Cell lines used:
DAOY human MB cells were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). DAOY cells were cultured as described in 4. Cell line authentication and cross contamination testing was performed by Multiplexion by single nucleotide polymorphism (SNP) profiling. ONS-76 cells were generously provided by Michael Taylor (SickKids, Canada). RT-112 (ACC418) was purchased from the Leibnitz-Institut DSMZ (Braunschweig, Germany). AGS (ATCC CRL-1739), DMS114 (ATCC CRL-2066), HCT116 (ATCC CCL-247), M059K (ATCC CRL-2365), SK-OV-3 (ATCC HTB-77), SNU-16 (ATCC® CRL5974) and SW 780 (ATCC CRL-2169) were purchased from LGC Standards GmbH, Wesel, Germany) and cultured according to LGC instructions.
Spheroid invasion assay (SIA) and automated cell dissemination counter (aCDc):
1000 - 2500 cells/well in 100 μl were seeded in 96 well Corning® Spheroid microplate (CLS4520, Sigma-Aldrich) or in cell-repellent 96 well microplate (650790, Greiner Bio-one). SIA and analysis was performed as described before 32. In brief: After spheroid formation (24 – 72 h after seeding), 70 µl medium was removed and replaced with the collagen mixture (2.5 mg/ml Pure Col Collagen I (Advanced Biomatrix)), DMEM 1x (from 10x stock, Sigma, D2429) and 0.4% Sodium bicarbonate (Sigma, S8761)), resulting in a final collagen concentration of 1.75 mg/ml. Embedded spheroids were stimulated with bFGF (100 ng/m)l, HGF (30 ng /ml) or EGF (20 ng/ml) and distance of invaded cells quantified 24 – 48 h after embedding.
3D-Cell viability (Cell TiterGlo) assay:
Cell viability was determined using CellTiter-Glo® 2D or 3D cell viability assays (#G9242, #G9682, Promega). 500 cells/25 µl were seeded in flat bottom (#781091, Greiner bio-one) or U-low adhesion (#4516, Corning) 384-well plate, 24 h prior to treatment. Increasing concentrations of compounds are deposited on cells using a HP Digital Drug Dispenser with DMSO total volume normalization. After 48 h, the CellTiter-Glo® 2D or 3D reagent was added (volume/volume) following manufacturer’s instructions. Plates were incubated at RT (room temperature) under agitation for 30 min and luminescence representing the number of viable cells was quantified with a Cytation 3 imaging reader (BioTek®). Experiments were performed independently three times with three technical replicas each.
Immunoblotting (IB):
Cells were serum starved o.n. and 10 µM compound or 1 µM (BGJ398) were added 3 h before stimulation with 100 ng/ml bFGF for 15 min. Cells were lysed in RIPA buffer containing protease inhibitor cocktail and lysates resolved by SDS-PAGE. Membranes were probed with primary antibodies against phospho-FRS2, FRS2, phospho-ERK1/2, ERK1/2, pAKT, AKT and tubulin. Integrated density of immuno-reactive bands was quantified using Adobe Photoshop CS3. Integrated densities of phospho bands relative to non-phospho bands of same protein were calculated and plotted as fold-change relative to untreated control conditions.
RNA expression analysis by RT-qPCR:
Cancer cells were seeded in 6-well plate for 24 h to reach 80-90% of confluency the following day. RNAs were extracted with the RNAeasy® plus mini kit (#74136, Qiagen) according to manufacturer’s instructions. 150 ng of RNA were used for reverse transcription in 20 µl reaction containing RNAse inhibitor and using the high-capacity cDNA reverse transcription kit (#4374967, Applied Biosystems, ThermoFisher Scientific). qPCR was performed on cDNA with the TaqManTM gene expression master mix (#4369016, ThermoFisher Scientific) using a 7900 HT fast real-time PCR system (Applied Biosystems). The relative expression level represented by the relative cDNA level was determined according to the standard curve method with a reference sample. Experiments were performed independently three times with two technical replicas.
Quantification and Statistical Analysis
Mean ± SEM are shown when means of three biological replicas are compared, mean and SD when three technical replicas are compared. Unpaired student’s t-test was used to test significance of differences between two samples. For all other analyses, one-way ANOVA repeated measures test using Bonferroni's Multiple Comparison with Prism software was performed. P-values or adjusted P-values < 0.05 were considered significant (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001). Where indicated, asterisks show statistical significances between control and test sample.
Chemical libraries:
F-Series library:
Collection of commercially available fast-delivery compounds of the suppliers Asinex Corp., ChemBridge Corp., ChemDiv, Enamine, Specs, UkrOrgSyntez Ltd. And Vitas-M Laboratory Ltd.
E compound series library:
PrestwickChemicalLibrary: Collection of 1280 FDA-approved compounds.
