Mechanistic Support for Combined MET and AR Blockade in Castration-Resistant Prostate Cancer

A recent phase III trial of the MET kinase inhibitor cabozantinib in men with castration-resistant prostate cancer (CRPC) failed to meet its primary survival end point; however, most men with CRPC have intact androgen receptor (AR) signaling. As previous work supports negative regulation of MET by AR signaling, we hypothesized that intact AR signaling may have limited the efficacy of cabozantinib in some of these patients. To assess the role of AR signaling on MET inhibition, we first performed an in silico analysis of human CRPC tissue samples stratified by AR signaling status (⁺ or ⁻), which identified MET expression as markedly increased in AR⁻ samples. In vitro, AR signaling inhibition in AR⁺ CRPC models increased MET expression and resulted in susceptibility to ligand (HGF) activation. Likewise, MET inhibition was only effective in blocking cancer phenotypes in cells with MET overexpression. Using multiple AR⁺ CRPC in vitro and in vivo models, we showed that combined cabozantinib and enzalutamide (AR antagonist) treatment was more efficacious than either inhibitor alone. These data provide a compelling rationale to combine AR and MET inhibition in CRPC and may explain the negative results of the phase III cabozantinib study in CRPC. Similarly, the expression of MET in AR⁻ disease, whether due to AR inhibition or loss of AR signaling, suggests potential utility for MET inhibition in select patients with AR therapy resistance and in AR⁻ prostate cancer.     

Calcein-effluxing human colon cancer cells are enriched for self-renewal capacity and depend on β-catenin

Putative cancer stem cells are a subpopulation of cancer cells that give rise to chemotherapy resistance and are therefore of prognostic and therapeutic interest, though their identification remains elusive in colon cancer due to lack of reliable and accurate markers. We previously identified a p53-dependent putative cancer stem cell population, the calcein low population (C^loP), based on their exclusive efflux of the fluorescent dye Calcein. This functional identification method enables comparative live cell studies of subpopulations without differential toxicity that occurs with traditional Hoechst methods, which has confounded conclusions and limited the utility of this cancer stem cell marker. In this study, we examined the cancer stem cell-like properties of the C^loP population in vivo in comparison with the parental and calcein-high population (C^hiP) in human colon cancer xenografts. Serial dilution xenograft experiments in NOD/SCID mice revealed that the C^loP is only marginally more tumorigenic compared to the C^hiP or parental cells. However, serial passage of these tumors revealed that the C^loP is uniquely enriched for self-renewal capacity in vivo compared to the other populations. Immunohistochemical analysis of these tumors revealed that the C^loP possesses increased levels of nuclear β-catenin and furthermore, siRNA-mediated knockdown of β-catenin significantly reduced the C^loP population. These findings highlight the C^loP as an important subpopulation of tumor cells that are exclusively endowed with the ability to self-renew and propagate tumors. The dependency of the C^loP on β-catenin provides a molecular explanation for this ability and suggests that this population can and should be therapeutically targeted by inhibition of Wnt signaling.     

Primary breast cancer patients with high risk clinicopathologic features have high percentages of bone marrow epithelial cells with ALDH activity and CD44⁺CD24lo cancer stem cell phenotype

Background

Cancer stem cells (CSCs) are purported to be epithelial tumour cells expressing CD44⁺CD24^lo that exhibit aldehyde dehydrogenase activity (Aldefluor⁺). We hypothesised that if CSCs are responsible for tumour dissemination, disseminated cells in the bone marrow (BM) would be positive for putative breast CSC markers. Therefore, we assessed the presence of Aldefluor⁺ epithelial (CD326⁺CD45^dim) cells for the presence of the CD44⁺CD24^lo phenotype in BM of patients with primary breast cancer (PBC).

Methods

BM aspirates were collected at the time of surgery from 66 patients with PBC. Thirty patients received neoadjuvant chemotherapy (NACT) prior to aspiration. BM was analysed for Aldefluor⁺ epithelial cells with or without CD44⁺CD24^lo expression by flow cytometry. BM aspirates from three healthy donors (HD) were subjected to identical processing and analyses and served as controls.

