- Email: [email protected]
Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1
Accepted Manuscript Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1 Louis Perdios, Tom D. Bunney, Sean C. Warren, Christopher Dunsby, Paul M.W. French, Edward W. Tate, Matilda Katan PII:
S2212-4926(15)30015-4
DOI:
10.1016/j.jbior.2015.09.002
Reference:
JBIOR 117
To appear in:
Advances in Biological Regulation
Received Date: 17 September 2015 Accepted Date: 22 September 2015
Please cite this article as: Perdios L, Bunney TD, Warren SC, Dunsby C, French PMW, Tate EW, Katan M, Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1, Advances in Biological Regulation (2015), doi: 10.1016/j.jbior.2015.09.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1
Edward W. Tate2 and Matilda Katan1*
1
RI PT
Louis Perdios1,2, Tom D. Bunney1, Sean C. Warren3, Christopher Dunsby3, Paul M. W. French3,
London, Gower St, London WC1E 6BT, UK 2
SC
Institute of Structural and Molecular Biology, Division of Biosciences, University College
Road, London SW7 2AZ, UK 3
M AN U
Department of Chemistry, Imperial College London, South Kensington Campus, Exhibition
Department of Physics, Imperial College London, South Kensington Campus, Exhibition Road,
EP
Address correspondence to: Matilda Katan, E-mail: [email protected]
AC C
*
TE D
London SW7 2AZ, UK
1
ACCEPTED MANUSCRIPT Time-resolved FRET reports FGFR1 dimerization and formation of a complex with
AC C
EP
TE D
M AN U
SC
RI PT
its effector PLCγ1
2
ACCEPTED MANUSCRIPT Abstract
In vitro and in vivo imaging of protein tyrosine kinase activity requires minimally invasive, molecularly precise optical probes to provide spatiotemporal mechanistic information of
RI PT
dimerization and complex formation with downstream effectors. We present here a construct with genetically encoded, site-specifically incorporated, bioorthogonal reporter that can be selectively labelled with exogenous fluorogenic probes to monitor the structure
SC
and function of fibroblast growth factor receptor (FGFR). GyrB.FGFR1KD.TC contains a coumermycin-induced artificial dimerizer (GyrB), FGFR1 kinase domain (KD) and a
M AN U
tetracysteine (TC) motif that enables fluorescent labelling with biarsenical dyes FlAsH-EDT2 and ReAsH-EDT2. We generated bimolecular system for time-resolved FRET (TR-FRET) studies, which pairs FlAsH-tagged GyrB.FGFR1KD.TC and N-terminal Src homology 2 (nSH2) domain of phospholipase Cγ (PLCγ), a downstream effector of FGFR1, fused to mTurquoise
TE D
fluorescent protein (mTFP). We demonstrated phosphorylation-dependent TR-FRET readout of complex formation between mTFP.nSH2 and GyrB.FGFR1KD.TC. By further application of TR-FRET, we also demonstrated formation of the GyrB.FGFR1KD.TC homodimer by
EP
coumermycin-induced dimerization. Herein, we present a spectroscopic FRET approach to facilitate and propagate studies that would provide structural and functional insights for
AC C
FGFR and other tyrosine kinases.
Keywords
fibroblast growth factor receptor; phospholipase Cγ; time-resolved FRET
3
ACCEPTED MANUSCRIPT 1. Introduction
Fibroblast growth factors (FGFs) and their receptors (FGFR1-4) play a critical role in many physiological processes including embryogenesis, wound healing, inflammation and
RI PT
angiogenesis as well as adult tissue homeostasis (Beenken and Mohammadi, 2009). Importantly, aberrant FGFs/FGFRs signaling has been causatively linked to several developmental syndromes and a broad range of human malignancies (Carter et al., 2015;
SC
Helsten et al., 2015). This involvement in the pathology of many cancer types provided a strong rationale for development of effective agents for these targets; consequently, there is
M AN U
a large ongoing effort to develop FGFR inhibitors as anticancer treatments (Touat et al., 2015).
FGFs mediate their diverse biological responses by binding to FGFRs, leading to their
TE D
dimerization and autophosphorylation; despite extensive insights into isolated extracellular and intracellular regions of FGFRs, overall process of dimerization is not clearly defined (Goetz and Mohammadi, 2013; Maruyama, 2014). Subsequent to dimerization, similar to
EP
other receptor tyrosine kinases (RTKs), activated FGFRs dimers recruit and phosphorylate a complement of signaling molecules that mediate the cellular activities of FGFs (Bradshaw et
AC C
al., 2015; Lemmon and Schlessinger, 2010; McCubrey et al., 2015). In case of FGFRs, the two main direct binding proteins that propagate signaling are the adaptor molecule FRS2 and phospholipase Cγ (PLCγ) (Beenken and Mohammadi, 2009; Eswarakumar et al., 2005). PLCγ is recognized as a key component downstream of RTKs involved in dynamic coupling of modifying enzymes with signaling lipids (Blind, 2014; Follo et al., 2015) and, furthermore, has been recently implicated in disease development through gain-of-function mutations (Koss et al., 2014).
4
ACCEPTED MANUSCRIPT Several lines of experimental evidence have demonstrated that FGFR is constitutively bound to FRS2 via FGFR juxtamembrane region, whereas PLCγ is recruited to the C-terminal tail of the receptor upon autophosphorylation of a highly conserved tyrosine (Y766 in case of FGFR1) (Eswarakumar et al., 2005). The main interacting domain in PLCγ is the N-terminal
RI PT
Src-homology 2 domain (nSH2) present within the PLCγ regulatory region (or γ-specific array, γSA); the interaction involves not only the canonical binding site centered on pY766 at the FGFR1 tale, but also a secondary binding site (Bae et al., 2009). Interestingly, the selectivity
SC
of PLCγ binding and signaling via activated FGFR1 are determined by interactions between this secondary binding site on nSH2 domain and a region in FGFR1 kinase domain. Thus,
M AN U
binding of nSH2 from PLCγ1 to two distinct sites on FGFR1 (the tale and kinase domain) represents a highly specific and high affinity (kD about 5 nM) event that is, overall, dependent on activation status of FGFR (Bae et al., 2009; Bunney et al., 2012). Measurements of this interaction using fluorescent reporters and FRET could provide a
TE D
sensitive method to assess activation of FGFR in vitro and in cells or an inhibition of this process in the presence of small molecule drugs.
EP
We here describe generation and characterisation of constructs that provide new tools to assess spatiotemporal mechanistic details of dimerization and downstream signaling. We
AC C
generated GyrB.FGFR1KD.TC construct comprising a coumermycin-induced artificial dimerizer (GyrB), FGFR1 kinase domain (KD) – exhibiting wild-type analogous autophosphorylation, substrate recruitment, and inhibitor response in kinase assays and chromatography measurements – and a tetracysteine (TC) motif that enables fluorescent labelling with biarsenicals FlAsH-EDT2 and ReAsH-EDT2. We conceived systems for TR-FRET studies, which pair either FlAsH- and ReAsH-tagged GyrB.FGFR1KD.TC or pair FlAsH-tagged GyrB.FGFR1KD.TC and nSH2 domain of PLCγ fused to mTurquoise fluorescent protein (mTFP). We demonstrated coumermycin-induced dimerization and phosphorylation-
5
ACCEPTED MANUSCRIPT dependent
TR-FRET
readout
of
complex
formation
between
mTFP.nSH2
and
GyrB.FGFR1KD.TC.
RI PT
2. Material and Methods
2.1 TR-FRET
SC
For fluorescence lifetime decay measurements in a quartz cuvette using a TR-FRET multidimensional spectrofluorometer. GyrB.FGFR1KD.TC (1 μM) tagged with FlAsH-EDT2
M AN U
(5μM and 50 μM) and its partner mTFP.nSH2 (1 μM) in 40 mM Tris-HCl (pH8.0), 20 mM MgCl2, 20 mM NaCl, 100 μM TCEP, and 100 μM Na3VO4 were incubated in the presence or absence of ATP (50μM) in at ambient conditions for 30 minutes. Excitation of the FRET pair was performed at 410 – 425 nm using the supercontinuum source, and fluorescence
TE D
emission detected at 430 – 520 nm (10 nm steps) in order to record the decrease in fluorescence lifetime of the donor mTFP. Analysis of lifetime and anisotropy decays was performed using TRFA data processor (Scientific Software Technologies Center, Minsk,
EP
Belarus).
AC C
Details for cloning, expression and purification of recombinant proteins, kinase assays, in vitro labelling of purified proteins and Supplementary Figures can be found in Supplementary Information.
6
ACCEPTED MANUSCRIPT 3. Results
3.1 Constructs for a TR-FRET approach to detect FGFR1 dimerization and complex
RI PT
formation with PLCγ
Accumulating evidence show that ligand-induced dimerization of RTK extracellular regions leads to activation of the intracellular KD, which, in turn, alters complex downstream
SC
signalling networks (Lemmon and Schlessinger, 2010). However, intact RTKs are difficult to generate and reconstruct into in vitro signaling systems. In order to study the molecular
M AN U
interactions of the FGFR1 in vitro, a dimerizer was incorporated at the N-terminus and a fluorescent self-labelling peptide sequence at the C-terminus of FGFR1KD (Figure 1 A, top). Our previous work (Bunney et al., 2012; Bunney et al., 2015) involved the construction of a synthetic open reading frame (ORF) encoding for a mutant form of human FGFR1 kinase
TE D
domain (KD), FGFR1KD [amino acids 457–774, (Y463, 583, 585F), (L457V, C488A, C584S)] that forms a central part of our new construct (Figure1 A, top). FGFR1 dimerization could be monitored in vitro by using an artificial dimerization motif at the N-terminus of the KD to
EP
mimic the extracellular portion of the wild-type protein. For this purpose, the N-terminal 24kDa subdomain of the B subunit of bacterial DNA gyrase (GyrB), which dimerizes in the
AC C
presence of the antibiotic coumermycin (Farrar et al., 1996; Liu et al., 2003), was incorporated into the construct, generating GyrB.FGFR1KD. Using recombinant cloning methods, an optimised TC motif (FLNCCPGCCMEP) (Spagnuolo et al., 2006) was introduced at the C-terminus of GyrB.FGFR1KD to generate GyrB.FGFR1KD.TC (Figure 1A, top). The protein was then isolated from transformed E.coli strain C41(DE3). The optimised FlAsH- or ReAsH- binding TC motif has been previously used to monitor protein structure in vitro and in mammalian cells (Luedtke et al., 2007; Martin et al., 2005; Roberti et al., 2007; Spille et al., 2011).
7
ACCEPTED MANUSCRIPT
The high affinity of the TC motif to the biarsenical dye (Adams and Tsien, 2008) suggests that the complex should be stable under typical denaturing conditions. The identity and integrity of TC-tagged protein was verified by survival of fluorescence after SDS-PAGE. The
RI PT
recombinant protein was incubated with FlAsH-EDT2 for 30 min at room temperature in a modified sample buffer for SDS-PAGE in which the typical thiol reductant (2mercaptoethanol) was substituted by tris(2-carboxy-ethyl)phosphine (TCEP). After
SC
electrophoresis, a single major fluorescent species running at the anticipated molecular weight for the GyrB.FGFR1KD.TC (60.1 kDa) was visible by UV illumination (Figure 1B). Excess
M AN U
FlAsH-EDT2, which could result in increased levels of background fluorescence, is removed as it migrates with the dye front. Subsequent staining with Coomassie Blue confirmed the specific binding of dye to the TC-containing protein. This result suggests that GyrB.FGFR1KD.TC tagged with FlAsH-EDT2 (or ReAsH-EDT2) can be studied in a mixture of
TE D
proteins or in a crude lysate. The identity of the fully labelled protein was further confirmed by native mass spectroscopy (Figure 1C), in which the GyrB.FGFR1KD.TC/FlAsH-EDT2 complex was clearly observed as a single peak at the anticipated molecular weight (found: 62543.0,
EP
calc.: 62539.0). Based on these data, labelling of GyrB.FGFR1KD.TC is suitable for further
AC C
FRET experiments to test dimerization.
As a second model system for RT-FRET experiments, the FGFR1 KD and its downstream target PLCγ were used; it has been previously established that a stable complex is formed between the C-terminus of activated FGFR1KD and two different sites on the surface of the nSH2 from PLCγ regulatory region (Figure 1A, bottom) (Bae et al., 2009). The canonical phosphorylation-dependent binding site consists of a positively charged patch on the surface of the nSH2 domain that binds to pY766 and a hydrophobic pocket in the FGFR1 C-terminal tail. A second binding site of the nSH2 forms mediates high affinity binding by
8
ACCEPTED MANUSCRIPT forming contacts with the backside of the C-lobe of the KD away from the activation segment of FGFR1. The co-crystal structure of nSH2 of PLCγ with activated FGFR1KD (PDB code: 3GQI) (Bae et al., 2009), estimates a distance of 28.7 Å between nSH2 (N-terminus) and FGFR1KD (C-terminus), making the protein interaction measurable by FRET. As a FRET
RI PT
donor, cyan fluorescent protein (CFP) variant mTurquoise (mTFP) (excitation: 434 nm,
bottom).
SC
emission: 474 nm) fused to the nSH2 domain of PLCγ (mTFP.nSH2) was used (Figure 1A,
3.2 Kinase profiling of GyrB.FGFR1KD.TC demonstrates that its biochemical properties
M AN U
match those of the wild-type protein in vitro
To determine the in vitro tyrosine kinase activity of recombinant GyrB.FGFR1KD.TC, a bioluminescent, ADP detection assay was used (Zegzouti et al., 2009). The kinase substrates
TE D
used here were PolyE4Y1 and a part of PLCγ regulatory region, which comprises the two SH2 domains of PLCγ followed by a segment containing Y783 (Figure 1A, bottom). Inhibitors used here were PD173074, an established ATP-competitive inhibitor of FGFR1 (inhibitor constant,
EP
Ki: ~ 44 nM, IC50: ~ 22 nM) (Mohammadi et al., 1998) and AZD4547, a selective inhibitor of FGFR1 currently in clinical trials (Ki: 6.1 nM, IC50: 0.2 nM) (Bunney et al., 2015; Gavine et al.,
AC C
2012).
