A Phosphotyrosine Proteomic Screen Identifies Multiple Tyrosine Kinase Signaling Pathways Aberrantly Activated in Malignant Mesothelioma

Abstract

Malignant mesothelioma (MM) is a highly aggressive cancer that is refractory to all current chemotherapeutic regimens. Therefore, uncovering new rational therapeutic targets is imperative in the field. Tyrosine kinase signaling pathways are aberrantly activated in many human cancers and are currently being targeted for chemotherapeutic intervention. Thus, we sought to identify tyrosine kinases hyperactivated in MM. An unbiased phosphotyrosine proteomic screen was employed to identify tyrosine kinases activated in human MM cell lines. From this screen, we have identified novel signaling molecules, such as JAK1, STAT1, cortactin (CTTN), FER, p130Cas (BCAR1), SRC, and FYN as tyrosine phosphorylated in human MM cell lines. Additionally, STAT1 and SRC family kinases (SFK) were confirmed to be active in primary MM specimens. We also confirmed that known signal transduction pathways previously implicated in MM, such as EGFR and MET signaling axes, are coactivated in the majority of human MM specimens and cell lines tested. EGFR, MET, and SFK appear to be coactivated in a significant proportion of MM cell lines, and dual inhibition of these kinases was demonstrated to be more efficacious for inhibiting MM cell viability and downstream effector signaling than inhibition of a single tyrosine kinase. Consequently, these data suggest that tyrosine kinase inhibitor monotherapy may not represent an efficacious strategy for the treatment of MM due to multiple tyrosine kinases potentially signaling redundantly to cellular pathways involved in tumor cell survival and proliferation.

Keywords

tyrosine kinases asbestos mesothelioma proteomics coactivation

Introduction

Malignant mesothelioma (MM) is an aggressive cancer that develops from mesothelial cells lining the pleural, peritoneal, and pericardial cavities. Approximately 2500 MM cases are diagnosed annually in the United States, and the disease is causally associated with exposure to asbestos.^1-3 Current chemotherapy regimens, which include combined treatment with cisplatin and premetrexed, only improve median survival rates by approximately 3 months when compared to single-agent premetrexed or cisplatin therapies.^2,3 Surgical resection, where a large portion of the diffuse tumor is removed, is theoretically the most effective treatment; however, microscopic portions of the tumor inevitably remain and continue to grow.⁴ As a direct result of the lack of effective treatment options for MM patients, the disease prognosis remains fatal in virtually all cases. Thus, discovering new molecular targets for therapeutic intervention remains a focus of many laboratories studying MM and other refractory cancers.

Our understanding of the molecular and cellular mechanisms that drive MM has improved in the past 2 decades. We now know that many genetic and cell signaling perturbations are involved in the development and progression of MM. Inactivation of tumor suppressor genes such as NF2 and CDKN2A/ARF (p16 ^INK4A/p14 ^ARF) occur in a large proportion of MMs.^5-9 Moreover, engineered deficiency of these tumor suppressors in knockout mice accelerates asbestos-induced MM, confirming the importance of inactivation of these tumor suppressors to the malignant phenotype.^10-12

Recent work has also identified cellular signaling axes that contribute to cell survival, proliferation, and chemoresistance of MM cells. Our laboratory identified the phosphatidylinositol-3 kinase (PI3K)–AKT-mammalian target of the rapamycin (mTOR) pathway as being hyperactivated in a high percentage of human and murine MMs and demonstrated that inhibiting this pathway pharmacologically increases sensitivity to cisplatin.¹³ The mTOR tributary of PI3K-AKT signaling appears to be particularly important in mediating survival signals in MM, especially in tumors with high baseline expression of activated p70 S6-kinase,^14,15 a downstream effector involved in protein translation. Other studies have shown that asbestos exposure stimulates epidermal growth factor receptor (EGFR) signaling and activation of the mitogen-activated protein kinase (MAPK) cascade in primary mesothelial cell cultures.^16,17 The MAPK signaling axis itself is hyperactive in MM cells and tumors.^18-21 Thus, inhibitors to the PI3K-AKT-mTOR and MAPK pathways are currently being evaluated as potential therapies in preclinical models of MM.

The PI3K-AKT-mTOR and MAPK pathways are regulated by various upstream kinases, including receptor and nonreceptor tyrosine kinases.²² The tyrosine kinase family represents a large class of proteins whose aberrant activity has been linked to many human cancers.²³ Recent studies have focused on identifying novel receptor and nonreceptor tyrosine kinases activated in a variety of cancers, including glioblastoma, non–small cell lung cancer, and colorectal cancer.^24-27 The identified kinases, including MET, EGFR, PDGFR, IGFR, and CSF1R, are currently being evaluated preclinically as potential therapeutic targets for these cancers.^24-27 With these new findings in mind, we initiated experiments to identify tyrosine kinase family members activated in MM. Using an unbiased phosphotyrosine proteome screen, we identified two receptor tyrosine kinases, EGFR and MET, previously reported to be hyperactivated in MM, as well as nonreceptor tyrosine kinases SRC, FYN, JAK1, and FER; the transcription factor STAT1; and the scaffolding proteins cortactin (CTTN) and p130Cas (breast cancer anti–estrogen resistance 1, BCAR1) as tyrosine phosphorylated proteins in MM cells. We further demonstrated that STAT1 phosphorylation is relatively common and specific to MM cell lines and tumors when compared to nontransformed mesothelial cells and normal mesothelial tissue. The SRC family kinases (SFK), including SRC and FYN, were commonly activated in MM cells and primary tumor specimens, and we determined that dual tyrosine kinase inhibitor (TKI) therapy is more effective at inhibiting downstream effector signaling and cell viability than TKI monotherapies in MM cell lines with concomitant activation of multiple tyrosine kinases. Together, these data indicate that many tyrosine kinase signaling axes are concomitantly activated in MM cells and that mono-TKI therapies may not be efficacious for the treatment of MM.

Results

Phosphotyrosine proteomic screen identifies tyrosine kinase signaling pathways in MM

To determine if tyrosine kinases are hyperactivated in MM cells, we evaluated the relative levels of pan-phosphotyrosine proteins in human MM cell lines in comparison to normal mesothelial cells. As shown in Figure 1A, we observed strong phosphotyrosine protein levels in the majority of MM cell lines. In contrast, normal mesothelial cells had undetectable levels of phosphotyrosine proteins (Fig. 1A). We also observed strong phosphotyrosine immunohistochemical (IHC) staining in MM tumors, with modest expression observed in normal mesothelial lining from lung pleura (Fig. 1B).

