Search for: All records

Abstract Lipid nanoparticles (LNPs) are the most clinically advanced nonviral RNA-delivery vehicles, though challenges remain in fully understanding how LNPs interact with biological systems.In vivo, proteins form an associated corona on LNPs that redefines their physicochemical properties and influences delivery outcomes. Despite its importance, the LNP protein corona is challenging to study owing to the technical difficulty of selectively recovering soft nanoparticles from biological samples. Herein, we developed a quantitative, label-free mass spectrometry-based proteomics approach to characterize the protein corona on LNPs. Critically, this protein corona isolation workflow avoids artifacts introduced by the presence of endogenous nanoparticles in human biofluids. We applied continuous density gradient ultracentrifugation for protein-LNP complex isolation, with mass spectrometry for protein identification normalized to protein composition in the biofluid alone. With this approach, we quantify proteins consistently enriched in the LNP corona including vitronectin, C-reactive protein, and alpha-2-macroglobulin. We explore the impact of these corona proteins on cell uptake and mRNA expression in HepG2 human liver cells, and find that, surprisingly, increased levels of cell uptake do not correlate with increased mRNA expression in part likely due to protein corona-induced lysosomal trafficking of LNPs. Our results underscore the need to consider the protein corona in the design of LNP-based therapeutics. Abstract Figure 

Neural Machine Translation (NMT) systems are known to degrade when confronted with noisy data, especially when the system is trained only on clean data. In this paper, we show that augmenting training data with sentences containing artificially-introduced grammatical errors can make the system more robust to such errors. In combination with an automatic grammar error correction system, we can recover 1.0 BLEU out of 2.4 BLEU lost due to grammatical errors. We also present a set of Spanish translations of the JFLEG grammar error correction corpus, which allows for testing NMT robustness to real grammatical errors.