Glucose-6-phosphate dehydrogenase blockade potentiates tyrosine kinase inhibitor effect on breast cancer cells through autophagy perturbation
- 210 Downloads
Glucose-6-phospate dehydrogenase (G6PD) is the limiting enzyme of the pentose phosphate pathway (PPP) correlated to cancer progression and drug resistance. We previously showed that G6PD inhibition leads to Endoplasmic Reticulum (ER) stress often associated to autophagy deregulation. The latter can be induced by target-based agents such as Lapatinib, an anti-HER2 tyrosine kinase inhibitor (TKI) largely used in breast cancer treatment.
Here we investigate whether G6PD inhibition causes autophagy alteration, which can potentiate Lapatinib effect on cancer cells. Immunofluorescence and flow cytometry for LC3B and lysosomes tracker were used to study autophagy in cells treated with lapatinib and/or G6PD inhibitors (polydatin). Immunoblots for LC3B and p62 were performed to confirm autophagy flux analyses together with puncta and colocalization studies. We generated a cell line overexpressing G6PD and performed synergism studies on cell growth inhibition induced by Lapatinib and Polydatin using the median effect by Chou-Talay. Synergism studies were additionally validated with apoptosis analysis by annexin V/PI staining in the presence or absence of autophagy blockers.
We found that the inhibition of G6PD induced endoplasmic reticulum stress, which was responsible for the deregulation of autophagy flux. Indeed, G6PD blockade caused a consistent increase of autophagosomes formation independently from mTOR status. Cells engineered to overexpress G6PD became resilient to autophagy and resistant to lapatinib. On the other hand, G6PD inhibition synergistically increased lapatinib-induced cytotoxic effect on cancer cells, while autophagy blockade abolished this effect. Finally, in silico studies showed a significant correlation between G6PD expression and tumour relapse/resistance in patients.
These results point out that autophagy and PPP are crucial players in TKI resistance, and highlight a peculiar vulnerability of breast cancer cells, where impairment of metabolic pathways and autophagy could be used to reinforce TKI efficacy in cancer treatment.
KeywordsPentose phosphate pathway Autophagy Lapatinib ER stress TKI Breast cancer Polydatin
Epidermal growth factor receptor
Human epidermal growth factor receptor 2
Intensity correlation quotient
Inositol-requiring enzyme 1
Lysosomal-associated membrane protein 1
Microtubule-associated proteins 1A/1B light chain 3B
Nicotinamide adenine dinucleotide phosphate
Protein kinase R (PKR)-like endoplasmic reticulum kinase
Pentose phosphate pathway
Reactive oxygen species
Tyrosine kinase inhibitor
Unfolded Protein Response
In recent years, metabolic deregulations have been studied as prognostic factors for tumours and as potential targets for innovative treatments [1, 2, 3, 4]. Pentose phosphate pathway (PPP), a cytoplasmic metabolic process parallel to glycolysis, synthesizes the nucleotide precursor ribose-5-phosphate and produces NADPH, the reduced form of NADP+ (the nicotinamide adenine dinucleotide phosphate), that is an essential cofactor for the synthesis of lipids and maintenance of redox balance of the cell [5, 6]. NADPH counteracts oxidative stress produced in highly metabolizing cancer cells in form of reactive oxygen species (ROS). Glucose-6-phosphate dehydrogenase (G6PD) is the key enzyme of the PPP, it is often over-expressed in several types of cancer such as breast cancer, oesophageal carcinoma, renal cancer and is correlated to worse prognosis [7, 8, 9, 10]. Indeed, some of the main oncogenes and tumour suppressors such as p53 and K-Ras, can directly regulate the enzymes involved in this pathway [6, 11, 12]. Several studies showed that the inhibition of G6PD may result in the development of therapeutic strategies against tumour growth and metastasis [13, 14, 15].
