Molecular Toxicology

Biochemist Ian Wilson on toxicity of paracetamol, clinical trials, and safe drugs

videos | June 5, 2017

Molecular Toxicology is the science of understanding why things are toxic. So we have toxic foods, the air can be toxic if it’s very polluted, some industrial chemicals can be quite toxic, drugs can be toxic if taken in overdose or used inappropriately. Now, every molecule causes toxicity in different ways, what molecular toxicology tries to do is to understand the mechanisms of why these things are toxic, how they exert their effects. Firstly, because in the case of, for example, a drug, by understanding why it is toxic, you can look for antidotes to that toxicity and treatments, and secondly, you can hope that the next generation of drugs which will follow it you can design that toxicity out. So it’s really quite important for human health, for carcinogens, whether in food products or things like that, understanding whether the carcinogenicity is caused by interactions with DNA or by some other mechanisms. That’s important to know as well, so that’s roughly what molecular toxicology is.

Toxicology has been practiced from the middle of the last century at the very least, now we have a lot more of very useful analytical tools, so we have the science of the genome – genomics – we have the science of proteomics and proteins, we are also looking at metabolites and things like that. We can look not just for the mechanisms of toxicity, but also are there biomarkers appearing in the blood that we can measure to see if toxicity is happening, and to see if we are able to treat that toxicity if the treatments we’re using to counter toxicity are actually working.

Pharmacologist David Nichols on the psychedelics, hallucinogen persisting perceptual disorder, and therapeutic effects of LSD
It’s fundamentally quite important to save a life. Your understanding of why things are toxic, understanding how you can prevent that toxicity to make things better. Perhaps, a very good example of how molecular toxicology has worked in the last 40-50 years or so is to look at the drug paracetamol. Or acetaminophen, which, as it happens, is a very safe drug if used properly in therapeutically recommended doses, but an overdose can be quite toxic and it leads to liver failure. You can tell a drug is toxic by the fact that it causes liver failure, but understanding why that liver failure happened is more complicated. When you ingest too much paracetamol you overwhelm the body’s defenses, you go into a toxic metabolic pathway where the drug is converted by enzymes in your body into a reactive species which goes around attacking cellular components – proteins, things like that. You can defend against that with the molecules that are in the cells, but eventually these become depleted.

Unless you treat this problem, when somebody presents in the hospital saying they’ve overdosed with paracetamol they will die or they’ll need a liver transplant, neither which is an acceptable conclusion from taking just too much of a therapeutic drug. Now what we have discovered is that the mechanism of toxicity involves depletion of cellular glutathione. If you give a compound called an acetylcysteine, even if someone has taken a large dose, you can usually bring them back and save their livers and you don’t need a transplant.

The other thing that we’ve learned is that if you limit the dose of the drug by making it difficult to get large amounts so that people can’t overdose, you can avoid the toxicity in that way, so understanding the mechanism and seeing how it’s related to cellular glutathione and cellular damage caused by these reactive metabolites really has helped us to come up with antidotes and methods for treating this type of toxicity. In classical toxicity testing we test drugs like paracetamol in animal species and these provide reasonable predictors of whether something is going to be toxic in humans. But they’re not perfect. You only really find out whether a drug is toxic to a large population when you start treating a large population.

Paracetamol is quite a nice drug in terms of the predictability of its toxicity, because it’s what we call a dose-dependent toxin, so if you take enough you will lose your liver. Other drugs which are more complicated, idiosyncratic toxins, which means that it is much more difficult to predict who will get toxicity, because it depends on a number of factors that are individual to those subjects. This type of toxicity is very difficult to pick up in animal populations or even in early clinical trials because we define an idiosyncratic toxin as something that perhaps causes an adverse drug reaction in one in 10,000 people. You don’t get to dose 10,000 people in drug discovery and drug development, maybe 2 or 3 thousand, and if you get one adverse drug reaction in three thousand, you won’t know whether that is going to translate to a population or whether that individual just happened to be ill to start off with.

So Idiosyncratic toxins are much more problematic. You can monitor for this sort of toxicity by looking for raised liver enzymes in blood. Most drugs that are withdrawn after they have been launched are hepatotoxins. There are well-known liver enzymes which appear in the blood if you’re getting liver toxicity and these are looked for in the early clinical trials and in the late-stage clinical trials, and you hope to pick up any paracetamol-like toxins very quickly, and, you know, quite a lot of drugs fail late in development, because they start showing this sort of toxic in long-term dosing at therapeutic levels.

