Quantitative proteomics of breast tumors: Tissue quality assessment to clinical biomarkers

Liquid chromatography-selected reaction monitoring mass spectrometry (LC-SRM) is not only a proven tool for clinical chemistry, but also a versatile method to enhance the capability to quantify biomarkers for tumor biology research. As the treatment of cancer continues to evolve, the ability to assess multiple biomarkers to assign cancer phenotypes based on the genetic background and the signaling of the individual tumor becomes paramount to our ability to treat the patient. In breast cancer, the American Society of Clinical Oncology has defined biomarkers for patient assessment to guide selection of therapy: estrogen receptor, progesterone receptor, and the HER2/Neu receptor tyrosine kinase; therefore, these proteins were selected for LC-SRM assay development. Detailed molecular characterization of these proteins is necessary for patient treatment, so expression and phosphorylation assays have been developed and applied. In addition, other LC-SRM assays were developed to further evaluate tumor biology (e.g. Ki-67 for proliferation and vimentin for tumor aggressiveness related to the epithelial-to-mesenchymal transition). These measurements combined with biomarkers for tissue quality and histological content are implemented in a three-tier multiplexed assay platform, which is translated from cell line models into frozen tumor tissues banked from breast cancer patients.

Understanding the Characteristics of Mass Spectrometry Data through the use of Simulation

Background

Mass spectrometry is actively being used to discover disease-related proteomic patterns in complex mixtures of proteins derived from tissue samples or from easily obtained biological fluids. The potential importance of these clinical applications has made the development of better methods for processing and analyzing the data an active area of research. It is, however, difficult to determine which methods are better without knowing the true biochemical composition of the samples used in the experiments.

Methods

We developed a mathematical model based on the physics of a simple MALDI-TOF mass spectrometer with time-lag focusing. Using this model, we implemented a statistical simulation of mass spectra. We used the simulation to explore some of the basic operating characteristics of MALDI or SELDI instruments.

Results

The simulation reproduced several characteristics of actual instruments. We found that the relative mass error is affected by the time discretization of the detector (about 0.01%) and the spread of initial velocities (about 0.1%). The accuracy of calibration based on external standards decays rapidly outside the range spanned by the calibrants. Natural isotope distributions play a major role in broadening peaks associated with individual proteins. The area of a peak is a more accurate measure of its size than the height.

Conclusions

The model described here is capable of simulating realistic mass spectra. The simulation should become a useful tool for generating spectra where the true inputs are known, allowing researchers to evaluate the performance of new methods for processing and analyzing mass spectra.

Availability

http://bioinformatics.mdanderson.org/cromwell.html

Parametric Power Spectral Density Analysis of Noise from Instrumentation in MALDI TOF Mass Spectrometry

Noise in mass spectrometry can interfere with identification of the biochemical substances in the sample. For example, the electric motors and circuits inside the mass spectrometer or in nearby equipment generate random noise that may distort the true shape of mass spectra. This paper presents a stochastic signal processing approach to analyzing noise from electrical noise sources (i.e., noise from instrumentation) in MALDI TOF mass spectrometry. Noise from instrumentation was hypothesized to be a mixture of thermal noise, 1/f noise, and electric or magnetic interference in the instrument. Parametric power spectral density estimation was conducted to derive the power distribution of noise from instrumentation with respect to frequencies. As expected, the experimental results show that noise from instrumentation contains 1/f noise and prominent periodic components in addition to thermal noise. These periodic components imply that the mass spectrometers used in this study may not be completely shielded from the internal or external electrical noise sources. However, according to a simulation study of human plasma mass spectra, noise from instrumentation does not seem to affect mass spectra significantly. In conclusion, analysis of noise from instrumentation using stochastic signal processing here provides an intuitive perspective on how to quantify noise in mass spectrometry through spectral modeling.

Abstract 1565: OnPLS-based integrative proteogenomics analysis of lung squamous cell cancer

Introduction: Multivariate projection methods such as PCA and PLS has been widely applied for analysis of biological and chemical data. OnPLS is a recent extension to these methods suitable for integrative analysis of omics data. With OnPLS it is possible to compare multiple omics datasets to identify joint variation and variation locally unique for each of the studied datasets. OnPLS is a new approach for truly integrative analysis of omics data to be contrasted to commonly applied approaches limiting analysis to 1) comparing findings from individually analyzed blocks of data 2) pairwise comparison of individual probes.

Experimental: A Java based implementation of OnPLS was used for the statistical modeling. 116 lung squamous cell cancer samples were characterized using gene expression profiling and global proteomics. The OnPLS model was applied to jointly model variation between mRNA and protein expression. Enrichment analysis of factor loadings was performed using the Enrichr tools to identify biological mechanisms explained by the different joint and unique components of the OnPLS model.

Results: Using a cross-validation procedure the model with the highest predictive ability was calculated having two joint components and one locally unique component for each of the proteomics and gene expression datasets. The model explained 21.9% of the variation in the expression data and 26.1% of the variation in the proteomics data. The first joint component captures the highest degree of common variation between mRNA and protein activity. From the mRNA data, this component is related to immune infiltrates, especially monocytes and B-cells, whereas this component is related to extracellular matrix activity from the protein data. This suggests covariance of mRNA immune-related gene expression and extracellular matrix-related protein expression. As expected, local variation specific to the protein measurements involved regulation of protein activation and processing. mRNA-specific variation is related to keratinization, a key process in squamous cell cancer.

