Fibroblast growth factor receptor 1 (FGFR-1), also known asbasic fibroblast growth factor receptor 1,fms-related tyrosine kinase-2 / Pfeiffer syndrome, andCD331, is areceptor tyrosine kinase whose ligands are specific members of thefibroblast growth factor family. FGFR-1 has been shown to be associated withPfeiffer syndrome,[5] andclonal eosinophilias.[6]
TheFGFR1 gene is located on human chromosome 8 at position p11.23 (i.e. 8p11.23), has 24 exons, and codes for aPrecursor mRNA that isalternatively spliced at exons 8A or 8B thereby generating twomRNAs coding for two FGFR1isoforms, FGFR1-IIIb (also termed FGFR1b) and FGFR1-IIIc (also termed FGFR1c), respectively. Although these two isoforms have different tissue distributions and FGF-binding affinities, FGFR1-IIIc appears responsible for most of functions of the FGFR1 gene while FGFR1-IIIb appears to have only a minor, somewhat redundant functional role.[7][8] There are four other members of theFGFR1 gene family:FGFR2,FGFR3,FGFR4, andFibroblast growth factor receptor-like 1 (FGFRL1). TheFGFR1 gene, similar to theFGFR2-4 genes are commonly activated in human cancers as a result of theirduplication,fusion with other genes, andpoint mutation; they are therefore classified asproto-oncogenes.[9]
FGFR1 is a member of thefibroblast growth factor receptor (FGFR) family, which in addition to FGFR1, includes FGFR2, FGFR3, FGFR4, and FGFRL1. FGFR1-4 arecell surface membrane receptors that possesstyrosine kinase activity. A full-length representative of these four receptors consists of an extracellular region composed of threeimmunoglobulin-like domains which bind their properligands, thefibroblast growth factors (FGFs), a single hydrophobic stretch which passes through the cell's surface membrane, and a cytoplasmic tyrosine kinase domain. When bonded to FGFs, these receptors formdimers with any one of the four other FGFRs and thencross-phosphorylate keytyrosine residues on their dimer partners. These newly phosphorylated sites bind cytosolic docking proteins such asFRS2,PRKCG andGRB2 which proceed to activatecell signaling pathways that lead tocellular differentiation, growth, proliferation, prolonged survival, migration, and other functions. FGFRL1 lacks a prominent intracellular domain and tyrosine kinase activity; it may serve as a decoy receptor by binding with and thereby diluting the action of FGFs.[9][10] There are 18 known FGFs that bind to and activate one or more of the FGFRs: FGF1 to FGF10 and FGF16 to FGF23. Fourteen of these, FGF1 to FGF6, FGF8, FGF10, FGF17, and FGF19 to FGF23 bind and activate FGFR1.[11] FGFs binding to FGFR1 is promoted by their interaction with cell surfaceheparan sulfate proteoglycans and, with respect to FGF19, FGF20, and FGR23, thetransmembrane proteinKlotho.[11]
FGFR1, when bound to a proper FGF, elicits cellular responses by activating signaling pathways that include the:a)Phospholipase C/PI3K/AKT,b)Ras subfamily/ERK,c)Protein kinase C,d)IP3-induced raising of cytosolic Ca2+, ande) Ca2+/calmodulin-activated elements and pathways. The exact pathways and elements activated depend on the cell type being stimulated plus other factors such as the stimulated cells microenvironment and previous as well as concurrent history of stimulation[9][10]
Activation of the gamma isoforms ofphospholipase C (PLCγ) (seePLCG1 andPLCG2 illustrates one mechanism by which FGFR1 activates cell stimulating pathways. Following its binding of a proper FGF and subsequent pairing with another FGFR, FGFR1 becomesphosphorylated by its partner FGFR on a highly conserved tyrosine residue (Y766) at its C-terminal. This creates a binding or "docking" site to recruit PLCγ via PLCγ tandemnSH2 and cSH2 domains and then phosphorylate PLCγ. By being phosphorylated PLCγ is relieved of its auto-inhibition structure and becomes active in metabolizing nearbyPhosphatidylinositol 4,5-bisphosphate (PIP2) to twosecondary messengers,inositol 1,4,5-trisphosphate (IP3) anddiacyglycerol (DAG). These secondary messengers proceed to mobilize other cell-signaling and cell-activating agents: IP3 elevates cytosolic Ca2+ and thereby variousCa2+-sensitive elements while DAG activates variousprotein kinase C isoforms.[11]