NCCR54k: Collection of around 54k compounds intended to represent the commercially-available chemical space.
ChemicalDversityExtension: Collection of around 14k compounds that extends NCCR54k with molecule that were initially excluded for various reasons (e.g. molecular weight, number of stereo-centers)
PPI: Collection of around 5k compounds targeted at disrupting protein-protein interactions.
NaturalProducts: Collection of around 2.5k commercially available natural compounds. These compounds are enantiomerically pure but their stereo chemistry is not always known. Therefore, several of them appear to have identical structure (because of the unspecified stereocenters) but the suppliers guarantee that different catalog IDs correspond to different compounds.
In silico screening methods:
Relevant PPIs were analyzed by iPRED analysis based on the only known structure of the FRS2-PTB 1XR0. Calculations were performed via the online available iPRED-tool. Virtual screening was performed with LigandScout based on PDB 1XR0. The co-crystallized peptide was removed, the surface analyzed for pockets by the in-built pocket calculator. Two shallow pockets the first formed by the alpha-helix (A94-M105) and the beta-sheet (L33-L47), the second being complementary to the FGFR-peptide sequence Ipep13-Rpep16 (FGFR_PEP: HSQMAVHKLAKSIPLRRQVTVS), were identified and isolated. Amino acids were charged at pH 7 according to their pKA and energy minimized. Apo Site grids were calculated with a maximum of 4 H-bond acceptors, 3 H-bond donors, 2 positively ionizable groups, 1 negatively ionizable group, 2 aromatic groups and 4 hydrophobic moieties. Accessibility was considered in the calculation. The generated pharmacophore was run against a database containing all fast-delivery compounds of the suppliers Asinex Corp., ChemBridge Corp., ChemDiv, Enamine, Specs, UkrOrgSyntez Ltd. And Vitas-M Laboratory Ltd. with an overall compound number of 3.5 million (F-series) and against a collection of compounds from libraries listed above (E-series). Screening was performed under the Pharmacophore Fit-model with maximally 4 pharmacophoric features being omittable in order to be considered. Hits from both pockets were then combined, ranked by the provided score and selected based upon chemical diversity. Fragment-based screening and docking was performed with LigandScout vers. 4.2.1 (Inte:Ligand, Vienna, Austria) and data analyzed with Microsoft Office Excel 365 (Microsoft Corp., Seattle, USA).
Nano-Differential Scanning Fluorimetry (nanoDSF):
Assay establishment and pilot nanoDSF experiments were performed with a Tycho® system. Each experiment was performed by mixing the protein and ligand of interest at a 1:1 ratio and subsequent incubation for 15-30 min at room temperature. Controls of protein alone and ligand alone in identical buffer compositions as for the complex were measured accordingly. The temperature gradient ranged from 35 to 95°C, and the heating rate was 30°C/min. Inflection temperatures TI were calculated from the first derivative of the spectra obtained at the ratio of the signals at 350 nm to those at 330 nm. ΔTI of a complex was calculated as following: ΔTI,complex = TI,complex – TI,protein
The in-built software version 1.1.5.668 (NanoTemper Technologies, Munich, Germany) and Microsoft Office Excel 365 (Microsoft Corp., Seattle, USA) was used for data analysis. Data was plotted with GraphPad Prism vers. 8.3.0 (GraphPad Software, San Diego, USA).
Purified FRS2_PTB or GB1_FRS2_PTB protein tagged with 6x Histidine residues and Guanine nucleotide-binding protein subunit beta (GB1) were diluted in the protein buffer (100 mM sodium phosphate, 50 mM NaCl, 0.5 mM EDTA, 50 mM arginine, 1 mM TCEP, pH 7.0) to 30 μM, GB-1 to 40 μM. FGFR_PEP, compounds or paracetamol (PARA) as negative control were dissolved at 1 mM in the protein buffer supplemented with 10% DMSO. Compound and protein were mixed at 1:1 volume ratio yielding a final concentration of 15 μM and 20 μM for FRS2_PTB/GB1_FRS2_PTB and GB1, respectively, as well as 500 μM for FGFR_PEP or compounds. The mixture was incubated for 15 min. Each compound was measured in triplicates. Protein controls were measured in six replicates in the beginning. The quality of the screen was assessed by the Z-factor. nanoDSF validation studies were performed on a Prometheus® system in high sensitivity capillaries. Samples were heated with 1°C/min from 20 to 95°C. Each compound was measured in triplicates. GB1-FRS2_PTB and GB1 were diluted in the protein buffer (100 mM sodium phosphate, 50 mM NaCl, 0.5 mM EDTA, 50 mM arginine, 1 mM TCEP, pH 7.0) to 30 μM. Compounds F3.3, F3.4, F3.18 as well as the F3.18 analogues were dissolved in 100% DMSO at 50 or 100 mM and further diluted to 1 mM with a final DMSO-concentration of 100%. FGFR_PEP was dissolved at 1 mM in the protein buffer supplemented with 10% DMSO. Compound and protein were mixed at 1:1 volume ratio yielding final concentrations of 15 μM protein and 500 μM for the compounds. FGFR_PEP was measured against both GB1-FRS2_PTB and GB1, all other compounds only against GB1-FRS2_PTB. The mixture was incubated for 15 min before measurement. The data was analyzed for both the ratio of signals at 330 and 350 nm and at 350 nm alone.
