Multiple myeloma phosphotyrosine proteomic profile associated with FGFR3 expression, ligand activation, and drug inhibition

  1. Jonathan R. St-Germaina,b,
  2. Paul Taylora,c,
  3. Jiefei Tonga,c,
  4. Lily L. Jina,b,
  5. Ana Nikolica,c,
  6. Ian I. Stewartd,
  7. Robert M. Ewingd,e,
  8. Moyez Dharseed,
  9. Zhihua Lic,f,
  10. Suzanne Trudelc,f and
  11. Michael F. Morana,b,c,g,1
  1. aMolecular Structure and Function Program, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada M5G 1X8;
  2. bDepartment of Molecular Genetics, University of Toronto, Toronto, ON, Canada M5S 1A8;
  3. cMcLaughlin Centre for Molecular Medicine and
  4. gBanting and Best Department of Medical Research, University of Toronto, Toronto, ON, Canada M5G 1L7;
  5. dInfochromics Inc., Toronto, ON, Canada M5G 1L7;
  6. eCenter for Proteomics and Bioinformatics, School of Medicine, Case Western Reserve University, Cleveland, OH 44106; and
  7. fHematology-Oncology, Princess Margaret Hospital, Toronto, ON, Canada M5G 1X8

  1. Communicated by Tony Pawson, Samuel Lunenfeld Research Institute, Toronto, ON, Canada, September 28, 2009 (received for review July 24, 2008)

 
Next Section

Abstract

Signaling by growth factor receptor tyrosine kinases is manifest through networks of proteins that are substrates and/or bind to the activated receptors. FGF receptor-3 (FGFR3) is a drug target in a subset of human multiple myelomas (MM) and is mutationally activated in some cervical and colon and many bladder cancers and in certain skeletal dysplasias. To define the FGFR3 network in multiple myeloma, mass spectrometry was used to identify and quantify phosphotyrosine (pY) sites modulated by FGFR3 activation and inhibition in myeloma-derived KMS11 cells. Label-free quantification of peptide ion currents indicated the activation of FGFR3 by phosphorylation of tandem tyrosines in the kinase domain activation loop when cellular pY phosphatases were inhibited by pervanadate. Among the 175 proteins that accumulated pY in response to pervanadate was a subset of 52 including FGFR3 that contained a total of 61 pY sites that were sensitive to inhibition by the FGFR3 inhibitor PD173074. The FGFR3 isoform containing the tandem pY motif in its activation loop was targeted by PD173074. Forty of the drug-sensitive pY sites, including two located within the 35-residue cytoplasmic domain of the transmembrane growth factor binding proteoglycan (and multiple myeloma biomarker) Syndecan-1/CD138, were also stimulated in cells treated with the ligand FGF1, providing additional validation of their link to FGFR3. The identification of these overlapping sets of co-modulated tyrosine phosphorylations presents an outline of an FGFR3 network in the MM model and demonstrates the potential for pharmacodynamic monitoring by label-free quantitative phospho-proteomics.

Tyrosine (Y) phosphorylation is a key mechanism of cell regulation, and tyrosine kinases are frequently activated in cancers (1). FGF receptor-3 (FGFR3) is a receptor tyrosine kinase (RTK) and drug target in a subset of human multiple myelomas (MM) that contain the t(4;14) translocation, which is responsible for the aberrant expression of FGFR3 in these tumors (2). Mutations that activate FGFR3 are prevalent in bladder cancer, have been observed in t(4;14) MM, cervical and colon cancers, and are associated with skeletal dysplasias (3). A fundamental question in MM is the nature of phosphotyrosine (pY) signaling networks in tumors that express FGFR3.

RTKs such as FGFR3 function to a large extent through ligand-induced autophosphorylation, which facilitates the activation of downstream effectors (4). Autophosphorylation of FGFR1 proceeds sequentially, involving one, and then both adjacent tyrosines in the kinase domain activation loop (AL), which causes an approximately 50-fold and 1,000-fold activation of kinase activity, respectively, followed by the modification of other receptor tyrosines (5). FGFR3 also contains tandem tyrosines in its AL (6), and has four additional Y sites including pY760 and pY724, which are linked to downstream signaling effector proteins (7, 8). Systematic in vitro investigations of pY-dependent protein interactions of RTKs with proteins containing SH2 and PTB domains (9, 10) and quantitative analyses of global (i.e., pY, pS, and pT) protein phosphorylation in cells (11) and tissue (12) have broadened our understanding of the signaling potential of RTKs and of the EGFR in particular. However, details of the FGFR3 signaling networks in MM and during human development remain to be fully characterized.

MS-based methods to characterize protein phosphorylation have been reviewed (13–15). The measurement of RTK network-associated phosphorylation in cultured cells (11, 16) or tissue (12) has typically involved labeling strategies and relative measurements or label-free semiquantitative spectral counting (17). A challenge in the definition of RTK-mediated signaling networks is that even in cells expressing activated Y kinases, protein-pY accounts for only less than 2% of protein phosphorylations (11). Treatment of cells with the nonselective pY phosphatase (PTP) inhibitor pervanadate broadly potentiates pY modifications (18, 19), which are naturally occurring (20), and this strategy has been shown to render latent kinase/phosphatase substrates amenable for MS analysis (21).

In the present study, high-resolution MS ion currents derived from unlabeled pY-containing peptides were quantified to identify phosphorylations co-modulated along with FGFR3 following treatment of t(4;14) MM cells with activators (pervanadate, FGF1) and an inhibitor (PD173074) of the kinase. This uncovered a set of proteins whose tyrosine phosphorylation was co-modulated with the receptor and therefore constitutes an outline of the FGFR3 network in the myeloma model.

