Identification of new FGF1 binding partners-Implications for its intracellular function

Abstract

Besides its classical mode of action through activation of specific receptors at the cell surface, fibroblast growth factor 1 (FGF1) can also cross the cellular membrane and translocate into the cytosol and further to the nucleus. The mechanism of this translocation is described partially, but the role of FGF1 inside the cell remains unknown. The aim of our work was to identify novel binding partners of FGF1 to predict its intracellular functions. We combined three methods of identification of such partners based on different principles: yeast two-hybrid screen and mass spectrometry (MS) analysis of complexes obtained by Tandem Affinity Purification (TAP) or by co-precipitation from cell lysate using recombinant FGF1. Altogether, we identified twenty novel intracellular proteins interacting with FGF1. For selected proteins, their direct interaction with FGF1 was confirmed by pull-down assays and SPR measurements. Interestingly, half of the proteins found are involved in processes related to cell viability, such as apoptosis, cell proliferation, and cell cycle regulation. Thus, our study indicates that the role of intracellular FGF1 is to protect the cell against stress conditions by providing an additional signal for cell survival, independently of receptor-activated signaling cascades.

Keywords: FGF1; Tandem Affinity Purification; apoptosis; binding partners; cell survival; intracellular function; translocation.

Figure 1

Three methods for identification of partner proteins: Tandem Affinity Purification (TAP) (A), co‐precipitation using recombinant FGF1 as bait and cell lysate (B), and yeast two‐hybrid screen (C).

Figure 2

Schematic presentation of TRIP11 and UACA identified as FGF1 partners using yeast two‐hybrid screen. Blue lines indicate selected interaction domains (SID) identified within these proteins.

Figure 3

Verification of intracellular FGF1 partners by co‐precipitation. A: BJ cell lysate was incubated with SBP‐FGF1 and then with dynabeads, or with dynabeads alone. Protein complexes were eluted from the beads with sample buffer, resolved by SDS‐PAGE and subjected to Western blotting using specific antibodies as indicated. B: HEK 293 cells transiently transfected with myc‐FGF1 were lysed and then subjected to immunoprecipitation followed by SDS‐PAGE and Western blotting with an anti‐nucleolin, anti‐p53, anti‐nucleophosmin, or anti‐UACA antibody. As controls non‐transfected cells were used.

Figure 4

Confirmation of direct interaction of FGF1 with HSP90α, MVP (113–474), nucleolin‐C (284–707), UACA (899–1403), p53, and nucleophosmin. A: Recombinant potential FGF1 binding partners were incubated with SBP‐FGF1 and then Streptavidin‐coated dynabeads or with dynabeads alone. Protein complexes were eluted from the beads with sample buffer, resolved by SDS‐PAGE and subjected to Western blotting using specific antibodies. B: Recombinant HSP90α, nucleolin‐C, or p53 were immobilized on carboxymethylated dextran sensor chip CM5 at the level of 500–1000 RU (response units). Recombinant FGF1 was injected at increasing concentrations at a flow rate of 30 μL/min and specific protein–protein interactions were measured.

Figure 5

FGFR‐independent anti‐apoptotic activity of exogenous FGF1. NIH3T3 cells were serum starved for 24 h and then treated for 6 h with FGF1 (200 ng/mL) and heparin (10 U/mL) in the presence of the specific FGFR tyrosine kinase inhibitor, 100 nM PD173084. Apoptosis was measured using ApoLive‐Glo Multiplex Assay (Promega) Next, caspase‐3/7 activity, reflecting the apoptosis progression, along with cell viability were measured using ApoLive‐Glo Multiplex Assay (Promega). Caspase‐3/7 activity data were divided by cell viability values, then normalized toward the cells untreated with FGF1 (control), and denoted as relative caspase‐3/7 activity. Graphs represent the mean ± SD of four independent experiments (***P < 0.001).

Figure 6

Gene ontology analysis of proteins identified as FGF1 binding partners. The classification was performed according to the GO annotation for “biological process” using UniProtKB database. Note that some proteins belong to more than one class.