Working out the step by step molecular mechanism to explain why some PS ASOs of all chemical classes are cytotoxic led directly to medicinal chemical solutions that are enhancing the therapeutic index of many PS ASOs significantly by dramatically reducing their potential cytotoxic effects without materially altering their potency (Shen, W., et.al., NAR, 2018; Migawa, M., et.al., NAR, 2019; Vasquez, G., et.al., NAR, 2021; Anderson, B., et. al., NAR, 2021) and representatives of this new generation of PS ASO designs are nearing clinical testing. Having addressed the potential cytotoxicity of some PS ASOs , we turned our attention toward establishing a deeper understanding of why some PS ASOs can activate the innate immune system.
In the first in a series of manuscripts (Pollack, A., et.al., NAR, 2022), we showed that though all PS ASOs bind to TLR9, only a fraction are full TLR9 agonists ( induce maximal activation of the innate immune system) while others are partial agonists or can even be competitive inhibitors. Further, we showed that there were extra-and intracellular proteins that can bind to PS ASOs, some of which exacerbated the innate immune activation others that inhibit innate immune activation. Finally, we showed that the kinetics of activation could be either rapid or slow.
In the next of the series of papers to publish (Pollack, et.al. 2023 , NAT in press) , we explored the structure activity relationships (SAR ) of PS ASOs that activate the innate immune system and found that the SAR is very complex and that the number and placement of PS moieties, the types, numbers and placement of 2’ modifications and phosphate substitutes all can affect the innate immune activation potential of PS ASOs. In contrast to cytotoxicity, no clear direction as to what position in a PS ASO can be modified to ablate or reduce innate immune activation. As a general rule, again differing from cytotoxicity, placement of 2’modifications in the 3’ side of the gap reduced activation while putting 2’ modifications in the 5’ portion of the gap worsened innate activation, but there were many exceptions to even this basic rule. The SAR of PS replacement with other phosphate analogs was even more diffuse as was the effects of sequence. These results demonstrate that it is possible to understand the molecular mechanisms of TLR9-induced innate immune signaling, but we still have a lot to learn.
Stay tuned for our next paper in this series in which we evaluate models that might explain some of our observations and begin to explore the means by which innate signaling is actively terminated.