Science in Society Archive

Super-Toxic Cocktails

Concentrations of pesticides insufficient to cause harm individually become highly toxic in combinations that occur in our environment. Current testing regimes are failing to cope with such lethal cocktails Prof. Peter Saunders

This article has been submitted to the UK Food Standards Agency and European Commission’s Panel on Food Additives, Flavourings, Processing Aids and Food Contact Materials

Synthetic chemicals are all around us, in the air we breathe, in the water we drink, in the food we eat, in the objects we touch. Many of them are hazardous, some more than others, and it is not easy to know how safe any of them is. Nowadays chemicals are tested when they are first introduced, but while this may pick up most of the acute effects that happen rapidly, it is likely to miss chronic effects that take longer to appear, such as the harm done by smoking. Chemicals that have been used for a long time will not have been subjected to the same sort of scrutiny as more recent ones, and we cannot assume that they are safe just because they are familiar. Lead was in common use for centuries before it was recognised to be dangerous.

What makes the problem of safeguarding health and the environment even harder is that testing is almost always carried out one chemical at a time. In real life, we encounter them in combinations, and it is well known when two chemicals act together, what happens may be quite different from the sum of their separate effects (see Box). In particular, a chemical that has been judged safe when tested on its own may be toxic in combination with certain others [1].

Kinds of Interaction
Synergism The combined effect of two or more chemicals is > predicted by dose additionAlcohol and carbon tetrachloride acting on the liver
AntagonismThe combined effect of two or more chemicals is < predicted by dose addition

Arsenic and lead acting on the kidneys and the blood

PotentiationA chemical that is not itself toxic increases the effect of one that isIsopropynol makes carbon tetra-chloride more toxic to the liver
InhibitionA non-toxic dose of one chemical decreases the toxic effect of anotherSome antidotes act this way

Unfortunately, it is simply not possible to test all possible combinations of chemicals. The EU’s scheme to register, evaluate and authorise the use of chemicals (REACH) proposes to prioritise 30 000 chemicals. That is only a small fraction of those in use; and yet it is estimated to cost €2.3 billion over the 11 years to do the testing; though it is also estimated that the potential health benefits over 30 years will be roughly €50 billion [2].

If we were to set out to test all pairs of chemicals, or all combinations of three, or four, the number of trials required would rapidly become astronomical. But that doesn’t mean we can ignore the issue. On the contrary, we have to do the best we can to understand the additional hazards that can arise from mixtures. And even if we believe that a particular chemical is safe at the dosage or concentration that we have specified, we still have to ask whether it might be toxic when it acts together with others.

The effect of pesticides on salmon

Salmon stocks in the American Pacific Northwest have been in decline for many years. There are a number of reasons for this, and one of them is almost certainly the pesticides frequently detected in the rivers salmon live in.

Two classes of pesticide in common use are organophosphates and N-methyl carbonates.  Both inhibit the enzyme acteylcholinesterase (AChE) and so interfere with neurotransmission in fish, and in humans as well. Combinations of them are often found in the streams in which the salmon live, and the tolerance level for the pesticides has to take this into account.

The US Environment Protection Agency (EPA) recommends a dose-additive approach, i.e. they assume that the toxicity of the mixture is just the sum of the toxicities of the components. A group of researchers based at the Northwest Fisheries Center in Seattle and at the University of Washington have tested this assumption and found that while it is true in vitro (in the test tube),  it does not always hold in vivo, i.e. in living salmon [3, 4].

They measured the effects of pairings of five commonly used pesticides: the organophosphates diazinon, malathion and chlopyrifos and the carbamates carbaryl and carbofruna. For some pairings of insecticides, the effect of two together is considerably greater than you would predict by adding the two separate effects together. They also found that several combinations of organophosphates were lethal at concentrations that were sublethal in single-chemical trials.

