‘There are better poisons if you really want to kill someone’

The chemical weapons expert who led the OPCW’s mission to Salisbury after the Novichok attack on the Skripals explains Alexey Navalny’s situation

Russia’s most prominent opposition figure, Alexey Navalny, has been in a coma for more than two weeks. On August 20, his flight home to Moscow was forced to make an emergency landing in Omsk after he became violently ill. Russian doctors treated Navalny for roughly two days before he was transferred abroad in an air ambulance to the Charité Clinic in Berlin, where specialists found evidence that he’d been poisoned with cholinesterase inhibitors. Physicians have been unable to identify the exact substance responsible for Navalny’s condition, but German officials announced on September 2 that experts have collected “unequivocal evidence” that he was poisoned with a substance similar in composition to the nerve agent Novichok. To understand more about Navalny’s poisoning, Meduza science editor Alexander Ershov spoke to Marc-Michael Blum, a biochemist who studies decontamination, countermeasures, and mitigation of chemical warfare agents. In 2018, following the Novichok poisoning of Sergey and Yulia Skirpal in England, Dr. Blum led the team sent by the Organization for the Prohibition of Chemical Weapons to Salisbury and Amesbury.

We currently know very little for certain about what happened to Alexey Navalny. Perhaps the most substantive information comes from the Charité Clinic’s press release on August 24: “Clinical findings indicate poisoning with a substance from the group of cholinesterase inhibitors. The specific substance involved remains unknown, and a further series of comprehensive testing has been initiated.” Can you explain what that means? How is it possible to detect that someone was poisoned without identifying the poison itself?

Yes, it’s possible. You take a blood sample and in the blood you have cholinesterase. And there are tests where you can check if they work. Cholinesterase is an enzyme — it helps to speed up a certain chemical reaction [vital for the transmission of nerve impulses] — and there is a test where you are not using acetylcholine but acetylthiocholine (very similar) and the cholinesterases work on that, as well. What you get is a product that you can react with something else and it creates a yellow color. That’s a classical test for activity. With that, you measure the activity of the cholinesterase in the blood plasma. What they [at the Charité Clinic] probably saw was significantly reduced activity level.

It’s only when you get to almost zero levels that your symptoms become severe. With the symptoms Navalny showed and the fact that they had to put him in an induced coma and the fact that they probably tested the cholinesterase in his blood and probably saw very low levels, that triggered the press release saying: we see it as a cholinesterase inhibitor, but we don’t know which one. Because they just tested the enzyme activity. Now, in the second step, you’re looking for the poison.

It’s very hard to say, of course, but what’s the likelihood in your view that the exact substance will be identified?

If your cholinesterases are in fact inhibited and most of it is inhibited, it means something is sitting inside or on that enzyme. It’s still there because, if it’s not there, then it would be active. When you measure blood that shows big inhibition of the cholinesterases, the poison is there. If the poison were already gone, the activity would return. That’s a good sign that the poison can be found. There is a cholinesterase in blood that we mainly look at called butyrylcholinesterase that’s soluble and easy to work with. If you have an inhibitor, it will inhibit that cholinesterase, as well. Then you look at it and there are two possibilities: some of the inhibitors are actually reversible, meaning that they stick like glue but can come off again, and some others like nerve agents actually bind to the enzymes and you cannot break that anymore. It’s really just like super glue. It’s irreversible.

Then you look for the fragments from these poisons that sit on the enzyme. So you take the enzyme, you cut it into small, little pieces, and there’s a very characteristic piece and, on this piece, you expect to find something sitting on it. Then you use mass spectrometry [an analytical technique that measures the mass-to-charge ratio of ions] and you know the mass of the fragment if it’s normal and then you expect that fragment also to be heavier because something is on it. And from these exact masses, you can deduce what’s on it. Once you think you’ve found something, ideally, if you have that poison available, you take some blood plasma from somebody else, and in the test tube you add it and you also measure this. If it looks identical, you have confirmation that you found the poison.

The doctors who treated Navalny in Russia say they tested him for poisoning with cholinesterase inhibitors but couldn’t get confirmation. Is there any rational, scientific explanation for the disparity between what Russian and German doctors have said? Maybe the physicians in Omsk didn’t have sensitive enough mass spectrometers? Is that possible?