To optimise assay conditions, we first determined the ATP concentration that equals the Km value for GyrB.FGFR1KD.TC (Figure 2A). The data were fitted to the Michaelis-Menten equation, thereby determining the GyrB.FGFR1KD.TC ATP Km to be 126 μΜ. After identifying appropriate substrates and an optimal ATP concentration, the dependence of signal linearity to kinase concentration was established in a series of GyrB.FGFR1KD.TC titrations with PolyE4Y1 (Figure 2B) and PLCγ SA WT (Supplementary Figure S3A). Here, we have chosen to
9
ACCEPTED MANUSCRIPT use 0.3 nM and 30 nM of GyrB.FGFR1KD.TC with inhibitors AZD4547 and PD173074 respectively, to guarantee sufficiently high assay signal as well as substrate conversion below 70%. In the last step of the optimisation process, the contribution of the autophosphorylation state of the kinase on the total assay signal was examined in the
(Figure
2C),
resulting
in
higher
levels
signal
readout
corresponding
to
SC
autophosphorylation.
of
RI PT
absence of substrate. Increasing enzyme concentration were titrated with 150 μΜ ATP
In order to validate the optimised GyrB.FGFR1KD.TC assay, the kinase activity was quantified
M AN U
with increasing concentrations of inhibitor AZD4547 (Figure 2D) and reference inhibitor PD173074 (Supplementary Figure S3B). For AZD4547, the IC50 value of 0.32 nM was in good agreement with the published value of 0.20 nM. For PD173074, the IC50obs was 13 nM, an estimation that diverges slightly from the reported value of ~22 nM. This may be ascribed to
TE D
dissimilarities between the assay format used here and the one in literature, where a radiometric readout assay format and full length FGFR1 were used to measure the IC50
EP
values.
3.3 Coumermycin-induced dimerization of FlAsH and ReAsH labelled GyrB.FGFR1KD.TC
AC C
proteins results in a FRET readout
Extracellular dimerization of FGFRs involves the formation of 2:2:2 FGF:FGFR:heparin ternary complex. The recombinant proteins used in this study contain the intracellular KD activity domain, which is suitable for enzyme activity assays, but do not contain the extracellular domain of FGFR1. Attempts to get multi-domain FGFR constructs were not pursued as they show low expression levels in E.coli, due to inadequate solubility and misfolding.
10
ACCEPTED MANUSCRIPT To achieve artificial dimerization of GyrB.FGFR1KD.TC, a model involving FGF ligand-mimic coumermycin, that induces dimerization of GyrB, was used. Coumermycin binds GyrB at a stoichiometric ratio 1:2, whereas another antibody, novobiocin, binds GyrB at 1:1 ratio (Ali et al., 1993) and was thereby suitable as a control of obtaining exclusively monomeric
RI PT
protein population. To first test this dimerization model biochemically, analytical sizeexclusion chromatography (SEC) was used before and following incubation with antibiotics. Samples incubated with coumermycin (GyrB to antibiotic molar ratio, 1.5:1) exhibited major
SC
peaks that corresponded to the presence of predominantly the dimer (Supplementary Figure S4B). The retention time of the coumermycin-bound dimer was equivalent to the retention
(Supplementary Figure S1A).
M AN U
time of species of 120 kDa when compared to SEC standard molecular weight markers
Coumermycin-induced dimerization was subsequently investigated by TR-FRET. In these
TE D
studies FlAsH-EDT2 is used as the donor to fluorescently tag a population of GyrB.FGFR1KD.TC while red-emitting biarsenical ReAsH-EDT2 (excitation: 593 nm, emission: 608 nm), which binds to the C-terminal TC motif of GyrB.FGFR1KD.TC (Supplementary Figure
EP
S2), is issued as the acceptor to tag another population of the same protein (Figure 3A). The suitability of FlAsH-EDT2 and ReAsH-EDT2 as a donor acceptor pair was demonstrated in
AC C
vitro by bulk spectrofluorimetry (Figure 3B), which highlighted that the two dyes have ideal
spectral overlap for FRET to occur. The change in donor emission in the presence and absence of coumermycin was analysed by fitting the decays using single exponential models (Figure 3C). Addition of coumermycin to the mixture of proteins tagged for FlAsH-EDT2 and ReAsH-EDT2 resulted in a decrease in lifetime of ~60 ps for the sample (Figure 3D). Thus, under conditions of dimerization, the overall dye proximity increases resulting in faster lifetime decay of the donor and FRET. Since this is triggered by addition of coumermycin to
11
ACCEPTED MANUSCRIPT the sample in solution, FRET can be attributed to the formation of dimerization complexes between FlAsH-tagged proteins and ReAsH-tagged proteins.
3.4.
TR-FRET
detection
of
the
complex-formation
between
FlAsH-labelled
RI PT
GyrB.FGFR1KD.TC and mTFP.nSH2
For TR-FRET measurements, GyrB.FGFR1KD.TC was reacted with FlAsH-EDT2 and then
SC
incubated with mTFP.nSH2 for 30 min at room temperature (Figure 4A). The suitability of FlAsH-EDT2 as a donor and mTFP as an acceptor was demonstrated in vitro by bulk
M AN U
spectrofluorimetry (Figure 4B). Next, fluorescence lifetime decays of the samples were recorded in the presence and absence of ATP. To interpret these data, exponential decay models were fitted to the decays; it was apparent that both mTFP.nSH2 by itself and the protein pair were well fit to a single exponential decay (Figure 4C). Addition of ATP to the
TE D
incubated protein pair resulted in a large decrease in lifetime of ~500 ps for the sample (Figure 4D). The fluorescence decay observed reflects the increase in spatial proximity of the donor and acceptor fluorophores, which was attributed to direct interaction between the
EP
nSH2 domain of PLCγ and activated FGFR1KD.
AC C
The interaction between GyrB.FGFR1KD.TC and mTFP.nSH2 was also confirmed using SEC; the incubated proteins eluted at the same retention time (elution volume 14.7 mL) indicating formation of a GyrB.FGFR1KD.TC–mTFP.nSH2 complex (Supplementary Figure S4A).
12
ACCEPTED MANUSCRIPT 4. Discussion
We here demonstrate, using a TR-FRET approach, the suitability of GyrB dimerizer for studies of FGFR signaling and application of bimolecular mTFP/FlAsH.TC system for
RI PT
monitoring the phosphorylation-dependent complex formation between the nSH2 domain of PLCγ and functional GyrB.FGFR1KD.TC.
SC
Use of non-physiological dimerization elements has been widely used to study signal transduction cascades triggered by dimerization (Fegan et al., 2010). This approach has also
M AN U
been applied to FGFR; for example, a construct incorporating FKBPv at the C-terminus of the FGFR KD and dimerization in the presence of a drug (AP20187) provided a basis for specific, inducible and ligand independent activation of FGFR1 signalling in cells and mouse models (Acevedo et al., 2007; Freeman et al., 2003; Tomlinson et al., 2012; Welm et al., 2002). We
TE D
here demonstrate that GyrB allows not only previously observed dimerization in cells (Liu et al., 2003), but also generation of purified, functional FGFR-GyrB fusion proteins (Figures 1 and 3) for further structural and in vitro studies of elements critical for dimerization and
EP
high-throughput screens for disruption of this key activation step.
AC C
Interactions between RTKs and SH2 domains of downstream effectors have been explored for their potential to report signaling events in cells and thus provide a platform for drug discovery that would overcome one of the common problems based on kinase assays in vitro: the identification of potent hits in vitro with poor cellular activity. One recent EGFR biosensor that allows measurements of EGFR clustering as a readout of activation in cells is based on binding of the green fluorescent protein fused to two tandem SH2 domains from adapter protein Grb2; it was subsequently shown that this system provides an important high throughput system for drug discovery performed in a cellular context (Antczak et al.,
13
ACCEPTED MANUSCRIPT 2012). We here suggest that the highly specific, high affinity interaction between FGFR1 and nSH2 domain from PLCγ (Bae et al., 2009; Bunney et al., 2012) can similarly be used in drug discovery. We generated a FRET pair for TR-FRET detection of the complex-formation between FlAsH-labelled GyrB.FGFR1KD.TC and mTFP.nSH2 (Figure 4). The decrease in
RI PT
fluorescence lifetime of mTFP depends exclusively on its molecular proximity to the TC motif; hence, FRET occurrence can be attributed to binding of mTFP.nSH2 to activated GyrB.FGFR1KD.TC. The dramatic decrease in fluorescent lifetime of the donor may be
SC
ascribed to the high affinity binding of the nSH2 domain to the activated FGFR1KD. In vitro experiments with isolated FRET pairs have provided a straightforward readout of the
M AN U
protein-protein interaction and can be adapted for screening of inhibitors that directly affect kinase activity as well as interaction between FGFR and PLCγ. Furthermore, this approach can be developed for cellular FLIM-FRET. Using constructs that ensure membrane localization of GyrB.FGFR1KD.TC and intracellular labelling of this protein in the presence of
TE D
FlAsH, subsequent activation in response to addition of coumermycin will result in binding of
EP
the fluorescent fusion protein mTFP.nSH2 expressed in the same cell and the FRET readout.
5. Acknowledgments
AC C
The MK laboratory is supported by Cancer Research UK (A16567) and LP by Imperial College London Doctoral Training Centre.
6. References Acevedo, V. D., Gangula, R. D., Freeman, K. W., Li, R., Zhang, Y., Wang, F., Ayala, G. E., Peterson, L. E., Ittmann, M., and Spencer, D. M., 2007. Inducible FGFR-1 activation leads to
14
ACCEPTED MANUSCRIPT irreversible prostate adenocarcinoma and an epithelial-to-mesenchymal transition. Cancer cell 12, 559-571. Adams, S. R., and Tsien, R. Y., 2008. Preparation of the membrane-permeant biarsenicals FlAsH-EDT2 and ReAsH-EDT2 for fluorescent labeling of tetracysteine-tagged proteins.
RI PT
Nature protocols 3, 1527-1534. Ali, J. A., Jackson, A. P., Howells, A. J., and Maxwell, A., 1993. The 43-kilodalton N-terminal fragment of the DNA gyrase B protein hydrolyzes ATP and binds coumarin drugs.
SC
Biochemistry 32, 2717-2724.
Antczak, C., Bermingham, A., Calder, P., Malkov, D., Song, K., Fetter, J., and Djaballah, H.,
M AN U
2012. Domain-based biosensor assay to screen for epidermal growth factor receptor modulators in live cells. Assay and drug development technologies 10, 24-36. Bae, J. H., Lew, E. D., Yuzawa, S., Tome, F., Lax, I., and Schlessinger, J., 2009. The selectivity of receptor tyrosine kinase signaling is controlled by a secondary SH2 domain binding site. Cell
TE D
138, 514-524.
Beenken, A., and Mohammadi, M., 2009. The FGF family: biology, pathophysiology and therapy. Nature reviews Drug discovery 8, 235-253.
EP
Blind, R. D., 2014. Disentangling biological signaling networks by dynamic coupling of signaling lipids to modifying enzymes. Advances in biological regulation 54, 25-38.
AC C
Bradshaw, R. A., Pundavela, J., Biarc, J., Chalkley, R. J., Burlingame, A. L., and Hondermarck, H., 2015. NGF and ProNGF: Regulation of neuronal and neoplastic responses through receptor signaling. Advances in biological regulation 58, 16-27. Bunney, T. D., Esposito, D., Mas-Droux, C., Lamber, E., Baxendale, R. W., Martins, M., Cole, A., Svergun, D., Driscoll, P. C., and Katan, M., 2012. Structural and functional integration of the PLCgamma interaction domains critical for regulatory mechanisms and signaling deregulation. Structure 20, 2062-2075.
15
ACCEPTED MANUSCRIPT Bunney, T. D., Wan, S., Thiyagarajan, N., Sutto, L., Williams, S. V., Ashford, P., Koss, H., Knowles, M. A., Gervasio, F. L., Coveney, P. V., and Katan, M., 2015. The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study. EBioMedicine 2, 194-204.
RI PT
Carter, E. P., Fearon, A. E., and Grose, R. P., 2015. Careless talk costs lives: fibroblast growth factor receptor signalling and the consequences of pathway malfunction. Trends in cell biology 25, 221-233.
SC
Eswarakumar, V. P., Lax, I., and Schlessinger, J., 2005. Cellular signaling by fibroblast growth factor receptors. Cytokine & growth factor reviews 16, 139-149.
M AN U
Farrar, M. A., Alberol-Ila, J., and Perlmutter, R. M., 1996. Activation of the Raf-1 kinase cascade by coumermycin-induced dimerization. Nature 383, 178-181. Fegan, A., White, B., Carlson, J. C., and Wagner, C. R., 2010. Chemically controlled protein assembly: techniques and applications. Chemical reviews 110, 3315-3336.
TE D
Follo, M. Y., Manzoli, L., Poli, A., McCubrey, J. A., and Cocco, L., 2015. PLC and PI3K/Akt/mTOR signalling in disease and cancer. Advances in biological regulation 57, 10-16. Freeman, K. W., Welm, B. E., Gangula, R. D., Rosen, J. M., Ittmann, M., Greenberg, N. M., and
EP
Spencer, D. M., 2003. Inducible prostate intraepithelial neoplasia with reversible hyperplasia in conditional FGFR1-expressing mice. Cancer research 63, 8256-8263.
AC C
Gavine, P. R., Mooney, L., Kilgour, E., Thomas, A. P., Al-Kadhimi, K., Beck, S., Rooney, C., Coleman, T., Baker, D., Mellor, M. J., et al., 2012. AZD4547: an orally bioavailable, potent, and selective inhibitor of the fibroblast growth factor receptor tyrosine kinase family. Cancer research 72, 2045-2056.
Goetz, R., and Mohammadi, M., 2013. Exploring mechanisms of FGF signalling through the lens of structural biology. Nature reviews Molecular cell biology 14, 166-180.