Conclusion: OnPLS offers an interesting approach for integrative analysis of omics data. Applying this approach to proteo-genomics data of lung squamous cell cancers suggest similar patterns of activity is represented in protein and gene expression data, however the biological processes associated with this activity may be distinct.

Citation Format: Fredrik Pettersson, Paul A. Stewart, Robbert J. Slebos, Eric A. Welsh, Ling Cen, Yonghong Zhang, Zhihua Chen, Chia-Ho Cheng, Guolin Zhang, Bin Fang, Victoria Izumi, Sean Yoder, Katherine Fellows, Y Ann Chen, Jamie K. Teer, Steven Eschrich, John M. Koomen, Anders Berglund, Eric B. Haura. OnPLS-based integrative proteogenomics analysis of lung squamous cell cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1565. doi:10.1158/1538-7445.AM2017-1565

Abstract 205: Underlying mechanisms of genome-proteome discordance in squamous cell lung cancer

Introduction: Genomic analyses have yielded a tremendous amount of data on the genetic changes in lung cancers, but translating these experiments into actionable information benefitting lung squamous cell carcinoma (SQLC) patients has proven more difficult. Studies by the NCI Clinical Proteomic Tumor Analysis Consortium (CPTAC), our group, and others have demonstrated that gene and protein expression show only moderate correlation, demonstrating limitations in explaining phenotypic changes from genomics alone. These findings indicate a clear need for integrative proteogenomics to better understand tumor biology, especially in a complex disease like SQLC.

Experimental: We have assembled a comprehensive proteogenomic dataset including DNA copy number (Affymetrix CytoScan HD Assay), targeted exome sequencing (Agilent Comprehensive Cancer Panel), RNA-sequencing (Illumina NextSeq), and shotgun proteomics (Q Exactive LC-MS/MS) on 116 surgically resected SQLC tumor samples with extensive clinical and follow up data.

Results: We have identified 6584 high confidence proteins from preliminary proteomic analysis. After quality control filtering, we utilized 5562 gene-protein pairs for further analysis. Clustering of patient RNA expression in this patient cohort has been unable to fully reproduce the molecular classification previously published for SQLC. Furthermore, proteomic results indicate yet another potential classification strategy selecting patient subgroups that differ at protein level. We observed a 0.29 median Spearman’s correlation of 5562 gene-protein pairs. There were 2781 highly correlated gene-protein pairs (greater than median) and 2781 poorly correlated gene-protein pairs (less than median) including 773 anti-correlated gene-protein pairs (less than 0). We hypothesized that poorly correlated gene-protein pairs could be functionally related in a pathway-dependent manner. Enrichment analysis of poorly correlated proteins identified pathways related to mRNA processing, growth factor signaling (EGFR, FGFR), and nonsense-mediated decay (NMD). Interestingly, there were 9 frequently mutated SQLC genes in the low correlation gene-protein pairs but only 3 in the highly correlated pairs. We found three distinct patient subgroups by clustering poorly correlated proteins. Analysis of these subgroups showed differentially expressed pathways related to mRNA processing, ubiquitination, and NMD.

Conclusion: Differential modulation of the proteome outside of genomic regulation may suggest important regulatory mechanisms in cancer and give new insights into treating SQLC. Analysis of poorly correlated gene-protein pairs suggests certain pathways are dysregulated in cancer, and ongoing DNA analysis and future analyses involving miRNAs, RNA-binding proteins, and the ubiquitin proteome system will help elucidate our preliminary findings.

Citation Format: Paul A. Stewart, Robbert J. Slebos, Eric A. Welsh, Ling Cen, Yonghong Zhang, Zhihua Chen, Chia-Ho Cheng, Fredrik Pettersson, Anders Berglund, Guolin Zhang, Bin Fang, Victoria Izumi, Sean Yoder, Katherine Fellows, Ann Chen, Jamie K. Teer, Steven A. Eschrich, John M. Koomen, Eric B. Haura. Underlying mechanisms of genome-proteome discordance in squamous cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 205. doi:10.1158/1538-7445.AM2017-205

Abstract 221: Integrated functional proteomics of MET/VEGFR inhibitors reveals complex mechanism of action of foretinib in NSCLC

BACKGROUND: Foretinib (FORE) and cabozantinib (CABO) are two MET/VEGFR inhibitors with similar chemical structures. CABO is FDA-approved for medullary thyroid and renal cancer; in addition, it is in clinical trials for treatment of non-small cell lung cancer (NSCLC). Through an unbiased viability screen we have observed potent cellular activity of FORE, but not CABO, in several NSCLC cell lines. We have previously shown that most NSCLC cell lines are insensitive to MET or VEGFR inhibition, suggesting off-target activity of FORE in these cells. The aim of this project is to identify the mechanism of action of FORE in NSCLC and design an optimized combination therapy.

METHODS: Cellular viability assays were done using CellTiter-Glo, cell cycle analysis by flow cytometry. Western blotting was performed to evaluate the induction of apoptosis through PARP1 and caspase cleavage, as well changes in signaling. We synthesized FORE and CABO analogues and performed differential quantitative chemical and phosphoproteomics to determine the target kinase profile and pathway effects in NSCLC cells. Changes in gene expression upon drug treatment were measured by RNA-seq. RNAi in combination with pharmacological inhibitors was performed to interrogate targets and pathways.

