In vitro transcription using bacteriophage T7 RNA polymerase (T7 RNAP) is the gold-standard platform for RNA production in both research and therapeutic applications. Despite its high processivity and promoter specificity, T7 RNAP generates multiple RNA by-products, including double-stranded RNA, 3'-extended transcripts, abortive RNAs, and prematurely terminated products. These impurities reduce RNA yield, complicate downstream purification, and raise safety concerns for RNA-based therapeutics by activating adverse innate immune pathways. Although reaction optimization and downstream purification strategies can mitigate these issues, they typically involve trade-offs between RNA purity and yield. Enzyme engineering has therefore emerged as a powerful upstream strategy to suppress by-product formation at its molecular origin. Here, we synthesize current knowledge on the structural and mechanistic basis of T7 RNAP by-product formation and systematically review engineering strategies to improve RNA purity. T7 RNAP variants are classified according to their underlying mechanisms of action, including enhanced thermostability, reduced non-specific template binding, smoother initiation-to-elongation transition, reduced premature termination, and template-biased polymerase designs. This analysis identifies general principles governing the trade-off between specificity and processivity and highlights synergistic combinations of mutations that improve RNA purity without compromising transcriptional efficiency. We conclude by discussing the remaining challenges for engineering T7 RNAP to meet the stringent purity requirements of next-generation RNA therapeutics.
Pauline Hermans, Y.I. Serikova, José Castillo et al.· 0 citations
Accurately predicting the effects of pharmacogenomic variants is essential for the development of personalized therapeutic strategies, as genetic variability can influence drug response differently across patients. Here, we assessed several computational approaches using a dataset of pharmacogenomic variants with either clinical annotations or functional characterization by deep mutational scanning, compiled from the literature, with an additional focus on CYP2C9, a clinically relevant drug-metabolizing enzyme. Our results show that, despite recent methodological advances, substantial room for improvement remains. In particular, current methods struggle to distinguish gain-of-function variants associated with increased drug clearance and fast-metabolizer phenotypes from neutral variants, whereas loss-of-function variants that reduce drug clearance are predicted more accurately. The integration of structural and evolutionary information appears to be a key strategy for improving performance, with the coevolution-based StructureDCA method achieving the highest accuracy compared with classical genetic variant-effect predictors and recent deep learning approaches, including the pathogenic-variant predictor AlphaMissense and general protein language model–based methods. Finally, our results indicate that computational models can complement in vitro experiments in clinical variant interpretation, as StructureDCA predictions showed better agreement with clinically annotated phenotypes than large-scale deep mutational scanning data in several cases.
F. Pucci, Pauline Hermans, Matsvei Tsishyn et al.· bioRxiv· 0 citations
