Thalidomide and its analogs differentially target Fibroblast Growth

Subscriber access provided by WEBSTER UNIV

Article

Thalidomide and its analogs differentially target Fibroblast Growth Factor Receptors: Thalidomide suppresses FGFR gene expression while pomalidomide dampens FGFR2 activity Lakshmikirupa Sundaresan, Pavitra Kumar, Jeganathan Manivannan, Uma Maheswari Balaguru, Dharanibalan Kasiviswanathan, Vimal Veeriah, Sharmila Anishetty, and Suvro Chatterjee Chem. Res. Toxicol., Just Accepted Manuscript • DOI: 10.1021/acs.chemrestox.8b00286 • Publication Date (Web): 05 Mar 2019 Downloaded from http://pubs.acs.org on March 5, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Thalidomide and its analogs differentially target Fibroblast Growth Factor Receptors: Thalidomide suppresses FGFR gene expression while pomalidomide dampens FGFR2 activity

Lakshmikirupa Sundaresan γ1,2, Pavitra Kumar γ1, Jeganathan Manivannan1, Uma Maheswari Balaguru1, Dharanibalan Kasiviswanathan2, Vimal Veeriah1, Sharmila Anishetty2, Suvro Chatterjee1,2* AU-KBC Research Centre, Anna University and 2Department of Biotechnology,

1

Anna University, Chennai

*Corresponding author: Dr. Suvro Chatterjee. AU-KBC Research Centre & Centre for Biotechnology MIT Campus of Anna University, Chromepet, Chennai-600044, India Tel.: +91 44 2223 4885 × 48 / +91 44 2223 2711 × 48; Fax: +91 44 2223 1034 E-mail address: [email protected] ORCID ID: 0000-0003-4413-9760

1 ACS Paragon Plus Environment

Chemical Research in Toxicology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 35

For TOC only THALIDOMIDE LENALIDOMIDE

Differential regulation

Differential regulation POMALIDOMIDE

FGFR signaling SOX2, SOX9, SMAD3, SMAD4 TEAD4, TCF4

FGFR2

FGFR2

FGFR1 FGFR2

FGF signaling

E2F1?

Eye development

Brain development

Limb development EMBRYO DEVELOPMENT

FGFR1 FGFR2

CANCER

Angiogenesis ENDOTHELIUM


Page 3 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract Thalidomide is an infamous teratogen and it is continuously being explored for its anti-cancer properties. Fibroblast growth factor receptors (FGFRs) are implicated in embryo development and cancer pathophysiology. With striking similarities observed between FGFR implicated conditions and thalidomide embryopathy, we hypothesized thalidomide targets FGFRs. We utilized three different cell lines and chicken embryo model to investigate the effects of thalidomide and analogs on FGFR expression. We performed molecular docking, KINOMEscan® analysis and kinase activity assays to study the drug-protein interactions. The expression of FGFR1 and FGFR2 was differentially regulated by all the three drugs in cells as well as in developing organs. Transcriptome analysis of thalidomidetreated chick embryo strongly suggests the modulation of FGFR signaling and key transcription factors. Corroboration with previous studies suggests that thalidomide might affect FGFR expression through the transcription factor, E2F1. At the protein level, molecular docking predicted all the three analogs to interact with lysine residue at 517th and 508th positions of FGFR2 and FGFR3 respectively. This lysine coordinates the ATP binding site of FGFR, thus hinting the possible perturbation of FGFR activity by thalidomide. Kinome analysis revealed that kinase activities of FGFR2 and FGFR3 (G697C) reduced by 31% and 65% respectively in the presence of 10µM thalidomide. Further, we checked and confirmed that the analogs inhibited the FGFR2 kinase activity in a dose-dependent manner. This study suggests that FGFRs could be potential targets of thalidomide and the two analogs, and also endorse the link between the teratogenicity and anti-tumor activities of the drugs.

Keywords: Fibroblast growth factor receptors, thalidomide, lenalidomide, pomalidomide, Kinase activity, teratogenicity



Page 4 of 35

1. Introduction Thalidomide, which was abandoned for its teratogenic properties, has been successfully repurposed for the treatment of multiple myeloma. Thalidomide’s teratogenic manifestations include limb deformities, craniofacial malformations, neurodevelopmental disorders and several internal organ defects (Lenz and Knapp, 1962). Thalidomide and its analogs, lenalidomide and pomalidomide are currently being tested in more than 2000 clinical trials (www.clinicaltrials.gov) suggesting huge their potential in cancer treatment. Thalidomide and its analogs, lenalidomide and pomalidomide, being small molecules, might interact with many proteins. However, very few protein targets of thalidomide are known to date. Earlier, we reported that thalidomide attenuated the activity of soluble guanylyl cyclase (sGC) (Majumder et al., 2009). Cereblon (CRBN), a protein believed to be implicated in diverse roles was the second target to be reported (Ito et al., 2010). A recent work by Rashid et al identified tubulin as the next direct target of thalidomide (Rashid et al., 2015). Thalidomide is well-known for its anti-angiogenic properties (D’Amato et al., 1994) and it does so by reducing FGF2 gene expression (Stephens et al., 2000). Thalidomide affects limb development by down-regulating the gene expression of FGF2 (Stephens et al., 2000) and FGF8 in chick (Therapontos et al., 2009) and rabbit embryos (Kawamura et al., 2014). Fibroblast growth factors (FGFs) and their corresponding receptors (FGFRs) play a major role in physiology, embryonic development and tumor pathophysiology (Turner and Grose, 2010a). The FGF family has 22 ligands known so far acting through 4 tyrosine kinase receptors namely FGFR1, FGFR2, FGFR3 and FGFR4 (Turner and Grose, 2010a). The specificity of the FGF-FGFR interactions is highly controlled in a tissue-specific manner (Partanen et al., 1991). Aberrant signalling in the FGF pathway has been implicated in more than 40 disease conditions, including breast, myeloma, endometrial, cervical, gastric, genitourinary cancers (Cohen, 2003; Turner and Grose, 2010a). FGFR mutations lead to various developmental disorders including Bent bone dysplasia syndrome and Apert syndrome (Wilkie et al., 1995). We have summarized the major similarities which suggest possible links between FGFR implicated syndromes and thalidomide embryopathy in Table S1. With FGFs, embryonic development, angiogenesis and cancer as common links, we hypothesized that FGFRs could be direct targets of thalidomide. Although implications of thalidomide have been documented in FGF biology, there is no known study reporting the FGFRs as direct targets of thalidomide. For a better understanding of FGFR modulation, we included other two analogs of thalidomide, lenalidomide and pomalidomide in our study. The present study offers the first experimental evidence that FGFR family might serve as direct targets of thalidomide and its analogs, both at transcriptional and protein activity levels.


Page 5 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2. Materials and Methods 2.1 Materials Dulbecco’s modified medium (DMEM), fetal bovine serum (FBS), penicillin and streptomycin were purchased from PAN Biotech. Thalidomide, Lenalidomide and Pomalidomide were from Toronto Research Chemicals, Canada. DMSO was purchased from Sigma. Rabbit polyclonal anti-FGFR2 antibody was procured from Abcam. Goat anti-rabbit FITC conjugated IgG antibody was obtained from Genei, Bangalore. All other chemicals used in the study were of the reagent grade and obtained commercially. 2.2 Methods The overall study design is illustrated in Figure S1. In silico tools were used to obtain a predictive relationship between the drugs and the receptors, followed by assessment of the effects of drugs on FGFRs at the gene expression and protein activity levels. 2.2.1 In silico target protein identification using STITCH STITCH (Search Tool for Interacting Chemicals) (Szklarczyk et al., 2016) predicts interactions between small molecules and proteins. The database contains 9,64,763 proteins from 2,031 organisms (as on 22.12.2018). To understand the interactions of thalidomide with proteins, the term “Thalidomide” “was given as input. To eliminate false positive targets, a confidence score >0.8 and not more than 50 interactors were used as input parameters. 2.2.2 Relationship between FGFR and thalidomide BioGraph (www.biograph.be) is a service that integrates different knowledge bases, relates two different entities and ranks genes or compounds (Liekens et al., 2011). The terms, “thalidomide” and “FGFR2” were given as inputs and the links were obtained as a graph. Information on lenalidomide and pomalidomide was not available in the database. 2.2.3 Chick embryos, Drug treatment and RT-PCR Fertilized brown leghorn eggs were obtained from Poultry Research Centre, Potheri, Chennai. Research on chicken embryos was conducted in accordance with the prior approval from the Institutional Biosafety and Ethical Committee (IBEC) of the AU-KBC Research Centre. We followed the protocol and concentrations for drug administration that we had published earlier (Veeriah et al., 2017). The eggs were maintained in a humidified incubator at 37⁰C. The embryos were staged as per the Hamilton Hamburger (HH) staging method and treated with 20 μL of 1mg/mL DMSO (vehicle control) or thalidomide or lenalidomide or pomalidomide (n=5).

