A PRRX1 Signature Identifies TIM-3 and VISTA as Potential Immune Checkpoint Targets in a Subgroup of Microsatellite Stable Colorectal Cancer Liver Metastases

Disease recurrence and drug resistance are major challenges in the clinical management of patients with colorectal cancer liver metastases (CLM), and because tumors are generally microsatellite stable (MSS), responses to immune therapies are poor. The mesenchymal phenotype is overrepresented in treatment-resistant cancers and is associated with an immunosuppressed microenvironment. The aim of this work was to molecularly identify and characterize a mesenchymal subgroup of MSS CLM to identify novel therapeutic approaches. We here generated a mesenchymal gene expression signature by analysis of resection specimens from 38 patients with CLM using ranked expression level of the epithelial-to-mesenchymal transition-related transcription factor PRRX1. Downstream pathway analysis based on the resulting gene signature was performed and independent, publicly available datasets were used to validate the findings. A subgroup comprising 16% of the analyzed CLM samples were classified as mesenchymal, or belonging to the PRRX1 high group. Analysis of the PRRX1 signature genes revealed a distinct immunosuppressive phenotype with high expression of immune checkpoints HAVCR2/TIM-3 and VISTA, in addition to the M2 macrophage marker CD163. The findings were convincingly validated in datasets from three external CLM cohorts. Upregulation of immune checkpoints HAVCR2/TIM-3 and VISTA in the PRRX1 high subgroup is a novel finding, and suggests immune evasion beyond the PD-1/PD-L1 axis, which may contribute to poor response to PD-1/PD-L1-directed immune therapy in MSS colorectal cancer. Importantly, these checkpoints represent potential novel opportunities for immune-based therapy approaches in a subset of MSS CLM.

Significance

CLM is an important cause of colorectal cancer mortality where the majority of patients have yet to benefit from immunotherapies. In this study of gene expression profiling analyses, we uncovered novel immune checkpoint targets in a subgroup of patients with MSS CLMs harboring a mesenchymal phenotype.

Protein Signature Predicts Response to Neoadjuvant Treatment With Chemotherapy and Bevacizumab in HER2-Negative Breast Cancers

PURPOSE

Antiangiogenic therapy using bevacizumab has proven effective for a number of cancers; however, in breast cancer (BC), there is an unmet need to identify patients who benefit from such treatment.

PATIENTS AND METHODS

In the NeoAva phase II clinical trial, patients (N = 132) with large (≥ 25 mm) human epidermal growth factor receptor 2 (HER2)-negative primary tumors were randomly assigned 1:1 to treatment with neoadjuvant chemotherapy (CTx) alone or in combination with bevacizumab (Bev plus CTx). The ratio of the tumor size after relative to before treatment was calculated into a continuous response scale. Tumor biopsies taken prior to neoadjuvant treatment were analyzed by reverse-phase protein arrays (RPPA) for expression levels of 210 BC-relevant (phospho-) proteins. Lasso regression was used to derive a predictor of tumor shrinkage from the expression of selected proteins prior to treatment.

RESULTS

We identified a nine-protein signature score named vascular endothelial growth factor inhibition response predictor (ViRP) for use in the Bev plus CTx treatment arm able to predict with accuracy pathologic complete response (pCR) (area under the curve [AUC] = 0.85; 95% CI, 0.74 to 0.97) and low residual cancer burden (RCB 0/I) (AUC = 0.80; 95% CI, 0.68 to 0.93). The ViRP score was significantly lower in patients with pCR (P < .001) and in patients with low RCB (P < .001). The ViRP score was internally validated on mRNA data and the resultant surrogate mRNA ViRP score significantly separated the pCR patients (P = .016). Similarly, the mRNA ViRP score was validated (P < .001) in an independent phase II clinical trial (PROMIX).

CONCLUSION

Our ViRP score, integrating the expression of nine proteins and validated on mRNA data both internally and in an independent clinical trial, may be used to increase the likelihood of benefit from treatment with bevacizumab combined with chemotherapy in patients with HER2-negative BC.

Targeting AXL and the DNA Damage Response Pathway as a Novel Therapeutic Strategy in Melanoma

Receptor tyrosine kinase AXL is found upregulated in various types of cancer, including melanoma, and correlates with an aggressive cancer phenotype, inducing cell proliferation and epithelial-to-mesenchymal transition. In addition, AXL has recently been linked to chemotherapy resistance, and inhibition of AXL is found to increase DNA damage and reduce expression of DNA repair proteins. In light of this, we aimed to investigate whether targeting AXL together with DNA damage response proteins would be therapeutically beneficial. Using melanoma cell lines, we observed that combined reduction of AXL and CHK1/CHK2 signaling decreased proliferation, deregulated cell-cycle progression, increased apoptosis, and reduced expression of DNA damage response proteins. Enhanced therapeutic effect of combined treatment, as compared with mono-treatment, was further observed in a patient-derived xenograft model and, of particular interest, when applying a three-dimensional ex vivo spheroid drug sensitivity assay on tumor cells harvested directly from 27 patients with melanoma lymph node metastases. Together, these results indicate that targeting AXL together with the DNA damage response pathway could be a promising treatment strategy in melanoma, and that further investigations in patient groups lacking treatment alternatives should be pursued.

