What distinguishes mutation from normal genetic variation

(8) Coronavirus update: Viruses always mutate

What are the consequences if the virus changes its genetic information? The virologist Christian Drosten says: Vaccines are probably not affected, but it could become more contagious.

"If the virus mutates, it becomes dangerous" - that was one of those fearful sentences that have often been uttered since the beginning of the year in connection with the coronavirus epidemic. Last week we spoke briefly about the subject of evolution. I will summarize the core message : Viruses mutating is a completely normal process, especially in connection with selection: which variant survives and which does not?

We are talking about these and other topics again today with Christian Drosten, the head of virology at the Berlin Charité.

The main questions of the episode at a glance

There are reports of a Chinese study that two variants of SARS Coronavirus 2 dominate - one more aggressive and one less aggressive. What does this mean for us?

What does a mutation mean for this coronavirus epidemic?

Do mutations have an impact on immunity, for example after an illness?

How fast do such mutations take place, how fast can they be?

If we can further identify and isolate chains of infection, is the virus possibly losing fitness?

What can you say about the situation in the hospitals right now?

Ten days ago you said you would still be traveling to Italy. How do you assess that today?


Podcast: Coronavirus update

Korinna Hennig: There are reports circulating about a Chinese study according to which two variants of the SARS coronavirus-2 already dominate - one more aggressive and one less aggressive. What does this mean for us?

Christian Drosten: Yes, of course you have to look at this study first if you want to form an opinion about it. And you can see: The virus populations, which initially looked like a wild bunch in Wuhan, have now become structured. Over time - because we now have more sequences - the entire bundle of these viruses can be divided into two large subunits. The Chinese scientists who published this weren't the first to see it. It has long been known that these two subunits exist. But they have now come up with a name for it and called one subunit “S variant” and the other “L variant” or “L types”. And they ultimately say that the S-type is an older virus type and the L-type is the newer and more aggressive virus type.

Korinna Hennig: What does more aggressive mean in this context - more infectious? Or make you sick, so to speak?

Christian Drosten: Yes, that's one of the many things mixed up in this study. So, first of all: It is very difficult, if not impossible, to say on the basis of sequences alone whether a virus is dangerous or not dangerous, whether it replicates more or less, or whether it makes more or less people sick. We can't tell from the genomes - apart from very, very rare exceptions. And that's not the case here. So we can't read something like that from the genetic make-up here.

Viruses mutate differently

And then in this study it is also the case that many misunderstandings have arisen. I'm not entirely sure if the authors who did this study really have any experience with virus evolution. I always read from the text - also from the language, from the words that are used there - that this is actually a background knowledge that comes from population genetics, from animal genetics. And there certain things are interpreted that are often seen in viruses, but which, viewed on a very long evolutionary scale, do not occur so often in animals. That would be remarkable in animals. In the case of viruses, they are self-evident and meaningless, but they are then interpreted as if the viruses observed here were something like animal subspecies.

And there you can see that there are several mutations - and they always occur at the same time. This simultaneity must mean something. This is nonsense. This simultaneity means nothing. This is simply the diversification of early virus diversity. And here we work under the expectation that mutations occurring at the same time (and this is the case in animal evolution in population genetic studies) must cooperate with one another and define a certain phenotype, for example shape the appearance of an animal. Otherwise recombination, i.e. sexuality in animal evolution, would blur such traces of simultaneity.

Download this script as a PDF

We speak of "linkage de-equilibrium", but we would not expect that at all with viruses in this situation. So what you see here is a trivial matter of course that is then interpreted.

And then the interpretation is an even bigger problem. Here you can see when which virus occurs and what you can see. In the early phase of the epidemic in Wuhan it was mainly the S-type that occurred and then more of the L-type in the later phase, and the older S-type then also spread internationally. And now something is made of it. And it is said: Well, you can see that over time more and more people have died in Wuhan, and there are also high case mortality rates outside of Wuhan. So the L-type must be the more dangerous one, who also makes more people seriously ill, so that more people also die.

But none of this is controlled at all. No correction factor has been incorporated for the fact that the epidemic itself developed in parallel and that the number of reports is completely different, that the reporting system has developed differently and that the hospital area in Wuhan was then overloaded. And if we have the time, I would maybe talk a little later about what could happen to the hospital sector in Germany. But that's just a side note.

