|Title||Selection and characterization of DNA aptamers|
|Source||University. Promotor(en): John van der Oost; Hauke Smidt. - S.l. : s.n. - ISBN 9789461735645 - 152|
|Publication type||Dissertation, internally prepared|
|Keyword(s)||aptameren - dna - selectie - nucleotiden - technieken - eiwit - rna - aptamers - selection - nucleotides - techniques - protein|
This thesis focusses on the selection and characterisation of DNA aptamers and the various aspects related to their selection from large pools of randomized oligonucleotides. Aptamers are affinity tools that can specifically recognize and bind predefined target molecules; this ability, however, is not exclusively associated with aptamers. Antibodies are the most successful affinity tools used today, but alternative affinity tools such as aptamers, engineered binding proteins and molecular imprinted polymers are emerging as sound alternatives. A comparison of their properties is described in Chapter 1. The strength and specificity of the interaction between an affinity tool and its target molecule is an important feature. Generally, an affinity tool should have a high affinity for its target and should be highly specific in order to be useful for research or commercial purposes. One highly advanced method to characterise the interaction between an affinity tool and its target molecule makes use of a Surface Plasmon Resonance (SPR)-based biosensor. Although SPR is an optical phenomenon, in depth knowledge of the physics behind this phenomenon is not required to operate an SPR-based biosensor. Experiments should be performed in a correct way, and therefore it is important to understand how experimental parameters, such as flow rate, ligand density, surface preparation, and reagent quality either improve or adversely affect data quality. Experimental considerations, as well as methods for proper data analysis are discussed in Chapter 2. Data generated within the framework of the 2011 Global Label-free Interaction Benchmark study serves as a typical example.
The ability of aptamers to bind a specific target originates from an intricate interplay between the oligonucleotide sequence and the three dimensional structure that this sequence allows to form. In Chapter 3 this is illustrated by the selection and characterisation of streptavidin-binding aptamers. Five aptamer families were identified, sharing a similar secondary structure. Although slight variations at the actual sequence level are present, two guanines are completely conserved. Using site-specific mutagenesis it was demonstrated that these guanines are essential for streptavidin binding. Binding kinetics and the dissociation constant of each aptamer was determined by SPR and were all within the range of 35-375 nM. Two aptamers can bind one streptavidin tetramer at the same time, as was shown by native mass spectrometry analysis. In addition, the three dimensional structure of the most abundant aptamer was modelled and manually docked to the streptavidin structure, in order to gain more insight in the molecular basis of the interaction. To extend this knowledge even further, crystallisation trails, aiming to obtain a co-crystal structure for the streptavidin-aptamer complex, were performed, and are described in Chapter 4. Unfortunately, these trials did only yield protein crystals, instead of the desired streptavidin-aptamer complex. Therefore, alternative experimental and computational approaches were investigated that could be used to study aptamer-protein interactions. Combining techniques as SPR, small-angle X-ray scattering (SAXS), isothermal titration calorimetry (ITC), and Dynamic light scattering (DLS) could be considered as an alternative to X-ray crystallography. In addition, some of these techniques may provide information on the dynamics of complex formation, whereas crystallography gives a time- and position-averaged image.
Besides streptavidin, another protein, SpaC, was subjected to aptamer selection in this thesis. SpaC is a subunit of pili present on the probiotic Gram-positive bacterium Lactobacillus rhamnosus GG and contains a binding domain for human-mucus. Presence of this binding domain is considered an advantage, because it is already designed to interact with other molecules. Successful production and purification of recombinant SpaC protein is described in Chapter 5, as well as the characterisation of DNA oligonucleotides enriched during subsequent selection rounds. Sequence analysis revealed that specific oligonucleotides are indeed enriched. Furthermore, results of pilot SPR experiments indicated that they bind specifically to SpaC, but more detailed experiments are required to unambiguously demonstrate this.
The dynamics of aptamer enrichment are poorly understood. To address this issue and to gain a more fundamental insight in the aptamer selection process, a multiplexed high throughput sequencing effort was started, which is described in Chapter 6. In this approach samples of 70 selection rounds, derived from 8 distinct aptamer selection experiments, were barcoded, pooled together and sequenced; over 84 million paired-end reads were obtained and analysed. Samples enriched to bind streptavidin show a decrease in α-diversity across subsequent selection rounds. Interestingly, large differences were found between the composition of fractions enriched by affinity elution and thermal elution. Moreover, a small scale comparison of two clone libraries showed that affinity elution, which is expected to enrich more specific binders, also specifically enriches rapid binders.
Supportive SPR experiments have made an important contribution throughout this thesis. The main focus in Chapter 7, however, is on a new application of SPR. The development of a capture approach for supercoiled plasmid DNA, using a triple helix forming oligonucleotide, is described. It could be demonstrated that plasmid DNA can indeed be captured and that SPR can subsequently be used to derive kinetic parameters of a specific interaction with a plasmid. In this particular case the interaction between Lac repressor and its plasmid-based operator was characterised, showing that the association and dissociation rates are ~18 times lower, but that the affinity is the same, when compared to binding to linear operator DNA. This difference underscores the importance of using a DNA substrate with a physiologically relevant topology for studying DNA-protein interactions.