Recent publication on the 2.5 Å crystal structure of PLCγ in complex with FGFR1 kinase (PDB: 3GQI) provides new insights in understanding the molecular mechanism of FGFR1's recruitment of PLCγ by its SH2 domains. Figure 1 on the extreme right shows the PLCγ-FGFR1 kinase complex with the c-SH2 domain colored in red, n-SH2 domain colored in blue, and the interdomain linker colored in yellow. The structure contains typical SH2 domain, with twoα-helices and threeantiparallel β-strands in each SH2 domain. In this complex, the phosphorylated tyrosine (pY766) on the C-terminal tail of FGFR1 kinase binds preferentially to the nSH2 domain of PLCγ. The phosphorylation of tyrosine residue 766 on FGFR1 kinase forms hydrogen bonds with the n-SH2 to stabilize the complex. Hydrogen bonds in the binding pocket help to stabilize the PLCγ-FGFR1 kinase complex. The water molecule as shown mediates the interaction of asparagine 647 (N647) and aspartate 768 (D768) to further increase the binding affinity of the n-SH2 and FGFR1 kinase complex. (Figure 2). The phosphorylation of tyrosine 653 and tyrosine 654 in the active kinase conformation causes a large conformation change in the activation segment of FGFR1 kinase. Threonine 658 is moved by 24Å from the inactive form (Figure 3.) to the activated form of FGFR1 kinase (Figure 4.). The movement causes the closed conformation in the inactive form to open to enable substrate binding. It also allows the open conformation to coordinate Mg2+ with AMP-PCP (analog of ATP). In addition, pY653 and pY654 in the active form helps to maintain the open conformation of the SH2 and FGFR1 kinase complex. However, the mechanism by which the phosphorylation at Y653 and Y654 helps to recruit the SH2 domain to its C-terminal tail upon phosphorylation of Y766 remains elusive. Figure 5 shows the overlay structure of active and inactive forms of FGFR1 kinase. Figure 6 shows the dots and contacts on phosphorylated tyrosine residues 653 and 654. Green dots show highly favorable contacts between pY653 and pY654 with surrounding residues. Red spikes show unfavorable contacts in the activation segment. The figure is generated through Molprobity extension on Pymol.
The tyrosine kinase region of FGFR1 binds to the N-SH2 domain of PLCγ primarily through charged amino acids. Arginine residue (R609) on the N-SH2 domain forms a salt bridge to aspartate 755 (D755) on the FGFR1 domain. The acid base pairs located in the middle of the interface are nearly parallel to each other, indicating a highly favorable interaction. The N-SH2 domain makes an additional polar contact through water-mediated interaction that takes place between the N-SH2 domain and the FGFR1 kinase region. The arginine residue 609 (R609) on the FGFR1 kinase also forms a salt bridge to the aspartate residue (D594) on the N-SH2 domain. The acid-base pair interacts with each other carry out areduction–oxidation reaction that stabilizes the complex (Figure 7). Previous studies have done to elucidate the binding affinity of the n-SH2 domain with the FGFR1 kinase complex by mutating these phenylalanine or valine amino acids. The results from isothermal titration calorimetry indicated that the binding affinity of the complex decreased by 3 to 6-fold, without affecting the phosphorylation of the tyrosine residues.[12]
FGF-induced activation of FGFR1 also stimulates the activation ofsprouty proteinsSPRY1,SPRY2,SPRY3, and/orSPRY4 which in turn interact with GRB2,SOS1, and/orc-Raf to reduce or inhibit further cell stimulation by activated FGFR1 as well as other tyrosine kinase receptors such as theEpidermal growth factor receptor. These interactions serve asnegative feedback loops to limit the extent of cellular activation.[11]