RESULTS: FORE showed greater potency in NSCLC cell lines than CABO with regard to inhibition of viability and induction of apoptosis. FORE decreased phosphorylation of AKT and ERK. Chemical and phosphoproteomics revealed several kinases, such as MEK and MAP4K5, to bind preferentially by FORE over CABO that differentially affect the adherens junction and MAPK signaling pathways. Target validation showed differential inhibition of MEK1/2 and MAP4K5. Cellular validation with RNAi in combination with pharmacological inhibitors suggested that MEK1/2, MAP4K5 and IGF1R are involved in the mechanism of action of foretinib in NSCLC cells. RNA-seq pathway analysis furthermore suggested regulation of chromatin organization and Wnt pathway signaling by foretinib.

CONCLUSION: Our results suggest that the difference in the efficacy between FORE and CABO is related to polypharmacology of FORE, which simultaneously targets IGF1R, MEK1/2 and MAP4K5. This difference results in divergence in signaling pathway inhibition and induces distinct effects in NSCLC. The establishment of FORE targets and signaling pathways can lead to optimized combination therapy for NSCLC and identification of new actionable kinases in lung cancer cells.

Citation Format: Natalia J. Sumi, Bin Fang, Lily L. Rix, Muhammad Ayaz, Fumi Kinose, Eric A. Welsh, Steven A. Eschrich, Harshani R. Lawrence, John M. Koomen, Eric B. Haura, Uwe Rix. Integrated functional proteomics of MET/VEGFR inhibitors reveals complex mechanism of action of foretinib in NSCLC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 221. doi:10.1158/1538-7445.AM2017-221

Unification of de novo and acquired ibrutinib resistance in mantle cell lymphoma

The novel Bruton's tyrosine kinase inhibitor ibrutinib has demonstrated high response rates in B-cell lymphomas; however, a growing number of ibrutinib-treated patients relapse with resistance and fulminant progression. Using chemical proteomics and an organotypic cell-based drug screening assay, we determine the functional role of the tumour microenvironment (TME) in ibrutinib activity and acquired ibrutinib resistance. We demonstrate that MCL cells develop ibrutinib resistance through evolutionary processes driven by dynamic feedback between MCL cells and TME, leading to kinome adaptive reprogramming, bypassing the effect of ibrutinib and reciprocal activation of PI3K-AKT-mTOR and integrin-β1 signalling. Combinatorial disruption of B-cell receptor signalling and PI3K-AKT-mTOR axis leads to release of MCL cells from TME, reversal of drug resistance and enhanced anti-MCL activity in MCL patient samples and patient-derived xenograft models. This study unifies TME-mediated de novo and acquired drug resistance mechanisms and provides a novel combination therapeutic strategy against MCL and other B-cell malignancies.

Ibrutinib has demonstrated high response rates in B-cell lymphomas but a lot of ibrutinib-treated patients relapse with resistance. This study unified TME-mediated de novo and acquired drug resistance through B-cell receptor signalling and PI3K-AKT-mTOR axis and provides a combination therapeutic strategy against B-cell malignancies.

Activity-Based Protein Profiling Shows Heterogeneous Signaling Adaptations to BRAF Inhibition

Patients with BRAF V600E mutant melanoma are typically treated with targeted BRAF kinase inhibitors, such as vemurafenib and dabrafenib. Although these drugs are initially effective, they are not curative. Most of the focus to date has been upon genetic mechanisms of acquired resistance; therefore, we must better understand the global signaling adaptations that mediate escape from BRAF inhibition. In the current study, we have used activity-based protein profiling (ABPP) with ATP-analogue probes to enrich kinases and other enzyme classes that contribute to BRAF inhibitor (BRAFi) resistance in four paired isogenic BRAFi-naïve/resistant cell line models. Our analysis showed these cell line models, which also differ in their PTEN status, have considerable heterogeneity in their kinase ATP probe uptake in comparing both naïve cells and adaptations to chronic drug exposure. A number of kinases including FAK1, SLK, and TAOK2 had increased ATP probe uptake in BRAFi resistant cells, while KHS1 (M4K5) and BRAF had decreased ATP probe uptake in the BRAFi-resistant cells. Gene ontology (GO) enrichment analysis revealed BRAFi resistance is associated with a significant enhancement in ATP probe uptake in proteins implicated in cytoskeletal organization and adhesion, and decreases in ATP probe uptake in proteins associated with cell metabolic processes. The ABPP approach was able to identify key phenotypic mediators critical for each BRAFi resistant cell line. Together, these data show that common phenotypic adaptations to BRAF inhibition can be mediated through very different signaling networks, suggesting considerable redundancy within the signaling of BRAF mutant melanoma cells.

Proteome-wide Profiling of Clinical PARP Inhibitors Reveals Compound-Specific Secondary Targets

Poly(ADP-ribose) polymerase (PARP) inhibitors (PARPi) are a promising class of targeted cancer drugs, but their individual target profiles beyond the PARP family, which could result in differential clinical use or toxicity, are unknown. Using an unbiased, mass spectrometry-based chemical proteomics approach, we generated a comparative proteome-wide target map of the four clinical PARPi, olaparib, veliparib, niraparib, and rucaparib. PARPi as a class displayed high target selectivity. However, in addition to the canonical targets PARP1, PARP2, and several of their binding partners, we also identified hexose-6-phosphate dehydrogenase (H6PD) and deoxycytidine kinase (DCK) as previously unrecognized targets of rucaparib and niraparib, respectively. Subsequent functional validation suggested that inhibition of DCK by niraparib could have detrimental effects when combined with nucleoside analog pro-drugs. H6PD silencing can cause apoptosis and further sensitize cells to PARPi, suggesting that H6PD may be, in addition to its established role in metabolic disorders, a new anticancer target.

