Abstract 1116: HDAC8 activity regulates stress induced phenotype switching in melanoma

Background: Melanoma cells show great plasticity and can switch from a proliferative, drug-sensitive state to an invasive, drug-resistant phenotype through mechanisms that remain poorly defined.

Hypothesis: Transcriptional and chromatin regulatory factors alter the gene expression states of cell in response to targeted therapy treatment and allow melanoma cells to persist and grow in the face of initially effective therapy. These factors/pathways may represent additional targets for anti-melanoma therapy.

Methods: BRAF mutant melanoma cells were exposed to the BRAF inhibitor vemurafenib to generate cells that persisted and grew in the face of continued drug treatment. RNA-Seq, CHIP seq, ATAC-Seq and mass spectroscopy analyses were performed on persistent/resistant cells in comparison to parental cells. Additionally, histone deacetylase (HDAC) 8 and EP300 expression was enforced in drug naïve melanoma cell lines and the effects determined.

Results: HDAC8 (but not other HDACs) was consistently up-regulated in BRAFi resistant melanoma cells. Increased expression of HDAC8 reduced cell growth, increased p53 activation and suppressed microphthalmia-associated transcription factor MITF, suppressing the melanocyte lineage program driven, leading to increased drug resistance. Introduction of HDAC8 or its active mutant (S39A) increased the tolerance of melanoma cells to multiples stresses (UV-irradiation, hypoxia, and BRAF-MEK inhibitor therapy), reduced their proliferation and increased their invasive potential in in vitro and in vivo models. Gene expression analyses indicate genes associated with a proliferative, drug-sensitive phenotype (e.g. DCT, MITF, MLANA, SOX10) are downregulated in HDAC8 active cells while genes correlative with an invasive, drug-resistant phenotype are upregulated (e.g. AXL, EGFR, TGFB1, WNT5A). ATAC-Seq analysis showed that HDAC8 regulates chromatin structure and decreased accessibility in the DNA binding sites of MITF and Sox10 while increasing accessibility in the DNA binding sites of c-Jun and TEADs (e.g. YAP1). Additionally, ChIP-Seq showed an increase in HDAC8 activity decreased H3K27 acetylation in the promoter region of MITF and MITF regulated genes (e.g. DCT, MLANA, PMEL, TYR). Mass spectrometry-based acetylomics and co-immunoprecipitation experiments demonstrated that active HDAC8 associated with and deacetylated the histone acetyltransferase EP300, a known coactivator of MITF in melanocytes. Increased expression of EP300 increased the sensitivity of melanoma cells to UV-irradiation, hypoxia and BRAF-MEK inhibitor therapy.

Conclusions: HDAC8 associated with EP300 and mediated the switch to an invasive, stress resistant state in melanoma cells by suppressing MITF and increasing c-Jun driven transcriptional programs. Targeting HDAC8 could be a strategy to limit plasticity in advanced melanoma, increasing the efficacy of current therapies.

Citation Format: Michael Emmons, Richard Bennett, Alberto Riva, Bin Fang, Edward Seto, john koomen, Jonathan Licht, Keiran Smalley. HDAC8 activity regulates stress induced phenotype switching in melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1116.

Abstract 378: HDAC inhibition enhances MEK antagonist therapy in uveal melanoma through combined blockade of YAP, AKT and RTK signaling

