Mice cast light on missing heritability

RichardMott
Dr Richard Mott

In recent years, genome-wide association studies have uncovered many new associations between genetic variants and disease. However, in most cases the amount of variability explained by genetics – the heritability – is less than studies of inheritance in families had predicted. Now Richard Mott and his colleagues at the Centre have found that the impact of most genetic variants depends crucially on whether they are inherited from the male or the female parent.

The phenomenon of ‘genetic imprinting’ or more generally ‘parent-of-origin effects’, in which the expression of a gene depends on whether offspring inherit it from the mother or the father, is known to apply only to some 100 genes in mammals. However some traits, such as growth, also show parent-of-origin effects.  To explore the question of parent-of-origin effects on complex traits (which depend on the combined actions of many genes), Mott and his colleagues turned to a well-studied population of genetically heterogeneous laboratory mice, whose pedigrees, genotypes and physical characteristics have been recorded. The results are just published in the journal Cell.

They measured 97 traits, most of which were not thought to be imprinted, and found that contrary to expectation, the heritability of almost all showed parent-of-origin effects. This means that identical alleles shared by two offspring but inherited from parents of different sexes generally contribute less to heritable variation than do identical alleles inherited from parents of the same sex.

Importantly, family structure also contributes to this effect. Siblings were more alike than non-siblings, as would be expected as they had the same parents, but they were also much more likely than non-siblings to share alleles inherited from a parent of the same sex. ‘This is a purely statistical consequence of the rules of Mendelian inheritance’, says Mott. ‘However, we were surprised to find that the components of heritability attributed to allele sharing from parents of different sexes was about half that from parents of the same sex.’ This result means that parent-of-origin effects will have added to estimates of heritability derived from studies of twins.

In a further step, the researchers bred mice in which genes that were not known to be imprinted but showed parent-of-origin effects had been knocked out in one or other parent. Those offspring that inherited one copy of the knockout variant not only showed different expression of the gene in question, but also signs that the control of gene expression was disrupted across the whole genome. Many other genes were expressed from only one chromosome, crucially including known imprinted genes.

 ‘It seems that this dysregulation of imprinting may generate the parent-of-origin effects observed’, says Mott. ‘We now need to see if these findings can be replicated in human studies.’

 

Mott, R et al, The Architecture of Parent of Origin Effects in Mice, Cell (2014), 16 January, pp 332-342.