Figure 1.

An unbiased phosphotyrosine proteomic screen identifies tyrosine kinases aberrantly activated in malignant mesothelioma (MM). (A) Western blot analysis of phosphotyrosine protein expression (PY20) in human MM cell lines compared to normal mesothelial cells. (B) Immunohistochemistry of phosphotyrosine protein expression in primary MMs (PY20) compared to that of normal mesothelial lining (arrows) from normal lung pleura. (C) Scheme for purifying and identifying phosphotyrosine proteins in MM cells. (D) Coomassie-stained gel of phosphotyrosine proteins immunopurified from MM cell lysates. (E) Validation of mass spectrometry queries by co-immunoprecipitation Western blot analysis.

From these initial experiments, we hypothesized that multiple tyrosine kinase signaling pathways may be aberrantly activated in MM. To identify the phosphotyrosine proteins observed in the MM cell lines, we immunoprecipitated proteins from MM cell lysate with the same pan-phosphotyrosine antibody used for the immunoblot analysis in Figure 1A (i.e., antibody PY20—scheme shown in Fig. 1C). Proteins pulled down with the phosphotyrosine antibody or control IgG antibody were separated by SDS–gel electrophoresis and stained with Coomassie blue (Fig. 1D). All unique bands were excised from the gel, processed, and run by liquid chromatography/tandem mass spectrometry (LC/MS/MS) to identify specific proteins.

Proteins identified from the mass spectrometry analysis were further analyzed using criteria that ensured that (a) the protein of interest was a known phosphotyrosine accepting protein, (b) the protein had adequate peptide coverage, and (c) the protein came from a Coomassie-stained band that corresponded closely to the molecular weight of that protein. Screen hits were further validated by evaluating the tyrosine phosphorylation status of the protein in MM cell lysates (Fig. 1E). Nine validated phosphotyrosine proteins identified from the proteomic analysis are shown in Table 1 and Figure 1E. Two of the proteins identified, EGFR and MET, are receptor tyrosine kinases previously reported to be aberrantly activated in MM cells,^28,29 thus validating the methodology of our screen. We also identified novel nonreceptor tyrosine kinases SRC, FYN, FER, and JAK1. In addition to tyrosine kinases, the transcription factor Signal Transducers and Activators of Transcription 1 (STAT1) and scaffolding proteins cortactin and p130Cas were also found to be tyrosine phosphorylated in MM cells (Table 1 and Fig. 1E).

Table 1.

Mass Spectrometry of Results of Proteins of Interest from Phosphotyrosine Immunoprecipitation/Coomassie Gel

Protein Identity	SwissProt Entry	Gene Name	Sequence Coverage (%)	Mascot Score
Epidermal growth factor receptor	P00533	EGFR	13	215
Hepatocyte growth factor receptor	P08581	MET	9	215
Proto-oncogene tyrosine-protein kinase Src	P12931	SRC	15	114
Signal transducer and activator of transcription 1-alpha/beta	P42224	STAT1	11	81
Tyrosine-protein kinase JAK1	P23458	JAK1	7	81
Src substrate cortactin	Q14247	CTTN	5	72
Proto-oncogene tyrosine-protein kinase Fyn	P06241	FYN	11	61
Proto-oncogene tyrosine-protein kinase FER	P16591	FER	4	52
Breast cancer anti–estrogen resistance protein 1	p56945	BCAR1	2	42

STAT1 signaling axis is aberrantly activated in MM

Tyrosine phosphorylation of STAT1 allows nuclear translocation and transcriptional activation of genes that negatively regulate cell proliferation and survival.³⁰ STAT1 is also an important transcriptional regulator of inflammation. Some cancers, including MM,³¹ are thought to require inflammation to promote tumorigenesis. Recently, STAT1 phosphorylation was demonstrated to be required for colitis-induced gastric carcinogenesis in mice.³² Thus, STAT1 phosphorylation-mediated regulation of inflammation may be an important step in MM development, in addition to STAT1’s role as a tumor suppressor. As such, we were interested in evaluating if STAT1 phosphorylation is a common event in MM.

We first evaluated the tyrosine phosphorylation status of STAT1 in 4 MM cell lines, Meso 6, 12, 43, and H-Meso, when compared to 2 nontransformed mesothelial cells, LP9-hTERT and 7086. Interestingly, we observed higher levels of P-STAT1 in 3 of the 4 MM cell lines in comparison to both control cell lines (Fig. 2A). All cell lines, MM and nontransformed mesothelial cell cultures, expressed total STAT1 protein at similar levels, suggesting that STAT1 phosphorylation is specific to the MM cells. We also observed strong nuclear P-STAT1 staining in primary MM specimens, with no staining observed in normal mesothelial lining (Fig. 2B).

Figure 2.

STAT1 is frequently phosphorylated in malignant mesothelioma (MM). (A) Western blot analysis of STAT1 phosphorylation (Y701) and total STAT1 levels in MM versus normal control cells. Actin is included as a loading control. (B) Immunohistochemistry of P-STAT1 (Y701) expression in primary MMs versus normal mesothelial lining (arrows) from lung pleura. Nuclear P-STAT1 staining is marked by arrows. (C) Immunoblot analysis of P-JAK1 (Y1022/1023), total JAK1, P-STAT1 (Y701), and total STAT1 in a panel of human MM cell lines. Actin is included as a loading control.

To assess the frequency of STAT1 phosphorylation in MM, we evaluated both JAK1, a proteomic screen hit and known upstream regulator of STAT1, and STAT1 phosphorylation in a panel of human MM cell lines. STAT1 was phosphorylated in 8 of 13 MM cell lines tested, and JAK1 was phosphorylated in 5 of the 13 cell lines (Fig. 2C). Interestingly, JAK1 phosphorylation did not always correlate with STAT1 phosphorylation (e.g., Meso 6 and 22), suggesting that additional upstream regulators contribute to STAT1 activation (Fig. 2C). Together, these data demonstrate that the STAT1 signaling pathway is frequently activated in MM cells.

Src family kinases (SFK) are hyperactivated in MM

Two SFK members, SRC and FYN, were validated as tyrosine-phosphorylated proteins in MM cells (Fig. 1). Interestingly, YES, another SFK member, was also identified by mass spectrometry as a phosphotyrosine protein in MM cells, but it was ruled out by our aforementioned criteria. We have since confirmed YES phosphorylation in 4 MM cell lines using antibody arrays (data not shown). Thus, it appears that multiple SFK are strongly phosphorylated in MM cells and may represent novel therapeutic targets in this disease.