Previous SectionNext Section

Results

Initial experiments tested the rationale of using pervanadate, FGF1, and a small molecule inhibitor of FGFR3 to uncover protein pY sites associated with FGFR3 activity in cell models. Anti-pY Western blots indicated the MM line KMS12PE, which is negative for the t(4;14) translocation and does not express FGFR3, presented barely detectable cellular pY-containing proteins (Fig. 1A). LP1 cells harbor the t(4;14) translocation and express wild-type (WT) FGFR3 and had only slightly elevated pY protein signals compared to KMS12PE. OPM2 and KMS11 are positive for the t(4;14) translocation and express variants of FGFR3 (K650E and Y373C, respectively) that are constitutively activated to some extent (22). All these MM cells produced a pY signal that was considerably less than the signal obtained with A431 cells (Fig. 1A, lane 5), which express greater than one million wild-type EGF receptors per cell (23). This result demonstrated the relative low level of pY in MM cells compared with the A431 system, which has proven amenable to phosphorylation analysis by MS (12, 24), and the challenge that represents the study of pY signaling in MM cells.

Fig. 1.
View larger version:
Fig. 1.

Phosphorylation in MM cells, and modulation by pervanadate, FGF, and PD173074 (PD). (A) Anti-pY Western blot (WB) of lysates from MM-derived cells (as indicated) that differ in their expression of FGFR3: KMS12PE no expression; LP1, wild-type; OPM2, K650E; KMS11, Y373C. Lane 5 is the adenocarcinoma A431 that expresses approximately 1 M EGFR/cell. Lanes 6–8 are serum-starved KMS11 cells either untreated (lane 5), FGF-stimulated (lane 6), or treated with pervanadate (lane 7). (B) FGFR3 was isolated by immunoprecipitation from KMS11 cells that had been treated with PD173074, FGF, and pervanadate as indicated, and then Western blotted for pY (upper panel) or FGFR3 (lower panel), and then imaged for chemiluminescence. Shown are representative results of repeated experiments. (C) Venn diagram showing nonredundant pY containing peptides (and proteins in parentheses) from KMS11 cells treated without or with pervanadate (VO4) to potentiate cellular protein tyrosine phosphorylation.

To explore the possibility to potentiate protein-pY levels in KMS11 cells, they were tested by treatment with FGF1 and pervanadate. As shown in Fig. 1A, KMS11 cells were responsive to FGF1 stimulation (compare lanes 6 and 7), and very minor differences in pY were observed in KMS11 cells growing in the presence of serum (Fig. 1A, lane 4) and when serum-deprived (lane 6). However, a dramatic elevation in total protein pY was achieved by treatment of KMS11 with pervanadate (lane 8), which indicated that the normally relatively low level of pY seen in KMS11 was maintained in part by PTP activity.

The pyrido-(2,3-d)-pyrimidine-class compound PD173074 inhibits FGFR family members (25) and the proliferation of FGFR3-expressing t(4;14) MM cell lines including KMS11, but not MM cell lines that do not express FGFR3 (26, 27). FGFR3 was to some extent constitutively tyrosine-phosphorylated in KMS11 cells (Fig. 1B, lane 1) as expected (22), and there was a minor increase in anti-pY reactivity following FGF1 treatment (lane 3). A more pronounced increase was seen when the cells were treated with pervanadate (lane 5). PD173074 caused the elimination of detectable FGFR3 pY associated with FGF1 (lanes 2 and 6) and lowered pY levels in the pervanadate-treated cells (Fig. 1B, lane 4).

The qualitative results depicted in Fig. 1 provided a rationale for a phospho-proteomic analysis of the FGFR3 network in MM, as represented by the KMS11 cell model. It was reasoned that pY levels would be generally elevated by pervanadate, and that modifications linked to FGFR3 activity would be susceptible to inhibition by PD173074 and stimulation by FGF treatments. To characterize the tyrosine phospho-proteome of KMS11 cells, pY profiles were generated with cells grown in serum-containing medium with and without pervanadate treatment. In these experiments, whole cell protein preparations (2 mg protein) were digested with trypsin, extracted with anti-pY, and analyzed by LC-MS/MS (28). Analysis of KMS11 cultures grown in serum-containing medium without pervanadate treatment uncovered 51 unique pY-peptides assigned to 34 unique proteins (Table S1). Expectedly, pY-containing peptides from FGFR3 were identified, as were pY-peptides from proteins known to be associated with FGFR3 or FGFR-family signaling including PLC, SHC1, and STAT3 (Table S1). As summarized in Fig. 1C, pervanadate effectively potentiated cellular pY levels. A total of 267 unique pY-peptides assigned to 175 proteins were identified in the pervanadate-treated KMS11 cells (Table S1). Within this set were 44 of the 51 pY-peptides (and 29 of 34 pY proteins) identified in the non-pervanadate-treated cells (Fig. 1C). To the best of our knowledge based on database searches (i.e., phosphosite.org, phospho.elm.eu.org, phosida.com, humanproteinpedia.org), 23 of these pY sites, associated with 23 different proteins, have not been reported previously (listed in Table S1).

Tandem MS analysis uncovered evidence for FGFR3 as the predominant activated Y kinase in KMS11. MS/MS events associated with 11 protein-tyrosine kinases in pervanadate-treated KMS11 cells were detected. Phosphopeptides from FGFR3, JAK1, and TYK2 we observed in each of three independent experiments, and peptides representing FGFR3 were most frequently observed. Of the 54 FGFR3-derived pY spectra counted in these samples, 21 were from the kinase domain AL, containing pY at one or both Y647 and Y648. These results are consistent with FGFR3 being the major, driving kinase in KMS11 (2, 27, 29). Consistent with this, pY residues from several proteins associated with signaling, trafficking, or regulation of RTKs were observed (Table S1) and which reflect a broad range of cellular structures, activities, and processes (Fig. S1). By comparison with the Human Protein Reference Database (hprd.org), 49 of the KMS11 pY proteins are known to interact with RTKs, including 41 that bind a member of the ErbB family (35 with EGFR), 19 that bind IGFR1, and 16 known to interact with an FGFR family kinase (Table S2). FGFR3 pY residues were identified at positions 577, 647, 648, and 724. Phosphorylation at Y760 was not observed, likely because its predicted tryptic peptide, at 5,785 Da, was too large to be efficiently resolved. However, the FGFR3-to-PLCγ signaling axis (7) ascribed to pY760 was evident, since tyrosine phosphorylated PLCγ1 (pY472, pY481, pY977) and PLCγ2 (pY759, pY1216) were observed.

