Bruce, your follow-up question to #1993 is a good one. A lot of scientists are concerned covid-19
already had, for whatever reason, multiple routes of entry into the human cell (ie beyond ACE2) at the time of release or that it can or will quickly evolve by point mutation the common additional mechanisms that other ssRNA viruses use to get into mammalian cells, either because of replicative advantage or selective pressure from partially effective drugs and immunological approaches.
The viral load in one NEJM German patient was reported as ~100,000,000 mature viroids per single ml of phlegm. At those titers, rare events happen all the time.
It is not plausible that a de novo (innovative) route will evolve by point mutation over a time span relevant to the current pandemic. There is no mechanistic counterpart in RNA viruses to MDR in hospital bacteria which arises from shared plasmids. However single point mutations in, say the spike protein, could provide resistance to multiple unrelated drugs targeting the same site.
Multiple routes of entry will vastly complicate the current Hail Mary approach of tossing already-approved or compassionate-use experimental drugs that target binding sites on either the infecting viroid or receptor proteins external to the cell surface. For example, chase through the thirty million biomedical abstracts at PubMed to see about lisinopril (200 million US Rx per year) which is known to target ACE1 so maybe perturbs the homolog ACE2; it’s been considered before but it doesn’t even bind ACE2 much less block covid-19.
https://www.ncbi.nlm.nih.gov/pubmed/But who knows, maybe something will work out. Pro-drugs like remdesivir don’t act on surface receptors; they have to be taken up and partly metabolized (not CYP450 here); remdesivir is a ribosyl adenylate analog that reversibly inhibits the RdRp replicase while evading proof-reading excision. It is not specific to covid-19 or even coronaviruses; resistance has evolved previously in related RNA viruses.
Talk about rush-rush clinical trials and long-shot vaccines is primarily
panic mitigation. The last thing the hospitals want right now is an ER full of worried well. The last thing FDA wants is another thalidomide. The last thing govt wants is rebuttal of the official narrative.
The current information environment is largely a mix of inapplicable older journal articles on other viruses, rushed preprints, hurry-up peer-review publications, unsupervised manuscript archives including Chinese-language only, and expert opinion from twitter blowhards..
Fortunately Bruce is asking about something we can
easily check for ourselves without outside assumptions other than accuracy of posted genome sequences.
The covid-19 protease story is involved overall but not difficult, lots come into play during the life cycle. Furin is a known quantity, a human serine proprotein convertase encoded on human chr 15 whose catalytic activity may have been
co-opted by covid-19 for its own activation agenda.
It’s mainly about the furin binding site on a target protein, wrongly described at wikipedia as the four residue, basic amino acid motif RXR/KR of arginines and lysine in the primary amino acid sequence. However that motif alone gives rise to numerous false positives and false negatives, so we’ll be needing to feed the covid-19 sequence to a 20-residue motif bioinformatic tool called PiTou that uses a hybrid experimental / hidden Markov chain to find furin sites much more accurately.
It's not possible to get from RXR/KR to a twenty residue motif by incremental point mutations unless the rest was already very close (as 4 to the 16th is too large with intermediate stage utility).
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3281273/Cleavage though is just downstream of the core tetrapeptide which has to be exposed for the furin enzyme to get at a valid substrate. This motif almost always will be on the surface because the 3-4 positive electric charges are about impossible to bury from entropic considerations. The exceptions are internal salt bridges (eg opsins) or an oligomeric binding partner that offsets the positive charges with negatives. However the 20 amino acid requirement really raises the ante.
In the case of the covid-19 spike protein, it is a homomeric trimer with known atomic level structure in both the inactive and activated configurations so we can check whether the RXR/KR etc is exposed. Even so, there could be steric hindrance in the intact virion. This has to be studied with cryoEM as few viruses beyond tobacco mosaic virus are crystallizable for xray structure elucidation. Proteolytic cleavage (unlike drug binding) is always irreversible because of energetics (55M water).
Poll