Breast cancer is the most common type of cancer in women and is associated with high mortality that is due to tumour aggressiveness and drug resistance. This cancer is classified according to the expression of oestrogen receptor (ERec), progesterone receptor (PR), and/or human epidermal growth factor receptor 2 (HER-2). The expression of these receptors determines the phenotype of the breast cancer and thus defines the therapeutic strategy to follow [16, 17]. Lapatinib, an orally administered small-molecule, is a dual tyrosine kinase inhibitor (TKI) that targets both HER-2 receptor and the epidermal growth factor receptor (EGFR), and which has been widely used for the treatment of breast cancer . This inhibitor induces apoptosis and autophagy in cancer cells. However, resistance to lapatinib treatment is also observed. The resistance can be caused by alterations in autophagosome and autolysosome proteins, suggesting a potential role of autophagy [19, 20, 21]. Furthermore, the increase of the antioxidant metabolic pathway has been associated with TKI resistance in cancer. This suggests that the strategies based on the targeting of antioxidant metabolic pathways may improve the efficiency of TKI-based therapy . In this study, we investigated the interplay between G6PD, ER stress and autophagy, and highlighted new possible strategies to improve the effect of TKIs in treatment of breast cancer overcoming drug resistance.
Materials and methods
Chemicals, cell culture and in vitro treatment
All chemicals were purchased from Sigma-Aldrich (St. Louis, USA) unless otherwise specified. Trans-polydatin, with a purity grade higher than 99%, was been supplied by Ghimas spa (Casalecchio, Bologna, Italy). Selective inhibitors of IRE1α (4μ8C) and PERK (GSK 2606414) were obtained from Tocris Bioscience (Bristol, United Kingdom). Lapatinib (Tyverb®) was obtained from GSK (Verona, Italy). MCF7 cells were purchased from ATCC. Cells were cultured in DMEM (Gibco, NY, USA) supplemented with 2 mM glutamine, 100 IU/mL penicillin, 100 μg/mL streptomycin (Invitrogen, Carlsbad, CA), and 10% heat-inactivated foetal bovine serum (FBS) (Gibco, NY, USA) at 37 °C in a humidified atmosphere under 5% CO2. All cell lines were kept mycoplasma free, checking was performed every three months.
Cell viability assay
Cell viability was measured by the colorimetric 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were seeded in 96-well plates at a density of 104 cells per well, then they were treated with 100 μL of 1 mg/mL MTT (Sigma) in DMEM medium containing 10% foetal bovine serum for 4 h at 37 °C. The medium was then replaced with 200 μL of DMSO and shaken for 15 min, then absorbance at 540 nm was measured using a microplate ELISA reader with DMSO used as the blank. To quantify the synergistic or antagonist effect of the drugs combinations, CompuSyn software was used .
After 24 h treatment with PD at various concentrations or 30 μM chloroquine (autophagy positive control), cells were washed in PBS and fixed with 4% paraformaldehyde (PFA) solution and permeabilized with 0.1% TRITON -X/PBS solution, then was performed a blocking in 1% BSA for 1 hat RT. Cells were incubated with LC3B antibody (Cell Signalling, USA) and Anti-SQSTM1/p62 (ab56416, Abcam, Cambridge, UK) in PBS for 30 min. Secondary antibodies were added after a PBS wash in the same conditions. Cells were incubated in a 1:500 solution of 10 mg/mL Hoechst (Invitrogen) in PBS for 10 min in the dark. To stain endoplasmic reticulum cells were incubated with 200 nM ER-Tracker Blue-White DPX in PBS solution for 20 min at 37 °C. For positive control cells were exposed for 16 h to 5 μg/mL tunicamycin. Images were collected under a fluorescence microscope (EVOS FL Cell Imaging System, Thermo Scientific, Rockford, USA). To stain lysosomes, cells were incubated with 60 nM LysoTracker (Thermo Fisher Scientific, USA) for 45 min at 37 °C. ImageJ (Fiji plugin) software was used for the calculation of Puncta and Intensity correlation quotient (ICQ).
For intracellular staining cells were fixed with Fix and Perm Reagent A (Invitrogen) for 20 min, and then resuspended in Fix and Perm Reagent B (Invitrogen) for 30 min containing primary LC3B antibody (Cell Signalling, USA).