16
ACCEPTED MANUSCRIPT Helsten, T., Schwaederle, M., and Kurzrock, R., 2015. Fibroblast growth factor receptor signaling in hereditary and neoplastic disease: biologic and clinical implications. Cancer metastasis reviews. Koss, H., Bunney, T. D., Behjati, S., and Katan, M., 2014. Dysfunction of phospholipase
RI PT
Cgamma in immune disorders and cancer. Trends in biochemical sciences 39, 603-611. Lemmon, M. A., and Schlessinger, J., 2010. Cell signaling by receptor tyrosine kinases. Cell 141, 1117-1134.
SC
Liu, G., Bafico, A., Harris, V. K., and Aaronson, S. A., 2003. A novel mechanism for Wnt activation of canonical signaling through the LRP6 receptor. Molecular and cellular biology
M AN U
23, 5825-5835.
Luedtke, N. W., Dexter, R. J., Fried, D. B., and Schepartz, A., 2007. Surveying polypeptide and protein domain conformation and association with FlAsH and ReAsH. Nature chemical biology 3, 779-784.
TE D
Martin, B. R., Giepmans, B. N., Adams, S. R., and Tsien, R. Y., 2005. Mammalian cell-based optimization of the biarsenical-binding tetracysteine motif for improved fluorescence and affinity. Nature biotechnology 23, 1308-1314.
EP
Maruyama, I. N., 2014. Mechanisms of activation of receptor tyrosine kinases: monomers or dimers. Cells 3, 304-330.
AC C
McCubrey, J. A., Abrams, S. L., Fitzgerald, T. L., Cocco, L., Martelli, A. M., Montalto, G., Cervello, M., Scalisi, A., Candido, S., Libra, M., and Steelman, L. S., 2015. Roles of signaling pathways in drug resistance, cancer initiating cells and cancer progression and metastasis. Advances in biological regulation 57, 75-101. Mohammadi, M., Froum, S., Hamby, J. M., Schroeder, M. C., Panek, R. L., Lu, G. H., Eliseenkova, A. V., Green, D., Schlessinger, J., and Hubbard, S. R., 1998. Crystal structure of an angiogenesis inhibitor bound to the FGF receptor tyrosine kinase domain. The EMBO journal 17, 5896-5904.
17
ACCEPTED MANUSCRIPT Roberti, M. J., Bertoncini, C. W., Klement, R., Jares-Erijman, E. A., and Jovin, T. M., 2007. Fluorescence imaging of amyloid formation in living cells by a functional, tetracysteinetagged alpha-synuclein. Nature methods 4, 345-351. Spagnuolo, C. C., Vermeij, R. J., and Jares-Erijman, E. A., 2006. Improved photostable FRET-
American Chemical Society 128, 12040-12041.
RI PT
competent biarsenical-tetracysteine probes based on fluorinated fluoresceins. Journal of the
Spille, J. H., Zurn, A., Hoffmann, C., Lohse, M. J., and Harms, G. S., 2011. Rotational diffusion
SC
of the alpha(2a) adrenergic receptor revealed by FlAsH labeling in living cells. Biophysical journal 100, 1139-1148.
M AN U
Tomlinson, D. C., Knowles, M. A., and Speirs, V., 2012. Mechanisms of FGFR3 actions in endocrine resistant breast cancer. International journal of cancer Journal international du cancer 130, 2857-2866.
Touat, M., Ileana, E., Postel-Vinay, S., Andre, F., and Soria, J. C., 2015. Targeting FGFR
TE D
Signaling in Cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 21, 2684-2694.
Welm, B. E., Freeman, K. W., Chen, M., Contreras, A., Spencer, D. M., and Rosen, J. M., 2002.
EP
Inducible dimerization of FGFR1: development of a mouse model to analyze progressive transformation of the mammary gland. The Journal of cell biology 157, 703-714.
AC C
Zegzouti, H., Zdanovskaia, M., Hsiao, K., and Goueli, S. A., 2009. ADP-Glo: A Bioluminescent and homogeneous ADP monitoring assay for kinases. Assay and drug development technologies 7, 560-572.
18
ACCEPTED MANUSCRIPT Figure Legends
Figure 1. Representation of the constructs used in dimerization and complex formation of FGFR1KD
RI PT
(A) Representation of the wild type FGFR1 and PLCγ as well as domains of protein constructs used in this study. FGFR1 (top) comprises Ig-like domains 1, 2, and 3 (D1, 2, and 3, respectively), trans-membrane domain (TM) and intracellular kinase domain (KD). KD (with
SC
mutations shown as yellow dots) is incorporated in GyrB.FGFR1KD.TC that also includes an N-terminal GyrB and a C-terminal TC motif. PLCγ (bottom) contains domains common to PLC
M AN U
family: N-terminal pleckstrin homology domain (PH), EF–hands (EF), catalytic domain (PLC, X-box and Y-box, shown as two parts) and C2 calcium/lipid-binding domain (C2). PLCγ Specific Array (PLCγ SA) comprises a split PH domain (PH), two SH2 domains (nSH2 and cSH2) and the SH3 domain. mTurquoise.nSH2 fusion protein (mTFP.nSH2) comprises an
TE D
mTurquoise (mTFP) and an nSH2 domain.
(B) Specific and quantitative labelling of GyrB.FGFR1KD.TC upon incubation with FlAsH-EDT2 demonstrated by SDS-PAGE (Coomassie staining) (top) and in-gel fluorescence (bottom).
EP
(C) Mass spectrum of the full length intact GyrB.FGFR1KD.TC before (left) and after (right)
AC C
labelling with FlAsH-EDT2.
Figure 2. Kinase profiling of GyrB.FGFR1KD.TC by bioluminescent ADP detection assays. (A) Determination of ATP Km for GyrB.FGFR1KD.TC by titrating with 10 – 240 μM ATP and 0.2 μg/μl PolyE4Y1.
(B) Examination of the dependence of signal linearity to kinase concentration by titration of 0 – 1.1 μM GyrB.FGFR1KD.TC with 150 μM ATP and 0.2 μg/μl PolyE4Y1. (C) Autophosphorylation assay at varying concentrations of 0 – 1.1 μM enzyme and 150 μM ATP.
19
ACCEPTED MANUSCRIPT (D) Inhibitor dose response titration of GyrB.FGFR1KD.TC in the presence of 150 μM ATP, 0.2 μg/μl PolyE4Y1 peptide and inhibitor AZD4547 at varying concentrations of 0.002 – 20 nM (FGFR1 IC50 = 0.2 nM). Kinase inhibition was measured by detecting the decrease in phosphorylation of PolyE4Y1 after 1 hour. Maximal activity of GyrB.FGFR1KD.TC and the
RI PT
background signal were measured in the absence of inhibitor AZD4547. GyrB.FGFR1KD.TC activity was expressed by calculating the ratio between the background-corrected assay signals in the absence and presence of the indicated inhibitor concentrations. Curve fitting
SC
for inhibitor dose response titration was performed using GraphPad Prism® sigmoidal doseresponse software.
M AN U
In all panels error bars indicate the standard deviations (SDs) of three replicates.
Figure 3. Coumermycin-induced dimerization of GyrB.FGFR1KD.TC observed by FRET
induced dimerization.
TE D
(A) Representation of the construct used for the TR-FRET investigating coumermycin-
(B) Spectral overlap as precondition for FRET. The spectral overlap integral is indicated by
EP
the purple area. The region of overlap between the excited donor's emission spectrum (green line) and the acceptor's absorption spectrum (orange line) has to be significant for
AC C
FRET to occur.
(C) Single exponential fit of the fluorescence decay (normalised photon count) over time from fluorescence lifetime decay measurements in a quartz cuvette using a TR-FRET multidimensional spectrofluorometer. (D) Donor lifetime decrease upon dimerization showing statistical significance assessed using data from a minimum of three repeats by a two-tailed unpaired t-test (***0.0001
20
ACCEPTED MANUSCRIPT Figure 4. Complex formation between GyrB.FGFR1KD.TC and mTFP.nSH2 observed by FRET (A) Representation of the constructs used for the TR-FRET examining complex formation. (B) Spectral overlap (purple area) between the excited donor's emission spectrum (green line) and the acceptor's absorption spectrum (orange line).
from fluorescence lifetime decay measurements.
RI PT
(C) Single exponential fit of the fluorescence decay (normalised photon count) over time
(D) Donor lifetime decrease upon complex formation showing statistical significance
SC
assessed using data from a minimum of three repeats by a two-tailed unpaired t-test
AC C
EP
TE D
M AN U
(*0.0001
A
B
GyrB.FGFR1KD.TC
FGFR1
FlAsH-EDT2 (μM) 5
10
50 100
-
kDa 75
D1
D2
D3
TM
KD
ACCEPTED MANUSCRIPT 50
GyrB.FGFR1KD.TC
Y653
464 GyrB
Y654
Y766
FlAsH-EDT2 37
774
KD
TC
N
75
C
C488A Y583,585F,C584S
37
C
PLCγSA PH
EF
PLC
P
nSH2
cSH2
SH3 H
PLC
C2
550 nSH2
100
GyrB.FGFR1KD.TC/FlAsH found:62543.0 calc.:62539.0
60000
62500
0 65000 60000 Mass, m/z
62500
65000 Mass, m/z
EP
TE D
0
AC C
mTFP
657
M AN U
mTFP.nSH2
% Intensity
100
GyrB.FGFR1KD.TC found:62065.0 calc.:62061.0
SC
PLCγ
RI PT
50
Figure 1
A
B
6000
4000
3000 MANUSCRIPT ACCEPTED
4000 3000 2000
Km, ATP: 126.3 μM
1000
C
0
50
100
150
200
2000
1000
0
250
ATP, μM
0.01
0.1 log [GyrB.FGFR1KD.TC], μM
D
150
350
100 50
0.1
1
50
0 -3
TE D
log [GyrB.FGFR1KD.TC], μM
EP
0
SC
150
100
M AN U
Luminescence, RLU
200
AC C
Luminescence, RLU
300 250
1
RI PT
0
Luminescence, RLU
Luminescence, RLU
5000
-2
0 -1 log [AZD4547], nM
IC50 = 0.32
1
Figure 2
2
A
B
Y654
Tyr Kinase N C488A Y583,585F,C584S
400
D 1
Fluorescent lifetime (ps)
GyrB.FGFR1KD.TC - dimeriser + dimeriser
0.5
20 30 Decay, ns
40
50
700
4380 4360 4340 4320 4300
Coumermycin (dimeriser)
GyrB.FGFR1KD.TC -
+
M AN U
10
500 600 Wavelength, nm
4400
TE D
0
EP
0
AC C
Normalised photon count
C
0
RI PT
GyrB
Y653
FlAsH-EDT2 emission ReAsH-EDT2 excitation
SC
464
Relative Intensity units
GyrB.FGFR1KD.TC
FlAsH-EDT2 1000 or ReAsH-EDT2 750 Y766 ACCEPTED MANUSCRIPT 774 500 TC 250 C
Figure 3
B
GyrB.FGFR1KD.TC
Y653
464
Y654
Y766
774
N
C
C488A Y583,585F,C584S mTFP.nSH2
550
1
P 0.5
400 500 600 Wavelength, nm
700
50
4000 3800 3600 3400 GyrB.FGFR1KD.TC with mTFP.nSH2 ATP +
SC
40
4200
TE D
M AN U
20 30 Decay, ns
EP
10
0 300
D Fluorescent lifetime (ps)
GyrB.FGFR1KD.TC + mTFP.nSH2
0
250
nSH2
N
0
500
657
AC C
Normalised photon count
C
750
RI PT
mTFP
mTFP emission FlAsH-EDT2 excitation
1000
ACCEPTED MANUSCRIPT TC
Tyr Kinase
GyrB
FlAsH-EDT2
Relative Intensity units
A
Figure 4
ACCEPTED MANUSCRIPT
Supplementary material
•
Supplementary Experimental Procedures
•
Supplementary References
•
Supplementary Figure Legends
•
Supplementary Figures S1, S2, S3 and S4
AC C
EP
TE D
M AN U
SC
RI PT
Table of contents
ACCEPTED MANUSCRIPT Supplementary Experimental Procedures Expression of recombinant proteins For expression of His6.Stag.3C.GyrB.FGFR1KD.TC {which encodes N-‐terminally hexahistidine tag, S-‐tag,
RI PT
3C protease recognition sequence, DNA gyrase B (GyrB) domain, FGFR1KD-‐3F-‐0P [457–774, (Y463, 583, 585F) (L457V, C488A, C584S)] and fluorescein arsenical hairpin binder (FlAsH) amino acid sequence (FLNCCPGCCMEP)} from pTriExTM4 (Novagen) vector backbone. Constructs were heat-‐shock transformed into E.coli strain C41 (DE3) with pTriExTM4-‐ His6.Stag.3C.GyrB.FGFR1KD.TC and pDUET-‐Cdc37-‐λPPase (which encodes lambda protein phosphatase (λPPase), an enzyme that hydrolyses phosphorylated
SC
tyrosines and Hsp90 co-‐chaperone Cdc37, which promotes the stability of kinase domains). Cells were recovered in 1 ml of SOC media for 1 h at 37 ºC, and plated on terrific broth (TB) agar plates
M AN U
supplemented with ampicillin (50 μg/mL), spectinomycin (50 μg/mL) and 10 mM glucose. Colonies were inoculated into 500 ml of TB with ampicillin (50 μg/mL) and spectinomycin (50 μg/mL) and grown to an OD600 of 1.0. Cultures were cooled to 15°C and expression was induced with 100 μM IPTG for approximately 16 hours. Cells were harvested by centrifugation and frozen at -‐80 ºC until processed.
