Tonight’s news suggests the coronavirus outbreak may have originated from pangolins. Details are very sketchy, almost nothing at all beyond a claim of high similarity. My impression is this claim comes from a press release by South China Agricultural University. Initial discussion online suggests this is a more likely candidate for an intermediate host than previous speculations. It is also ‘science by press release’, where claims are made in advance of publication of preprints, research papers or data.
Pangolins are also known as ‘spiny anteaters’, small mammals covered in scales that feed on ants and termites. They’re nocturnal, and roll themselves into defensive balls when provoked.
Pangolins worldwide are a threatened species. From wikipedia,
Pangolins are in high demand for Chinese traditional medicine in southern China and Vietnam because their scales are believed to have medicinal properties. Their meat is also considered a delicacy. 100,000 are estimated to be trafficked a year to China and Vietnam, amounting to over one million over the past decade. This makes it the most trafficked animal in the world.
It’s interesting to see the wikipedia entry mention, ‘They are susceptible to diseases such as pneumonia’ (also noted in earlier research).
There is promotion that the third Saturday in February should be World Pangolin Day. This year that is February 15th.
This (translated) WeChat blog post is as close as I can get to an original source at the time of writing,
This report lacks any detail, but some suggestions can be gleaned from other reports.
CCTV AsiaPacific reports,
After analyzing over 1,000 metagenome samples, researchers from China identified #pangolins as a potential intermediate host of the virus. Through molecular biological detection, the researchers found out that the positive rate of the β #coronavirus in pangolins is 70 percent.
Metagenomics is also known as environmental genomics. Rather than taking DNA from an plant or animal (or whatever else), researchers take samples of the environment—dirt, swabs of surfaces, and so on, then directly sequence whatever DNA is in them. From this they then try work backwards to see what species’ DNA was in their samples.
You could for example, take sample of handrails in the New York subway stations, then try work out what bacteria are on the handrails.
By ‘the positive rate’, I suspect they mean how often pangolins were found to have this virus. This would imply they’ve tested a number of pangolins to find out if this virus is common in them.
(The ‘β’ likely refers to that there are four main human lineages of coronaviruses referred to by the first four greek letters, α (alpha), β (beta), γ (gamma), and δ (delta). SARS and MERS are both β coronaviruses.)
Edward Holmes on Twitter says this work is ‘Supported by a study I am involved in’, and that the preprint is in preparation. A longer introduction to the claim can be found in Xiaowei Jiang’s tweets.
Photographs taken at the Wuhan market before it was closed at the end of last year show cages packed with snakes, porcupines and foxes. Media said about 50 types of wild animal were on sale at the market, including endangered pangolins.
Why is this more likely?
The early reports lack detail, but a few clues suggest this is a more likely candidate than earlier claims –
1. The high sequence identity, 99%. This sounds as if it is sequence identity across the full length of the coronavirus genome. In earlier claims, leaving aside questionable biology, there were enough differences to require speculative guesses as to how differences came about. With very high identity that is less of an issue. We’ll still need to know what the specific differences are, view them within lineages, and if they make biological sense.
2. Informal comments suggest several different, but closely related coronavirus genomes similar to 2019-nCov may have been found: ‘The data I am seeing makes it look like pangolins have multiple CoV lineages that are close to 2019-nCoV. If so, they might be better classed as a ‘reservoir’. Need a closer look though.’ It’s what we might see in a stable host species for the virus, with different varieties of the virus developing within them.
3. Earlier discussion offers pangolin as a candidate, based on a different project published late last year identifying sendai and coronavirus infections in pangolin. The similarities reported there are complex enough that this remained an open question. (More on this below.)
4. Mammals are a more likely source for a spillover host than snakes (or fish) as their biology is more similar, and many mammals are known to be infected by coronaviruses.
Not so fast
The discussion in point 3 (above, by Matthew Wong in Joe Petrosino’s lab at Baylor College of Medicine) has a few tantalising hints, but also some unresolved questions.