Results

Patients with triple-receptor-negative (TN) tumours had a significantly higher median percentage of CD44⁺CD24^lo CSC within Aldefluor⁺ epithelial cell population than patients with other immunohistochemical subtypes (P = 0.018). Patients with TN tumours or with pN2 or higher pathologic nodal status were more likely to have a proportion of CD44⁺CD24^lo CSC within Aldefluor⁺ epithelial cell population above the highest level of HD. Furthermore, patients who received NACT were more likely to have percentages of Aldefluor⁺ epithelial cells than the highest level of HD (P = 0.004).

Conclusion

The percentage of CD44⁺CD24^lo CSC in the BM is higher in PBC patients with high risk tumour features. The selection or enrichment of Aldefluor⁺ epithelial cells by NACT may represent an opportunity to target these cells with novel therapies.     

Cell type of origin as well as genetic alterations contribute to breast cancer phenotypes

Breast cancer is classified into different subtypes that are associated with different patient survival outcomes, underscoring the importance of understanding the role of precursor cell and genetic alterations in determining tumor subtypes. In this study, we evaluated the oncogenic phenotype of two distinct mammary stem/progenitor cell types designated as K5⁺/K19⁻ or K5⁺/K19⁺ upon introduction of identical combinations of oncogenes-mutant H-Ras (mRas) and mutant p53 (mp53), together with either wild-type ErbB2(wtErbB2) or wild-type EGFR (wtEGFR). We examined their tumor forming and metastasis potential, using both in-vitro and in-vivo assays. Both the combinations efficiently transformed K5⁺/K19⁻ or K5⁺/K19⁺ cells. Xenograft tumors formed by these cells were histologically heterogeneous, with variable proportions of luminal, basal-like and claudin-low type components depending on the cell types and oncogene combinations. Notably, K5⁺/K19⁻ cells transformed with mRas/mp53/wtEGFR combination had a significantly longer latency for primary tumor development than other cell lines but more lung metastasis incidence than same cells expressing mRas/mp53/wtErbB2. K5⁺/K19⁺ cells exhibit shorter overall tumor latency, and high metastatic potential than K5⁺/K19⁻ cells, suggesting that these K19⁺ progenitors are more susceptible to oncogenesis and metastasis. Our results suggest that both genetic alterations and cell type of origin contribute to oncogenic phenotype of breast tumors.     

Distinct breast cancer stem/progenitor cell populations require either HIF1α or loss of PHD3 to expand under hypoxic conditions

The heterogeneous nature of breast cancer is a result of intrinsic tumor complexity and also of the tumor microenvironment, which is known to be hypoxic. We found that hypoxia expands different breast stem/progenitor cell populations (cells with increased aldehyde dehydrogenase activity (Aldefluor⁺), high mammosphere formation capacity and CD44⁺CD24^−/low cells) both in primary normal epithelial and tumor cells. The presence of the estrogen receptor (ER) limits hypoxia-dependent CD44⁺CD24^−/low cell expansion. We further show that the hypoxia-driven cancer stem-like cell enrichment results from a dedifferentiation process. The enhanced mammosphere formation and Aldefluor⁺ cell content observed in breast cancer cells relies on hypoxia-inducible factor 1α (HIF1α). In contrast, the CD44⁺CD24^−/low population expansion is HIF1α independent and requires prolyl hydroxylase 3 (PHD3) downregulation, which mimics hypoxic conditions, leading to reduced CD24 expression through activation of NFkB signaling. These studies show that hypoxic conditions expand CSC populations through distinct molecular mechanisms. Thus, potential therapies that combine current treatments for breast cancer with drugs that target CSC should take into account the heterogeneity of the CSC subpopulations.     