In any case, all these influences were simply accepted and equated with the viruses. And then a cautious conclusion was made - in a scientific tone - that it said it could be that the L variant is associated with more severe trajectories. And then in a press release it became: “The L variant is deadlier, and it has now emerged.” And then this scientific article also speculates that the L variant was provoked by selection pressure caused by the isolation measures in Wuhan - and at the latest by this point in time this discussion will turn into a freak show. So it's almost entertaining what's going on here. And that could now be subsumed under “dangerous half-knowledge” on the part of the few who have scientifically interpreted this here.

Half knowledge becomes ten percent knowledge

And what then often happens in science communication and especially in social media, where that was once again exaggerated: People who have even less than half knowledge, maybe ten percent knowledge, peddle it and make it even more sensational . And that's something very dangerous that we really need to avoid right now. I also see that with other things that I say here in the podcast, for example. Then there are people on Twitter who say, “Ah, you did that wrong. You're a bungler. ”And then any numbers that I have mentioned somewhere are simply converted using some formulas. And numbers are generated from this, which are then displayed dramatically. And that is then tweeted again. And that's just awful! You can hardly stop that.

But it's something that's happening now. And you have to live with that. But that makes classifying the situation and dealing realistically with such an epidemic even more difficult for decision-makers. It's no use. If you paint the devil on the wall in social media and then think: now the drama is big enough. Now politics will probably give in and then let public life come to a complete standstill. At some point a stimulus and attention threshold is simply exceeded among decision-makers. And also with the normal population. You can't think you're adding up any dramatic numbers and then all the people follow you. This is utter nonsense. That only makes you unbelievable.

The glossary for the Corona podcast

What is an aerosol? What is a cell culture? Our glossary explains the most important technical terms from our podcast with virologists Christian Drosten and Sandra Ciesek. more

Korinna Hennig: We want to counteract the dangerous half-knowledge a little in this podcast and therefore explain a little more in context. I already said that you have to get to the basics, obviously. A few technical terms have already been mentioned. Briefly: Sequence refers to the genetic information of the virus.

Christian Drosten: Right, that's the genome of the virus, about 30,000 bases, RNA in that case - we sequence that. But at the DNA level, it comes out the same.

Korinna Hennig: We have already heard that we should neglect this message about S and L types. But the big issue of virus mutation - what significance does it have for this coronavirus epidemic? Especially for the question of whether this changes the properties of the virus for us humans?

Christian Drosten: So, we can count on the virus to mutate. We can already see that. There are already mutations in the genome everywhere, and we can also rely on the virus to change its properties in the process. But we cannot say whether this has already happened. Because to do this, we would have to study the virus under controlled conditions for a while - i.e. not under conditions of any number of reports that are not controlled, where in some cases even criteria are changed and where clinical situations change - these field conditions are not suitable for this.

We have to study the coronavirus in the laboratory for months

We have to create laboratory conditions. So we have to take the viruses and classify them phylogenetically. So we have to sequence the genome and say: where in the family tree is this virus now? And where are mutations in this genome that are of interest to us, that might have changed something? So for such a coarser classification that you say: Aha, where is it in the family tree, is that a virus that emerged recently? Or is that one of the original diversity that existed at the beginning? So the question of whether it is a new or an old virus and whether the new viruses are more dangerous or harmless than the old ones.

To do this, you have to examine these viruses in the laboratory. You have to isolate the viruses in cell culture and then just look, they replicate differently, are they differently sensitive, for example to human antibodies from serum, are they differently sensitive to cytokines, for example, to interferon? These are substances that our cells release to defend themselves against viruses in general. And we can simulate all of that in the laboratory. And we do that too. We’re starting this here in our lab these days.

We have already obtained a number of virus isolates. And we have experience with such studies. For the MERS virus, I am in the process of putting such a similar study together for publication. And then we saw that the MERS virus was also changing in terms of its appearance. We are now doing something like that for the SARS-2 virus as well. But this is work that takes months. You don't just do that quickly, it is associated with major disruptive factors. And you have to look very carefully in the laboratory and check yourself again and again through control experiments - and that's why it takes months.

Korinna Hennig: What kind of effects are at least conceivable in theory? If the virus mutates, changes its genetic information - then there are variants, it is not a new virus. Does this have an impact on immunity, for example after an illness?

