Abstract

Gastrointestinal stromal tumors (GIST) are the most common mesenchymal tumors in the gastrointestinal tract. Most GIST patients respond to imatinib, yet will eventually exhibit resistance. The mechanisms of imatinib resistance have not yet been fully understood. Some mechanisms of secondary resistance are known to be heterogeneous: acquisition of secondary mutation in the KIT or PDGFRA, genomic amplification of KIT and over-expression of the protein and activation of other receptor tyrosine kinases. To explore additional mechanisms of imatinib resistance, we generated imatinib resistant cells from the imatinib sensitive KIT exon 11 mutant GIST-T1 cells, which have deletion of Gly561-Glu579 including Tyr568, a binding site for SRC kinases, by exposing them to increasing concentrations of imatinib for 6 months. Imatinib resistant cells (GIST-T1 IR) became resistant to imatinib with IC50 of 5–7 μM imatinib. GIST-T1 IR cells showed no new mutations in KIT, PDGFRA, PKC𝛉 and JAK2. KIT and its downstream intermediates including AKT, JAK2 remained hyper-activated in GIST-T1 IR with the presence of 1 μM imatinib compared with those in the parental cells. DNA microarray showed over-expression of NEDD9 (neural precursor cell expressed, developmentally down-regulated 9) in resistant cells with 532 fold higher than that in parental cells. NEDD9 also known as HEF1 (human enhancer of filamentation 1) or Cas-L (Crk-associated substrate lymphocyte type) is a docking protein at focal adhesion sites, playing a pivotal role in integrin-mediated signaling which regulates many biological functions, such as cell adhesion, migration, apoptosis, proliferation and differentiation. These cellular functions are known to be related to organogenesis, cancer metastasis, and immune response. FAK (focal adhesion kinase) and SRC kinases are responsible for NEDD9 phosphorylation upon integrin ligation. In GIST-T1 IR cells NEDD9 over-expression and SRC activation were also demonstrated by Western blotting, compared with those in the parental cells. Although imatinib (1 μM) or PP1 (a SRC inhibitor, 25 μM) alone was not enough to suppress SRC activation and its downstream intermediates in GIST-T1 IR cells, the combination of 1 μM imatinib and 10 μM PP1 was able to suppress SRC activation in GIST-T1 IR cells showing a synergistic effect on cell proliferation. Our results for the first time show the causative role of NEDD9 over-expression and SRC activation in imatinib-resistant GIST. These findings can help us to understand the mechanisms of imatinib resistance and the role of SRC inhibitors in treatment of GIST. To achieve durable therapeutic effect in patients with GIST, SRC kinases have to be targeted. Since the combination of KIT and SRC inhibitors completely inhibited SRC phosphorylation showing synergistic effect on cell proliferation compared with those of imatinib or PP1 alone in GIST-T1 IR, this combination should be requisite in treatment regimen of GIST, especially in secondary imatinib-resistant GIST.

Author notes

Disclosure: No relevant conflicts of interest to declare.