To further connect pY peptides with FGFR3, their modulation by PD173074 was examined. High-resolution MS (LTQ-Orbitrap) extracted ion current (XIC) chromatograms of pY-peptide ions were used to quantify their abundance as a function of pervanadate and PD173074 treatments. PD173074 inhibits FGFR-family kinases with an IC50 of ≈22 nM and at approximately 10-fold higher concentration VEGFR2 and at approximately 1,000-fold higher concentration PDGFR, SRC, EGFR, INSR, MEK, and PKCα/β/γ (25). We found no evidence for the expression/phosphorylation of VEGFR2 or FGFR kinases other than FGFR3 in the myeloma cell lines investigated (Table S1). However, we cannot exclude the possibility that PD173074 affected kinases other than FGFR3 in KMS11 cells.

FGFR3 AL phosphopeptides containing a single pY or the di-pY motif were quantified by measurement of their respective XICs from PD173074-treated and untreated cells (Fig. 2 and Fig. S2 for the corresponding MS and MS/MS spectra and Table S3 for XIC values). The singly phosphorylated AL was detected as two distinct peptides depending on whether trypsin proteolysis occurred at K649 or K650. We suspect that proteolysis at K649 was hindered by the proximal phospho amino acids. Consistent with this interpretation, the doubly phosphorylated AL peptide was only detected as a K650 cleavage product. Pervanadate increased the abundance of both the singly and doubly phosphorylated AL isoforms of FGFR3 by more than 6-fold (Fig. 2 and Table S3). Interestingly, treatment with PD173074 reduced the level of the doubly phosphorylated isoform back to control levels, whereas the singly phosphorylated isoform remained elevated approximately 5-fold above the control level. The effect of PD173074 on the two singly phosphorylated trypsin products (ending in K649 and K650) were roughly equivalent (Table S4). These data indicated the activation of FGFR3 kinase activity by pervanadate and inhibition by PD173074.

Fig. 2.
View larger version:
Fig. 2.

Label-free MS-based quantification of FGFR3 activation loop phosphorylation in KMS11 cells. Extracted MS ion currents for singly (left) and doubly (right) phosphorylated isoforms of the indicated FGFR3 activation loop tryptic peptide are shown (arrows). KMS11 cells were grown in medium with serum as a control (CON), treated with pervanadate (VO4), or pervanadate (VO4) plus PD173074 as indicated.

In order to identify tyrosine phosphorylations that were co-modulated with FGFR3, algorithms (30) were used to compare across experimental samples ion currents associated with individual peptides (chrompeaks) (see Materials and Methods and SI Text). Samples typically produced more than 1,200 discernible chrompeaks with >400 common to control, pervanadate, and pervanadate-plus-drug samples, and used for normalization. In two independent experiments, chrompeaks shared across the three samples and with Mascot scores greater than or equal to 20 had a mean CV of 0.31 (SEM 0.02; n = 198) and 0.37 (SEM 0.01, n = 252), respectively (SI Text and Figs. S3–S6). The analysis of the two independent experiments resulted in the identification and quantification of 218 pY-containing peptides, including a subset representing 61 pY sites from 52 different proteins that decreased in abundance at least 2-fold as a consequence of PD173074 in one or both repeats of the experiment (Table S5 and Table S6).

To further validate the drug-affected pY sites as associated with FGFR3, pY-peptides were tested for increased abundance in response to FGF1 treatment of FGFR3-expressing MM cell lines as summarized in Table 1. The experiment was repeated eight times including four repetitions with KMS11 and two each with the LP1 and OPM2 lines (detailed in Table S7). Since the level of cellular protein-pY was much less in FGF1-stimulated cells than pervanadate-treated cells (Fig. 1), data analysis with the FGF1 experiments focused on the 61 pY sites affected by PD173074 in KMS11, and resulted from the manual calculation of integrated XICs. This resulted in the verification of 40 pY sites (from 34 proteins), which were identified in FGF1-stimulated cells (Table S7).

To generate an outline model of the FGFR3 network based on co-modulated pY sites, the results for the PD173074 and FGF1 experiments were compared by Venn analysis (Fig. S7). Reproducibility between the repeated experiments involving PD173074 extended over 70%: 45 of the 61 drug-modulated pY sites that were decreased at least 2-fold were found as such in both independent experiments. The overlap between the 61 drug-modulated pY sites and those observed following FGF1 stimulation was 65% (40 of 61). Thirty-three pY sites were affected in each of the three experiments, and 52 overlapped in at least two (Fig. S7). Based on these measures of reproducibility and experimental verification, an outline of the FGFR3 network was defined to include the 44 proteins containing the 52 experimentally overlapping pY sites (Table 1).

View this table:
Table 1.