Mice genetically engineered to lack a functionalFgfr1 gene (ortholog of the humanFGFR1 gene) die in utero before 10.5 days of gestation. Embryos exhibit extensive deficiencies in the development and organization ofmesoderm-derived tissues and themusculoskeletal system. TheFgfr1 gene appears critical for thetruncation of embryonic structures and formation of muscle and bone tissues and thereby the normal formation of limbs, skull, outer, middle, and inner ear,neural tube, tail, and lower spine as well as normal hearing.[11][13][14]
Hereditary mutations in theFGFR1 gene are associated with various congenital malformations of themusculoskeletal system. Interstitial deletions at human chromosome 8p12-p11, arginine to a stopnonsense mutation at FGFR1 amino acid 622 (annotated as R622X), and numerous otherautosomal dominant inactivating mutations inFGFR1 are responsible for ~10% of the cases ofKallmann syndrome. This syndrome is a form ofhypogonadotropic hypogonadism associated in a varying percentage of cases withanosmia orhyposmia;cleft palate and other craniofacial defects; andscoliosis and other musculoskeletal malformations. An activating mutation in FGFR1 viz., P232R (proline-to-arginine substitution in the protein's 232nd amino acid), is responsible for the Type 1 or classic form of Pfeiffer syndrome, a disease characterized bycraniosynostosis and mid-face deformities. A tyrosine-to-cysteine substitution mutation in the 372nd amino acid of FGFR1 (Y372C) is responsible for some cases of Osteoglophonic dysplasia. This mutation results incraniosynostosis, mandibularprognathism,hypertelorism,brachydactyly, and inter-phalangeal joint fusion. Other inherited defects associated with 'FGFR1 mutations likewise involve musculoskeletal malformations: these include theJackson–Weiss syndrome (proline to arg substitution at amino acid 252),Antley-Bixler syndrome (isoleucine-to-threonine at amino acid 300 (I300T), andtrigonocephaly (mutation the same as the one for the Antley-Bixler syndrome viz., I300T).[10][11][15]
Somatic mutations andepigenetic changes in the expression of theFGFR1 gene occur in and are thought to contribute to various types of lung, breast, hematological, and other types of cancers.
Amplification of theFGFR1 gene (four or more copies) is present in 9 to 22% of patients withnon-small-cell lung carcinoma (NSCLC).FGFR1 amplification was highly correlated with a history of tobacco smoking and proved to be the single largest prognostic factor in a cohort of patients suffering this disease. About 1% of patients with other types of lung cancer show amplifications in FGFR1.[9][10][16][17]
Amplification ofFGFR1 also occurs in ~10% ofestrogen receptor positive breast cancers, particularly of theluminal subtype B form of breast cancer. The presence ofFGFR1 amplification has been correlated with resistance tohormone blocking therapy and found to be a poor prognostic factor in the disease.[9][10]
In certain rare hematological cancers, thefusion ofFGFR1 with various other genes due toChromosomal translocations orInterstitial deletions create genes that encode chimeric FGFR1Fusion proteins. These proteins have continuously active FGFR1-derivedtyrosine kinase and thereby continuously stimulated the cell growth and proliferation. These mutations occur in the early stages ofmyeloid and/orlymphoid cell lines and are the cause of or contribute to the development and progression of certain types ofhematological malignancies that have increased numbers of circulating bloodeosinophils, increased numbers ofbone marrow eosinophils, and/or theinfiltration of eosinophils into tissues. These neoplasms were initially regarded aseosinophilias,hypereosinophilias,Myeloid leukemias,myeloproliferative neoplasms,myeloid sarcomas,lymphoid leukemias, ornon-Hodgkin lymphomas. Based on their association with eosinophils, unique genetic mutations, and known or potential sensitivity totyrosine kinase inhibitor therapy, they are now being classified together asclonal eosinophilias.