CDK4/6 and autophagy inhibitors synergistically induce senescence in Rb positive cytoplasmic cyclin E negative cancers

Abstract

Deregulation of the cell cycle machinery is a hallmark of cancer. While CDK4/6 inhibitors are FDA approved (palbociclib) for treating advanced estrogen receptor-positive breast cancer, two major clinical challenges remain: (i) adverse events leading to therapy discontinuation and (ii) lack of reliable biomarkers. Here we report that breast cancer cells activate autophagy in response to palbociclib, and that the combination of autophagy and CDK4/6 inhibitors induces irreversible growth inhibition and senescence in vitro, and diminishes growth of cell line and patient-derived xenograft tumours in vivo. Furthermore, intact G1/S transition (Rb-positive and low-molecular-weight isoform of cyclin E (cytoplasmic)-negative) is a reliable prognostic biomarker in ER positive breast cancer patients, and predictive of preclinical sensitivity to this drug combination. Inhibition of CDK4/6 and autophagy is also synergistic in other solid cancers with an intact G1/S checkpoint, providing a novel and promising biomarker-driven combination therapeutic strategy to treat breast and other solid tumours.

Figure 1. CDK4/6 inhibition induces ROS-mediated senescence and autophagy.

(a) IC₅₀ values of palbociclib in ER+ (MCF7, T47D, ZR75-1) and HMEC (MCF10A) cells treated with increasing concentrations of palbociclib (0.01–12 μM). MCF7, T47D and MCF10A cells were treated with DMSO (Cnt) or palbociclib for 6 days, allowed to recover (rel) and subjected to (b) Clonogenic assay, (c) Cell counting and (d) SA-ß galactosidase activity measurement with representative images. (e) Impact of combined siRNA knockdown of CDK4 and CDK6 along with treatment with DMSO, 1 or 5 μM palbociclib for 6 days. (f) Measurement of MDC-positive acidic vesicles, including autophagosomes, by flow cytometry in MCF7 and T47D cells treated siRNA against CDK4/6. (g) Western blot for LC3B I, II and p62 upon treatment with palbociclib for 6 days. (h) Representative TEM microphotograph of cells treated with DMSO (Cnt) or 1 μM palbociclib for 6 days. Red arrows indicate double-membraned autophagosomes. Scale bars, 500 nm. (i) Western blot of LC3B and p62 in MCF7 and T47D cells treated with a combination of 25 μM CQ for 1 h and palbociclib for 6 days. (j,k) Quantification of GFP-LC3 puncta (j) and RFP-GFP-LC3 puncta (k) and representative images in MCF7 cells treated with 1 μM palbociclib and/or 25 μM CQ for 48 h. (l,m) Cellular ROS measurement and quantification (MFI) of ROS levels in MCF7 and T47D cells upon transfection with siRNA against CDK4 and CDK6 (l) or treatment with palbociclib for 6 days (m). (n,o) LC3B and p62 protein levels (n) and MDC +ve cells (o) upon combined treatment with 10 mM NAC or 0.1 mM Trolox and 5 μM palbociclib for 6 days. All data represent mean±s.d. from three independent experiments; P values were calculated in comparison with cells treated with DMSO (Control) unless indicated. NS: P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Figure 2. Autophagy inhibition sensitizes cells to CDK4/6 inhibition.

(a) Western blot for Beclin-1 and Atg-5 in MCF7 and T47D cells after transfection with Scrambled (Scr), Beclin-1 or Atg5 shRNA. MCF7 and T47D cells with Beclin-1 or Atg5 knocked down were treated with palbociclib for 6 days and subjected to (b) dose–response assay after 6 days of recovery and (c) SA-ß galactosidase measurement. P values calculated in comparison with SCR −1 μM palbociclib. MCF7 and T47D cells were treated with combination of palbociclib and 15 μM HCQ for 6 days. Cells were allowed to recover for 4 or 6 days in drug-free media and subjected to (d) Cell counting, (e) clonogenic assay, (f) cellular ROS measurement and (g) SA-ß galactosidase measurement. (h) Clonogenic assay to measure impact of combined siRNA against CDK4/6 and treatment with 15 μM HCQ for 6 days. MCF7 and T47D cells were treated with 5 μM abemaciclib combined with 15 μM HCQ for 6 days and recovery for 4 or 6 days and subjected to (i) cell counting (j) measurement of total apoptotic cells (early apoptosis: Annexin V+/propidium iodide− and late apoptosis: Annexin V+/propidium iodide+) and (k) clonogenic assay. (l) Clonogenic assay in MCF7 and T47D cells treated with 10 μM spautin-1 or 1 nM bafilomycin A1 (Baf-A1) combined with 1 μM palbociclib for 6 days and recovery for 6 days. (m,n) Cell counting to assess growth of cells treated with 1 μM palbociclib and 7.5 μM Lys-05 (m) or 10 μM Spautin-1 (n) for 6 days and recovery for 4 days. All data represent mean±s.d. from three independent experiments; NS: P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Figure 3. Palbociclib synergizes with autophagy inhibition in vivo.