Microscale thermophoresis (MST):
Protein labeling for MST was performed with the 2nd Generation BLUE-NHS dye. GB1-FRS2_PTB- was labelled at 20 μM with 60 μM dye and incubation for 30 min in the dark at room temperature. The labelling was performed in the protein buffer without arginine supplementation. It was rebuffered to protein buffer (100 mM sodium phosphate, 50 mM NaCl, 0.5 mM EDTA, 50 mM arginine, 1 mM TCEP, pH 7.0). The dye was subsequently removed by gravitational flow chromatography and the protein concentration determined by UV-spectroscopy.
The assay was established with the FGFR_PEP. The peptide was dissolved in protein buffer with and without 10% DMSO supplementation in a 1:1 serial dilution from 1 mM to 61.04 nM. A total volume of 10 μl of 50 nM labelled protein was added to 10 μl of the peptide dilution series for a final labelled protein concentration of 25 nM. The samples were incubated for 15 min at RT. Premium-coated capillaries were used, excitation power was set at 20%, MST-power to 40% (4 K temperature gradient) with a laser-on time of 20 s and a laser-off time of 3 s. Temperature was set to 25°C. Each measurement was repeated twice. The interaction was measured in duplicates for the non-DMSO-supplemented buffer and in triplicates for DMSO-supplemented buffer. The compounds F3.3, F3.14, F3.18 as well as the analogs 18.2, 18.7, 18.8 and 18.9 were dissolved in 100% at 50 or 100 mM and further diluted to 1 mM with a final DMSO-concentration of 100%. The compounds were diluted in a 1:1 serial dilution from 1 mM to 61.04 nM in protein buffer supplemented with 10% DMSO. 10 μl of 50 nM labeled protein was added to 10 μl of each compound dilution for a final labeled protein concentration of 25 nM and a DMSO-concentration of 5%. The samples were incubated for 15 min, and MST assessed as described above. Each measurement was repeated twice. The interaction was measured in two independent duplicates. LIFC was observed in all measurements. Subsequently, 10 μl of compounds F3.3, F3.14 and F3.18 at 2 mM, 1 mM, 500 and 250 μM were incubated with 10 μl of 50 nM dye to yield final compound concentrations from 1 mM to 125 μM in a 1:1 dilution series and 25 nM dye concentration. The mixture was then measured for their initial fluorescence once.
MST-experiments were performed with MO.Control vers. 1.6 (NanoTemper Technologies, Munich, Germany) and data analyzed with MO.Analysis vers. 2.3 (NanoTemper Technologies, Munich, Germany). Data was plotted with GraphPad Prism vers. 8.3.0 (GraphPad Software, San Diego, USA).
Expression and Purification:
The PTB domain of human FRS2, residues 15 to 135, were cloned into a pEM3BT2 vector as a fusion with an N-terminal 6x histidine tag, GB1 with or without FGFR1 peptide (FGFR_PEP HSQMAVHKLAKSIPLRRQVTVS) and was then used to transform BL21 (DE3) cells for plasmid DNA amplification and protein expression. To induce protein expression, the transformed clones were cultured at 37°C in labelled M9 media supplemented with 100 µg/ml of ampicillin using 15N-NH4Cl and 13C-glucose as the sole sources of 15N and 13C. When the culture reached an OD600 of 0.6 to 0.8, protein expression was induced by adding 1 mM of IPTG at 25°C overnight.
The bacterial cells were collected by centrifugation at 5000 x g for 15 min and the pellets lysed by adding 10 mg/ml of lysozyme and sonication. The lysates were then centrifuged for 30 min at 18,000 rpm at 4°C. The supernatants were filtered using a 0.2 µm membrane and passed on an equilibrated, 5 ml Ni-NTA column using ÄKTA prime at 4°C. The protein/resin complex was then washed with 5 column volumes each using low salt wash buffer and high salt wash buffer. The protein was eluted from the resin using the elution buffer containing 100 mM Tris, 50 mM NaCl and 500 mM (pH 7.0) imidazole. Following SDS-PAGE analysis of the eluted fractions, the protein was dialyzed overnight against the buffer containing 50 mM phosphate buffer, 50 mM NaCl, 1 mM TCEP and 0.5 mM EDTA, pH 7.0 at 4°C. The protein fractions were then pooled and concentrated to 250 µM using Amicon Ultra – 4 (10 KDa cutoff) centrifugal filters. The final protein concentration was determined by measuring absorbance at 280 nm using a nanodrop instrument and further used for NMR analysis.
