Genome-wide CRISPR screen identifies LGALS2 as an oxidative stress-responsive gene with an inhibitory function on colon tumor growth

Reactive oxygen species (ROS) are free radicals containing oxygen. They can actively react with DNA, RNA, proteins, and lipids to cause damage, resulting in cell injury or even cell death.  ROS are mainly produced in the mitochondria as a byproduct of oxidative phosphorylation. Cells have established their own systems to tightly monitor and scavenge ROS to avoid cellular damage. However, cells could be exposed to high levels of ROS when this balance is disrupted. Some cells may even undergo apoptosis. At moderate concentrations of ROS, some cells may transform and eventually form tumors. The tumor microenvironment is different from the normal cellular microenvironment and benefit from mildly elevated ROS through activation of proliferation and survival pathways, such as STAT3 (Signal transducer and activator of transcription 3) and MAPK (Mitogen-activated protein kinase). In almost all cancers, ROS is found to be elevated. Thus, leading to our question: how are reactive oxygen species regulated in tumor cells?


The CRISPR/Cas9 system is a powerful genome editing tool adapted from the innate immune system of bacteria. It has been increasingly used in countless biomedical studies after its discovery. Besides in academia, many new medical companies have been using CRISPR to solve complex disease problems. 

To address our question about ROS, we employed the genome-wide CRISPR knockout screening system using Cas9-expressing HEK293 transduced with GeCKO v2 gRNA library A, containing 6.5x104 gRNAs, targeting over 20,000 human genes, and established a HEK293 library with each gRNA represented in over 1000 cells. The gRNAs, amplified from the HEK293 gRNA library, treated with sub-lethal H2O2, were analyzed by next-generation sequencing (NGS). We found 625 gRNA enriched and 345 gRNAs depleted in the cells treated with H2O2.  Several of the targets were previously identified with H2O2 treatment, indicating that our screening was successful.


The gene LGALS2 (encoding Gal2) is of interest to us as it is predominantly expressed in the gastrointestinal (GI) tract and downregulated in human colon tumors. Colorectal cancer is the third leading cause of cancer-related deaths in the United States. Our further investigation showed that Gal2 loss promoted tumor growth in a mouse colorectal cancer model, likely through regulating the STAT3 signaling pathway. Our results demonstrated that Gal2 plays a suppressive role in colon tumor growth and highlights the therapeutic potential of Gal2 in colon cancer.


I would like to thank my postdoctoral mentor Dr. Han, who provided great support and guidance for my study, and also my colleagues, who finished the study after I left the lab.