Recently, this remarkable paper went up on @biorxivpreprint – the authors wrote that CRISPR-induced null mutations trigger a pervasive up-regulation of homologs of the deleted gene. It has huge implications for CRISPR technology as well as our understanding of genetic diseases.
My lab makes a lot of null mutations in cancer cells, so it had immediate relevance for our work. I wanted to see if I could observe this phenomenon, so I searched the Gene Expression Omnibus for RNA-Seq experiments using knockout cancer cell lines.
In short, we see no evidence that homologue up-regulation is a common consequence of null mutations in cancer. Here’s an experiment in which ARID1A is knocked out in colon cancer cells, but there’s no up-regulation of its close homologue ARID1B (elifesciences.org/articles/30506). 
Here’s an experiment in which KRAS is knocked out in mouse lung cancer cells, but there’s no up-regulation of HRAS or NRAS (pnas.org/content/114/14…).
I looked at 10 separate experiments, and you can certainly find instances in which a homolog of a targeted gene is up, but it does not appear to be a common phenomenon. This raises the very exciting possibility that cancer cells have lost this compensatory pathway.
I’m interested in hearing what other people think about this phenomenon. We have a freezer full of CRISPR-KO cancer lines to study, and understanding this phenomenon seems crucial for getting a complete picture of the genetic architecture of cancer cells.