NMR spectroscopy:
NMR experiments were recorded using 250 µM solutions of proteins in 50 mM phosphate buffer, 50 mM NaCl, 1 mM TCEP and 0.5 mM EDTA, pH 7.0. All NMR experiments were recorded at 310 K on Bruker Avance-Neo 600 or 700 MHz spectrometers equipped with cryoprobes. For backbone assignment of the GB1-FRS2_PTB_FGFR_PEP fusion protein a ([79% 2H, 99% [13C,15N])-labeled sample was used. Experiments were selected from the Bruker standard pulse sequence library. [15N,1H]-HSQC experiments were of the sensitivity-enhance type50 and comprised spectral widths of 14(F2, 1H) and 36 (F1, 15N) ppm with 1024*128 complex data points. HNCACB / HN(CO)CACB spectra were used to link sequential amide groups via Cα/Cβ resonances. In addition, sequential connection was confirmed via common carbonyl resonances using the HNCO and HN(CA)CO experiments. The usage of the HN(CACO)NH experiment proved to be particularly useful by establishing sequential contacts in cases where the Cα/Cβ correlations were missing by directly establishing sequential 15N connectivity. All spectra were processed in TOPSPIN using linear prediction and spectra were analyzed in the program CARA.
STD and WaterLOGSY experiments were measured at 7ºC on a Bruker AvNEO 600 spectrometer equipped with a TCI cryoprobe. In the STD experiments saturation was achieved by applying a series of Gaussian-shaped pulses at a RF field strength of 54 Hz (max.) for 3 s at -0.5 ppm (-40 ppm for the reference experiment) and 512 scans were accumulated. The WaterLOGSY was measured with 256 scans and a mixing time of 1.5 s. A 7.5 ms Gaussian-shaped 180° pulse was used to control the water magnetization. For both experiments samples of 25 µM GB1-FRS2_PTB with a 20-fold excess of ligand (500 µM) were used or with the ligand alone for the control experiment.
Thermal Protein Profiling (TPP) in intact cells:
Two 15 cm dishes with 0.5 ´ 106 DAOY cells per dish were grown in full medium o.n. Prior to compound treatment, cells were washed once in PBS. Complete growth medium containing 10 µM compound was then added and cells incubated for another 90 min at regular culture conditions. Medium was removed, cells washed with PBS containing 10 µM compound and detached using trypsin containing 10 µM compound. Trypsin was neutralized using regular growth medium containing 10 µM compound and cells were collected by centrifugation (300 x g, 25°C, 5 min). After one more wash in PBS containing 10 µM compound, cell pellet was resuspended in 220 µl PBS containing 10 µM compound. Ten 20 µl aliquots were dispensed in a pre-cooled PCR plate, which was sealed with aluminum foil and placed on ice until heat treatments. Heat treatment was performed in a gradient PCR machine. The following temperatures were used: 40.5, 43.6, 46.2, 48.8. 51.2, 53.8, 56.4, 58.3, 61.7 and 73.8°C for three min. Plate with samples was kept at RT for 3 min and then kept on ice until centrifugation for 2 min at 500 g at RT.
Cells were lysed by adding 30 µl TPP lysis buffer (PBS, 1.33% IGEPAL, 1.66 mM MgCl2, 1.66 ´ cOmplete protease Inhibitors, 1.66 ´ PhosStop, 416.6 U/ml benzonase) per sample, resulting in a final concentration of 0.8% IPEGAL, 1 mM MgCl2, 1 ´ cOmplete protease Inhibitors, 1 ´ PhosStop, 250 U/ml benzonase. Samples were incubated 1h at 4°C, with shaking at 500 rpm. Samples are filtered using a MultiScreenHTS HV-Filterplate (Merck, MSHVN4510). Membrane wells were prewetted with 50 µl PBS and centrifuged at 2000 x g, RT, 3 min. The lysates were centrifuged at 2000 g, RT, 3 min and 40 µl of each supernatant was transferred to the filter plate and centrifuged at 500 g at RT for 5 min. Flow-through of samples was collected in a fresh 96-well plate and kept on ice. 20 µl of samples was snap-frozen and stored at -80°C for further analysis by mass-spectrometry. 5 µl was used for protein concentration determination using BCA assay.