FGFR3 network proteins identified as co-modulated along with FGFR3 by PD173074 and FGF1 ligand

The proteins co-modulated with FGFR3 were subjectively categorized according to cellular function, such as those involved in phosphorylation or signaling, endocytosis, and adhesion/cytoskeleton, which together accounted for 28 of the 44 proteins. Thirteen of the proteins are previously identified as RTK-interacting (Table 1). Additional proteins known or implicated in FGFR3 or RTK signaling and verified as expressed and tyrosine-phosphorylated in pervanadate-treated KMS11 (Table S1), but not co-modulated by the criteria set out above, included the ubiquitin ligase c-CBL, the adaptor proteins SHC1, GAB1, and IRS2, and the MAP kinases ERK1/2, which were shown previously to be constitutively phosphorylated and susceptible to FGF-stimulation (22) and FGFR3 inhibition (29) in KMS11. A SHC1-derived peptide containing the GRB2 SH2 binding pY site at position 427 decreased less than 2-fold following PD173074 (ratio 0.74 ± 0.07; Table S4), and was therefore not included in Table 1. However, the SHC1 pY427 peptide increased 16.5-fold (±9.4, n = 4) in response to FGF1 stimulation (Table S7). This suggests that SHC1 is modulated by both FGFR3 and PD173074-insensitive kinases in MM cells. STAT1 was not detected, while STAT3 phosphorylations at Y539 and Y705 were found, but not modulated by PD173074 or FGF1, consistent with Ronchetti et al. (22).

Previous SectionNext Section

Discussion

The experimental strategy to outline the FGFR3 network in the KMS11 model involved the identification of protein-pY sites modulated in concert with FGFR3. In many tumor types, including even myelomas that overexpress FGFR3 such as KMS11, the level of protein-pY is low compared to many well studied model systems (e.g., Fig. 1). By using pervanadate treatment, cellular protein-pY levels were effectively potentiated, as has been commonly observed (18–21). However, since pervanadate is a nonselective PTP inhibitor, it was important to establish the activation of FGFR3 in the system and then to identify within the larger set of pervanadate-associated pY sites, those linked to FGFR3.

Several pieces of data were consistent with the notion that FGFR3 is a driver or dominant tyrosine kinase in KMS11, and this is consistent with phenotypic data indicating G1 growth arrest, apoptosis, differentiation, and xenograft tumor regression associated with FGFR3 inhibitors in myeloma cells (26, 27). By semiquantitative spectral counting, FGFR3-derived pY peptides were more prevalent than for all other Y kinases combined, and more than one-third of them (21 of 54) were from the kinase domain activation loop, indicative of an activated kinase. Additionally, more precise quantification of AL-derived extracted ion currents confirmed that a fraction of FGFR3 became highly catalytically activated as a consequence of pervanadate treatment and that the receptor is normally inhibited by cellular PTP activity. This interpretation assumes FGFR3 activity is regulated by tandem phosphorylation within the AL similar to FGFR1 (5) and is consistent with evidence that signaling mechanisms are conserved among the FGFR family (31).

The identification and quantification of FGFR3 AL phosphorylation (Fig. 2 and Fig. S2 and Table S3) illustrates the potential utility of pY-directed phospho-proteomics to measure drug pharmacodynamics, since it provided a measure of drug target modulation, and insight into drug mechanisms. For example, PD173074 appeared more selective for the doubly phosphorylated, and therefore most highly catalytically activated isoform of FGFR3 (Table S3), suggesting singly and doubly phosphorylated FGFR3 have distinct structures differentiated by PD173074. A total of 26 protein kinases were detected as pY peptides in pervanadate-treated KMS11 cells (Table S1). Seventeen of these were seen as AL peptides, including the MAP kinases ERK1, ERK2, and p38α. This further illustrates the potential of the pY profiling approach to identify activated protein kinases and reveal potential signaling pathways. This was recently realized in subsets of non-small cell lung carcinomas that were found to express activated PDGFRα and other kinases (17) and in glioblastomas found to express activated Met (32).

The co-modulation of pY sites by FGF1 and/or PD173074 indicated they are linked to FGFR3 activity and are therefore proposed to be part of an FGFR3 network. Even though the evidence suggested FGFR3 was the most prevalent activated kinase in the pervanadate-stimulated cells, the 40 FGF1-associated pY sites represents only a subset of the 61 pY peptides that were sensitive to PD173074. This may indicate that some of the phosphorylations associated with pervanadate and PD173074 are not part of the normal cellular response to ligand-mediated FGFR3 activation. It is also likely that additional FGF1-induced phosphorylations were beyond the limits of detection of the analytical platform. The co-modulated pY sites may be direct substrates of FGFR3 or of kinases and PTPs modulated downstream of it. They are a diverse set of proteins involved in various cellular processes, such as signal transduction, RTK endocytosis, and cell adhesion (Table 1). As discussed below, some are implicated in MM pathology, prognostics, or therapeutics, while others are uncharacterized and/or not previously associated with RTK or FGFR functions. Thirteen network proteins are known RTK-binding proteins (Table 1), and this observation reinforces the dogma that diversity in signal transduction is achieved through overlapping sets of interacting proteins and substrates assembled into context-specific permutations (33).

IGF1R is an established target in MM (34). Activation loop tryptic peptides from the insulin and IGF-1 receptors are identical, and therefore we could not distinguish which (or both) of these receptors was modulated along with FGFR3. It was unlikely the PD173074 modulation of IGF1R/INSR was direct, since it was stimulated by FGF, which provides evidence of crosstalk with FGFR3 (Table 1). We did not detect RSK2 pY529, which was observed previously as an FGFR3 substrate and binding site for ERK (35). However, RSK2 pY707, identified previously as downstream of FGFR1 (36), was found diminished more than 20-fold by PD173074 and stimulated by FGF1 (Table 1). Y707 is a key residue in the autoinhibition of the C-terminal kinase domain of RSK2 (37), and structural analysis suggests its phosphorylation would activate RSK2 (38). Therefore, RSK2 appears tightly regulated by FGFR3 through distinct mechanisms involving phosphorylations at Y529 (35) and Y707. Proteins implicated in the regulation of the Ras-ERK pathway were found modulated along with FGFR3 including SHC1 (described above) and DOK2. DOK2 pY299 is a Ras-GAP binding site and was modulated by FGF1 and PD173074. Some proteins implicated in FGFR-Ras signaling including FRS2, Shp2, and PI3K were not observed in this study, which suggests the FGFR3 network model based on pY profiling alone is incomplete and that more comprehensive pY profiling and data integration is required to fully recognize cell/tumor-specific differences in FGFR3 networks.

