Drug Repurposing: Lithium Chloride as a PI3K Mutant Selective Inhibitor

Drug repurposing is the extension of the clinical use of a known drug to new diseases and exploits either the activity of the drug against its intended primary target in other tissues or secondary off-targets1. This is an important strategy that significantly reduces the cost and timelines associated with the development of novel compounds because known drugs have already been evaluated and optimized in humans and have established safety profiles. Notable examples of repurposed drugs include the use of Sildenafil (Viagra) for erectile dysfunction and Thalidomide for the treatment of cancer and leprosy.

Historically, alternative uses for known drugs have been discovered by chance in the clinic. However, recently there has been a move towards the systematic screening of approved drugs using cell line platforms1. This approach was exemplified in work by Professor Ben Park and colleagues at John Hopkins University who used Horizon Discovery’s Cell Lines to discover a new use for the bipolar disorder drug Lithium chloride in the treatment of cancers carrying PI3K mutations2.

Horizon Discovery’s suite of Cell Lines contain patient relevant mutations, engineered at their native gene loci. These are an ideal tool for drug repurposing applications because they more accurately reflect the disease state compared to the over-expression of mutant oncogenes from transgenes that have been randomly inserted into the genome, an approach that often leads to artefacts. 

Results and Discussion

The p110α catalytic subunit of PI3K is frequently mutated in a range of cancers with two of the most common oncogenic mutations being E545K and H1047R. To identify novel therapeutic targets in tumor cells carrying oncogenic PI3K Gustin et al examined the effect that introduction of the E545K or H1047R mutations had on key signaling pathways in non-tumorigenic MCF10A human breast epithelial Cell Lines by Western blot2. This revealed that the presence of oncogenic PI3K leads to elevated levels of phosphorylated GSK3β suggestive of upregulation of this pathway (Figure 1).

The above observation raised the possibility that cells expressing oncogenic PI3K may be more sensitive to GSKβ inhibition. To test this Gustin et al treated cells with the GSKβ inhibitors lithium chloride and SB216763, which revealed that treatment with either compound selectively reduced the growth of PI3K mutant MCF10A cells (Figure 2) ². 


This differential sensitivity to treatment with LiCl was confirmed using a second HCT116 human colorectal cancer cell line pair2, which express either a single wild type or mutant H1047R allele (Figure 3).

To confirm these findings in vivo Gustin et al. performed mouse xenograft experiments using the HCT116 isogenic pair to assess the effect of lithium chloride2. Mice were treated orally to achieve serum levels comparable to those in human patients treated with lithium chloride for bipolar disorder. Consistent with the in vitro results these experiments revealed that oncogenic PI3K-expressing tumors exhibited significantly less tumor growth than wild type-expressing cells (Figure 3).



These results from Gustin et al. show that GSK3β is a key effector of mutant PI3Kα and that the approved drug Lithium chloride selectively inhibits the growth of cells expressing oncogenic PI3Kα2. This ‘re-profiled’ candidate cancer drug, which has an established safety profile, is now directly entering clinical trials for efficacy in breast cancers carrying mutant PI3Kα. This study validates the power of Cell Lines carrying patient-relevant oncogenic mutations for target validation, drug discovery and as an important tool for identifying new applications for known drugs. 

PI3K Cell Lines  Reposition drugs using high throughput combination screening


  1. Huang et al., Sci Transl Med 2011 (3) p80ps16
  2. Gustin et al., PNAS 2009 (106) p2835
Back to top