One does not have to look far today to encounter concerns around our increasing exposure to ‘chemicals’. However, every substance is in fact a chemical or a mixture of chemicals, including all our food, drink and our own bodies. What people usually mean is that they want to avoid any ‘synthetic’ chemicals (those that don’t exist in nature), as they consider ‘natural’ chemicals safer. There appears to be some logic behind this preference, as certainly some classes of artificial chemicals, such as polychlorinated biphenyls (PCBs) and anthropocentric contaminants such as polycyclic aromatic (PAHs), can lead to long term environmental and human health problems.
However, overall the evidence is not clear cut, as there are plenty of harmful ‘natural’ compounds out there. In order to look at the dangers of chemicals in detail, we need to look at how toxicity is determined and human health risk assessments data derived.
It is not always obvious as to whether a compound should be classed as ‘natural’. Vitamins occur in nature, but those found in vitamin pills are typically produced in chemical factories, the ultimate precursor often being crude oil. The synthetic vitamins are the same compounds as vitamins that occur in nature, so their source makes no difference to their toxicity.
However, we don’t have a 100% complete list of substances that do occur naturally. It might be that a substance is produced in a laboratory or factory, but is later discovered arising naturally in some organism. An example would be methyl iodide, which was first made in the laboratory back in the 19th century, but was only detected in the oceans and atmosphere in the 1970s. The compound is produced by various organisms, such as marine algae and seaweed. Again, the compound is considered a potential carcinogen, but is used as a fumigant and soil-sterilizing agent in some countries (not the UK). No doubt there are other laboratory and industrial compounds that will be discovered to have natural sources in the future. Of course, such a discovery will not make them any safer. Human and ecological risk assessments have to take account of the actual hazards of chemicals, regardless of whether or not they are found in nature.
The well-known toxicologist Bruce Ames, in a 1990 paper, argued that there was little overall difference between ‘natural’ and ‘synthetic’ chemicals with regard to long-term effects. A similar proportion of both sets of chemicals exhibited adverse chronic effects at very high doses. Thus, taking the broad picture, there is no need to fear ‘synthetic’ chemicals in general. However, this does not mean that bans on particular synthetic chemicals are not justified. DDT and similar polychlorinated insecticides caused great harm to the populations of birds of prey. Since they were phased out in the 1970s and 80s, the populations have recovered.
‘Exposure’ is whenever a receptor comes into contact with a chemical, whether natural or synthetic, either by eating, drinking, breathing it in, via dermal absorption or inhalation. Toxicologists study the biological effects of exposure on an organism (or tissue cells), known as the response. There are two types of toxicity: acute toxicity (will exposure to this substance do immediate harm?) and chronic toxicity (will long-term exposure cause harm?). 'Dose' can assume a number of meanings, but for the purposes of this post it will be defined as the concentration received by the subject.
Acute toxicity is traditionally determined through toxicological risk assessment by administering relatively high doses of a substance to laboratory animals (often rats or mice). Typically, this will be by feeding (oral toxicity), but may be by breathing in a volatile substance, or skin exposure, if data is needed on these modes of absorption. The response (i.e. the biological changes caused by a substance) of a group of animals is then measured. It’s useful to know the highest level at which no effect is noted (NOAEL, no observed adverse effect level). The relationship between dose and response can be investigated. Typically, the higher the dose, the greater the response.
The amount of a substance producing a lethal response in 50 percent of test subjects is designated as the LD50. The LD50 is normally given in mg of substance per kg of test animal (since large animals require more of a chemical to kill them). The smaller the LD50 the greater the toxicity.
Although such figures give useful information on toxicity, there are a small proportion of chemicals where the human toxicity is significantly different to that of animals, so occasionally LD50 results are not a good guide to human toxicity. Nowadays, there is much work being carried out on replacing animal tests with other alternatives. However, since they have been previously determined for a vast range of substances, it’s still useful to examine the LD50 response to various compounds to give an idea of their relative acute toxicity.
An LD50 test is a particular type bioassay used to determine biological activity (in this case chemical toxicity). Another example would be examining the effect of a substance on a tissue culture of human or animal cells. Here the use of live animals is avoided, and there may exist the possibility of using human cells. However, effects only seen with a whole animal may be missed. For this reason human health risk assessments are often based on data from animal studies.