A set of proteins in the FGFR3 network was implicated in endocytosis. TOM1L2, initially identified in Blagoev et al. (39) as part of the EGFR network, contains VHS (Vps27/HRS/STAM) and GAT (GGA/Tom1) domains implicated in vesicular trafficking and shown to modulate Src family kinases and interact with endosome components downstream of the EGFR and PDGFR (40–43). HRS interacts with STAM2, contains a VHS domain, and is involved in the ubiquitin-dependent trafficking of RTKs (44, 45). These two proteins, along with STAM, CBL, and the 5′-phosphoinositol phosphatase SHIP2 were previously identified as Y phosphorylated and part of the EGFR network (46). SHIP2 is linked to RTK signaling through its SH2 domain, is an effector of EGFR endocytosis (47), and considered a drug target in breast cancer, where its expression was linked to EGFR inhibitor sensitivity in MDA-231 cells (48). Epsin-4 interacts with clathrin and the AP-2 complex (including AP2B1 identified herein). Also co-modulated with FGFR3 was the small GTPase Rab11, which modulates endosome trafficking, including EGFR recycling (49, 50). The identification of this set of proteins is an indication that FGFR3 is subject to endocytosis similar to other RTKs and that the coordination of this process may be affected by Y phosphorylation downstream of the receptor. This information may be relevant in the development and targeting of FGFR3-driven tumors since endocytosis is a critical factor in the growth, metastasis, and therapeutics of certain cancers (51).

A group of FGFR3 network proteins have roles in cell-cell or cell-matrix interactions or are associated with the cytoskeleton (Table 1). Connexin-43 pY313 was modulated by PD173074 and observed previously in non-small cell lung carcinoma (NSCLC) cells and tissue and as a pY modification highly responsive to the expression level of the oncogenic vIII variant of the EGFR (32). Tensin-3 contains SH2 and PTB domains, couples EGFR activity to actin- and integrin-based adhesion sites, and plays a key role in mammary cell migration (52). SLAMF7 is an integral membrane protein implicated in human B lymphocyte proliferation, and its extracellular engagement is associated with the secretion of cytokine growth factors (53).

The proteoglycan Syndecan-1 (SDC1; also known as CD138) is considered both a biomarker of poor prognosis (54) and a target in MM (55). It is shed from MM cells into the microenvironment, and it acts as a sink that binds heparin-sulfate-binding ligands including FGF2, HB-EGF, VEGF, and HGF, and therefore may be an effector of RTKs that promotes angiogenesis, tumor growth, and metastasis. The observed co-modulation of Syndecan-1 phosphorylation at two tyrosines within its 35-residue cytoplasmic tail suggests that a functional link may exist through which FGFR3 modulates autocrine and paracrine signals in MM. Another identified FGFR3 network protein that may affect FGFR3 activation by ligand is PTTG1IP, which has transforming activity in NIH 3T3 cells and binds the transcription factor PTTG leading to FGF2 expression associated with thyroid cancer (56). We speculate PTTG1IP, which is overexpressed in hematopoietic malignancies (57), may affect production of FGF2, which is a ligand for the FGFR3c isoform that is expressed in MM (4, 58) including KMS11 (22, 59). The role of SH2D4A is unknown, but owing to its SH2 domain may function through phosphorylation-mediated protein interactions. SLC7A6 is involved in amino acid uptake, which has been shown to be inhibited in certain cultured cell systems by pervanadate (60). Relatively uncharacterized network proteins include CDV3, which is upregulated in cells overexpressing ErbB2 (61), the known phosphoprotein CK059, and the membrane protein TM63B.

The results presented in this study provide insight into the function of FGFR3 in the KMS11 model of MM. The “co-modulation” strategy to combine general phosphatase inhibition with specific kinase modulators to identify phosphorylations linked to a given kinase may prove useful in the definition of additional signaling networks. This approach coupled with label-free MS quantification may have particular utility to identify activated kinases and monitor their modulation in tumors and animal models. Defining the molecular details of how FGFR3 network proteins are linked to FGFR3 and the cellular phenotypes associated with their association may provide insight into MM and FGFR family signaling.

Previous SectionNext Section

Materials and Methods

Human MM cells were cultured as described previously (27), and pY peptide enrichment was according to Rush et al. (28). LC-MS/MS involved nano-LC (Proxeon Biosystems A/S) coupled to an LTQ-Orbitrap instrument (Thermo Scientific). Mass spectra were processed with Proteomarker software (Infochromics) (30). Charge determination and deisotoping resulted in a list of monoisotopic mass peaks for each retention time point in the LC/MS gradient. These were grouped across contiguous time points, yielding chrompeaks to which MS/MS precursor ions and top-ranked peptide sequences were linked via mass and time tolerances. Time alignment and intensity normalization were applied; treated samples were normalized against the untreated control in each experiment. Normalized chrompeak areas were used for relative quantification expressed as ratios. SI Text contains a complete description of the materials and methods, details on identified peptides, and statistical analyses.

Previous SectionNext Section

Acknowledgments

This work was supported by the Canada Research Chairs Program, Canadian Institutes for Health Research, and Canadian Cancer Society Research Institute (to M.F.M.).

Previous SectionNext Section

Footnotes

  • 1To whom correspondence should be addressed. E-mail: m.moran{at}utoronto.ca

  • Author contributions: M.F.M. designed research; J.R.S., P.T., and J.T. performed research; L.L.J., A.N., I.I.S., R.M.E., M.D., Z.L., S.T., and M.F.M. contributed new reagents/analytic tools; J.R.S., P.T., J.T., L.L.J., I.I.S., R.M.E., M.D., S.T., and M.F.M. analyzed data; and J.R.S. and M.F.M. wrote the paper.