Glutamine to proline conversion is associated with response to glutaminase inhibition in breast cancer

INTRODUCTION

Glutaminase inhibitors target cancer cells by blocking the conversion of glutamine to glutamate, thereby potentially interfering with anaplerosis and synthesis of amino acids and glutathione. The drug CB-839 has shown promising effects in preclinical experiments and is currently undergoing clinical trials in several human malignancies, including triple-negative breast cancer (TNBC). However, response to glutaminase inhibitors is variable and there is a need for identification of predictive response biomarkers. The aim of this study was to determine how glutamine is utilized in two patient-derived xenograft (PDX) models of breast cancer representing luminal-like/ER+ (MAS98.06) and basal-like/triple-negative (MAS98.12) breast cancer and to explore the metabolic effects of CB-839 treatment.

EXPERIMENTAL

MAS98.06 and MAS98.12 PDX mice received CB-839 (200 mg/kg) or drug vehicle two times daily p.o. for up to 28 days (n = 5 per group), and the effect on tumor growth was evaluated. Expression of 60 genes and seven glutaminolysis key enzymes were determined using gene expression microarray analysis and immunohistochemistry (IHC), respectively, in untreated tumors. Uptake and conversion of glutamine were determined in the PDX models using HR MAS MRS after i.v. infusion of [5-¹³C] glutamine when the models had received CB-839 (200 mg/kg) or vehicle for 2 days (n = 5 per group).

RESULTS

Tumor growth measurements showed that CB-839 significantly inhibited tumor growth in MAS98.06 tumors, but not in MAS98.12 tumors. Gene expression and IHC analysis indicated a higher proline synthesis from glutamine in untreated MAS98.06 tumors. This was confirmed by HR MAS MRS of untreated tumors demonstrating that MAS98.06 used glutamine to produce proline, glutamate, and alanine, and MAS98.12 to produce glutamate and lactate. In both models, treatment with CB-839 resulted in accumulation of glutamine. In addition, CB-839 caused depletion of alanine, proline, and glutamate ([1-13C] glutamate) in the MAS98.06 model.

CONCLUSION

Our findings indicate that TNBCs may not be universally sensitive to glutaminase inhibitors. The major difference in the metabolic fate of glutamine between responding MAS98.06 xenografts and non-responding MAS98.12 xenografts is the utilization of glutamine for production of proline. We therefore suggest that addiction to proline synthesis from glutamine is associated with response to CB-839 in breast cancer. The effect of glutaminase inhibition in two breast cancer patient-derived xenograft (PDX) models. ¹³C HR MAS MRS analysis of tumor tissue from CB-839-treated and untreated models receiving ¹³C-labeled glutamine ([5-¹³C] Gln) shows that the glutaminase inhibitor CB-839 is causing an accumulation of glutamine (arrow up) in two PDX models representing luminal-like breast cancer (MAS98.06) and basal-like breast cancer (MAS98.12). In MAS98.06 tumors, CB-839 is in addition causing depletion of proline ([5-¹³C] Pro), alanine ([1-¹³C] Ala), and glutamate ([1-¹³C] Glu), which could explain why CB-839 causes tumor growth inhibition in MAS98.06 tumors, but not in MAS98.12 tumors.

Basal-like breast cancer engages tumor-supportive macrophages via secreted factors induced by extracellular S100A4

The tumor microenvironment (TME) may influence both cancer progression and therapeutic response. In breast cancer, particularly in the aggressive triple‐negative/basal‐like subgroup, patient outcome is strongly associated with the tumor's inflammatory profile. Tumor‐associated macrophages (TAMs) are among the most abundant immune cells in the TME, shown to be linked to poor prognosis and therapeutic resistance. In this study, we investigated the effect of the metastasis‐ and inflammation‐associated microenvironmental factor S100A4 on breast cancer cells (BCCs) of different subtypes and explored their further interactions with myeloid cells. We demonstrated that extracellular S100A4 activates BCCs, particularly the basal‐like subtype, to elevate secretion of pro‐inflammatory cytokines. The secreted factors promoted conversion of monocytes to TAM‐like cells that exhibited protumorigenic activities: stimulated epithelial–mesenchymal transition, proliferation, chemoresistance, and motility in cancer cells. In conclusion, we have shown that extracellular S100A4 instigates a tumor‐supportive microenvironment, involving a network of cytokines and TAM‐like cells, which was particularly characteristic for basal‐like BCCs and potentiated their aggressive properties. The S100A4–BCC–TAM interaction cascade could be an important contributor to the aggressive behavior of this subtype and should be further explored for therapeutic targeting.