600 μL of Ni2+-‐NTA beads (Qiagen) were added to the extract and the mixture was incubated with
TE D
agitation for 1 h at 4 ºC. Beads were collected by centrifugation (10 min, 1000 g). The beads were three times resuspended in 30 mL wash buffer (20 mM Tris-‐HCl, 30 mM imidazole, 300 mM NaCl, pH 8.0) and spun down at 1000g. Subsequently, the beads were resuspended in 10 mL of wash buffer and transferred to a column. The protein was eluted with 3 ml of wash buffer supplemented with 200 mM
EP
imidazole and further purified by size-‐exclusion chromatography employing a HiLoad 16/60 Superdex 75 Prep Grade column (GE Life Sciences) at a flow rate of 1 mL/min (buffer: 20 mM Tris-‐HCl, 100 mM NaCl,
AC C
2 mM EDTA, pH 7.4). Fractions containing the protein were pooled and concentrated with an Amicon Ultra-‐15 3 kDa MWCO centrifugal filter device (Millipore). Purified proteins were analyzed by 10% SDS-‐ PAGE and their mass confirmed by mass spectrometry (see Supplementary Information).
Purification of recombinant proteins For purification of protein domains expressed in C41 (DE3) or BL21 (DE3). Pellets derived from 1 L cultures were thawed on ice and resuspended in 20 ml of chilled lysis buffer (25 mM Tris-‐HCl, 250 mM NaCl, 40 mM imidazole, 10 mM benzamidine, 100 μg/ml lysozyme, pH 8.0). Resuspension was accomplished by placing the pellets on an orbital shaker at 150 rpm at 4 °C for 30 minutes. A solution of 10% (v/v) Triton-‐X-‐100 and 1 K unitz of bovine pancreatic DNAse I was added and the lysate was placed
ACCEPTED MANUSCRIPT on the orbital shaker at 150 rpm at 4°C for 1 hour. Cell lysates were clarified by centrifugation at 13,000 rpm at 4°C for 1 hour in an SS34 rotor (Sorvall). The clarified lysate was loaded onto a 5-‐ml HisTrap (HiTrap Chelating HP, GE Healthcare) on an Akta Explorer system (GE Healthcare) using His Buffer A (25 mM Tris-‐HCl, 500 mM NaCl, 40 mM Imidazole, 1 mM TCEP, pH 8.0). Non-‐specific binding of other proteins was removed by washing the column with 10 column volumes of His Buffer A. His-‐tagged
RI PT
proteins were collected by eluting with linear imidazole gradient from His Buffer A to His Buffer B (25 mM Tris-‐HCl, 500 mM NaCl, 500 mM imidazole, 1 mM TCEP, pH 8.0) over 5 column volumes. Protein concentrations from the eluent were measured with a Nanodrop (ThermoScientific). The pooled fractions containing the target protein were further incubated with 10 μg of protease per 1 mg of protein (Ulp1 protease for His6-‐Sumo, and 3C protease for His6-‐3C) overnight at 4◦C to cleave of the N-‐
SC
terminal fusion tag. The protein/protease mix was dialysed overnight at 4 °C against 500 ml of Dialysis Buffer (25 mM Tris-‐HCl, 150 mM NaCl, 10 mM Imidazole and 1 mM TCEP, pH 8.0). Separation of the
M AN U
cleaved tag was achieved by reverse HisTrap chromatography over the 5 ml HisTrap column and material that did not bind was collected. Subsequently, proteins were further purified by strong quaternary ammonium (Q) anion exchange chromatography on a 5 ml HiTrap Q column (GE Healthcare) in Q Buffer A (25 mM Tris-‐HCl, 20 mM NaCl, 1 mM TCEP, pH 8.0) and eluted with a linear gradient of 10– 20 column volumes to 50% of Q Buffer B (25 mM Tris-‐HCl, 1 M NaCl, 1 mM TCEP, pH 8.0). Fractions containing the target protein were pooled, concentrated in spin concentrators (Vivaproducts) and then
TE D
further purified by size exclusion chromatography on a Superdex 200 10/300 GL column (GE Healthcare) using Gel Filtration Buffer (25 mM Tris-‐HCl, 150 mM NaCl, 1 mM TCEP, pH 8.0). Fractions were collected, combined and concentrated in spin concentrators, snap frozen in liquid nitrogen and stored at -‐80 °C. Purified proteins were analyzed by 10% SDS-‐PAGE.
EP
Cloning, expression and purification of recombinant proteins
AC C
PCR reactions were carried out on G-‐STORM thermal cycler, followed by digests using DpnI (NEB). The final reaction was transformed into E.coli XL10-‐Gold (Stratagene). Colonies were picked, plasmid prepared (QIAprep Spin Miniprep Kit, Qiagen) and sequenced after manipulation (BigDye3.1, ABI PRISM) to ensure sequence fidelity. Kinase assays In vitro kinase assays of recombinant proteins were carried out using ADP-‐Glo™ Kinase Assay (Promega).The assays were carried out at 25°C in 40 mM Tris-‐HCl pH 8.0, 20 mM MgCl2, 20 mM NaCl, 100 μM TCEP, and 100 μM Na3VO4, in a total volume of 60 μl (15 μl kinase reaction;15 μl ADP-‐Glo™
ACCEPTED MANUSCRIPT Reagent; 30 μl Kinase Detection Reagent) in solid white, flat-‐bottom 96-‐well plates. Poly (E4Y1) peptide (Sigma-‐Aldrich) and PLCγSA WT were used as substrates. The latter is an appropriate substrate since during physiological activation of PLCγ by FGFR, the interaction of activated FGFR1 is centred on the nSH2 of PLCγ, while phosphorylation occurs on residue Y783. Tyrosine phosphatase inhibitor sodium orthovanadate was used in the buffer system to preserve the protein tyrosyl phosphorylation state and
RI PT
allow activation of FGFR1KD in the absence of ligand HSPG (Gherzi 1988, Posner 1994). Kinase reactions were stopped after 30 minutes of incubation for enzyme titration and autophosphorylation reactions, and after 60 minutes for inhibitor dose response titrations. ATP-‐to-‐ADP standard conversion curves, Z-‐ values and signal-‐to-‐background ratio were calculated according to the manufacturer’s published procedure in order to assess the linearity of the assay and to calculate the amount of ADP produced
SC
during enzymatic titrations. Z' values were calculated for 5 – 100% ATP conversion using 50 µM ATP. For determinations of Km of the ATP (Km, ATP) concentration that should be used, reactions contained a
M AN U
peptide concentration of 0.2 μg/μl and ATP at varying concentrations of 10–240 μM. The assay signal in the presence of increasing ATP concentrations was measured and fitted to the Michaelis–Menten equation. The optimal kinase amount was determined when 50% of substrate conversion was achieved (half maximal reaction velocity, ½Vmax), and assays were carried out at the apparent Km, ATP, at 50 µM ATP. Negative control experiments were performed in the absence of ATP, substrate, enzyme and positive controls were performed with FGFR1KD WT (data not shown). Luminescence output was
TE D
recorded using BMG Labtech FLUOstar Optima plate reader at 520nm. Apparent IC50 values were calculated by detecting the decrease in phosphorylation of Poly (E4Y1) upon kinase inhibition using AZD4547 and PD173074. Curve fitting for inhibitor dose response titration was performed using GraphPad Prism® sigmoidal dose-‐response (variable slope) software. Apparent IC50 values were then
EP
used to measure the approximate inhibitor constant value (Ki) using the Cheng–Prusoff equation, which describes the dependencies between IC50 value and ATP concentration for ATP-‐competitive inhibitors.
AC C
Error bars indicate the SDs of three replicates.
SDS-‐PAGE and Western blotting Sample buffer (4x stock: 62.5 mM Tris.HCl, 2% SDS, 25% glycerol, 0.01% Bromophenol blue, 0.25 M DTT, pH 6.8) was added to protein samples to 1x. Protein denaturation was performed at 100°C for 5 minutes for optimal sample migration. Protein samples were separated by electrophoresis through sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-‐PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane (equilibrated in methanol) (Millipore) using a standard wet transfer method (buffer: 50 mM Tris-‐HCl, 380 mM glycine and 20% methanol). For pull-‐down assays, proteins were visualised by Amido Black staining. For immunoblotting, the membranes were blocked with 5% non-‐fat
ACCEPTED MANUSCRIPT dry milk (Marvel) in TBS pH 7.5 containing 0.1% Tween-‐20 (Sigma-‐Aldrich) (TBS-‐T) for 30-‐60 min with agitation. Membranes were incubated with the indicated primary antibodies for 1 hour at room temperature or overnight incubation at 4°C. Membranes were washed three times for 5 min in TBS-‐T followed by incubation with required conjugate diluted in 5% milk/TBS-‐T for 1 hour with agitation. Blots were subject to three washes in TBS-‐T and developed using with ECL prime detection kit (Pierce) and
RI PT
exposed to Hyperfilm ECL (Amersham Biosciences).
In-‐gel fluorescence of GyrB.FGFR1KD.TC with FlAsH-‐EDT2 or ReAsH-‐EDT2
FlAsH-‐EDT2 (10 – 100 μM) or ReAsH-‐EDT2 (10 – 50 μM) were added to Laemmli sample buffer [4x stock: 62.5 mM Tris.HCl pH 6.8, 2% SDS, 25% glycerol, 0.01% Bromophenol blue, 50 mM TCEP (neutralized
SC
with 3.5 equiv of Tris.HCl pH 8.0)]. The protein and sample buffer were kept at room temperature for 15-‐30 min and loaded onto the SDS-‐PAGE that was run as usual. The protein-‐FlAsH complexes were
M AN U
visualized on the unstained or unfixed gel by illumination with either an ultraviolet light-‐box or by using a CCD camera with appropriate filters for fluorescence of FlAsH (excitation: 511 nm, emission: 527 nm) or ReAsH (excitation: 593 nm, emission: 608 nm). Native Mass Spectroscopy
Samples were analysed by HPLC, using a Waters Premier quadrupole time of flight (Q-‐Tof) system with a
TE D
Waters 2795XC Separations Module and Waters Symmetry ® C18 column (2.1x100 mm, 3.5 μm). Mobile phase A (water + 0.1% formic acid) and mobile phase B (acetonitrile + 0.1% formic acid) were used with a flow rate of 600 μL/min, column temperature of 40 oC and injection volume of 5 – 10 μL (~20 μg of protein at ~1 mg/mL [5 μM]). Positive Ion Electrospray ionization mode on a Waters Micromass ® Q-‐Tof
EP
micro ™ was in operation. Data processing was performed using Advion ‘Chipsoft’ software.
AC C
Supplementary References
Gherzi, R. et al. DIRECT MODULATION OF INSULIN-‐RECEPTOR PROTEIN TYROSINE KINASE BY VANADATE AND ANTI-‐INSULIN RECEPTOR MONOCLONAL-‐ANTIBODIES. Biochemical and Biophysical Research Communications 152, 1474-‐1480, doi:10.1016/s0006-‐291x(88)80452-‐2 (1988). Posner, B. I. et al. PEROXOVANADIUM COMPOUNDS -‐ A NEW CLASS OF POTENT PHOSPHOTYROSINE PHOSPHATASE INHIBITORS WHICH ARE INSULIN MIMETICS. Journal of Biological Chemistry 269, 4596-‐ 4604 (1994).
ACCEPTED MANUSCRIPT Supplementary figure legends Supplementary Figure S1. (A) SEC molecular weight standards as determined on a Superdex 200 column. (B) Elution profile of the GyrB-‐FGFR1KD.TC monomer as determined on a Superdex 200 column.
RI PT
Supplementary Figure S2. Specific and quantitative labelling of GyrB.FGFR1KD.TC upon incubation with FlAsH-‐EDT2 demonstrated by SDS-‐PAGE (Coomassie staining and in-‐gel fluorescence).
Supplementary Figure S3. (A) (B) Inhibitor dose response titration of GyrB.FGFR1KD.TC in the presence of 50 μM ATP, 0.2 μg/μl PolyE4Y1 peptide and inhibitor PD173074 at varying concentrations of 0.8 – 500
SC
nM (FGFR1 IC50 = ~22 nM). Kinase inhibition was measured by detecting the decrease in phosphorylation of PolyE4Y1 after 1 hour. Maximal activity of GyrB.FGFR1KD.TC and background signal
M AN U
were measured in the absence of inhibitor PD173074.
Supplementary Figure S4. Complex formation of FGFR1KD with mTFP.nSH2 and artificial dimerization as observed by chromatography
(A) Complex formation; Elution profiles of the GyrB.FGFR1KD.TC-‐mTFP.nSH2 complex (blue) and the
excess nSH2 (pink) with 50 μM ATP/20 mM MgCl2 as determined on a Superdex 200 column. (B) Elution profiles of the GyrB.FGFR1KD.TC dimer (green) and the GyrB.FGFR1KD.TC monomer (yellow)
EP
AC C
TE D
with 125 μM coumermycin as determined on a Superdex 200 column.