The reason this was not observed in vitro is that when you do an experiment in the lab, you generally use only the chemicals you are studying, in this case the insecticides and AChE, together with any other reagents you need to make the system work. Organisms are far more complicated.  Anything you do can affect many more reactions than the one you are looking at, and this can have all sorts of consequences, including speeding up or slowing down the reaction you are modelling.

What exactly is happening in the salmon is not yet fully understood, but it is known that as well as inhibiting AChE, pesticides reduce the liver activity of another group of enzymes known as carboxylesterases (CaEs).  There is evidence that CaEs play an important role in the detoxification of many pesticides, and in particular, may prevent or delay the interaction between insecticides and AChE. If this indeed the cause of the synergy observed in the live salmon, that would explain why the effect was not observed in the in vitro experiments.

The researchers concluded that single chemical risk assessments are likely to underestimate the impacts of insecticides, and that mixtures of pesticides that have been commonly reported in salmon habitats may be posing a more important challenge for species recovery than previously anticipated. What their work also suggests is that similar results may apply in other situations and for other species, including humans.

Food colourings

As reported earlier [5] (Food Colouring Confirmed Bad for Children. Food Standards Agency Refuses to Act, SiS 36), it has been suspected for a long time that food colourings were at least partly responsible for the increase in attention deficit hyperactivity disorder (ADHD) in children since the War. Children will typically consume several colourings in the course of a day and there was evidence to suggest that combinations of them were having an effect on behaviour.

The evidence was not conclusive, however; so the UK Food Standards Agency (FSA) commissioned a group in Southampton University to carry out rigorous double blind trials. When the Southampton group found that there were significant effects, the FSA seemed very reluctant to accept the findings of the research that it had paid for.

First, they refused to do anything until the results had appeared in a peer reviewed journal. As far as we know, that accomplished nothing except to introduce a six month delay before anything was done.  When the results were finally published [6], the FSA placed its revised advice in an inconspicuous part of its website, stressed that there were many other factors in ADHD, and claimed, falsely, that the results applied only to children who already had a tendency to ADHD, when the actual results were for a typical group of children. And it was left to parents who were still worried to look at the labels on candies and decide for themselves [7].

Things have moved on. The FSA has now tacitly accepted the recommendation of the Southampton group that as the additives in question have no function other than cosmetic, it would be better not to use them in foods, especially those consumed by children. The colourings are still not banned, but the FSA has recommended that manufacturers should stop using them, and it points out with some satisfaction that many of them are doing so, though not all of them have. In particular the soft drinks manufacturers have rejected the FSA’s recommendation. As AG Barr, the manufacturer of IRN-BRU, pointed out when they announced their decision, all their ingredients remain in accordance with the relevant UK and EU rules and legislation [8], which is, of course, precisely why the rules and legislation should be changed.

The European Commission’s view

It is the European Commission, not the UK government that decides which additives are permitted in food sold within the UK. The Commission asked its Panel on Food Additives, Flavourings, Processing Aids and Food Contact Materials (AFC) to consider the Southampton research, and on the basis of their report, decided not to change the acceptable daily intake (ADI) for the colourings.

The reasons are depressing but not surprising, considering their source [9]. They state that McCann et al provide “limited evidence” that the mixtures of colourings had an effect on activity and attention in children; it seems that in their view, full proof is needed before something can be banned. They point out that no “biologically plausible mechanism for induction of behavioural effects from consumption of food additives” has been proposed: this is of course the same argument that the tobacco manufacturers used for years in the face of overwhelming epidemiological evidence against them [10]. And much the same wording is

used by Monsanto to dismiss statistically significant effects of genetically modified food in feeding trials that were subsequently reanalysed and faulted by independent scientists [11] (GM Maize MON 863 Toxic, SiS 34).

The AFC also criticises the Southampton study because as it worked with mixtures it was unable to “ascribe the observe effects to any of the individual components.”  The members of the Panel clearly missed the point: what McCann et al were studying were effects that are observed when the colourings are consumed in combination, as they are in real life. One might as well complain that no one has shown whether it is carbon, hydrogen or oxygen that makes us drunk when we consume too much alcohol.