Abstract 4329: Targetable nodes in fibroblast-supported melanoma cells that show resistance to BRAF inhibitors

Metastatic melanoma is notorious for the ability to change its phenotype in response to signals from the microenvironment, which might influence how melanoma responds to therapy. We have disclosed an association between fibroblast-induced phenotypic alterations in melanoma and resistance to the mutated BRAF inhibitor vemurafenib (BRAFi). This signifies the need to find other targets than BRAF to eliminate stroma-influenced melanoma cells. To approach this challenge, we performed proteomic analysis and cancer drug sensitivity screening, comparing fibroblast-supported versus non-supported melanoma cells. We showed that the effect of fibroblasts was critically dependent on cell-cell proximity, where melanoma cells get trapped in a fibronectin network, produced by adjacent fibroblasts. In such environment, melanoma cells down-regulate melanocytic programs (MITF-driven), gain mesenchymal features (AXL, PDGFR, fibronectin) and activate stress/inflammatory-response signaling pathways (JNK and STAT3). Altogether, this indicates fibroblast-induced melanoma transition to a de-differentiated, mesenchymal-like, pro-inflammatory phenotype. Melanoma cells with such phenotype were less responsive to BRAF/MAPK inhibitors and a number of other targeted drugs. However, they showed enhanced sensitivity to PI3K/mTOR inhibitors and, particularly, an inhibitor of GSK3b, stimulating Wnt/b-catenin signaling. Further, we employed flow cytometry to measure the levels of Ki67 and pS6 in single melanoma cells upon different conditions/treatments. Such analysis allowed discrimination of cell subpopulations representing a proliferative and a quiescent cellular state, and nicely reflected the influence of the tested drugs in the presence or absence of fibroblasts. We observed a subpopulation of proliferative pS6high/Ki67high melanoma cells, which remained after treatment with BRAFi if fibroblasts were present. This, fibroblast-protected BRAFi-resistant cell subpopulation, could be reduced/eliminated by PI3K or GSK3b inhibitors, verifying PI3K/GSK3 as potential targets in fibroblast-rich tumors. Currently, we are using mass cytometry (CyTOF) to further characterize cell subpopulations with respect to multiple markers related to cell signaling and immune interactions. Preliminary results indicate that not only signaling protein levels, but also levels of immunoregulatory proteins are altered in melanoma cells that get support from the fibroblasts. In conclusion, we demonstrate fibroblast-induced melanoma switching to a mesenchymal-like pro-inflammatory phenotype, which favors melanoma resistance to BRAF inhibitors, but sensitizes to inhibitors of PI3K/mTOR-associated signaling. CyTOF-analysis of complex tumor-stroma cell systems is used to search for additional strategies to target stroma-supported melanoma cells, either at the level of signaling, or immune interactions.

Citation Format: Kotryna Seip, Marco V. Haselager, Kjetil Jørgensen, Marco Albrecht, Mads H. Haugen, Eivind Valen Egeland, Philippe Lucarelli, Thomas Sauter, Olav Engebraaten, Gunhild M. Mælandsmo, Lina Prasmickaite. Targetable nodes in fibroblast-supported melanoma cells that show resistance to BRAF inhibitors [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 4329. doi:10.1158/1538-7445.AM2017-4329

Abstract 1813: Bevacizumab potentiates the proteomic response to neoadjuvant chemotherapy in breast cancer patients: Rppa exploration of consecutive tumor samples in the NeoAva randomized phase II trial

Antiangiogenic therapy using bevacizumab has proven effective for a number of cancers; however, in breast cancer there is an unmet need to identify patients that benefit from such treatment. Sampling of tumor biopsies before and during treatment, as well as at the time of surgery enables the assessment of response at multiple molecular levels. At the proteomic level reverse phase protein analysis (RPPA) support expression of numerous cancer associated proteins simultaneously, which can further be used to unravel molecular mechanisms associated with clinical response to bevacizumab treatment.

In this phase II clinical trial, patients with HER2 negative primary tumors of ≥25 mm were treated with neoadjuvant chemotherapy (4 x FEC100 + 12 weeks of taxane-based therapy) and randomized (1:1) to receive bevacizumab or not. Mammography, ultrasound and MR imaging were used for response evaluation, in addition to final pathology assessment. Tumor responses were evaluable in 132 patients; of which 66 received bevacizumab. Ratio of the tumor size at final pathology assessment, and at inclusion was calculated to obtain a continuous scale of response reflecting the percentage of tumor shrinkage in response to therapy. Tumor biopsies were removed before start of treatment, at week 12 at the start of taxane-based tharapy and at the time of surgery. Lysates from each sample was analyzed on reverse phase protein arrays (RPPA) for expression levels of 210 proteins of which 54 were phospho-specific.