APOSTL: An Interactive Galaxy Pipeline for Reproducible Analysis of Affinity Proteomics Data

With continuously increasing scale and depth of coverage in affinity proteomics (AP-MS) data, the analysis and visualization is becoming more challenging. A number of tools have been developed to identify high-confidence interactions; however, a cohesive and intuitive pipeline for analysis and visualization is still needed. Here we present Automated Processing of SAINT Templated Layouts (APOSTL), a freely available Galaxy-integrated software suite and analysis pipeline for reproducible, interactive analysis of AP-MS data. APOSTL contains a number of tools woven together using Galaxy workflows, which are intuitive for the user to move from raw data to publication-quality figures within a single interface. APOSTL is an evolving software project with the potential to customize individual analyses with additional Galaxy tools and widgets using the R web application framework, Shiny. The source code, data, and documentation are freely available from GitHub ( https://github.com/bornea/APOSTL ) and other sources.

Phosphoproteomics Reveals MAPK Inhibitors Enhance MET- and EGFR-Driven AKT Signaling in KRAS-Mutant Lung Cancer

Pathway inhibition of the RAS-driven MAPK pathway using small-molecule kinase inhibitors has been a key focus for treating cancers driven by oncogenic RAS, yet significant clinical responses are lacking. Feedback reactivation of ERK driven by drug-induced RAF activity has been suggested as one of the major drug resistance mechanisms, especially in the context of oncogenic RAS. To determine whether additional adaptive resistance mechanisms may coexist, we characterized global phosphoproteomic changes after MEK inhibitor selumetinib (AZD6244) treatment in KRAS-mutant A427 and A549 lung adenocarcinoma cell lines employing mass spectrometry-based phosphoproteomics. We identified 9,075 quantifiable unique phosphosites (corresponding to 3,346 unique phosphoproteins), of which 567 phosphosites were more abundant and 512 phosphosites were less abundant after MEK inhibition. Selumetinib increased phosphorylation of KSR-1, a scaffolding protein required for assembly of MAPK signaling complex, as well as altered phosphorylation of GEF-H1, a novel regulator of KSR-1 and implicated in RAS-driven MAPK activation. Moreover, selumetinib reduced inhibitory serine phosphorylation of MET at Ser985 and potentiated HGF- and EGF-induced AKT phosphorylation. These results were recapitulated by pan-RAF (LY3009120), MEK (GDC0623), and ERK (SCH772984) inhibitors, which are currently under early-phase clinical development against RAS-mutant cancers. Our results highlight the unique adaptive changes in MAPK scaffolding proteins (KSR-1, GEF-H1) and in RTK signaling, leading to enhanced PI3K-AKT signaling when the MAPK pathway is inhibited.

IMPLICATIONS

This study highlights the unique adaptive changes in MAPK scaffolding proteins (KSR-1, GEF-H1) and in RTK signaling, leading to enhanced PI3K/AKT signaling when the MAPK pathway is inhibited. Mol Cancer Res; 14(10); 1019-29. ©2016 AACR.

Abstract 3876: EphrinB3 and EphrinB4 receptors: potential therapeutic targets in glioblastoma

Introduction: Glioblastoma (GBM) is the most common primary malignant neoplasm of the central nervous system in adults with poor median overall survival despite advances in tumor. CD133 is a known putative cancer stem cell marker, and we aimed to identify markers in CD133+ GBM tumor initiating cell lines (CSCs) with an infiltrative phenotype that could serve as therapeutic targets.

Methods: Expression (mRNA) microarray datasets including 22,278 probes for known cell-surface markers in three CD133+ CSCs and 17 normal brain tissue lines were obtained. We identified genes with uniformly high mRNA expression, filtered for known localization at the cell-surface and for non- or low expression in normal brain. Ephrin B3 receptor (EphB3), ephrin B4 receptor (EphB4) and fibroblast growth factor receptor (FGFR1) were highly expressed. Protein expression was established by mass spectrometry using CD133+ cell line extracts and then assessed in 27 patient GBM (IDH-wildtype) tumor samples by immunohistochemistry. Expression was graded based of the percentage of cells positive and intensity of staining the tumor, interface and uninvolved brain. One-way ANOVA was used to test any group difference among any combination of two genes; Tukey honest significant difference method was used to adjust for p value for pairwise comparison in tumor locations. Spearman correlation analysis was applied to determine potential correlation co-expression EphB3 and EphB4. The Gamma statistic was used to identify correlation of intensity of staining with tumor location, and the Fisher's exact test was utilized to elucidate potential significance.

Results: High expression (> 50% of cells stained) was 7/27 for EphB3, 8/27 for EphB4, and 18/27 for FGFR1. EphB3 and EphB4 are significantly associated with tissue location based on percentage of cells stained in each location (p < 0.05). When compared to the uninvolved brain, EphB3 and EphB4 expression in the tumor and interface tissues were significantly higher (p<0.05). EphB3 expression was significantly correlated with EphB4 expression (p < 0.05) in tumor tissue, though the correlation was weak (r = 0.414) EphB3 and EphB4 intensity were positively associated with tissue location: 0.53 [0.21 - 0.86] and EphB4 0.45 [0.09 - 0.80] respectively. FGFR-1 was not associated with tumor location in terms of both percentage and intensity of staining.