  • The authors declare no conflict of interest.

  • Data deposition: Data described in this study are freely available through the Human Proteinpedia portal (humanproteinpedia.org), Tranche (proteomecommons.org/tranche), PRIDE (ebi.ac.uk/pride), and GPMDB (thegpm.org).

  • This article contains supporting information online at www.pnas.org/cgi/content/full/0910957106/DCSupplemental.

  • Freely available online through the PNAS open access option.

Previous Section
 

References

  1. ↵
    1. Blume-Jensen P,
    2. Hunter T
    (2001) Oncogenic kinase signaling. Nature 411:355–365.
    CrossRefMedline
  2. ↵
    1. Paterson JL,
    2. et al.
    (2004) Preclinical studies of fibroblast growth factor receptor 3 as a therapeutic target in multiple myeloma. Br J Haematol 124:595–603.
    CrossRefMedlineWeb of Science
  3. ↵
    1. Hideshima T,
    2. Bergsagel PL,
    3. Kuehl WM,
    4. Anderson KC
    (2004) Advances in biology of multiple myeloma: Clinical applications. Blood 104:607–618.
    Abstract/FREE Full Text
  4. ↵
    1. Eswarakumar VP,
    2. Lax I,
    3. Schlessinger J
    (2005) Cellular signaling by fibroblast growth factor receptors. Cytokine Growth Factor Rev 16:139–149.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Furdui CM,
    2. Lew ED,
    3. Schlessinger J,
    4. Anderson KS
    (2006) Autophosphorylation of FGFR1 kinase is mediated by a sequential and precisely ordered reaction. Mol Cell 21:711–717.
    CrossRefMedlineWeb of Science
  6. ↵
    1. Webster MK,
    2. D'Avis PY,
    3. Robertson SC,
    4. Donoghue DJ
    (1996) Profound ligand-independent kinase activation of fibroblast growth factor receptor 3 by the activation loop mutation responsible for a lethal skeletal dysplasia, thanatophoric dysplasia type II. Mol Cell Biol 16:4081–4087.
    Abstract
  7. ↵
    1. Chen J,
    2. et al.
    (2005) Constitutively activated FGFR3 mutants signal through PLCgamma-dependent and -independent pathways for hematopoietic transformation. Blood 106:328–337.
    Abstract/FREE Full Text
  8. ↵
    1. Hart KC,
    2. Robertson SC,
    3. Donoghue DJ
    (2001) Identification of tyrosine residues in constitutively activated fibroblast growth factor receptor 3 involved in mitogenesis Stat activation and phosphatidylinositol 3-kinase activation. Mol Biol Cell 12:931–942.
    Abstract/FREE Full Text
  9. ↵
    1. Jones RB,
    2. Gordus A,
    3. Krall JA,
    4. MacBeath G
    (2006) A quantitative protein interaction network for the ErbB receptors using protein microarrays. Nature 439:168–174.
    CrossRefMedline
  10. ↵
    1. Schulze WX,
    2. Deng L,
    3. Mann M
    (2005) Phosphotyrosine interactome of the ErbB-receptor kinase family. Mol Syst Biol 1:2005.0008.
    Medline
  11. ↵
    1. Olsen JV,
    2. et al.
    (2006) Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127:635–648.
    CrossRefMedlineWeb of Science
  12. ↵
    1. Stover D,
    2. et al.
    (2004) Differential phosphoprofiles of EGF and EGFR kinase inhibitor-treated human tumor cells and mouse xenografts. Clin Proteomics 1:69–80.
    CrossRef
  13. ↵
    1. Moran MF,
    2. Tong J,
    3. Taylor P,
    4. Ewing RM
    (2006) Emerging applications for phospho-proteomics in cancer molecular therapeutics. Biochim Biophys Acta 1766:230–241.
    Medline
  14. ↵
    1. Schmelzle K,
    2. White FM
    (2006) Phosphoproteomic approaches to elucidate cellular signaling networks. Curr Opin Biotechnol 17:406–414.
    CrossRefMedlineWeb of Science
  15. ↵
    1. Thingholm TE,
    2. Jensen ON,
    3. Larsen MR
    (2009) Analytical strategies for phosphoproteomics. Proteomics 9:1451–1468.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Zhang Y,
    2. et al.
    (2005) Time-resolved mass spectrometry of tyrosine phosphorylation sites in the epidermal growth factor receptor signaling network reveals dynamic modules. Mol Cell Proteomics 4:1240–1250.
    Abstract/FREE Full Text
  17. ↵
    1. Rikova K,
    2. et al.
    (2007) Global survey of phosphotyrosine signaling identifies oncogenic kinases in lung cancer. Cell 131:1190–1203.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Irish JM,
    2. et al.
    (2004) Single cell profiling of potentiated phospho-protein networks in cancer cells. Cell 118:217–228.
    CrossRefMedlineWeb of Science
  19. ↵
    1. Kamps MP,
    2. Sefton BM
    (1988) Most of the substrates of oncogenic viral tyrosine protein kinases can be phosphorylated by cellular tyrosine protein kinases in normal cells. Oncogene Res 3:105–115.
    MedlineWeb of Science
  20. ↵
    1. Pan C,
    2. Gnad F,
    3. Olsen JV,
    4. Mann M
    (2008) Quantitative phosphoproteome analysis of a mouse liver cell line reveals specificity of phosphatase inhibitors. Proteomics 8:4534–4546.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Boeri Erba E,
    2. et al.
    (2007) Quantitation of multisite EGF receptor phosphorylation using mass spectrometry and a novel normalization approach. J Proteome Res 6:2768–2785.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Ronchetti D,