The addition of bevacizumab to the chemotherapy do not alter proteomic response from week 0 to 25 to such extent that this patient group cluster naturally together. While the proteomic response from week 0 to 12 in both treatment arms had an overall similar profile regarding up- and down-regulated proteins, the combination treatment (FEC100 + bevacizumab) induced substantially more effect on the regulation of each protein. This suggests that bevacizumab treatment have the capability to potentiate the effects of the anthracyclin based chemotherapy from week 0 to 12. Conversely, from week 12-25 (taxane-based therapy + bevacizumab) this effect was lost or even reversed, possibly due to a de-vascularized and less accessible tumor. An exception to this observation was a few phospho-proteins that do seem to have sustained stronger regulation over the whole treatment period. We are in the process of analyzing in more detail the impact of phosphorylation and thus protein activation states on treatment response.

Deciphering molecular response and activity regulation at the proteomic level is a promising approach and may reveal novel knowledge with potential important clinical relevance.

Citation Format: Mads H. Haugen, Ole Christian Lingjaerde, Marit Krohn, Wei Zhao, Evita M. Lindholm, Laxmi Silwal-Pandit, Elin Borgen, Øystein Garred, Anne Fangberget, Marit M. Holmen, Ellen Schlichting, Helle K. Skjerven, Steinar Lundgren, Erik Wist, Bjørn Naume, Gunhild M. Maelandsmo, Yiling Lu, Anne-Lise Boerresen-Dale, Gordon B. Mills, Olav Engebraaten. Bevacizumab potentiates the proteomic response to neoadjuvant chemotherapy in breast cancer patients: Rppa exploration of consecutive tumor samples in the NeoAva randomized phase II trial [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 1813. doi:10.1158/1538-7445.AM2017-1813

Abstract P6-13-01: Proteomic response in breast cancer treated with neoadjuvant chemotherapy with and without bevacizumab: Reverse phase protein array (RPPA) results from NeoAva - A randomized phase II study

BACKGROUND: Patients with HER2 negative primary tumors of ≥25 mm were treated with neoadjuvant chemotherapy (4 x FEC100 + 12 weeks of taxane-based therapy) and randomized (1:1) to receive bevacizumab or not. Mammography, ultrasound and MR imaging were used for response evaluation, in addition to the final pathology assessment after surgery.

HYPOTHESIS: RPPA proteomic analyses support identification of molecular mechanisms associated with clinical response to bevacizumab treatment.

METHODS: Tumor responses were evaluable in 132 patients; of which 66 received bevacizumab. Ratio of the tumor size at final pathology assessment, and at inclusion was calculated to obtain a continuous scale of response reflecting the percentage of tumor shrinkage in response to therapy. Tumor material was obtained at screening, 12 weeks into treatment and at surgical removal of tumors at 25 weeks. Lysates from each sample was analyzed on reverse phase protein arrays (RPPA) for expression levels of 210 proteins of which 54 were phospho-specific.

RESULTS: Several proteins were found for which expression prior to treatment reflected a better response on tumor shrinkage in the combination treatment arm (chemotherapy+bevacizumab). The proteomic response from week 0 to 12 in both treatment arms had an overall similar profile regarding up- and down-regulated proteins; however, the combination treatment (FEC100 + bevacizumab) induced a more pronounced effect on regulation of each protein. This might reflect the capability of bevacizumab therapy to potentiate the effects of the anthracyclin based chemotherapy from week 0 to 12. Conversely, from week 12-25 (taxane-based therapy + bevacizumab) this effect was lost or even reversed, except for certain phosphoproteins where potentiation imposed by bevacizumab was sustained throughout the whole treatment period. We are in the process of analyzing the impact of phosphorylation and thus protein activation states on treatment response. Furthermore, tumors with low hormone receptor pathway score demonstrated a better response in the combination treatment (chemotherapy+bevacizumab). Additionally, in these good responders the hormone signaling pathway was significantly upregulated during treatment. Further investigations are conducted to determine if this was due to selective ablation of hormone receptor negative tumor cells, or a re-programming of the molecular phenotype of cells present prior to treatment. The above mentioned results have potentially important clinical relevance and will be further investigated with respect to subtypes and the biological pathways affected by antiangiogenic therapy.

Citation Format: Haugen MH, Lindgjærde OC, Krohn M, Zhao W, Lindholm EM, Silwal-Pandit L, Borgen E, Garred Ø, Fangberget A, Holmen MM, Schlichting E, Skjerven H, Lundgren S, Wist E, Naume B, Mælandsmo GM, Lu Y, Børresen-Dale A-L, Mills GB, Engebråten O. Proteomic response in breast cancer treated with neoadjuvant chemotherapy with and without bevacizumab: Reverse phase protein array (RPPA) results from NeoAva - A randomized phase II study [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-13-01.