(a) Percentage change in tumour volume (normalized to Day 0) upon treatment with vehicle or palbociclib (n≥4 tumours per group) daily for 7 days. Tumours were then harvested for (b) Tumour weight measurement, (c) western blot of cell cycle (phospho-Rb, Rb, FOXM1) and autophagy (LC3B, p62) proteins, (d) TEM analysis (red arrows indicate double-membraned autophagosomes; Scale bars, 500 nm) and (e) BrdU, 8-OHdG and SA-ß gal staining (Scale bars, 50 μm). (f,g) Percentage change in mean (f) or individual (g) tumour volumes (normalized to Day 0) upon treatment with Vehicle, 25 mg kg⁻¹ palbociclib, 60 mg kg⁻¹ HCQ or combination of palbociclib (25 mg kg⁻¹) and HCQ daily for 21 days (treatment phase) and recovery for 21 days (recovery phase) (n≥8 tumours per group). Data represented as mean±s.e.m. Tumours were harvested for (h) tumour weight measurement at end of treatment+recovery phase (n≥6 for each group), (i) western blot of cell cycle (phospho-Rb, Rb) and autophagy (LC3B, p62) proteins, (j) RPPA analysis and Pathway score of proteins in the cell cycle (n=10) and senescence (n=13) pathways (Error bars represent maximum and minimum values) and (k) BrdU, 8-OHdG and SA-ß gal staining with representative images end of recovery phase and quantitation (Scale bars, 50 μm). (l,m) Percentage change in mean (l) or individual (m) tumour volumes (normalized to Day 0) upon treatment with Vehicle, 5 mg kg⁻¹ palbociclib, 10 mg kg⁻¹ Lys-05 or combination of palbociclib and Lys-05daily for 21 days (treatment phase) and recovery phase of 14 days. Data represented as mean±s.e.m. n≥5 for each group. (n) Kaplan–Meier survival curve with death and tumours exceeding 1,000 mm³ as end point upon treatment as in (l). n≥5 for each group. All data are represented as mean±s.d. and P values were calculated in comparison to mice treated with vehicle unless indicated. NS: P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Figure 4. G1/S checkpoint predicts response to combination of palbociclib and autophagy inhibition.

(a) Heat map constructed using the Gene Set Enrichment Analysis program, denoting expression of genes from the Biocarta_Cell_Cycle gene set. (b) Representative TEM microphotograph of parental (Par) and Rb-knockdown MCF7 and T47D cells treated with DMSO or 1 μM palbociclib for 6 days. Scale bars, 500 nm. Rb-knockdown MCF7 and T47D cells were treated with the combination of 1 μM palbociclib and 15 μM HCQ for 6 days, allowed to recover for 6 or 4 days and subjected to (c) Clonogenic assay and (d) Cell counting. MCF7 and T47D cells overexpressing vector (Vec), full-length cyclin E (EL) or LMWE were treated with palbociclib for 6 days and subjected to (e) Dose–response studies to measure Palbociclib IC₅₀ values and (f) Cell counting in combination with 15 μM HCQ. (g,h) Measurement of cellular ROS levels and quantification (MFI; numbers on graph) in Rb-knockdown (g) LMWE overexpressing (h) MCF7 and T47D cells treated palbociclib for 6 days. (i,j) Cell counting (i) and clonogenic assay (k) to assess proliferation upon treatment with 1 μM palbociclib, 1 μM letrozole and 15 μM HCQ in the indicated combinations. (k,l) Impact of overexpressing LMWE in MCF7 aromatase-expressing cells measured by cell counting (k) and clonogenic assay (l) following treatment with 1 μM palbociclib, 1 μM letrozole and 15 μM HCQ in the indicated combinations. (m) IC₅₀ values of letrozole and anastrazole in parental versus letrozole or anastrazole resistant cells, respectively. MCF7 parental, letrazole and anastrazole resistant cells were treated with palbociclib for 6 days and subjected to (n) Dose–response studies to measure Palbociclib IC₅₀ values and (o) Cell counting in combination with 15 μM HCQ. All data represent mean±s.d. from three independent experiments; P values were calculated in comparison to cells treated with DMSO (Control) or parental unless indicated. NS: P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Figure 5. Synergism between palbociclib and autophagy inhibition in other solid tumours.

