Frequently asked questions
How can I use more than one inbred strain without increasing the total number of animals which I use?
Answer: If you normally have, say, eight animals in a treatment group you could have four of one strain and four of another or even two animals of each of four strains. This will not normally reduce the statistical power of the comparison between the control and treated groups. Indeed, in comparison with using eight outbred animals in each group it should lead to increased statistical power. The statistical analysis is a bit more complicated, you will need to use a two-way analysis of variance assuming you have measurement data, but it will also tell you whether the strains all respond in the same way. Any differences in response will be attributable to genetic factors, assuming the strains are of comparable age and come from the same environment.
Don’t most inbred strains get cancer?
Answer: Many inbred strains of mice such as AKR and C3H were developed for cancer research so were selectively bred for a high incidence of leukaemia or mammary cancer. Similarly, some strains have been developed for other characteristics such as high blood pressure or diabetes. However, many strains were produced by brother x sister mating without selection for any specific trait, and these are useful for general research. Of course each strain has its own set of inherited characteristics, just like each human, so care needs to be taken in selecting a strain suitable for each specific research project. In some cases it may be worth doing a small multi-strain study to find a strain suitable for a particular project.
Humans are genetically heterogeneous so aren’t genetically heterogeneous mice and rats a better model of humans?
Answer: Models do not need to be like the thing being modelled in every respect. We use mice and rats in research because they are small, prolific and economical to maintain (all differences from humans), while resembling humans in anatomy, physiology and biochemistry. Most people are studying some aspect of physiology, biochemistry, immunology, behaviour, neurology, etc. and do not require that the animals are a good genetic model of humans. The advantage of using isogenic strains is that experiments using them will be more powerful or can be done using fewer animals than if outbred animals are used. People using outbred stocks are rarely able to detect genetic variation in response, but those using isogenic strains can easily do so by using more than one strain without increasing the total number of animals used (see Multi-strain experiments)
My job is to screen chemicals for their toxic effects. What if the inbred strain I chose were to be genetically resistant to the chemical?
Answer: Genetically-determined resistance to a toxic chemical can pose serious problems. However, it is not confined to isogenic strains. Outbred stocks may also be resistant to some test chemicals. For example, the Han:Wistar outbred stock is extremely resistant to TCDD (Dioxin), and some Sprague-Dawley stocks are resistant to diethylstilboestrol (DES). Toxicological screening can best be done using several isogenic strains without increasing the total number of animals which are used. Such experiments will usually be statistically more powerful than those using a single outbred stock and have the added advantage that they show whether the response is under genetic control. Where susceptible and resistant strains are identified, further studies of the reason for the differences provide an additional way of studying toxic mechanisms. See Milti-strain experiments web page.
Is the use of outbred stocks ever justified?
Answer: Yes, such stocks have been used effectively in genetical research to map genes, and as a base population for selective breeding to develop new or improved animal models of human conditions. Also, in species other than mice or rats there is usually no practical alternative. Outbred stocks may also be used for within-animal experiments where differences between individuals do not increase experimental error. Occasionally an outbred stock will be found which has some useful condition not found in any inbred strain. In this case further research may be justified. However, in many such cases a good strategy is to develop some inbred strains from it. This is what has been done, for example, with the outbred BB Wistar rat which was found to get type I diabetes.
Is there any type of research for which inbred strains are unsuitable?
Answer:They are suitable for virtually all research, but they do have some limitations. They often breed poorly, so inbred strains may not be suitable for some reproductive studies such as teratology screening. F1 hybrids breed well so may be useful for such studies, but their offspring will be genetically segregating, so may be quite variable. Inbred strains are not suitable for selection experiments aimed at altering strains characteristics, because their characteristics are fixed and will not respond to selection. There are also some genetic studies, such as genetic mapping, where outbred stocks have been useful. Care should always be taken to choose a strain which is suitable for the planned research project.
Surely, the variability of toxicity obtained in less well defined animals is a strength in itself, not a problem, when trying to predict safety margin in the non-isogenic human population. ?
Answer: This is a common fallacy. Although outbred stocks are a bit more variable than inbred strains there are many well documented examples of outbred stocks which are resistant to toxic agents. For example, the outbred Han:Wistar is 1000 times more resistant (LD50) to dioxin compared with other strains (Tuomisto et al Toxicol. Appl. Pharmacol. 155:71-81, 1999) and Sprague-Dawley rats are resistant to diethylstilbestrol at levels which cause over 70% of tumours in other strains (Shellabarger, et al1978. JNCI 61:1505-1508.).Five stocks/strains treated with 3,2′-dimethyl-4-aminobiphenyl varied enormously in response. While some strains developed a high percentage of prostate tumours, the outbred Wistar stock was totally resistant and the CD (an outbred Sprague-Dawley) was nearly resistant (Shirai et al 1990, Carcinogenesis 11:793.) .
Thus, use of a single outbred stock may give entirely misleading results.
The only way to ensure a reasonable representation of susceptibility genotypes is to use small numbers of several isogenic strains in a factorial experimental design. This can be done without increasing the total number of animals (see Multi-strain Expts.) . It will result in much more powerful experiments, better able to detect toxic responses, and will also give an estimate of the extent of genetic variation in response. In the long-run it will also enable toxicologists to identify susceptibility genes in animals which may have homologs in humans. Also, comparison of susceptible and resistant strains can provide additional information on toxic mechanisms which may be important for extrapolating to humans.