The drugs were administered

through a small aperture made through the eggshell in the air sac using a sterile needle at the 24th hour 5 ACS Paragon Plus Environment


Page 6 of 35

of incubation. After six days of appropriate incubation, the embryos were dissected out and the individual organs, heart, brain, limb and eye were carefully isolated, pooled and further processed for RNA isolation using TRIzol method and RT-PCR was performed. Primers and temperature conditions used for RT-PCR have been summarized in Table S2. 2.2.4 Immunohistochemistry Eggs were treated with 20µL of vehicle control or thalidomide or lenalidomide or pomalidomide at 24h (n=5). Five embryos were assigned to each group and immunohistochemistry was performed as described elsewhere (Kumar et al., 2016). Briefly, the eggs were opened on 6th day of incubation, embryos dissected out, the limbs were isolated and fixed in 10% neutral buffered formalin. The fixed limbs were embedded in paraffin after dehydration with various concentrations of ethanol and xylene. Then 5µm sized sections were obtained with microtome and the sections were deparaffinized with xylene, followed by rehydration. Antigen retrieval was done using Tris-EDTA buffer followed by washing, blocking, primary and secondary antibody staining. The slides were stained with nuclear stain and after appropriate washing, they were visualized in fluorescent microscope. The fluorescence intensity was measured using ImageJ software and the intensity values were compared between treatment and DMSO control. 2.2.5 Thalidomide-treated embryo transcriptome analysis Transcriptome sequencing data of 6th day chick embryo treated with vehicle control or thalidomide GEO (Accession No. GSE69159) was leveraged to understand the effects of thalidomide on FGFR signaling in embryonic development. Sequencing methodology and bioinformatics analysis have been described in detail elsewhere (Kumar et al., 2016). The differentially expressed genes with a p value < 0.05 were considered to be significantly modulated. Further, enrichment analysis was performed using Enrichr (Kuleshov et al., 2016). 2.2.6 Cell culture and drug treatment We chose three different cell lines, an immortalized endothelial cell derived from HUVEC (EA.hy926), breast cancer (MCF-7) and cervical cancer cell lines (HeLa) as they express FGFRs at moderate, high and low levels respectively. EA.hy926 cell line was kindly gifted by Dr. C.J.S Edgell, University of North Carolina, Chapel Hill, NC, USA. MCF-7 and HeLa were purchased from NCCS, Pune. All the cell lines were maintained in DMEM supplemented with 10% FBS and 1% penicillinstreptomycin and anti-mycotic agent. The drugs were prepared in 5% DMSO (vehicle control). Following our previous studies, the cells were treated with 75µg/mL of the drugs as published earlier (Majumder et al., 2009) unless mentioned otherwise.


Page 7 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.2.7 Cell viability determination by MTT assay MTT assay was performed on the cells to determine the effects of the drugs on cell viability. Briefly, EA.hy926, MCF-7 and HeLa cells were plated in 96-well plates at 10,000 cells/well. After 12h serum starvation, the cells were treated with 75µg/mL of thalidomide, lenalidomide or pomalidomide or vehicle control (n=5). After 24h incubation, the treatments were removed and the cells were washed and incubated with MTT (5µg/ml final concentration) for 4h at 37ºC. After removal of MTT solution, the formazon crystals formed were dissolved in DMSO and the plate was read at 570nm using a microplate reader. 2.2.8 RT-PCR To study the effect of thalidomides on the expression of FGFR1, FGFR2 and FGFR3 in EA.hy926, HeLa and MCF-7 cells, we treated the cells with vehicle control or thalidomide or lenalidomide or pomalidomide for 24h (n=5), the total RNA was isolated by TRIzol® method and the gene expression was checked by RT-PCR. 2.2.9 Immunocytochemistry The cells were grown in 24-well plates and as they reached 60% confluence, they were treated with vehicle control or thalidomide or lenalidomide or pomalidomide for 24h (n=5). Then the cells were fixed in 4% paraformaldehyde and blocked with 5% BSA, followed by incubation of anti-FGFR2 antibody, overnight. After incubation with secondary antibody for 1h at 37ºC, the cells were stained with DAPI and imaging was done using Fluorescent microscope. 2.2.10 Molecular Docking Docking studies were carried out using Autodock 4.2 (Morris and Huey, 2009). The PDB structures of 1GJO for FGFR2 kinase domain, 3KY2 for FGFR1 kinase domain and 3GRW for FGFR3 were downloaded from Protein Data Bank (rcsb.org) (Berman et al., 2000). The sdf files of the small molecules, thalidomide (CID: 5426), lenalidomide (CID: 216326), pomalidomide (CID: 134780) and formononetin (CID: 5280378) were obtained from PubChem (Kim et al., 2016) and converted into pdb files using Openbabel (O’Boyle et al., 2011).Sdf file of ARQ069 (PDB ID: 3RH) was obtained from Protein Data Bank. The hetero atoms were removed and Kollman charges were added. PDBQT files were generated and blind docking was performed by selecting the entire macromolecule in the grid. The results were analyzed and visualized using BIOVIA Biodiscovery studio (Dassault Systèmes BIOVIA, 2016). 2.2.11 Homology Modelling The crystal structures of FGFR3 kinase domain wildtype and mutant, FGFR3 mutants were not available in the Protein Data Bank. Therefore, using the FASTA sequence obtained from NCBI, 7 ACS Paragon Plus Environment


Page 8 of 35

structural models were built based on the crystal structure of FGFR3 K650E kinase using Swiss homology modelling tool (Schwede et al., 2003) and further docking studies were carried out as mentioned previously. 2.2.12 Kinase inhibitor assay We measured the activity of FGFR2 kinase using CycLex® FGFR2 Kinase Assay/ Inhibitor Screening Kit. This is the most sensitive protocol for quantitative measurement of FGFR2 kinase activity, which can be used to assess both activators and inhibitors. With different concentrations of the drugs, we checked for the kinase activity. The IC50 values for FGFR2 inhibition by thalidomide, lenalidomide and pomalidomide were determined as per the manufacturer’s instructions. 2.2.13 Kinase activity and Kd value determination by KINOMEscan® screening To understand how thalidomide affected the activities of kinases at the global level, we performed KINOMEscan® screening using DiscoverX platform. This assay screens the activity of 460 kinases in presence of small molecules. The principle behind the method is that the ligands are bound to a solid support and their corresponding kinases are allowed to interact with the kinase. When a compound that binds the kinase sterically or allosterically is added, the binding of kinase to its ligand is inhibited. The captured kinases are measured by ultra-sensitive qPCR and the quantified kinase activity is compared with the control (no small molecule added). This method yields results in terms of thermodynamic interaction affinities (ATP concentration-independent) rather than IC50 values. The kinases related to FGFR pathway including FGFR1, FGFR2, FGFR2 (N550K), FGFR3 and FGFR3 (G697C) and FGFR4 were present in the assay panel. We also screened for dissociation constants (Kd values) of thalidomide (concentrations up to 30µM) against FGFR1, FGFR2, FGFR3 and FGFR4 using the KdELECT® assay platform. KdELECT® quantifies binding affinity of compound-kinase interactions. Further, to understand the processes affected by thalidomide, kinases whose activity was affected by at least 30% were subjected to enrichment analysis using STRING server (Szklarczyk et al., 2017). 2.2.14 Statistical analysis Sigmastat v3.5 was used to perform the statistical analyses. Data are expressed as mean +SEM. All the experiments were performed with n=5 unless specified otherwise. The statistical tests performed were one way ANOVA, followed by Tukey post-hoc or Holm-Sidak test wherever appropriate. p values < 0.05 were considered to be statistically significant.


