Saving their skin: Research to reduce animal testing

By Sophie Regnault
Royal Veterinary College, UK

This summary was highly commended by the judges for Access to Understanding 2015

Picture 1 of 1

Animal testing is expensive, difficult and, for many people, straddles an ethically uncomfortable line. A central tenet of science and medicine over the past half century has been the 3R’s (Reducing the number of animals in experiments, Refinement of experiments to do the least harm, and ultimately Replacing animal testing whenever possible) and European law reflects these morals, restricting animal testing and especially cosmetics testing.

In vitro experiments, where potentially toxic substances are tested on cells in a petri dish instead of whole animals, seem like they should be an ideal alternative. However traditional in vitro experimentation has some unfortunate limitations. The cells used often have mutations or other adaptations which allow them to thrive in the relatively unnatural environment of a test tube, but means they might react differently to the ‘normal’ cells in the body of an animal or person. The cells also sit in thin sheets on the bottom of a dish or float around in solutions rather than forming 3D organs, which again can change how they react to substances. Skin, for example, is a complex organ composed of different cell types organised into layers with an outer protective covering, but a thin sheet of identical cells in a dish doesn’t reflect this structure. Because of these reasons, in vitro tests tend to exaggerate the toxic effect of substances. Dump alcohol onto a petri dish of hapless cells and you might kill them, but spilling a drink on your arm doesn’t require a skin graft.

Scientists are trying to address the shortcomings of traditional in vitro testing. One method being developed is a more realistic living 3D skin replica complete with outer protective layer, made using freshly donated human skin cells. Because it more accurately reflects the structure and function of normal skin, it is hoped that this model is better at realistically mimicking the body’s reaction to potentially toxic substances.

One of the most important parts of developing new scientific methods such as this is validation – checking that the method and results are repeatable (and not just a lucky accident), and that the model actually does closely mimic the body’s reaction. As well as backing up previous research, validation studies also explore the characteristics of the new model by finding out exactly when it behaves realistically, where its weaknesses lie, and how it might be improved. Used correctly, these validation studies can help open up whole new areas in medicine and research.

One such study is that performed by Kate Chapman and her colleagues at Swansea University. They tested the new skin model’s response to known safe and toxic chemicals, and are the first to study its response to low doses (the most common level of chemical exposure). More excitingly, they are also the first to try measuring genotoxicity (a specific type of toxicity) automatically in this model.

Genotoxicity is the ability of a substance to cause DNA damage. A more genotoxic chemical will cause more DNA damage, which can lead to mutations and cancer. When a cell divides normally, the DNA is split between two nuclei (one for each new cell). A micronucleus – an extra small nucleus – can also form around any broken pieces of DNA. One way of gauging the amount of genetic damage to a cell is to count how many micronuclei the cell produces in response to chemical exposure. More micronuclei signal a more genotoxic substance.

Counting micronuclei manually is very time-consuming. It involves examining hundreds of cells under a microscope, identifying potential micronuclei, and deciding if they count (whilst keeping track of where you’ve already looked!). It can be subjective, with different counts for different observers, and prone to human error. Chapman and her colleagues trialled an automatic micronuclei counting system for samples from the new skin model, showing that the machine generally agrees with counts made by a human (and may even be better at finding micronuclei, though these results are less clear).

With the automated method, the authors were able to measure DNA damage in a large number of samples, fast and reliably, across a range of doses. Because a machine can count faster and more consistently than a person, it means that large number of samples can be evaluated. In turn, larger sample sizes allow more powerful statistical tests to be used to detect patterns in the data. In trialling their automated method, the authors have contributed to the total knowledge about a promising new in vitro model. They have made it easier for other scientists to work with it (by identifying some of the model’s quirks, like behaving better with acetone – paint thinner – rather than water) and make discoveries of their own. Ultimately, these authors and groups like them are bringing us a step closer to replacing animals in medicine and research.

This article describes the research published in:

Automation and validation of micronucleus detection in the 3D EpiDerm™ human reconstructed skin assay and correlation with 2D dose responses
E. Chapman, A. D. Thomas, J. W. Wills, S. Pfuhler, S. H. Doak, G. J. Jenkins Mutagenesis (2014) 29(3) 165-175

This article was selected for inclusion in the competition by the National Centre for the Replacement, Refinement and Reduction of Animals in Research.



Please give us your feedback.
Your feedback will not be visible to other users visiting this page.