To further investigate SFK activity in MM, we used an antibody that recognizes all SFK members when tyrosine phosphorylated at residue 416, an activating phosphorylation event for SFK. When comparing MM cells to nontransformed mesothelial cells by immunoblot analysis, we observed strong SFK phosphorylation in virtually all MM cell lines tested with very little P-SFK observed in control cells (Fig. 3A). Consistent with these results, we observed strong P-SFK IHC staining in MM tumor specimens and no detectable staining in normal mesothelial lining (Fig. 3B). Thus, SFK phosphorylation appears to be specific to MM cells and not to normal mesothelial cells.

Figure 3.

SFK are phosphorylated in malignant mesothelioma (MM). (A) Western blot analysis of P-SFK (Y416), SRC, and FYN levels in MM versus control mesothelial cells. Actin is included as a loading control. (B) Immunohistochemistry of P-SFK (Y416) expression in primary MMs versus normal mesothelial lining (arrows) from lung pleura. (C) Western blot analysis of P-SFK (416), SRC, and FYN expression in a panel of human MM cell lines. Actin is included as a loading control.

To determine if SFK activation is common in MM, we evaluated SFK phosphorylation in a panel of MM cell lines. SFK activation was observed in most (11 out 13) MM cell lines tested (Fig. 3C). SRC and FYN were also expressed in a large proportion of the MM cell lines (Fig. 3C). Taken together, these data validate SFK as tyrosine kinases that are frequently activated in MM.

EGFR and MET are coactivated in MM

EGFR and MET, 2 receptor tyrosine kinases previously implicated in MM, were also found to be tyrosine phosphorylated in MM cells from our proteomic screen (Fig. 1D,E). We evaluated the frequency of EGFR and MET phosphorylation in a panel of MM cell lines by immunoprecipitation/immunoblot analysis. Both EGFR and MET were coactivated in 8 of 11 cell lines tested (Fig. 4A). Consistent with this result, we also observed concomitant IHC staining for P-EGFR and P-MET in primary MM samples (Fig. 4B). When comparing SFK phosphorylation to EGFR and MET activation, we also found that half of the MM cell lines had concomitant activation of all 3 tyrosine kinases (Figs. 3B and 4A). Collectively, these data suggest that multiple tyrosine kinases that control cell proliferation and survival are concurrently activated in MM.

Figure 4.

EGFR and MET are concomitantly activated in malignant mesothelioma (MM). (A) Immunoblot analysis of P-EGFR, total EGFR, P-MET, and total MET in a panel of MM cell lines. Actin is included as a loading control. (B) Immunohistochemistry of P-EGFR (Y845) and P-MET (Y1230/1234/1235) expression in primary MM.

Combinatorial inhibition of EGFR, MET, and/or SFK decreases MM cell viability more effectively than mono-TKI treatments

Multiple tyrosine kinases are hyperactivated in many human cancers, and thus, small tyrosine kinase inhibitors have been exploited for the treatment of these diseases. EGFR, MET, and SFK can signal downstream to redundant signaling effectors such as PI3K-AKT-mTOR and MAPK signaling cascades. The aforementioned downstream effects are known to be hyperactivated in MM and have been shown to contribute to MM cell proliferation and survival. Because EGFR, MET, and SFK can all signal to these oncogenic pathways, mono-TKI therapy against one of these tyrosine kinases (e.g., EGFR) may not be effective if other tyrosine kinases such as MET or SFK signal redundantly to these same cell proliferation/survival pathways.

To evaluate if targeting multiple tyrosine kinases is more efficacious at inhibiting MM cell viability than mono-TKI inhibitors, we treated 2 MM cell lines (Meso 8 and Meso 43) that exhibit concomitant activation of the aforementioned tyrosine kinases with small-molecule inhibitors against EGFR (AG1478), MET (MET Kinase Inhibitor), and SFK (PP2) alone or in combination. Before conducting the experiments with various drug combinations, we first determined the lowest concentration needed for each small-molecule inhibitor to inhibit the activity of EGFR, MET, and SFK, respectively, in both Meso 8 and Meso 43 (Fig. 5A). Interestingly, the concentration required to inhibit phosphorylation of EGFR, MET, and SFK with AG1478, MET Kinase Inhibitor, and PP2, respectively, was consistently lower than the IC₅₀ concentration calculated for each drug (Fig. 5A), suggesting that these drugs may inhibit cell viability through off-target effects at high concentrations. Based on the initial cell viability data, Meso 8 and 43 cells were then treated with doses of AG1478 (designated treatment E in Fig. 5B,C), MET Kinase Inhibitor (M), and PP2 (P) that resulted in a loss of cell viability of 20% to 40% when used as a single agent. Cells were treated with AG1478 (15 µM), MET Kinase Inhibitor (2 µM), PP2 (5 µM), or dual combinations of these inhibitors (EM, MP, EP) for 72 hours, and cell viability was evaluated by MTT assay. As shown in Figure 5B, EGFR and MET coinhibition (EM) was slightly more efficient at inhibiting cell viability compared to EGFR or MET alone in Meso 8 cells (P = 0.025). Similarly, MET and SFK inhibition (MP) also inhibited cell viability more efficiently than inhibition of MET or SFK alone in both cell lines tested (Fig. 5B; Meso 8: M or P vs. MP, P = 0.025; Meso 43: M vs. MP, P = 0.012). Coinhibition of EGFR and SFK was found to inhibit cell viability to the greatest extent, again more efficiently than inhibition of either EGFR or SFK alone (Fig. 5B; Meso 8: E or P vs. EP, P = 0.025; Meso 43: E or P vs. EP, P < 0.0004).

Figure 5.

Dual inhibition of concomitantly activated tyrosine kinases is more efficacious at inhibiting malignant mesothelioma (MM) cell viability and downstream effector signaling. (A) Meso 8 and Meso 43 cells were seeded on a 96-well plate and treated with increasing concentrations of AG1478 (EGFR inhibitor), MET Kinase Inhibitor, or PP2 (SFK inhibitor) for 72 hours, and then cell viability was determined by MTT assay. Western blot analysis was performed to evaluate a minimal concentration of each drug needed to inhibit EGFR, MET, and SFK phosphorylation, respectively (24 hours posttreatment). Total EGFR, MET, and Src, as well as actin, are included as loading controls. (B) Meso 8 and 43 cells were seeded on a 96-well plate and treated or not with AG1478 (E), MET Kinase Inhibitor (M), PP2 (P), AG1478 + MET Kinase Inhibitor (EM), MET Kinase Inhibitor + PP2 (MP), and AG1478 + PP2 (EP) for 72 hours, followed by assessment of cell viability by MTT assay. (C) Meso 8 and 43 cells were treated with the aforementioned drug combinations and were harvested 24 hours later for Western blot analysis. P-AKT, P-ERK1/2, and P-S6RP levels were evaluated to determine drug effects on AKT, MAPK, and mTOR activity, respectively. Total AKT, ERK1/2, and S6RP, as well as actin, are included as loading controls.