Once we start to look at numbers for toxicity, rather than just relying on gut instinct, we find plenty of natural substances with high toxicity. Some of the most toxic substances turn out to be naturally occurring compounds. Living creatures produce many toxins, often as defence compounds. They have had millions of years of evolution to perfect their chemical defences. An example is Aflatoxin B1, which is a toxin produced by moulds found on peanuts; it has an LD50 of only 5mg/kg (oral, rat, OSHA figure). Strychnine, from tree seeds, and the botulinum toxin, produced by bacteria, are other very toxic naturally occurring compounds, although the latter is used in very small doses for ‘Botox’ treatment. Many synthetic compounds have fairly high LD50 values. In contrast, the controversial synthetic herbicide compound glyphosate has an LD50 of 5,600mg/kg (oral, rat).
It is important to note that almost anything can be toxic given a high enough dose. Vitamins are essential for life, but some can be toxic if too much is taken. Vitamin A has an LD50 value of 2,000mg/kg (oral, rat), so has significant toxicity at very high doses. A human health risk assessment for vitamin A would conclude that harm can arise from either eating too much or too little.
Long term exposure to a substance can cause harmful effects, even when a substance has relatively low acute toxicity (high LD50). Chronic toxicity can be associated with serious conditions, such as cancer and adverse effects on reproduction. Chronic toxicity is generally established by long-term studies on rodents (typically 1-2 years). Data is occasionally available for humans from epidemiological studies or where long-term occupational exposure has resulted in disease. Asbestos (another naturally-occurring substance) is a well-known example of a substance with low acute toxicity, but serious long-term effects, including the risk of cancer.
Chronic toxicity testing is often carried out at very high levels, when the acute toxicity allows. Adverse effects, including cancer, found at these levels may have little bearing on the risk in real life situations, where the exposure is likely to be much lower. Compounds with high acute toxicity (low LD50) can’t be tested at high levels in chronic toxicity testing, as the animals would all die first. The results of chronic toxicity testing can vary, as different strains of animal, let alone different species, are susceptible to adverse effects to varying degrees. Thus, despite its low acute toxicity (high LD50), controversy exists about the long-term effects of the glyphosate herbicide, which may or may not be carcinogenic to animals at high doses. Animal testing has its limitations, particularly when chronic toxicological effects are relatively mild. Sometimes the statistics give ‘borderline’ results, which people tend to interpret according to their own prejudices.
Human and ecological risk assessments have to be carried out on a case by case basis, to determine whether the levels exposure to a particular substance causes harm. It is important to estimate the actual exposure under real life conditions, which may require much detailed study.
Thus a very hazardous substance may be low risk if it is safely contained, such that no exposure occurs. Conversely, chronic exposure to a relatively low hazard substance, such as alcohol, can lead to a high risk of human health damage over the years.
It is important to follow approved guidelines for ecological risk assessment (also known as environmental risk assessment). DEFRA produced a revised version of its ‘Guidelines for environmental risk assessment and management: Green leaves three’ in 2011. The guidelines suggest that ‘SPR assessments’ are carried out to determine likely exposure. Here S stands for the source of a hazardous substance, P the likely pathways it could travel in the environment, and R the possible receptors, which could include humans, animals, plants, or whole ecosystems. All these factors need to be considered in order to obtain a useful ecological risk assessment. It is important to reflect on all possible pathways when carrying out such assessments. Slurry from cattle may be fine spread on a field, but if washed into a stream by unusually heavy rain may cause great harm to wildlife, due to ‘natural’ chemicals, such as nitrates and ammonia, in the slurry.
Thus, environmental toxicology consultants carry out a great deal of detailed work in order to assess the risks of any particular chemical. There are no short cuts, and every substance needs to be examined on its merits using proper human or ecological risk assessment methods, regardless of whether or not it happens to occur in nature.
Ames, B.N. et al, ‘Nature's chemicals and synthetic chemicals: comparative toxicology’, Proc Natl Acad Sci U S A., 1990, vol. 87(19), pp. 7782–7786. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC54832/
OSHA website, ‘Aflatoxins’. https://www.osha.gov/dts/chemicalsampling/data/CH_217356.html
Cornell University Website, ‘The Dose Makes the Poison’. http://ei.cornell.edu/teacher/pdf/ATR/ATR_Chapter1_X.pdf