Around 85-90% of all uveal melanomas harbor driver mutations in GNAQ or GNA11 leading to constitutive activation of numerous signaling pathways, including the MAPK pathway. MEK inhibitors have been evaluated clinically for metastatic uveal melanoma, but the responses are short-lived and the mechanisms of adaptation are poorly understood. In the current study, we performed RNA-seq and activity-based protein profiling (ABPP) to define the adaptive response of uveal melanoma cells to MEK inhibition and to design more effective combination therapy strategies. These analyses showed that MEK inhibition caused cytoskeleton remodeling driven by cortactin/Rho-GTPases with an increase in YAP activity, which in turn allowed therapeutic escape. Cortactin knockdown decreased YAP activity in response to MEK inhibition, increased cell death in vitro and was associated with tumor shrinkage in vivo. The proteomic data showed that MEK inhibition increased HDAC activity and an increase in global protein deacetylation. Co-targeting of HDACs and MEK was associated with increased apoptosis, decreased survival in 2D and 3D cell culture assays and suppression of YAP signaling. As YAP was unlikely to be the only escape pathway, we performed kinome and RTK arrays and demonstrated MEK inhibition also increases ROR1/2 and IGF-1R phosphorylation, leading to downstream PI3K/AKT signaling. At a signal transduction level, the combination of a pan-HDAC inhibitor (panobinostat) with a MEK inhibitor (trametinib) blocked all of the adaptive signaling pathways we identified, including RTKs, AKT, YAP and cortactin. In vivo xenograft studies revealed the MEK/HDAC inhibitor combination to outperform either agent alone, leading to long-term decrease of tumor growth and the suppression of adaptive PI3K/AKT, cortactin and YAP signaling. These findings identify HDAC inhibitors as a promising combination partner for MEK inhibitors in advanced uveal melanoma that may lead to improved systemic responses.

Citation Format: Fernanda Faiao-Flores, Michael Emmons, Michael Durante, Biswarup Saha, Bin Fang, John Koomen, Srikumar Chellappan, Silvya Maria-Engler, Jonathan Licht, William Harbour, Keiran Smalley. HDAC inhibition enhances MEK antagonist therapy in uveal melanoma through combined blockade of YAP, AKT and RTK signaling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 378.

Abstract 3800: HDAC8 regulates plasticity and escape from therapy in BRAF mutant melanoma

Melanoma cells are highly plastic and have the ability to switch to a dedifferentiated, invasive phenotype in response to multiple stimuli. We here show that exposure to melanoma cell lines and patient specimens to multiple stresses including BRAF-MEK inhibitor therapy, hypoxia and UV-irradiation leads to an increase in HDAC8 expression/activity, and in turn, the adoption of a drug-resistant, invasive phenotype. Systems level analyses using mass spectrometry-based phosphoproteomics and RNA-Seq demonstrated HDAC8 to be involved in pathways that regulated cell cycle entry, ribosome function, RNA binding, regulation of the cytoskeleton and MAPK pathway signaling. Introduction of HDAC8 into drug-naïve melanoma cells conveyed resistance in vitro and in in vivo xenograft models. The HDAC8-mediated BRAF inhibitor resistance was mediated through upstream receptor tyrosine kinase (RTK) activation, Ras/CRAF/MEK/ERK signaling and the suppression of apoptosis through increased stabilization of Mcl-1 and the inhibition of pro-apoptotic BIM. Among the multiple RTKs increased by HDAC8, EGFR emerged as a direct transcriptional target, and it was found that EGFR inhibition could overcome HDAC8-mediated tolerance to BRAF inhibition. Although it is known that HDACs primarily function at the histone level, they can also regulate signaling through the modulation of cytoplasmic protein acteylation. In line with this, we observed that HDAC8 introduction decreased the acetylation of c-JUN, leading to an increase in its transcriptional activity and the enrichment for an AP-1 gene signature. Mutation of putative acetylation sites in c-JUN (K268R, K271R, K273R) reduced the transcriptional activation of c-JUN in melanoma cells and conveyed resistance to BRAF inhibition through increased MAPK pathway activity. In vivo xenograft studies confirmed the key role of HDAC8 in therapeutic adaptation, with either a pan-HDAC inhibitor (panobinostat) or an HDAC8 inhibitor (PCI-34051) enhancing the durability of response to BRAF inhibitor therapy. We have thus identified HDAC8 as a key driver of phenotype switching in melanoma that regulates the responses to multiple cellular stresses and conveys resistance to BRAF inhibition. Our studies demonstrate that isoform-specific HDAC8 inhibitors could be an excellent strategy to limit the adaptation of melanoma cells to multiple stresses and therapeutic interventions, including the BRAF-MEK inhibitor combination.