On its website, the UK FSA assures the British public that food additives are only allowed to be used in Europe if experts decide that they are necessary and safe. The two conditions have to be considered together, and indeed this is why both the Southampton group and also (if only implicitly) the FSA recommend the immediate withdrawal of the colourings but not of the preservative sodium benzoate.

The AFC, however, says nothing whatsoever in its report about whether the additives in question are necessary. It, and the European Commission, are thus applying a strong form of the anti-precautionary principle: as long as there is not conclusive evidence that the colourings are harmful, then they must be permitted, even if they serve no useful purpose [12] (Use and Abuse of the Precautionary Principle, ISIS Report)

Fortunately, the European Parliament is in the process of updating the EU rules for authorising food additives, flavourings and enzymes, and this gave it the opportunity to intervene. It has adopted a legislative package on four draft regulations, and included the statement: that a food additive must be authorised only if it safe to use, if there is a technological need for its use, if its use does not mislead the customer, and if it has advantages and benefits to customers. While the colourings in the Southampton study are not to be banned, foods containing them must be labelled not just with the relevant E number but also with the words [13] “may have an adverse effect on activity and attention in children.”  

Conclusion and recommendations

Chemicals are important in our daily lives but they can also be harmful. It is not always easy to know whether a chemical is hazardous or, if it is, at what concentration, which is why the EU has proposed the REACH scheme to test 30 000 chemicals that have been in use since before the current testing regimes were introduced.

It is even more difficult to know the effects of mixtures of chemicals, and it is clearly not possible to test all the combinations we may encounter. We need to develop a set of rules and procedures aimed at doing our best to reduce the hazard even if we cannot hope to eliminate it completely. As a start, we recommend the following:

· We must develop criteria that specify which combinations must be tested. At the very least we should follow the lead of the US Food Quality Protection Act, which directs the EPA to consider together the effects of different pesticides that act in similar ways. Chemicals with the same mechanism of action are not the only problem, but they are the most obvious.

· In particular, because the effect has been clearly demonstrated in pesticides, because mixtures of pesticides are so common in the environment, and because humans too can be affected by pesticides, the toxicity of those mixtures should be studied thoroughly, and for several generations. It is now becoming increasingly clear that many xenobiotics have epigenetic effects that straddle multiple generations after exposure [14] (see Epigenetic Toxicology, SiS 41).

· In vitro experiments are not adequate for testing for synergy, because for all but simple reactions the test tube is not an adequate model of an organism.

· Where epidemiological or other evidence suggests that some chemical has harmful effects, tests on that chemical alone are not sufficient to rule out the possibility. We should also look for possible synergistic effects.

· The difficulties in establishing or ruling out synergistic effects make it even harder to be confident that a chemical is not toxic. That is all the more reason for adopting a precautionary approach and banning chemicals that serve no useful function.

No amount of testing can ever ensure that no one ever suffers from the effect of a toxic chemical, still less of  mixtures of chemicals. That is not, however, an excuse for doing nothing. On the contrary, we must as a matter of urgency put in place measures that will reduce the danger as much as we can. Even simply recognising that mixtures can be more dangerous than would be predicted from a study of the individual components will help improve health and safety.

