Gaining depth on gain of function

Yesterday I posted “The fallacious Fauci strikes again.” My comments elicited a message from Michael S. Rogers, Assistant Professor of Surgery at Harvard Medical School and Research Associate in the Vascular Biology Program at Boston Children’s Hospital. I thought some readers might find it of interest. With his permission I am posting his message below the break.

* * * * *

The question of whether the 2017 PLoS Pathogens paper contains gain of function experiments has recently been in the news. Since I have significant experience (30 years) in the recombinant DNA techniques in question, I thought I might lay out what was done in the experiments and let you decide.

Using a firearms analogy, here’s what they did. They had a bunch of new lower receivers (Spike genes) for AR-15s (SARS-like coronaviruses) and wanted to know whether they could fire full-auto (infect human cells). They knew that if they tried each of them on a semi-automatic AR-15, they might wind up with an automatic weapon and would thereby violate the National Firearms Act (perform gain of function research). However, they already had a full-auto M16 (a virus called WIV1). So instead, they took each of the lower receivers and put it into their M16 to see if it could still shoot full auto. In two cases it could and in 6 cases it couldn’t.

What is more interesting about this paper is that it debunks most of the “the virus couldn’t have been made in a lab” papers. To use the gun analogy, it’s as if outside “experts” are saying “they couldn’t have made a new gun, there are no tool marks” while in this paper the authors show that they know how to make a gun without leaving tool marks. Here’s how. The papers from outside experts assume that a new virus would be made by restriction enzyme technology that leaves “scars” (4-8 base pair palindromes) where two pieces were joined.

Those papers look for the scars and, finding none, declare that the virus couldn’t be man-made. However, there are newer “scarless” cloning techniques (Gateway cloning, Gibson assembly, etc.) that don’t leave such marks. In this paper, the authors use one of these techniques (Gateway cloning) to put the new Spike genes into WIV1. Of course, the fact that it could have been done doesn’t mean that it was done.

Now, for more details on what was done with the potential gain of function experiments. Here’s the paper. In figure S7, the authors take the Spike gene (the part of the virus that attaches to cells and injects the viral RNA) from 8 newly-discovered bat SARS-like coronaviruses and puts them into a BAC (bacterial artificial chromosome) that codes for WIV1.

These new BACs are then introduced into cells where they generate “live” virus. This virus is then tested on human (HeLa) and monkey (Vero) cells to see if they can be infected. Two of the 8 new viruses were able to infect human cells as evidence by the red color (indicating the presence of virus protein) in the relevant panels in figure S7B. Whether this is gain-of-function research depends critically on the characteristics of WIV1. As is evident from both the control experiment in the paper (figure 7A, top row, note the red in the cells) and the authors’ 2013 Nature paper (Figure 4, Nature 503:535–538 ) WIV1 can infect human, bat, and civet cells.

Hope this helps.

Notice: All comments are subject to moderation. Our comments are intended to be a forum for civil discourse bearing on the subject under discussion. Commenters who stray beyond the bounds of civility or employ what we deem gratuitous vulgarity in a comment — including, but not limited to, “s***,” “f***,” “a*******,” or one of their many variants — will be banned without further notice in the sole discretion of the site moderator.