Secondary antibodies were added in Fix and Perm Reagent B (Invitrogen) in the same conditions. Apoptosis (Annexin V apoptosis detection kit, BD biosciences), CellROX assay (Thermo Fisher Scientific, USA), LysoTracker assay (Thermo Fisher Scientific, USA), were performed according to the manufacturer’s instructions. Cells were analyzed with a FACSAria III (BD Biosciences, San Jose, CA) or a BD Accuri Cytometer (BD Biosciences, San Jose, CA). Data were analysed by FlowJo V10 software (FlowJo LLC, USA).
p3-G6PD-t1 and negative control pCMV3-untagged-NCV (control) hygromycin-resistant plasmids were purchased from Sino Biological Inc. (Sino Biological, Beijing, China). MCF7 cells were stably transfected with Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, MA USA) according to the manufacturer’s instructions. Clones with upregulated expression of G6PD were selected with 100 μg/mL Hygromycin. Clones were screened by Western blot.
Protein extraction and Western blotting
Cells were lysed in 1x RIPA buffer (150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris. Cl pH 7.5) plus 1% protease inhibitor cocktail, 1% PMSF (200 mM) and 1% sodium orthovanadate (Santa Cruz Biotechnology, USA). Lysates were clarified by centrifugation at 8000 x g for 5 min at 4 °C and equal amounts of protein were fractionated by SDS-PAGE and subsequently transferred onto nitrocellulose membrane, immunoblots were visualized using Supersignal® West Pico Chemiluminescent substrate (Thermo Scientific, Rockford, USA). Proteins were detected with anti-Glucose 6 Phosphate Dehydrogenase (Novus Biologicals, USA); Anti-SQSTM1 / p62 (ab56416, Abcam, Cambridge, UK); anti-Phospho-Akt (Ser473) (9271, Cell Signalling, USA), anti-Akt (9272, Cell Signalling, USA); anti-phospho-mTOR (Ser2448) (2971, Cell Signalling, USA); anti-mTOR (2972, Cell Signalling, USA); anti-LC3B (2775, Cell Signalling, USA); anti-α-Tubulin Antibody (#2144 Cell Signalling Technology, UK); Anti-GAPDH (ab9485, Abcam, Cambridge, UK) were used for assessing loading.
In silico analysis
In silico validation was performed using gene expression profiles generated as part of the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC)  and Pawitan  studies. The gene expression profiles were generated using an Illumina HT-12 v3.0 Gene Expression BeadChip (METABRIC) and an Affymetrix Human Genome U133A/U133B (Pawitan) array. The gene expression of G6PD was selected and min-max normalised across the complete patient population of both datasets. The normalised gene expression was analysed using scatter plots for the differentiation of G6PD expression across breast cancer subtypes (Her2+ enriched, Basal-like, Luminal A, Luminal B and Normal-like) and disease recurrence. Unpaired t-test was used to analyse differences between sample groups. In addition, Kaplan-Meier curves were generated presenting disease-free survival (DFS) in relation to G6PD expression. Patient groups were separated according to median expression and survival curves were analysed using Mantel-Cox test. P-values below 0.05 were considered to be significant (p ≤ 0.05 = *, p ≤ 0.01 = **, p ≤ 0.001 = *** and p ≤ 0.0001 = ****). Analysis was performed using GraphPad Prism 8.