ACCEPTED MANUSCRIPT
Aggr
kDa
670 158 44 17
A280nm (mAU) 5
10 15 Volume (mL)
20
25
670 158 44 17
1.3
60 40 20 0
0
5
10 15 Volume (mL)
20
25
M AN U
SC
0
Aggr
kDa
100
0
Superdex 200 10/300 GL
1.3
200
EP
TE D
Supplementary Figure S1
AC C
A280nm (mAU)
B
Superdex 200 10/300 GL
RI PT
A
ACCEPTED MANUSCRIPT GyrB.FGFR1KD.TC
+
ReAsH-EDT2 (μM)
10 50 50
+
-
+ 0
75 50
RI PT
37 Coomassie 75
37
M AN U
Fluorescence
SC
50
AC C
EP
TE D
Supplementary Figure S2
ACCEPTED MANUSCRIPT
B Luminescence, RLU
4000
1500
3000 2000
1000
500
1000
0
0
1 2 log [PD173074], nM
SC
0.5
M AN U
-1.5 -1 -0.5 0 log [GyrB.FGFR1KD.TC], μM
TE D
-2
EP
0
RI PT
5000
AC C
Luminescence, RLU
A
Supplementary Figure S3
3
ACCEPTED MANUSCRIPT
670 158 44 17
A280nm (mAU)
40 20
5
10 15 Volume (mL)
20
25
60 40 20 0
670 158 44 17
1.3
0
5
10 15 Volume (mL)
20
AC C
EP
TE D
M AN U
0
Aggr
kDa
60
0
Superdex 200 10/300 GL
1.3
RI PT
Aggr
kDa
A280nm (mAU)
B
Superdex 200 10/300 GL
SC
A
Supplementary Figure S4
25
S2212-4926(15)30015-4
DOI:
10.1016/j.jbior.2015.09.002
Reference:
JBIOR 117
To appear in:
Advances in Biological Regulation
Received Date: 17 September 2015 Accepted Date: 22 September 2015
Please cite this article as: Perdios L, Bunney TD, Warren SC, Dunsby C, French PMW, Tate EW, Katan M, Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1, Advances in Biological Regulation (2015), doi: 10.1016/j.jbior.2015.09.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Time-resolved FRET reports FGFR1 dimerization and formation of a complex with its effector PLCγ1
Edward W. Tate2 and Matilda Katan1*
1
RI PT
Louis Perdios1,2, Tom D. Bunney1, Sean C. Warren3, Christopher Dunsby3, Paul M. W. French3,
London, Gower St, London WC1E 6BT, UK 2
SC
Institute of Structural and Molecular Biology, Division of Biosciences, University College
Road, London SW7 2AZ, UK 3
M AN U
Department of Chemistry, Imperial College London, South Kensington Campus, Exhibition
Department of Physics, Imperial College London, South Kensington Campus, Exhibition Road,
EP
Address correspondence to: Matilda Katan, E-mail: [email protected]
AC C
*
TE D
London SW7 2AZ, UK
1
ACCEPTED MANUSCRIPT Time-resolved FRET reports FGFR1 dimerization and formation of a complex with
AC C
EP
TE D
M AN U
SC
RI PT
its effector PLCγ1
2
ACCEPTED MANUSCRIPT Abstract
In vitro and in vivo imaging of protein tyrosine kinase activity requires minimally invasive, molecularly precise optical probes to provide spatiotemporal mechanistic information of
RI PT
dimerization and complex formation with downstream effectors. We present here a construct with genetically encoded, site-specifically incorporated, bioorthogonal reporter that can be selectively labelled with exogenous fluorogenic probes to monitor the structure
SC
and function of fibroblast growth factor receptor (FGFR). GyrB.FGFR1KD.TC contains a coumermycin-induced artificial dimerizer (GyrB), FGFR1 kinase domain (KD) and a
M AN U
tetracysteine (TC) motif that enables fluorescent labelling with biarsenical dyes FlAsH-EDT2 and ReAsH-EDT2. We generated bimolecular system for time-resolved FRET (TR-FRET) studies, which pairs FlAsH-tagged GyrB.FGFR1KD.TC and N-terminal Src homology 2 (nSH2) domain of phospholipase Cγ (PLCγ), a downstream effector of FGFR1, fused to mTurquoise
TE D
fluorescent protein (mTFP). We demonstrated phosphorylation-dependent TR-FRET readout of complex formation between mTFP.nSH2 and GyrB.FGFR1KD.TC. By further application of TR-FRET, we also demonstrated formation of the GyrB.FGFR1KD.TC homodimer by
EP
coumermycin-induced dimerization. Herein, we present a spectroscopic FRET approach to facilitate and propagate studies that would provide structural and functional insights for
AC C
FGFR and other tyrosine kinases.
Keywords
fibroblast growth factor receptor; phospholipase Cγ; time-resolved FRET
3
ACCEPTED MANUSCRIPT 1. Introduction
Fibroblast growth factors (FGFs) and their receptors (FGFR1-4) play a critical role in many physiological processes including embryogenesis, wound healing, inflammation and
RI PT
angiogenesis as well as adult tissue homeostasis (Beenken and Mohammadi, 2009). Importantly, aberrant FGFs/FGFRs signaling has been causatively linked to several developmental syndromes and a broad range of human malignancies (Carter et al., 2015;
SC
Helsten et al., 2015). This involvement in the pathology of many cancer types provided a strong rationale for development of effective agents for these targets; consequently, there is
M AN U
a large ongoing effort to develop FGFR inhibitors as anticancer treatments (Touat et al., 2015).
FGFs mediate their diverse biological responses by binding to FGFRs, leading to their
TE D
dimerization and autophosphorylation; despite extensive insights into isolated extracellular and intracellular regions of FGFRs, overall process of dimerization is not clearly defined (Goetz and Mohammadi, 2013; Maruyama, 2014). Subsequent to dimerization, similar to
EP
other receptor tyrosine kinases (RTKs), activated FGFRs dimers recruit and phosphorylate a complement of signaling molecules that mediate the cellular activities of FGFs (Bradshaw et
AC C
al., 2015; Lemmon and Schlessinger, 2010; McCubrey et al., 2015). In case of FGFRs, the two main direct binding proteins that propagate signaling are the adaptor molecule FRS2 and phospholipase Cγ (PLCγ) (Beenken and Mohammadi, 2009; Eswarakumar et al., 2005). PLCγ is recognized as a key component downstream of RTKs involved in dynamic coupling of modifying enzymes with signaling lipids (Blind, 2014; Follo et al., 2015) and, furthermore, has been recently implicated in disease development through gain-of-function mutations (Koss et al., 2014).
4
ACCEPTED MANUSCRIPT Several lines of experimental evidence have demonstrated that FGFR is constitutively bound to FRS2 via FGFR juxtamembrane region, whereas PLCγ is recruited to the C-terminal tail of the receptor upon autophosphorylation of a highly conserved tyrosine (Y766 in case of FGFR1) (Eswarakumar et al., 2005). The main interacting domain in PLCγ is the N-terminal
RI PT
Src-homology 2 domain (nSH2) present within the PLCγ regulatory region (or γ-specific array, γSA); the interaction involves not only the canonical binding site centered on pY766 at the FGFR1 tale, but also a secondary binding site (Bae et al., 2009). Interestingly, the selectivity
SC
of PLCγ binding and signaling via activated FGFR1 are determined by interactions between this secondary binding site on nSH2 domain and a region in FGFR1 kinase domain. Thus,
M AN U
binding of nSH2 from PLCγ1 to two distinct sites on FGFR1 (the tale and kinase domain) represents a highly specific and high affinity (kD about 5 nM) event that is, overall, dependent on activation status of FGFR (Bae et al., 2009; Bunney et al., 2012). Measurements of this interaction using fluorescent reporters and FRET could provide a
TE D
sensitive method to assess activation of FGFR in vitro and in cells or an inhibition of this process in the presence of small molecule drugs.
EP
We here describe generation and characterisation of constructs that provide new tools to assess spatiotemporal mechanistic details of dimerization and downstream signaling. We
AC C
generated GyrB.FGFR1KD.TC construct comprising a coumermycin-induced artificial dimerizer (GyrB), FGFR1 kinase domain (KD) – exhibiting wild-type analogous autophosphorylation, substrate recruitment, and inhibitor response in kinase assays and chromatography measurements – and a tetracysteine (TC) motif that enables fluorescent labelling with biarsenicals FlAsH-EDT2 and ReAsH-EDT2. We conceived systems for TR-FRET studies, which pair either FlAsH- and ReAsH-tagged GyrB.FGFR1KD.TC or pair FlAsH-tagged GyrB.FGFR1KD.TC and nSH2 domain of PLCγ fused to mTurquoise fluorescent protein (mTFP). We demonstrated coumermycin-induced dimerization and phosphorylation-
5
ACCEPTED MANUSCRIPT dependent
TR-FRET
readout
of
complex
formation
between
mTFP.nSH2
and
GyrB.FGFR1KD.TC.
RI PT
2. Material and Methods
2.1 TR-FRET
SC
For fluorescence lifetime decay measurements in a quartz cuvette using a TR-FRET multidimensional spectrofluorometer. GyrB.FGFR1KD.TC (1 μM) tagged with FlAsH-EDT2
M AN U
(5μM and 50 μM) and its partner mTFP.nSH2 (1 μM) in 40 mM Tris-HCl (pH8.0), 20 mM MgCl2, 20 mM NaCl, 100 μM TCEP, and 100 μM Na3VO4 were incubated in the presence or absence of ATP (50μM) in at ambient conditions for 30 minutes. Excitation of the FRET pair was performed at 410 – 425 nm using the supercontinuum source, and fluorescence
TE D
emission detected at 430 – 520 nm (10 nm steps) in order to record the decrease in fluorescence lifetime of the donor mTFP. Analysis of lifetime and anisotropy decays was performed using TRFA data processor (Scientific Software Technologies Center, Minsk,
EP
Belarus).
AC C
Details for cloning, expression and purification of recombinant proteins, kinase assays, in vitro labelling of purified proteins and Supplementary Figures can be found in Supplementary Information.
6
ACCEPTED MANUSCRIPT 3. Results
3.1 Constructs for a TR-FRET approach to detect FGFR1 dimerization and complex
RI PT
formation with PLCγ
Accumulating evidence show that ligand-induced dimerization of RTK extracellular regions leads to activation of the intracellular KD, which, in turn, alters complex downstream
SC
signalling networks (Lemmon and Schlessinger, 2010). However, intact RTKs are difficult to generate and reconstruct into in vitro signaling systems. In order to study the molecular
M AN U
interactions of the FGFR1 in vitro, a dimerizer was incorporated at the N-terminus and a fluorescent self-labelling peptide sequence at the C-terminus of FGFR1KD (Figure 1 A, top). Our previous work (Bunney et al., 2012; Bunney et al., 2015) involved the construction of a synthetic open reading frame (ORF) encoding for a mutant form of human FGFR1 kinase
TE D
domain (KD), FGFR1KD [amino acids 457–774, (Y463, 583, 585F), (L457V, C488A, C584S)] that forms a central part of our new construct (Figure1 A, top). FGFR1 dimerization could be monitored in vitro by using an artificial dimerization motif at the N-terminus of the KD to
EP
mimic the extracellular portion of the wild-type protein. For this purpose, the N-terminal 24kDa subdomain of the B subunit of bacterial DNA gyrase (GyrB), which dimerizes in the
AC C
presence of the antibiotic coumermycin (Farrar et al., 1996; Liu et al., 2003), was incorporated into the construct, generating GyrB.FGFR1KD. Using recombinant cloning methods, an optimised TC motif (FLNCCPGCCMEP) (Spagnuolo et al., 2006) was introduced at the C-terminus of GyrB.FGFR1KD to generate GyrB.FGFR1KD.TC (Figure 1A, top). The protein was then isolated from transformed E.coli strain C41(DE3). The optimised FlAsH- or ReAsH- binding TC motif has been previously used to monitor protein structure in vitro and in mammalian cells (Luedtke et al., 2007; Martin et al., 2005; Roberti et al., 2007; Spille et al., 2011).
7
ACCEPTED MANUSCRIPT
The high affinity of the TC motif to the biarsenical dye (Adams and Tsien, 2008) suggests that the complex should be stable under typical denaturing conditions. The identity and integrity of TC-tagged protein was verified by survival of fluorescence after SDS-PAGE. The
RI PT
recombinant protein was incubated with FlAsH-EDT2 for 30 min at room temperature in a modified sample buffer for SDS-PAGE in which the typical thiol reductant (2mercaptoethanol) was substituted by tris(2-carboxy-ethyl)phosphine (TCEP). After
SC
electrophoresis, a single major fluorescent species running at the anticipated molecular weight for the GyrB.FGFR1KD.TC (60.1 kDa) was visible by UV illumination (Figure 1B). Excess
M AN U
FlAsH-EDT2, which could result in increased levels of background fluorescence, is removed as it migrates with the dye front. Subsequent staining with Coomassie Blue confirmed the specific binding of dye to the TC-containing protein. This result suggests that GyrB.FGFR1KD.TC tagged with FlAsH-EDT2 (or ReAsH-EDT2) can be studied in a mixture of
TE D
proteins or in a crude lysate. The identity of the fully labelled protein was further confirmed by native mass spectroscopy (Figure 1C), in which the GyrB.FGFR1KD.TC/FlAsH-EDT2 complex was clearly observed as a single peak at the anticipated molecular weight (found: 62543.0,
EP
calc.: 62539.0). Based on these data, labelling of GyrB.FGFR1KD.TC is suitable for further
AC C
FRET experiments to test dimerization.
As a second model system for RT-FRET experiments, the FGFR1 KD and its downstream target PLCγ were used; it has been previously established that a stable complex is formed between the C-terminus of activated FGFR1KD and two different sites on the surface of the nSH2 from PLCγ regulatory region (Figure 1A, bottom) (Bae et al., 2009). The canonical phosphorylation-dependent binding site consists of a positively charged patch on the surface of the nSH2 domain that binds to pY766 and a hydrophobic pocket in the FGFR1 C-terminal tail. A second binding site of the nSH2 forms mediates high affinity binding by
8
ACCEPTED MANUSCRIPT forming contacts with the backside of the C-lobe of the KD away from the activation segment of FGFR1. The co-crystal structure of nSH2 of PLCγ with activated FGFR1KD (PDB code: 3GQI) (Bae et al., 2009), estimates a distance of 28.7 Å between nSH2 (N-terminus) and FGFR1KD (C-terminus), making the protein interaction measurable by FRET. As a FRET
RI PT
donor, cyan fluorescent protein (CFP) variant mTurquoise (mTFP) (excitation: 434 nm,
bottom).
SC
emission: 474 nm) fused to the nSH2 domain of PLCγ (mTFP.nSH2) was used (Figure 1A,
3.2 Kinase profiling of GyrB.FGFR1KD.TC demonstrates that its biochemical properties
M AN U
match those of the wild-type protein in vitro
To determine the in vitro tyrosine kinase activity of recombinant GyrB.FGFR1KD.TC, a bioluminescent, ADP detection assay was used (Zegzouti et al., 2009). The kinase substrates
TE D
used here were PolyE4Y1 and a part of PLCγ regulatory region, which comprises the two SH2 domains of PLCγ followed by a segment containing Y783 (Figure 1A, bottom). Inhibitors used here were PD173074, an established ATP-competitive inhibitor of FGFR1 (inhibitor constant,
EP
Ki: ~ 44 nM, IC50: ~ 22 nM) (Mohammadi et al., 1998) and AZD4547, a selective inhibitor of FGFR1 currently in clinical trials (Ki: 6.1 nM, IC50: 0.2 nM) (Bunney et al., 2015; Gavine et al.,
AC C
2012).