Prospective identification of tumorigenic osteosarcoma cancer stem cells in OS99‐1 cells based on high aldehyde dehydrogenase activity

High aldehyde dehydrogenase (ALDH) activity has recently been used to identify tumorigenic cell fractions in many cancer types. Herein we hypothesized that a subpopulation of cells with cancer stem cells (CSCs) properties could be identified in established human osteosarcoma cell lines based on high ALDH activity. We previously showed that a subpopulation of cells with high ALDH activity were present in 4 selected human osteosarcoma cell lines, of which a significantly higher ALDH activity was present in the OS99‐1 cell line that was originally derived from a highly aggressive primary human osteosarcoma. Using a xenograft model in which OS99‐1 cells were grown in NOD/SCID mice, we identified a highly tumorigenic subpopulation of osteosarcoma cells based on their high ALDH activity. Cells with high ALDH activity (ALDH^br cells) from the OS99‐1 xenografts were much less frequent, averaging 3% of the entire tumor population, compared to those isolated directly from the OS99‐1 cell line. ALDH^br cells from the xenograft were enriched with greater tumorigenicity compared to their counterparts with low ALDH activity (ALDH^lo cells), generating new tumors with as few as 100 cells in vivo. The highly tumorigenic ALDH^br cells illustrated the stem cell characteristics of self‐renewal, the ability to produce differentiated progeny and increased expression of stem cell marker genes OCT3/4A, Nanog and Sox‐2. The isolation of osteosarcoma CSCs by their high ALDH activity may provide new insight into the study of osteosarcoma‐initiating cells and may potentially have therapeutic implications for human osteosarcoma.     

Carboxyl-terminal domain of MUC16 imparts tumorigenic and metastatic functions through nuclear translocation of JAK2 to pancreatic cancer cells

MUC16 (CA125) is a type-I transmembrane glycoprotein that is up-regulated in multiple cancers including pancreatic cancer (PC). However, the existence and role of carboxyl-terminal MUC16 generated following its cleavage in PC is unknown. Our previous study using a systematic dual-epitope tagged domain deletion approach of carboxyl-terminal MUC16 has demonstrated the generation of a 17-kDa cleaved MUC16 (MUC16-Cter). Here, we demonstrate the functional significance of MUC16-Cter in PC using the dual-epitope tagged version (N-terminal FLAG- and C-terminal HA-tag) of 114 carboxyl-terminal residues of MUC16 (F114HA). In vitro analyses using F114HA transfected MiaPaCa-2 and T3M4 cells showed enhanced proliferation, motility and increased accumulation of cells in the G2/M phase with apoptosis resistance, a feature associated with cancer stem cells (CSCs). This was supported by enrichment of ALDH⁺ CSCs along with enhanced drug-resistance. Mechanistically, we demonstrate a novel function of MUC16-Cter that promotes nuclear translocation of JAK2 resulting in phosphorylation of Histone-3 up-regulating stemness-specific genes LMO2 and NANOG. Jak2 dependence was demonstrated using Jak2^+/+ and Jak2^−/− cells. Using eGFP-Luciferase labeled cells, we demonstrate enhanced tumorigenic and metastatic potential of MUC16-Cter in vivo. Taken together, we demonstrate that MUC16-Cter mediated enrichment of CSCs is partly responsible for tumorigenic, metastatic and drug-resistant properties of PC cells.     

Phosphorylation of the androgen receptor at Ser81 is co‐sustained by CDK1 and CDK9 and leads to AR‐mediated transactivation in prostate cancer

Androgen receptor (AR) is the principal molecule in prostate cancer (PCa) etiology and therapy. AR re‐activation still remains a major challenge during treatment of castration‐resistant prostate cancer (CRPC) tumors that relapse after castration therapies. Recent reports have indicated the enrichment of Ser81‐phosphorylated AR (pS81) in the nucleus of CRPC cells, and CDK1 and CDK9 as the kinases phosphorylating AR at S81. In the current study we showed that pS81 is preferentially localized in the nucleus in both rapid biopsy metastatic CRPC samples and PCa xenografts, and nuclear pS81 localization is correlated with AR transactivation in tumor xenografts. Chromatin immunoprecipitation (ChIP) analysis demonstrated an alignment of S81 phosphorylation and AR‐mediated transactivation with the chromatin locus openness. Moreover, pS81‐specific ChIP‐Seq showed a disproportional occupancy of pS81 on AR‐activated promoters, while 3C‐ChIP assays further indicated an enrichment of pS81 at the PSA enhancer‐promoter loop, a known AR activating hub. In the latter, CDK9 was shown to modulate the transactivation of the AR and RNA Pol II. Indeed, ChIP and re‐ChIP assays also confirmed that AR‐dependent activation of the PSA enhancer and promoter mediated by pS81 was coupled with activation of Pol II and the pTEFb complex. Mechanistically, we determined that CDK1 and CDK9 sustained the pS81 AR modification in the soluble and chromatin‐bound fractions of PCa cells, respectively. Finally, we demonstrated that CDK1 activity was maintained throughout the cell cycle, and that CDK1 inhibitors restored androgen sensitivity in CRPC tumor cells. Based on these findings, CDK1 and CDK9 could be targeted as pS81 kinases in patients with CRPC, either alone or in conjunction with direct AR antagonists.     