Page 9 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3. Results 3.1 In silico analysis 3.1.1 STITCH predicted all the FGF receptors as interactors Using STITCH online database (accessed on 13.12.2017), drug targets of thalidomide were extracted. “CRBN” was obtained as the first hit with a score of 0.988, followed by VEGFA with 0.985. FGFR2 was 13th target in the list with a score of “0.854”. We obtained all the FGFRs as interactors of thalidomide. FGFR2 had the highest interaction score of 0.854 among the four, followed by FGFR1 with a score of 0.73. The scores of FGFR3 and FGFR4 were 0.59 and 0.47 respectively. In addition, FGF2 was the third hit with a score of 0.984 and FGF8 with a score of 0.869. Figure S2 illustrates the evidence view of STITCH network. 3.1.2 Biograph related thalidomide to FGFR2 among top 0.93% candidates Biograph derives relationship between gene and compounds. We had FGFR2 and thalidomide as inputs. For FGFR2, Biograph assigned thalidomide, a rank of 79 out of 8508 compounds which falls under top 0.93%. The size of the phenotypes reflects the strength of association. In the connections projected between FGFR2 gene and thalidomide, the major link was congenital limb deformities, followed by liver cirrhosis. The next link was FGF2 gene where FGFR2 is connected to FGF2 through interaction and thalidomide affects the expression and secretion of FGF2. Another link was KDR gene which is affected by thalidomide whose activity is influenced by FGFR2. ACP5 was another predicted link, which is linked through bone morphogenesis to FGFR2. Other key terms that were found in common with the branching involved in embryonic placenta morphogenesis, craniofacial dysostosis, craniofacial abnormalities, craniosynostosis (Figure S3). The data for lenalidomide and pomalidomide were not available in the database. STITCH network, Pocketome (Table S7) and Biograph offers a strong indication that thalidomide interferes with FGFRs. 3.2 In vivo model – Gene expression studies As thalidomide is a well-known teratogen, we wanted to check whether thalidomide treatment affects the FGFR expression during embryonic development in chick embryos. The expression of the receptors in the whole embryo as well as in individual organs, limb, brain, heart and eye were studied during the developmental phase HH29 of the chick embryo. HH29 was chosen as the organ development is almost complete during that particular stage. 3.2.1 Differential regulation of FGFR expression in chick embryo by thalidomide Thalidomide-treated embryos were checked for FGFR expression. FGFR1 was down-regulated in the whole embryo, limb and brain by 3-fold, 4.6-fold and 6.3-fold respectively. FGFR2 was underexpressed in the limb and brain by 7-fold and 11-fold respectively. FGFR3 was down-regulated in the 9 ACS Paragon Plus Environment


Page 10 of 35

whole embryo, brain and eye by 3-fold and 11.2-fold respectively. Notably, all the three receptors were affected in brain. Interestingly, none of the receptor expression was significantly affected in heart due to the drug treatments (Figure 1A,B,C). 3.2.2 Lenalidomide affected FGFR expression in chick embryo Under lenalidomide treatment, FGFR1 was up-regulated in the whole embryo, brain and heart by 3.4, 9.4 and 3-fold respectively. FGFR2 was down-regulated by 8.2 and 10-fold in whole embryo and eye respectively. FGFR3 was not affected in any of the organs (Figure 1B). 3.2.3 FGFR1 and FGFR2 were up-regulated by pomalidomide in chick embryo In pomalidomide-treated embryos, FGFR1 was up-regulated in most of the organs. Whole embryo, brain, heart and eye showed 5.1, 17, 3.7 and 3.5-fold increase in FGFR1 expression. FGFR2 was also up-regulated by 4.4 and 5.2-fold in whole embryo and brain. FGFR3 expression was 78-fold downregulated in the whole embryo under pomalidomide treatment and limb showed a 4.9-fold increase in FGFR3 expression (Figure 1C). 3.2.4 Thalidomide affected FGFR2 protein levels in developing limb of chick embryo The limb sections were analyzed for FGFR2 expression under the drug treatments. Compared to the control, FGFR2 expression was significantly less in thalidomide and lenalidomide. As compared to vehicle-treated limb, thalidomide-treated limb showed a significant decrease in the expression (Figure 2). Lenalidomide and pomalidomide did not show significant changes in the FGFR2 expression. 3.2.5 Transcriptome Analysis of Thalidomide Treated Embryos Transcriptome sequencing of whole chick embryo on 6th day of incubation, pre-treated with vehicle control or thalidomide, was performed to understand how thalidomide affects FGFR expression and associated signaling during development. The expression of FGFR2 was not affected significantly, while FGFR3 was significantly upregulated in thalidomide-treated embryos. FGFR1 and FGFR4 expression were not detected. mRNA expression of FGFR ligands, FGF 1,2,4,12,16,19 and 20 and FGF binding proteins, FGFBP1 and FGFBP2 were reduced while the expression of ligands, FGF4, 6,8 and 23 were observed to be upregulated. Many genes implicated in limb, eye and other master regulators of development were found to be differentially regulated in thalidomide treated embryos (Figure 3A). Enrichment analysis of the differentially expressed genes showed that pathways, “Signaling by FGFR1 in disease”, “Signaling by FGFR3 fusions in cancer”, “Signaling by FGFR3 in disease”, “Signaling by FGFR3 point mutants in cancer” along with transcription factors, SOX2, STAT3, TEAD4, TCF4 and SMAD4 were upregulated. Cell cycle related biological processes were also observed to be modulated under thalidomide treatment (Figure 3B). Down-regulated genes 10 ACS Paragon Plus Environment

Page 11 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enriched terms, “FGFR knockdown” and transcription factors including SOX2, TEAD4 and TCF4 (Figure 3C). 3.3 In vitro model 3.3.1 Thalidomide and its analogs affected cell viability significantly Thalidomide reduced the cell viability of EA.hy926 cells significantly at the tested concentration whereas other two cell lines, MCF-7 and HeLa were not significantly affected. Lenalidomide significantly reduced cell viability of EA.hy926 and MCF-7. Interestingly, Pomalidomide significantly affected the viability of all the three cell lines including HeLa (Figure 4). The next question was how thalidomide and the two analogs affected the FGFR expression in cells. Since thalidomide is known to affect angiogenesis, we chose EA.hy926, an immortalized endothelial cell line derived from HUVEC (Human umbilical vein endothelial cell line). In addition, thalidomide is currently considered a potential candidate as anti-cancer agent. Hence we checked the expression of FGFRs in cancer cell lines, HeLa and MCF-7. The expression of FGFRs under the three treatments conditions in the three cell lines have been summarized in Table S3. 3.3.2 Thalidomide and lenalidomide affected FGFR expression in EA.hy926 The expression of FGFR1 was significantly up-regulated while FGFR2 was down-regulated in both thalidomide and lenalidomide treated endothelial cells (Figure 5A). In addition, thalidomide treated cells showed a decrease in FGFR3 expression. Immunofluorescence data also followed similar FGFR2 expression pattern, when the cells were treated with thalidomide and its analogs (Figure 5D). Pomalidomide did not affect the expression of any of the receptors at mRNA level (Figure 5A) and at the protein level as well (Figure 5D). 3.3.3 FGFR expression was most affected in MCF-7 by thalidomide, lenalidomide and pomalidomide The morphology of MCF-7 was perturbed strongly under thalidomide treatment (Figure S4). There was a striking reduction in FGFR1 mRNA expression under all the treatments in MCF-7 (Figure 5B). Lenalidomide was found to decrease the expression of all the three receptors. FGFR2 was downregulated in thalidomide and lenalidomide-treated cells (Figure 5B). Similar results were observed at the protein level (Figure 5E). FGFR2 protein expression was reduced significantly by thalidomide followed by lenalidomide whereas no significant difference was observed in pomalidomide-treated cells (Figure 5E).