To optimise assay conditions, we first determined the ATP concentration that equals the Km value for GyrB.FGFR1KD.TC (Figure 2A). The data were fitted to the Michaelis-Menten equation, thereby determining the GyrB.FGFR1KD.TC ATP Km to be 126 μΜ. After identifying appropriate substrates and an optimal ATP concentration, the dependence of signal linearity to kinase concentration was established in a series of GyrB.FGFR1KD.TC titrations with PolyE4Y1 (Figure 2B) and PLCγ SA WT (Supplementary Figure S3A). Here, we have chosen to
9
ACCEPTED MANUSCRIPT use 0.3 nM and 30 nM of GyrB.FGFR1KD.TC with inhibitors AZD4547 and PD173074 respectively, to guarantee sufficiently high assay signal as well as substrate conversion below 70%. In the last step of the optimisation process, the contribution of the autophosphorylation state of the kinase on the total assay signal was examined in the
(Figure
2C),
resulting
in
higher
levels
signal
readout
corresponding
to
SC
autophosphorylation.
of
RI PT
absence of substrate. Increasing enzyme concentration were titrated with 150 μΜ ATP
In order to validate the optimised GyrB.FGFR1KD.TC assay, the kinase activity was quantified
M AN U
with increasing concentrations of inhibitor AZD4547 (Figure 2D) and reference inhibitor PD173074 (Supplementary Figure S3B). For AZD4547, the IC50 value of 0.32 nM was in good agreement with the published value of 0.20 nM. For PD173074, the IC50obs was 13 nM, an estimation that diverges slightly from the reported value of ~22 nM. This may be ascribed to
TE D
dissimilarities between the assay format used here and the one in literature, where a radiometric readout assay format and full length FGFR1 were used to measure the IC50
EP
values.
3.3 Coumermycin-induced dimerization of FlAsH and ReAsH labelled GyrB.FGFR1KD.TC
AC C
proteins results in a FRET readout
Extracellular dimerization of FGFRs involves the formation of 2:2:2 FGF:FGFR:heparin ternary complex. The recombinant proteins used in this study contain the intracellular KD activity domain, which is suitable for enzyme activity assays, but do not contain the extracellular domain of FGFR1. Attempts to get multi-domain FGFR constructs were not pursued as they show low expression levels in E.coli, due to inadequate solubility and misfolding.
10
ACCEPTED MANUSCRIPT To achieve artificial dimerization of GyrB.FGFR1KD.TC, a model involving FGF ligand-mimic coumermycin, that induces dimerization of GyrB, was used. Coumermycin binds GyrB at a stoichiometric ratio 1:2, whereas another antibody, novobiocin, binds GyrB at 1:1 ratio (Ali et al., 1993) and was thereby suitable as a control of obtaining exclusively monomeric
RI PT
protein population. To first test this dimerization model biochemically, analytical sizeexclusion chromatography (SEC) was used before and following incubation with antibiotics. Samples incubated with coumermycin (GyrB to antibiotic molar ratio, 1.5:1) exhibited major
SC
peaks that corresponded to the presence of predominantly the dimer (Supplementary Figure S4B). The retention time of the coumermycin-bound dimer was equivalent to the retention
(Supplementary Figure S1A).
M AN U
time of species of 120 kDa when compared to SEC standard molecular weight markers
Coumermycin-induced dimerization was subsequently investigated by TR-FRET. In these
TE D
studies FlAsH-EDT2 is used as the donor to fluorescently tag a population of GyrB.FGFR1KD.TC while red-emitting biarsenical ReAsH-EDT2 (excitation: 593 nm, emission: 608 nm), which binds to the C-terminal TC motif of GyrB.FGFR1KD.TC (Supplementary Figure
EP
S2), is issued as the acceptor to tag another population of the same protein (Figure 3A). The suitability of FlAsH-EDT2 and ReAsH-EDT2 as a donor acceptor pair was demonstrated in
AC C
vitro by bulk spectrofluorimetry (Figure 3B), which highlighted that the two dyes have ideal
spectral overlap for FRET to occur. The change in donor emission in the presence and absence of coumermycin was analysed by fitting the decays using single exponential models (Figure 3C). Addition of coumermycin to the mixture of proteins tagged for FlAsH-EDT2 and ReAsH-EDT2 resulted in a decrease in lifetime of ~60 ps for the sample (Figure 3D). Thus, under conditions of dimerization, the overall dye proximity increases resulting in faster lifetime decay of the donor and FRET. Since this is triggered by addition of coumermycin to
11
ACCEPTED MANUSCRIPT the sample in solution, FRET can be attributed to the formation of dimerization complexes between FlAsH-tagged proteins and ReAsH-tagged proteins.
3.4.
TR-FRET
detection
of
the
complex-formation
between
FlAsH-labelled
RI PT
GyrB.FGFR1KD.TC and mTFP.nSH2
For TR-FRET measurements, GyrB.FGFR1KD.TC was reacted with FlAsH-EDT2 and then
SC
incubated with mTFP.nSH2 for 30 min at room temperature (Figure 4A). The suitability of FlAsH-EDT2 as a donor and mTFP as an acceptor was demonstrated in vitro by bulk
M AN U
spectrofluorimetry (Figure 4B). Next, fluorescence lifetime decays of the samples were recorded in the presence and absence of ATP. To interpret these data, exponential decay models were fitted to the decays; it was apparent that both mTFP.nSH2 by itself and the protein pair were well fit to a single exponential decay (Figure 4C). Addition of ATP to the
TE D
incubated protein pair resulted in a large decrease in lifetime of ~500 ps for the sample (Figure 4D). The fluorescence decay observed reflects the increase in spatial proximity of the donor and acceptor fluorophores, which was attributed to direct interaction between the
EP
nSH2 domain of PLCγ and activated FGFR1KD.
AC C
The interaction between GyrB.FGFR1KD.TC and mTFP.nSH2 was also confirmed using SEC; the incubated proteins eluted at the same retention time (elution volume 14.7 mL) indicating formation of a GyrB.FGFR1KD.TC–mTFP.nSH2 complex (Supplementary Figure S4A).
12
ACCEPTED MANUSCRIPT 4. Discussion
We here demonstrate, using a TR-FRET approach, the suitability of GyrB dimerizer for studies of FGFR signaling and application of bimolecular mTFP/FlAsH.TC system for
RI PT
monitoring the phosphorylation-dependent complex formation between the nSH2 domain of PLCγ and functional GyrB.FGFR1KD.TC.
SC
Use of non-physiological dimerization elements has been widely used to study signal transduction cascades triggered by dimerization (Fegan et al., 2010). This approach has also
M AN U
been applied to FGFR; for example, a construct incorporating FKBPv at the C-terminus of the FGFR KD and dimerization in the presence of a drug (AP20187) provided a basis for specific, inducible and ligand independent activation of FGFR1 signalling in cells and mouse models (Acevedo et al., 2007; Freeman et al., 2003; Tomlinson et al., 2012; Welm et al., 2002). We
TE D
here demonstrate that GyrB allows not only previously observed dimerization in cells (Liu et al., 2003), but also generation of purified, functional FGFR-GyrB fusion proteins (Figures 1 and 3) for further structural and in vitro studies of elements critical for dimerization and
EP
high-throughput screens for disruption of this key activation step.
AC C
Interactions between RTKs and SH2 domains of downstream effectors have been explored for their potential to report signaling events in cells and thus provide a platform for drug discovery that would overcome one of the common problems based on kinase assays in vitro: the identification of potent hits in vitro with poor cellular activity. One recent EGFR biosensor that allows measurements of EGFR clustering as a readout of activation in cells is based on binding of the green fluorescent protein fused to two tandem SH2 domains from adapter protein Grb2; it was subsequently shown that this system provides an important high throughput system for drug discovery performed in a cellular context (Antczak et al.,
13
ACCEPTED MANUSCRIPT 2012). We here suggest that the highly specific, high affinity interaction between FGFR1 and nSH2 domain from PLCγ (Bae et al., 2009; Bunney et al., 2012) can similarly be used in drug discovery. We generated a FRET pair for TR-FRET detection of the complex-formation between FlAsH-labelled GyrB.FGFR1KD.TC and mTFP.nSH2 (Figure 4). The decrease in
RI PT
fluorescence lifetime of mTFP depends exclusively on its molecular proximity to the TC motif; hence, FRET occurrence can be attributed to binding of mTFP.nSH2 to activated GyrB.FGFR1KD.TC. The dramatic decrease in fluorescent lifetime of the donor may be
SC
ascribed to the high affinity binding of the nSH2 domain to the activated FGFR1KD. In vitro experiments with isolated FRET pairs have provided a straightforward readout of the
M AN U
protein-protein interaction and can be adapted for screening of inhibitors that directly affect kinase activity as well as interaction between FGFR and PLCγ. Furthermore, this approach can be developed for cellular FLIM-FRET. Using constructs that ensure membrane localization of GyrB.FGFR1KD.TC and intracellular labelling of this protein in the presence of
TE D
FlAsH, subsequent activation in response to addition of coumermycin will result in binding of
EP
the fluorescent fusion protein mTFP.nSH2 expressed in the same cell and the FRET readout.
5. Acknowledgments
AC C
The MK laboratory is supported by Cancer Research UK (A16567) and LP by Imperial College London Doctoral Training Centre.
6. References Acevedo, V. D., Gangula, R. D., Freeman, K. W., Li, R., Zhang, Y., Wang, F., Ayala, G. E., Peterson, L. E., Ittmann, M., and Spencer, D. M., 2007. Inducible FGFR-1 activation leads to
14
ACCEPTED MANUSCRIPT irreversible prostate adenocarcinoma and an epithelial-to-mesenchymal transition. Cancer cell 12, 559-571. Adams, S. R., and Tsien, R. Y., 2008. Preparation of the membrane-permeant biarsenicals FlAsH-EDT2 and ReAsH-EDT2 for fluorescent labeling of tetracysteine-tagged proteins.
RI PT
Nature protocols 3, 1527-1534. Ali, J. A., Jackson, A. P., Howells, A. J., and Maxwell, A., 1993. The 43-kilodalton N-terminal fragment of the DNA gyrase B protein hydrolyzes ATP and binds coumarin drugs.
SC
Biochemistry 32, 2717-2724.
Antczak, C., Bermingham, A., Calder, P., Malkov, D., Song, K., Fetter, J., and Djaballah, H.,
M AN U
2012. Domain-based biosensor assay to screen for epidermal growth factor receptor modulators in live cells. Assay and drug development technologies 10, 24-36. Bae, J. H., Lew, E. D., Yuzawa, S., Tome, F., Lax, I., and Schlessinger, J., 2009. The selectivity of receptor tyrosine kinase signaling is controlled by a secondary SH2 domain binding site. Cell
TE D
138, 514-524.
Beenken, A., and Mohammadi, M., 2009. The FGF family: biology, pathophysiology and therapy. Nature reviews Drug discovery 8, 235-253.
EP
Blind, R. D., 2014. Disentangling biological signaling networks by dynamic coupling of signaling lipids to modifying enzymes. Advances in biological regulation 54, 25-38.
AC C
Bradshaw, R. A., Pundavela, J., Biarc, J., Chalkley, R. J., Burlingame, A. L., and Hondermarck, H., 2015. NGF and ProNGF: Regulation of neuronal and neoplastic responses through receptor signaling. Advances in biological regulation 58, 16-27. Bunney, T. D., Esposito, D., Mas-Droux, C., Lamber, E., Baxendale, R. W., Martins, M., Cole, A., Svergun, D., Driscoll, P. C., and Katan, M., 2012. Structural and functional integration of the PLCgamma interaction domains critical for regulatory mechanisms and signaling deregulation. Structure 20, 2062-2075.
15
ACCEPTED MANUSCRIPT Bunney, T. D., Wan, S., Thiyagarajan, N., Sutto, L., Williams, S. V., Ashford, P., Koss, H., Knowles, M. A., Gervasio, F. L., Coveney, P. V., and Katan, M., 2015. The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study. EBioMedicine 2, 194-204.
RI PT
Carter, E. P., Fearon, A. E., and Grose, R. P., 2015. Careless talk costs lives: fibroblast growth factor receptor signalling and the consequences of pathway malfunction. Trends in cell biology 25, 221-233.
SC
Eswarakumar, V. P., Lax, I., and Schlessinger, J., 2005. Cellular signaling by fibroblast growth factor receptors. Cytokine & growth factor reviews 16, 139-149.
M AN U
Farrar, M. A., Alberol-Ila, J., and Perlmutter, R. M., 1996. Activation of the Raf-1 kinase cascade by coumermycin-induced dimerization. Nature 383, 178-181. Fegan, A., White, B., Carlson, J. C., and Wagner, C. R., 2010. Chemically controlled protein assembly: techniques and applications. Chemical reviews 110, 3315-3336.
TE D
Follo, M. Y., Manzoli, L., Poli, A., McCubrey, J. A., and Cocco, L., 2015. PLC and PI3K/Akt/mTOR signalling in disease and cancer. Advances in biological regulation 57, 10-16. Freeman, K. W., Welm, B. E., Gangula, R. D., Rosen, J. M., Ittmann, M., Greenberg, N. M., and
EP
Spencer, D. M., 2003. Inducible prostate intraepithelial neoplasia with reversible hyperplasia in conditional FGFR1-expressing mice. Cancer research 63, 8256-8263.
AC C
Gavine, P. R., Mooney, L., Kilgour, E., Thomas, A. P., Al-Kadhimi, K., Beck, S., Rooney, C., Coleman, T., Baker, D., Mellor, M. J., et al., 2012. AZD4547: an orally bioavailable, potent, and selective inhibitor of the fibroblast growth factor receptor tyrosine kinase family. Cancer research 72, 2045-2056.
Goetz, R., and Mohammadi, M., 2013. Exploring mechanisms of FGF signalling through the lens of structural biology. Nature reviews Molecular cell biology 14, 166-180.
