang et al., 2004). Such polymers are known to promote the stabilizing cage effect, retard the dissociation step, and increase local concentrations of analytes. Using a short coated capillary, the addition of 2% PEG allowed the aptamer-thrombin peak to be detected with low interference of the free aptamer peak (Fig. 15.8). A detection limit of 10 nmol/L was also obtained using the stabilizing effect of PEG.

On the basis of competitive experiments, this CE assay approach has been extended to the quantification of antithrombin III (Huang et al., 2004). As shown in Fig. 15.8, the increasing concentration of antithrombin III caused a decrease of the aptamer-thrombin complex associated with an increased peak of free apta-mer. This confirmed an earlier report that demonstrated that the binding of antithrombin III to thrombin induced a conformational change in thrombin that rendered the analyte binding to the aptamer unstable (Fredenburgh et al., 2001).

Finally, Krylov and co-authors have proposed a new method that allows the use of low-affinity aptamers as affinity probes in quantitative analysis of proteins (Berezovski et al., 2003). This is based on the non-equilibrium capillary electro-

Fig. 15.9 Electropherograms generated by the non-equilibrium capillary electrophoresis of the equilibrium mixture (NECEEM) of thrombin and antithrombin DNA aptamer. Peak 1 corresponds to the equilibrium fraction of free aptamer (A1 area). Exponential part 2 corresponds to the equilibrium fraction of the complex (A2 area). From the A1 and A2 areas, the dissociation constant Kd between the protein and the aptamer can be easily determined so that target quantification is obtained. The inset illustrates fitting of experimental data (black line) with the single-exponential function (red line). Reprinted from Berezovski et al. (2003) with permission.

Fig. 15.9 Electropherograms generated by the non-equilibrium capillary electrophoresis of the equilibrium mixture (NECEEM) of thrombin and antithrombin DNA aptamer. Peak 1 corresponds to the equilibrium fraction of free aptamer (A1 area). Exponential part 2 corresponds to the equilibrium fraction of the complex (A2 area). From the A1 and A2 areas, the dissociation constant Kd between the protein and the aptamer can be easily determined so that target quantification is obtained. The inset illustrates fitting of experimental data (black line) with the single-exponential function (red line). Reprinted from Berezovski et al. (2003) with permission.

phoresis of the equilibrium mixture (NECEEM) of a protein with its labeled aptamer. Generally, NECEEM generates an electropherogram with three characteristic features: two peaks that correspond to the free aptamer and protein-aptamer complex and an exponential curve, which is ascribed to the complex decaying during separation. The NECEEM method has been applied to the analysis of thrombin using its specific aptamer. The aptamer was indirectly labeled via an additional 16-mer oligonucleotide sequence that was associated to a complementary 15-mer sequence labeled with fluorescein. Using a relative long analysis time, the NECEEM electropherograms of the thrombin-aptamer mixtures showed a first peak corresponding to the free aptamer and an exponential curve associated with the monomolecular decay of the complex; the second peak, which should correspond to the thrombin-aptamer association, was not detected due to the complete dissociation of the complex (Fig. 15.9). From the areas of the first peak (A1) and the exponential part (A2), the dissociation constant was calculated and the unknown concentration of the protein was classically derived from this value. Using this approach, the detection limit of thrombin was found to be 60nmol/L, comparable to that reported previously for the equilibrium method. Such an NECEEM-based method appears to be of interest because it would be universally applicable to the analysis of proteins, even when the aptamer forms an unstable complex with its target.

340 | 15 Aptamers as Ligands for Affinity Chromatography and Capillary Electrophoresis Applications 15.4

Concluding Remarks

Although DNA and RNA aptamers have been exploited successfully in the separation, purification, and quantification of analytes using chromatography or capillary electrophoresis techniques, it some major drawbacks still remain that could limit broad practical applications in routine analysis.

As reported above, the well-known role of nucleases in RNA aptamer degradation is problematic, notably in chromatography or electrochromatography, where the stationary phases have to be reusable and then stable over time. This has been resolved by applying the mirror-image strategy (Klussmann et al., 1996), with special interest in the chiral separation field (Brumbt et al., 2005). Alternatively, the substitution of the 2'-hydroxyl on the ribose by a more stable functional group such as 2'-O-methyl or 2'-fluoro may protect such RNA aptamer ligands.

