Engineered small protein scaffold for a novel mesothelin-targeted therapy: a focus on pleural mesothelioma

In the last decades, targeted therapy has been proposed as a novel strategy to overcome tumor burden worldwide, exploiting monoclonal antibodies (mAbs, ~150 kDa) engineered to recognize tumor biomarkers for therapeutic purposes. Recently, small protein scaffolds have been also employed as targeting agents with similar functions of mAbs. Due to their molecular size (~10 kDa) and their biophysical properties, such protein structures can be used to improve drug delivery specifically in solid tumors. Mesothelin (MSLN) is a well-known tumor biomarker expressed at high levels on the surface of several solid cancers. Among these MSLN-positive tumors, pleural mesothelioma (PM) is a deadly cancer that arises from mesothelial cells lining the pleura and lacks effective treatments and diagnostic strategies. In this thesis, an engineered Fn3-based protein scaffold that recognizes MSLN with low nano-molar affinity and high specificity is proposed to target MSLN-positive cancers. Previous MSLN-binding Fn3s have been demonstrated to internalize upon binding and to enhance cancer cell killing effect. Here, a novel MSLN-binding Fn3 is further engineered for binding with MSLN against PM, testing Fn3 as a potential radionuclide carrier for diagnostic or therapeutic purposes. High affinity and specificity of MSLN/Fn3 binding is reported in newly established PM cell line via flow cytometry. Immunofluorescence assays showed that Fn3 colocalizes with MSLN on the PM cell membrane. To exploit Fn3 as a radio carrier, the protein was conjugated with DOTAGA metal chelator. Then, to optimize bioconjugation, MSLN-binding Fn3 was further engineered via site-directed mutagenesis to add a cysteine. Exploiting the newly added cysteine, bioconjugation was carried out via thiol maleimide chemistry and, the optimal 1:1 stoichiometry of DOTAGA:Fn3 was obtained. In addition, this novel variant maintained high affinity for MSLN on PM cells even when conjugated with DOTAGA. Furthermore, MSLN/Fn3 binding interaction was investigated through protein-protein docking and molecular dynamics. This analysis revealed that the N-terminus portion of MSLN in the membrane distal region is involved in the binding site with the protein scaffold. Fn3 loops are also reported to play a key role in MSLN/Fn3 binding. The predictions were validated by domain-level epitope mapping using yeast surface display. Overall, experimental data agree with computational modeling, showing that the binding site is located in the membrane-distal region of MSLN. In conclusion, the results reported in this thesis show MSLN-binding Fn3 as a potential radiopharmaceutical carrier. Fn3 can be conjugated with a radiometal chelator without lowering binding affinity for MSLN on cancer cells. Future research includes radiolabeling of Fn3 to test biodistribution in vitro and in vivo. Additionally, further investigation of the MSLN/Fn3 binding site could improve the knowledge about Fn3 biological effects in PM, and other MSLN-positive cancers. Part of this data is included in provisional patents discussing engineered proteins that bind mesothelin for use in cancer targeting.