(a) Western blot for Rb and Cyclin E in TNBC cell lines. (b) IC₅₀ values from drug response experiments in TNBC cell lines treated with palbociclib for 6 days. (c) Quantification of GFP-LC3 puncta and representative images in MDA-MB-231 cells treated with 1 μM palbociclib and/or 25 μM CQ for 48 h. TNBC cell lines were treated with combination of 1 μM palbociclib and 15 μM HCQ for 6 days and subjected to (d) Clonogenic and (e) SA-ß galactosidase assays. (f) Percentage change in volume (normalized to Day 0) of PDX tumours upon treatment with Vehicle, 25 mg kg⁻¹ palbociclib, 60 mg kg⁻¹ HCQ or combination of palbociclib (25 mg kg⁻¹) and HCQ daily for 21 days. Data represented as mean±s.e.m. n=4 for each group. (g,h) Kaplan–Meier survival curve with death and tumours exceeding 1,000 mm³ as end point (g) and weight (h) of PDX tumours treated as in (f). n=4 for each group. (i) Western blot of Rb and Cyclin E in ovarian, pancreatic (PDAC), lung, colon and prostate cancer cell lines. (j,k) IC₅₀ values from drug response experiments (j) and Clonogenic assay (k) in the mentioned cancer cell lines treated with palbociclib for 6 days and recovery for 6 days. (l,m) Cell counting (l) and clonogenic assay (m) in cancer cell lines treated with 1 μM palbociclib and 15 μM HCQ for 6 days and recovery for 4 or 6 days respectively. P value calculated in comparison with 1 μM palbociclib. (n) Correlation between palbociclib IC₅₀ values (from dose–response studies in all cancers) and levels of Rb and cyclin E proteins, with and without inhibition of autophagy (Beclin-1/Atg5 knockdown or HCQ treatment). (o) Schematic depicting the mechanism by which palbociclib inhibits growth of Rb+/LMWE− breast cancer cells by regulating ROS, autophagy and senescence. All data represent mean±s.d. from three independent experiments; NS: P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Figure 6. Rb and LMWE are prognostic markers of palbociclib in breast cancer patients.

(a) Alterations in Rb and cyclin E RNA levels in tumours from patients with breast (n=817), ovary (n=557), lung (n=230), pancreas (n=178), colon (n=379) or prostate (n=333) cancer taken from the TCGA database. (b) Percentage of breast cancer patients exhibiting alterations in Rb and LMWE as determined via immunohistochemical staining of tumour samples from the NCI patient cohort (n=879). (c) Proportion of 109 breast patients treated with palbociclib with Rb+/LMW−, Rb+/LMWE+ or Rb−/LMWE+ status and their disease progression (response to palbociclib). (d,e) Kaplan–Meier curves showing PFS duration (in months) among 109 patients with advanced ER+ breast cancer classified on the basis of their tumoral expression of Rb and cyclin E (d) and separated based on letrozole and fulvestrant (e). Survival curves were censored at disease progression or date of last follow-up. (f) Representative images from immunohistochemical analysis of Rb and LMWE in tumours from patients with ER+ breast cancer treated with palbociclib, classified on the basis of response. Scale bars equal 50 μm and insert scale bars equal 20 μm. (g) C-index of the multivariate cox model with progesterone receptor, prior therapy for metastatic disease only (without Rb) and the addition of Rb and LMWE.