Page 12 of 35

3.3.4 FGFR1 expression was up-regulated in HeLa cell by all the drug treatments HeLa showed up-regulation in FGFR1 expression under all the treatments. FGFR2 was not affected significantly by any of the treatments whereas pomalidomide increased FGFR3 expression (Figure 5C). The FGFR2 protein levels were very low in HeLa cells and the expression was not affected or up-regulated by any of the treatments in HeLa (Figure S5). 3.4 Protein interaction and functional assays 3.4.1 Docking studies predicted similarities in binding of drugs to FGFR2 with that of known inhibitors Next, we wanted to check whether thalidomide and its two analogs interact with the FGFRs at protein level. In this direction, we first opted for molecular docking studies. Docking results revealed the interaction of analogs with the receptors. The best binding positions of thalidomide, lenalidomide and pomalidomide with each of the receptor kinase domains have been depicted in Figure S8. Lenalidomide docked with FGFR2 kinase domain showed predicted binding energy and Ki to be 5.95 kcal/mol and 43.17 µM respectively. For pomalidomide and thalidomide predicted binding energies with FGFR2 kinase domain were -6.52 and -6.37 kcal/mol respectively. The predicted Ki values were 16.69µM and 21.56µM respectively. Formononetin and ARQ069, two known FGFR2 inhibitors had predicted binding energies of -6.19 kcal/mol and -7.12 kcal/mol respectively. Remarkably, all the three drugs and the two known FGFR2 inhibitors were predicted to interact with Lys517 and Leu531 of FGFR2. Leu647 was in common among thalidomide, pomalidomide and ARQ069. Pomalidomide was predicted to have 6 hydrogen bonds and 5 hydrophobic contacts with FGFR2. Pomalidomide shared residues, Gln494 and Met518 with formononetin and Phe492 with ARQ069 (Figure 6D). The predicted binding energies, Ki values and the interacting residues for FGFR2 docked with the three drugs can be found in Table 1. When the homology models of FGFR3 WT and G697C were docked with thalidomides, we observed that all the three drugs were predicted to bind to the same amino acid residues of wild type FGFR3 with different predicted interaction binding energies and inhibition constants (Figures 6B,C and Table S6). Thalidomide and pomalidomide were predicted to make hydrogen bonds with residues Lys508 of FGFR3 (Table S6). G697C mutation introduced a major shift in the interacting amino acids. In G697C mutant FGFR3 kinase, thalidomide was predicted to interact with Val555 through hydrophobic bond and interestingly, the G697C mutation abolished all the hydrogen bonds with the ligand, however created hydrophobic bonds with the same residues that were observed with the wild type kinase. In addition, lenalidomide showed extended interaction with two amino acids Leu478 and Gly479 with increase in predicted binding energies. G697C mutation was predicted to abolish the interaction of thalidomide with residues, Asn562, Ala634 and Arg621 and thalidomide showed extended interaction with Val555,


Page 13 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


which was not observed in wild type FGFR3. To compare the effects of G697C mutation, we docked thalidomides with another mutant of FGFR3 (K650E). K650E mutant FGFR3 also showed similar profile where Val555 was predicted to interact with thalidomide and lenalidomide. In addition to keeping the wild type interactions intact, lenalidomide was predicted to interact additionally with Asn562, Ala634 and Arg621. The interaction with Val555 was unique for thalidomide alone in G697C mutant kinase when compared to lenalidomide and pomalidomide (Table S6). Docking results of FGFR1 have been discussed in the Supplemental material (Table S5 & Figure S9). 3.4.2 KINOMEscan® analysis and KdELECT® analysis of FGFRs against Thalidomide KINOMEscan® assay yields percentage of control (POC) values, which denotes the activity of a kinase under a particular treatment with respect to control having 100% activity. We obtained POCs for 469 kinases against thalidomide. Lower the POC, higher the inhibition on a scale of 100. The POCs of FGFR1, FGFR2, FGFR3, FGFR3 (G697C), FGFR3 and FGFR4 were found to be 94, 69, 88, 35 and 94% respectively (Figure 7A). When we determined the Kd values individually for FGFR1, FGFR2, FGFR3 and FGFR4 against thalidomide, we found that the Kd value was greater than 30µM (data not shown). Enrichment analysis of affected kinases (wild type and mutant) showed that biological processes including “immune response”, “cell migration”, “positive regulation of MAPK cascade”, “Angiogenesis”, “Regulation of multicellular organismal process”, “Fibroblast growth factor receptor signaling pathway”, “cell cycle” and “positive regulation of endothelial cell proliferation” (Figure 7B). 3.4.3 Thalidomide, Lenalidomide and Pomalidomide affected the activity of FGFR2 kinase The results obtained from our docking studies and KINOMEscan® suggested that FGFR2 activity or function was most likely to be affected by thalidomide through direct interaction. Therefore, we further performed functional assay focussing on FGFR2. We carried out the FGFR2 activity assay using CycLex FGFR2 Kinase Assay/ Inhibitor Screening Kit to confirm the effects of thalidomide and its two analogs on FGFR2 function. All the three analogs decreased the FGFR2 kinase activity in a dose-dependent manner (Figure 7B). For the determination of IC50 values, the FGFR2 kinase activity was measured at different concentrations of the drug molecules. Out of the three drugs, pomalidomide showed the highest inhibition with a low IC50 value, followed by lenalidomide and thalidomide (Figure 7C). 4. Discussion Various studies have reported the effects of thalidomide on FGFs, however, to the best of our knowledge, there are no studies published demonstrating the direct association between FGFRs and thalidomide. Results from the in silico predictive tools suggested that there could be strong links between FGFRs and thalidomide. We moved on to investigate the effects of thalidomide at FGFR 13 ACS Paragon Plus Environment


Page 14 of 35

gene expression and protein function, which are not related to each other directly, however, examined both the aspects by which FGFR activity and downstream signalling can be perturbed. Thalidomide affects angiogenesis (D’Amato et al., 1994) by modulating bFGF (Stephens et al., 2000), and nitric oxide pathway (Majumder et al., 2009; Tamilarasan et al., 2006). In the present study, endothelial cells showed up-regulated FGFR1 and down-regulated FGFR2 thalidomide and lenalidomide treatment possibly causing an imbalance in the FGF signalling, which is a major player in angiogenesis. Thalidomide, in addition, was observed to down-regulate FGFR3 expression, whose role is believed to be compensatory to FGFR1/FGFR2 loss (Yang et al., 2015). Cell viability data correlated well with the expression pattern as the percentage of viable cells was lower under thalidomide and lenalidomide treatment as compared to pomalidomide (Figure 4). Pomalidomide does not affect angiogenesis in chicken and human endothelium (Mahony et al., 2013) and we observed that pomalidomide did not affect FGFR expression as significantly as the other two drugs. Our results suggest that thalidomide might affect angiogenesis by directly affecting the FGFR expression as well in addition to modulating FGFs in endothelium. FGFRs are known players in tumor pathology and inhibition of FGFRs seems promising in cancer treatment. Highly selective FGFR tyrosine kinase inhibitors including AZD4547, BGJ398, JNJ42756493, ASP5878 are currently in Phase I/ II clinical trials for the treatment of various cancer conditions (ClinicalTrials.gov n.d.). Alofanib, a selective FGFR2 inhibitor affected endothelial and cancer cell proliferation in addition to reducing in vivo tumor growth and blood vessel maturation (Tsimafeyeu et al., 2016). Formononetin, a selective FGFR2 inhibitor-induced cell cycle arrest in MCF-7, prevented tumor growth of breast cancer cells in nude mouse (Chen et al., 2011) and affected breast cancer xenografts significantly (Wu et al., 2015). The high similarity of binding profile of pomalidomide with that of formononetin suggests similar mechanisms of action. . FGFR1 upregulation in HeLa and down-regulation in MCF-7 seems to be contradictory. This may be attributed to the cell and tissue type specificity of the drug and its analogs. Also, in cancer, a number of genomic alterations are observed and this might result in varied responses to the drug treatments. Further investigation into breast cancer is required in the context of FGFRs which may explain the outcomes of current clinical trials with thalidomide and lenalidomide for breast cancer. As FGFRs are important players in embryonic development, their expression is highly temporal and region-specific. FGFR1 and FGFR2 knockouts at germ line level are embryonic lethal. Conditional knockouts of different exon regions lead to a range of limb deformities (Su et al., 2014). FGFR1 is expressed in developing limb bud mesenchyme whereas FGFR2 expression is restricted to endoderm including the apical ectodermal ridge (AER) which is essential for limb bud initiation (Li et al., 2005). FGFR3 is not expressed by the early limb buds (Walshe and