Thermal Protein Profiling (TPP) in lysates:
Four 15 cm dishes with 0.5 ´ 106 DAOY cells per dish were grown in full medium o.n. Medium was then removed, cells washed with PBS, detached using trypsin, collected in regular growth medium, centrifuged (300 g, 25°C, 5 min), washed with PBS and resuspended in 550 µl ice-cold PBS. Cells were lysed with three freeze-thaw cycles (freeze in liquid N2, thaw in 25°C heat block, vortex quickly) and lysates were placed in aliquots of 250 µl on ice. Compound or solvent was added to 10 µM final concentration and samples were incubated 20 min at 4°C under rotation. Ten 20 µl aliquots were dispensed in a pre-cooled PCR plate, which was sealed with aluminum foil and placed on ice until heat treatments. Heat treatment was performed in a gradient PCR machine with the following temperatures 40.5, 43.6, 46.2, 48.8. 51.2, 53.8, 56.4, 58.3, 61.7 and 73.8°C for three min. Plate with samples was kept at RT for 3 min and then kept on ice until centrifugation for 2 min at 500 x g at RT.
The soluble fraction of the lysates was recovered by addition of 30 µl TPP lysis buffer (PBS, 1.33% IGEPAL, 1.66 mM MgCl2, 1.66 ´ cOmplete protease Inhibitors, 1.66 ´ PhosStop, 416.6 U/ml benzonase) per sample, resulting in a final concentration of 0.8% IPEGAL, 1 mM MgCl2, 1 ´ cOmplete protease Inhibitors, 1 ´ PhosStop, 250 U/ml benzonase. Samples were incubated 1h at 4°C, with shaking at 500 rpm. Samples are filtered using a MultiScreenHTS HV-Filterplate (Merck, MSHVN4510). Membrane wells were prewetted with 50 µl PBS and centrifuged at 2000 g, RT, 3 min. The lysates were centrifuged at 2000 g, RT, 3 min and 40 µl of each supernatant was transferred to the filterplate and centrifuged at 500 g at RT for 5 min. Flow-through of samples was collected in a fresh 96-well plate and kept on ice. 20 µl of samples was snap-frozen and stored at -80°C for further analysis by mass-spectrometry. 5 µl was used for protein concentration determination using BCA assay.
Cellular thermal shift assay (CETSA) using IB:
For a cellular thermal shift assay (CETSA) run, 7.5 µl of the flowthrough samples of the TPP preparation was mixed with 3 µl 4x Laemmli buffer containing 50 mM DTT and boiled for 5 min at 95°C. 10 µl of each sample was assayed by immunoblot using antibodies against FRS2 and beta-tubulin.
Mass spectrometry – ESI MS:
Sample preparation: Protein samples were processed using a modified SP3 clean-up and digestion procedure on a KingFisher Flex (Thermo). In short, protein samples were reduced/alkylated using 2 mM TCEP, 15.6 mM CAA for 30 min at 60°C. Subsequently, samples were adjusted to 50% Ethanol and bound to 100 μg magnetic beads. Bound proteins were washed three times with 80% ethanol and on-bead digested overnight in 50 mM TEAB including Trypsin/Lys-C (1:50, enzyme:protein) at 37°C. Resulting peptides were recovered and dried to completeness.
For isobaric labelling, peptides were resuspended in 20 µl 50 mM TEAB and 0.1 mg TMT reagent was added in 5 µl anhydrous ACN and incubate for 1h at room temperature with shaking. Unused labelling reagent was quenched by adding 1.25 µl of 5% hydroxylamine solution and incubation for 15 min at RT. Quenched samples were pooled and cleaned using C18 solid phase extraction and subsequently dried down.
High pH reversed-phase (RP) peptide fractionation: Peptides were dissolved in 100 μl of 9 mM ammonium formate, and 90 μl were loaded onto a XBridge Peptide BEH C18 column (Waters, 130 Å, 3.5 µm, 1 mm x 100 mm) on an Agilent 1200 HPLC. The samples were separated into 64 fractions using a gradient of 2% ACN, 9 mM ammonium formate (pH 10) to 40% ACN, 9 mM ammonium formate (pH 10) in 60 min at a flow rate of 1 ml/min and concatenated into 8 fractions before drying the peptides to completeness.
LC-MS Data Acquisition: LC-MS analysis of pooled high-pH RP fractions was conducted on a Q Exactive HF (Thermo Fisher Scientific) mass spectrometer operated in-line with an ACQUITY UPLC M-Class (Waters). A 75 µm forward-trap elute configuration was used for peptide separation on a 25 cm x 75 μm, 1.8 μm HSS T3 analytical column (Waters). The separating linear gradient covered 5% ACN, 0.1% FA to 35% ACN, 0.1% FA in 90 min at a flow rate of 400 nl/min. Mass spectra were essentially recorded in data dependent mode (top25). MS2 spectra were acquired by isolating peptide precursor ions at 0.7 Da, followed by HCD fragmentation at an NCE of 33 (AGC target: 1e5, maxIT: 87 ms, R: 45000). Dynamic exclusion was set to 30 s and charge states of the type unassigned, 1, ≥6 was ignored.