Near the end of the introduction, they note that,
Most of the dead pangolins had a swollen lung which contained a frothy liquid, as well as the symptom of pulmonary fibrosis,
This study was of Malaya pangolins in the Guangdong Wildlife Rescue Center, in Canton, Southern China. The samples are of dead animals, but it’s not possible to say the viruses observed were the cause of death.
Some viruses enter cells by binding to a receptor protein on the surface of the cell. The discussion notes similarities in just one part of the protein that binds the receptor.
This partial similarity tentatively suggested a recombination between bat and pangolin coronaviruses to form the virus found in human patients. That would want to be explained by the new 99% identity data.
If pangolins are a stable host, then this is a different ultimate source than from bats. Against this, we’d expect reservoir hosts to be tolerant of their infections rather than susceptible to them.
We’re not expecting 100% identity as viruses generally evolve fairly rapidly, especially when adapting to a new species, but even so we’ll want to be able to explain the differences. Even with high identity, I’d prefer this set into context of a lineage tree (a phylogenetic tree). Lineages matter, and genetic differences are made clear when compared to lineages.
We need the paper
Without the evidence it’s ultimately an empty claim!
There’s also an annoying phrase biologists use: ‘consistent with’. The scanty claims I have seen so far are ‘consistent with’ pangolins being a candidate intermediate species for the coronavirus outbreak. That’s neither here nor there, but it is more promising than earlier claims. Reports over the next few days are likely to clarify this.
As I write, Nature has released a brief summary along similar lines. In it they report that, ‘Liu Yahong, president of the South China Agricultural University, told the press conference that the results would be published soon to help efforts to control the coronavirus.’
At least the sequence data could have been released at the time of the announcement, surely? I guess ‘whatever’. This is all a mad rush, after all. That said, I’ve never been a fan of science by press release. It’s happening too often with “exciting” discoveries. Even when it seems reasonably justified, as in this case, it allows claims and speculation to proceed ahead of critique. More on that another day, perhaps.
Perhaps this can give a name?
As an afterthought, perhaps we might soon get a sensible long-term name for the virus?
One frustration with this outbreak is that the virus has an awkward temporary name, 2019-nCov (2019 novel coronavirus). If pangolins proves to be the intermittent, intermediate or reservoir species, then perhaps this can give us a long-term, common-language name?
Other articles in Code for life
The 2019-nCov coronavirus outbreak: all together now This pulls together some of what scientists are working on, a few cautions about reading ‘fast science’, and a few suggested places to read.
The coronavirus outbreak: what is R0? You’ll hear a few statistics mentioned, one is R0. This explains what it is, and why it’s an ongoing estimate.
Finding platypus venom Researchers cleverly did not extract the venom, but created possible venoms by comparing the platypus genome with known venomous proteins and expressing the genes that matched.
Temperature-induced hearing loss This was a surprise to learn: a few rare individuals have lose their hearing when they have a high body temperature.
The inheritance of face recognition (should you blame your parents if you can’t recognise faces?) Prosopagnosia is surprisingly common and has fascinated me for years.
Kumara are transgenic They’re natural GMOs and serve as an illustration of how arbitrary calling something a GMO is.
If you’re looking for more reading on this outbreak, I’ve linked to my older pieces above, but check out this post collating all articles Sciblogs about this outbreak.
1. They’re one creature I’ve wanted to see while I am still in Borneo. A catch is they’re noturnal and live in the forests. I believe night safari at the Singapore Zoo has some, possibly also the zoo at Taiping.
2. I’m not all that keen on science by press release. Even if well-meant, as it likely does here, it lets news fly around the world without any means of critiquing it. Is this particular case there is a little earlier commentary that helps, but it’s still patchy stuff to judge this by.
3. Without digging into the data, this looks to be a similarity of the protein, not the genome, however. (The genetic code for identical proteins can differ. It’s more convincing if the genome code for the protein has strong identity, as that’s a better indication they’re from the same genetic stock.)
4. Arsenic DNA, anyone?! (This will almost certainly be sounder, but still.)
Pangolin, Manis javanica.
Source: Wikimedia Common, CC ASA 3.0 Unported.