Page 15 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Mason, 2000). FGFR selective inhibitors including NF449, PD161570 and PD173074 affected the limb development to different extents as they have different mechanisms of inhibition (Horakova et al., 2014). Interestingly, the limb bud from thalidomide-treated embryos showed down-regulation of both FGFR1 and FGFR2. The reciprocal loop regulation of FGFR2c, which is expressed in the mesenchyme between FGF8 and FGF10 is required for limb outgrowth and patterning during limb bud development (Moon and Capecchi, 2000; Xu et al., 1998). Shh is expressed in the anterior limb bud (Riddle et al., 1993) which also has been reported to be downregulated under thalidomide treatment (Ito et al., 2010). We suggest that the combined reduction of FGFR1 and FGFR2 in limb bud along with thalidomide-induced low FGF8 and Shh expression (Ito et al., 2010; Therapontos et al., 2009) interferes with the limb bud development resulting in various limb deformities. FGF8, FGF10 and other genes including Dkk1, Tbx5 (Knobloch et al., 2007; Kumar et al., 2016) playing significant roles in limb development are also affected by thalidomide. Vascular dynamics in limb development is a critical component. Limb defects occur due to the inhibition of vasculature under thalidomide exposure (Therapontos et al., 2009). FGFRs are present uniquely in angiogeneic template (Lee et al., 2000) and simultaneously in the neighbourhood growth cone in limb buds (Li et al., 2005). Our data suggest that in addition to affecting the blood vessel maturation, thalidomide affects the FGFRs as well. Mahony and colleagues reported that lenalidomide-treated chick embryos had a normal pattern of forelimb (proximal to distal) whereas the length of humerus, hand plate and digits was reduced in lenalidomide. No change in limb pattern was observed in pomalidomide treated chick embryos (Mahony et al., 2013). This is in consistent with our findings that lenalidomide and pomalidomide did not affect FGFR1 and FGFR2 expression possibly explaining the normal limb patterns observed in lenalidomide and pomalidomide-treated chick embryos (Mahony et al., 2013). FGF/FGFR signalling has a major role in the nervous system development including brain and eye formation (Turner and Grose, 2010b; Zhao et al., 2008). Developing brain from drug-treated chick embryos showed modulation of FGFR expression which warrants further detailed investigation of how these drugs affect brain development. Eyes from thalidomide treated embryos showed reduced FGFR3 expression. In our previous study, we reported that Pax6 and crystallins were down-regulated under thalidomide treatment (Kumar et al., 2016). This could be the result of perturbed FGFR signalling as FGFR deletions during lens development causes increased apoptosis, lack of fiber cell elongation, reduced expression of Pax6, Prox1, c-Maf and α-, β- and γ-crystallins (Zhao et al., 2008) which are modulated as well in thalidomide treatment (Kumar et al., 2016).



Page 16 of 35

Transcription factors are upstream regulators of gene expression control and FGFR1 expression has been reported to be controlled by E2F1 transcription factors (Tashiro et al., 2003). Thalidomide has been shown to affect TNF-alpha and Interleukins in several studies (Klausner et al., 1996). TNF-α has been shown to modulate FGFR2 expression through E2F1 (D’Amici et al 2014). In an earlier study, we observed that thalidomide and lenalidomide affect the E2F transcription factor family (Sundaresan et al., 2018). Corroborating support from all these individual observations strongly suggest that thalidomide by modulating E2F through its effects on TNFα and IL1β might affect FGFR expression. Transcriptome sequencing data of thalidomide-treated embryo captures the diseased FGFR signaling supporting our hypothesis that FGFRs serve as targets of thalidomide. Thalidomide affects the upstream regulators of FGFR, thereby affecting FGFR signaling in addition to causing an imbalance in the function of key transcription factors involved in limb, brain and eye development. This multiple effects observed might be responsible for thalidomide’s diverse effects on embryo development. Next, we checked the possibilities of direct protein-drug interactions. KINOMEscan® assay revealed the activity of a panel of kinases against thalidomide. Interestingly, thalidomide reduced FGFR3 (G697C) mutant activity to 35%. Although thalidomide and its analogs were predicted to interact with the Lys508, a residue that is crucial for FGFR3 kinase activity, through hydrogen bonds, they did not seem to affect the kinase activity of wild type FGFR3. Thalidomide affected the kinase activity of FGFR3 G697C mutant, the most, along with LCK among the 469 kinases tested. The G697 mutation along with K650 is a frequently mutated position in the kinase domain of FGFR3, which is unique to FGFR3 (Patani et al., 2016). When glycine is replaced by cysteine at the 697th position, there is high autophosphorylation in the kinase domain (Yan et al., 2005). Notably, three strong FGFR3 inhibitors, namely, IM-412 (Jung et al., 2015), ponatinib (Kim et al., 2015) and kaempferol (Lee et al., 2018) have been shown to interact with valine residue at position 555 and lysine at 508th position, which are the gatekeeper and ATP coordinating residues respectively (Azam et al., 2008; Bunney et al., 2015; Chell et al., 2013). Thalidomide did not seem to interact with Val555 of wild type FGFR3 whereas it was predicted to interact with the Val555 of G697C and K650E mutant kinases. This strongly suggests that thalidomide might affect FGFR3 kinase activity through its interaction with gatekeeper residue and Lys508 at the ATP binding site, possibly explaining the selective inhibition of FGFR3 mutant G697C in KINOMEscan® experiments. G697C mutation was reported to be associated with oral squamous carcinoma in the Japanese population (Yan et al., 2005) whereas there was no association found in the European population (Aubertin et al., 2007). In addition, G697C mutation has been implicated also in gallbladder cancer (Javle et al., 2014) and Spermatocytic seminoma (Goriely et al., 2009) suggesting few conditions in which thalidomide could be tested against. While thalidomide affected FGFR2 kinase activity in a dose-dependent manner, pull-down (Figure S7) and KINOMEscan® assays also suggest the possible binding and inhibition of thalidomide to FGFR2. Binding predictions for pomalidomide were comparable with that of potent FGFR2 16 ACS Paragon Plus Environment

Page 17 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


inhibitors, ARQ069 and formononetin. Strikingly, all the three drugs and the specific FGFR2 inhibitors, ARQ069 and formononetin were predicted to interact with Lys517 of FGFR2 (Figure 6D and Table 1). Notably, the drugs were predicted to bind to lys508 of FGFR3. Lys517 (FGFR2) and Lys508 (FGFR3) are essential for coordinating the phosphate groups of ATP for the activity of their respective kinases (Chen et al., 2008) and the drugs, therefore may affect the ATP binding, thus affecting the kinase activity as evident by the FGFR2 kinase inhibition assay which is ATP dependent (Figure 7). In addition, thalidomide and pomalidomide were predicted to interact with Leu647. Leu647 is a part of the activation loop (A-loop) that is responsible for maintaining the active conformation of FGFR2 (Chen et al., 2017). A-loop is one of the major hotspots for mutations resulting in human congenital skeletal disorders and cancers (Wilkie, 2005). Possibly, this binding of thalidomide to the hotspot might be responsible for its teratogenic property. The interactions of pomalidomide with FGFR2 seem to be stronger as evident by docking and kinase activity assays when compared to other analogs. Pomalidomide also exhibits differential teratogenic profiles and the embryotoxic effects of pomalidomide in chicken and zebrafish are relatively milder or none when compared to thalidomide (Mahony et al., 2013). However, pomalidomide causes teratogenic limb deformities and cardiac malformations in rats and rabbits (Smith RL, Fabro S, Schumacher H, 1965). Therefore, further studies on higher animals may unravel the role of FGFR modulation by pomalidomide and resolve this ambiguity. 5. Conclusion In this study, we show that thalidomide dampens FGFR biology by interfering simultaneously with gene expression and kinase activity both at the embryonic and the adult cellular level. Pull-down, insilico docking, KINOMEscan® binding and activity assays further strengthen the claim that thalidomide interacts with and affects FGFR2 activity. Pomalidomide exhibits strongest anti-FGFR2 effects at functional level. However, we propose further in-depth studies using X-ray crystallography, CRD, isotherm titration and thermophoresis to authenticate the concept with complete proof of evidence. These confirmatory studies would also decipher the core-mechanism of multi-dimensional effects of thalidomide on life; when all the down-stream events to FGFR are specifically affected with thalidomide treatment. The study promotes the thought that thalidomide and the analogs could be added to the next-generation specific inhibitors of FGFR. This study also suggests that the deregulation FGFRs can be one of the major links between thalidomide’s anti-angiogenic and antitumor activity. We conclude that thalidomide, lenalidomide and pomalidomide interfere with the expression and functions of FGFRs in a differential manner, which further accounts for their differential teratogenic, anti-angiogenic and anti-cancer effects of thalidomide analogs as reported earlier.