I don’t think so. The first step is testing for cholinesterase activity — quite a simple test. You take the plasma, you have your acetylthiocholine, and you do that in a little glass cuvette and it creates color. All you need is a photometer, which is very ordinary lab equipment. You have that everywhere. You don’t need super-sophisticated technology for that. With that, you can do the test. If they found cholinesterase inhibitors [in Berlin], it should have been found in Russia, as well.

What you hear now is speculation, on Twitter and so on, that maybe he was poisoned on the way [to Berlin] on top of this metabolic condition he had. Of course, that’s a possibility, but how likely is that? He was severely ill when he was found and taken off that airplane.

A question I ask myself is that the Russian hospital said he got atropine [a medication used as an antidote to certain types of nerve agent and pesticide poisonings] but they also say he had very low blood sugar. When you have very low blood sugar, normally your heartbeat goes very fast and you sweat. But atropine is given if your heartbeat is very low — you don’t give it if it’s really high. It’s very unusual. If they gave him atropine, there must have been a reason to give it, and I think they have not yet explained why they gave him atropine. You’re not just giving atropine to somebody who is severely ill just as a precaution. There must be a clear indication of why you’re giving a certain antidote or why you think you need that emergency medicine to stabilize the patient. I haven’t read anything saying why he received atropine. There’s no good explanation at this moment.

For a long time, Navalny’s transfer to Germany wasn’t permitted. The air ambulance that came to Omsk for him was stuck on the tarmac for hours and hours. As a result, some of Navalny’s supporters theorize that he was kept in Russia as long as possible to give the poison in his body time to disintegrate, making detection impossible. How plausible is that theory and how quickly do poisons break down inside the human body?

Some of these compounds hydrolyze quite fast in the body. The nerve agent sarin, for example, hydrolyzes very quickly. But it attaches to the enzyme permanently and you can always measure that. The body is remaking cholinesterases. If you’ve knocked out the cholinesterases, it takes about two months to get back to normal. Effectively, you can take samples three or four weeks after exposure and you’ll still find it. What goes quite fast are the metabolites in the urine — only two or three days and then it’s gone. But what sits on the protein, on the enzyme, that stays for a very long time. If it had been hydrolyzed and completely gone, Charité would not have measures inhibited cholinesterases.

So keeping him there just for the poison to disappear doesn’t really work in the case of this poison group. There are other poisons where waiting for a couple of hours is quite effective. Everyone says he was poisoned by drinking a cup of tea, but that is not proven. He could have had contact with a poison somewhere else: earlier on in his hotel, on the way to the airport, in the airport, or wherever. People focus on the tea, but we’ve recently had other poisonings like Sergey Skripal, for example. Yes, he went to town with his daughter and they had a pint of beer in a pub, yet the poisoning occurred somewhere else. For me, it’s not totally clear that it was the tea. It depends on what they will find in the end — if the tea theory is plausible or if it’s more likely that it happened somewhere else.

But if it happened somewhere else and he came in contact through the skin, then, of course, it’s also quite dangerous for everybody around him because he’s contaminated and he could potentially spread that contamination to other objects and people he touches.

Now that German specialists have determined that Navalny was poisoned by cholinesterase inhibitors, is there any chance they will be unable to identify the exact substance responsible?

Is it possible they won’t be able to identify the poison? Yes. Because there’s such a wide range of possibilities here and the concentrations you look at are very small, which means you have to look for every single possibility. First, you go through the usual suspects, the usual nerve agents, the most-used pesticides, and so on, and it can be quite a long list. Also, you would look to see if there’s still something in his urine or some hydrolysis product in the blood — a small molecule or a metabolite. Maybe they’d take skin samples to see if anything’s still there that might help support the analysis of the cholinesterase in the blood.

You’re looking from a lot of different angles. I’m still quite confident that they’ll find what it was, or at least the general class. One issue is that it isn’t the whole poison molecule that attaches to the cholinesterase. A part is lost. With sarin, for example, the fluorine is lost. So they might say the poison was something that looks like sarin, but instead of fluorine, there was cyanide or something. Or bromine — another possibility. You don’t know because that part of the molecule is lost, but the rest looks like sarin and it was powerful enough to attach to the cholinesterase, so you know it’s definitely a nerve agent. I’m actually quite confident that it will be found because his condition indicated quite severe poisoning, which means he probably ingested or absorbed quite a high dose.