Article first published 27/04/09


  1. Basic Concepts for Mixture Risk Assessment and references given there. Argonne National Laboratory (2005). Human Health Fact Sheet.
  2. European Commission Environment Directorate General (2007). REACH in Brief.
  3. Scholz, NL, NK Truelove, JS Labenia, DH Baldwin and TK Collier, (2006). Dose-additive inhibition of chinook salmon acetylcholinesterase activity by mixtures of organophosphate and carbanate insecticides. Environmental Toxicology and Chemistry 25, 1200-7.
  4. Laetz, CA, DH Baldwin, TK Collier, V Herbert, JD Stark and NL Scholz, (2009) The synergistic toxicity of pesticide mixtures: Implications for risk assessment and the conservation of endangered Pacific salmon. Environmental Health Perspectives 117 348-53.
  5. Saunders PT (2007) Food colouring confirmed bad for children: Food Standards Agency refuses to act. Science in Society 36, 30-31, 2007.
  6. McCann D, Barrett A, Cooper A, Crumpler D, Dalen L, Grimshaw K, Kitchin E, Lok K, Porteous L, Prince E, Sonuga-Barke E, Warner JO, Stevenson J. Food additives and hyperactive behaviour in 3-year-old and 8/9-year-old children in the community: a randomised, double-blinded, placebo-controlled trial. The Lancet (published online: September 6, 2007, DOI:10.1016/S0140-6736(07)61306-3).
  7. Food Standards Agency (undated) Intolerance to food additives.
  8. Mercer, C. UK: AG Barr resists FSA pressure on artificial colourings. t:
  9. AFC, (2008). Scientific opinion of the Panel on Food Additives, Flavourings, Processing Aids and Food Contact Materials (AFC) on a request from the Commission on the results of the study by McCann et al (2007) on the effect of some colours and sodium benzoate on children’s behaviour. The EFSA Journal 660. 1-5.
  10. On the other hand, when there is evidence of a causal relationship, industry and its allies tend to argue that randomised controlled trials are the only conclusive evidence. See, e.g., D. Healy, Let Them Eat Prozac, New York University Press, New York, 2004, p235.
  11. Ho MW. GM maize MON 853 toxic. Science in Society 34, 26, 2007.
  12. Saunders PT. Use and abuse of the precautionary principle. ISIS submission to US Advisory Committee on International Economic Policy (ACIEP) Biotechnology working Group, 13 July 2000,
  13. European Parliament (2008). Modernising the rules on food additives and labelling of azo dyes. (An “E number” is a code used within the EU to label food additives: the six colourings have E numbers E110, E104, E122, E129, E102 and E124.)
  14. Ho MW. Epigenetic toxicology. Science in Society 41, 12-15, 2009.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a cat?

There are 4 comments on this article so far. Add your comment above.

Rory Short Comment left 28th April 2009 04:04:02
I wonder whether it should not be necessary as an absolute first step for new chemicals that they be shown to have possible benefits for humans before any further work with them is permitted such as testing for any damaging efects as decribed in this article. If a new chemical appears, at this point in time, to have no possible benefits for humans it should be put totally on ice until the possibility of such benefits appears.

kyriaki zannettou Comment left 28th April 2009 21:09:53
what about the dangerous synthetic chemicals in personal care products and cosmetics, where almost there is no any official safety control

Sharon Medway Comment left 6th May 2009 17:05:44
I think the bigger issue is the necessary statement and the mislead the customer statement. For a very long time we would get Dr. Browns soda and I always noticed that it was clear, no matter the flavor. Now it is colored. Is it necessary to make cherry cola red? Does the red coloring mislead the consumer into thinking that the natural cherry creates that color? This is I think one of the main issues. We have these cocktails that no one even disputes. The last time I looked at a bag name brand "corn chips" there were over 30 ingredients listed. Why are all those things in there? Why doesn't anyone ask? Wouldn't it be cheaper for the company to remove those items and just sell us: corn mean, oil, cheese, paprika? There is no rash of world wide concern about paprika poisoning and effecting our health. We have the tools but why wont people use them? Why is it so hard?

Dr. Ilya Sandra Perlingieri Comment left 22nd May 2009 18:06:05
21May 2009 To Dr. Mae-Wan Ho and I-SIS: I thought you would be interested in this: “The American Academy Of Environmental Medicine Calls For Immediate Moratorium On Genetically Modified Foods” See: Also see F. William Engdahl’s article about it at: From: Dr. I. Perlingieri Author: “The Uterine Crisis”