    2. et al.
    (2001) Deregulated FGFR3 mutants in multiple myeloma cell lines with t(4;14): Comparative analysis of Y373C, K650E and the novel G384D mutations. Oncogene 20:3553–3562.
    CrossRefMedlineWeb of Science
  23. ↵
    1. Sunada H,
    2. Peacock J,
    3. Mendelsohn J
    (1991) Ligand induced internalization of epidermal growth factor receptors by A431 cells decreases at high cell densities in culture. Growth Factors 5:45–55.
    CrossRefMedline
  24. ↵
    1. Chen Y,
    2. et al.
    (2007) Phosphoproteomics identified Endofin, DCBLD2, and KIAA0582 as novel tyrosine phosphorylation targets of EGF signaling and Iressa in human cancer cells. Proteomics 7:2384–2397.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Mohammadi M,
    2. et al.
    (1998) Crystal structure of an angiogenesis inhibitor bound to the FGF receptor tyrosine kinase domain. EMBO J 17:5896–5904.
    CrossRefMedlineWeb of Science
  26. ↵
    1. Grand EK,
    2. Chase AJ,
    3. Heath C,
    4. Rahemtulla A,
    5. Cross NC
    (2004) Targeting FGFR3 in multiple myeloma: Inhibition of t(4;14)-positive cells by SU5402 and PD173074. Leukemia 18:962–966.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Trudel S,
    2. et al.
    (2004) Inhibition of fibroblast growth factor receptor 3 induces differentiation and apoptosis in t(4;14) myeloma. Blood 103:3521–3528.
    Abstract/FREE Full Text
  28. ↵
    1. Rush J,
    2. et al.
    (2005) Immunoaffinity profiling of tyrosine phosphorylation in cancer cells. Nat Biotechnol 23:94–101.
    CrossRefMedlineWeb of Science
  29. ↵
    1. Xin X,
    2. et al.
    (2006) CHIR-258 is efficacious in a newly developed fibroblast growth factor receptor 3-expressing orthotopic multiple myeloma model in mice. Clin Cancer Res 12:4908–4915.
    Abstract/FREE Full Text
  30. ↵
    1. Le Bihan T,
    2. et al.
    (2006) Differential analysis of membrane proteins in mouse fore- and hindbrain using a label-free approach. J Proteome Res 5:2701–2710.
    CrossRefMedlineWeb of Science
  31. ↵
    1. Hart KC,
    2. et al.
    (2000) Transformation and Stat activation by derivatives of FGFR1, FGFR3, and FGFR4. Oncogene 19:3309–3320.
    CrossRefMedlineWeb of Science
  32. ↵
    1. Huang PH,
    2. et al.
    (2007) Quantitative analysis of EGFRvIII cellular signaling networks reveals a combinatorial therapeutic strategy for glioblastoma. Proc Natl Acad Sci USA 104:12867–12872.
    Abstract/FREE Full Text
  33. ↵
    1. Pawson T
    (2004) Specificity in signal transduction: From phosphotyrosine-SH2 domain interactions to complex cellular systems. Cell 116:191–203.
    CrossRefMedlineWeb of Science
  34. ↵
    1. Mitsiades CS,
    2. et al.
    (2004) Inhibition of the insulin-like growth factor receptor-1 tyrosine kinase activity as a therapeutic strategy for multiple myeloma, other hematologic malignancies, and solid tumors. Cancer Cell 5:221–230.
    CrossRefMedlineWeb of Science
  35. ↵
    1. Kang S,
    2. et al.
    (2007) FGFR3 activates RSK2 to mediate hematopoietic transformation through tyrosine phosphorylation of RSK2 and activation of the MEK/ERK pathway. Cancer Cell 12:201–214.
    CrossRefMedlineWeb of Science
  36. ↵
    1. Hinsby AM,
    2. Olsen JV,
    3. Bennett KL,
    4. Mann M
    (2003) Signaling initiated by overexpression of the fibroblast growth factor receptor-1 investigated by mass spectrometry. Mol Cell Proteomics 2:29–36.
    Abstract/FREE Full Text
  37. ↵
    1. Poteet-Smith CE,
    2. Smith JA,
    3. Lannigan DA,
    4. Freed TA,
    5. Sturgill TW
    (1999) Generation of constitutively active p90 ribosomal S6 kinase in vivo. Implications for the mitogen-activated protein kinase-activated protein kinase family. J Biol Chem 274:22135–22138.
    Abstract/FREE Full Text
  38. ↵
    1. Malakhova M,
    2. et al.
    (2008) Structural basis for activation of the autoinhibitory C-terminal kinase domain of p90 RSK2. Nat Struct Mol Biol 15:112–113.
    CrossRefMedlineWeb of Science
  39. ↵
    1. Blagoev B,
    2. Ong SE,
    3. Kratchmarova I,
    4. Mann M
    (2004) Temporal analysis of phosphotyrosine-dependent signaling networks by quantitative proteomics. Nat Biotechnol 22:1139–1145.
    CrossRefMedlineWeb of Science
  40. ↵
    1. Collin G,
    2. Franco M,
    3. Simon V,
    4. Benistant C,
    5. Roche S
    (2007) The Tom1L1-clathrin heavy chain complex regulates membrane partitioning of the tyrosine kinase Src required for mitogenic and transforming activities. Mol Cell Biol 27:7631–7640.
    Abstract/FREE Full Text
  41. ↵
    1. Seykora JT,
    2. Mei L,
    3. Dotto GP,
    4. Stein PL
    (2002) 'Srcasm: A novel Src activating and signaling molecule. J Biol Chem 277:2812–2822.
    Abstract/FREE Full Text
  42. ↵
    1. Franco M,
    2. et al.
    (2006) The adaptor protein Tom1L1 is a negative regulator of Src mitogenic signaling induced by growth factors. Mol Cell Biol 26:1932–1947.