We also conducted in vitro studies of the effect of combining molecularly targeted drugs with two different chemotherapeutic agents. In these experiments, inhibition of EGFR, SRC, or MET showed no consistent effect on the sensitivity of mesothelioma cells to gemcitabine or doxorubicine (data not shown).

The tyrosine kinases EGFR, MET, and SFK can signal redundantly to downstream effector pathways such as the PI3K-AKT-mTOR and MAPK pathways that contribute to tumor cell survival and proliferation. Thus, a tumor cell with tyrosine kinases concomitantly activated may confer resistance to mono-TKI therapy due to redundant signaling to downstream effectors by other tyrosine kinases. To determine if dual-TKI treatment is more effective at inhibiting downstream signaling, Meso 8 and 43 cells were treated with the aforementioned drug combinations, and immunoblot analysis with phospho-specific antibodies against AKT, ERK1/2, and S6RP was performed to determine relative effects on downstream effector signaling. Dual inhibition of EGFR and MET, MET and SFK, and EGFR and SFK was more efficient at inhibiting AKT phosphorylation than mono-TKI treatments in Meso 8, whereas inhibition of both EGFR and SFK repressed AKT phosphorylation more than any other mono-TKI or dual-TKI treatment in Meso 43, as assessed by immunoblot analysis (Fig. 5C). Interestingly, EGFR and SFK alone inhibited AKT phosphorylation in Meso 43 but not in Meso 8 (Fig. 5C). MET inhibition was ineffective at inhibiting AKT phosphorylation in either cell line.

Another pathway downstream of EGFR, MET, and SFK implicated in MM tumor cell survival/proliferation is the MAPK signaling cascade. To evaluate the effect of each drug combination on MAPK signaling, ERK1/2 phosphorylation was evaluated by Western blot analysis. In both of the MM cell lines tested, dual inhibition of EGFR/MET, MET/SFK, and EGFR/SFK was more efficient at inhibiting ERK1/2 phosphorylation than mono-TKI treatments (Fig. 5C). Dual inhibition of EGFR and SFK (EP) was the most effective treatment for inhibition of ERK1/2 phosphorylation in both cell lines, similar to what was observed with regard to AKT phosphorylation (Fig. 5C).

A third downstream effector pathway regulated by EGFR, MET, and SFK found hyperactivated in MM specimens is the mTOR pathway. mTOR signaling has recently been linked to MM tumor cell survival and chemoresistance.^13-15 To evaluate the effect of the drug combination treatments on mTOR signaling, S6 Ribosomal Protein (S6RP) phosphorylation was evaluated by immunoblotting. Dual inhibition of MET/SFK and EGFR/SFK was more efficient at inhibiting S6RP phosphorylation than any other treatment (Fig. 5C). Consistent with AKT and ERK1/2 phosphorylation, dual inhibition of EGFR/SFK (EP) was the most effective treatment to inhibit mTOR signaling. Collectively, these data suggest that dual inhibition of EGFR and SFK can effectively inhibit PI3K-AKT, MAPK, and mTOR signaling in MM cells.

Discussion

Role of STAT1 activation in cancer, including MM

STAT1 is an important mediator of the interferon gamma–induced transcriptional program.³⁰ Upon cytokine receptor activation, STAT1 is phosphorylated and homo- or heterodimerizes before translocating to the nucleus, where it regulates the expression of genes that contribute to cellular arrest, apoptosis, and inflammation.³⁰ STAT1 is directly phosphorylated by JAK1 and SRC, both identified through our screen as being activated in MM cells, as well as by JAK2.^30,33 Interestingly, JAK1 phosphorylation did not directly correlate with STAT1 phosphorylation in the panel of MM cell lines examined, suggesting that other upstream regulators such as SRC are also involved (Fig. 2C). Based on its ability to negatively regulate cell proliferation and/or survival, STAT1 is considered to be a tumor suppressor. STAT3 and STAT5, on the other hand, are considered to be oncoproteins because they positively regulate cell proliferation and survival upon activation.³⁰

Recently, however, STAT1 activation has been shown to promote radioresistance and tumorigenesis.^32,34-36 Inhibiting STAT1 in both renal cell and squamous cell carcinomas was shown to increase radiosensitization.^34,35 STAT1 was also identified as an important tumor promoter in a mouse model of leukemia.³⁶ Moreover, both STAT1 and STAT3 were recently reported to be required for inflammation-associated gastric tumorigenesis in a mouse model of this disease.³² Thus, in certain cancers, STAT1 activation can promote tumor formation and resistance to therapies, which contrasts with its putative role as a tumor suppressor.

Our data indicate that STAT1 tyrosine phosphorylation is a common finding in MM cell lines (Fig. 2A). Furthermore, IHC staining for nuclear P-STAT1 was found in MM tumors but was not detected in normal mesothelial lining (Fig. 2C). Together, these data suggest that STAT1 activation may contribute to MM tumorigenesis. Interestingly, activation of STAT1-survivin signaling has been shown to be associated with a poor prognosis in MM patients.³⁷ Asbestos, the major cause of MM, has been demonstrated to cause inflammation through the Nalp3 inflammasome in mice.³⁸ The role of inflammation in MM progression is still unclear. However, a case could be made that if inflammation drives MM development, then STAT1 signaling may be required for the development of MM, similar to its role in inflammation-associated gastric carcinogenesis.³²

SFK are hyperactivated in MM

SFK, such as SRC and FYN, are nonreceptor tyrosine kinases that govern many cellular processes, including proliferation and survival.^39-42 SFK can regulate both the PI3K-AKT-mTOR and MAPK pathways in various cell types.^43,44 Not surprisingly, SFK have been found to be dysregulated in many human solid tumors, including breast, gastrointestinal, lung, and ovarian carcinomas.^{40,42,43,45-48} SFK activation has also been observed in leukemias, lymphomas, and myelomas.^43,49 SFK inhibitors have been developed, and at least 3 such drugs are currently in phase I clinical trials.^43,50,51 Having identified 2 SFK as being hyperactivated in human MM cell lines, SRC and FYN, we hypothesize that these kinases may contribute to MM cell survival and proliferation, potentially via signaling through the PI3K-AKT-mTOR and MAPK pathways. Germane to this, a recent study demonstrated that the dual SRC/BCR-ABL inhibitor dasatinib caused cytotoxic effects in vitro against MM cell lines.⁵²