[6] These mutations are described by connecting the chromosome site for theFGFR1 gene, 8p11 (i.e. human chromosome 8's short arm [i.e. p] at position 11) with another gene such as theMYO18A whose site is 17q11 (i.e human chromosome 17's long arm [i.e. q] at position 11) to yield the fusion gene annotated as t(8;17)(p11;q11). TheseFGFR1 mutations along with the chromosomal location ofFGFR1A's partner gene and the annotation of the fused gene are given in the following table.[18][19][20]
| Gene | locus | notation | gene | locus | notation | Gene | locus | notation | gene | locus | notation | gene | locus | notation | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MYO18A | 17q11 | t(8;17)(p11;q11) | CPSF6 | 12q15 | t(8;12)(p11;q15) | TPR | 1q25 | t(1;8)(q25p11;; | HERV-K | 10q13 | t(8;13)(p11-q13) | FGFR1OP2 | 12p11 | t(8;12)(p11;q12) | ||||
| ZMYM2 | 13q12 | t(8;13)(p11;q12) | CUTL1 | 7q22 | t(7;8)(q22;p11) | SQSTM1 | 5q35 | t(5;8)(q35;p11 | RANBP2 | 2q13 | t(2;8)(q13;p11) | LRRFIP1 | 2q37 | t(8;2)(p11;q37) | ||||
| CNTRL | 9q33 | t(8;9)(p11;q33) | FGFR1OP | 6q27 | t(6;8)(q27;p11) | BCR | 22q11 | t(8;22)(p11;q11 | NUP98 | 11p15 | t(8;11)(p11-p15) | MYST3 | 8p11.21 | multiple[21] | ||||
| CEP110 | 16p12 | t(8;16)(p11;p12) |
These cancers are sometimes termed8p11myeloproliferative syndromes based on the chromosomal location of theFGFR1 gene. Translocations involvingZMYM2,CNTRL, and FGFR1OP2 are the most common forms of these 8p11 syndromes. In general, patients with any of these diseases have an average age of 44 and present with fatigue,night sweats, weight loss, fever,lymphadenopathy, and enlarged liver and/or spleen. They typically evidence hematological features of themyeloproliferative syndrome with moderate to greatly elevated levels of blood and bone marrow eosinophils. However, patients bearing:a)ZMYM2-FGFR1 fusion genes often present asT-cell lymphomas with spreading to non-lymphoid tissue;b)FGFR1-BCR fusion genes usually present aschronic myelogenous leukemias;c)CEP110 fusion genes may present as achronic myelomonocytic leukemia with involvement of tonsil; andd)FGFR1-BCR orFGFR1-MYST3 fusion genes often present with little or no eosinophilia. Diagnosis requires conventionalcytogenetics usingFluorescence in situ hybridization#Variations on probes and analysis forFGFR1.[19][21]
Unlike many other myeloid neoplasms with eosinophil such as those caused byPlatelet-derived growth factor receptor A orplatelet-derived growth factor receptor B fusion genes, the myelodysplasia syndromes caused byFGFR1 fusion genes in general do not respond totyrosine kinase inhibitors, are aggressive and rapidly progressive, and require treatment withchemotherapy agents followed bybone marrow transplantion in order to improve survival.[19][18] The tyrosine kinase inhibitorPonatinib has been used as mono-therapy and subsequently used in combination with intensive chemotherapy to treat the myelodysplasia caused by theFGFR1-BCR fusion gene.[19]
Phosphaturic mesenchymal tumors is characterized by a hypervascular proliferation of apparently non-malignant spindled cells associated with a variable amount of ‘smudgy’ calcified matrix but a small subset of these tumors exhibit malignant histological features and may behave in a clinically malignant fashion. In a series of 15 patients with this disease, 9 were found to have tumors that bore fusions between theFGFR1 gene and theFN1 gene located on human chromosome 2 at position q35.[22] TheFGFR1-FN1 fusion gene was again identified in 16 of 39 (41%) patients with phosphaturic mesenchymal tumors.[23] The role of the(2;8)(35;11)FGFR1-FN1 fusion gene in this disease is not known.
