Optimization of SELEX: Comparison of different methods for monitoring the progress of in vitro selection of aptamers

Highlights

• Aptamer selection is often hindered due to accompanying adverse effects.

• Assays are described for measuring aptamer pool complexity and binding affinity.

• Methods are compared using anti-streptavidin aptamer selection as model.

Abstract

Oligonucleotide aptamers are selected from libraries typically comprising up to 10¹⁵ different sequences by an iterative process of binding, separation, amplification and purification, called SELEX. During this process, the diversity of the oligonucleotide pool decreases until, presumably, only sequences with highest binding affinities towards chosen targets remain. This selection technique is time-consuming, labor-intensive and expensive. Though well posed in principles, the SELEX procedure is noise sensitive, due to amplification of unspecific-binding sequences, and it is not surprising that aptamer selection is often not successful in practice. In view of that, a follow-up of the progress of selection during its course with simple yet reliable methods is necessary. In this paper, we describe five independent assays to estimate the sequence complexity of SELEX pools including qualitative restriction fragment length polymorphism analysis, melting curve analysis, quantitative fluorescence intensity measurements of bound ssDNA, real time PCR quantification and pool dissociation constant analysis during the progress of aptamer selection against streptavidin. Properties and features of each method are discussed and compared. Pool dissociation constant analysis and sequencing serve as reference methods.

Introduction

Aptamers are short single stranded nucleic acids, selected in vitro to bind to specific targets. Since their discovery in 1990 [1], [2], aptamers have been engineered for numerous considerably different molecular targets ranging from simple organic [2], [3] and inorganic [4] molecules to large and complex structures such as carbohydrates [5], proteins [1], [6], [7] and even whole organisms [8]. Studies on aptamers emerge rapidly since its discovery as they show comparable binding characteristics to monoclonal antibodies and offer simultaneously many advantages that surpass limitations in the use of antibodies. Some of the features of aptamers making them ideal tools for research, diagnosis and possibly therapy include the accurate, large scale, automated, low cost and reproducible synthesis of aptamer sequences and uniform labeling for instance with biotin or fluorescence reporters, their functional immobilization on various substrates, long-term chemical stability achieved by chemical modifications of bases or backbone of aptamer oligonucleotides, low immunogenicity, and ability to recover native active conformation after denaturation [9].

Systematic evolution of ligands by exponential enrichment (SELEX) is a procedure used to isolate aptamers for particular molecular target from a synthetically derived oligonucleotide library with a typical selection screen of 10¹⁴–10¹⁵ unique/different sequences [1], [2]. This high-flux screening technique is comprised of several rounds of target binding, separation of target bound from unbound oligonucleotides and elution, also called partitioning, and further PCR amplification and purification of selected oligonucleotides. Several rounds of selection are performed to obtain high affinity, target-specific aptamers before cloning, sequencing and, finally, characterization. Multiple parameters affect the required number of iterations, including target structure and charge, library design and stochastic events during selection, partitioning and amplification [10], [11]. For these reasons, it is not unlikely that repeating SELEX procedures for same targets with only slightly changing one or even none of the parameters may lead to different result. An extensive list of streptavidin specific aptamers which have been selected in different SELEX runs, groups and with more or less different methods supports this conjecture (see [12], [13], [14] and references therein as well as the results of our SELEX in Section 3.6 and Appendix A of Supplementary material 2). In the effective selection process, the sequence diversity of oligonucleotide library is drastically reduced and it is expected, that the overall affinity of selected oligonucleotides increases in each successive SELEX round, until an upper limit, determined by the affinity of the strongest binders in the initial random library, is reached. By the term “effective” we here refer to the case where the target affinity binding dominates the selection over the accompanying adverse effects. Given this fact, the pool stops converging when aptamers, specific for distinct binding sites, are not competing each other anymore. At this point it is reasonable to progressively increase the selection and partitioning stringency by introducing the counter (negative) selection, increasing the number of washing steps, adding polyanionic competitors and decreasing the target concentration in binding reaction or decreasing the target-pool incubation time [15]. It is important to note that PCR amplification of large oligonucleotide libraries with random region should be carefully optimized since over-amplification can lead to by-product formation and subsequently to a complete loss of the product [16].

Efficient following of the progression of SELEX is thus essential for successful selection of aptamers with good binding affinities. In order to address this issue we investigated different methods, in parallel with determination of pool's K_D and sequencing, to monitor the convergence and composition of random library during selection of streptavidin-binding sequences. In this paper, we highlight the most efficient, simple and fast approach to follow the evolution of the oligonucleotide pool during SELEX, regardless of the characteristics of the chosen SELEX protocol or target. However, measuring binding activities of aptamer pools is still the most reliable and unambiguous technique to follow the course of enrichment through successive rounds of selection, if background effects are considered correctly, and binding activity is unambiguously defined and understood.