Transforming potential of alternatively spliced variants of fibroblast growth factor receptor 2 in human mammary epithelial cells.

Fibroblast growth factor receptor 2 (FGFR2) is gene amplified and alternatively spliced in the SUM-52 human breast cancer cell line. SUM-52 cells over express nine distinct isoforms of the receptor tyrosine kinase FGFR2. The isoforms differ in expression of immunoglobulin-like domains, presence or absence of the acid box, and carboxyl termini. All three potential carboxyl termini, C1, C2, and C3, are represented among FGFR2 isoforms expressed by the SUM-52 cells. We have focused on two otherwise identical FGFR2 variants, differing only in their carboxyl termini (C1 or C3). FGFR2-C1 is the version normally expressed by human mammary epithelial (HME) cells, whereas FGFR2-C3 expression has only been found in tumors. We over expressed the full-length FGFR2-C1 and C3 isoforms in HME cells, and both receptors induced transformed phenotypes of growth factor independence and anchorage independent growth. Strikingly, our results suggest that expression of FGFR2-C3 leads to more potent transformation than FGFR2-C1. HME cells expressing FGFR2-C3 are invasive of basement membrane and activate robust levels of downstream signaling pathways, consistent with the tumor-specific expression of the FGFR2-C3 variant. By contrast, expression of the FGFR2-C1 isoform did not confer invasiveness or ligand-independent signal transduction. Specifically, FGFR2-C3 expressing cells exhibit constitutively high levels of tyrosine phosphorylation of the adapter protein FGF receptor substrate 2 (FRS2), which are also elevated in the presence of stimulation with the FGFR2 ligand keratinocyte growth factor (KGF). In contrast, FGFR2-C1 cells require constant stimulation with KGF to achieve and maintain similar levels of FRS2 activation. Tyrosine 769 is the only phosphorylated tyrosine present on the C1 carboxyl terminus and absent from C3. Site-mutation of Tyr769 did not yield an FGFR2-C1 receptor with similar signaling activity as FGFR2-C3, suggesting that interactions involving Tyr769 are not inhibitory to FGFR2-C1 signal transduction. Overall, this work strongly suggests that FGFR2 variants including the C3 carboxyl terminus are more potent signaling molecules than FGFR2-C1 isoforms. Therefore, amplification and alternative splicing leading to aberrant expression of FGFR2-C3 variants contributed to the transformation of the SUM-2 breast cancer cells.PhDBiological SciencesCellular biologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/125715/2/3208515.pd