Elevated expression of FGFR1 protein was detected in 10 of 10 humanRhabdomyosarcoma tumors and 4 of 4 human cell lines derived from rhabdomyocarcoma. The tumor cases included 6 cases ofAlveolar rhabdomyosarcoma, 2 cases ofEmbryonal rhabdomyosarcoma, and 2 cases ofpleomorphic rhabdomyosarcoma. Rhabdomyosarcoma is a highly malignant form of cancer that develops from immature skeletal muscle cell precursors viz.,myoblastss that have failed to fullydifferentiate. FGFR1 activation causes myoblast to proliferate while inhibiting their differentiation, dual effects that may lead to the assumption of a malignantphenotype by these cells. The 10 human rhabdomyosarcoma tumor exhibited decreased levels of methylation ofCpG islandsupstream of the first FGFR1exon. CpG islands commonly function to silence expression of adjacent genes while theirmethylation inhibits this silencing. Hypomethylation of CpG islands upstream of FGFR1 is hypothesized to be at least in part responsible for the over-expression of FGFR1 by and malignant behavior of these rhabdomyosarcoma tumors.[24] In addition, a single case of rhabdomyosarcoma tumor was found express co-amplifiedFOXO1 gene at 13q14 andFGFR1 gene at 8p11, i.e. t(8;13)(p11;q14), suggesting the formation, amplification, and malignant activity of a chimericalFOXO1-FGFR1 fusion gene by this tumor.[9][25]
Acquired abnormalities if theFGFR1 gene are found in: ~14% of urinary bladderTransitional cell carcinomas (almost all are amplifications); ~10% of squamous cellHead and neck cancers (~80% amplifications, 20% other mutations); ~7% ofendometrial cancers (half amplifications, half other types of mutations); ~6% ofprostate cancers (half amplifications, half other mutations); ~5% of ovarianPapillary serous cystadenocarcinoma (almost all amplifications); ~5% ofcolorectal cancers (~60 amplifications, 40% other mutations); ~4% ofsarcomas (mostly amplifications); <3% ofGlioblastomas (Fusion ofFGFR1 andTACC1 (8p11) gene); <3% ofSalivary gland cancer (all amplifications); and <2% in certain other cancers.[11][26][27]
FGFR-targeted drugs exert direct as well as indirect anticancer effects, due to the fact that FGFRs on cancer cells and endothelial cells are involved in tumorigenesis and vasculogenesis, respectively.[9] FGFR therapeutics are active as FGF affects numerous features of cancers, such asinvasiveness,stemness andcellular survival. Primary among such drugs are antagonists. Small molecules that fit between the ATP binding pockets of thetyrosine kinase domains of the receptors. For FGFR1, numerous such small molecules have been approved for targeting the TKI ATP pocket. These includedovitinib andbrivanib. The table below provides the IC50 (nanomolar) of small-molecule compounds targeting FGFRs.[9]
| PD173074 | Dovitinib | Ki23057 | Lenvatinib | Brivanib | Nintedanib | Ponatinib | MK-2461 | Lucitanib | AZD4547 |
|---|---|---|---|---|---|---|---|---|---|
| 26 | 8 | NA | 46 | 148 | 69 | 2.2 | 65 | 18 | 0.2 |
FGFR1 mutation in breast and lung cancer as a result ofgenetic over-amplification is effectively targeted using dovitinib andponatinib, respectively.[28] Drug resistance is a highly relevant topic in the field of drug development for FGFR targets. FGFR inhibitors allow for the increase of tumor sensitivity to cytotoxic anticancer drugs such aspaclitaxel, andetoposide in human cancer cells, thereby decreasingantiapoptotic potential based on faulty FGFR activation.[9] Since FGF signaling inhibition dramatically reduces revascularization, it interferes with one of the hallmarks of cancers,angiogenesis. It also reduces tumor burden in human tumors that depend onautocrine FGF signaling, based on FGF2 upregulation following the common VEGFR-2 therapy for breast cancer. Thus, FGFR1 can act synergistically with therapies to cut off cancer clonal resurgence by eliminating potential pathways of future relapse. Moreover, FGF signaling inhibition dramatically reduces revascularization.[29][30]
FGFR inhibitors have been predicted to be effective on relapsed tumors because of the clonal evolution of an FGFR-activated minor subpopulation after therapy targeted to EGFRs or VEGFRs. Because there are multiple mechanisms of action for FGFR inhibitors to overcome drug resistance in human cancer, FGFR-targeted therapy might be a promising strategy for the treatment of refractory cancer.[31]
AZD4547 has undergone a phase II clinical trial in gastric cancer and reported some results.[32]
Lucitanib is an inhibitor ofFGFR1 andFGFR2 and has undergone clinical trials for advanced solid tumors.[33]
Dovitinib (TKI258), an inhibitor of FGFR1,FGFR2, andFGFR3, has had a clinical trial on FGFR-amplified breast cancers.[28]
Fibroblast growth factor receptor 1 has been shown tointeract with:
This article incorporates text from theUnited States National Library of Medicine, which is in thepublic domain.
PDB gallery | |
|---|---|
|