Christian Drosten: Yes, that would be a function that could change on the virus. But we simply ask ourselves: Are functions changing and, if so, in which direction? And then we just have to realize that we have selection. The variations in replication, i.e. the multiplication of the virus genome, always offers new mutations. Almost all of these mutations are actually not good for the virus. They interfere with the virus or hinder the virus. And whenever a competing viral line is present, natural selection will remove the inferior viral line because the competing viral line will multiply more successfully. It has more offspring, i.e. more infected patients in the next generation. And that is why such a virus, once it first circulates widely in the population, and when there are competing virus lines in the population, then the virus will always be selected for better transferability. A virus will not be selected for a higher death rate or anything like that. Unless this higher death rate has an indirect effect on better transmissibility - because this virus does not want anything else. Nothing else is selected than transferability. Because transferability is fitness, as we say in evolutionary biology - i.e. the number of offspring per generation. With animals one would simply say: Aha, a pig of this breed gets an average of seven piglets and a pig of that breed gets twelve piglets, then fitness is probably higher in the latter case.

What progeny is there cannot be counted so directly in the case of viruses. We have laboratory methods, but that's all limping. Because these are epidemic viruses, we can simply count the number of infected people right away, which is a good surrogate for fitness.

Korinna Hennig: But that means that it is conceivable that the virus will become more contagious, to put it in layman's terms, once it has mutated?

Christian Drosten: Correct. Such a virus is then selected for a higher quality of being contagious.

Korinna Hennig: Does this also have an impact on vaccine development? We have already discussed them here, for example why it takes so long. Now a long process is under way, at some point you will be in clinical studies, with the testing of the vaccines. Then the virus mutated. Has everything been for the cat then, in layman's terms?

Christian Drosten: So, I'm pretty sure that in this case, a vaccine we're making today is still the right vaccine in a year and a half. I don't think the virus will really change any of the properties that affect its vaccine sensitivity during this time. We don't have to expect that here.

Korinna Hennig: The day before yesterday they clearly explained that immunity against a certain coronavirus can be related to that against another, related virus - using the example of the corona cattle virus, which causes diarrhea in cattle and does not affect people because people are infected with a “commercially available” corona cold virus (ie not the current SARS-2) to such an extent in the population that there is herd immunity. Is this topic also related to mutations, can that influence it?

Christian Drosten: Everything is always related to everything. But at this point, for the sake of simplicity, I would first say: No. That's actually a different discussion. When we get to the mutations, the next thing we have to discuss now: What can happen if such a virus is selected for transferability? In other words, what is the use of the virus? What can the virus do to increase its transmissibility? Now let's assume the original SARS virus: It's a virus that mainly replicates in the lungs. A virus like this can suddenly start to replicate in the throat through mutation processes. And that would surely be strongly advocated by evolution. In other words, that would be highly selective. Simply because this inevitably increases transferability.

The virus has already optimized itself

That is the consideration: the virus is now neck-to-neck and no longer has to travel from lungs to lungs. So shorter way.

Korinna Hennig: That is already the case with this virus.

Christian Drosten: It looks like this virus has already done it. Well, this card has already been drawn. At this point, the virus will probably only be able to optimize itself a little.Perhaps the virus would benefit if it next started to infect the nasal mucosa, and the dry cough and scratchy throat would suddenly turn into a runny nose. Then many more people would have their nasal secretions stuck to their hands all the time and would also infect many more people through contact transmission. So that would be a conceivable mutation that evolution would select.

Korinna Hennig: Because you said, Mr. Drosten, you don't yet know whether and how this has already happened - in theory: How quickly do such mutations take place or how quick can they be?

