DIY Virus Editing - The Easy Way
A new reverse-genetics platform for SARS-CoV-2. Or, "I survived Clonathon 2022, and all I got was this lousy T-shirt"
New: Easy virus replication, recombination, and mutation!
The new preprint, out of the University of Basel:1
Overall, this strategy will facilitate recombinant SARS-CoV-2 rescue and accelerate its manipulation. Using our protocol, newly emerging variants can quickly be engineered to further elucidate its biology.
Make new variants faster!
H—hooray…!
As the title indicates, Kipfer and co. have made a gene-to-virus (“reverse genetics”) platform for SARS-CoV-2 that is “cloning-free.” So, no clones here. Put down the clones; walk away from the clones. All clones are strictly verboten; no exceptions even for certified Emotional Support Clones.
The Old Way: An obscure point of incredible confusion
Viruses are creatures of cycles. Infection, replication, repeat. Some, including myriad likely ancient bacterial viruses (phages), have evolved sophisticated means of pausing their cycles and going dormant within the DNA of their host cell. But the others survive only in a perpetual leap from cell to cell to cell, like tiny rain-turtles throwing their eggs upward out of falling drops — they must constantly keep renewing, or perish.
Studying objects that can only exist in motion is difficult. One cannot simply wave an Elmo doll at a virus, to get it to pause for the camera.
For RNA viruses, what helps is to convert the “turtle” — the genes inside the viral “egg” — into complementary DNA, and nature already provides molecules that will do this (via RNA viruses that already copy their genes back to DNA, such as HIV). Then, the “only question” is how to convert the cDNA-copy back into a viable, readable RNA molecule — which, if put into a cell, is akin to pressing “play” in the animation when the turtle is making a new egg. Of course, you also want to be able to copy, and modify, your DNA, just as you would a project on your computer.
In practice the logistical considerations are legion; but all that has ever been important to understand is that as long as these problems are solved, then what results is a DNA-based platform capable of storing, modifying, and growing “real virus.” All that needs to be achieved is going from DNA copy of the virus genes (in however many segments and copies) to a whole, readable RNA molecule inside a cell, and the animation restarts as the turtle making a new egg. There is nothing “different” about the virus from that part, bar any genetic modifications made in the lab.
This has always been my critique of the “GOF Bad!” trope — by hyperbolizing the possible unintended consequences of research designed to understand pathogenicity of viruses, it directs the spotlight in a very particular location, the hypothetical “leak” from the front-stage where viruses are researched, leaving in the dark the back-stage reality: Such research that would lead to a “leak” is built on a technology that renders deliberate viral modification and release possible.
At all events, it shouldn’t be too interesting to anyone but a practitioner how the logistical constraints of going from DNA back to virus are solved; it’s basically cooking, in that things from nature are heated up and spun around, except with 1,000 steps for every recipe.
And yet the details of how this is done have generated incredible confusion the last six months
The source of this confusion is no big secret: JJ Couey’s late discovery of the term “infectious clones.” While I would posit that “infectious clones” only needs to be understood as the DVD-extras content of how researchers solve the problem of going from DNA back to virus, Couey spun the term into an entirely new theory of the nature of SARS-CoV-2 and our reaction to it: “Gain of Purity.” I have previously given a timeline of my own interaction with the theory in the introduction to this post; it contains references and links to his own material:
A longer-running battle has persisted between Couey and Kevin McCairn, with the latter bringing his usual combination of technical acumen and gleefully horrible table manners to the plate. The final, complete critique of Couey’s “theory,” featuring Charles Rixey, appeared last month:
If the reader is somehow still unfamiliar with “Gain of Purity,” be assured there is no characterization I could give it that would not be deemed a misreading.
But what has been largely understood by followers of the theory, is that somehow “infectious clones” are lab-made, high-purity versions of virus-like-things that can’t persist in nature. Note that much of Couey’s theory spirals into claims that don’t even relate or depend on the necessity of “infectious clones” — as just described — being a valid concept. These other claims, e.g. “PCR just picks up background coronaviruses,” are sometimes also embraced by those influenced by his work; sometimes not. The whole thing is like a school lunch buffet of irrelevant juxtapositions, and one can take whatever beliefs they want for their own. I only wish to focus on “clones.”