    Abstract/FREE Full Text
  43. ↵
    1. Puertollano R
    (2005) Interactions of TOM1L1 with the multivesicular body sorting machinery. J Biol Chem 280:9258–9264.
    Abstract/FREE Full Text
  44. ↵
    1. Abella JV,
    2. et al.
    (2005) Met/Hepatocyte growth factor receptor ubiquitination suppresses transformation and is required for Hrs phosphorylation. Mol Cell Biol 25:9632–9645.
    Abstract/FREE Full Text
  45. ↵
    1. Stern KA,
    2. et al.
    (2007) Epidermal growth factor receptor fate is controlled by Hrs tyrosine phosphorylation sites that regulate Hrs degradation. Mol Cell Biol 27:888–898.
    Abstract/FREE Full Text
  46. ↵
    1. Steen H,
    2. Kuster B,
    3. Fernandez M,
    4. Pandey A,
    5. Mann M
    (2002) Tyrosine phosphorylation mapping of the epidermal growth factor receptor signaling pathway. J Biol Chem 277:1031–1039.
    Abstract/FREE Full Text
  47. ↵
    1. Prasad NK,
    2. Decker SJ
    (2005) SH2-containing 5′-inositol phosphatase, SHIP2, regulates cytoskeleton organization and ligand-dependent down-regulation of the epidermal growth factor receptor. J Biol Chem 280:13129–13136.
    Abstract/FREE Full Text
  48. ↵
    1. Prasad NK,
    2. Tandon M,
    3. Badve S,
    4. Snyder PW,
    5. Nakshatri H
    (2008) Phosphoinositol phosphatase SHIP2 promotes cancer development and metastasis coupled with alterations in EGF receptor turnover. Carcinogenesis 29:25–34.
    Abstract/FREE Full Text
  49. ↵
    1. Palmieri D,
    2. Bouadis A,
    3. Ronchetti R,
    4. Merino MJ,
    5. Steeg PS
    (2006) Rab11a differentially modulates epidermal growth factor-induced proliferation and motility in immortal breast cells. Breast Cancer Res Treat 100:127–137.
    CrossRefMedlineWeb of Science
  50. ↵
    1. Ullrich O,
    2. Reinsch S,
    3. Urbe S,
    4. Zerial M,
    5. Parton RG
    (1996) Rab11 regulates recycling through the pericentriolar recycling endosome. J Cell Biol 135:913–924.
    Abstract/FREE Full Text
  51. ↵
    1. Mosesson Y,
    2. Mills GB,
    3. Yarden Y
    (2008) Derailed endocytosis: An emerging feature of cancer. Nat Rev Cancer 8:835–850.
    CrossRefMedlineWeb of Science
  52. ↵
    1. Katz M,
    2. et al.
    (2007) A reciprocal tensin-3-cten switch mediates EGF-driven mammary cell migration. Nat Cell Biol 9:961–969.
    CrossRefMedlineWeb of Science
  53. ↵
    1. Lee JK,
    2. Mathew SO,
    3. Vaidya SV,
    4. Kumaresan PR,
    5. Mathew PA
    (2007) CS1 (CRACC, CD319) induces proliferation and autocrine cytokine expression on human B lymphocytes. J Immunol 179:4672–4678.
    Abstract/FREE Full Text
  54. ↵
    1. Seidel C,
    2. et al.
    (2000) Serum syndecan-1: A new independent prognostic marker in multiple myeloma. Blood 95:388–392.
    Abstract/FREE Full Text
  55. ↵
    1. Yang Y,
    2. et al.
    (2007) The syndecan-1 heparan sulfate proteoglycan is a viable target for myeloma therapy. Blood 110:2041–2048.
    Abstract/FREE Full Text
  56. ↵
    1. Stratford AL,
    2. et al.
    (2005) Pituitary tumor transforming gene binding factor: A novel transforming gene in thyroid tumorigenesis. J Clin Endocrinol Metab 90:4341–4349.
    Abstract/FREE Full Text
  57. ↵
    1. Dominguez A,
    2. et al.
    (1998) hpttg, a human homologue of rat pttg, is overexpressed in hematopoietic neoplasms. Evidence for a transcriptional activation function of hPTTG. Oncogene 17:2187–2193.
    CrossRefMedlineWeb of Science
  58. ↵
    1. Bernard-Pierrot I,
    2. et al.
    (2006) Oncogenic properties of the mutated forms of fibroblast growth factor receptor 3b. Carcinogenesis 27:740–747.
    Abstract/FREE Full Text
  59. ↵
    1. Otsuki T,
    2. et al.
    (1999) Expression of fibroblast growth factor and FGF-receptor family genes in human myeloma cells, including lines possessing t(4;14)(q16.3;q32.3) and FGFR3 translocation. Int J Oncol 15:1205–1212.
    MedlineWeb of Science
  60. ↵
    1. Tsiani E,
    2. Abdullah N,
    3. Fantus IG
    (1997) Insulin-mimetic agents vanadate and pervanadate stimulate glucose but inhibit amino acid uptake. Am J Physiol 272:C156–C162.
    MedlineWeb of Science
  61. ↵
    1. Fukumaru S,
    2. et al.
    (2002) Novel mRNA molecules are induced in hypertrophied ventricles of carnitine-deficient mice and belong to a family of up-regulated gene in cells overexpressing c-erbB-2. Biochim Biophys Acta 1577:437–444.
    Medline
« Previous | Next Article »Table of Contents
OPEN ACCESS ARTICLE

This Article

  1. Published online before print November 9, 2009, doi: 10.1073/pnas.0910957106 PNAS November 24, 2009 vol. 106 no. 47 20127-20132
  1. Free via Open Access: OA
  2. OA Abstract
  3. Figures Only
  4. » Full Text
  5. Full Text (PDF)
  6. Supporting Information

Readers Also Viewed

Classifications

Navigate This Article

Link: Advanced Search

This Week's Issue

  1. February 9, 2010, 107 (6)
  1. Current Issue
  1. Alert me to new issues of PNAS

Most