Before I explain how scientists test the toxicity of a chemical, it is important to reiterate a concept: the dose makes the poison. For the most, there is a relationship between the quantity of chemical an organism is exposed to and the nature and degree of harmful effects. This is called a dose-response relationship (see Figure 2).
Is there a difference between dose and concentration?
Yes, there is. To best understand the difference, consider water. If you put 5 milligrams of a chemical in one liter of water, the concentration is 5 milligrams/liter. If you consume 0.5 liters of this water, the dose you get is 2.5 milligrams (5mg/L * 0.5L).
In the rest of this post, dose can mean both dose and concentration.
The older kind of toxicity testing
Most often, the measure of toxicity is death. Though scientists also look at how growth, development, reproduction and behavior are affected. Most experiments are carried out on lab animals.
Scientists look at how animals are affected by short and long term exposure to a chemical. For short exposures, they calculate the dose that either kills or causes any other effect in 50% of the treated population. For long exposures, usually the lowest dose that causes a toxic effect or the highest dose that doesn’t cause a toxic effect are obtained. (Ideally though, it is better to generate a dose-response relationship.)
The newer kind of toxicity testing
While the above methods are still largely used, alternative methods that reduce vertebrate testing and are more humane are gaining ground.
Doing toxicity testing on non-vertebrates and/or vertebrate cell cultures (cells isolated from vertebrates and grown artificially) and extrapolating the results to whole vertebrates in one such way. However, it is important to know how different species respond and how cells and tissues differ from whole organisms.
Instead of giving animals high doses of a chemical and killing them, scientists nowadays give low doses and sample animal blood or other tissues and check for levels of enzymes and proteins that could be affected. Also increasingly, computer models are being used to predict the toxicity of a chemical by studying the relationship between a chemical’s structure and its biological activity. Mathematical models to predict effects of chemicals at or above population levels are also being used.
There are millions of species and it is possible to test only a minuscule number. Also, some species can be difficult to obtain, rear, or study. Hence, surrogate species for toxicity testing have been identified for humans, birds, plants, fish, terrestrial invertebrates etc. These surrogate species are tested and uncertainty factors are added to extrapolate the results to the original species. For example, the surrogate species for humans are rodents. If the highest dose that doesn’t cause a toxic effect in rodents is 100 mg of the chemical/kg body weight, then the maximum allowable safe dose for humans is 0.1 mg of the chemical/kg body weight. An uncertainty factor of 1000 (10 x 10 x 10) was added to account for differences between species (rodents and humans), within species (between humans) and for the increased sensitivity of the children and the elderly.
Principles of Ecotoxicology (4th Edition) by C.H. Walker, R.M. Sibly, S.P. Hopkin and D.B. Peakall
U.S. Environmental Protection Agency website (https://www.epa.gov/)