Conclusion: In terms of percentage and intensity of staining at the uninvolved brain, interface and normal tissue, EphB3 and EphB4 were significantly associated with those three tissue locations. EphB3 and EphB4 were highly expressed in tumor and interface tissue relative to the normal tissue. Co-expression of EphB3 and EphB4 in tumor tissue was correlated. Evaluation with clinical parameters may disclose subsets of these tumors with varying infiltrative potential.

Citation Format: Radhika Sreeraman Kumar, Robert J. B. Macaulay, Hannah C. Rutherford, Natalie Barkey, Jiannong Li, Jongphil Kim, John M. Koomen, David L. Morse. EphrinB3 and EphrinB4 receptors: potential therapeutic targets in glioblastoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3876.

PAXIP1 Potentiates the Combination of WEE1 Inhibitor AZD1775 and Platinum Agents in Lung Cancer

The DNA damage response (DDR) involves a complex network of signaling events mediated by modular protein domains such as the BRCA1 C-terminal (BRCT) domain. Thus, proteins that interact with BRCT domains and are a part of the DDR constitute potential targets for sensitization to DNA-damaging chemotherapy agents. We performed a pharmacologic screen to evaluate 17 kinases, identified in a BRCT-mediated interaction network as targets to enhance platinum-based chemotherapy in lung cancer. Inhibition of mitotic kinase WEE1 was found to have the most effective response in combination with platinum compounds in lung cancer cell lines. In the BRCT-mediated interaction network, WEE1 was found in complex with PAXIP1, a protein containing six BRCT domains involved in transcription and in the cellular response to DNA damage. We show that PAXIP1 BRCT domains regulate WEE1-mediated phosphorylation of CDK1. Furthermore, ectopic expression of PAXIP1 promotes enhanced caspase-3-mediated apoptosis in cells treated with WEE1 inhibitor AZD1775 (formerly, MK-1775) and cisplatin compared with cells treated with AZD1775 alone. Cell lines and patient-derived xenograft models expressing both PAXIP1 and WEE1 exhibited synergistic effects of AZD1775 and cisplatin. In summary, PAXIP1 is involved in sensitizing lung cancer cells to the WEE1 inhibitor AZD1775 in combination with platinum-based treatment. We propose that WEE1 and PAXIP1 levels may be used as mechanism-based biomarkers of response when WEE1 inhibitor AZD1775 is combined with DNA-damaging agents. Mol Cancer Ther; 15(7); 1669-81. ©2016 AACR.

Inhibition of PI3K/Akt/mTOR signaling in PI3KR2-overexpressing colon cancer stem cells reduces tumor growth due to apoptosis

In sporadic colon cancer, colon cancer stem cells (CCSCs) initiate tumorigenesis and may contribute to late disease recurrences and metastases. We previously showed that aldehyde dehydrogenase (ALDH) activity (as indicated by the ALDEFLUOR^® assay) is an effective marker for highly enriching CCSCs for further evaluation. Here, we used comparative transcriptome and proteome approaches to identify signaling pathways overrepresented in the CCSC population. We found overexpression of several components of the phosphoinositide 3-kinase (PI3K)/Akt/mechanistic target of rapamycin (mTOR) signaling pathway, including PI3KR2, a regulatory subunit of PI3K. LY294002, a PI3K inhibitor, defined the contribution of the PI3K/Akt/mTOR signaling pathway in CCSCs. LY294002-treated CCSCs showed decreases in proliferation, sphere formation and self-renewal, in phosphorylation-dependent activation of Akt, and in expression of cyclin D1. Inhibition of PI3K in vivo reduced tumorigenicity, increased detection of cleaved caspase 3, an indicator of apoptosis, and elevated expression of the inflammatory chemokine, CXCL8. Collectively, these results indicate that PI3K/Akt/mTOR signaling controls CCSC proliferation and CCSC survival, and suggests that it would be useful to develop therapeutic agents that target this signaling pathway.

Tyrosine Kinase Signaling in Clear Cell and Papillary Renal Cell Carcinoma Revealed by Mass Spectrometry-Based Phosphotyrosine Proteomics

PURPOSE

Targeted therapies in renal cell carcinoma (RCC) are limited by acquired resistance. Novel therapeutic targets are needed to combat resistance and, ideally, target the unique biology of RCC subtypes.

EXPERIMENTAL DESIGN

Tyrosine kinases provide critical oncogenic signaling and their inhibition has significantly impacted cancer care. To describe a landscape of tyrosine kinase activity in RCC that could inform novel therapeutic strategies, we performed a mass spectrometry-based system-wide survey of tyrosine phosphorylation in 10 RCC cell lines as well as 15 clear cell and 15 papillary RCC human tumors. To prioritize identified tyrosine kinases for further analysis, a 63 tyrosine kinase inhibitor (TKI) drug screen was performed.

RESULTS

Among the cell lines, 28 unique tyrosine phosphosites were identified across 19 kinases and phosphatases including EGFR, MET, JAK2, and FAK in nearly all samples. Multiple FAK TKIs decreased cell viability by at least 50% and inhibited RCC cell line adhesion, invasion, and proliferation. Among the tumors, 49 unique tyrosine phosphosites were identified across 44 kinases and phosphatases. FAK pY576/7 was found in all tumors and many cell lines, whereas DDR1 pY792/6 was preferentially enriched in the papillary RCC tumors. Both tyrosine kinases are capable of transmitting signals from the extracellular matrix and emerged as novel RCC therapeutic targets.

CONCLUSIONS

Tyrosine kinase profiling informs novel therapeutic strategies in RCC and highlights the unique biology among kidney cancer subtypes. Clin Cancer Res; 22(22); 5605-16. ©2016 AACR.

