Candidate biomarkers for treatment benefit from sunitinib in patients with advanced renal cell carcinoma using mass spectrometry-based (phospho)proteomics.

The tyrosine kinase inhibitor sunitinib is an effective first-line treatment for patients with advanced renal cell carcinoma (RCC). Hypothesizing that a functional read-out by mass spectrometry-based (phospho, p-)proteomics will identify predictive biomarkers for treatment outcome of sunitinib, tumor tissues of 26 RCC patients were analyzed. Eight patients had primary resistant (RES) and 18 sensitive (SENS) RCC. A 78 phosphosite signature (p < 0.05, fold-change > 2) was identified; 22 p-sites were upregulated in RES (unique in RES: BCAR3, NOP58, EIF4A2, GDI1) and 56 in SENS (35 unique). EIF4A1/EIF4A2 were differentially expressed in RES at the (p-)proteome and, in an independent cohort, transcriptome level. Inferred kinase activity of MAPK3 (p = 0.026) and EGFR (p = 0.045) as determined by INKA was higher in SENS. Posttranslational modifications signature enrichment analysis showed that different p-site-centric signatures were enriched (p < 0.05), of which FGF1 and prolactin pathways in RES and, in SENS, vanadate and thrombin treatment pathways, were most significant. In conclusion, the RCC (phospho)proteome revealed differential p-sites and kinase activities associated with sunitinib resistance and sensitivity. Independent validation is warranted to develop an assay for upfront identification of patients who are intrinsically resistant to sunitinib.

Phosphotyrosine-based Phosphoproteomics for Target Identification and Drug Response Prediction in AML Cell Lines.

Acute myeloid leukemia (AML) is a clonal disorder arising from hematopoietic myeloid progenitors. Aberrantly activated tyrosine kinases (TK) are involved in leukemogenesis and are associated with poor treatment outcome. Kinase inhibitor (KI) treatment has shown promise in improving patient outcome in AML. However, inhibitor selection for patients is suboptimal.In a preclinical effort to address KI selection, we analyzed a panel of 16 AML cell lines using phosphotyrosine (pY) enrichment-based, label-free phosphoproteomics. The Integrative Inferred Kinase Activity (INKA) algorithm was used to identify hyperphosphorylated, active kinases as candidates for KI treatment, and efficacy of selected KIs was tested.Heterogeneous signaling was observed with between 241 and 2764 phosphopeptides detected per cell line. Of 4853 identified phosphopeptides with 4229 phosphosites, 4459 phosphopeptides (4430 pY) were linked to 3605 class I sites (3525 pY). INKA analysis in single cell lines successfully pinpointed driver kinases (PDGFRA, JAK2, KIT and FLT3) corresponding with activating mutations present in these cell lines. Furthermore, potential receptor tyrosine kinase (RTK) drivers, undetected by standard molecular analyses, were identified in four cell lines (FGFR1 in KG-1 and KG-1a, PDGFRA in Kasumi-3, and FLT3 in MM6). These cell lines proved highly sensitive to specific KIs. Six AML cell lines without a clear RTK driver showed evidence of MAPK1/3 activation, indicative of the presence of activating upstream RAS mutations. Importantly, FLT3 phosphorylation was demonstrated in two clinical AML samples with a FLT3 internal tandem duplication (ITD) mutation.Our data show the potential of pY-phosphoproteomics and INKA analysis to provide insight in AML TK signaling and identify hyperactive kinases as potential targets for treatment in AML cell lines. These results warrant future investigation of clinical samples to further our understanding of TK phosphorylation in relation to clinical response in the individual patient.

INKA, an integrative data analysis pipeline for phosphoproteomic inference of active kinases.

Identifying hyperactive kinases in cancer is crucial for individualized treatment with specific inhibitors. Kinase activity can be discerned from global protein phosphorylation profiles obtained with mass spectrometry-based phosphoproteomics. A major challenge is to relate such profiles to specific hyperactive kinases fueling growth/progression of individual tumors. Hitherto, the focus has been on phosphorylation of either kinases or their substrates. Here, we combined label-free kinase-centric and substrate-centric information in an Integrative Inferred Kinase Activity (INKA) analysis. This multipronged, stringent analysis enables ranking of kinase activity and visualization of kinase-substrate networks in a single biological sample. To demonstrate utility, we analyzed (i) cancer cell lines with known oncogenes, (ii) cell lines in a differential setting (wild-type versus mutant, +/- drug), (iii) pre- and on-treatment tumor needle biopsies, (iv) cancer cell panel with available drug sensitivity data, and (v) patient-derived tumor xenografts with INKA-guided drug selection and testing. These analyses show superior performance of INKA over its components and substrate-based single-sample tool KARP, and underscore target potential of high-ranking kinases, encouraging further exploration of INKA's functional and clinical value.

Ferritin heavy chain in triple negative breast cancer: a favorable prognostic marker that relates to a cluster of differentiation 8 positive (CD8+) effector T-cell response.

