Commentary: Why are there difficulties in controlling genetic variability?

Abstract

More than 20 years ago, K. Gartner published a paper in the journal Laboratory Animals. According to the author: ‘Reduction of genetic variability by using inbred strains and reduction of environmental variability by highly standardized husbandry in laboratory animals did not remarkably reduce the range of random variability in quantitative biological traits.’ As far as I can see, this statement is correct and has certainly survived the test of time. Similar conclusions, although in a different form, were drawn by other researchers. For instance, Fitch and Atchley wrote: ‘Genetic variation at 97 loci in ten commonly used inbred strains of mice is greatly in excess of that expected under current assumptions.’ Ruvinsky et al. found that even those traits which are the closest to genes, like electrophoretic mobility of enzyme glucose 6-phosphate dehydrogenase, varied within parthenogenetic Daphnia pulex clones. There is plenty of other evidence supporting the conclusion expressed in the Gartner paper and only the brief nature of these commentaries prevent me from going into details. Gartner rightly deduced that environmental variation could not be significant enough to explain the results. He also believed that ‘Due to the lack of genetic variability in inbred strains all the phenotypic variability should be environmentally induced.’ At the time, such statements were quite common but nevertheless incorrect. Now it is clear that even identical monozygotic twins do have some genetic differences and all inbred strains also have residual heterozygosity. One way or another, Gartner suggested there was a so-called ‘third component’ that was particularly influential in the situation he investigated and ‘may originate from ooplasmic differences’. While such a difference might exist and could contribute to the phenomenon, our current knowledge about numerous factors causing variability is much greater and briefly mentioned below. Gartner also thought that the lack of uniformity even in inbred animals ‘seems to be an arrangement supporting natural selection’. No doubt genetic variation is essential for effective selection, but one should hardly entertain a view that random variation was arranged especially for this purpose alone. This is rather an integral feature of any biological unit. My recently published book ‘Genetics and Randomness’ provides broad characterization of various factors causing randomness in the biological world. Here we just refer to the processes that commonly generate phenotypic variability in inbred strains of laboratory animals or individuals with identical zygotic genotype. Among them are novel mutations acquired during development of multicellular individuals, random somatic recombination, alternative splicing, stochastic variation in gene activity, random gene inactivation events and several less understood phenomena. Some level of heterozygosity, which cannot be brought to zero even in highly inbred strains, will also enhance phenotypic variation. Importantly, a sense of limits existing in nature is not always highlighted in biological publications. Why would one expect that certain experimental procedures might lead to complete disappearance of phenotypic variation? An alternative question could be asked whether it is possible at all to remove phenotypic variation from populations. The answer to this question, as discussed in the above-mentioned book, is a resounding ‘NO’. Hopefully in the future, biological thinking will embrace the idea of randomness much more. No matter how far science advances, the proportion of what is knowable to what is random will remain unchanged, and attempts to ignore this critical threshold are futile at best. With the revolutionary explosion in genetic information discovery, it is crucially important to recognize the underlying limitations of scientific prediction in genetics.