Christian Drosten: The theory is much, much more complicated than it is so often, we might come up with it next. I would finish one more thought: We have now said that there are certain things that are related to what we call the tropism of a virus. So the question is, which tissue does the virus actually attack? Which fabric is preferred? And then we had already said: lungs, throat, nose. So a tropism change or a tropism shift. That could be thought of as a phenotypic change. We have genotype and phenotype. Genotype is the sequence from which we can read very little. The phenotype is the observation of the organism. How the organism appears would be a change in phenotype. Now, however, viruses don't change their phenotype very often. That was the case, for example, with the Ebola epidemic. People who are not very familiar with viruses have speculated again on social media that the Ebola virus might soon become airborne - and then we are all lost. Virological experts said very early on: No, no, no, that won't happen anytime soon. There are hardly any examples of this in the evolution of viruses. I can only underline that here. Such phenotype changes are rarely observed! What can normally be observed in virus evolution are the smallest gradual shifts in things that ultimately go hand in hand with the level of replication, i.e. with the intensity of reproduction, of such a virus. For example, changes in the way such a virus does something against the general defense barriers of the cell. We already discussed the interferon system. Or there can also be optimizations in the way the virus cooperates with certain properties of the cell. Viruses are cell pirates! But this piracy can work well or badly. This can still be increased in efficiency.

Korinna Hennig: So the adaptation to the host?

Christian Drosten: Right, that's exactly what we're talking about, and these are small, gradual shifts that become visible over time. And because these are such small gradual changes, it is also so difficult to record them in the laboratory. And that is precisely why it is so difficult to record this on such rough data as epidemiological data. That is why fallacies so often come out of it. That's why you have to be very skeptical at the moment if you see any reports about it in the media and even in scientific publications.

And then we have to come back to another theoretical aspect. And that is the perception that evolution actually always involves positive selection. So what we discussed earlier. There are competing virus lines in the population, and that is why the virus line that has a higher fitness and a higher transferability will always prevail. Incidentally, for the experts among the audience: I am well aware that I am confusing transferability and fitness and many other terms here. These are just simplifications. You have to do that, otherwise it will no longer be understandable for a general audience what we are talking about here. So forgive me for the inaccuracies in my terms. In any case, a big consideration is that if we have this competition from viruses in the population, then in the end an easily transmissible virus always comes out or the virus does not lose any transmissibility. The wild type is stabilized when the wild type already has a high degree of transferability.

There is no competition with other virus lines

But now we have a completely different situation here. We assume here, throughout our discussion, that viruses compete with each other as if we were in a seasonal situation - as if the whole population were full of virus, as if the virus was already established. But that's not it at all. We have early chains of transmission here. We have a situation here where a virus has come out of an animal reservoir and is now starting to spread through the world. But in all of the countries in which we have just entered the virus at the moment, these are recent epidemic events. And there is still no competition. If I get infected and infect my wife, and she then passes it on to her acquaintance and the acquaintance back to her husband and the husband to three work colleagues: Then that's a transmission tree that we can trace on a piece of paper. And in my family and in the group of colleagues where the virus is brought in, there is no other virus at all. There is only this, and this virus can stay or go away. There is no competition with any other line of virus. Evolutionary biologists have a term for this phenomenon. This term is called “Founder Effect”. You can see that in animals too.

For example, let's talk about any birds. The birds live on a large land mass, and a piece of the land mass breaks off. This land mass now becomes independent and swims over to another island and the birds continue to reproduce there. But unfortunately it turned out that those breeding pairs who established a new population there lay fewer eggs on average than the original population on the large land mass (founder effect). It was just a stupid coincidence. So a breeding pair came to this island that is not so fertile. Under normal circumstances, the offspring of this breeding pair would disappear from the population within a few generations because they have less fitness and fewer offspring. In the population they are then more and more in the minority, while that does not happen in the island situation! Since they continue to multiply as they are - with the low level of fitness they have. This means that fitness losses are possible within the framework of a founder effect.

Korinna Hennig: If we can further identify and isolate chains of infection, does that mean, conversely, that the virus may even then lose fitness?

Christian Drosten: That is correct, that is the practical epidemiological conclusion from saying: Yes, it is still worthwhile to rarify infection events. But unfortunately you have to say that at the moment we are already in such a situation: Even in Germany we would already have virus lines spreading elsewhere, which would then take over the field. In reality, we are now at a point where we will have competition between virus lines. We don't have any competition locally, but in a few months we will have the same virus all over Germany because an unaffected virus line from another location will then take over the field. So unfortunately we are, I believe, already in a situation where this benefit of the founder effect can hardly be seen.