Ferritin heavy chain (FTH1) is a 21-kDa subunit of the ferritin complex, known for its role in iron metabolism, and which has recently been identified as a favorable prognostic protein for triple negative breast cancer (TNBC) patients. Currently, it is not well understood how FTH1 contributes to an anti-tumor response. Here, we explored whether expression and cellular compartmentalization of FTH1 correlates to an effective immune response in TNBC patients. Analysis of the tumor tissue transcriptome, complemented with in silico pathway analysis, revealed that FTH1 was an integral part of an immunomodulatory network of cytokine signaling, adaptive immunity, and cell death. These findings were confirmed using mass spectrometry (MS)-derived proteomic data, and immunohistochemical staining of tissue microarrays. We observed that FTH1 is localized in both the cytoplasm and/or nucleus of cancer cells. However, high cytoplasmic (c) FTH1 was associated with favorable prognosis (Log-rank p = 0.001), whereas nuclear (n) FTH1 staining was associated with adverse prognosis (Log-rank p = 0.019). cFTH1 staining significantly correlated with total FTH1 expression in TNBC tissue samples, as measured by MS analysis (Rs = 0.473, p = 0.0007), but nFTH1 staining did not (Rs = 0.197, p = 0.1801). Notably, IFN γ-producing CD8+ effector T cells, but not CD4+ T cells, were preferentially enriched in tumors with high expression of cFTH1 (p = 0.02). Collectively, our data provide evidence toward new immune regulatory properties of FTH1 in TNBC, which may facilitate development of novel therapeutic targets.

Phosphoproteomics of patient-derived xenografts identifies targets and markers associated with sensitivity and resistance to EGFR blockade in colorectal cancer.

Epidermal growth factor receptor (EGFR) is a well-exploited therapeutic target in metastatic colorectal cancer (mCRC). Unfortunately, not all patients benefit from current EGFR inhibitors. Mass spectrometry-based proteomics and phosphoproteomics were performed on 30 genomically and pharmacologically characterized mCRC patient-derived xenografts (PDXs) to investigate the molecular basis of response to EGFR blockade and identify alternative drug targets to overcome resistance. Both the tyrosine and global phosphoproteome as well as the proteome harbored distinctive response signatures. We found that increased pathway activity related to mitogen-activated protein kinase (MAPK) inhibition and abundant tyrosine phosphorylation of cell junction proteins, such as CXADR and CLDN1/3, in sensitive tumors, whereas epithelial-mesenchymal transition and increased MAPK and AKT signaling were more prevalent in resistant tumors. Furthermore, the ranking of kinase activities in single samples confirmed the driver activity of ERBB2, EGFR, and MET in cetuximab-resistant tumors. This analysis also revealed high kinase activity of several members of the Src and ephrin kinase family in 2 CRC PDX models with genomically unexplained resistance. Inhibition of these hyperactive kinases, alone or in combination with cetuximab, resulted in growth inhibition of ex vivo PDX-derived organoids and in vivo PDXs. Together, these findings highlight the potential value of phosphoproteomics to improve our understanding of anti-EGFR treatment and response prediction in mCRC and bring to the forefront alternative drug targets in cetuximab-resistant tumors.

Kinase Inhibitor Treatment of Patients with Advanced Cancer Results in High Tumor Drug Concentrations and in Specific Alterations of the Tumor Phosphoproteome.

Identification of predictive biomarkers for targeted therapies requires information on drug exposure at the target site as well as its effect on the signaling context of a tumor. To obtain more insight in the clinical mechanism of action of protein kinase inhibitors (PKIs), we studied tumor drug concentrations of protein kinase inhibitors (PKIs) and their effect on the tyrosine-(pTyr)-phosphoproteome in patients with advanced cancer. Tumor biopsies were obtained from 31 patients with advanced cancer before and after 2 weeks of treatment with sorafenib (SOR), erlotinib (ERL), dasatinib (DAS), vemurafenib (VEM), sunitinib (SUN) or everolimus (EVE). Tumor concentrations were determined by LC-MS/MS. pTyr-phosphoproteomics was performed by pTyr-immunoprecipitation followed by LC-MS/MS. Median tumor concentrations were 2-10 µM for SOR, ERL, DAS, SUN, EVE and >1 mM for VEM. These were 2-178 × higher than median plasma concentrations. Unsupervised hierarchical clustering of pTyr-phosphopeptide intensities revealed patient-specific clustering of pre- and on-treatment profiles. Drug-specific alterations of peptide phosphorylation was demonstrated by marginal overlap of robustly up- and downregulated phosphopeptides. These findings demonstrate that tumor drug concentrations are higher than anticipated and result in drug specific alterations of the phosphoproteome. Further development of phosphoproteomics-based personalized medicine is warranted.

Proteomic Analysis of miR-195 and miR-497 Replacement Reveals Potential Candidates that Increase Sensitivity to Oxaliplatin in MSI/P53wt Colorectal Cancer Cells.