Activity-Based Proteomics Reveals Heterogeneous Kinome and ATP-Binding Proteome Responses to MEK Inhibition in KRAS Mutant Lung Cancer

One way cancer cells can escape from targeted agents is through their ability to evade drug effects by rapidly rewiring signaling networks. Many protein classes, such as kinases and metabolic enzymes, are regulated by ATP binding and hydrolysis. We hypothesized that a system-level profiling of drug-induced alterations in ATP-binding proteomes could offer novel insights into adaptive responses. Here, we mapped global ATP-binding proteomes perturbed by two clinical MEK inhibitors, AZD6244 and MEK162, in KRAS mutant lung cancer cells as a model system harnessing a desthiobiotin-ATP probe coupled with LC-MS/MS. We observed strikingly unique ATP-binding proteome responses to MEK inhibition, which revealed heterogeneous drug-induced pathway signatures in each cell line. We also identified diverse kinome responses, indicating each cell adapts to MEK inhibition in unique ways. Despite the heterogeneity of kinome responses, decreased probe labeling of mitotic kinases and an increase of kinases linked to autophagy were identified to be common responses. Taken together, our study revealed a diversity of adaptive ATP-binding proteome and kinome responses to MEK inhibition in KRAS mutant lung cancer cells, and our study further demonstrated the utility of our approach to identify potential candidates of targetable ATP-binding enzymes involved in adaptive resistance and to develop rational drug combinations.

Fibronectin induction abrogates the BRAF inhibitor response of BRAF V600E/PTEN-null melanoma cells

The mechanisms by which some melanoma cells adapt to Serine/threonine-protein kinase B-Raf (BRAF) inhibitor therapy are incompletely understood. In the present study, we used mass spectrometry-based phosphoproteomics to determine how BRAF inhibition remodeled the signaling network of melanoma cell lines that were BRAF mutant and PTEN null. Short-term BRAF inhibition was associated with marked changes in fibronectin-based adhesion signaling that were PTEN dependent. These effects were recapitulated through BRAF siRNA knockdown and following treatment with chemotherapeutic drugs. Increased fibronectin expression was also observed in mouse xenograft models as well as specimens from melanoma patients undergoing BRAF inhibitor treatment. Analysis of a melanoma tissue microarray showed loss of PTEN expression to predict for a lower overall survival, with a trend for even lower survival being seen when loss of fibronectin was included in the analysis. Mechanistically, the induction of fibronectin limited the responses of these PTEN-null melanoma cell lines to vemurafenib, with enhanced cytotoxicity observed following the knockdown of either fibronectin or its receptor α5β1 integrin. This in turn abrogated the cytotoxic response to BRAF inhibition via increased AKT signaling, which prevented the induction of cell death by maintaining the expression of the pro-survival protein Mcl-1. The protection conveyed by the induction of FN expression could be overcome through combined treatment with a BRAF and PI3K inhibitor.

Inhibition of Shp2 suppresses mutant EGFR-induced lung tumors in transgenic mouse model of lung adenocarcinoma

Epidermal growth factor receptor (EGFR) mutants drive lung tumorigenesis and are targeted for therapy. However, resistance to EGFR inhibitors has been observed, in which the mutant EGFR remains active. Thus, it is important to uncover mediators of EGFR mutant-driven lung tumors to develop new treatment strategies. The protein tyrosine phosphatase (PTP) Shp2 mediates EGF signaling. Nevertheless, it is unclear if Shp2 is activated by oncogenic EGFR mutants in lung carcinoma or if inhibiting the Shp2 PTP activity can suppress EGFR mutant-induced lung adenocarcinoma. Here, we generated transgenic mice containing a doxycycline (Dox)-inducible PTP-defective Shp2 mutant (tetO-Shp2CSDA). Using the rat Clara cell secretory protein (CCSP)-rtTA-directed transgene expression in the type II lung pneumocytes of transgenic mice, we found that the Gab1-Shp2 pathway was activated by EGFRL858R in the lungs of transgenic mice. Consistently, the Gab1-Shp2 pathway was activated in human lung adenocarcinoma cells containing mutant EGFR. Importantly, Shp2CSDA inhibited EGFRL858R-induced lung adenocarcinoma in transgenic animals. Analysis of lung tissues showed that Shp2CSDA suppressed Gab1 tyrosine phosphorylation and Gab1-Shp2 association, suggesting that Shp2 modulates a positive feedback loop to regulate its own activity. These results show that inhibition of the Shp2 PTP activity impairs mutant EGFR signaling and suppresses EGFRL858R-driven lung adenocarcinoma.

Target Identification in Small Cell Lung Cancer via Integrated Phenotypic Screening and Activity-Based Protein Profiling

To overcome hurdles in identifying key kinases in small cell lung cancer (SCLC), we integrated a target-agnostic phenotypic screen of kinase inhibitors with target identification using activity-based protein profiling (ABPP) in which a desthiobiotin-ATP probe was used. We screened 21 SCLC cell lines with known c-MYC amplification status for alterations in viability using a chemical library of 235 small-molecule kinase inhibitors. One screen hit compound was interrogated with ABPP, and, through this approach, we reidentified Aurora kinase B as a critical kinase in MYC-amplified SCLC cells. We next extended the platform to a second compound that had activity in SCLC cell lines lacking c-MYC amplification and identified TANK-binding kinase 1, a kinase that affects cell viability, polo-like kinase-1 signaling, G2-M arrest, and apoptosis in SCLC cells lacking MYC amplification. These results demonstrate that phenotypic screening combined with ABPP can identify key disease drivers, suggesting that this approach, which combines new chemical probes and disease cell screens, has the potential to identify other important targets in other cancer types. Mol Cancer Ther; 15(2); 334-42. ©2016 AACR.

