Peptide-based NTA(Ni)-nanodiscs for studying membrane enhanced FGFR1 kinase activities

Tyrosine autophosphorylation plays a crucial regulatory role in the kinase activities of fibroblast growth factor receptors (FGFRs), and in the recruitment and activation of downstream intracellular signaling pathways. Biophysical and biochemical investigations of FGFR kinase domains in membrane environments offer key insights into phosphorylation mechanisms. Hence, we constructed nickel chelating nanodiscs based on a 22-residue peptide. The spontaneous anchoring of N-terminal His6-tagged FGFR1c kinase domain (FGFR1K) onto peptide nanodiscs grants FGFR1K orientations occurring on native plasma membranes. Following membrane incorporation, the autophosphorylation of FGFR1K, as exemplified by Y653 and Y654 in the A-loop and the total tyrosine phosphorylation, increase significantly. This in vitro reconstitution system may be applicable to studies of other membrane associated phenomena.


56
To investigate the mechanisms of RTK autophosphorylation, many biological and biophysical 57 studies have been performed using analyses of isolated kinase domains in aqueous solution (Klein 58 et al. 2015;Kobashigawa et al. 2015). Yet other studies of RTKs show that isolated kinase domains 59 fail to reproduce in vivo observations (Bae et al. 2010). This largely reflects the importance of cell 60 membrane localization of kinase domains, which is essential for promoting receptor dimerization 61 and cooperative cross-receptor autophosphorylation.

62
Given the importance of protein-membrane interactions, accurate experimental mimicry of 63 membrane anchors is necessary for studies of RTK kinase domains. A strategy for binding poly-111 exclusion chromatography with a Superdex 200 GL 16/60 column (GE Healthcare) at a flow rate 112 of 1ml/min at 4°C. 20 mM Tris, pH 8.0, 400 mM NaCl was used as the running buffer. Target 113 fractions were pooled and concentrated, finally stored at −80°C.
117 Subsequently, 22-amino acid peptides were dissolved in buffer containing 40 mM potassium 118 phosphate, (pH 7.4), to make a 10 mg/mL peptide stock solutions. Peptide and lipids were then 119 mixed at indicated ratios and incubated at 50°C for 10 min and then at room temperature for 10 120 min. This procedure was repeated 3 to 5 times until the solution became clear, and three freeze and 121 thaw cycles were then performed between −80°C and room temperature to homogenize the  The specimens for transmission electron microscopy (TEM) was prepared by negative-stain 128 technique. Carbon-coated copper grids (200 mesh) were cleaned for 5 min in plasma cleaner.
129 Nanodisc samples were incubated on the grids for 1 min, and the excess was removed by blotting 130 with filter paper. The grids were then stained with phosphorotungstic acid for 20 s and excess 131 solution was removed by blotting. Air-dried samples were stored in a desiccator until observation. 132 All TEM measurements were carried out using a FEI Tecnai Spirit tandem electron microscope 133 (TEM; 100 kV). The particle diameter distribution and discoidal shape analysis were based on the 134 2D class averaging using EMAN2. For each peptide nanodiscs, at least 3000 particles were 135 selected and classified. 144 Reactions were allowed to proceed at room temperature for indicated times and were then 145 quenched using 50 mM EDTA. Subsequently, 6× loading buffer was added to each sample and 146 proteins were loaded onto polyacrylamide-sodium dodecyl sulfate gels. After electrophoretic 147 separation, proteins were transferred to nitrocellulose membranes, and these were then blocked 148 using 5% BSA in TBS-T. Proteins were probed with various antibodies and the resulting blots 149 were visualized using chemiluminescence autoradiography. The results were quantitated by 150 ImageJ analysis. Data represent the average of at least four experiments. Statistical tests were 151 conducted using either unpaired two-tailed Student's t test or one-way analysis of variance with 152 post-hoc Dunnett's test. Error bars depict the S.D. P values of *p < 0.05, **p < 0.01, and ***p < 153 0.001 were considered to be statistically significant.  Fig. S1). TEM of 22A nanodiscs (Fig. 2B-D) 177 provide a direct evidence that peptide nanodiscs form disc-like assembly. The appearance of 178 nanodisc stacking at molar ratio of 1:1 (Supplementary Fig. S2) was an artifact associated with 179 negative staining sample treatment, which were also reported in other study (R et al. 2017). The  Fig. S7), the phosphorylation level is rather low 249 (Fig. 3C), and occurs only when two kinase domains diffuse and collide with appropriate 250 orientation relative to each other. The presence of NTA(Ni)-nanodiscs, on the other hand, restricts 251 kinase domains to diffuse in a 2D space, and increases the chances of kinase collision and reaction. 252 Moreover, the membrane anchoring also helps in coordinating the kinase enzyme and kinase 253 substrate, into the appropriate orientations for phosphorylation. From our data, only NTA(Ni)-254 nanodiscs at a lipid:peptide ratio of 1:1 exhibited significant enhancements of kinase 255 oligomerization and activation, indicating the necessary of sufficient membrane surface for 256 autophosphorylation. The flexibility and convenience of peptide nanodiscs in tuning membrane 257 areas, therefore make it a versatile membrane mimic system in studying various membrane 258 associated phenomena.

259
Autophosphorylation of the tyrosine residues of FGFR1K is a precisely ordered process and is 260 kinetically driven, as shown in mass spectroscopy analyses ). These studies 261 suggest that each step of the process is accompanied by a different orientation of catalytic kinase 262 relative to substrate kinase. FGFR phosphorylation kinetics on the membrane is likely more 263 complicated, warranting further studies to determine the order of phosphorylation reactions on 264 membranes, and how membranes affect phosphorylation kinetics. Although the present western 265 blotting analyses of the kinetics of autophosphorylation at tyrosine sites on nanodiscs are semi 266 quantitative, our nanodiscs significantly increased phosphorylation levels of all tyrosine residues 267 and those at Y653 and Y654. Moreover, in the presence of nanodiscs, Y653 phosphorylation 268 plateaued after 1 h, whereas the Y654 phosphorylation continue to increase (Fig. 4), indicating 269 phosphorylation of Y653 happens earlier than that of Y654. This is consistent with previous report 270 that Y653 is the first site to be phosphorylated and stimulates kinase activity by 50-to 100-fold 271 ). However, due to the complexity of the autophosphorylation mechanism and the is required for tyrosine autophosphorylation of fibroblast growth factor receptor in living cells.