On January 10, 2020, Chinese health authorities published the sequence of the Covid-19 virus. At that time, it was already a month after the first identified cases of this new disease were hospitalized in Wuhan, China, and two weeks before the first cases were admitted to French hospitals. Viral genetics experts around the world quickly pointed out several strange characteristics of SARS-CoV-2, the Covid-19 virus. One question nagged at them: could the virus have emerged following an infection caused by an animal species (zoonosis), or did it come from manipulation at the Wuhan Institute of Virology (WIV), where world experts in coronaviruses work? But as in any investigation, even scientific ones, clues can be misleading.
A fragment identical to that of a pangolin coronavirus
At the end of the Covid-19 virus's Spike protein, the fragment responsible for host cell recognition is of interest to researchers. Called RBM, for Receptor Binding Motif, this sequence recognizes the ACE2 protein in human cells to enable infection. However, the RBM of SARS-CoV-2 is unlike that of any other known SARS virus and appears to be particularly well-suited to infecting human cells. The only sequence published prior to SARS-CoV-2 that is from the same viral lineage is a pangolin coronavirus sequence, published in October 2019 by Chinese researchers from Guangdong and Guangxi (Chinese provinces far from Wuhan, editor's note),” he explains to Science and Future Alexandre Hassanin, a mammalian phylogeneticist at Sorbonne University, remarks. Surprisingly, the researcher notes, the RBM of this virus is almost identical to that of SARS-CoV-2, with 68 amino acids (the "building blocks" of proteins) identical out of 69. So, the RBD which at first seemed strange and therefore suspicious actually existed naturally", reasons in The Conversation CNRS researcher Florence Débarre, who advocates the theory of zoonotic emergence. If it emerged naturally once, why not twice, these scientists postulate. But a second, related element undermines this conclusion.
The furin cleavage site, four “letters” that sow doubt
"I decide to look at the viral sequence more closely, and I see things that worry me, notably the furin cleavage site", says Kristian Andersen, viral emergence expert at the Scripps Institute (United States), during of a hearing to the United States Congress. On its surface, SARS-CoV-2 carries a protein called Spike, which recognizes the ACE2 protein, the gateway to our cells. But to pass through this gateway efficiently, the Spike protein must be cut ("cleaved") by a protease – an enzyme that cuts proteins – such as furin. However, the Spike of the Covid-19 virus has a cleavage site for furin, a human enzyme whose cutting transforms Spike from a safety pin to a key. Compared to SARS-CoV-1 (2003 epidemic), which does not have one, and to all sequenced sarbecoviruses (named after the group of coronaviruses to which the Covid-19 virus belongs), SARS-CoV-2 shows an insertion of 12 nucleotides (the "letters" that make up the genetic sequences), encoding the four amino acids of this furin site: RRAR.
Other viruses with a furin site exist, including HIV, Ebola, H5N1 avian flu, and MERS, the coronavirus that caused an epidemic in 2012 that originated in the Middle East. But these examples are genetically very distant from the Covid-19 virus. None of the hundreds of known genomes in the sarbecovirus group possess this furin site. It was at the time of this insertion that the virus acquired its pandemic potential.", underlines with Science and Future virologist Etienne Decroly, research director at the CNRS and author of "Experiments in Virology - Benefits and Risks" (Editions Quæ). Although regularly exposed to sarbecoviruses, which therefore lack a furin site, only 4% of the populations of Yunnan, a southern region of China, who live near bat caves, have antibodies against them, confirming the low transmission capacity of these bat viruses between humans.
A spontaneous appearance of the furin site is possible but unlikely
Initially, the researchers imagined that this furin site could have appeared spontaneously, knowing that SARS-CoV-2 mutates very quickly. Our hypothesis was that it would have circulated within human or terrestrial mammal populations, and that at some point this site would have been created by mutation, would be selected, and would emerge. This is exactly what happens with influenza viruses. (bird flu, editor's note),” he explains to Science and Future Marc Eloit, virologist and former director of the Pathogen Discovery Laboratory at the Pasteur Institute. But when he and his teams tried to experimentally mimic this natural evolution in the laboratory with six "passages" (transmissions) between transgenic mice or human cell cultures or even monkeys, no furin cleavage site emerged. Given the low transmissibility of the virus without this site, its spontaneous appearance would have had to be earlier than this to make its natural emergence likely.
There is another option that would explain the natural emergence of the furin site in SARS-CoV-2. If not by mutation, then by recombination.", reasons Marc Eloit. Where mutations are one-off errors that accumulate over generations of viruses, recombinations involve the accidental integration of an entire sequence of external genetic material. But for this to be plausible, the sequences surrounding the one that was mistakenly integrated must be similar enough for the enzyme responsible for synthesizing the viral genome to mistake them. The suspect is a furin site found in a human protein in our upper respiratory mucosa, called aENaC. But the researchers found no sequence homology, invalidating this possibility. " I no longer believe in this hypothesis.", concludes Marc Eloit. Another option: the donor organism would be a coronavirus carrying this furin site, currently unknown, which would have co-infected at the same time an intermediate animal species, also unidentified.
Between improbable biological event and human creation
The difficulty in this case is that in biology, even improbable things can exceptionally happen. Thus, and even if it is extremely rare, sequences can integrate without necessarily homologies on either side and explain a natural emergence of the furin site. This would have required a co-infection of the same bat by the progenitor (parent) virus of SARS-CoV-2 and another virus carrying the furin site, and for recombination (exchange) to take place during viral reproduction. Improbable, especially given the total absence of any trace terrestrial mammal hosts infected by a progenitor of SARS-CoV-2, but not formally impossible. Another hypothesis is that sometimes it is repetition errors made during the multiplication of viruses and their genetic material that lead to the appearance of new sequences. But there is no trace of this in SARS-CoV-2. I've done a lot of work on insertions and deletions during evolution, and here we have 12 nucleotides that come from nowhere and create a site that is completely identical to the human protein. hasENaC on eight amino acids, which is very unlikely", supports Alexandre Hassanin.
Certainly, "tAny emergence of an epidemic is statistically unlikely", specifies Marc Eloit. But just like field epidemiologist Renaud Piarroux, who heads the parasitology and mycology department at the Pitié Salpêtrière Hospital (Paris), he weighs the probability that this emergence will occur precisely in 2019 in Wuhan. We haven't had a coronavirus pandemic since 1889 (the “Russian flu”, which is said to have caused a million deaths, editor’s note), so the probability is not that it will take place in Wuhan, but in Wuhan in 2019, where one of the only two laboratories in the world with expertise in crossing the species barrier by SARS-type viruses is located“, analyzes Renaud Piarroux. “ Having a natural emergence of a coronavirus pandemic in Wuhan in 2019 that had the same profile as what was proposed by the DEFUSE project is an extremely unlikely event.. » The DEFUSE project had been submitted in 2018 for funding by the NGO EcoHealth Alliance and should have been carried out, in particular, at the WIV; it proposed genetic manipulations of viruses strongly reminiscent of SARS-CoV-2, including the insertion of the furin cleavage site, a year before its appearance. If rage emerged in the 15the arrondissement of Paris, one would probably start by looking towards the Pasteur Institute before imagining that rabid foxes are walking along Avenue de Suffren", says Marc Eloit ironically.
Read episode 2 soon : how the DEFUSE project fueled the theory of a virus released from the laboratory.
English 
French