(Just some of) the problems with “clones”
The reason DNA<>virus platforms can make high purity virus, is that the reassembled viral gene, when introduced into a cell, tends to instruct the cell to replicate itself with fidelity. Several passages of a virus “rescued” in this manner can go by with no detected changes from the source DNA genome. None of the work in maintaining purity is being done by the “clone” after transfection into the cell that results in viral replication; if and when impurity appears, this reflects a lack of fidelity in the DNA<>virus platform. We will see the same thing in the new study not involving “clones.”
The consequence of 1) is that all of Couey’s arguments about RNA virus genetic infidelity are based on a massively exaggerated notion of error rates, and thus have no value as arguments (they are misunderstandings, and as such should not ask us to believe anything about anything else).
Specifically, there is nothing about the ability to grow “pure” virus in a lab, capable of a highly infectious dose, that should ask us to believe that highly infectious doses do not naturally occur in viral transmission all the time. Humans got very sick from flu long before there were bio-labs. Couey asks us to believe (based on misunderstanding) this is impossible; but it literally happened.
Circling back to 1: If real-world viruses couldn’t make highly infectious doses at high fidelity, because of mutations and defects in the “swarm,” then how could cDNA be any use for virus research? Putting a viral gene into cells would, by the same flaws asserted to taint natural RNA virus replication, fail to give you viruses matching the source DNA in cell passage. (But Couey’s point is to be nonspecific at all times.)
And so regarding “clones,” Couey is inverting meaning:
In the lab, “clones” work because RNA viruses replicate with some fidelity. In Couey’s theory, “clones” work because RNA viruses don’t replicate with fidelity (but then how can “clones” work?).
In the lab, an “infectious clone” only has value because it is successful at making real virus. In Couey’s theory, “clones” have value because viruses don’t, and something besides a virus is needed to explain sporadic outbreaks of severe disease.
Circumventing these types of pointless misunderstandings is why I have repeatedly referred to these systems as “DNA<>virus platforms,” rather than the term of art “infectious clone,” because like all terms of arts it describes a mechanism of an action rather than the meaning of an action.
Science, like any hobby, is replete with esoteric terminology, and as an outsider and critic I take these plain-speak liberties all the time.
The New Way: PCR to 30kb virus in one
Herein we present an elaborated strategy for the rapid and straightforward rescue of recombinant plus-stranded RNA-viruses with high sequence fidelity, using the example of SARS-CoV-2. The strategy called CLEVER (CLoning-free and Exchangeable system for Virus Engineering and Rescue) is based on the intracellular recombination of transfected overlapping DNA fragments allowing the direct mutagenesis within the initial PCR-amplification step.
Kipfer, et al. provide a history of coronavirus DNA<>virus platforms:
Principally, the targeted engineering of RNA viruses to study viral biology requires the conversion of the RNA genome into a cDNA copy before it can be manipulated. Based on the large size of coronavirus genomes of between 27 and 32 kb, and the presence of typical sequences as nucleotide runs or so-called “poison sequences” that are hard to amplify in bacteria, it took quite some time before the first full-length coronavirus cDNA clone was published in 2000 (Almazán et al., 2000). The initially described bacterial artificial chromosome (BAC) cloning technology was quickly followed by other but similarly laborious techniques to yield infectious cDNA clones of coronaviruses, such as in vitro ligation or vaccinia-based cloning techniques (Thiel et al., 2001; Yount et al., 2000).
During the emergence of severe acute respiratory syndrome virus (SARS-CoV) in 2003 or Middle East respiratory syndrome virus (MERS-CoV) in 2012, applicable coronavirus reverse genetics methods remained largely unchanged. When SARS-CoV-2 emerged in 2019 previously established coronavirus reverse genetics methods including in vitro ligation and BAC cloning got rapidly adapted to SARS-CoV-2 (Fahnøe et al., 2022; Xie et al., 2020; Ye et al., 2020). Then, additional methods established for other RNA viruses became available: Transformation-associated recombination (TAR)-cloning in yeast (Thi Nhu Thao et al., 2020) and the "circular polymerase extension reaction" (CPER) (Amarilla et al., 2021; Torii et al., 2021). However, TAR still requires multiple steps of in vitro manipulation, and CPER still relies on the in vitro assembly and subsequent transfection of the 30 kb viral full-length product.
In short:
Converting an entire 30kb coronavirus gene to cDNA is difficult, because length.
Subsequent methods divide the gene into smaller bits to be ligated back together.
Most methods still require re-establishing a 30kb molecule to transfect into a cell that acts as a birthing ward for virus.