Page 18 of 35

Abbreviations: FGFR - Fibroblast Growth Factor Receptor, CRBN –Cereblon, sGC - solubleGuanylyl cyclise, DMSO - Dimethylsulphoxide, HH stage - Hamilton Hamburger, AER- Apical ectodermal ridge

Acknowledgement This work was partially supported by University Grant Commission Faculty Recharge Programme (UGC-FRP); Government of India to SC. LS acknowledges financial support from Department of Biotechnology-Junior Research Fellowship programme (DBT-JRF), Government of India. PK is thankful to the financial support from University Grant Commission- Senior Research Fellowship programme (UGC-SRF), Government of India. Author contributions LS and PK contributed equally to this work. LS, PK, UMS, DK, VV performed experiments. LS, PK, JM, participated in data analysis and interpretation. LS wrote the manuscript. SA contributed to data analysis, interpretation and critical revision of the manuscript. SC conceived the study, and critically revised the manuscript for intellectual content. All authors approved the final version of the manuscript. Conflicts of interest The authors declare that there are none.

References: (1) Aubertin, J., Tourpin, S., Janot, F., Ahomadegbe, J.C., and Radvanyi, F. (2007) Analysis of fibroblast growth factor receptor 3 G697C mutation in oral squamous cell carcinomas. Int. J. Cancer 120, 2058–2059. (2) Azam, M., Seeliger, M.A., Gray, N.S., Kuriyan, J., and Daley, G.Q. (2008) Activation of tyrosine kinases by mutation of the gatekeeper threonine. Nat. Struct. Mol. Biol. 15, 1109–18. (3) Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N., Weissig, H., Shindyalov, I.N., and Bourne, P.E. (2000) The protein data bank. Nucleic Acids Res. 28, 235–242. (4) 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. (5) Chell, V., Balmanno, K., Little, A.S., Wilson, M., Andrews, S., Blockley, L., Hampson, M., Gavine, P.R., and Cook, S.J. (2013) Tumour cell responses to new fibroblast growth factor receptor tyrosine kinase inhibitors and identification of a gatekeeper mutation in FGFR3 as a mechanism of acquired resistance. Oncogene 32, 3059–3070. (6) Chen, H., Marsiglia, W.M., Cho, M.-K., Huang, Z., Deng, J., Blais, S.P., Gai, W., Bhattacharya, S., Neubert, T.A., Traaseth, N.J., and Mohammadi, M. (2017) Elucidation of a foursite allosteric network in fibroblast growth factor receptor tyrosine kinases. eLife 6. e21137. 18 ACS Paragon Plus Environment

Page 19 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(7) Chen, H., Xu, C.-F., Ma, J., Eliseenkova, A. V, Li, W., Pollock, P.M., Pitteloud, N., Miller, W.T., Neubert, T. a, and Mohammadi, M. (2008) A crystallographic snapshot of tyrosine transphosphorylation in action. Proc. Natl. Acad. Sci. U. S. A. 105, 19660–19665. (8) Chen, J., Zeng, J., Xin, M., Huang, W., and Chen, X. (2011) Formononetin induces cell cycle arrest of human breast cancer cells via IGF1/PI3K/Akt pathways in vitro and in vivo. Horm. Metab. Res. 43, 681–686. (9) Cohen, M.M., 2003. Neoplasms associated with alterations in fibroblast growth factor receptors. Am. J. Med. Genet. 119A, 97–100. (10) D’Amato, R.J., Loughnan, M.S., Flynn, E., and Folkman, J. (1994) Thalidomide is an inhibitor of angiogenesis. Proc. Natl. Acad. Sci. 91, 4082–4085. (11) Dassault Systèmes BIOVIA, 2016. Discovery Studio Modeling Environment, Release 2017. (12) Goriely, A., Hansen, R.M.S., Taylor, I.B., Olesen, I.A., Jacobsen, G.K., McGowan, S.J., Pfeifer, S.P., McVean, G.A.T., Meyts, E.R., and Wilkie, A.O.M. (2009) Activating mutations in FGFR3 and HRAS reveal a shared genetic origin for congenital disorders and testicular tumors. Nat. Genet. 41, 1247–1252. (13) Horakova, D., Cela, P., Krejci, P., Balek, L., Moravcova Balkova, S., Matalova, E., and Buchtova, M. (2014) Effect of FGFR inhibitors on chicken limb development. Dev. Growth Differ. 56, 555–572. (14) Ito, T., Ando, H., Suzuki, T., Ogura, T., Hotta, K., Imamura, Y., Yamaguchi, Y., and Handa, H. (2010) Identification of a Primary Target of Thalidomide Teratogenicity. Science. 327, 1345– 1350. (15) Javle, M., Rashid, A., Churi, C., Kar, S., Zuo, M., Eterovic, A.K., Nogueras-Gonzalez, G.M., Janku, F., Shroff, R.T., Aloia, T.A., Vauthey, J.N., Curley, S., Mills, G. and Roa, I. (2014) Molecular characterization of gallbladder cancer using somatic mutation profiling. Hum. Pathol. 45, 701–708. (16) Jung, S.-Y., Yi, J.Y., Kim, M.H., Song, K.H., Kang, S.M., Ahn, J., Hwang, S.G., Nam, K.Y., and Song, J.Y. (2015) IM-412 inhibits the invasion of human breast carcinoma cells by blocking FGFR-mediated signaling. Oncol. Rep. 34, 2731–2737. (17) Kawamura, Y., Shirotsuka, Y., Awatsuji, H., Matsumoto, K. and Sato, K. 2014. Common nature in the effects of thalidomide on embryo-fetal development in Kbl: JW and Kbl: NZW rabbits. Congenit. Anom. (Kyoto). 54, 41–53. (18) Kim, D.-H., Kwak, Y., Doo Kim, N., Sim, T., 2015. Cancer Biology & Therapy Antitumor effects and molecular mechanisms of ponatinib on endometrial cancer cells harboring activating FGFR2 mutations Antitumor effects and molecular mechanisms of ponatinib on endometrial cancer cells harboring activating FGFR2 mutations. Cancer Biol. Ther. 17, 65-78. (19) Kim, S., Thiessen, P.A., Bolton, E.E., Chen, J., Fu, G., Gindulyte, A., Han, L., He, J., He, S., Shoemaker, B.A., Wang, J., Yu, B., Zhang, J. and Bryant, S.H. (2016). PubChem substance and compound databases. Nucleic Acids Res. 44, D1202–D1213. (20) Klausner, J.D., Freedman, V.H., and Kaplan, G. (1996) Thalidomide as an Anti-TNF-a Inhibitor: Implications for Clinical Use, Clin. Immunol. Immunopathol. 81, 219 - 293. (21) Knobloch, J., Shaughnessy, J.D., and Ruther, U. (2007) Thalidomide induces limb deformities by perturbing the Bmp/Dkk1/Wnt signaling pathway. FASEB J. 21, 1410–1421. (22) Kuleshov, M. V., Jones, M.R., Rouillard, A.D., Fernandez, N.F., Duan, Q., Wang, Z., Koplev, S., Jenkins, S.L., Jagodnik, K.M., Lachmann, A., McDermott, M.G., Monteiro, C.D., Gundersen, G.W., and Ma’ayan, A. (2016) Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97. (23) Kumar, P., Kasiviswanathan, D., Sundaresan, L., Kathirvel, P., Veeriah, V., Dutta, P., Sankaranarayanan, K., Gupta, R., and Chatterjee, S., 2016. Harvesting clues from genome wide 19 ACS Paragon Plus Environment