Data analysis: Peptide and protein identification, as well as protein-level quantification relative to the lowest temperature (TMT channel 126), was conducted using Proteome Discoverer 2.5.0.4. Melting curve fitting was conducted using the Bioconductor TPP package 3.13.
In vitro absorption, distribution, metabolism, and excretion – toxicity (ADMET) assays:
In vitro ADME-T assays were performed by Cyprotex (Cyprotex Discovery, Cheshire, UK). Semi-Thermodynamic solubility: The shortlisted hits were added to a 96-well plate in quadruplicate and 1 x PBS was added to give a maximum concentration of 100 µM. The solution was agitated at ambient temperature overnight. The solutions were then centrifuged for 30 min at 3000 rpm at RT. The supernatant was removed and centrifuged for a further 30 min under the same conditions. An aliquot of the resulting supernatant was diluted in 50% methanol in water (containing an internal standard for MS analysis) prior to analysis by LC-MS/MS. A standard curve was produced by diluting the 10 mM DMSO stock with DMSO to give concentrations of 1 mM and 0.1 mM. These stocks were diluted in in 50% methanol in water (containing an internal standard for MS analysis) prior to analysis by LC-MS/MS. The solubility of the shortlisted hits was calculated from a linear or quadratic fit of the standard curve.
Microsomal metabolic stability: Pooled liver microsomes were purchased from Xenotech (H0500) and were stored according to manufacturer’s instructions prior to use. Microsomes (final protein concentration 0.5 mg/ml), 0.1 M phosphate buffer pH 7.4 and shortlisted hits (final substrate concentration 1 µM; final DMSO concentration 0.25%) were pre-incubated at 37°C prior to addition of NADPH (final concentration 1 mM) to initiate the reaction. A minus co-factor control incubation was included for each compound tested where 0.1 M phosphate buffer pH 7.4 was added instead of NADPH. Two control compounds were included with each species. All incubations were performed singularly for each shortlisted hit. Each compound was incubated for 0, 5, 15, 30 and 45 min. The minus co-factor was incubated for 45 min only. The reactions were stopped by transferring incubate into acetonitrile at the appropriate time points in a 1:3 ratio. The termination plates were centrifuged at 3000 rpm for 20 min at 4°C to precipitate the protein. Following protein precipitation, the supernatant were combined in cassette of up to 4 compounds, internal standard for MS analysis was added and samples were analyzed using LC-MS/MS. From a plot of In peak area ratio (compound peak area/internal standard peak area) against time, the gradient of the lines were determined. Subsequently, half-life and intrinsic clearance were calculated using the following equations.
Elimination rate constant (k) = (- gradient)
Half-life (t1/2) (min) = 0.693/k
Intrinsic clearance (CLint) (µl/min/mg protein) - Vx 0.693 / t1/2
Where V = Incubation volume (µl) / Microsomal protein (mg)
CaCo-2 permeability (Bi-directional): Caco-2 cells obtained from the ATCC were used between passage numbers 40 - 60. Cells were seeded onto Millipore Multiscreen Transwell plates at 1 x 105 cells/cm2. The cells were cultured in DMEM and media was changed every two or three days. On day 20, the permeability study was performed. Cell culture and assay incubations were carried out at 37ºC in an atmosphere of 5% CO2 with a relative humidity of 95 %. On the day of the assay, the monolayers were prepared by rinsing both apical and basolateral surfaces twice with Hanks Balanced Salt Solution (HBSS) at the desired pH warmed to 37°C. Cells were then incubated with HBSS at the desired pH in both apical and basolateral compartments for 40 min to stabilize physiological parameters.
The dosing solutions were prepared by diluting the test compound with assay buffer to give a final compound concentration of 10 μM (final DMSO concentration of 1% v/v). The fluorescent integrity marker lucifer yellow was also included in the dosing solution. Analytical standards were prepared from test compound DMSO dilutions and transferred to buffer, maintaining a 1% v/v DMSO concentration.
For assessment of A-B permeability, HBSS was removed from the apical compartment and replaced with test compound dosing solution. The apical compartment insert was then placed into a companion plate containing fresh buffer (containing 1% v/v DMSO). For assessment of B-A permeability, HBSS was removed from the companion plate and replaced with test compound dosing solution. Fresh buffer (containing 1% v/v DMSO) was added to the apical compartment insert, which was then placed into the companion plate.
At 120 min, the apical compartment inserts and the companion plates were separated, and apical and basolateral samples were diluted for analysis.
Compound permeability was assessed in duplicate. Compounds of known permeability characteristics were run as controls on each assay plate.