The Agaricus blazei-Based Mushroom Extract, Andosan™, Protects against Intestinal Tumorigenesis in the A/J Min/+ Mouse

Background

The novel A/J Min/+ mouse, which is a model for human Familial Adenomatous Polyposis (FAP), develops spontaneously multiple adenocarcinomas in the colon as well as in the small intestine. Agaricus blazei Murill (AbM) is an edible Basidiomycetes mushroom that has been used in traditional medicine against cancer and other diseases. The mushroom contains immunomodulating β-glucans and is shown to have antitumor effects in murine cancer models. Andosan^™ is a water extract based on AbM (82%), but it also contains the medicinal Basidiomycetes mushrooms Hericeum erinaceus and Grifola frondosa.

Methods and findings

Tap water with 10% Andosan^™ was provided as the only drinking water for 15 or 22 weeks to A/J Min/+ mice and A/J wild-type mice (one single-nucleotide polymorphism (SNP) difference), which then were exsanguinated and their intestines preserved in formaldehyde and the serum frozen. The intestines were examined blindly by microscopy and also stained for the tumor-associated protease, legumain. Serum cytokines (pro- and anti-inflammatory, Th1-, Th2 -and Th17 type) were measured by Luminex multiplex analysis. Andosan^™ treated A/J Min/+ mice had a significantly lower number of adenocarcinomas in the intestines, as well as a 60% significantly reduced intestinal tumor load (number of tumors x size) compared to control. There was also reduced legumain expression in intestines from Andosan^™ treated animals. Moreover, Andosan^™ had a significant cytotoxic effect correlating with apoptosis on the human cancer colon cell line, Caco-2, in vitro. When examining serum from both A/J Min/+ and wild type mice, there was a significant increase in anti-tumor Th1 type and pro-inflammatory cytokines in the Andosan^™ treated mice.

Conclusions

The results from this mouse model for colorectal cancer shows significant protection of orally administered Andosan^™ against development of intestinal cancer. This is supported by the finding of less legumain in intestines of Andosan^™ treated mice and increased systemic Th1 cytokine response. The mechanism is probably both immuno-modulatory and growth inhibition of tumor cells by induction of apoptosis.

Abstract 3268: Proteomic response in breast cancer treated with neoadjuvant chemotherapy with and without bevacizumab: Reverse phase protein array (RPPA) results from NeoAva - a randomized phase II study

BACKGROUND: In this phase II clinical trial, patients with HER2 negative primary tumors of ≥25 mm were treated with neoadjuvant chemotherapy (4 x FEC100 + 12 weeks of taxane-based therapy) and randomized (1:1) to receive bevacizumab or no bevacizumab. Mammography, ultrasound and MR imaging were used for response evaluation, in addition to final pathology assessment.

HYPOTHESIS: RPPA proteomic analyses support identification of molecular mechanisms associated with clinical response to bevacizumab treatment.

METHODS: Tumor responses were evaluable in 132 patients; of which 66 received bevacizumab. Ratio of the tumor size at final pathology assessment, and at inclusion was calculated to obtain a continuous scale of response reflecting the percentage of tumor shrinkage in response to therapy. Tumor material was obtained at screening, 12 weeks into treatment and at surgical removal of tumors at 25 weeks. Lysates from each sample was analyzed on reverse phase protein arrays (RPPA) for expression levels of 210 proteins of which 54 were phospho-specific. Data from protein analyses was compared to previously generated mRNA expression data.

RESULTS: Several proteins were found for which expression prior to treatment (week 0) reflected a better response on tumor shrinkage in the combination treatment arm (chemotherapy+bevacizumab): E.g. good responders had lower PDGFR-beta expression, and this was also observed at the mRNA level, while this result was not identified in the mono treatment arm (chemotherapy alone) on either level. The proteomic response from week 0 to 12 in both treatment arms had an overall similar profile regarding up- and down-regulated proteins; however, the combination treatment (FEC100 + bevacizumab) induced substantially more effect on regulation of each protein. This might reflect the capability of bevacizumab treatment to potentiate the effects of the anthracyclin based chemotherapy from week 0 to 12. Conversely, from week 12-25 (taxane-based therapy + bevacizumab) this effect was lost or even reversed, and reveals a possible need for further studies investigating changes in protein expression and correlation to response of a given treatment. Of particular interest were proteins that switched direction of regulation between the FEC and taxane-based regimes, however, these effects were not confined to the combination treatment and thus probably not due to the added bevacizumab. We are in the process of analyzing the impact of phosphorylation and thus protein activation states on treatment response. The above mentioned results have potentially important clinical relevance and will be further investigated with respect to subtypes and the biological pathways affected by antiangiogenic therapy.