In order to identify a poison, do you need to know more than its chemical formula? Or do you need to have a sample of the substance itself? How long might it take to complete all this analysis?

You can do some interpretation of the spectra you find. Ideally, once you know what you think it is, either you have it in your stocks and you can take it from the drawer and use it, or alternatively, if you don’t have it, there’s the possibility of making it — synthesizing it in very small amounts. Just enough to carry out your tests. That would take a few days extra, of course.

In principle, the whole thing is a three-step process: the first step is screening. You just screen to see what it might be. The second step is identification. Of course, because this is a very high-profile case, you want to be sure for the third step, which is confirmation. You confirm with another method, you use a reference chemical, and only when you have that are you probably confident enough to put it in a report. I would guess that we’ll see it this week or maybe next week.

So, by next week, we should know exactly what kind of poison we’re talking about?

At the latest. If we don’t hear by then, it’s more likely that we will not know what it was. Also, I don’t know who makes the decision to make that information public. Once the chemical is identified, I’m very sure the information will be given to the treating doctors because it might affect his therapy, and they will also probably tell his wife. By that time, it will probably be made public, I would guess. But it’s unknown how they’ll go about this. The laboratory talks to the Charité, but there might be some consultations about when to make it public because it’s a political case.

At this point, though, we can at least be certain that we’re looking at a chemical warfare agent that’s difficult to obtain and not an ordinary pesticide that acts in some similar way?

Some of the pesticides are organophosphorus compounds, just like the nerve agents — a bit different in structure, but the big difference between them is the toxicity. You need to ingest a lot more of the pesticide to have these severe toxic effects. Let’s take something very poisonous: VX. A little drop on your skin could kill you. If you take one of the usual pesticides, a drop will not kill you. The difference in toxicity might easily be a thousandfold. There are really big differences.

If you look back in history at how the nerve agents were discovered, it was normally work on new pesticides and insecticides. Some of these turned out to be very effective but unfortunately too toxic for use because they were really toxic to humans. Those developments would then go to military use. They’d say, “Ah, this is probably a very good warfare agent.” And those that are effective against insects and not that toxic to humans would go for civilian use and become pesticides.

There are also some other cholinesterase inhibitors. For example, there are even some medicine therapeutics used against Alzheimer’s, and there are carbamates — a nonphosphorus class of inhibitors. All these substances can also have such effects. That’s why it’s hard to find — you have to look at so many different classes of chemicals. At the moment, given what we know, it’s really very difficult to speculate. If it was a normal pesticide, you’d think there must have been a lot in that tea! (If the tea hypothesis is right.) It wouldn’t be enough just to touch some surface contaminated with a pesticide — that wouldn’t be enough to kill you.

That said, there are ways to enhance chemicals’ penetration through your skin. That’s also a possibility. There are still so many unknowns. I think the next thing to do is to wait for the identification. Once we have an identification of what the chemical is, then we go forward and ask if it’s still likely that the tea was the source of the poisoning. Between the tea and the moment he really dropped down on the airplane, there was a bit more than an hour. Is that realistic [if the tea theory is correct]? Even a lot of pesticides act faster. After 10 or 15 minutes, especially on an empty stomach, people would start to show symptoms.

But what we need to know now is the identity of the chemical.

Why does Navalny remain in a medically induced coma if his condition is stable? What are the doctors waiting for?

In principle, they want to see his cholinesterase levels go up again. If you’re really knocked out on the cholinesterase symptoms you show, even with atropine, there will be the typical symptoms: you cannot control your muscles, you need ventilation, you cannot breathe on your own, and being intubated for the whole time is quite harsh on the patient if he is conscious.

Normally, you sedate people and put them in a shallow coma. Without pain, they can tolerate the ventilation much better. Also, you don’t take them out of the coma abruptly, they just slowly stop administering the medicine responsible for the coma. And you monitor the patient’s reaction. If you see that he’s in a lot of pain or he’s getting spasms, you’ll keep him in the coma, measuring his cholinesterase in parallel. If he’s back to 10 or 15 percent, you can start taking him out again. In the end, that decision lies with the treating physician — they’re the guys who really know. Basically, it’s about reducing pain. Being ventilated is not a very pleasant experience, so it makes sense to keep the body down. [Navalny’s] coma might easily last for a couple more weeks.