G6PD inhibition induces autophagy in breast cancer cells
Autophagy is dependent on ER stress
G6PD overexpression reduces autophagy and induces resistance to Lapatinib
Autophagy mediates cell death and synergism
G6PD expression is correlated to disease-free survival (DFS) and recurrence in breast cancer patients
In this manuscript, we describe for the first time that the inhibition of G6PD causes an activation of autophagic flux, which synergistically increases the cytotoxic effect of Lapatinib on breast cancer cells. Pentose phosphate pathway (PPP) is a major player in glucose catabolism that results in the production of NADPH, necessary to control redox balance, lipid metabolism and nucleotide precursors [6, 40]. G6PD is the limiting enzyme of the PPP and it is finely regulated following the cell redox state and metabolic needs. PPP activity and G6PD itself are often upregulated in cancer and are associated with aggressiveness, drug resistance and poor prognosis [7, 8, 9]. Thus, inhibition of this pathway has been pointed out as a major goal for the definition of new cancer treatments . Indeed, targeting G6PD results in cancer cytotoxicity, reduction of metastases and restoration of sensitivity to drug [11, 12, 26, 35]. Nevertheless, the choice among G6PD inhibitors is very limited and debated . Recently our group discovered that the natural molecule polydatin directly inhibits G6PD causing oxidative stress, endoplasmic reticulum stress and apoptosis in cancer cells . Polydatin is a glucoside of resveratrol that has been studied for many years and for different pathological conditions, including inflammation and cancer [42, 43, 44]. MCF7 is most widely used breast cancer cell line for in vitro research . PPP and G6PD have been proven to play an important role in their metabolism [15, 46, 47] as well as they have been used to study lapatinib effect and resistance especially in combination with other drugs [48, 49]. Here, we show that, G6PD inhibition induced an increase of both intracytoplasmic vesicles (puncta) that colocalize with Lysotracker, a marker of acidic compartments, and LC3B which is associated to autophagosomes. During autophagy induction, LC3-I is converted into LC3-II with a concurrent decrease in p62. LC3B activation was confirmed also by immunoblotting.
In the study of autophagy flux, inhibition of lysosomal degradation by chloroquine or bafilomycin A causes accumulation of both LC3-II and p62, and this reflects the amount of LC3-II and p62 that would have been degraded by autophagy over the treatment period [50, 51, 52]. In Fig. 2c the increase of p62 and LC3B in cells co-treated with chloroquine is evident. Other studies using metabolic inhibitor show an increase in autophagic flux leading to cells death , this corroborate our strategy to use metabolic autophagy inducer to potentiate anticancer drug effect.
mTOR is a master regulator of cell metabolism by controlling autophagy, among other processes. Typically, when mTOR is inactive or inhibited, autophagy occurs. Therefore, most of the molecules that influence autophagy acts directly or indirectly on the mTOR pathway . Resveratrol, which can be produced from polydatin degradation, has been described as mTOR inhibitor [28, 55, 56]; thus, to exclude a direct effect of polydatin on this pathway, we analysed the activation of mTOR and its upstream regulator AKT. We showed that both AKT and mTOR are not inhibited. A possible case in which autophagy is induced independently from mTOR is represented by the unfolded protein response process that is initiated following ER stress [57, 58]. Indeed, both IRE1 and PERK, which are the principal mediator of UPR, have been described to induce autophagy independently from mTOR status [39, 59]. Our results confirmed  that G6PD inhibition caused ER stress and that, if UPR was inhibited by blocking IRE1 or PERK, polydatin was not still able to induce autophagy.