We found that SRC and FYN are frequently phosphorylated in MM. SFK activation was observed in more than 50% of the MM cell lines evaluated. Interestingly, SFK were found to be phosphorylated in MM cells that also exhibit hyperactivation of EGFR and MET. EGFR and MET can positively regulate SFK in many human cancer cells and often require SFK signaling to promote cell proliferation and survival.^{40,42,46,48,53} Conversely, SFK can positively regulate EGFR and MET signaling to promote cancer cell proliferation and survival.^25,54

EGFR and MET are concomitantly activated in MM

EGFR and MET are 2 receptor tyrosine kinases that have been found to be overexpressed, mutated, and hyperactivated in many human cancers. Consequently, pharmacological inhibitors and antagonistic antibodies against each receptor have been explored as potential therapies for many tumor types.^55,56 EGFR and MET are independently hyperactivated in MM cells and tumors.^{28,29,55,57,58} The ligands for each receptor, EGF and HGF, respectively, are elevated in cells and fluids derived from pleural lavages of MM patients.^59,60 EGFR and MET can signal independently to redundant pathways in MM cells, including the PI3K-AKT-mTOR and MAPK pathways. A recent independent investigation has also found coactivation of EGFR and MET in MM and demonstrated that dual inhibition of these kinases was more efficacious at inhibiting cell viability and proliferation than TKI-monotherapies directed against individual kinases.⁶¹

Coactivation of EGFR and MET has also been observed in other human tumor types. In lung adenocarcinoma cells, EGFR and MET can be concomitantly activated, with MET signaling allowing for resistance to EGFR inhibitors erlotinib and gefitinib by providing a redundant survival signal to the PI3K-AKT and MAPK pathways.^62,63 This result is consistent with the notion that MET coactivation may allow for MM cell survival by signaling to redundant EGFR regulated pathways, such as the PI3K-AKT-mTOR and MAPK pathways.

In addition to their redundant roles contributing to tumor cell survival, EGFR and MET have been shown to positively regulate the activity of one another in a variety of cell lines.^64-67 In human glioma cells, HGF-MET signaling stimulated EGFR phosphorylation in a transcription-dependent manner.⁶⁷ Conversely, EGFR signaling allowed for constitutive MET phosphorylation in human hepatoma cell lines, human epidermoid carcinoma cell lines, and mammary epithelial cells.⁶⁴ It is not clear from our study whether EGFR activation is required for MET signaling or if MET activation contributes to EGFR signaling. Determining the codependence of EGFR and MET signaling and evaluating the efficacy of combinatorial targeting of both receptor tyrosine kinases for the treatment of MM will be the focus of future studies.

Concomitant activation of tyrosine kinases in human cancers, including MM

Recent studies investigating the efficacy of TKIs in the treatment of glioblastoma demonstrated that more than one tyrosine kinase is often hyperactive in these tumors and that these kinases can signal downstream to redundant effector pathways that contribute to tumor cell survival.²⁷ Importantly, concomitant activation of tyrosine kinases can allow tumor cells to acquire resistance to TKI monotherapies.²⁷ We found that multiple tyrosine kinases, EGFR, MET, and SFK, are concomitantly activated in many MM cells lines (Figs. 3 and 4). These kinases have also been reported to be coactivated in other human cancer types.^{25-27,41,46,53,54,65,68} EGFR, MET and SFK can positively regulate one another and can signal downstream to the same pathways, including the PI3K-AKT-mTOR and MAPK signaling cascades.^{25-27,40,41,46,48,65,68,69} EGFR and SRC have been shown to activate MET to allow serum-independent growth in human bladder carcinoma cells.⁴¹ MET and SFK were recently shown to compensate for loss of EGFR activation in breast cancer cells, providing redundant survival signaling and allowing tumor cells to escape targeted therapy against EGFR.⁵⁴ Furthermore, MET was found to be activated in glioblastoma cells with hyperactive EGFRvIII receptor, and coactivation of MET by EGFRvIII was found to be important for tumor cell viability.⁶⁵ Dual MET/EGFR combinatorial inhibition was shown to decrease erlotinib-resistant lung cancer cell survival both in vitro and in vivo.⁶² Thus, EGFR, MET, and SFK represent tyrosine kinases that are often concomitantly activated in human cancers and can signal redundantly to regulate tumor cell survival and resistance to single TKI therapies.

Using a phosphotyrosine proteomic screen, we were able to identify multiple tyrosine kinases aberrantly activated in MM cells and tumor specimens. Because these various tyrosine kinases are coactivated in MM cells, we hypothesize that TKI monotherapies will be ineffective in the treatment of MM. Consistent with this notion, single-agent erlotinib therapy had no efficacy in a phase II clinical trial with MM patients.²⁰ We had previously proposed that resistance to erlotinib in MM patients may be linked to other mechanisms that allow for continual signaling through the PI3K-AKT-mTOR pathway in the presence of erlotinib.²⁰ Data presented here demonstrate that inhibition of multiple tyrosine kinases is more effective at inhibiting cell viability and downstream effector signaling pathways in MM cell lines. Collectively, these data suggest that combinatorial TKI therapy might represent a promising approach for the treatment of MM.

Materials and Methods

Cell culture

Human MM cell lines were established from surgically explanted primary tumors as described previously⁷⁰ and grown in RPMI 1640 with 10% fetal bovine serum (FBS), supplemented with L-glutamine and penicillin/streptomycin. Nontransformed mesothelial cell culture LP9 (AG# 7086—referred to as 7086 herein) and derivative cell line LP9-hTERT were grown in M199/MCDB106 + 15% fetal calf serum (FCS) + 0.4 mg/mL hydrocortisone + 10 ng/mL EGF. The 7086 and LP9-hTERT cells used for Western blot analysis were at passage 6 or earlier.