Citation Format: Michael Emmons, Fernanda Flores, John Koomen, Edward Seto, Jane Messina, Eric Lau, Jonathan Licht, Keiran Smalley. HDAC8 regulates plasticity and escape from therapy in BRAF mutant melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3800.

Abstract 3025: Ligand-independent EphA2 signaling drives an amoeboid phenotype that promotes melanoma brain metastasis development

The majority of melanoma patients treated with BRAF inhibitors (BRAFi) ultimately develop resistance and fail on therapy. It has previously been shown that BRAFi resistance is highly correlated with the adoption of a strongly invasive phenotype through ligand-independent EphA2 signaling.

In this study, we used comprehensive mass spectrometry-based proteomic approaches to delineate the global signaling changes associated with the aggressive phenotype driven by ligand-independent EphA2 signaling. Preliminary data have identified new signaling adaptations associated with a mesenchymal-to-amoeboid transition (MAT) phenotype, which was confirmed in 3D collagen cultures. Functional experiments demonstrated that the amoeboid phenotype promoted cell migration and invasion, which was mediated through the interaction of EphA2 and CDC42. These findings were confirmed by expressing a constitutively activated form of CDC42. BRAFi resistant cell lines also exhibited the MAT phenotype and were more invasive compared to their treatment-naïve counterparts, in line with their dependence upon ligand-independent EphA2 signaling following BRAFi selection pressure.

To further demonstrate the metastatic potential of these amoeboid cells driven by ligand-independent EpHA2 signaling in vivo, we performed intracardiac injections in mice with cell lines expressing either EpHA2 S897A (inactive) or S897E (constitutively active phosphomimetic). Interestingly, we found a preferential homing of EphA2 S897E cells to the brain, but no difference in metastasis to other organs including the lung and liver. An analysis of brain metastasis specimens from patients failing BRAF and BRAF/MEK inhibitor therapy showed strong staining for the amoeboid phenotype marker EphA2.

To investigate whether the amoeboid phenotype may confer a survival advantage in circulation, we carried out shear stress assays. These experiments demonstrated that the amoeboid phenotype driven by ligand-independent EphA2 signaling promoted the survival of melanoma cells under shear stress. Cell attachment assays, trans-endothelial invasion assays and vascular permeability assays show that these cells are better suited to attach to and permeate an endothelial monolayer. We further show that inhibiting PI3K reversed the amoeboid phenotype and limited the EphA2-driven invasive capacity.

In summary, we show for the first time that BRAFi resistance is associated with the adoption of an MAT phenotype that increases the metastatic seeding of melanoma cells to the brain, which can be reversed through inhibition of PI3K signaling.

Citation Format: Chao Zhang, Inna Smalley, Ritin Sharma, Michael Emmons, Jane Messina, John Koomen, Keiran Smalley. Ligand-independent EphA2 signaling drives an amoeboid phenotype that promotes melanoma brain metastasis development [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3025.

Abstract 294: The role of the histone deacetylase HDAC 8 in mediating BRAF/MEK inhibitor metastatic dissemination and resistance