Abstract LB-C12: Functional proteomics elucidates signaling adaptation driven by combination therapy in BRAF mutant melanoma cell line models

Targeted kinase inhibition is a promising treatment modality for melanoma, which has significantly increased available clinical strategies and improved survival outcomes for patients, whose tumors harbor the BRAF V600E mutation. However, responses are transient and multiple mechanisms for drug resistance can contribute to therapeutic escape. Therefore, a novel multiplexed proteomics approach provides an optimal choice for comprehensively dissecting multiple mechanisms of resistance to BRAFi and combination therapy within one experiment.

A large-scale quantitative expression proteomics and phosphoproteomics analysis (SysQuant) was carried out on BRAF V600E mutant melanoma cell lines treated with two different clinically relevant kinase inhibitor combinations: 1) BRAFi/MEKi and 2) BRAFi/PI3Ki. Changes in protein expression and phosphorylation in response to each treatment are determined by comparison to vehicle controls. Cell line models all harbor BRAF V600E mutations, but differ in PTEN status (A375 is WT, while WM793 is PTEN null). Briefly, cells were treated with either drug combination (as above) and harvested at 1hr, 6hrs, 24hrs and 48hrs post-treatment. Control cells were treated with DMSO. Samples were lysed, reduced, alkylated and digested with trypsin. Tryptic peptides from each sample were chemically labeled or “barcoded” with TMT-10plex reagents (4 BRAFi/MEKi time points, 4 BRAFi/PI3Ki time points, 1 DMSO control and a pooled reference sample) and combined for LC-MS/MS. After peptide fractionation with strong cation exchange chromatography and phosphopeptide enrichment, LC-MS/MS peptide sequencing and relative quantification was performed using an Orbitrap Fusion mass spectrometer (Thermo). Raw MS data was searched by Proteome Discoverer and analyzed by in-house R-scripts and Perseus statistical software package. Pathway analysis was done in GeneGO (Metacore).

The SysQuant workflow identified >9,000 protein groups and >17,000 unique phosphosites per cell line across different treatments. Principal component analysis of the phosphoproteomics data revealed signaling differences across different treatment conditions and drug combinations are mainly driven by BRAFi and the time post-treatment. K-means clustering was also used to examine trends in the data; as an example, this technique could be used to track signaling changes that correlate with reduction and recovery in ERK signaling. The SysQuant approach provides a systems view of global signaling changes occurring in response to drug treatment. The ability to multiplex samples with TMT allows quantitative deduction of protein expression and phosphorylation patterns that are common or unique in different cell lines, time post-treatment and the effect of combination treatments.

This proteomics approach ties protein expression and phosphorylation status in response to combination therapy and generates several hypotheses for further testing with the goal of developing novel combination therapy strategies.

Citation Format: Ritin Sharma, Inna Fedorenko, Sasa Koncarevic, Vikram Mitra, Stefan Selzer, Gitte Boehm, Ian Pike, Keiran Smalley, John M. Koomen. Functional proteomics elucidates signaling adaptation driven by combination therapy in BRAF mutant melanoma cell line models. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr LB-C12.

A Pilot Proteogenomic Study with Data Integration Identifies MCT1 and GLUT1 as Prognostic Markers in Lung Adenocarcinoma

We performed a pilot proteogenomic study to compare lung adenocarcinoma to lung squamous cell carcinoma using quantitative proteomics (6-plex TMT) combined with a customized Affymetrix GeneChip. Using MaxQuant software, we identified 51,001 unique peptides that mapped to 7,241 unique proteins and from these identified 6,373 genes with matching protein expression for further analysis. We found a minor correlation between gene expression and protein expression; both datasets were able to independently recapitulate known differences between the adenocarcinoma and squamous cell carcinoma subtypes. We found 565 proteins and 629 genes to be differentially expressed between adenocarcinoma and squamous cell carcinoma, with 113 of these consistently differentially expressed at both the gene and protein levels. We then compared our results to published adenocarcinoma versus squamous cell carcinoma proteomic data that we also processed with MaxQuant. We selected two proteins consistently overexpressed in squamous cell carcinoma in all studies, MCT1 (SLC16A1) and GLUT1 (SLC2A1), for further investigation. We found differential expression of these same proteins at the gene level in our study as well as in other public gene expression datasets. These findings combined with survival analysis of public datasets suggest that MCT1 and GLUT1 may be potential prognostic markers in adenocarcinoma and druggable targets in squamous cell carcinoma. Data are available via ProteomeXchange with identifier PXD002622.