Most patients with advanced colorectal cancer (CRC) eventually develop resistance to systemic combination therapy. miR-195-5p and miR-497-5p are downregulated in CRC tissues and associated with drug resistance. Sensitization to 5-FU, oxaliplatin, and irinotecan by transfection with miR-195-5p and miR-497-5p mimics was studied using cell viability and clonogenic assays in cell lines HCT116, RKO, DLD-1, and SW480. In addition, proteomic analysis of transfected cells was implemented to identify potential targets. Significantly altered proteins were subjected to STRING (protein-protein interaction networks) database analysis to study the potential mechanisms of drug resistance. Cell viability analysis of transfected cells revealed increased sensitivity to oxaliplatin in microsatellite instable (MSI)/P53 wild-type HCT116 and RKO cells. HCT116 transfected cells formed significantly fewer colonies when treated with oxaliplatin. In sensitized cells, proteomic analysis showed 158 and 202 proteins with significantly altered expression after transfection with miR-195-5p and miR-497-5p mimics respectively, of which CHUK and LUZP1 proved to be coinciding downregulated proteins. Resistance mechanisms of these proteins may be associated with nuclear factor kappa-B signaling and G1 cell-cycle arrest. In conclusion, miR-195-5p and miR-497-5p replacement enhanced sensitivity to oxaliplatin in treatment naïve MSI/P53 wild-type CRC cells. Proteomic analysis revealed potential miRNA targets associated with the cell-cycle which possibly bare a relation with chemotherapy sensitivity.

Phosphotyrosine-based-phosphoproteomics scaled-down to biopsy level for analysis of individual tumor biology and treatment selection.

Mass spectrometry-based phosphoproteomics of cancer cell and tissue lysates provides insight in aberrantly activated signaling pathways and potential drug targets. For improved understanding of individual patient's tumor biology and to allow selection of tyrosine kinase inhibitors in individual patients, phosphoproteomics of small clinical samples should be feasible and reproducible. We aimed to scale down a pTyr-phosphopeptide enrichment protocol to biopsy-level protein input and assess reproducibility and applicability to tumor needle biopsies. To this end, phosphopeptide immunoprecipitation using anti-phosphotyrosine beads was performed using 10, 5 and 1mg protein input from lysates of colorectal cancer (CRC) cell line HCT116. Multiple needle biopsies from 7 human CRC resection specimens were analyzed at the 1mg-level. The total number of phosphopeptides captured and detected by LC-MS/MS ranged from 681 at 10mg input to 471 at 1mg HCT116 protein. ID-reproducibility ranged from 60.5% at 10mg to 43.9% at 1mg. Per 1mg-level biopsy sample, >200 phosphopeptides were identified with 57% ID-reproducibility between paired tumor biopsies. Unsupervised analysis clustered biopsies from individual patients together and revealed known and potential therapeutic targets. SIGNIFICANCE: This study demonstrates the feasibility of label-free pTyr-phosphoproteomics at the tumor biopsy level based on reproducible analyses using 1mg of protein input. The considerable number of identified phosphopeptides at this level is attributed to an effective down-scaled immuno-affinity protocol as well as to the application of ID propagation in the data processing and analysis steps. Unsupervised cluster analysis reveals patient-specific profiles. Together, these findings pave the way for clinical trials in which pTyr-phosphoproteomics will be performed on pre- and on-treatment biopsies. Such studies will improve our understanding of individual tumor biology and may enable future pTyr-phosphoproteomics-based personalized medicine.

Peptide-mediated 'miniprep' isolation of extracellular vesicles is suitable for high-throughput proteomics.

Extracellular vesicles (EVs) are cell-secreted membrane vesicles enclosed by a lipid bilayer derived from endosomes or from the plasma membrane. Since EVs are released into body fluids, and their cargo includes tissue-specific and disease-related molecules, they represent a rich source for disease biomarkers. However, standard ultracentrifugation methods for EV isolation are laborious, time-consuming, and require high inputs. Ghosh and co-workers recently described an isolation method utilizing Heat Shock Protein (HSP)-binding peptide Vn96 to aggregate HSP-decorated EVs, which can be performed at small 'miniprep' scale. Based on microscopic, immunoblot, and RNA sequencing analyses this method compared well with ultracentrifugation-mediated EV isolation, but a detailed proteomic comparison was lacking. Therefore, we compared both methods using label-free proteomics of replicate EV isolations from HT-29 cell-conditioned medium. Despite a 30-fold different scale (ultracentrifugation: 60 ml/Vn96-mediated aggregation: 2 ml) both methods yielded comparable numbers of identified proteins (3115/3085), with similar reproducibility of identification (72.5%/75.5%) and spectral count-based quantification (average CV: 31%/27%). EV fractions obtained with either method contained established EV markers and proteins linked to vesicle-related gene ontologies. Thus, Vn96 peptide-mediated aggregation is an advantageous, simple and rapid approach for EV isolation from small biological samples, enabling high-throughput analysis in a biomarker discovery setting.