Citation Format: Mads H. Haugen, Ole Christian Lingjaerde, Marit Krohn, Evita M. Lindholm, Laxmi Silwal-Pandit, Elin Borgen, Øystein Garred, Anne Fangberget, Marit M. Holmen, Ellen Schlichting, Helle Skjerven, Steinar Lundgren, Erik Wist, Bjoern Naume, Gunhild M. Maelandsmo, Yiling Lu, Anne-Lise Boerresen-Dale, Gordon B. Mills, Olav Engebraaten. Proteomic response in breast cancer treated with neoadjuvant chemotherapy with and without bevacizumab: Reverse phase protein array (RPPA) results from NeoAva - a randomized phase II study. [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 3268.

Fibroblast-induced switching to the mesenchymal-like phenotype and PI3K/mTOR signaling protects melanoma cells from BRAF inhibitors

The knowledge on how tumor-associated stroma influences efficacy of anti-cancer therapy just started to emerge. Here we show that lung fibroblasts reduce melanoma sensitivity to the BRAF inhibitor (BRAFi) vemurafenib only if the two cell types are in close proximity. In the presence of fibroblasts, the adjacent melanoma cells acquire de-differentiated mesenchymal-like phenotype. Upon treatment with BRAFi, such melanoma cells maintain high levels of phospho ribosomal protein S6 (pS6), i.e. active mTOR signaling, which is suppressed in the BRAFi sensitive cells without stromal contacts. Inhibitors of PI3K/mTOR in combination with BRAFi eradicate pS6high cell subpopulations and potentiate anti-cancer effects in melanoma protected by the fibroblasts. mTOR and BRAF co-inhibition also delayed the development of early-stage lung metastases in vivo. In conclusion, we demonstrate that upon influence from fibroblasts, melanoma cells undergo a phenotype switch to the mesenchymal state, which can support PI3K/mTOR signaling. The lost sensitivity to BRAFi in such cells can be overcome by co-targeting PI3K/mTOR. This knowledge could be explored for designing BRAFi combination therapies aiming to eliminate both stroma-protected and non-protected counterparts of metastases.

Construction of a plasmid coding for green fluorescent protein tagged cathepsin L and data on expression in colorectal carcinoma cells

The endo-lysosomal cysteine cathepsin L has recently been shown to have moonlighting activities in that its unexpected nuclear localization in colorectal carcinoma cells is involved in cell cycle progression (Tamhane et al., 2015) [1]. Here, we show data on the construction and sequence of a plasmid coding for human cathepsin L tagged with an enhanced green fluorescent protein (phCL-EGFP) in which the fluorescent protein is covalently attached to the C-terminus of the protease. The plasmid was used for transfection of HCT116 colorectal carcinoma cells, while data from non-transfected and pEGFP-N1-transfected cells is also shown. Immunoblotting data of lysates from non-transfected controls and HCT116 cells transfected with pEGFP-N1 and phCL-EGFP, showed stable expression of cathepsin L-enhanced green fluorescent protein chimeras, while endogenous cathepsin L protein amounts exceed those of hCL-EGFP chimeras. An effect of phCL-EGFP expression on proliferation and metabolic states of HCT116 cells at 24 h post-transfection was observed.

Enrichment of nuclear S100A4 during G2/M in colorectal cancer cells: possible association with cyclin B1 and centrosomes

S100A4 promotes metastasis in several types of cancer, but the involved molecular mechanisms are still incompletely described. The protein is associated with a wide variety of biological functions and it locates to different subcellular compartments, including nuclei, cytoplasm and extracellular space. Nuclear expression of S100A4 has been associated with more advanced disease stage as well as poor outcome in colorectal cancer (CRC). The present study was initiated to investigate the nuclear function of S100A4 and thereby unravel potential biological mechanisms linking nuclear expression to a more aggressive phenotype. CRC cell lines show heterogeneity in nuclear S100A4 expression and preliminary experiments revealed cells in G2/M to have increased nuclear accumulation compared to G1 and S cells, respectively. Synchronization experiments validated nuclear S100A4 expression to be most prominent in the G2/M phase, but manipulating nuclear levels of S100A4 using lentiviral modified cells failed to induce changes in cell cycle distribution and proliferation. Proximity ligation assay did, however, demonstrate proximity between S100A4 and cyclin B1 in vitro, while confocal microscopy showed S100A4 to localize to areas corresponding to centrosomes in mitotic cells prior to chromosome segregation. This might indicate a novel and uncharacterized function of the metastasis-associated protein in CRC cells.