They’ve said his condition is improving, that it’s severe but stable, which means it’s not life-threatening anymore. But “severe” probably means we’re keeping him in that coma for the time being.

That amount of time is roughly what the body needs to synthesize enough cholinesterase to fulfill its function as neurotransmitters?

Yes, the body has to remake it. Atropine doesn’t help with the blocked enzyme. It only helps by working against the symptoms. There’s a second class of therapeutics called oximes (pralidoxime, obidoxime) and some organophosphate compounds that can reactivate the cholinesterase. If that works, they can help get you out of that condition, but it only works with certain compounds and not with others. If it doesn’t work, you simply have to wait until (a) all the poison is gone from your body and cannot inhibit the newly made cholinesterase, and (b) your body reproduces [the cholinesterase], which can take several weeks.

With each of these failed poisonings, we see theories that the culprits wanted only to frighten their victims, not assassinate them. Is it possible to administer poisons in concentrations calculated exactly to bring people close to death without killing them?

Assuming it’s a military nerve agent, I would say it’s definitely an attempt to kill. You can’t fine-tune it. With these kinds of compounds, you basically have a curve where you start seeing effects at a certain dose and then they die at the next dose. With the nerve agents, this window is extremely small. Between the first symptoms and death, the window is very narrow. It’s wider with the pesticides. You’ll see symptoms, but you’ll still need a lot more of the stuff to actually kill you. That’s also true for some other inhibitors. That’s why I say it’s important to see which poison was used.

If this was a military nerve agent, you would say it was, in all likelihood, a real attempt to kill him. If it was something else, you could still speculate that maybe it was just a warning. On the other hand, it doesn’t really look like a warning, given the severe condition he was in.

If the use of chemical warfare agents is always attempted murder, why have these attempts failed time and again? It brings to mind, of course, the most striking recent example: the attack against the Skripals in England.

It’s very cynical to say this, but there are better poisons if you really want to poison a single person and kill him. There are poisons that are much harder to detect and faster to act. Just because something is used as a chemical warfare agent, where you think about battlefield use and using it in grenades — the poison there is different from the perfect poison to assassinate somebody, where you need a different profile. [Navalny’s poisoning] is very exotic. You could also say, if you want to kill somebody, people might just disappear, have a car accident, whatever — or just they’re just shot. By using poison, there’s probably an additional message to bring attention to the case.

In the case of the Skripals, one has to say that both he and his daughter were extremely lucky that the ambulance arrived as fast as it did when they collapsed, and gave them treatment and moved them to the ICU so quickly. There are numbers out there saying that they probably would be dead if the ambulance had arrived 10 minutes later.

Of course, we’ve also had very exotic poisonings that succeeded: [former KGB and FSB officer Alexander] Litvinenko with the polonium — again, very exotic. Why go through all the hassle of using polonium, which is very hard to get? If you simply want to kill somebody, there are probably easier ways.

It’s very hard to say. Maybe it’s also — and this is pure speculation — that someone wants to carry out an assassination but if he or she survives, at least the message has been sent. It might also be simply a lack of experience on the operator’s part. Whoever did it — the people who actually gave him the poison or put it somewhere — are probably not the experts in its poisonous properties. They were probably told what to do and they used a bit too much or a bit too little.

I think it’s always very difficult when using a contact poison — something you have to touch. You never know what will happen next. The person touches it with his hand and, for some reason, he thinks: I should wash my hands. Then most of the poison is off again. Or others might simply not notice [the contact point] and miss it entirely. It’s very unpredictable. If you put something in a drink and you know the person will drink it completely, then that’s a bit easier.

But all that is pure speculation. We should wait to find out what it is. That might enlighten the situation a bit about what’s behind it. Chemical analysis alone will never be able to show who the perpetrator is. You need police work for that. Even if you find small amounts of poison in his body, the concentrations are so low that it doesn’t have a fingerprint or signature where you can say: oh, based on these properties or impurities, it was probably made in this or that way. That might be possible if you have a lot and you can analyze the pure substance, but that’s implausible in a person.

Interview by Alexander Ershov