Abstract 668: Phosphoproteomic characterization of selumetinib action in KRAS mutant lung cancer

In the field of lung cancer therapeutics, KRAS-mutant lung cancer is a significant problem due to limited therapeutic options and poor prognosis. Pathway inhibition of the RAS-driven RAF-MEK-ERK pathway using small-molecule kinase inhibitors has been a key focus for treating cancers driven by oncogenic RAS; however, significant clinical responses are lacking. Our initial experiments assessed the efficacy of the clinical MEK inhibitor AZD6244 (selumetinib) in a panel of non-small cell lung cancer cell lines harboring oncogenic KRAS mutations. None of the cell lines showed pronounced apoptosis nor did we observe significant reduction in cell viability in nano-molar range (<1 μM). Given that cancer cells can rapidly rewire signaling cascades to adapt to loss of survival signaling, we hypothesized that a phosphoproteomics approach would provide further insight into adaptive signaling alteration associated with AZD6244. To this aim, we assessed global phosphoproteome change after AZD6244 treatment (1 μM, 24hr) in A427 and A549 lung adenocarcinoma cell lines harboring oncogenic KRAS mutations. We identified 7,730 unique phosphopeptides (corresponding to 3,137 unique phosphoproteins) after removing low-confidence phosphopeptides. We defined phosphopeptide significantly altered by AZD6244 as one that has had quantity changed by two-fold, corresponding to one standard deviation of fold change within the data. This resulted in 882 increased and 831 decreased phosphopeptides after MEK inhibition. As expected, AZD6244 decreased ERK phosphorylation and induced feedback increase of MEK and B- and C-RAF phosphorylation. In addition, we observed that AZD6244 increased tyrosine phosphorylation (Y1172) and decreased serine phosphorylation (S1045) of EGFR, which are involved in kinase activation and receptor trafficking, respectively. We also found reduced inhibitory serine phosphorylation of MET (S985), suggesting MEK inhibition could lead to MET activation by relieving inhibitory serine phosphorylation. These observations raise the possibility that MEK inhibition leads to feedback activation of RTKs, as well as its downstream activation (e.g., wild-type RAS). Notably, AZD6244 increased phosphorylation of KSR-1, a scaffolding protein required for assembly of MAPK signaling complex as well as phosphorylation of GEF-H1, a recently reportedly novel regulator of KSR-1 required for proper activation of RAS-driven MAPK activation. Collectively, our phosphoproteomics results indicate altered phosphorylation of a series of signaling proteins in MAPK signaling cascade, presumably contributing feedback activation of RAF, MEK and ERK reactivation following MEK inhibition. We are currently characterizing the importance of GEF-H1 and KSR-1 in lung cancer cell growth and survival and their role in adaptive signaling change associated with MEK inhibition. Updated work will be presented.

Citation Format: Jae-Young Kim, Eric A. Welsh, Bin Fang, Fumi Kinose, John M. Koomen, Eric B. Haura. Phosphoproteomic characterization of selumetinib action in KRAS mutant lung cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 668. doi:10.1158/1538-7445.AM2015-668

Abstract 2006: Kinases and adaptive signaling contribute to drug resistance in BRAF mutant melanoma

Melanoma, the most lethal form of skin cancer, is marked by numerous genetic modifications, including point mutations as well as overexpression and deletion of genes. Intrinsic and acquired resistance to BRAF V600E targeted therapies (BRAFi) in metastatic melanoma patients further underscores the need for global profiling of melanoma circuitry at the functional level. Therefore, activity-based protein profiling (ABPP) and phosphoproteomics was carried out to decipher steady state differences in global signaling mechanisms in naïve and BRAFi resistant melanoma cell lines with BRAF V600E mutations.

Four cell lines (A375, 1205Lu, WM164 and WM793) were selected to evaluate different molecular backgrounds of BRAF mutation based on their PTEN status (either WT or null). For each cell type, both the naïve and BRAFi resistant lines were analyzed via ABPP as well as chemical labeling with tandem mass tags (TMT) prior to discovery phosphoproteomics. While the ABPP approach mined for kinases, phosphoproteomics identified STY phosphorylated peptides providing information on signaling via kinase substrates. LC-MS/MS discovery proteomics (RSLC and Q Exactive, Thermo) identified and relatively quantified all peptides observed in ABPP and TMT phosphoproteomics experiments. MaxQuant was used for data evaluation; preliminary statistical analyses were performed in Perseus to select significant differences for pathway mapping (GeneGO, Metacore) and follow up experiments using siRNA or pharmacological inhibition. Adaptive responses to combination treatment were also explored using the SysQuant workflow for quantitative expression analysis and phosphoproteomics. These experiments served as a basis for comparison for a pilot project of 12 metastatic tumors from BRAF mutant melanoma patients selected for comparison of good and poor survival outcomes.

ABPP measurements on different cell line models (A375, 1205Lu, WM793 and WM164) reveal significant differences in ATP uptake of proteins in the resistant cell line model compared to its naïve counterpart. For example, in the 1205Lu cell line several proteins including EGFR, p38alpha, DNA-PK formed an interconnected pathway. Overall we identified between 2,000-2,800 proteins in each cell line with ∼150 kinases. Isobaric labeling coupled to phosphoproteomics identified ∼1,600 quantifiable proteins with ∼4,000 phosphorylation sites. Phosphoproteomics revealed concomitant increase in phosphorylation levels of the substrates acted upon by kinases showing higher ATP uptake in ABPP measurements. For example, in 1205Lu BRAFi resistant cells, CDK1, CDK2 and DNA-PK showed higher ATP uptake and their substrates SSK1, DPYSL3, and vimentin showed higher phosphorylation levels.

The complementary nature of the two functional proteomics approaches provided holistic overview of signalling network in melanoma and enabled selection of targets for follow-up studies.

Citation Format: Ritin Sharma, Manali Phadke, David Britton, Ian Pike, Keiran Smalley, John M. Koomen. Kinases and adaptive signaling contribute to drug resistance in BRAF mutant melanoma. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2006. doi:10.1158/1538-7445.AM2015-2006