Tyrosine kinase inhibitors are a class of drugs that are largely used in clinical settings to treat different cancers including breast carcinoma. The work of Gregory et al.  pointed out that G6PD is responsible for resistance to FMS-like tyrosine kinase 3 inhibitors in acute myeloid leukaemia due to increased redox metabolism. Among TKIs that are used to treat breast cancer, lapatinib has been described to cause cancer cells cytotoxicity by inducing autophagy [19, 34]. Therefore, we hypothesized that G6PD might play a role in modulating lapatinib effect on cancer cells. For this purpose, we created a cell line overexpressing G6PD and showed that they are resistant to autophagy induced by either lapatinib or polydatin. On the other hand, cells that received the mock plasmid showed a strong induction of autophagy, especially with the combination of the two drugs. These results were confirmed by viability analyses where the MCF7G6PD+ were less sensitive to lapatinib. Moreover, the analysis of synergism between polydatin and lapatinib showed that the two molecules were highly synergistic in the MCF7mock at both 24 h and 48 h after treatment, while this effect was less evident or absent on the MCF7G6PD+. These results were confirmed by Annexin V/PI analysis. Moreover, to confirm the role of autophagy in cancer cell death, we analysed both synergism and apoptosis in the presence of 3-metyladenin (3-MA), which is a widely used autophagy inhibitor [37, 60]. Differently from chloroquine and bafilomycin A, 3-MA inhibits autophagosomes formation at early stages. For these reasons, 3-MA has been widely used in similar experiments [38, 60]. In these experimental conditions, synergism was completely prevented and cell viability increased from about 27% in polydatin/lapatinib-treated cells to about 50% in the presence of 3-MA. In silico studies on two publicly available patient’s databases showed a significant association between G6PD and HER2+ patients and an inverse correlation between G6PD expression and DFS. These data were on line with other reports that used different databases [7, 8]. Possible explanation of the synergistic mechanism showed here involve autophagic cell death. Shimuzu et al. showed that Bcl-2 family of protein, which is a regulator of apoptosis, control autophagic cell death by binding to Beclin-1 and APG5 . Moreover, Ros induce JNK phosphorylation can activate autophagy by interfering with the interaction between Beclin-1 and bcl-2 . In addition, another mechanism that could be involved in this process is a novel programmed autophagic cells death depend on iron called ferroptosis. This involved an increase of lipid peroxides and excessive degradation of ferritin and NCOA4 [62, 63]. Future studies are necessary to understand the precise role of this mechanisms and pathways in the context described in this manuscript.
Taken together, our results led to the following conclusions: i. G6PD blockade caused autophagy through ER stress; ii. G6PD activity influenced lapatinib effect on cancer cells by preventing autophagy; iii. G6PD inhibitors such as polydatin might be used to increase lapatinib effect on breast cancer.
Cancer is more and more evidenced as a complex disease in which metabolic pathways play a fundamental role, either for the growth, metastases or resistance to treatments. It becomes evident that targeting these pathways is necessary to develop effective and resolute strategies. Here we provide an example in which the co-targeting of PPP and tyrosine kinase receptors lead to synergistic effects on breast cancer cells. This approach could be easily translated into clinical setting.
We would like to thank Dr. Paola Stiuso for kindly providing Polydatin as well as Prof Gianpietro Ravagnan for productive discussions. Moreover, we thank the association ARTOI “Associazione Ricerca Terapie Oncologiche Integrate” for supporting travel expenses to VD for the participation to the 1st Annual World Congress of Digestive Deseases-2015, Nanjing, China in December 2015. The collaboration between the University of Campania “Luigi Vanvitelli” and the Nottingham Trent University have been supported by the project CA16122 – BIONECA.
This study was supported by a grant from the "Programma VALERE 2017" of the University of Campania "Luigi Vanvitelli" (Naples, Italy).
Availability of data and materials
In this current study 2 publicly available datasets, namely METABRIC and Pawitan, were used.
The METABRIC dataset is available at the European Genome-phenome Archive under the accession number EGAS00000000098 (https://www.ebi.ac.uk/ega/studies/EGAS00000000098).
The Pawitan dataset is available the Gene Expression Omnibus data repository under the accession number GSE1456 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE1456).
Single genes of both datasets can be accessed through CANCERTOOL (http://web.bioinformatics.cicbiogune.es/CANCERTOOL/BCaBasicAnalyses.html).
LM, MLN, AL and F. Paino performed the experiments; LM, VT and VD designed the study; TR, MC, TR, SW and F. Papaccio interpreted the data; LM and VD wrote the manuscript; TR and GP proofed the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 16.Shi H, Zhang W, Zhi Q, Jiang M. Lapatinib resistance in HER2+ cancers: latest findings and new concepts on molecular mechanisms. Tumour Biol. 2016.Google Scholar
- 48.Emde AM, Maslak K, Liu H, Reles AE, Possinger K, Eucker J. Combination of fulvestrant and lapatinib in non-HER2-overexpressing and adriamycin-resistant breast cancer cell lines. J Clin Oncol. 2007;25(18_suppl):14050.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.