Cancer stem cell-driven efficacy of trastuzumab (Herceptin): towards a reclassification of clinically HER2-positive breast carcinomas

Clinically HER2+ (cHER2+) breast cancer (BC) can no longer be considered a single BC disease entity in terms of trastuzumab responsiveness. Here we propose a framework for predicting the response of cHER2+ to trastuzumab that integrates the molecular distinctions of intrinsic BC subtypes with recent knowledge on cancer stem cell (CSC) biology. First, we consider that two interchangeable populations of epithelial-like, aldehyde dehydrogenase (ALDH)-expressing and mesenchymal-like, CD44⁺CD24^−/low CSCs can be found in significantly different proportions across all intrinsic BC subtypes. Second, we overlap all the intrinsic subtypes across cHER2+ BC to obtain a continuum of mixed phenotypes in which one extreme exhibits a high identity with ALDH+ CSCs and the other extreme exhibits a high preponderance of CD44⁺CD24^−/low CSCs. The differential enrichment of trastuzumab-responsive ALDH+ CSCs versus trastuzumab-refractory CD44⁺CD24^−/low CSCs can explain both the clinical behavior and the primary efficacy of trastuzumab in each molecular subtype of cHER2+ (i.e., HER2-enriched/cHER2+, luminal A/cHER2+, luminal B/cHER2+, basal/cHER2+, and claudin-low/cHER2+). The intrinsic plasticity determining the epigenetic ability of cHER2+ tumors to switch between epithelial and mesenchymal CSC states will vary across the continuum of mixed phenotypes, thus dictating their intratumoral heterogeneity and, hence, their evolutionary response to trastuzumab. Because CD44⁺CD24^−/low mesenchymal-like CSCs distinctively possess a highly endocytic activity, the otherwise irrelevant HER2 can open the door to a type of “Trojan horse” approach by employing antibody-drug conjugates such as T-DM1, which will allow a rapid and CSC-targeted delivery of cytotoxic drugs to therapeutically manage trastuzumab-unresponsive basal/cHER2+ BC. Contrary to the current dichotomous model used clinically, our model proposes that a reclassification of cHER2+ tumors based on the spectrum of molecular BC subtypes might inform on their CSC-determined sensitivity to trastuzumab, thus providing a better delineation of the predictive value of cHER2+ in BC by incorporating CSCs-driven intra-tumor heterogeneity into clinical decisions.     

AMP-activated protein kinase is dispensable for maintaining ATP levels and for survival following inhibition of glycolysis,but promotes tumour engraftment of Ras-transformed fibroblasts

Lactic acid generated by highly glycolytic tumours is exported by the MonoCarboxylate Transporters, MCT1 and MCT4, to maintain pHi and energy homeostasis. We report that MCT1 inhibition combined with Mct4 gene disruption severely reduced glycolysis and tumour growth without affecting ATP levels. Because of the key role of the 5′-AMP-activated protein kinase (AMPK) in energy homeostasis, we hypothesized that targeting glycolysis (MCT-blockade) in AMPK-null (Ampk^−/−) cells should kill tumour cells from ‘ATP crisis’. We show that Ampk^−/−-Ras-transformed mouse embryonic fibroblasts (MEFs) maintained ATP levels and viability when glycolysis was inhibited. In MCT-inhibited MEFs treated with OXPHOS inhibitors the ATP level and viability collapsed in both Ampk^+/+ and Ampk^−/− cells. We therefore propose that the intracellular acidification resulting from lactic acid sequestration mimicks AMPK by blocking mTORC1, a major component of an ATP consuming pathway, thereby preventing ‘ATP crisis’. Finally we showed that genetic disruption of Mct4 and/or Ampk dramatically reduced tumourigenicity in a xenograft mouse model suggesting a crucialrolefor these two actors in establishment of tumours in a nutrient-deprived environment. These findings demonstrated that blockade of lactate transport is an efficient anti-cancer strategy that highlights the potential in targeting Mct4 in a context of impaired AMPK activity.     

