Inducible expression is often desirable for functional genetic testing to establish clear cause and effect for a specific phenotype. A particular phenotype can be demonstrably linked to expression of a specific cDNA by showing it disappears when transcription is suppressed. Although less direct, similar logic applies to shRNA expression, where a phenotype appears when an shRNA is expressed and disappears when the same shRNA is repressed. To facilitate these types of analyses, we developed inducible versions of both the H1 and U6 RNA polymerase III promoters using the tetracycline repressor element. These promoter constructs were optimized so that the addition of tetracycline (actually, a tetracycline-analog doxycycline) induces expression of the shRNA by inhibiting the tetracycline-element-specific repressor (TetR) from binding and blocking transcription. An example of induced repression of GFP can be seen in the figure below.
We routinely use tetracycline-inducible shRNA promoters to validate the effectiveness of individual shRNA sequences identified in our screenings. In particular, with potentially lethal shRNA that target essential genes for cell viability, it is almost a requirement to prevent expression of the shRNA until the cells are established so that the phenotype of cell arrest, necrosis, or apoptosis can be clearly observed and specifically linked to expression of the shRNA. Since the purpose of a majority of our screens is to identify essential genes required for cell proliferation, these inducible shRNA constructs are essential for validating the identified "hits."
For general functional shRNA screening, however, inducibility is not always desirable. Although it is often assumed by research groups with whom we interact that a library with an inducible shRNA promoter would produce more reproducible and quantitative hits from a genetic screen, our experience indicates this is not necessarily the case. For example, with viability screens, where we are simply looking for which shRNA sequences inhibit cell proliferation (i.e., shRNAs that are depleted in the overall cell population after several divisions), the need to induce the expression of the library by adding doxycycline to the cells complicates the screening procedure and can introduce some unnecessary variation in the system. Of course, with some screens, it may be preferable or even necessary to use an inducible library. For example, screens to identify genes which repress a particular reporter may be easier to carry out when shRNA expression in the library is repressed until sometime after infection and selection of a baseline population. Also, for in vivo screening, using an inducible library may be almost essential so that significant library shRNA expression and selection does not occur until the cells are established in the mouse model. However, there is no clear benefit to including a defined induction step in any particular screen.
As with most experimental options, choosing between inducible vs. constitutive shRNA expression for a library requires careful consideration of the experimental setup. While the disadvantages are not always so obvious, there are often unforeseen drawbacks in adding seemingly small variables into what is already a technically challenging assay.