Engineered VPg saRNA achieves cap-independent, low-immunogenic and precise encoding of therapeutic proteins in vivo.

HPLC-MS analysis of nucleosidylated VPg was performed on a total of 2 samples, including unmodified VPg (n=1) and nucleosidylated VPg (n=1) species with 3 technical replicates. Samples were prepared by diluting purified VPg or in vitro nucleotidylation reaction mixtures in LC-MS-grade water, centrifuging at 12,000 ×g to remove particulates, and injecting 1 μl of each preparation onto a CAPCELL CORE AQ S2.7 column (2.1 mm ×150 mm). HPLC separation was carried out on a Waters Alliance 2795 system equipped with a 2996 PDA detector using 10 mmol/L ammonium phosphate as the mobile phase at a flow rate of 400 μl/min, allowing up to 200 μg of VPg per injection without loss of resolution. Fractions corresponding to individual nucleosidylated VPg species were manually collected and stored at -80 °C for subsequent LC-MS analysis. Mass spectrometry was performed on a Waters Q-TOF Micro instrument equipped with a dual electrospray ionization (ESI) lockspray source, acquiring data in positive-ion mode under identical conditions for sample eluates and a VPg reference infused at 2 μl/min. Source parameters included a source temperature of 120 °C, desolvation temperature of 300 °C, and cone voltages of 25 V (reference) and 20 V (sample). Raw spectra were processed and deconvoluted using MassLynx 4.1 with the MaxEnt1 Transform algorithm to obtain zero-charge intact molecular masses. Because nucleotidylated VPg yields intact mass shifts rather than proteolytic peptides, peptide-database searches, missed-cleavage criteria, FDR filtering and peptide-scoring parameters were not applicable. Instead, VPg nucleosidylation states were identified by theoretical-to-observed mass matching, with VPg-pU confirmed by an intact mass increase of 303.17 Da corresponding to uridylylation at Tyr30(Y30). All LC-MS analyses were repeated independently at least 3 times to ensure reproducibility.