Identifying and targeting cancer stem cells in leiomyosarcoma: prognostic impact and role to overcome secondary resistance to PI3K/mTOR inhibition

Abstract

Background: Leiomyosarcoma (LMS) is one of the most frequent soft tissue sarcoma subtypes and is characterized by a consistent deregulation of the PI3K/mTOR pathway. Cancer stem cells (CSCs) have been poorly studied in soft tissue sarcomas. In this study, we aimed to evaluate the association between CSCs, the outcome of LMS patients, and the resistance to PI3K/mTOR pathway inhibition.

Methods: We investigated the relationships between aldehyde dehydrogenase 1 (ALDH1) expression, a cancer stem cell marker, and the outcome of LMS patients in two independent cohorts. We assessed the impact of CSCs in resistance to PI3K/mTOR pathway inhibition using LMS cell lines, a xenograft mouse model, and human tumor samples.

Results: We found that enhanced ALDH1 activity is a hallmark of LMS stem cells and is an independent prognostic factor. We also identified that secondary resistance to PI3K/mTOR pathway inhibition was associated with the expansion of LMS CSCs. Interestingly, we found that EZH2 inhibition, a catalytic component of polycomb repressive complex which plays a critical role in stem cell maintenance, restored sensitivity to PI3K/mTOR pathway inhibition. Importantly, we confirmed the clinical relevance of our findings by analyzing tumor samples from patients who showed secondary resistance after treatment with a PI3Kα inhibitor.

Conclusions: Altogether, our findings suggest that CSCs have a strong impact on the outcome of patients with LMS and that combining PI3K/mTOR and EZH2 inhibitors may represent a promising strategy in this setting.

Keywords: ALDH1; BEZ235; Cancer stem cell; EZH2; Leiomyosarcoma.

Fig. 1

Study of expression of ALDH1 in patients with sarcoma. a Kaplan-Meier curves of overall survival according to expression levels of ALDH1 from 145 LMS. b Kaplan-Meier curves of overall survival according to expression levels of ALDH1 from ICGC cohort (89 LMS). Red line, expression higher than the median expression level; green line: expression lower than the median. The result of the logrank test is indicated in each graph

Fig. 2

Anti-tumor effect of dual PI3K/mTOR inhibition in parental and resistant LMS cell lines. a Growth curves indicating growth inhibition of the three parental (IB112, IB134, and IB136) and resistant (resIB112, resIB134, and resIB136) LMS cell lines after BEZ235 treatment for 72 h. IC₅₀ is indicated in μM. b Percentage of apoptotic cells after BEZ235 treatment for 72 h. Data are presented as the mean ± SEM of three independent experiments. *resistant p < 0.05 vs. parental; **resistant p < 0.01 vs. parental; ***resistant p < 0.001 vs. parental (two-way ANOVA). c Representative signal intensities, obtained by western blotting, for p-S6RP^S240/244 were normalized to those for total S6RP in each cell line. Data are presented as the mean ± SEM of three independent experiments

Fig. 3

Anti-tumor effect of dual PI3K/mTOR inhibition in IB136 and resIB136-derived xenografts in Ragγ2C−/− mice. a Tumor volume progression curves over 3 weeks of BEZ235 treatment. Mice were randomly assigned to receive 40 mg/kg of BEZ235 or vehicle. The data points represent an average from 8 mice (bars, SEM). *p < 0.05; ***p < 0.001, two-way ANOVA. b Kaplan-Meier curves for tumor doubling times. c Immunohistochemical staining images of tumor samples treated with anti-Ki-67 and anti-pS6RP^ser240/244 antibodies (objective magnification, × 10). Percentages correspond to positively stained cells estimated by a pathologist

Fig. 4

Heatmap of differentially expressed genes between the IB136 and resIB136-derived xenografts. a Heatmap based on mean, normalized expression of general differentially expressed genes. b Heatmap based on mean, normalized expression of stem cell signaling components. The red and green colors denote high and low intensities, respectively. c Immunohistochemical staining images of tumor samples treated with anti-SOX2 and anti-ALDH1 antibodies (objective magnification, × 10). Percentages correspond to positively stained cells estimated by a pathologist. Endothelial cells (positive control) are indicated by black arrows

