Applications of Next Generation Sequencing in the Field of Chemical Biology

Chemical biology is the study of biological systems accelerated primarily by chemical methods and the assistance of technologies from biology. Powerful techniques such as Next Generation Sequencing (NGS) made huge impacts on this field and paved the way for new technologies such as DNA encoded libraries that dramatically accelerated the discovery of new protein ligands. We present here two NGS applications in chemical biology that expand and offer a new viewpoint on existing ideas.
DNA alkylating drugs such as temozolomide (TMZ) or streptozotocin (STZ) are known to be electrophiles that modify DNA and cause the cell death of fast replicating cells. Their impact on RNA has been studied less thoroughly, as RNA is typically regarded as being short-lived. We adapted sequencing techniques used to probe secondary and tertiary structures of RNA to measure the damage caused by alkylating drugs. With this approach, we have shown that TMZ and STZ alkylate RNA selectively at G in a non-sequence dependent manner. We have validated our approach by comparing these treatments to the effect of the electrophilic small molecule dimethylsulfate (DMS). Furthermore, we have measured for the first time the alkylation profile of trimethylsilyl diazomethane by RNA sequencing and showed that it reacts similarly to DMS, both in vitro and in vivo. We believe that this method is widely applicable and can be used to study the alkylation patterns of more complex natural product.
We applied the knowledge of NGS and bioinformatics and transferred it to DNA encoded libraries (DELs), culminating in a new open-source toolkit written in python. This toolkit enabled and accelerated our studies of DELs by offering a convenient command-line interface to create these combinatorial libraries in situ. Furthermore, the toolkit assists in evaluating the NGS results by offering easy-to-use tools and helps to understand the results by generating quick standard reports.
Finally, we created a DNA encoded library where we diversified a DNA-bound azide with a library of alkynes by a copper(I)-catalysed cycloaddition, which we subsequently encoded by splint-ligation. We used the same azide to generate a second sublibrary by reducing it to an amine and diversifying it with carboxylic acids. We encoded for these acids using the same codon space. Then, we introduced a unique identifier for each sublibrary encoding for the reaction type used, a concept we termed reaction encoding. This approach increases the diversity of one point by utilising several small sets of differently functionalised building blocks, making the purchase of large building block libraries unnecessary. Furthermore, it allows analysing structure-activity relationships not only across building blocks but also across connectivity types.