Page 20 of 35

transcriptome analysis for exploring thalidomide mediated anomalies in eye development of chick embryo: Nitric oxide rectifies the thalidomide mediated anomalies by swinging back the system to normal transcriptome pattern. Biochimie 121, 253–267. (24) Lee, C.J., Moon, S.J., Jeong, J. H., Lee, S., Lee, M.H., Yoo, S.M., Lee, H.S., Kang, H.C., Lee, J.Y., Lee, W.S., Lee, H. J., Kim, E.K., Jhun, J.Y., Cho, M.L., Min, J.K., and Cho, Y.Y. (2018) Kaempferol targeting on the fibroblast growth factor receptor 3-ribosomal S6 kinase 2 signaling axis prevents the development of rheumatoid arthritis. Cell Death Dis. 9, 401. (25) Lee, S.H., Schloss, D.J., and Swain, J.L. (2000) Maintenace of vascular integrity in the embryo requires signaling through the fibroblast growth factor receptor. J. Biol. Chem. 275, 33679– 33687. (26) Lenz, W., and Knapp, K. (1962) Thalidomide embryopathy. Arch. Environ. Health 5, 14–19. (27) Li, C., Xu, X., Nelson, D.K., Williams, T., Kuehn, M.R., and Deng, C.X. (2005) FGFR1 function at the earliest stages of mouse limb development plays an indispensable role in subsequent autopod morphogenesis. Development 132, 4755–4764. (28) Liekens, A.M.L., De Knijf, J., Daelemans, W., Goethals, B., De Rijk, P., and Del-Favero, J. (2011) BioGraph: Unsupervised biomedical knowledge discovery via automated hypothesis generation. Genome Biol. 12, R57. (29) Mahony, C., Erskine, L., Niven, J., Greig, N.H., Figg, W.D., and Vargesson, N. (2013) Pomalidomide is nonteratogenic in chicken and zebrafish embryos and nonneurotoxic in vitro. Proc. Natl. Acad. Sci. 110, 12703–12708. (30) Majumder, S., Rajaram, M., Muley, A., Reddy, H.S., Tamilarasan, K.P., Kolluru, G.K., Sinha, S., Siamwala, J.H., Gupta, R., Ilavarasan, R., Venkataraman, S., Sivakumar, K.C., Anishetty, S., Kumar, P.G., and Chatterjee, S. (2009) Thalidomide attenuates nitric oxide-driven angiogenesis by interacting with soluble guanylyl cyclase. Br. J. Pharmacol. 158, 1720–1734. (31) Moon, A.M., and Capecchi, M.R. (2000) Fgf8 is required for outgrowth and patterning of the limbs. Nat. Genet. 26, 455–459. (32) Morris, G., and Huey, R. (2009) AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 30, 2785–2791. (33) O’Boyle, N.M., Banck, M., James, C.A., Morley, C., Vandermeersch, T., and Hutchison, G.R. (2011) Open Babel: An Open chemical toolbox. J. Cheminform. 3, 33. (34) Partanen, J., Makela, T.P., Eerola, E., Korhonen, J., Hirvonen, H., Claesson-Welsh, L., and Alitalo, K. (1991) FGFR-4, a novel acidic fibroblast growth factor receptor with a distinct expression pattern. EMBO J. 10, 1347–1354. (35) Patani, H., Bunney, T.D., Thiyagarajan, N., Norman, R.A., Ogg, D., Breed, J., Ashford, P., Potterton, A., Edwards, M., Williams, S. V, Thomson, G.S., Pang, C.S.M., Knowles, M.A., Breeze, A.L., Orengo, C., Phillips, C., and Katan, M. (2016) Landscape of activating cancer mutations in FGFR kinases and their differential responses to inhibitors in clinical use. Oncotarget 7, 24252–68. (36) Rashid, A., Kuppa, A., Kunwar, A., and Panda, D. (2015) Thalidomide (5HPP-33) Suppresses Microtubule Dynamics and Depolymerizes the Microtubule Network by Binding at the Vinblastine Binding Site on Tubulin. Biochemistry 54, 2149–2159. (37) Riddle, R.D., Johnson, R.L., Laufer, E., and Tabin, C. (1993) Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 1401–16. (38) Schwede, T., Kopp, J., Guex, N., Peitsch, M.C. (2003) SWISS-MODEL: An automated protein homology-modeling server. Nucleic Acids Res. 31, 3381–5. (39) Smith, R.L., Fabro, S., and Schumacher, H. W. (1965) Studies on the relationship between the chemical structure and embryotoxic activity of thalidomide and related compounds. In: Robson, J.M., Sullivan, F., Smith, R.L., editors. Embryopathic Activity of Drugs, Biological Council Symposium. London: Churchill. 20 ACS Paragon Plus Environment

Page 21 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(40) Stephens, T.D., Bunde, C.J.W., and Fillmore, B.J. (2000) Mechanism of action in thalidomide teratogenesis. Biochem. Pharmacol. 59, 1489-1499. (41) Su, N., Jin, M., and Chen, L. (2014) Role of FGF/FGFR signaling in skeletal development and homeostasis: Learning from mouse models. Bone Res. 2, 14003. (42) Sundaresan, L., Kumar, P., and Chatterjee, S. (2018) Mechanistic insights into the differential effects of thalidomide and lenalidomide in metastatic prostate cancer. Futur. Oncol. fon-2018-0090. (43) Szklarczyk, D., Morris, J.H., Cook, H., Kuhn, M., Wyder, S., Simonovic, M., Santos, A., Doncheva, N.T., Roth, A., Bork, P., Jensen, L.J., and von Mering, C. (2017) The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible. Nucleic Acids Res. 45, D362–D368. (44) Szklarczyk, D., Santos, A., Von Mering, C., Jensen, L.J., Bork, P., and Kuhn, M. (2016) STITCH 5: Augmenting protein-chemical interaction networks with tissue and affinity data. Nucleic Acids Res. 44, D380–D384. (45) Tamilarasan, K.P., Kolluru, G.K., Rajaram, M., Indhumathy, M., Saranya, R., and Chatterjee, S. (2006) Thalidomide attenuates nitric oxide mediated angiogenesis by blocking migration of endothelial cells. BMC Cell Biol. 7, 17. (46) Tashiro, E., Minato, Y., Maruki, H., Asagiri, M., and Imoto, M. (2003) Regulation of FGF receptor-2 expression by transcription factor E2F-1. Oncogene 22, 5630–5635. (47) Therapontos, C., Erskine, L., Gardner, E.R., Figg, W.D., and Vargesson, N. (2009) Thalidomide induces limb defects by preventing angiogenic outgrowth during early limb formation. Proc. Natl. Acad. Sci. 106, 8573–8578. (48) Tsimafeyeu, I., Ludes-Meyers, J., Stepanova, E., Daeyaert, F., Khochenkov, D., Joose, J.B., Solomko, E., Van Akene, K., Peretolchina, N., Yin, W., Ryabaya, O., Byakhov, M., and Tjulandin, S. (2016) Targeting FGFR2 with alofanib (RPT835) shows potent activity in tumour models. Eur. J. Cancer 61, 20–28. (49) Turner, N., and Grose, R. (2010) Fibroblast growth factor signalling: from development to cancer. Nat. Rev. Cancer 10, 116–129. (50) Veeriah, V., Kumar, P., Sundaresan, L., Mafitha, Z., Gupta, R., Saran, U., Manivannan, J., and Chatterjee, S. (2017) Transcriptomic Analysis of Thalidomide Challenged Chick Embryo Suggests Possible Link between Impaired Vasculogenesis and Defective Organogenesis. Chem. Res. Toxicol. 30, 1883–1896. (51) Walshe, J., and Mason, I. (2000) Expression of FGFR1, FGFR2 and FGFR3 during early neural development in the chick embryo. Mech. Dev. 90, 103–110. (52) Wilkie, A.O.M. (2005) Bad bones, absent smell, selfish testes: The pleiotropic consequences of human FGF receptor mutations. Cytokine Growth Factor Rev. 16, 187-203. (53) Wilkie, A.O.M., Slaney, S.F., Oldridge, M., Poole, M.D., Ashworth, G.J., Hockley, A.D., Hayward, R.D., David, D.J., Pulleyn, L.J., Rutland, P., Malcolm, S., Winter, R.M., and Reardon, W. (1995) Apert syndrome results from localized mutations of FGFR2 and is allelic with Crouzon syndrome. Nat. Genet. 9, 165–172. (54) Wu, X.Y., Xu, H., Wu, Z.F., Chen, C., Liu, J.Y., Wu, G.N., Yao, X.Q., Liu, F.K., Li, G., and Shen, L. (2015) Formononetin, a novel FGFR2 inhibitor, potently inhibits angiogenesis and tumor growth in preclinical models. Oncotarget 6, 44563–44578. (55) Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R.I., Ornitz, D.M., Leder, P., and Deng, C. (1998) Fibroblast growth factor receptor 2 (FGFR2)-mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 125, 753–765. (56) Yan, Z., Hiraishi, Y., Wang, H., Sumi, K.S., Hayashido, Y., Toratani, S., Kan, M., Sato, J.D., and Okamoto, T. (2005) Constitutive activating mutation of the FGFR3b in oral squamous cell carcinomas. Int. J. Cancer 117, 166–168. 21 ACS Paragon Plus Environment


Page 22 of 35

(57) Yang, X., Liaw, L., Prudovsky, I., Brooks, P.C., Vary, C., Oxburgh, L., and Friesel, R. (2015) Fibroblast Growth Factor Signaling in the Vasculature. Curr. Atheroscler. Rep. 17, 509. (58) Zhao, H., Yang, T., Madakashira, B.P., Thiels, C.A., Bechtle, C.A., Garcia, C.M., Zhang, H., Yu, K., Ornitz, D.M., Beebe, D.C., and Robinson, M.L. (2008) Fibroblast growth factor receptor signaling is essential for lens fiber cell differentiation. Dev. Biol. 318, 276–288.