Western blot analysis and co-immunoprecipitation

MM cells were trypsinized and harvested by centrifugation, washed in phosphate-buffered saline (PBS), and transferred to lysis buffer (50 mM Tris-Cl [pH 8.0], 150 mM NaCl, 1% Triton X-100 containing proteinase inhibitor cocktail and phosphatase inhibitor cocktail II; Sigma Aldrich, St. Louis, MO). The cell lysates were clarified by centrifugation at 14,000 g for 10 minutes, and the protein concentrations of the whole-cell extracts were determined using a Bio-Rad protein assay kit (Bio-Rad, Hercules, CA). Then, 50 µg of whole-cell extract for each cell line was separated on a 4% to 12% Bis-Tris NuPAGE gel and transferred to nitrocellulose for immunoblotting. All NuPAGE products and gels were purchased from Invitrogen (Carlsbad, CA). Primary antibodies were used for Western blot analysis at a dilution of 1:1000, with incubation overnight. Anti-MET (C12), anti-EGFR (1005), anti-p130Cas (M-72), anti-cortactin (A-4), anti-p-Tyr (PY20), anti-β-actin (I-19), anti-P-ERK (E4), and anti-c-SRC (B-12) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); anti-phospho (P)–SRC (Y416) family, anti-P-JAK1 (Y1022/1023), anti-JAK1, anti-STAT1, anti-EGFR (Y845), anti-P-Tyr-100, anti-P-AKT (S473), anti-P-S6RP (S235/236), anti-S6RP, anti-ERK1/2, and anti-Fyn were purchased from Cell Signaling Technology (Beverly, MA); anti-P-MET (pYpYpY^{1230/1234/1235}) was purchased from Invitrogen, and anti-P-STAT1 (Y701) was a gift from G. Rall (Fox Chase Cancer Center). For co-immunoprecipitation of phosphotyrosine proteins from MM cell lysates, agarose-conjugated anti-P-Tyr (PY20) was employed (Santa Cruz Biotechnology).

Immunohistochemistry

An MM tissue microarray (TMA) was prepared by the Histopathology Facility at Fox Chase Cancer Center in accordance with guidelines of our Institutional Review Board (protocol 00-812). The TMA included 15 MM specimens spotted in duplicate along with 4 normal lung tissue samples and 2 normal kidney tissue samples used as controls. TMAs were deparaffinized and rehydrated prior to antibody staining. Anti-P-STAT1 (Y701—dilution 1:50), anti-P-EGFR (Y845—dilution 1:10), anti-P-MET (pYpYpY^{1230/1234/1235}—dilution 1:50), and anti-P-SRC family (Y416—dilution 1:50) were incubated overnight with TMAs after antigen retrieval with 10 mM sodium citrate buffer (pH 6.0). Anti-P-Tyr-100 (dilution 1:100) was incubated overnight after antigen retrieval with 1 mM EDTA (pH 8.0). TMAs were counterstained with hematoxylin. All images were taken on a Nikon Eclipse E600 microscope equipped with a Nikon Digital Camera DXM1200 at a magnification of 40x. All images were acquired using Nikon ACT-1 version 2 software.

Scaled-up phosphotyrosine immunoprecipitation

Of total protein lysate, 20 mg from human MM cell line Meso 8 was used to immunoprecipitate phosphotyrosine-containing proteins or nonspecific proteins (control IgG). Agarose-conjugated anti-phosphotyrosine (PY20) antibody and control mouse IgG was allowed to incubate with the lysate overnight while rotating at 4°C. Immunoprecipitated proteins were centrifuged and washed 6 times with lysis buffer before being reduced and denatured in NuPAGE LDS Sample Buffer (Invitrogen). Immunoprecipitated proteins were separated on a 4% to 12% NuPAGE Bis-Tris Gel and subsequently stained for 30 minutes with Brilliant Coomassie G Blue stain (Sigma Aldrich). The gel was washed in distilled water overnight while rotating at room temperature.

Mass spectrometry analysis

Gel bands containing proteins were digested with trypsin and analyzed by nano LC/MS/MS on a QSTAR XL mass spectrometer (Applied Biosystems/MDS Sciex, Foster City, CA) as previously described in detail.⁷¹ For discovery of more proteins and peptides in each gel slice and to overcome the possibility of false negatives due to undersampling of coeluting peptides, an exclusion list composed of the peptides sequenced in the first LC/MS/MS run was assembled and used to direct another LC/MS/MS run to sequence only new peptides in the same sample. The two peak lists were combined for database searching for protein identification. The mass spectrum was processed by Analyst QS and then submitted to SwissProt protein database, release 54.1, using MASCOT 2.2 (Matrix Sciences, London, UK). Fixed modification of carbamidomethylcysteine, variable oxidation of methionine, and one trypsin miss were allowed for protein identification in LC/MS/MS. The protein identifications in Table 1 required more than one peptide for each protein. False discovery rate due to coincidence in the database was less than 3.5% for individual peptides as judged by hits at a decoy database containing randomized sequences in each entry.

MTT cell viability assay

Meso 8 or Meso 43 MM cells were each seeded on a 96-well plate, with 5000 cells per well. After 24 hours, various drug concentrations and combinations were added to individual wells. MTT reagent was added to the plate 72 hours posttreatment, and absorbance was determined at 490 nm. EGFR inhibitor AG1478, MET Kinase Inhibitor, and SFK inhibitor PP2 were purchased from Calbiochem (EMD, Darmstadt, Germany). P values were determined using the Wilcoxon test (one-sided); P values of ≤0.05 were considered statistically significant.

Footnotes

The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

The following Fox Chase Cancer Center shared facilities were used in this study: Biotechnology, Cell Culture, Histopathology, and Biostatistics & Bioinformatics Facilities.

Grant support: This work was supported by the Mesothelioma Applied Research Foundation, National Cancer Institute (grant numbers CA-114047, CA-009035, and CA-06927), Local No. 14 Mesothelioma Fund of the International Association of Heat and Frost Insulators & Allied Workers in memory of Hank Vaughan and Alice Haas, Pew Charitable Trust, Kresge Foundation, and Pennsylvania Department of Health. The Department of Health disclaims responsibility for any analyses, interpretations, or conclusions.

References

Bridda

Padoan

Mencarelli

Frego

. Peritoneal mesothelioma: a review. MedGenMed 2007;9:32.

Ellis

Davies

Evans

Haynes

Lloyd

. The use of chemotherapy in patients with advanced malignant pleural mesothelioma: a systematic review and practice guideline. J Thorac Oncol 2006;1:591-601.

Janne

Wozniak

Belani

. Pemetrexed alone or in combination with cisplatin in previously treated malignant pleural mesothelioma: outcomes from a phase IIIB expanded access program. J Thorac Oncol 2006;1:506-12.

Ismail-Khan

Robinson

Williams

Jr Garrett

Bepler

Simon

. Malignant pleural mesothelioma: a comprehensive review. Cancer Control 2006;13:255-63.