Culture of Medicine professor, David S. Jones, on the age of genomics, alternative ways to help society, and the use of current knowledge
The idiosyncratic ones are still a problem. Now, I am confident that although they are idiosyncratic at the moment, that’s because we don’t understand the molecular toxicology behind them. If something causes a toxicity, there has to be a reason. It’s only idiosyncratic until you know what that reason is, and then you can hope to find predictive markers that will individualize treatment to that patient.

One of the things that we hope to do by understanding toxicity, especially the idiosyncratic toxicity, is to be able to walk up to an individual and using the knowledge we have of that individual say “this drug is for you and it’s fine. It will make you feel better, it will treat your disease, and the chances of you getting toxicity as a result of this drug are really low” while for someone else we might say “well actually, your metabolism is likely to produce a bad effect of this drugs so rather than this drug we will give you this one”. One of the problems (Paracelsus defined this in the 16th century) is that everything is toxic, it’s just a matter of the dose. So you can die if you drink too much water, this is an example of toxicity which we all know. The aim though is that people should not suffer adverse reactions that are unacceptable for their condition in terms of their disease.

Molecular toxicology is the study of the toxicity. What we’re doing when we’re studying molecular toxicity is we try to find a molecular basis for the toxicity of the drug, or the food substance, or environmental pollutants, which might be genetic – it might damage genes, these are genotoxins. They’re really difficult to sort out because there’s no safe level for genotoxins, or there may be paracetamol-like toxins, where actually if you are careful with the dose they’re perfectly safe. Or they may be more idiosyncratic toxins where you look at the risk benefit, if you stand a one in 10,000 chance of getting a toxicity and you have a disease that will kill you in three months, most people would take the one if 10,000 chance to take the drug. No one wants to get a liver failure because they have a headache or die because they have a headache. It was a risk and a balance here.

There’s no such thing as a perfectly safe drug. You don’t take drugs unless you are ill so it’s the risk-benefit. What the molecular toxicology tries to do is to understand why compounds are toxic. Because if you know why a compound is toxic that can highlight what might be a vulnerable population. Probably the main questions in molecular toxicology at the moment are how do we solve the problem of idiosyncratic toxicity, drugs like paracetamol and related things. We have a pretty good idea how they work and they’re predictable. We need to be smarter at picking genotoxins up, and there are a whole variety of methods coming along based on the new -omics technologies that will help us with that.

Microbiologist Kim Lewis on multidrug resistance pump, prodrugs, and Metronidazole
The other really beneficial thing would be when we can really deliver a personalized medicine strategy for individuals. We are talking about drugs, but it applies to foods as well, sort of personalized feeding strategy that will minimize the toxic effects of dietary components. If we can pick out from a crowd the one person in 10,000 who will suffer from taking that drug, the remaining 9999 can continue to benefit from it. If we take a precautionary principle and say one person in 10.000 is too much, then 9999 lose that benefit. We throw away drugs that otherwise would be beneficial, and for society to be able to provide that level of benefit would be huge.
By learning more about populations and more about individuals within those populations we will be able to deliver something like true personalized medicine rather than “you look like that sort of person, you’ll be all right on this drug, why don’t you take it for 6 months and we will see?”. One of the things we’re trying to do in molecular toxicology is to come up with better models. You can’t throw human beings to find out whether something is toxic. It’s unethical, so we need better in vivo and in vitro models. The future lies in making models that truly are predictive and predict for humans rather than for animals.

Professor, Faculty of Medicine, Department of Surgery & Cancer, Imperial College London; Chair in Drug Metabolism and Molecular Toxicology, Imperial College London
Did you like it? Share it with your friends!
    Published items
    To be published soon

    Most viewed

  • 1
    Tim Reynolds
  • 2
    Peter Kivisto
  • 3
    Simon White
  • 4
    Greg Towers
  • 5
    Sophie Scott
  • 6
    Tristram Wyatt
  • 7
    David Rubinsztein
  • 8
    Joanna D. Haigh
  • 9
    Michael Thomas
  • New