Page 23 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table Table 1 FGFR2 docking with thalidomide, lenalidomide, pomalidomide

S.No Drug

Hydrogen bonds

Hydrophobic interactions

Residue

1

2

3

4

5

Distance (Å) 2.77 Lys517, Leu531 Thalidomide Lys517 2.63` Gly646 Leu647 3.14 Glu534 3.2 Leu531 Lenalidomide Lys517 2.80 Arg630 3.13 Asn631 3.16 Leu531 3,17 2.97 Phe492, Gly493, Lys517, Pomalidomide Cys491 2.92 Leu531, Leu647 Phe492 2.79 Gln494 3.00 Lys517 Met518 3.26 Asp530 3.35 Glu489, Gly490, Phe492, ARQ069 Gly488 2.67 Val495, Lys517, Leu531 Phe492 3.18 Gly646 3.2 3.32 Lys517 Formononetin Asn 631 2.13 Gly490, Lys517,Leu531 Gln494 2.67 Met518 1.90 Asp626 3.07 Bold residues denote residues that are in common with the known FGFR2 inhibitors



Page 24 of 35

Figures Figure 1


Page 25 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2

E



Figure 3

A

Fold change

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 35

B

C


Page 27 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4



Page 28 of 35

Figure 5


Page 29 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5D



Page 30 of 35

Figure 5E


Page 31 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6A

Figure 6B&C

B

C



Figure 6D

Figure 7A

Kinase activity of FGFRs

Percentage of control

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 35

Figure 7B 32 ACS Paragon Plus Environment

Page 33 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7C



Page 34 of 35

Figure legends Figure 1: mRNA expression of FGFRs in organs – FGFR expression in organs from embryos treated with thalidomide (A), lenalidomide (B) and pomalidomide (C) determined by RT-PCR. “*” denotes more than 2-fold change in the expression. n=5. Figure 2: A-D. Morphology and histology of limb bud from chick embryos- Morphology and Histology of limb bud isolated from 4th day embryo A. Normal morphology of limb bud before sectioning. Panel B shows the H&E stained of the cross sections. C& D. Immunohistochemistry for FGFR2 protein in limb bud cross sections (C) DAPI and (D) FGFR2 expression. (n=5). E. FGFR2 protein expression in limb bud sections from 10th day chick embryo. * p< 0.05 vs control; # p < 0.05 vs Pomalidomide. n=3. Figure 3: Transcriptome analysis of thalidomide challenged chick embryo. A. Expression of FGFR and FGFs from thalidomide treated chick embryo. * indicates > 2-fold change difference. B. Enrichment analysis of up-regulated genes. Upregulated pathways, processes and transcription factors enriched under thalidomide treatment C. Enrichment analysis of down-regulated genes Downregulated pathways and transcription factors enriched in thalidomide treated embryos. Figure 4: Cell viability assay: Thalidomide reduced the cell viability of EA.hy926 cells significantly whereas lenalidomide affected the viability of EA.hy926 and MCF-7 cell lines. Pomalidomide reduced the cell viability of all the three cell lines significantly.* p< 0.05 vs control; # p < 0.05 vs thalidomide; & p < 0.05 vs thalidomide and lenalidomide. Figure 5: A, B & C. FGFR expression in cell lines- FGFR1, FGFR2 and FGFR3 expression in EA.hy926, MCF-7 and HeLa treated with thalidomide, lenalidomide and pomalidomide for 24h respectively determined by RT-PCR. Fold change of the genes with respect to GAPDH as housekeeping gene. (n=5)* indicates a 2-fold change in the expression. D. Immunofluorescence – FGFR2 expression in EA.hy926 cells EA.hy926 cells were treated with thalidomide or lenalidomide or pomalidomide or vehicle control for 24h and FGFR2 expression was measured by immunofluorescence. Thalidomide and lenalidomide-treated cells showed a significant reduction in FGFR2 expression. n=5; * p < 0.05 vs control; # p < 0.05 vs pomalidomide. Error bars indicate SEM. E. Immunofluorescence – FGFR2 expression in MCF-7 cells - FGFR expression in MCF-7 treated with thalidomide or lenalidomide or pomalidomide or vehicle control for 24h. Thalidomide and lenalidomide-treated cells showed a significant reduction in FGFR2 expression. n=5; * p < 0.05 vs control; # p < 0.05 vs pomalidomide. Error bars indicate SEM. Figure 6: Molecular docking of FGFRs with thalidomides. A. Predicted binding energies of FGFRs with thalidomides. Thal, Len, Pom, ARQ, Form and Asp stand for thalidomide, lenalidomide, pomalidomide, ARQ069, formononetin and aspirin respectively. B & C. Comparison of predicted interaction energies and inhibition constants of drugs with FGFR3 wild type and mutants. Binding of pomalidomide to FGFR3 was predicted to be more prominent among the three drugs. D. Interaction of thalidomide, lenalidomide, pomalidomide and ARQ069 with FGFR2 - 3D representation of thalidomide, lenalidomide, pomalidomide and ARQ069 bound to FGFR2. Ligand structure is highlighted in yellow color. Hydrogen bonds and hydrophobic bonds are illustrated in the figure. The green dashed lines represent hydrogen bonds; Pink dashed lines represent hydrophobic contacts; Red dashed line indicates unfavorable acceptor-acceptor bonds; Light blue indicates Carbon-hydrogen bond.


Page 35 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7: Protein-drug interaction studies. A. Percentage of active kinase as determined by KINOMEscan® analysis –The percentage of control values of thalidomide against FGF receptors against 10µM thalidomide as determined by KINOMEscan® analysis B. Thalidomide modulated Biological Processes enriched by wild type and mutant kinases C. FGFR2 kinase activity assay. Dosedependent reduction in FGFR2 kinase activity by the drugs with maximum inhibition by pomalidomide. Error bars indicate SEM. (n= 3).

Supplementary information List of figures Figure S1: Overall experimental design of the study. Figure S2: STITCH network of drug-protein interaction Figure S3: Biograph relationship between FGFR2 and thalidomide Figure S4: Morphology of MCF-7 cells under the drug treatments Figure S5:FGFR2 immunofluorescence in HeLa cells Supplementary Methods & Results Figure S6: Fluorescent anisotropy Figure S7: Pull down assay Figure S8: Best conformations of FGFR2 with thalidomide and analogs Figure S9: 3D representation of the Thalidomide, lenalidomide and pomalidomide bound to FGFR1. Figure S10: Comparison of experimental IC50 and predicted Ki values List of Tables Table S1: Similarities between FGFR implicated syndromes and Thalidomide-mediated teratogenecity Table S2: RT-PCR conditions and primer sequences Table S3: Summary of the expression data of FGFRs in EAhy.926, MCF-7 and HeLa cells under three drugs from Figure 4. Table S4: Consolidated table for FGFR expression in organs Table S5: Interaction of FGFR1 with thalidomide and analogs Table S6 FGFR3 docked with thalidomide and analogs Table S7: Pocketome prediction linking thalidomide with FGFR2.


Thalidomide and its analogs differentially target Fibroblast Growth

Recommend Documents