Bacterial production of human discoidin domain receptor 2 (DDR2) kinase and its gatekeeper mutants to assess activation by phosphorylation

Abstract

Discoidin domain receptor 2 (DDR2) belongs to a subfamily of receptor tyrosine kinases (RTKs) known as discoidin domain receptors (DDRs). DDRs are characterized by an extracellular discoidin homology domain and being activated by collagen resulting in a delayed but sustained activation. DDR2 overexpression was reported in some tumors, atherosclerosis as well as a number of inflammatory and fibrotic conditions. Recently 3 kinase inhibitor drugs (imatinib, nilotinib and dasatinib) were reported to inhibit DDR2 kinase activity suggesting it might have a role in their therapeutic effect. However minimal information is available on DDR2 downstream signaling. Identification of DDR2 substrates is essential to understand the molecular mechanisms underlying the associated diseases and predict side effects and resistance to kinase inhibitors. Identification of kinase substrates is a challenging task as all kinases use ATP to phosphorylate their substrates in addition to the low stoichiometry of phosphorylation events. Kevan Shokat lab designed a chemical genetic approach for substrate identification, where kinases are engineered to accept synthetic ATP analogs (A*TP) and kinase inhibitors that do not bind to the wild types (analog sensitive kinases). The first aim of the present study was to construct a series of kinase domain mutations and a functional analog sensitive kinase to help in the identification of DDR2 substrates. Currently DDR2 protein is produced in insect and mammalian cells, which are expensive and time-consuming expression systems. The second aim was to develop a protocol for DDR2 kinase production using the less demanding bacterial system. A series of analog sensitive, drug resistant, activating and kinase dead mutations were constructed and the kinase domains were expressed in Rosetta pLysS bacterial strain as glutathione-S-transferase (GST) fusion proteins. The proteins were produced in soluble forms with milligram yields. However preliminary kinase and autophosphorylation reactions of the purified proteins revealed low kinase activities. The results show that if the bacterial expression system is to be pursued, in vitro activation using the upstream Src kinase or Src co-expression is essential to produce a highly active DDR2 kinase.