Oncogenic driver FGFR3-TACC3 requires five coiled-coil heptads for activation and disulfide bond formation for stability.

FGFR3-TACC3 represents an oncogenic fusion protein frequently identified in glioblastoma, lung cancer, bladder cancer, oral cancer, head and neck squamous cell carcinoma, gallbladder cancer, and cervical cancer. Various exon breakpoints of FGFR3-TACC3 have been identified in cancers; these were analyzed to determine the minimum contribution of TACC3 for activation of the FGFR3-TACC3 fusion protein. While TACC3 exons 11 and 12 are dispensable for activity, our results show that FGFR3-TACC3 requires exons 13-16 for biological activity. A detailed analysis of exon 13, which consists of 8 heptads forming a coiled coil, further defined the minimal region for biological activity as consisting of 5 heptads from exon 13, in addition to exons 14-16. These conclusions were supported by transformation assays of biological activity, examination of MAPK pathway activation, analysis of disulfide-bonded FGFR3-TACC3, and by examination of the Endoglycosidase H-resistant portion of FGFR3-TACC3. These results demonstrate that clinically identified FGFR3-TACC3 fusion proteins differ in their biological activity, depending upon the specific breakpoint. This study further suggests the TACC3 dimerization domain of FGFR3-TACC3 as a novel target in treating FGFR translocation driven cancers.

Typical achondroplasia secondary to a unique insertional variant of FGFR3 with in vitro demonstration of its effect on FGFR3 function.

We describe an individual in whom clinical and radiographic features are typical for achondroplasia, but in whom the common variants of FGFR3 that result in achondroplasia are absent. Whole exome sequencing demonstrated a novel, de novo 6 base pair tandem duplication in FGFR3 that results in the insertion of Ser-Phe after position Leu324. in vitro studies showed that this variant results in aberrant dimerization, excessive spontaneous phosphorylation of FGFR3 dimers and excessive, ligand-independent tyrosine kinase activity. Together, these data suggest that this variant leads to constitutive disulfide bond-mediated dimerization, and that this, surprisingly, occurs to an extent similar to the neonatal lethal thanatophoric dysplasia type I Ser249Cys variant.

Oncogenic driver FGFR3-TACC3 is dependent on membrane trafficking and ERK signaling.

Fusion proteins resulting from chromosomal translocations have been identified as oncogenic drivers in many cancers, allowing them to serve as potential drug targets in clinical practice. The genes encoding FGFRs, Fibroblast Growth Factor Receptors, are commonly involved in such translocations, with the FGFR3-TACC3 fusion protein frequently identified in many cancers, including glioblastoma, cervical cancer, bladder cancer, nasopharyngeal carcinoma, and lung adenocarcinoma among others. FGFR3-TACC3 retains the entire extracellular domain and most of the kinase domain of FGFR3, with its C-terminal domain fused to TACC3. We examine here the effects of targeting FGFR3-TACC3 to different subcellular localizations by appending either a nuclear localization signal (NLS) or a myristylation signal, or by deletion of the normal signal sequence. We demonstrate that the oncogenic effects of FGFR3-TACC3 require either entrance to the secretory pathway or plasma membrane localization, leading to overactivation of canonical MAPK/ERK pathways. We also examined the effects of different translocation breakpoints in FGFR3-TACC3, comparing fusion at TACC3 exon 11 with fusion at exon 8. Transformation resulting from FGFR3-TACC3 was not affected by association with the canonical TACC3-interacting proteins Aurora-A, clathrin, and ch-TOG. We have shown that kinase inhibitors for MEK (Trametinib) and FGFR (BGJ398) are effective in blocking cell transformation and MAPK pathway upregulation. The development of personalized medicines will be essential in treating patients who harbor oncogenic drivers such as FGFR3-TACC3.