Then the authors mention a much simpler method previously only used for a smaller virus:
Another DNA-based method has been described by Lamballerie et al. in 2014 for the much smaller flavivirus model of about 10kb (Aubry et al., 2014). This method, termed “infectious subgenomic amplicons” (ISA), allows transfected overlapping DNA fragments to recombine within the eukaryotic cell into a full-length genome copy.
Here the cell only needs to be transfected with PCR-amplified fragments of the virus gene — they auto-recombine, and the virus gene results. The “clonal” middle-man is cut out of the equation. What Kipfer, et al. go on to report is successfully making an ISA platform for SARS-CoV-2, and finding the result to be high-fidelity and versatile.
The CLEVER platform
Cells sit constantly at the ready to repair and recombine DNA at all times. CLEVER feeds cells four2 PCR fragments spanning the 5’ (front)-end elements (akin to DNA "tags" before the viral gene which promote reading the latter), the SARS-CoV-2 gene itself, and 3’ (rear)-end elements, and these are reassembled by the cells into a DNA molecule that is expressed as the viral gene in RNA form, the same way your own genes express mRNA for proteins (but simpler). The cells, after doing all this, are centrifuged and the stuff at the top is inoculated onto Vero E6 cells to see if they start to die. They do.
No extra steps (sometimes, for other platforms, you need to express accessory viral proteins that help the RNA perform) are required. This also makes verifying viral recovery easier, as detection of SARS-CoV-2 Nucleocapsid protein can be used (which would not work if you had to express it to help the RNA to perform in the first place). The solution to custom-replicating and manipulating the longest RNA virus has turned out to be simple all along. Just PCR it.
Verifying fidelity
8 rescued viruses (with a marker-mutation to distinguish between wild SARS-CoV-2) were passaged twice to avoid contamination (carry-over) with the starting DNA, and then sequenced. 1 of the 8 was found to have a single mutation (“SNP”), with 13% of reads disagreeing with the consensus sequence. That’s 1 mutation for 231,040 bases (8 x 29,880). As in the previous section, this speaks to the fidelity of the virus itself — the replicating in the passage stage is being done by the virus, not by the input PCR-amplified DNA. But it also reflects on the fidelity of the PCR-amplification. Both have to be of decent fidelity in order for 231,039 of 231,040 bases to stay on script.
It’s important to understand, from the perspective of the claims in “Gain of Purity,” that this does not mean that there were no smaller minorities of mutant viruses — such as a variant that has infected a single cell and now is having that mutation multiplied. In fact, the authors sequence 5 single-cell virus clones as well, and 2 / 5 have mutations. But by definition the relegation of mutants to single cells does not imply that extra “purity” can be “gained” by some external means — it arises naturally in viral replication, or else there would be no “consensus” sequence.
Verifying mutability
Finally and unsurprisingly, the authors verify that their new platform facilitates record-speed experimentation with the virus. They create chimeric virus with BA.1 and BA.5 spike protein on a Wuhan backbone; they point-introduce N501Y and D614 to enhance and degrade spike function; they delete a gene (Orf3); they add a fragment. All of this is simply done via primer design modifications; it’s a Swiss-army-knife for virus genes. None of it makes something new possible; it only makes the technology simpler and easier.
And that’s…
Probably not such a big deal. One wonders why someone hasn’t thought to try it before — to “cut out the middleman.” In fact, maybe someone already has (as in, with SARS-CoV-2).
But hopefully, by having a new, simpler method of converting DNA for a virus into real virus, the question about there was ever some mysterious, incredible portent to the use of “clones” in existing methods can be put to bed.
If you derived value from this post, please drop a few coins in your fact-barista’s tip jar.
Kipfer, ET. et al. “Rapid cloning-free mutagenesis of new SARS-CoV-2 variants using a novel reverse genetics platform.” biorxiv.org
The initial proof-of-concept was with eight fragments; performance with four was faster and just as accurate.
Think I understand what’s being said here, but does that mean that a theoretical baddie could just put these 4 PCR made DNA fragments into us separately and we ( our cells) would combine them to make a nice new virus that would then spread onwards in the traditional way?
If so, how do they get the 4 different bits into our cells. We’d notice being injected with DNA wrapped in LNP. Would inhalable ‘ exosomes’ work - but how would they make those? Or do they just make the original one in the lab this way and then release it to do its magic the traditional way?
Does that mean SARS-Cov3~-Cov10 may also come out?