Methods for mapping the genetic alterations that underlie complex traits are getting ever more sophisticated, but modelling these often subtle changes is usually left to rather crude manipulations. In mice, for example, the expression of transgenes that are microinjected into a one-cell embryo is unpredictable as it depends on how many copies are integrated into the genome and where in the genome the integration occurs. To obtain a more predictable and less intrusive means of controlling gene expression, Kakoki and colleagues turned to the 3′-untranslated region (UTR), which has been shown many times before to influence the stability of a gene's mRNA. Their aim was to find defined changes in the 3′ UTR that would allow the expression of a gene to be altered in situ in a predictable way.
To test their idea, they created mouse embryonic stem (ES) cells in which a GFP gene was inserted into an endogenous locus. Altering the 3′ sequences of the transgene allowed them to assess the effect of various 3′ regions on GFP expression by monitoring the level of fluorescence emitted by the cells. Protein expression (which correlated with mRNA levels) varied over a 100-fold range according to whether the 3′ sequences were derived from say, the
Fos
gene, which has a very unstable message, or from the bovine growth hormone, which has a very stable message. These data could then be used to introduce systematic changes into a 3′ UTR, while being careful to retain features that are specific to that gene: for example, the inclusion of a GU/U-rich element downstream of the 3′ UTR increased expression levels by 2–3-fold.
This is a preview of subscription content, access via your institution