A gene the virus no longer uses sticks around anyway, and NIH called that work dangerous.
A jarring letter flagged some of his lab's virus evolution work, quoting it, as dangerous gain-of-function research. After some back and forth the NIH determined that NIH funds cannot be used for one set of experiments, and Elde confirms no NIH money will support that work going forward.
Answered around 6:02Through the back and forth it became clear the concern was passaging MHV in a macrophage cell line. Elde calls the stipulation on the renewal wild governmental overreach: it sits totally outside peer review, the expert reviewers never flagged it as a problem, and there is no appeal process.
Answered around 1:04:24A Science news piece, NIH Suspends Dozens of Pathogen Studies Over Gain-of-Function Concerns, describes 40 projects meeting the gain-of-function definition, suspended while erring on the side of caution, plus another 170 flagged for potential suspension or termination. Elde thinks his was in that pile of 170.
Answered around 8:21Hidden Evolutionary Constraints Dictate the Retention of Coronavirus Accessory Genes, published at the end of last year in Current Biology. Stephen Goldstein, a postdoc in Elde's lab, drove it; contributors include Zoe Hilbert, Diane Downauer, and Nima Moshiri, with viruses and cells from Susan Weiss and Sunny Shin at Penn.
Answered around 14:52MHV, a mouse virus known for 75 years that has never infected a human and has infected a lot of mice. The lab works with it under BSL-2, biosafety level two, in cell lines rather than whole animals, with protocols, training and protective equipment agreed with environmental health and safety.
Answered around 42:42Doing only the control, passaging in macrophages, the mutated NS2 gene in the ORF2 region did not go away. Less is more evolution predicted natural selection would quickly delete a sequence no longer encoding a working protein. Instead it sat there like a rock, in both cell backgrounds.
Answered around 44:47ACAP7, the host phosphodiesterase put into the ORF4 location, persisted in macrophages, where the OAS and RNase L pathway runs strong and that enzymatic activity is needed. In L2 fibroblasts, which do not express much OAS or RNase L, selection is relaxed and the gene was lost.
Answered around 46:22Elde thinks the genome folds up, forming RNA secondary structure that is read out in translation or in replication and transcription. Unpublished assays show a ton of secondary structure not only at the five prime and three prime UTRs but throughout the body of that open reading frame.
Answered around 32:25Yes. Searching the millions of sequences in GISAID, ORF8 by and large sticks around despite debilitating mutations: the B.1.17 variant had a stop codon at position 27, XBB.1.X at position 8. Deletions eat at the sequence in populations, then weeks later most of them have it back.
Answered around 58:07Long read sequencing of the populations showed shifts in subgenomic RNA abundance. In four of five clones, by passage 10, the subgenomic RNA encoding spike was up around five-fold in the fibroblast passaged viruses. More spike may help infect the next cell, but would attract antibodies in an animal.
Answered around 51:14“So it violated this notion of less is more evolution.”
“So to think that we could engineer a virus is ridiculous, and it shows a lack of understanding of the field.”
“They've shifted that onto the researchers.”
“That's why the NIH model is so beautiful.”
“Hidden Constraints in Coronavirus Genes.” https://traffic.libsyn.com/secure/twievo/TWiEVO117.mp3. Transcript summary by WhipScribe, https://whipscribe.com/transcript-pages/hidden-constraints-in-coronavirus-genes.html.