A simple method for eliminating fixed‐region interference of aptamer binding during SELEX

Abstract

Standard libraries for systematic evolution of ligands by exponential enrichment (SELEX) typically utilize flanking regions that facilitate amplification of aptamers recovered from each selection round. Here, we show that these flanking sequences can bias the selection process, due in part to their ability to interfere with the fold or function of aptamers localized within the random region of the library sequence. We then address this problem by investigating the use of complementary oligonucleotides as a means to block aptamer interference by each flanking region. Isothermal titration calorimetry (ITC) studies are combined with fold predictions to both define the various interference mechanisms and assess the ability of added complementary oligonucleotides to ameliorate them. The proposed blocking strategy is thereby refined and then applied to standard library forms of benchmark aptamers against human α-thrombin, streptavidin, and vascular endothelial growth factor (VEGF). In each case, ITC data show that the new method effectively removes fixed-region mediated interference effects so that the natural binding affinity of the benchmark aptamer is completely restored. We further show that the binding affinities of properly functioning aptamers within a selection library are not affected by the blocking protocol, and that the method can be applied to various common library formats comprised of different flanking region sequences. Finally, we present a rapid and inexpensive qPCR-based method for determining the mean binding affinity of retained aptamer pools and use it to show that introduction of the pre-blocking method into the standard SELEX protocol results in retention of high-affinity aptamers that would otherwise be lost during the first round of selection. Significant enrichment of the available pool of high-affinity aptamers is thereby achieved in the first few rounds of selection. By eliminating single-strand (aptamer-like) structures within or involving the fixed regions, the technique is therefore shown to isolate aptamer sequence and function within the desired random region of the library members, and thereby provide a new selection method that is complementary to other available SELEX protocols.

Hi‐Fi SELEX: A high‐fidelity digital‐PCR based therapeutic aptamer discovery platform

“…Hi‐Fi SELEX works to improve round‐by‐round PE values (where PE equals the ratio of aptamers removed in the wash fractions to those collected in the elution fractions) and overall selection efficiencies, in part through careful control of the manner by which the library and the target are presented, equilibrated and washed. To begin, a “competent” SELEX library is prepared using a novel method we recently reported that anneals a complementary oligonucleotide to each fixed region to eliminate the various ways in which those common sequences can interfere with the fold or function of aptamers localized within the random region of the library (Ouellet et al, ). In that work, we show the method is applicable to common SELEX library formats comprised of different flanking region sequences, and results in recovery of function and enrichment of tight‐binding library‐formatted aptamers otherwise lost due to fold disruption, steric hindrance, or other fixed‐region mediated interference mechanisms.…”

“…Finally, library incubation conditions and times were screened and optimal conditions thereby identified (1 nmole resuspended in 1× AF buffer pH 7.5 with 0.005% Tween‐20 for 60 min at 25°C with gentle (300 rpm) agitation) so as to permit members to approach equilibrium partitioning (Figure S4A; Supporting Information) in each selection round while minimizing non‐specific adsorption (Figure S4B; Supporting Information). The combined impact of our novel target and library presentation method on aptamer selection was then assessed (Table ) by measuring round‐by‐round bulk affinities of retained pools of aptamers selected against human α‐thrombin using either a conventional SELEX protocol employing magnetic MyOne Carboxylic Dynabeads (Ouellet et al, ) or the platform described above coupled with the bulk PCR and λ‐exonuclease treatment typically used in SELEX to amplify and reconstitute each retained pool of ssDNA library members. For each system, results are shown for screening of both competent (supplemented with both 5′ and 3′ blocking agents) and standard 80‐mer ssDNA libraries to parse out the performance contribution of the novel library‐presentation elements of our approach.…”

“…The 80‐mer library was combinatorially synthesized by mixing the DNA nucleotides A:C:G:T at a molar ratio of 3:3:2:2.4 to achieve equal probability incorporation of each nucleotide in the core region (Pollard et al, ). The 20‐mer flanking sequences and associated primers (Table S‐I; Supporting Information) are ones commonly employed in conventional SELEX (Janssen et al, ; Ouellet et al, ; Vanbrabant et al, ), and were utilized here due to their suitability for SYBR green qPCR amplification (Janssen et al, ). For this work, both flanking sequences had their reverse complement synthesized, with the complement to the 5′ and 3′ flanking sequences denoted in Table S‐I (Supporting Information) as 5′‐Comp and 3′‐Comp, respectively.…”

“…Mean equilibrium dissociation binding constants ( K d ) of retained aptamer pools were determined by a qPCR‐based method previously shown to yield identical results to Isothermal Titration Calorimetry (ITC)‐based measurements (Ouellet et al, ). As described above, the protein target was immobilized in 16 wells of a NUNC Amino Immobilizer plate.…”

“…The expectation is that the multi‐round selection process will be sufficiently stringent to identify a relatively small pool of specific high‐affinity sequences (Stoltenburg et al, ). However, aptamer discovery by SELEX, including the various modified forms that have been described (Kong and Byun, ; Kulbachinskiy, ), can fail due to a number of mechanisms, including (i) interferences from the common regions flanking each random sequence within the library (Ouellet et al, ), (ii) poor partition efficiencies resulting from non‐specific retention of sequences (Berezovski et al, ; Weng et al, ), (iii) excessive accumulation of amplification artifacts and library‐regeneration biases that reduce yields and hamper enrichment of a high‐affinity aptamer pool (Musheev and Krylov, ), and (iv) inadvertent isolation of candidate aptamers that lack sufficient specificity for the target due, in part, to a binding mechanism dominated by ion‐exchange. Together, these factors are known to increase the number of rounds required to identify high‐affinity members (Levine and Nilsen‐Hamilton, ; Wang et al, ) or, in more extreme cases, completely thwart the enrichment and discovery of aptamers against therapeutically relevant targets (Jolma et al, ).…”