16
ACCEPTED MANUSCRIPT Helsten, T., Schwaederle, M., and Kurzrock, R., 2015. Fibroblast growth factor receptor signaling in hereditary and neoplastic disease: biologic and clinical implications. Cancer metastasis reviews. Koss, H., Bunney, T. D., Behjati, S., and Katan, M., 2014. Dysfunction of phospholipase
RI PT
Cgamma in immune disorders and cancer. Trends in biochemical sciences 39, 603-611. Lemmon, M. A., and Schlessinger, J., 2010. Cell signaling by receptor tyrosine kinases. Cell 141, 1117-1134.
SC
Liu, G., Bafico, A., Harris, V. K., and Aaronson, S. A., 2003. A novel mechanism for Wnt activation of canonical signaling through the LRP6 receptor. Molecular and cellular biology
M AN U
23, 5825-5835.
Luedtke, N. W., Dexter, R. J., Fried, D. B., and Schepartz, A., 2007. Surveying polypeptide and protein domain conformation and association with FlAsH and ReAsH. Nature chemical biology 3, 779-784.
TE D
Martin, B. R., Giepmans, B. N., Adams, S. R., and Tsien, R. Y., 2005. Mammalian cell-based optimization of the biarsenical-binding tetracysteine motif for improved fluorescence and affinity. Nature biotechnology 23, 1308-1314.
EP
Maruyama, I. N., 2014. Mechanisms of activation of receptor tyrosine kinases: monomers or dimers. Cells 3, 304-330.
AC C
McCubrey, J. A., Abrams, S. L., Fitzgerald, T. L., Cocco, L., Martelli, A. M., Montalto, G., Cervello, M., Scalisi, A., Candido, S., Libra, M., and Steelman, L. S., 2015. Roles of signaling pathways in drug resistance, cancer initiating cells and cancer progression and metastasis. Advances in biological regulation 57, 75-101. Mohammadi, M., Froum, S., Hamby, J. M., Schroeder, M. C., Panek, R. L., Lu, G. H., Eliseenkova, A. V., Green, D., Schlessinger, J., and Hubbard, S. R., 1998. Crystal structure of an angiogenesis inhibitor bound to the FGF receptor tyrosine kinase domain. The EMBO journal 17, 5896-5904.
17
ACCEPTED MANUSCRIPT Roberti, M. J., Bertoncini, C. W., Klement, R., Jares-Erijman, E. A., and Jovin, T. M., 2007. Fluorescence imaging of amyloid formation in living cells by a functional, tetracysteinetagged alpha-synuclein. Nature methods 4, 345-351. Spagnuolo, C. C., Vermeij, R. J., and Jares-Erijman, E. A., 2006. Improved photostable FRET-
American Chemical Society 128, 12040-12041.
RI PT
competent biarsenical-tetracysteine probes based on fluorinated fluoresceins. Journal of the
Spille, J. H., Zurn, A., Hoffmann, C., Lohse, M. J., and Harms, G. S., 2011. Rotational diffusion
SC
of the alpha(2a) adrenergic receptor revealed by FlAsH labeling in living cells. Biophysical journal 100, 1139-1148.
M AN U
Tomlinson, D. C., Knowles, M. A., and Speirs, V., 2012. Mechanisms of FGFR3 actions in endocrine resistant breast cancer. International journal of cancer Journal international du cancer 130, 2857-2866.
Touat, M., Ileana, E., Postel-Vinay, S., Andre, F., and Soria, J. C., 2015. Targeting FGFR
TE D
Signaling in Cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 21, 2684-2694.
Welm, B. E., Freeman, K. W., Chen, M., Contreras, A., Spencer, D. M., and Rosen, J. M., 2002.
EP
Inducible dimerization of FGFR1: development of a mouse model to analyze progressive transformation of the mammary gland. The Journal of cell biology 157, 703-714.
AC C
Zegzouti, H., Zdanovskaia, M., Hsiao, K., and Goueli, S. A., 2009. ADP-Glo: A Bioluminescent and homogeneous ADP monitoring assay for kinases. Assay and drug development technologies 7, 560-572.
18
ACCEPTED MANUSCRIPT Figure Legends
Figure 1. Representation of the constructs used in dimerization and complex formation of FGFR1KD
RI PT
(A) Representation of the wild type FGFR1 and PLCγ as well as domains of protein constructs used in this study. FGFR1 (top) comprises Ig-like domains 1, 2, and 3 (D1, 2, and 3, respectively), trans-membrane domain (TM) and intracellular kinase domain (KD). KD (with
SC
mutations shown as yellow dots) is incorporated in GyrB.FGFR1KD.TC that also includes an N-terminal GyrB and a C-terminal TC motif. PLCγ (bottom) contains domains common to PLC
M AN U
family: N-terminal pleckstrin homology domain (PH), EF–hands (EF), catalytic domain (PLC, X-box and Y-box, shown as two parts) and C2 calcium/lipid-binding domain (C2). PLCγ Specific Array (PLCγ SA) comprises a split PH domain (PH), two SH2 domains (nSH2 and cSH2) and the SH3 domain. mTurquoise.nSH2 fusion protein (mTFP.nSH2) comprises an
TE D
mTurquoise (mTFP) and an nSH2 domain.
(B) Specific and quantitative labelling of GyrB.FGFR1KD.TC upon incubation with FlAsH-EDT2 demonstrated by SDS-PAGE (Coomassie staining) (top) and in-gel fluorescence (bottom).
EP
(C) Mass spectrum of the full length intact GyrB.FGFR1KD.TC before (left) and after (right)
AC C
labelling with FlAsH-EDT2.
Figure 2. Kinase profiling of GyrB.FGFR1KD.TC by bioluminescent ADP detection assays. (A) Determination of ATP Km for GyrB.FGFR1KD.TC by titrating with 10 – 240 μM ATP and 0.2 μg/μl PolyE4Y1.
(B) Examination of the dependence of signal linearity to kinase concentration by titration of 0 – 1.1 μM GyrB.FGFR1KD.TC with 150 μM ATP and 0.2 μg/μl PolyE4Y1. (C) Autophosphorylation assay at varying concentrations of 0 – 1.1 μM enzyme and 150 μM ATP.
19
ACCEPTED MANUSCRIPT (D) Inhibitor dose response titration of GyrB.FGFR1KD.TC in the presence of 150 μM ATP, 0.2 μg/μl PolyE4Y1 peptide and inhibitor AZD4547 at varying concentrations of 0.002 – 20 nM (FGFR1 IC50 = 0.2 nM). Kinase inhibition was measured by detecting the decrease in phosphorylation of PolyE4Y1 after 1 hour. Maximal activity of GyrB.FGFR1KD.TC and the
RI PT
background signal were measured in the absence of inhibitor AZD4547. GyrB.FGFR1KD.TC activity was expressed by calculating the ratio between the background-corrected assay signals in the absence and presence of the indicated inhibitor concentrations. Curve fitting
SC
for inhibitor dose response titration was performed using GraphPad Prism® sigmoidal doseresponse software.
M AN U
In all panels error bars indicate the standard deviations (SDs) of three replicates.
Figure 3. Coumermycin-induced dimerization of GyrB.FGFR1KD.TC observed by FRET
induced dimerization.
TE D
(A) Representation of the construct used for the TR-FRET investigating coumermycin-
(B) Spectral overlap as precondition for FRET. The spectral overlap integral is indicated by
EP
the purple area. The region of overlap between the excited donor's emission spectrum (green line) and the acceptor's absorption spectrum (orange line) has to be significant for
AC C
FRET to occur.
(C) Single exponential fit of the fluorescence decay (normalised photon count) over time from fluorescence lifetime decay measurements in a quartz cuvette using a TR-FRET multidimensional spectrofluorometer. (D) Donor lifetime decrease upon dimerization showing statistical significance assessed using data from a minimum of three repeats by a two-tailed unpaired t-test (***0.0001
20
ACCEPTED MANUSCRIPT Figure 4. Complex formation between GyrB.FGFR1KD.TC and mTFP.nSH2 observed by FRET (A) Representation of the constructs used for the TR-FRET examining complex formation. (B) Spectral overlap (purple area) between the excited donor's emission spectrum (green line) and the acceptor's absorption spectrum (orange line).
from fluorescence lifetime decay measurements.
RI PT
(C) Single exponential fit of the fluorescence decay (normalised photon count) over time
(D) Donor lifetime decrease upon complex formation showing statistical significance
SC
assessed using data from a minimum of three repeats by a two-tailed unpaired t-test
AC C
EP
TE D
M AN U
(*0.0001
A
B
GyrB.FGFR1KD.TC
FGFR1
FlAsH-EDT2 (μM) 5
10
50 100
-
kDa 75
D1
D2
D3
TM
KD
ACCEPTED MANUSCRIPT 50
GyrB.FGFR1KD.TC
Y653
464 GyrB
Y654
Y766
FlAsH-EDT2 37
774
KD
TC
N
75
C
C488A Y583,585F,C584S
37
C
PLCγSA PH
EF
PLC
P
nSH2
cSH2
SH3 H
PLC
C2
550 nSH2
100
GyrB.FGFR1KD.TC/FlAsH found:62543.0 calc.:62539.0
60000
62500
0 65000 60000 Mass, m/z
62500
65000 Mass, m/z
EP
TE D
0
AC C
mTFP
657
M AN U
mTFP.nSH2
% Intensity
100
GyrB.FGFR1KD.TC found:62065.0 calc.:62061.0
SC
PLCγ
RI PT
50
Figure 1
A
B
6000
4000
3000 MANUSCRIPT ACCEPTED
4000 3000 2000
Km, ATP: 126.3 μM
1000
C
0
50
100
150
200
2000
1000
0
250
ATP, μM
0.01
0.1 log [GyrB.FGFR1KD.TC], μM
D
150
350
100 50
0.1
1
50
0 -3
TE D
log [GyrB.FGFR1KD.TC], μM
EP
0
SC
150
100
M AN U
Luminescence, RLU
200
AC C
Luminescence, RLU
300 250
1
RI PT
0
Luminescence, RLU
Luminescence, RLU
5000
-2
0 -1 log [AZD4547], nM
IC50 = 0.32
1
Figure 2
2
A
B
Y654
Tyr Kinase N C488A Y583,585F,C584S
400
D 1
Fluorescent lifetime (ps)
GyrB.FGFR1KD.TC - dimeriser + dimeriser
0.5
20 30 Decay, ns
40
50
700
4380 4360 4340 4320 4300
Coumermycin (dimeriser)
GyrB.FGFR1KD.TC -
+
M AN U
10
500 600 Wavelength, nm
4400
TE D
0
EP
0
AC C
Normalised photon count
C
0
RI PT
GyrB
Y653
FlAsH-EDT2 emission ReAsH-EDT2 excitation
SC
464
Relative Intensity units
GyrB.FGFR1KD.TC
FlAsH-EDT2 1000 or ReAsH-EDT2 750 Y766 ACCEPTED MANUSCRIPT 774 500 TC 250 C
Figure 3
B
GyrB.FGFR1KD.TC
Y653
464
Y654
Y766
774
N
C
C488A Y583,585F,C584S mTFP.nSH2
550
1
P 0.5
400 500 600 Wavelength, nm
700
50
4000 3800 3600 3400 GyrB.FGFR1KD.TC with mTFP.nSH2 ATP +
SC
40
4200
TE D
M AN U
20 30 Decay, ns
EP
10
0 300
D Fluorescent lifetime (ps)
GyrB.FGFR1KD.TC + mTFP.nSH2
0
250
nSH2
N
0
500
657
AC C
Normalised photon count
C
750
RI PT
mTFP
mTFP emission FlAsH-EDT2 excitation
1000
ACCEPTED MANUSCRIPT TC
Tyr Kinase
GyrB
FlAsH-EDT2
Relative Intensity units
A
Figure 4
ACCEPTED MANUSCRIPT
Supplementary material
•
Supplementary Experimental Procedures
•
Supplementary References
•
Supplementary Figure Legends
•
Supplementary Figures S1, S2, S3 and S4
AC C
EP
TE D
M AN U
SC
RI PT
Table of contents
ACCEPTED MANUSCRIPT Supplementary Experimental Procedures Expression of recombinant proteins For expression of His6.Stag.3C.GyrB.FGFR1KD.TC {which encodes N-‐terminally hexahistidine tag, S-‐tag,
RI PT
3C protease recognition sequence, DNA gyrase B (GyrB) domain, FGFR1KD-‐3F-‐0P [457–774, (Y463, 583, 585F) (L457V, C488A, C584S)] and fluorescein arsenical hairpin binder (FlAsH) amino acid sequence (FLNCCPGCCMEP)} from pTriExTM4 (Novagen) vector backbone. Constructs were heat-‐shock transformed into E.coli strain C41 (DE3) with pTriExTM4-‐ His6.Stag.3C.GyrB.FGFR1KD.TC and pDUET-‐Cdc37-‐λPPase (which encodes lambda protein phosphatase (λPPase), an enzyme that hydrolyses phosphorylated
SC
tyrosines and Hsp90 co-‐chaperone Cdc37, which promotes the stability of kinase domains). Cells were recovered in 1 ml of SOC media for 1 h at 37 ºC, and plated on terrific broth (TB) agar plates
M AN U
supplemented with ampicillin (50 μg/mL), spectinomycin (50 μg/mL) and 10 mM glucose. Colonies were inoculated into 500 ml of TB with ampicillin (50 μg/mL) and spectinomycin (50 μg/mL) and grown to an OD600 of 1.0. Cultures were cooled to 15°C and expression was induced with 100 μM IPTG for approximately 16 hours. Cells were harvested by centrifugation and frozen at -‐80 ºC until processed.
600 μL of Ni2+-‐NTA beads (Qiagen) were added to the extract and the mixture was incubated with
TE D
agitation for 1 h at 4 ºC. Beads were collected by centrifugation (10 min, 1000 g). The beads were three times resuspended in 30 mL wash buffer (20 mM Tris-‐HCl, 30 mM imidazole, 300 mM NaCl, pH 8.0) and spun down at 1000g. Subsequently, the beads were resuspended in 10 mL of wash buffer and transferred to a column. The protein was eluted with 3 ml of wash buffer supplemented with 200 mM
EP
imidazole and further purified by size-‐exclusion chromatography employing a HiLoad 16/60 Superdex 75 Prep Grade column (GE Life Sciences) at a flow rate of 1 mL/min (buffer: 20 mM Tris-‐HCl, 100 mM NaCl,
AC C
2 mM EDTA, pH 7.4). Fractions containing the protein were pooled and concentrated with an Amicon Ultra-‐15 3 kDa MWCO centrifugal filter device (Millipore). Purified proteins were analyzed by 10% SDS-‐ PAGE and their mass confirmed by mass spectrometry (see Supplementary Information).