Triple negative breast cancers comprise a highly tumorigenic cell subpopulation detectable by its high responsiveness to a Sox2 regulatory region 2 (SRR2) reporter

We have recently described a novel phenotypic dichotomy within estrogen receptor-positive breast cancer cells; the cell subset responsive to a Sox2 regulatory region (SRR2) reporter (RR cells) are significantly more tumorigenic than the reporter unresponsive (RU) cells. Here, we report that a similar phenomenon also exists in triple negative breast cancer (TNBC), with RR cells more tumorigenic than RU cells. First, examination of all 3 TNBC cell lines stably infected with the SRR2 reporter revealed the presence of a cell subset exhibiting reporter activity. Second, RU and RR cells purified by flow cytometry showed that RR cells expressed higher levels of CD44, generated more spheres in a limiting dilution mammosphere formation assay, and formed larger and more complex structures in Matrigel. Third, within the CD44^High/CD24⁻ tumor-initiating cell population derived from MDA-MB-231, RR cells were significantly more tumorigenic than RU cells in an in vivo SCID/Beige xenograft mouse model. Examination of 4 TNBC tumors from patients also revealed the presence of a RR cell subset, ranging from 1.1-3.8%. To conclude, we described a novel phenotypic heterogeneity within TNBC, and the SRR2 reporter responsiveness is a useful marker for identifying a highly tumorigenic cell subset within the CD44^High/CD24⁻tumor-initiating cell population.     

Helios,and not FoxP3, is the marker of activated Tregs expressing GARP/LAP

Regulatory T cells (Tregs) are key players of immune regulation/dysregulation both in physiological and pathophysiological settings. Despite significant advances in understanding Treg function, there is still a pressing need to define reliable and specific markers that can distinguish different Treg subpopulations. Herein we show for the first time that markers of activated Tregs [latency associated peptide (LAP) and glycoprotein A repetitions predominant (GARP, or LRRC32)] are expressed on CD4⁺FoxP3⁻ T cells expressing Helios (FoxP3⁻Helios⁺) in the steady state. Following TCR activation, GARP/LAP are up-regulated on CD4⁺Helios⁺ T cells regardless of FoxP3 expression (FoxP3^+/−Helios⁺). We show that CD4⁺GARP^+/−LAP⁺ Tregs make IL-10 immunosuppressive cytokine but not IFN-γ effector cytokine. Further characterization of FoxP3/Helios subpopulations showed that FoxP3⁺Helios⁺ Tregs proliferate in vitro significantly less than FoxP3⁺Helios⁻ Tregs upon TCR stimulation. Unlike FoxP3⁺Helios⁻ Tregs, FoxP3⁺Helios⁺ Tregs secrete IL-10 but not IFN-γ or IL-2, confirming they are bona fide Tregs with immunosuppressive characteristics. Taken together, Helios, and not FoxP3, is the marker of activated Tregs expressing GARP/LAP, and FoxP3⁺Helios⁺ Tregs have more suppressive characteristics, compared with FoxP3⁺Helios⁻ Tregs. Our work implies that therapeutic modalities for treating autoimmune and inflammatory diseases, allergies and graft rejection should be designed to induce and/or expand FoxP3⁺Helios⁺ Tregs, while therapies against cancers or infectious diseases should avoid such expansion/induction.     

HDAC inhibitor confers radiosensitivity to prostate stem-like cells

Background:

Radiotherapy can be an effective treatment for prostate cancer, but radiorecurrent tumours do develop. Considering prostate cancer heterogeneity, we hypothesised that primitive stem-like cells may constitute the radiation-resistant fraction.