Nuclear cathepsin L activity is required for cell cycle progression of colorectal carcinoma cells

Prominent tasks of cysteine cathepsins involve endo-lysosomal proteolysis and turnover of extracellular matrix constituents or plasma membrane proteins for maintenance of intestinal homeostasis. Here we report on enhanced levels and altered subcellular localization of distinct cysteine cathepsins in adenocarcinoma tissue in comparison to adjacent normal colon. Immunofluorescence and immunoblotting investigations revealed the presence of cathepsin L in the nuclear compartment in addition to its expected endo-lysosomal localization in colorectal carcinoma cells. Cathepsin L was represented as the full-length protein in the nuclei of HCT116 cells from which stefin B, a potent cathepsin L inhibitor, was absent. Fluorescence activated cell sorting analyses with synchronized cell cultures revealed deceleration of cell cycle progression of HCT116 cells upon inhibition of cathepsin L activity, while expression of cathepsin L-enhanced green fluorescent protein chimeras accelerated S-phase entry. We conclude that the activity of cathepsin L is high in the nucleus of colorectal carcinoma cells because of lacking stefin B inhibitory activity. Furthermore, we hypothesize that nuclear cathepsin L accelerates cell cycle progression of HCT116 cells thereby supporting the notion that cysteine cathepsins may play significant roles in carcinogenesis due to deregulated trafficking.

Proteolysis mediated by cysteine cathepsins and legumain-recent advances and cell biological challenges

Proteases play essential roles in protein degradation, protein processing, and extracellular matrix remodeling in all cell types and tissues. They are also involved in protein turnover for maintenance of homeostasis and protein activation or inactivation for cell signaling. Proteases range in function and specificity, with some performing distinct substrate cleavages, while others accomplish proteolysis of a wide range of substrates. As such, different cell types use specialized molecular mechanisms to regulate the localization of proteases and their function within the compartments to which they are destined. Here, we focus on the cysteine family of cathepsin proteases and legumain, which act predominately within the endo-lysosomal pathway. In particular, recent knowledge on cysteine cathepsins and their primary regulator legumain is scrutinized in terms of their trafficking to endo-lysosomal compartments and other less recognized cellular locations. We further explore the mechanisms that regulate these processes and point to pathological cases which arise from detours taken by these proteases. Moreover, the emerging biological roles of specific forms and variants of cysteine cathepsins and legumain are discussed. These may be decisive, pathogenic, or even deadly when localizing to unusual cellular compartments in their enzymatically active form, because they may exert unexpected effects by alternative substrate cleavage. Hence, we propose future perspectives for addressing the actions of cysteine cathepsins and legumain as well as their specific forms and variants. The increasing knowledge in non-canonical aspects of cysteine cathepsin- and legumain-mediated proteolysis may prove valuable for developing new strategies to utilize these versatile proteases in therapeutic approaches.

Nuclear legumain activity in colorectal cancer

The cysteine protease legumain is involved in several biological and pathological processes, and the protease has been found over-expressed and associated with an invasive and metastatic phenotype in a number of solid tumors. Consequently, legumain has been proposed as a prognostic marker for certain cancers, and a potential therapeutic target. Nevertheless, details on how legumain advances malignant progression along with regulation of its proteolytic activity are unclear. In the present work, legumain expression was examined in colorectal cancer cell lines. Substantial differences in amounts of pro- and active legumain forms, along with distinct intracellular distribution patterns, were observed in HCT116 and SW620 cells and corresponding subcutaneous xenografts. Legumain is thought to be located and processed towards its active form primarily in the endo-lysosomes; however, the subcellular distribution remains largely unexplored. By analyzing subcellular fractions, a proteolytically active form of legumain was found in the nucleus of both cell lines, in addition to the canonical endo-lysosomal residency. In situ analyses of legumain expression and activity confirmed the endo-lysosomal and nuclear localizations in cultured cells and, importantly, also in sections from xenografts and biopsies from colorectal cancer patients. In the HCT116 and SW620 cell lines nuclear legumain was found to make up approximately 13% and 17% of the total legumain, respectively. In similarity with previous studies on nuclear variants of related cysteine proteases, legumain was shown to process histone H3.1. The discovery of nuclear localized legumain launches an entirely novel arena of legumain biology and functions in cancer.