Bianchi

Mitsunaga

Cheng

. High frequency of inactivating mutations in the neurofibromatosis type 2 gene (NF2) in primary malignant mesotheliomas. Proc Natl Acad Sci U S A 1995;92:10854-8.

Cheng

Jhanwar

Klein

. p16 alterations and deletion mapping of 9p21-p22 in malignant mesothelioma. Cancer Res 1994;54:5547-51.

Hirao

Bueno

Chen

Gordon

Heilig

Kelsey

. Alterations of the p16(INK4) locus in human malignant mesothelial tumors. Carcinogenesis 2002;23:1127-30.

Pei

Kruger

Testa

. High-resolution analysis of 9p loss in human cancer cells using single nucleotide polymorphism-based mapping arrays. Cancer Genet Cytogenet 2006;170:65-8.

Sekido

Pass

Bader

. Neurofibromatosis type 2 (NF2) gene is somatically mutated in mesothelioma but not in lung cancer. Cancer Res 1995;55:1227-31.

10.

Altomare

Vaslet

Skele

. A mouse model recapitulating molecular features of human mesothelioma. Cancer Res 2005;65:8090-5.

11.

Jongsma

van Montfort

Vooijs

. A conditional mouse model for malignant mesothelioma. Cancer Cell 2008;13:261-71.

12.

Altomare

Menges

Pei

. Activated TNF-α/NF-κB signaling via down-regulation of Fas-associated factor 1 in asbestos-induced mesotheliomas from Arf knockout mice. Proc Natl Acad Sci U S A 2009;106:3420-5.

13.

Altomare

You

Xiao

. Human and mouse mesotheliomas exhibit elevated AKT/PKB activity, which can be targeted pharmacologically to inhibit tumor cell growth. Oncogene 2005;24:6080-9.

14.

Barbone

Yang

Morgan

Gaudino

Broaddus

. Mammalian target of rapamycin contributes to the acquired apoptotic resistance of human mesothelioma multicellular spheroids. J Biol Chem 2008;283:13021-30.

15.

Wilson

Barbone

Yang

. mTOR mediates survival signals in malignant mesothelioma grown as tumor fragment spheroids. Am J Respir Cell Mol Biol 2008;39:576-83.

16.

Pache

Janssen

Walsh

. Increased epidermal growth factor-receptor protein in a human mesothelial cell line in response to long asbestos fibers. Am J Pathol 1998;152:333-40.

17.

Zanella

Posada

Tritton

Mossman

. Asbestos causes stimulation of the extracellular signal-regulated kinase 1 mitogen-activated protein kinase cascade after phosphorylation of the epidermal growth factor receptor. Cancer Res 1996;56:5334-8.

18.

de Melo

Gerbase

Curran

Pache

. Phosphorylated extracellular signal-regulated kinases are significantly increased in malignant mesothelioma. J Histochem Cytochem 2006;54:855-61.

19.

Frey

Wali

Pass

Lonardo

. Osteopontin is linked to p65 and MMP-9 expression in pulmonary adenocarcinoma but not in malignant pleural mesothelioma. Histopathology 2007;50:720-6.

20.

Garland

Rankin

Gandara

. Phase II study of erlotinib in patients with malignant pleural mesothelioma: a Southwest Oncology Group Study. J Clin Oncol 2007;25:2406-13.

21.

Vintman

Nielsen

Berner

Reich

Davidson

. Mitogen-activated protein kinase expression and activation does not differentiate benign from malignant mesothelial cells. Cancer 2005;103:2427-33.

22.

Blume-Jensen

Hunter

. Oncogenic kinase signalling. Nature 2001;411:355-65.

23.

Traxler

. Tyrosine kinases as targets in cancer therapy: successes and failures. Expert Opin Ther Targets 2003;7:215-34.

24.

Dierck

Machida

Voigt

. Quantitative multiplexed profiling of cellular signaling networks using phosphotyrosine-specific DNA-tagged SH2 domains. Nat Methods 2006;3:737-44.

25.

Emaduddin

Bicknell

Bodmer

Feller

. Cell growth, global phosphotyrosine elevation, and c-Met phosphorylation through Src family kinases in colorectal cancer cells. Proc Natl Acad Sci U S A 2008;105:2358-62.

26.

Rikova

Guo

Zeng

. Global survey of phosphotyrosine signaling identifies oncogenic kinases in lung cancer. Cell 2007;131:1190-203.

27.

Stommel

Kimmelman

Ying

. Coactivation of receptor tyrosine kinases affects the response of tumor cells to targeted therapies. Science 2007;318:287-90.

28.

Jagadeeswaran

Seiwert

. Functional analysis of c-Met/hepatocyte growth factor pathway in malignant pleural mesothelioma. Cancer Res 2006;66:352-61.

29.

Okuda

Sasaki

Kawano

. Epidermal growth factor receptor gene mutation, amplification and protein expression in malignant pleural mesothelioma. J Cancer Res Clin Oncol 2008;134:1105-11.

30.

Calo

Migliavacca

Bazan

. STAT proteins: from normal control of cellular events to tumorigenesis. J Cell Physiol 2003;197:157-68.

31.

Bianchi

. Malignant mesothelioma: global incidence and relationship with asbestos. Ind Health 2007;45:379-87.

32.

Ernst

Najdovska

Grail

. STAT3 and STAT1 mediate IL-11-dependent and inflammation-associated gastric tumorigenesis in gp130 receptor mutant mice. J Clin Invest 2008;118:1727-38.

33.

Cirri

Chiarugi

Marra

. c-Src activates both STAT1 and STAT3 in PDGF-stimulated NIH3T3 cells. Biochem Biophys Res Commun 1997;239:493-7.

34.

Hui

Tretiakova

Zhang

. Radiosensitization by inhibiting STAT1 in renal cell carcinoma. Int J Radiat Oncol Biol Phys 2009;73:288-95.

35.

Khodarev

Minn

Efimova

. Signal transducer and activator of transcription 1 regulates both cytotoxic and prosurvival functions in tumor cells. Cancer Res 2007;67:9214-20.

36.

Kovacic

Stoiber

Moriggl

. STAT1 acts as a tumor promoter for leukemia development. Cancer Cell 2006;10:77-87.

37.

Kothmaier

Quehenberger

Halbwedl

. EGFR and PDGFR differentially promote growth in malignant epithelioid mesothelioma of short and long term survivors. Thorax 2008;63:345-51.

38.

Dostert

Petrilli

Van Bruggen

Steele

Mossman

Tschopp

. Innate immune activation through Nalp3 inflammasome sensing of asbestos and silica. Science 2008;320:674-7.