The stationary phases in high-performance liquid chromatography (HPLC) or CEC are commonly characterized by a limited binding capacity, due possibly to the presence of inactive/different conformers at a given temperature, and slow mass transfer kinetics related, at least in part, to the target binding-dependent conformational changes of the ligand (Michaud et al., 2003; Brumbt et al., 2005). This makes the binding properties of the immobilized aptamer highly concentration dependent, with severe limitations of the efficiency performances and peak shape. This is exemplified in Fig. 15.5A for the separation of the arginine enantiomers on an anti-d-arginine RNA chiral stationary phase. In addition, due to the relatively high cost of the aptamers, the applications have been limited to miniaturized systems in most cases, including chips, capillary electrophoresis, and micro/nano HPLC. These constraints probably preclude, at least at the present time, any applications at the preparative level.

Finally, another major problem is related to the SELEX methodology. Although highly efficient, this requires highly sophisticated equipment, expensive reagents, and can be relatively time-consuming. Furthermore, the number of aptamers directed against small molecules of importance such as the drugs is relatively restrained at this time. However, improved methods of selection have been recently developed (Murphy et al., 2003; Mendosa and Bowser, 2005) which could allow rapid generation of new aptamers of great interest for the development of efficient separation analytical tools in the pharmaceutical or biochemical fields.

References

Berezovski, M., R. Nutiu, Y. Li, S. N. Krylov (2003). Affinity analysis of a protein-apta-mer complex using nonequilibrium capillary electrophoresis of equilibrium mixtures. Anal Chem 75, 1382-1386.

Brumbt, A., C. Ravelet, C. Grosset, A. Ravel, A. Villet, E. Peyrin (2005). Chiral stationary phase based on a biostable L-RNA aptamer. Anal Chem 77, 1993-1998.

Buchanan, D. D., E. E. Jameson, J. Perlette, A. Malik, R. T. Kennedy (2003). Effect of buffer, electric field, and separation time on detection of aptamer-ligand complexes for affinity probe capillary electrophoresis. Electrophoresis 24, 1375-1382.

Charles, J. A. M., L. B. McGown (2002). Separation of Trp-Arg- and Arg-Trp using G-quartet-forming DNA oligonucleotides in open-tubular capillary electrochromatogra-phy. Electrophoresis 23, 1599-1604.

Cho, S., S. Lee, W. Chung, Y. Kim, Y. Lee, B. Kim (2004). Microbead-based affinity chro-matography chip using RNA aptamer modified with photocleavable linker. Elec-trophoresis 25, 3730-3739.

Chung, W., M. Kim, S. Cho, S. Park, J. Kim, Y. Kim, B. Kim, Y. Lee (2005). Microaffinity purification of proteins based on photolytic elution: toward an efficient microbead affinity chromatography on a chip. Electro-phoresis 26, 694-702.

Clark, S. L., V. T. Remcho (2003a). Electro-chromatographic retention studies on a flavin-binding RNA aptamer sorbent. Anal Chem 75, 5692-5696.

Clark, S. L., V. T. Remcho (2003b). Open tubular liquid chromatographic separations using an aptamer stationary phase. J Sep Sci 26, 1451-1454.

Connor, A. C., L. B. McGown. Aptamer stationary phase for protein capture in affinity capillary chromatography. J Chromatogr A (in press).

Davis, K. A., B. Abrams, Y. Lin, S. D. Jayasena (1996). Use of a high affinity DNA ligand in flow cytometry. Nucleic Acids Res 24, 702706.

Deng, Q., I. German, D. Buchanan, R. T. Kennedy (2001). Retention and separation of adenosine and analogues by affinity chromatography with an aptamer stationary phase. Anal Chem 73, 5415-5421.

Deng, Q., C. J. Watson, R. T. Kennedy (2003). Aptamer affinity chromatography for rapid assay of adenosine in microdialysis samples collected in vivo. J Chromatogr A 1005, 123-130.

Dick Jr., L. W, B. J. Swinteck, L. B. McGown (2004). Albumins as a model system for investigating separations of closely related proteins on DNA stationary phases in capillary electrochromatography. Anal Chim Acta 519, 197-205.

Ellington, A. D., J. W. Szostak (1990). In vitro selection of RNA molecules that bind specific ligands. Nature 346, 818-822.

Famulok, M (2005). Allosteric aptamers and aptazymes as probes for screening approaches. Curr Opin Mol Ther 7, 137-143.