Fig. 5

Measurement of ALDH1 activity and its role in tumorigenicity and resistance. a Percentage of parental and resistant cells with ALDH1 activity as determined by flow cytometry. The SKBR3 cell line represents a positive control in the Aldefluor assay. Data are presented as the mean ± SEM of three independent experiments. *resistant p < 0.05 vs. parental; **resistant p < 0.01 vs. parental (two-tailed Student’s t tests). b Percentage of viable ALDH1^high and ALDH1^low FACS-sorted BEZ235-resistant cells after 72 h of 0.3 μM of BEZ235 treatment. The results represent the mean ± SEM. **p < 0.01; ***p < 0.001, one-way ANOVA. c Resistant LMS cells were sorted by FACS based on ALDH1 activity and the positive and negative subpopulations of sorted cells were submitted to in vitro and in vivo tumorsphere assays. Representative images of cell sorting by FACS and phase contrast microscopy of tumorspheres formed by resIB136 cells after 15 days of culture in nonadherent conditions in vitro. Scale bars, 150 μM. d Number of tumorspheres formed after 15 days in ALDH1^high and ALDH1^low subpopulation. e Number of tumorspheres formed after 15 days by cells dissociated from tumorspheres. The results represent the mean ± SEM. **p < 0.01; ***p < 0.001, one-way ANOVA

Fig. 6

Effect of EPZ011989 pretreatment on dual PI3K/mTOR sensitivity and on the CSC-like subpopulation in resistant leiomyosarcoma cell lines. a Immunohistochemical staining images of BEZ235-resistant LMS cell line pellets treated with anti-H3K27Me3 antibody (objective magnification, × 10). Percentage corresponds to positively stained cells estimated by a pathologist. b Growth curves indicating growth inhibition of the three resistant (resIB112, resIB134, and resIB136) LMS cell lines after BEZ235 treatment for 72 h. Resistant + EPZ011989 are the cells which were pretreated for 1 week with 10 μM of EPZ011989. IC₅₀ is indicated in μM. p value was calculated by one-way ANOVA. c Percentage of cells with ALDH1 activity, as determined by flow cytometry on resistant cell lines exposed to 10 μM of EPZ011989 for 1 week. Data are presented as the mean ± SEM of three independent experiments. **p < 0.01, two-tailed Student’s t tests

Fig. 7

Effect of BEZ235 and EPZ011989 on resIB136-derived xenografts in Ragγ2C−/− mice and study of expression of H3K27Me3 in patients with sarcoma. a Tumor volume progression curves during 5 weeks of treatment. Mice were randomly pretreated with vehicle or 125 mg/kg of EPZ011989 BID for 2 weeks. Then, in the EPZ011989-pretreated group, mice were randomly treated for 3 weeks with 40 mg/kg of BEZ235 or 125 mg/kg of EPZ011989 BID. The data points represent an average from eight mice (bars, SEM). *p < 0.05, two-way ANOVA. b Kaplan-Meier curves for tumor doubling times. c Immunohistochemical staining images of tumor samples treated with anti-Ki-67, anti-pS6RP^ser240/244, anti-SOX2, and anti-H3K27Me3 antibodies (objective magnification, × 10). Percentages correspond to positively stained cells estimated by a pathologist. Endothelial cells (positive control) are indicated by black arrows

Fig. 8

Study of expression of H3K27Me3 and heatmap of differentially expressed genes in patients with sarcoma. a Heatmap based on mean, normalized expression of general differentially expressed genes. b Heatmap based on mean, normalized expression of stem cell signaling components. The red and green colors denote high and low intensities, respectively. c Immunohistochemical staining images of three pairs of primary and PI3Kα inhibitor-resistant sarcoma tumor samples treated with anti-H3K27Me3 antibody (objective magnification, × 10). Percentages correspond to positively stained cells estimated by a pathologist. Endothelial cells (positive control) are indicated by black arrows