BRAF/MEK inhibitor resistance frequently results in increased phosphorylation of EphA2 at S897, which increases the invasive potential of melanoma cells in a ligand-independent manner. Through use of an unbiased phosphoproteomic screen, we identified HDAC phosphorylation as also being increased following the acquisition of BRAF/MEK inhibitor resistance. Treatment with the pan-HDAC inhibitor LBH 589 inhibited EphA2 phosphorylation at S897 and reduced tumor invasion, indicating a role for HDACs in the regulation of the resistance-associated metastatic phenotype. A comparison of HDAC protein levels in matched BRAF inhibitor sensitive/resistant cell lines showed that HDAC6, HDAC8, and HDAC11 protein levels were more highly expressed in BRAF resistant cell lines with HDAC8 being upregulated in 5 of 6 cell line pairs. shRNA knockdown of HDAC8 in the BRAF/MEK inhibitor resistant cell line WM164RR resulted in a decreased in pEphA2 as well as a decrease in pAKT at S473, which is known to directly bind to and phosphorylate EphA2 at S897. Reduced expression of HDAC8 in WM164RR cells resulted in a decrease in the invasive potential of the resistant phenotype as well as a resensitization of these cells to the BRAF inhibitor vemurafenib. HDAC8 overexpression in drug naïve cells led to increased melanoma cell invasion in a 3D spheroid and a matrigel invasion model. Overexpression of HDAC8 was also found to decrease the level of apoptosis following BRAF inhibitor treatment and was associated with increased number of clones in a colony formation assay. Conversely, knocking down HDAC8 with shRNA led to an increase in vemurafenib-mediated apoptosis. Mechanistically, it was found that HDAC8 overexpression increased the speed of adaptive MAPK signaling following BRAF inhibition and that this in turn limited the expression of pro-apoptotic BIM. Overall, these results indicate a dual role for HDAC8 in mediating BRAF/MEK inhibitor resistance and metastatic dissemination. Increased levels of HDAC8 caused an increase in invasion as well as a decrease in vemurafenib-induced apoptosis in BRAF inhibitor sensitive cell lines. Our studies suggest that selective HDAC8 inhibition may be one strategy to limit adaptation to both BRAF and BRAF/MEK inhibitor therapy.

Citation Format: Michael Emmons, Rashed Rab, Ritin Sharma, John Koomen, Keiran Smalley. The role of the histone deacetylase HDAC 8 in mediating BRAF/MEK inhibitor metastatic dissemination and resistance. [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 294.

Abstract 4439: Ero1L is a determinant of resistance to bortezomib and collateral sensitivity to MTI-101-induced cell death in myeloma

Our laboratory previously reported that the novel first-in-class peptidometitc, referred to as MTI-101, binds a CD44/ITGA4-containing complex and induces necrotic cell death in multiple myeloma cells. In this presentation, we worked towards further elucidating the mechanism of action of MTI-101. To this end, we examined changes in gene expression between samples that are either resistant or sensitive to MTI-101 using both an isogenic acquired resistant cell line model and primary myeloma patient specimens. Using cell line models we show that the acquisition of resistance to this class of compounds correlates with changes in genes predicted to regulate Ca2+ homeostasis. Changes included decreased expression of the IP3 receptor(IP3R) isoform 3, phospholipase C (PLC)-β, and SERCA3 channel, as well as decreased expression of the plasma membrane Ca2+ channels TRPC1 and TRPM7. Moreover, we show that MTI-101 treatment evokes a robust and sustained increase in intracellular Ca2+ levels, a phenotype that as predicted was attenuated in the drug-resistant cell line. Blocking Ca2+ release from the IP3R as well as blocking influx of Ca2+ from plasma membrane channels inhibited the MTI-101 induced Ca2+ response. We are also able to inhibit MTI-101 induced cell death by pharmacologically blocking release of Ca2+ from the IP3 receptor and blocking uptake by the mitochondria. Importantly, and consistent with what we previously reported with the linear analog HYD-1, MTI-101 is more potent in specimens obtained from relapsed patients compared to newly diagnosed patients. This differential response to MTI-101 treatment was used to compare baseline GEP in these primary myeloma samples. Using this strategy we show that in primary multiple myeloma specimens, the sensitivity to MTI-101 significantly correlates with increased levels of the thiol oxidase Ero1L. Ero1L has been shown by others to regulate Ca2+ release from the IP3R. Mechanistically, we demonstrate that reducing the expression of Ero1L is sufficient to confer resistance to MTI-101, a finding that correlates with attenuation of MTI-101-induced Ca2+ levels. Moreover, reducing the expression of Ero1L was sufficient to cause increased sensitivity to bortezomib-induced cell death. Importantly, a retrospective analysis of relapsed specimens showed that increased Ero1L levels were found to be a poor prognostic marker for response to bortezomib. Together, our data indicate that Ero1L is a prognostic marker for resistance to bortezomib and may account for the increased sensitivity to MTI-101 in relapsed multiple myeloma specimens. As MTI-101 induces cell death by the deregulation of Ca2+, we speculate that alterations in Ca2+ signaling may play a role in the progression of multiple myeloma.