Fredenburgh, J. C., A. R. Stafford, J. I. Weitz Dagger (2001). Conformational changes in thrombin when complexed by serpins. J Biol Chem 276, 44828-44834.

Geiger A., P. Burgstaller, H. von der Eltz, A Roeder, M. Famulok (1996). RNA aptamers that bind L-arginine with sub-micromolar dissociation constants and high enantios-electivity. Nucleic Acids Res 24, 1029-1036.

German, I. D. D. Buchanan, R. T. Kennedy (1998). Aptamers as ligands in affinity probe capillary electrophoresis. Anal Chem 70, 4540-4545.

Huang, C., Z. Cao, H. Chang, W. Tan (2004). Protein-protein interaction studies based on molecular aptamers by affinity capillary electrophoresis. Anal Chem 76, 6973-6981.

Ito, Y., S. Fujita, N. Kawazoe, Y. Imanishi (1998). Competitive binding assay for thyr-oxine using in vitro selected oligonucleo-tides. Anal Chem 70, 3510-3512.

Jayasena, S. D. (1999). Aptamers: an emerging class of molecules that rival antibodies in diagnostics. Clin Chem 45, 1628.

Klussmann, S., A. Nolte, R. Bald, A. Erdmann, J.P. Furste (1996). Mirror-image RNA that binds D-adenosine. Nat Biotechnol 14, 1112-1115.

Kotia, R. B., L. Li, L. B. McGown (2000). Separation of nontarget compounds by DNA aptamers. Anal Chem 72, 827-831.

Mendonsa, S. D., M. T. Bowser (2005). In vitro selection of aptamers with affinity for neuropeptide Y using capillary electrophoresis. J Am Chem Soc 127, 9382-9383.

Michaud, E., E. Jourdan, A. Villet, A. Ravel, C. Grosset, E. Peyrin (2003). A DNA aptamer as a New Target-Specific Chiral Selector for HPLC. J Am Chem Soc 125, 8672-8679.

Michaud, M., E. Jourdan, C. Ravelet, A. Villet, A. Ravel, C. Grosset, E. Peyrin (2004). Immobilized DNA aptamers as target-specific chiral stationary phases for resolution of nucleoside and amino acid derivative en-antiomers. Anal Chem 76, 1015-1020.

Murphy, M. B., S. T. Fuller, P. M. Richardson, S. A. Doyle (2003). An improved method for in vitro evolution of aptamers and applications in protein detection and purification. Nucleic Acids Res 31, e110.

O'Sullivan, C. K. (2002). Aptasensors - the future of biosensing? Anal Bioanal Chem 372, 44-48.

Pavski, V., X. C. Le (2001). Detection of human immunodeficiency virus type 1 reverse transcriptase using aptamers as probes in affinity capillary electrophoresis. Anal Chem 73, 6070-6076.

Potyrailo, R. A., R. C. Conrad, A. D. Ellington, G. M. Hieftje (1998). Adapting selected nucleic acid ligands (aptamers) to biosensors. Anal Chem 70, 3419-3425.

Ravelet, C., Boulkedid Rym, A. Ravel, C. Grosset, A. Villet, J. Fize, E. Peyrin (2005).

A L-RNA aptamer chiral stationary phase for resolution of target and related compounds. J Chromatogr A 1076, 62-70.

Rehder, M. A., L. B. Mc Gown (2001). Open-tubular capillary electrochromatography of bovine ß-lactoglobulin variants A and B using an aptamer stationary phase. Electrophoresis 22, 3759-3764.

Rehder-Silinski, M. A., L. B. McGown (2003). Capillary electrochromatographic separation of bovine milk proteins using a G-quartet DNA stationary phase. J Chromatogr A 1008, 233-245.

Robertson, D. L., G. F. Joyce (1990). Selection in vitro of an RNA enzyme that specifically cleaves single-stranded DNA. Nature 344, 467-468.

Romig, T. S., C. Bell, D. W. Drolet (1999). Aptamer affinity chromatography: combinatorial chemistry applied to protein purification. J Chromatogr B 731, 275-284.

Tuerk, C., L. Gold (1990). Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249, 505-510.

Vo, T. U., L. B. McGown (2004). Selectivity of quadruplex DNA stationary phases toward amino acids in homodipeptides and alanyl dipeptides. Electrophoresis 25, 1230-1236.

Wang, H., M. Lu, X. C. Le. DNA-driven focusing for protein-DNA binding assays using capillary electrophoresis. Anal Chem (in press).

0 0

Post a comment