Related to someone's elses disbeliefs in how proteins produced from a mRNA vaccine end up in B cells; in theory the process could be more complicated than "naked" egress (which arguably does have simplicity in its favor). There's at least one paper that discussed spike proteins observed egressing in extracellular vesicles (following mRNA vaccination) but it doesn't really say whether this is the common route. And from the general gist of that paper, EVs seem to interfere with detection, so I'm guessing that's not the usual route. For yet another alternative, I could imagine that transfected cells first get killed by CD8 T cells and it's mostly this debris that gets picked up by B cells.
Interestingly, on the last angle, one review mentioned that
Comparative studies performed by our research group and others demonstrated that nucleoside-modified mRNA LNPs outperformed unmodified mRNA LNPs in translation capacity, but failed to elicit successful CTL [cytotoxic T cell] responses due to the drastically reduced production of type I IFNs [119, 157].
So it seems that the mRNA vaccines currently in clinical use for infectious diseases like Covid-19 (which do use nucleoside-modified mRNA) are pretty unsuccessful at killing cells, which makes the 3rd pathway I was considering above somewhat unlikely as a major contributor. (The research group which wrote that review seems focused on cancer immunotherapeutics, and the two cited studies definitely are about that. The lack of CTL activation is a big deal for cancer therapy.)
So, what direct evidence is there for how proteins most often come out cells following mRNA vaccination, particularly in nucleoside-modified vaccines?