Purification of recombinant proteins For purification of protein domains expressed in C41 (DE3) or BL21 (DE3). Pellets derived from 1 L cultures were thawed on ice and resuspended in 20 ml of chilled lysis buffer (25 mM Tris-‐HCl, 250 mM NaCl, 40 mM imidazole, 10 mM benzamidine, 100 μg/ml lysozyme, pH 8.0). Resuspension was accomplished by placing the pellets on an orbital shaker at 150 rpm at 4 °C for 30 minutes. A solution of 10% (v/v) Triton-‐X-‐100 and 1 K unitz of bovine pancreatic DNAse I was added and the lysate was placed
ACCEPTED MANUSCRIPT on the orbital shaker at 150 rpm at 4°C for 1 hour. Cell lysates were clarified by centrifugation at 13,000 rpm at 4°C for 1 hour in an SS34 rotor (Sorvall). The clarified lysate was loaded onto a 5-‐ml HisTrap (HiTrap Chelating HP, GE Healthcare) on an Akta Explorer system (GE Healthcare) using His Buffer A (25 mM Tris-‐HCl, 500 mM NaCl, 40 mM Imidazole, 1 mM TCEP, pH 8.0). Non-‐specific binding of other proteins was removed by washing the column with 10 column volumes of His Buffer A. His-‐tagged
RI PT
proteins were collected by eluting with linear imidazole gradient from His Buffer A to His Buffer B (25 mM Tris-‐HCl, 500 mM NaCl, 500 mM imidazole, 1 mM TCEP, pH 8.0) over 5 column volumes. Protein concentrations from the eluent were measured with a Nanodrop (ThermoScientific). The pooled fractions containing the target protein were further incubated with 10 μg of protease per 1 mg of protein (Ulp1 protease for His6-‐Sumo, and 3C protease for His6-‐3C) overnight at 4◦C to cleave of the N-‐
SC
terminal fusion tag. The protein/protease mix was dialysed overnight at 4 °C against 500 ml of Dialysis Buffer (25 mM Tris-‐HCl, 150 mM NaCl, 10 mM Imidazole and 1 mM TCEP, pH 8.0). Separation of the
M AN U
cleaved tag was achieved by reverse HisTrap chromatography over the 5 ml HisTrap column and material that did not bind was collected. Subsequently, proteins were further purified by strong quaternary ammonium (Q) anion exchange chromatography on a 5 ml HiTrap Q column (GE Healthcare) in Q Buffer A (25 mM Tris-‐HCl, 20 mM NaCl, 1 mM TCEP, pH 8.0) and eluted with a linear gradient of 10– 20 column volumes to 50% of Q Buffer B (25 mM Tris-‐HCl, 1 M NaCl, 1 mM TCEP, pH 8.0). Fractions containing the target protein were pooled, concentrated in spin concentrators (Vivaproducts) and then
TE D
further purified by size exclusion chromatography on a Superdex 200 10/300 GL column (GE Healthcare) using Gel Filtration Buffer (25 mM Tris-‐HCl, 150 mM NaCl, 1 mM TCEP, pH 8.0). Fractions were collected, combined and concentrated in spin concentrators, snap frozen in liquid nitrogen and stored at -‐80 °C. Purified proteins were analyzed by 10% SDS-‐PAGE.
EP
Cloning, expression and purification of recombinant proteins
AC C
PCR reactions were carried out on G-‐STORM thermal cycler, followed by digests using DpnI (NEB). The final reaction was transformed into E.coli XL10-‐Gold (Stratagene). Colonies were picked, plasmid prepared (QIAprep Spin Miniprep Kit, Qiagen) and sequenced after manipulation (BigDye3.1, ABI PRISM) to ensure sequence fidelity. Kinase assays In vitro kinase assays of recombinant proteins were carried out using ADP-‐Glo™ Kinase Assay (Promega).The assays were carried out at 25°C in 40 mM Tris-‐HCl pH 8.0, 20 mM MgCl2, 20 mM NaCl, 100 μM TCEP, and 100 μM Na3VO4, in a total volume of 60 μl (15 μl kinase reaction;15 μl ADP-‐Glo™
ACCEPTED MANUSCRIPT Reagent; 30 μl Kinase Detection Reagent) in solid white, flat-‐bottom 96-‐well plates. Poly (E4Y1) peptide (Sigma-‐Aldrich) and PLCγSA WT were used as substrates. The latter is an appropriate substrate since during physiological activation of PLCγ by FGFR, the interaction of activated FGFR1 is centred on the nSH2 of PLCγ, while phosphorylation occurs on residue Y783. Tyrosine phosphatase inhibitor sodium orthovanadate was used in the buffer system to preserve the protein tyrosyl phosphorylation state and
RI PT
allow activation of FGFR1KD in the absence of ligand HSPG (Gherzi 1988, Posner 1994). Kinase reactions were stopped after 30 minutes of incubation for enzyme titration and autophosphorylation reactions, and after 60 minutes for inhibitor dose response titrations. ATP-‐to-‐ADP standard conversion curves, Z-‐ values and signal-‐to-‐background ratio were calculated according to the manufacturer’s published procedure in order to assess the linearity of the assay and to calculate the amount of ADP produced
SC
during enzymatic titrations. Z' values were calculated for 5 – 100% ATP conversion using 50 µM ATP. For determinations of Km of the ATP (Km, ATP) concentration that should be used, reactions contained a
M AN U
peptide concentration of 0.2 μg/μl and ATP at varying concentrations of 10–240 μM. The assay signal in the presence of increasing ATP concentrations was measured and fitted to the Michaelis–Menten equation. The optimal kinase amount was determined when 50% of substrate conversion was achieved (half maximal reaction velocity, ½Vmax), and assays were carried out at the apparent Km, ATP, at 50 µM ATP. Negative control experiments were performed in the absence of ATP, substrate, enzyme and positive controls were performed with FGFR1KD WT (data not shown). Luminescence output was
TE D
recorded using BMG Labtech FLUOstar Optima plate reader at 520nm. Apparent IC50 values were calculated by detecting the decrease in phosphorylation of Poly (E4Y1) upon kinase inhibition using AZD4547 and PD173074. Curve fitting for inhibitor dose response titration was performed using GraphPad Prism® sigmoidal dose-‐response (variable slope) software. Apparent IC50 values were then
EP
used to measure the approximate inhibitor constant value (Ki) using the Cheng–Prusoff equation, which describes the dependencies between IC50 value and ATP concentration for ATP-‐competitive inhibitors.
AC C
Error bars indicate the SDs of three replicates.
SDS-‐PAGE and Western blotting Sample buffer (4x stock: 62.5 mM Tris.HCl, 2% SDS, 25% glycerol, 0.01% Bromophenol blue, 0.25 M DTT, pH 6.8) was added to protein samples to 1x. Protein denaturation was performed at 100°C for 5 minutes for optimal sample migration. Protein samples were separated by electrophoresis through sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-‐PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane (equilibrated in methanol) (Millipore) using a standard wet transfer method (buffer: 50 mM Tris-‐HCl, 380 mM glycine and 20% methanol). For pull-‐down assays, proteins were visualised by Amido Black staining. For immunoblotting, the membranes were blocked with 5% non-‐fat
ACCEPTED MANUSCRIPT dry milk (Marvel) in TBS pH 7.5 containing 0.1% Tween-‐20 (Sigma-‐Aldrich) (TBS-‐T) for 30-‐60 min with agitation. Membranes were incubated with the indicated primary antibodies for 1 hour at room temperature or overnight incubation at 4°C. Membranes were washed three times for 5 min in TBS-‐T followed by incubation with required conjugate diluted in 5% milk/TBS-‐T for 1 hour with agitation. Blots were subject to three washes in TBS-‐T and developed using with ECL prime detection kit (Pierce) and
RI PT
exposed to Hyperfilm ECL (Amersham Biosciences).
In-‐gel fluorescence of GyrB.FGFR1KD.TC with FlAsH-‐EDT2 or ReAsH-‐EDT2
FlAsH-‐EDT2 (10 – 100 μM) or ReAsH-‐EDT2 (10 – 50 μM) were added to Laemmli sample buffer [4x stock: 62.5 mM Tris.HCl pH 6.8, 2% SDS, 25% glycerol, 0.01% Bromophenol blue, 50 mM TCEP (neutralized
SC
with 3.5 equiv of Tris.HCl pH 8.0)]. The protein and sample buffer were kept at room temperature for 15-‐30 min and loaded onto the SDS-‐PAGE that was run as usual. The protein-‐FlAsH complexes were
M AN U
visualized on the unstained or unfixed gel by illumination with either an ultraviolet light-‐box or by using a CCD camera with appropriate filters for fluorescence of FlAsH (excitation: 511 nm, emission: 527 nm) or ReAsH (excitation: 593 nm, emission: 608 nm). Native Mass Spectroscopy
Samples were analysed by HPLC, using a Waters Premier quadrupole time of flight (Q-‐Tof) system with a
TE D
Waters 2795XC Separations Module and Waters Symmetry ® C18 column (2.1x100 mm, 3.5 μm). Mobile phase A (water + 0.1% formic acid) and mobile phase B (acetonitrile + 0.1% formic acid) were used with a flow rate of 600 μL/min, column temperature of 40 oC and injection volume of 5 – 10 μL (~20 μg of protein at ~1 mg/mL [5 μM]). Positive Ion Electrospray ionization mode on a Waters Micromass ® Q-‐Tof
EP
micro ™ was in operation. Data processing was performed using Advion ‘Chipsoft’ software.
AC C
Supplementary References
Gherzi, R. et al. DIRECT MODULATION OF INSULIN-‐RECEPTOR PROTEIN TYROSINE KINASE BY VANADATE AND ANTI-‐INSULIN RECEPTOR MONOCLONAL-‐ANTIBODIES. Biochemical and Biophysical Research Communications 152, 1474-‐1480, doi:10.1016/s0006-‐291x(88)80452-‐2 (1988). Posner, B. I. et al. PEROXOVANADIUM COMPOUNDS -‐ A NEW CLASS OF POTENT PHOSPHOTYROSINE PHOSPHATASE INHIBITORS WHICH ARE INSULIN MIMETICS. Journal of Biological Chemistry 269, 4596-‐ 4604 (1994).
ACCEPTED MANUSCRIPT Supplementary figure legends Supplementary Figure S1. (A) SEC molecular weight standards as determined on a Superdex 200 column. (B) Elution profile of the GyrB-‐FGFR1KD.TC monomer as determined on a Superdex 200 column.
RI PT
Supplementary Figure S2. Specific and quantitative labelling of GyrB.FGFR1KD.TC upon incubation with FlAsH-‐EDT2 demonstrated by SDS-‐PAGE (Coomassie staining and in-‐gel fluorescence).
Supplementary Figure S3. (A) (B) Inhibitor dose response titration of GyrB.FGFR1KD.TC in the presence of 50 μM ATP, 0.2 μg/μl PolyE4Y1 peptide and inhibitor PD173074 at varying concentrations of 0.8 – 500
SC
nM (FGFR1 IC50 = ~22 nM). Kinase inhibition was measured by detecting the decrease in phosphorylation of PolyE4Y1 after 1 hour. Maximal activity of GyrB.FGFR1KD.TC and background signal
M AN U
were measured in the absence of inhibitor PD173074.
Supplementary Figure S4. Complex formation of FGFR1KD with mTFP.nSH2 and artificial dimerization as observed by chromatography
(A) Complex formation; Elution profiles of the GyrB.FGFR1KD.TC-‐mTFP.nSH2 complex (blue) and the
excess nSH2 (pink) with 50 μM ATP/20 mM MgCl2 as determined on a Superdex 200 column. (B) Elution profiles of the GyrB.FGFR1KD.TC dimer (green) and the GyrB.FGFR1KD.TC monomer (yellow)
EP
AC C
TE D
with 125 μM coumermycin as determined on a Superdex 200 column.
ACCEPTED MANUSCRIPT
Aggr
kDa
670 158 44 17
A280nm (mAU) 5
10 15 Volume (mL)
20
25
670 158 44 17
1.3
60 40 20 0
0
5
10 15 Volume (mL)
20
25
M AN U
SC
0
Aggr
kDa
100
0
Superdex 200 10/300 GL
1.3
200
EP
TE D
Supplementary Figure S1
AC C
A280nm (mAU)
B
Superdex 200 10/300 GL
RI PT
A
ACCEPTED MANUSCRIPT GyrB.FGFR1KD.TC
+
ReAsH-EDT2 (μM)
10 50 50
+
-
+ 0
75 50
RI PT
37 Coomassie 75
37
M AN U
Fluorescence
SC
50
AC C
EP
TE D
Supplementary Figure S2
ACCEPTED MANUSCRIPT
B Luminescence, RLU
4000
1500
3000 2000
1000
500
1000
0
0
1 2 log [PD173074], nM
SC
0.5
M AN U
-1.5 -1 -0.5 0 log [GyrB.FGFR1KD.TC], μM
TE D
-2
EP
0
RI PT
5000
AC C
Luminescence, RLU
A
Supplementary Figure S3
3
ACCEPTED MANUSCRIPT
670 158 44 17
A280nm (mAU)
40 20
5
10 15 Volume (mL)
20
25
60 40 20 0
670 158 44 17
1.3
0
5
10 15 Volume (mL)
20
AC C
EP
TE D
M AN U
0
Aggr
kDa
60
0
Superdex 200 10/300 GL
1.3
RI PT
Aggr
kDa
A280nm (mAU)
B
Superdex 200 10/300 GL
SC
A
Supplementary Figure S4
25