Intra- and extracellular regulation of activity and processing of legumain by cystatin E/M

Legumain, an asparaginyl endopeptidase, is up-regulated in tumour and tumour-associated cells, and is linked to the processing of cathepsin B, L, and proMMP-2. Although legumain is mainly localized to the endosomal/lysosomal compartments, legumain has been reported to be localized extracellularly in the tumour microenvironment and associated with extracellular matrix and cell surfaces. The most potent endogenous inhibitor of legumain is cystatin E/M, which is a secreted protein synthesised with an export signal. Therefore, we investigated the cellular interplay between legumain and cystatin E/M. As a cell model, HEK293 cells were transfected with legumain cDNA, cystatin E/M cDNA, or both, and over-expressing monoclonal cell lines were selected (termed M38L, M4C, and M3CL, respectively). Secretion of prolegumain from M38L cells was inhibited by treatment with brefeldin A, whereas bafilomycin A1 enhanced the secretion. Cellular processing of prolegumain to the 46 and 36 kDa enzymatically active forms was reduced by treatment with either substance alone. M38L cells showed increased, but M4C cells decreased, cathepsin L processing suggesting a crucial involvement of legumain activity. Furthermore, we observed internalization of cystatin E/M and subsequently decreased intracellular legumain activity. Also, prolegumain was shown to internalize followed by increased intracellular legumain processing and activation. In addition, in M4C cells incomplete processing of the internalized prolegumain was observed, as well as nuclear localized cystatin E/M. Furthermore, auto-activation of secreted prolegumain was inhibited by cystatin E/M, which for the first time shows a regulatory role of cystatin E/M in controlling both intra- and extracellular legumain activity.

Cystatin E/M suppresses legumain activity and invasion of human melanoma

Background

High activity of cysteine proteases such as legumain and the cathepsins have been shown to facilitate growth and invasion of a variety of tumor types. In breast cancer, several recent studies have indicated that loss of the cysteine protease inhibitor cystatin E/M leads to increased growth and metastasis. Although cystatin E/M is normally expressed in the skin, its role in cysteine protease regulation and progression of malignant melanoma has not been studied.

Methods

A panel of various non-melanoma and melanoma cell lines was used. Cystatin E/M and C were analyzed in cell media by immunoblotting and ELISA. Legumain, cathepsin B and L were analyzed in cell lysates by immunoblotting and their enzymatic activities were analyzed by peptide substrates. Two melanoma cell lines lacking detectable secretion of cystatin E/M were transfected with a cystatin E/M expression plasmid (pCST6), and migration and invasiveness were studied by a Matrigel invasion assay.

Results

Cystatin E/M was undetectable in media from all established melanoma cell lines examined, whereas strong immunobands were detected in two of five primary melanoma lines and in two of six lines derived from patients with metastatic disease. Among the four melanoma lines secreting cystatin E/M, the glycosylated form (17 kD) was predominant compared to the non-glycosylated form (14 kD). Legumain, cathepsin B and L were expressed and active in most of the cell lines, although at low levels in the melanomas expressing cystatin E/M. In the melanoma lines where cystatin E/M was secreted, cystatin C was generally absent or expressed at a very low level. When melanoma cells lacking secretion of cystatin E/M were transfected with pCST6, their intracellular legumain activity was significantly inhibited. In contrast, cathepsin B activity was not affected. Furthermore, invasion was suppressed in cystatin E/M over-expressing melanoma cell lines as measured by the transwell Matrigel assay.

Conclusions

These results suggest that the level of cystatin E/M regulates legumain activity and hence the invasive potential of human melanoma cells.

The metastasis-associated protein S100A4 exists in several charged variants suggesting the presence of posttranslational modifications

Background

S100A4 is a metastasis-associated protein which has been linked to multiple cellular events, and has been identified extracellularly, in the cytoplasm and in the nucleus of tumor cells; however, the biological implications of subcellular location are unknown. Associations between a variety of posttranslational protein modifications and altered biological functions of proteins are becoming increasingly evident. Identification and characterization of posttranslationally modified S100A4 variants could thus contribute to elucidating the mechanisms for the many cellular functions that have been reported for this protein, and might eventually lead to the identification of novel drugable targets.

Methods

S100A4 was immuoprecipitated from a panel of in vitro and in vivo sources using a monoclonal antibody and the samples were separated by 2D-PAGE. Gels were analyzed by western blot and silver staining, and subsequently, several of the observed spots were identified as S100A4 by the use of MALDI-TOF and MALDI-TOF/TOF.

Results

A characteristic pattern of spots was observed when S100A4 was separated by 2D-PAGE suggesting the presence of at least three charge variants. These charge variants were verified as S100A4 both by western immunoblotting and mass spectrometry, and almost identical patterns were observed in samples from different tissues and subcellular compartments. Interestingly, recombinant S100A4 displayed a similar pattern on 2D-PAGE, but with different quantitative distribution between the observed spots.

Conclusion

Endogenously expressed S100A4 were shown to exist in several charge variants, which indicates the presence of posttranslational modifications altering the net charge of the protein. The different variants were present in all subcellular compartments and tissues/cell lines examined, suggesting that the described charge variants is a universal phenomenon, and cannot explain the localization of S100A4 in different subcellular compartments. However, the identity of the specific posttranslational modification and its potential contribution to the many reported biological events induced by S100A4, are subject to further studies.