But we made such observations at the beginning of the SARS epidemic, for example. Then we know that the virus lost a protein by a stupid coincidence. In our laboratory experiment, we returned this gene to the SARS virus, which it lost in evolution, in the early human-to-human chain. And we see: The virus replicates more and better than before, and we can draw the conclusion from this: Perhaps the SARS virus lost portability at the beginning of its spread. And maybe that's why the SARS epidemic went so smoothly. It could be one of several reasons. And with the SARS-2 virus, nothing like that happened, at least nothing conspicuous that an entire gene would be lost.

Korinna Hennig: Mr. Drosten, if we promise something, we have to keep it. At the beginning you said you wanted to say something about the hospital situation.

Christian Drosten: Well, I can now see that the time, the half hour, is over now. And I think we should move within that timeframe as well. Otherwise it will be difficult to understand at some point. But I want to say something else - because I'm getting more and more comments on social media that the podcast is glossing over. The assessment that Drosten is spreading here is that it is a fabric softener assessment and in reality a big problem is rolling towards us here. And then some earlier statements are quoted as saying that I would have said the virus was just a cold - and the context is lost again. I want to repeat again - I say that a lot, and so do many other experts: There is no reason for individuals to panic here. For the individual, this is first and foremost a common cold. But for society, and especially for our medical system, it is a huge challenge when these many common colds all occur at the same time. That is the real challenge - for hospitals, especially for difficult cases. We have full waiting areas, you can't keep up with the testing and so on.

We also need to talk about hospital beds

But we have another problem entirely, and that hasn't been discussed in this podcast. And then maybe I would discuss that in the next week. That is the availability of hospital beds and especially in intensive care units. So intensive care beds and ventilation places. We have to talk about it, because that is also an important aspect of this assessment of what is in store for us. We also have to talk about possible scenarios, at what speed is it coming now? It is true that I have often said in the past that it is all just a matter of speed. And if it's going slowly, it's not a problem at all.

But of course we also have to talk about what happens when things run fast and what does fast mean. Here we come into a speculative area where even modelers can no longer help. So it is of no use if we now say that we are calculating such an exponential function here - then we can calculate that in April or May so and so many percent of the population will be infected. This is extremely difficult because we cannot model what the contact networks look like in the population. And that's why all these simple calculations of any exponential spread functions are too rough and won't capture reality. In addition, there are additional effects that we cannot factor in because we do not know those for this virus. An important disruptive effect: There are the temperature and other things that change in the transmission networks in the population when it gets warmer and when people distance themselves more because they are more outside. None of this can be modeled. And that's why we have to say at a certain point, now we just have to go through scenarios. It could come this way or it could come this way - and what does that mean in each case. Maybe that's something we should be looking at for the next week.

Korinna Hennig: That has already been noted, there are also many listeners' questions on the topics of “hospital situation” and “warmer temperatures”. A short request anyway, because we received so many questions from the audience. When we started the podcast ten days ago, you said you would still be traveling to Italy. The dynamic has changed a bit. Many listeners ask what your assessment of travel is now.

Christian Drosten: I find this individual travel advice very exhausting because you then refer to something that cannot achieve the resolution that you need. For a consultation, I would first have to reply: Where exactly should I go? And then I would have to look up what exactly it looks like at that location at the moment. And that's the difference between travel medical advice and this podcast. And that's why I wouldn't really like to answer these questions here. But I can tell you that a good friend of mine asked me this morning: I want to go to the football stadium with the kids on the weekend. What am I doing there? So I wrote back to him by SMS: The probability that a broadcast will take place in the stadium is possibly almost one hundred percent. That it affects you and your children, maybe at 0.0001 percent. I can not say more about that. And that's exactly where we are in this dichotomy between the individual and society. On the one hand, I say that I see large football stadiums full of people very critically, especially in the Rhineland, where there is apparently a lot of infection activity with the virus. At the same time the question, will you come to the stadium? If I were a football fan, I would probably say: Well, statistically it probably won't affect me personally. That's the problem.

Korinna Hennig: We notice again: We are clarifying the big picture and want to provide clarification so that it becomes easier to form an opinion for yourself and then to carry out individual travel advice with other people. Christian Drosten, thank you again today for a lot of clarification and patience with our questions. Even if you probably don't get much from your weekend during these times: We'll allow you at least two days off at this point and then we'll hear from you again in the next week.

Coronavirus update: all consequences

The virologist Christian Drosten provides expert knowledge in the podcast Coronavirus Update - together with virologist Sandra Ciesek. Here is an overview of all the episodes. more