Test and control compounds were quantified by LC-MS/MS cassette analysis using a 7-point calibration with appropriate dilution of the samples. The starting concentration (C0) was determined from the dosing solution and the experimental recovery calculated from C0 and both apical and basolateral compartment concentrations. The integrity of the monolayer throughout the experiment was checked by monitoring lucifer yellow permeation using fluorimetric analysis.
The permeability coefficient (Papp) for each compound was calculated from the following equation:
Papp= (dQ/dtC0 ×A)
where dQ/dt was the rate of permeation of the drug across the cells, C0 was the donor compartment concentration at time zero and A was the area of the cell monolayer. C0 was obtained from analysis of the dosing solution.
Efflux ratio (ER) was calculated from mean A-B and B-A data. This was derived from:
ER= Papp(B−A)Papp(A−B)
Three control compounds were screened alongside the test compounds, atenolol (human absorption 50 %), propranolol (human absorption 90 %) and talinolol (a substrate for P-glycoprotein).
Cell Viability (Cytotoxicity testing using HepG2): HepG2 human hepatocellular carcinoma cells were plated on 96-well tissue culture polystyrene plates for 24 h prior to dosing of the cells. The shortlisted hits were diluted in DMSO and serial dilutions are made 1% DMSO in growth media. Compounds at 8 concentrations in triplicate was then incubated for 72 h. Appropriate blanks and controls were run alongside the assay. One h prior to the end of the incubation period, the cells were loaded with MTT [yellow; 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide], the plates were dried and re-solubilized using DMSO. The plates were then scanned at 570 nm using a plate reader.
The minimum effective concentration was determined from the lowest concentration whose mean value exceeds the significance level, provided either a clear dose-response relationship was observed, or at least two consecutive concentration points were above the significance level.
In vivo bioavailability analysis with E12, F3.14 and F18.7:
In vivo bioavailability studies were performed by Pharmacology Discovery Services Taiwan, Ltd., (New Taipei City, Taiwan). Male ICR mice weighing 20 - 30 g were provided by BioLasco Taiwan (under Charles River Laboratories Licensee). Animals were acclimated for 3 days prior to use and were confirmed with good health. All animals were maintained in a hygienic environment with controlled temperature (20 - 24ºC), humidity (30% - 70%) and 12 hours light/dark cycles. Free access to sterilized standard lab diet [MFG (Oriental Yeast Co., Ltd., Japan)] and autoclaved tap water. A pharmacokinetic (PK) study was performed in male ICR mice following intravenous (IV) and oral (PO) administration of compound F18.7 and E12 and IV administration of the test compound F3.14. The three test compounds (F18.7, E12 and F3.14) were formulated in 1% dimethyl sulfoxide (DMSO)/10% Solutol® HS-15/ phosphate buffered saline (PBS) at 0.2 mg/mL and 1 mg/mL for IV and PO administration, respectively. The dosing volumes were 5 ml/kg for IV and 10 ml/kg for PO. The dose was 1 mg/kg for IV and 10 mg/kg for PO routes. The plasma samples were collected and at 3, 10, 30, 60, 120, 240, 480, and 1440 min after IV and 10, 30, 60, 120, 240, 360, 480, and 1440 min after PO administration. T1/2 was calculated as follows:
In vivo single dose and multi dose MTD studies with F18.7:
In vivo MTD studies were performed by Pharmacology Discovery Services Taiwan (New Taipei City, Taiwan) Phase 1: Single-dose maximum tolerated dose (MTD): F18.7 was administered PO to groups of 2 male and 2 female (23 ± 3 g) ICR mice. Animals received an initial dose of 30 mg/kg. If the animals had no significant adverse effects within 60 min after treatment, the dose for the next cohort was increased. If one or more animals died or had significant adverse effects within 60 min after treatment, the dose for the next cohort was decreased. The testing stopped when all animals survived at the upper bound (200 mg/kg), or lower bound (3 mg/kg) had been reached. Full clinical examinations and body weight change were assessed. At each dose level, animals were observed, and mortality was noted daily after compound administration for three days. Animals were observed for the presence of acute toxic symptoms (mortality, convulsions, tremors, muscle relaxation, and sedation) and autonomic effects (diarrhea, salivation, lacrimation, vasodilation, piloerection, etc.) during the first 60 min and again at 120 min after administration. Body weights were recorded pre-dose and at 72 hours.