Methods:

Primary cultures were derived from patients undergoing resection for prostate cancer or benign prostatic hyperplasia. After short-term culture, three populations of cells were sorted, reflecting the prostate epithelial hierarchy, namely stem-like cells (SCs, α₂β₁integrin^hi/CD133⁺), transit-amplifying (TA, α₂β₁integrin^hi/CD133⁻) and committed basal (CB, α₂β₁integrin^lo) cells. Radiosensitivity was measured by colony-forming efficiency (CFE) and DNA damage by comet assay and DNA damage foci quantification. Immunofluorescence and flow cytometry were used to measure heterochromatin. The HDAC (histone deacetylase) inhibitor Trichostatin A was used as a radiosensitiser.

Results:

Stem-like cells had increased CFE post irradiation compared with the more differentiated cells (TA and CB). The SC population sustained fewer lethal double-strand breaks than either TA or CB cells, which correlated with SCs being less proliferative and having increased levels of heterochromatin. Finally, treatment with an HDAC inhibitor sensitised the SCs to radiation.

Interpretation:

Prostate SCs are more radioresistant than more differentiated cell populations. We suggest that the primitive cells survive radiation therapy and that pre-treatment with HDAC inhibitors may sensitise this resistant fraction.     

Clonogenic Multiple Myeloma Cells have Shared stemness Signature Associated with Patient Survival

Multiple myeloma is the abnormal clonal expansion of post germinal B cells in the bone marrow. It was previously reported that clonogenic myeloma cells are CD138⁻. Human MM cell lines RPMI8226 and NCI H929 contained 2-5% of CD138⁻ population. In this study, we showed that CD138⁻ cells have increased ALDH1 activity, a hallmark of normal and neoplastic stem cells. CD138⁻ALDH⁺ cells were more clonogenic than CD138⁺ALDH⁻ cells and only CD138⁻ cells differentiated into CD138⁺ population. In vivo tumor initiation and clonogenic potentials of the CD138⁻ population was confirmed using NOG mice. We derived a gene expression signature from functionally validated and enriched CD138⁻ clonogenic population from MM cell lines and validated these in patient samples. This data showed that CD138⁻ cells had an enriched expression of genes that are expressed in normal and malignant stem cells. Differentially expressed genes included components of the polycomb repressor complex (PRC) and their targets. Inhibition of PRC by DZNep showed differential effect on CD138⁻ and CD138⁺ populations. The ‘stemness’ signature derived from clonogenic CD138⁻ cells overlap significantly with signatures of common progenitor cells, hematopoietic stem cells, and Leukemic stem cells and is associated with poorer survival in different clinical datasets.     

Small-molecule inhibition of BRD4 as a new potent approach to eliminate leukemic stem- and progenitor cells in acute myeloid leukemia (AML)

Acute myeloid leukemia (AML) is a life-threatening stem cell disease characterized by uncontrolled proliferation and accumulation of myeloblasts. Using an advanced RNAi screen-approach in an AML mouse model we have recently identified the epigenetic ‘reader’ BRD4 as a promising target in AML. In the current study, we asked whether inhibition of BRD4 by a small-molecule inhibitor, JQ1, leads to growth-inhibition and apoptosis in primary human AML stem- and progenitor cells. Primary cell samples were obtained from 37 patients with freshly diagnosed AML (n=23) or refractory AML (n=14). BRD4 was found to be expressed at the mRNA and protein level in unfractionated AML cells as well as in highly enriched CD34⁺/CD38⁻ and CD34⁺/CD38⁺ stem- and progenitor cells in all patients examined. In unfractionated leukemic cells, submicromolar concentrations of JQ1 induced major growth-inhibitory effects (IC₅₀ 0.05-0.5 μM) in most samples, including cells derived from relapsed or refractory patients. In addition, JQ1 was found to induce apoptosis in CD34⁺/CD38⁻ and CD34⁺/CD38⁺ stem- and progenitor cells in all donors examined as evidenced by combined surface/Annexin-V staining. Moreover, we were able to show that JQ1 synergizes with ARA-C in inducing growth inhibition in AML cells. Together, the BRD4-targeting drug JQ1 exerts major anti-leukemic effects in a broad range of human AML subtypes, including relapsed and refractory patients and all relevant stem- and progenitor cell compartments, including CD34⁺/CD38⁻ and CD34⁺/CD38⁺ AML cells. These results characterize BRD4-inhibition as a promising new therapeutic approach in AML which should be further investigated in clinical trials.