Citation Format: Michael Emmons, Lori Hazlehurst, Steven Escherich, Mark McLaughlin, Javier Cuevas, Daniel Sullivan. Ero1L is a determinant of resistance to bortezomib and collateral sensitivity to MTI-101-induced cell death in myeloma. [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 4439. doi:10.1158/1538-7445.AM2015-4439

Abstract 1547: Acquisition of resistance to the β1 integrin antagonist HYD1 correlates with decreased α4, α6, αV and β1 integrin expression and decreased adhesion to fibronectin, VCAM-1 and vitronectin

Our laboratory recently reported that the D amino acid-containing β1 integrin inhibitory peptide HYD1 causes caspase independent cell death in multiple myeloma (MM) cell lines as a single agent in vitro and in vivo (Nair et al Molecular Cancer Therapeutics, 2009). Exposure of MM cells to HYD1 induces autophagy, but autophagy was shown to contribute to survival following HYD1 treatment. HYD1 treatment induces cell death in MM cell lines indicative of necrosis. These characteristics included: (a) an increase in the levels of reactive oxygen species (ROS), (b) ATP depletion and (c) loss of mitochondrial membrane potential. To further characterize the mechanism of HYD1 induced cell death an isogenic drug resistant variant was developed. This variant, referred to as H929-60, was developed by chronically exposing parental H929 cells to increasing doses of HYD1. Acquired resistance correlated with decreased HYD1 induced TO-PRO-3 positivity, decreased levels of ROS, reduced ATP depletion and reduced mmp loss compared to the H929 cell line. Additionally, the drug resistant variant H929-60 cells displayed decreased cell surface binding of FAM-HYD1 compared to H929 cells. These data indicate that decreased expression of the binding target(s) of HYD1 may contribute to drug resistance. Gene expression profiling (GEP) was used as an unbiased approach to identify changes in gene expression that may contribute to HYD1 resistance. Differentially expressed genes were mined in silico using the pathway analysis programs Ingenuity and GeneGo. The integrin signaling hub showed substantial changes in gene expression in the drug resistant variant compared to the parental cell line. Specifically, α4, α6, αV and β1 integrins were all down regulated along with the integrin associated protein CD47 in the HYD1 resistant cell line. These changes were consistent with decreased protein expression of integrins on the cell membrane and decreased functional adhesion to fibronectin, vitronectin and VCAM-1. Reducing the expression of α4 integrin using shRNA in the parental H929 cell line conferred partial resistance to HYD1 induced cell death. Together our data show that the acquisition of resistance to HYD1 correlated with decreased integrin expression and functional binding to extracellular matrixes. Future studies aim to determine whether a decrease in α6, αV and β1 integrins on the cell surface contributes to HYD1 resistance.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 1547.

