Proteins determine the fate of PS ASOs in biological systems

Though we have known that PS ASOs bind to proteins since 1989 and, in fact, one of the reasons we chose phosphorothioate to replace the phosphodiester inter-nucleotide linkage was the belief that they would bind to plasma proteins thereby preventing rapid clearance in urine via glomerular filtration, the focus of ASO medicinal chemistry was on the language of ASO-nucleic acid interaction. That is, we focused on learning how to enhance the affinity and specificity of PS ASO interactions with RNA and were successful. Thousands of PS ASO chemical modifications were made and tested, creating a strong working knowledge of how our drugs interact with their cognate sequences, “receptors” in highly structured ribonucleoprotein complexes. We then used the insights gained to dramatically improve the performance of PS ASOs. We increased potency many orders of magnitude, increased stability, thus greatly prolonging the duration of action, reduced the incidence and severity of a number of adverse events, learned to design PS ASOs to take advantage of a wide range of post-RNA binding mechanisms and enabled essentially all routes of administration. Then, several years ago the observation that GalNAc conjugation to RNA targeted drugs mediated targeted delivery to hepatocytes stimulated a broad interest in identifying ligand and receptor systems that might facilitate targeted delivery to other cells and tissues. Progress in this area of research continues.

My interest in better understanding PS-ASO-protein interactions and the potential consequences of those interactions arose quite independently of the interest in targeted delivery. For many years I focused my group’s efforts on developing a detailed understanding of the molecular mechanisms of action of PS ASO and using that knowledge to design better performing antisense drugs. We continue that effort and continue to report exciting results.  However, I came to the conclusion that in order to thoroughly understand the molecular mechanism by which PS ASOs produce their observed, we needed to develop a much deeper understanding of PS ASO-protein interactions for several reasons. First, though there was little direct evidence, I believed that cellular uptake and subcellular distribution had to be mediated by PS ASO-proteins interactions and that indirect approaches such as siRNA screens would never yield a meaningful answer. Second, we already knew that proteins such as RNase H1 are essential for the pharmacology of PS ASOs, that sub-cellular localization affected PS ASO activity and that there were proteins that could alter the activity of RNase H1. I suspected that many unidentified cellular proteins might alter the molecular pharmacology of PS ASOs.  Finally, at Ionis, we had invested many years of intense efforts to understand why some ASO sequences were toxic while others were not. The debate centered on the question of whether most toxic PS ASO sequences were toxic because of binding to non-cognate sequences and RNase H1 mediated cleavage of non-target RNAs or some other unidentified mechanism was responsible. Certainly, the “off-target” hypothesis made a great deal of sense, but basic research my group had performed on RNase H1 and other lines of evidence suggested to me that we needed to consider other hypotheses.

As is often the case, new assays greatly facilitated our research. The first breakthrough derived from a biotin pull-down assay was developed by Liang Liang in my group. With that assay we were able to identify and characterize a number of intracellular proteins that bind PS ASOs and begin to understand the structure activity relationships (SAR) of these interactions. Next, Tim Vickers in my group developed modified NanoBRET and NanoBiT assays that greatly accelerated the development of in-depth understanding of the chemical characteristics of PS ASO protein interactions and for the first time to visualize those interactions in live cells.

We have now identified a number of proteins at the cell surface that mediate cellular uptake of PS ASOs (we suspect that there are more to be found). We know the major proteins and endosomal pathways responsible for productive (pharmacologically active) uptake and cellular distribution. We have identified quite a number of proteins that alter PS ASO pharmacological activities and are making progress in understanding the molecular mechanisms involved. We have defined a step-by-step molecular mechanism that explains the toxicities of 95% of the toxic PS ASOs in all chemical classes, cell types and organs studied to date. Remarkably, we then showed that straightforward chemical modifications could reduce or ablate these toxicities with little effect on potency. Of course, work continues to explore the SAR. We selected a number of important model proteins on which we sought to understand the basic chemistry of these interactions. Finally, we showed PS ASOs can alter the fates of many of the proteins with which they interact.

In the review just published online in Nucleic Acid Research (Crooke, Vickers and Liang, NAR, 2020 doi: 10.1093), we summarize and codify what we have learned. If you are interested in understanding the molecular mechanisms by which PS ASOs produce the effects we observe, you may find this review of interest.