39.

Roskoski

Jr . Src kinase regulation by phosphorylation and dephosphorylation. Biochem Biophys Res Commun 2005;331:1-14.

40.

Rahimi

Hung

Tremblay

Saulnier

Elliott

. c-Src kinase activity is required for hepatocyte growth factor-induced motility and anchorage-independent growth of mammary carcinoma cells. J Biol Chem 1998;273:33714-21.

41.

Yamamoto

Mammadova

Song

Fukami

Sato

. Tyrosine phosphorylation of p145met mediated by EGFR and Src is required for serum-independent survival of human bladder carcinoma cells. J Cell Sci 2006;119:4623-33.

42.

Zhang

Kalyankrishna

Wislez

. SRC-family kinases are activated in non-small cell lung cancer and promote the survival of epidermal growth factor receptor-dependent cell lines. Am J Pathol 2007;170:366-76.

43.

Alvarez

Kantarjian

Cortes

. The role of Src in solid and hematologic malignancies: development of new-generation Src inhibitors. Cancer 2006;107:1918-29.

44.

Johnson

Gallick

. SRC family nonreceptor tyrosine kinases as molecular targets for cancer therapy. Anticancer Agents Med Chem 2007;7:651-9.

45.

Han

Landen

Trevino

. Antiangiogenic and antitumor effects of SRC inhibition in ovarian carcinoma. Cancer Res 2006;66:8633-9.

46.

Herynk

Zhang

Parikh

Gallick

. Activation of Src by c-Met overexpression mediates metastatic properties of colorectal carcinoma cells. J Exp Ther Oncol 2007;6:205-17.

47.

Hiscox

Morgan

Green

Barrow

Gee

Nicholson

. Elevated Src activity promotes cellular invasion and motility in tamoxifen resistant breast cancer cells. Breast Cancer Res Treat 2006;97:263-74.

48.

Mao

Irby

Coppola

. Activation of c-Src by receptor tyrosine kinases in human colon cancer cells with high metastatic potential. Oncogene 1997;15:3083-90.

49.

Summy

Gallick

. Treatment for advanced tumors: SRC reclaims center stage. Clin Cancer Res 2006;12:1398-401.

50.

Cortes

Rousselot

Kim

. Dasatinib induces complete hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in blast crisis. Blood 2007;109:3207-13.

51.

Quintas-Cardama

Kantarjian

Jones

. Dasatinib (BMS-354825) is active in Philadelphia chromosome-positive chronic myelogenous leukemia after imatinib and nilotinib (AMN107) therapy failure. Blood 2007;109:497-9.

52.

Tsao

Saigal

. Inhibition of c-Src expression and activation in malignant pleural mesothelioma tissues leads to apoptosis, cell cycle arrest, and decreased migration and invasion. Mol Cancer Ther 2007;6:1962-72.

53.

Yeh

Cheng

. EGFR mutants found in non-small cell lung cancer show different levels of sensitivity to suppression of Src: implications in targeting therapy. Oncogene 2008;27:957-65.

54.

Mueller

Hunter

Ethier

Boerner

. Met and c-Src cooperate to compensate for loss of epidermal growth factor receptor kinase activity in breast cancer cells. Cancer Res 2008;68:3314-22.

55.

Toschi

Janne

. Single-agent and combination therapeutic strategies to inhibit hepatocyte growth factor/MET signaling in cancer. Clin Cancer Res 2008;14:5941-6.

56.

Zhang

Berezov

Wang

. ErbB receptors: from oncogenes to targeted cancer therapies. J Clin Invest 2007;117:2051-8.

57.

Janne

Taffaro

Salgia

Johnson

. Inhibition of epidermal growth factor receptor signaling in malignant pleural mesothelioma. Cancer Res 2002;62:5242-7.

58.

Mukohara

Civiello

Davis

. Inhibition of the met receptor in mesothelioma. Clin Cancer Res 2005;11:8122-30.

59.

Adamson

Bakowska

. KGF and HGF are growth factors for mesothelial cells in pleural lavage fluid after intratracheal asbestos. Exp Lung Res 2001;27:605-16.

60.

Cesario

Catassi

Festi

. Farnesyltransferase inhibitors and human malignant pleural mesothelioma: a first-step comparative translational study. Clin Cancer Res 2005;11:2026-37.

61.

Kawaguchi

Murakami

Taniguchi

. Combined inhibition of MET and EGFR suppresses proliferation of malignant mesothelioma cells. Carcinogenesis 2009;30:1097-105.

62.

Tang

Jiang

. Dual MET-EGFR combinatorial inhibition against T790M-EGFR-mediated erlotinib-resistant lung cancer. Br J Cancer 2008;99:911-22.

63.

Yano

Wang

. Hepatocyte growth factor induces gefitinib resistance of lung adenocarcinoma with epidermal growth factor receptor-activating mutations. Cancer Res 2008;68:9479-87.

64.

Bonine-Summers

Aakre

Brown

. Epidermal growth factor receptor plays a significant role in hepatocyte growth factor mediated biological responses in mammary epithelial cells. Cancer Biol Ther 2007;6:561-70.

65.

Huang

Mukasa

Bonavia

. Quantitative analysis of EGFRvIII cellular signaling networks reveals a combinatorial therapeutic strategy for glioblastoma. Proc Natl Acad Sci U S A 2007;104:12867-72.

66.

Stolz

Esplen

Dorko

Michalopoulos

Strom

. Cross-talk between epidermal growth factor receptor and c-Met signal pathways in transformed cells. J Biol Chem 2000;275:8806-11.

67.

Reznik

Sang

. Transcription-dependent epidermal growth factor receptor activation by hepatocyte growth factor. Mol Cancer Res 2008;6:139-50.

68.

Guo

Villen

Kornhauser

. Signaling networks assembled by oncogenic EGFR and c-Met. Proc Natl Acad Sci U S A 2008;105:692-7.

69.

Popsueva

Poteryaev

Arighi

. GDNF promotes tubulogenesis of GFRalpha1-expressing MDCK cells by Src-mediated phosphorylation of Met receptor tyrosine kinase. J Cell Biol 2003;161:119-29.

70.

Taguchi

Jhanwar

Siegfried

Keller

Testa

. Recurrent deletions of specific chromosomal sites in 1p, 3p, 6q, and 9p in human malignant mesothelioma. Cancer Res 1993;53:4349-55.

71.

Patel

Liu

. Proteomic analyses of pancreatic cyst fluids. Pancreas 2009;38:e33-42.