The next dose level was determined by the following scheme:
30 mg/kg, if no death, 100 mg/kg, if no death, 200 mg/kg (upper bound)
30 mg/kg, if no death, 100 mg/kg, if death, 50 mg/kg
30 mg/kg, if death, 10 mg/kg, if death, 3 mg/kg (lower bound)
30 mg/kg, if death, 10 mg/kg, if no death, 17 mg/kg
Phase 2: F18.7 was administered PO at 200 mg/kg, bid x5, 1h interval, 200 mg/kg, qd x5, and 150 mg/kg, bid x5, 8-hour interval to groups of 3 female CB.17 SCID mice (7 ± 1 week-old) for assessment of possible adverse effects. Animals were observed for the presence of acute toxic symptoms (mortality, convulsions, tremors, muscle relaxation, sedation) and autonomic effects (diarrhea, salivation, lacrimation, vasodilation, piloerection) during the first 50 (after 1st daily dose) or 60 (2nd daily dose) min after each dose. Body weights were recorded once daily for 8 days. The animals were observed for mortality twice daily for 8 days. In addition, plasma samples were collected at 0.5, 1, 2 and 5.3 hour(s) after the second dose on Day 4 (150 mg/kg, PO; bid x5 group) and at 0.5, 1, 2 and 6 hour(s) after the final dose on Day 5 (200 mg/kg, PO; qd x5 group). The exposure levels (ng/mL) of F18.7 in plasma samples were then determined by LC-MS/MS. AUClast was calculated from the area under the plasma concentration-time curve from time zero to time of last measurable concentration.
In vivo PK studies with F18.7:
In vivo PK studies were performed by Pharmacology Discovery Services Taiwan (New Taipei City, Taiwan). F18.7 (30, 100, and 200 mg/kg, PO) was further administered to groups of 2 male and 2 female (23 ± 3 g) satellite ICR mice in PK study; the plasma and liver samples were harvested at 0.5 hour after administration. In addition, one additional group was dosed at 30 mg/kg and the plasma samples were harvested at 1 and 2 hour(s) after the treatment. The body and liver weights were recorded. The exposure levels (ng/ml or ng/g) of F18.7 in plasma and liver samples were then determined by Liquid Chromatograph Tandem Mass Spectrometer (LC-MS/MS), and the plasma: liver ratios were calculated.
Anti-tumor activity in mouse models:
In vivo mouse models were generated and experiment with SK-OV-3 and AGS models performed by EPO, experimental pharmacology & oncology Berlin-Buch GMBH (Berlin, Germany). AGS model: AGS cells were cultured in F12-K medium (10%FCS, 1%PS) until P11. Freshly isolated cells (1x106) were subcutaneously transplanted to left flank of 8-week-old female mice in PBS/Matrigel (1:1). After achieving aTV group mean of 0.139 and 0.140cm³ mice were randomly assigned to three study groups with n=6 animals (TV: min. 0.106 / max. 0.229cm³) and were daily treated with either the vehicle (1% DMSO/ 10% solutol/ PBS), F18.7 (200 mg/kg in 1% DMSO / 10% solutol / PBS) or BGJ398 (10 mg/kg, in 2:1 PEG300/D5W). Tumor diameters were determined by calliper measurements 2x weekly, body weight (BW) was measured as parameter for tolerability 2x weekly.
SKOV3 model: SKOV-3 cells were cultured in McCoy´s 5A (26600-023, ThermoFisher) media (10%FCS) until P7. Freshly isolated cells (1x106) with a viability of 98% were subcutaneously transplanted to left flank of 8-week-old female mice in PBS/Matrigel (1:1). After achieving aTV group mean of 0.138 and 0.143cm³ mice were randomly assigned to three study groups with n=8 animals (TV: min. 0.102 / max. 0.216cm³) after 48 h post cell transplantation and were daily treated either the vehicle (1% DMSO/ 10% solutol/ PBS), F18.7 (200 mg/kg in 1% DMSO / 10% solutol / PBS) or BGJ398 (10 mg/kg, in 2:1 PEG300/D5W). Tumor diameters were determined by calliper measurements 2x weekly, body weight (BW) was measured as parameter for tolerability 2x weekly.
Immunohistochemistry (IHC):
IHC of tumor sections was performed by Sophistolab (Muttenz, Switzerland) on a Lecia BondMax instrument using Refine HRP-Kits (Leica DS9800). All buffer-solutions were purchased from Lecia Microsystems Newcastle, Ltd and used according to the manufacturer’s guidelines. Paraffin-slides were de-waxed, pre-treated and incubated as follows: ER-solution 2 for 10 min at 95°C, ER-solution 2 for 20 min at 100°C and ER-solution 2 for 30 min at 100°C. pFRS2 ; Rabbit anti-Phospho FRS2 (PhosphoY 436, Abcam Limited; ab193363), dilution 1:150; phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (D13.14.4E) Rabbit mAb (Cell Signaling Technology #4370), dilution 1:1600. The IHC images were captured digitally using a Nikon Epifluorescence Eclipse Ti2 equipped with a Nikon DS-Ri2 color and a Nikon DS-Qi2 monochrome camera.