Topoisomerase II beta levels are a determinant of melphalan-induced DNA crosslinks and sensitivity to cell death

The role of topoisomerase (topo) II in DNA repair has yet to be fully elucidated. Current evidence suggesting a role for topo II in the repair of DNA damage has been obtained by using in vitro model systems or inferred from correlative data in drug resistant cell lines. In this study we directly examined the role of topo IIalpha and beta in mediating the repair of melphalan-induced crosslinks in cellular DNA. To accomplish this, we used siRNA technology to knock down either topo IIalpha or beta in human chronic myelogenous leukemia K562 and histiocytic lymphoma U937 cell line. Our data demonstrate that topo IIbeta levels, (but not alpha), are a determinant of melphalan-induced crosslinks and sensitivity to melphalan. Furthermore, we show that knocking down topo IIbeta inhibits the repair of melphalan-induced crosslinks in K562 cells. These studies represent the first direct evidence that topo IIbeta participates in the repair of DNA damage induced by an alkylating agent in cellular DNA. Finally, these results suggest non-redundant roles for these two isoforms in mediating repair of DNA crosslinks.

Cell Adhesion to Fibronectin (CAM-DR) Influences Acquired Mitoxantrone Resistance in U937 Cells

Cell adhesion to fibronectin is known to confer a temporally related cell adhesion-mediated drug resistance (CAM-DR). However, it is unknown whether cell adhesion during drug selection influences the more permanent form of acquired drug resistance. To examine this question, we compared the acquisition of mitoxantrone resistance in U937 cells adhered to fibronectin versus cells selected in a traditional suspension culture. Our data show that acquired drug resistance levels of resistance to mitoxantrone are 2- to 3-fold greater for cells adhered to fibronectin compared with cells in suspension culture. We also compared mechanism(s) of resistance associated with drug selection in suspension versus fibronectin-adherent cultures. Drug resistance in both suspension and fibronectin-adhered cultures correlated with reduced drug-induced DNA damage and diminished topoisomerase II levels and activity; however, mechanisms regulating topoisomerase II levels differed depending on culture conditions. In suspension cultures, a reduction in topoisomerase IIbeta levels was detected at both RNA and protein levels. Furthermore, the decreased expression of topoisomerase IIbeta mRNA levels correlated with decreased expression of NF-YA. In contrast, in spite of no changes in NF-YA or topoisomerase IIbeta RNA expression, topoisomerase IIbeta protein levels were decreased in fibronectin-adherent, drug-resistant cells. In addition, topoisomerase IIalpha protein levels (but not RNA levels) were reduced in drug resistance cells selected on fibronectin; however, no change in topoisomerase IIalpha was observed in cells selected with mitoxantrone in suspension culture. Taken together, our results suggest that the development of drug resistance models must consider interactions with the microenvironment to identify clinically relevant targets and mechanisms associated with acquired drug resistance.

Measuring simultaneous fluxes from soil of N2O and N2 in the field Using the 15N-gas "nonequilibium" technique

Our purpose was to measure simultaneous fluxes from soil of both N2O and N2 from the same plot in the field using the 15N-gas "nonequilibrium" technique (i.e., the "Hauck" technique) as used previously for N2. We accommodated analysis of N2O by modifying the head amplifier of our mass spectrometer. Our system accurately measured the 15N enrichments of labeled soil slurries for both N2 and N2O. In the field, we measured flux of N2 and N2O during soil denitrification from a 15N-labeled plot of winter wheat. Nine chamber incubations were conducted over 4 days. N2 flux ranged from below detection limit (<0.022 g x m(-2) x d(-1)) to 0.055 g x m(-2) x d(-1). N2O flux ranged from 0.0002 to 0.0027 g N2O-N x m(-2) x d(-1), with a detection limit of 1.0 x 10(-6) g N2O-N x m(-2) x d(-1). For N2O flux, the 15N-gas technique and gas chromatography technique agreed well (r = 0.98). The 15N enrichment of the soil mineral pool undergoing denitrification, measured nondestructively using the N2O data, dropped from about 0.82 to 0.72 atom fraction 15N over 4 days. Applying the 15N-gas nonequilibrium technique to N2O complements its use for 15N-N2 analysis when studying the relative production of N2O and N2 during denitrification.