CIB1 depletion with docetaxel or TRAIL enhances triple-negative breast cancer cell death
Patients diagnosed with triple negative breast cancer (TNBC) have limited treatment options and often suffer from resistance and toxicity due to chemotherapy. We previously found that depleting calcium and integrin-binding protein 1 (CIB1) induces cell death selectively in TNBC cells, while sparing normal cells. Therefore, we asked whether CIB1 depletion further enhances tumor-specific killing when combined with either the commonly used chemotherapeutic, docetaxel, or the cell death-inducing ligand, TRAIL.
We targeted CIB1 by RNA interference in MDA-MB-436, MDA-MB-231, MDA-MB-468, docetaxel-resistant MDA-MB-436 TNBC cells and ME16C normal breast epithelial cells alone or combination with docetaxel or TRAIL. Cell death was quantified via trypan blue exclusion using flow cytometry and cell death mechanisms were analyzed by Western blotting. Cell surface levels of TRAIL receptors were measured by flow cytometry analysis.
CIB1 depletion combined with docetaxel significantly enhanced tumor-specific cell death relative to each treatment alone. The enhanced cell death strongly correlated with caspase-8 activation, a hallmark of death receptor-mediated apoptosis. The death receptor TRAIL-R2 was upregulated in response to CIB1 depletion, which sensitized TNBC cells to the ligand TRAIL, resulting in a synergistic increase in cell death. In addition to death receptor-mediated apoptosis, both combination treatments activated a non-apoptotic mechanism, called paraptosis. Interestingly, these combination treatments also induced nearly complete death of docetaxel-resistant MDA-MB-436 cells, again via apoptosis and paraptosis. In contrast, neither combination treatment induced cell death in normal ME16C cells.
Novel combinations of CIB1 depletion with docetaxel or TRAIL selectively enhance naive and docetaxel-resistant TNBC cell death while sparing normal cell. Therefore, combination therapies that target CIB1 could prove to be a safe and durable strategy for treatment of TNBC and potentially other cancers.
KeywordsCIB1 TRAIL Apoptosis Triple-negative breast cancer Chemoresistance
triple-negative breast cancer
protein kinase B
extracellular regulated kinase
calcium integrin-binding protein 1
tumor necrosis factor-related apoptosis-inducing ligand
epidermal growth factor receptor
mitogen-activated protein kinase
death receptor 4/5
ALG-2-interacting protein X
short hairpin RNA
fluorescence-activated cell sorting
microtubule-associated proteins 1A/1B light chain 3B
phosphorylated receptor-interacting serine/threonine-protein kinase 1
cellular FLICE-inhibitory protein
X-linked inhibitor of apoptosis
inhibitor of apoptosis
glyceraldehyde 3-phosphate dehydrogenase
poly (ADP-ribose) polymerase
tetraethyl benzimidazolyl carbocyanine iodide
Approximately 15–20% of all breast cancer deaths in the U.S. occur in patients diagnosed with triple-negative breast cancer (TNBC), a subtype defined by a lack of the estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (Ref. ). Due to lack of these targetable cell surface receptors, radiation, surgery, and chemotherapy remain the current standard of care [1, 2]. While a subset of TNBC patients respond initially to chemotherapy, they often suffer from toxicity and acquired resistance, resulting in cancer recurrence and metastasis [2, 3, 4]. Therefore, there is a critical need for efficacious therapies that limit toxicity and overcome resistance for TNBC patients.
Recent efforts to improve clinical efficacy of chemotherapies have shifted towards combining them with targeted approaches to lower effective chemotherapeutic doses while maintaining therapeutic response [1, 2]. For instance, in clinical trials that targeted the epidermal growth factor receptor (EGFR), which is commonly upregulated/activated in TNBC, there was no improvement in chemotherapeutic efficacy [5, 6, 7]. Moreover, while ongoing clinical trials are testing PI3K–AKT and MEK–ERK signaling pathways downstream of EGFR, toxicity remains unresolved with a modest to no increase in patient survival (NCT02423603, NCT01964924).
Our lab previously found that calcium and integrin-binding protein 1 (CIB1), an intracellular protein that regulates highly oncogenic PI3K–AKT and MEK–ERK signaling, may represent a viable target in TNBC [8, 9]. We showed that CIB1 depletion simultaneously inhibits AKT and ERK activation and induces significant cell death in approximately 70% of TNBC cell lines tested and in an in vivo xenograft model . Common to TNBC cell lines that are sensitive to CIB1 depletion is elevated AKT activation. Thus, CIB1 depletion does not cause cell death in TNBC and normal cell lines that exhibit low basal AKT activity . Here, we evaluated cell death in TNBC versus normal breast epithelial cell lines using novel combination treatments involving CIB1 depletion with and without the commonly used chemotherapeutic docetaxel. We found that CIB1 depletion enhanced docetaxel-induced cell death selectively in TNBC over normal cells largely via a death receptor-mediated, as opposed to mitochondrial-mediated apoptosis.
TNF-related apoptosis-inducing ligand (TRAIL), a death receptor ligand, was once thought to be a promising anti-cancer agent because of its selectivity for killing tumor but not normal cells; however, it was not pursued due to innate or acquired resistance driven by dysfunctional TRAIL receptors [10, 11]. Interestingly, there is a growing interest in combining chemotherapeutic drugs or targeted approaches with TRAIL to overcome this resistance [12, 13, 14]. Here we find that CIB1 depletion in combination with the death receptor ligand TRAIL potentiates TNBC-selective cell death, likely due to the upregulation of TRAIL receptor-2 (TRAIL-R2/DR5) in CIB1-depleted TNBC cells. Thus, TRAIL in combination with CIB1 depletion may represent a novel mechanism to sensitize TRAIL-resistant TNBC cells.
Chemotherapy not only causes toxicity, but also often leads to cancer recurrence . Since recurrence driven by drug resistance is often associated with dysfunctional apoptotic mechanisms , one approach to circumvent resistance is to induce non-apoptotic cell death [4, 16]. Here, we identified paraptosis as a likely non-apoptotic mode of cell death induced by CIB1 depletion, due to the observed cellular swelling and intracellular vacuolization that are morphological hallmarks of paraptosis [17, 18]. Currently, there are no well-defined molecular mechanisms known to regulate paraptosis. However, the intracellular protein ALG-2-interacting protein X (Alix) has been shown to inhibit paraptosis by preventing cytoplasmic vacuolization, potentially by regulating endosomal sorting and fusion with other organelles [17, 19]. Separate studies showed that known inducers of paraptosis such as withaferin A and reactive oxygen species decrease the expression of Alix in breast cancer cells [18, 20]. Therefore, loss of Alix expression is considered a molecular marker of paraptosis [17, 18, 20, 21]. In addition to downregulation of Alix, insulin-like growth factor I receptor (IGF-1R) tyrosine kinase and JNK activity were found to induce paraptosis . Here we report that CIB1 depletion alone or in combination with docetaxel/TRAIL not only restores apoptotic signaling but also induces paraptosis in docetaxel-resistant TNBC cells. Combination therapies involving CIB1 targeting could therefore provide safe and durable strategies for treatment of TNBC and potentially other cancers.
Materials and methods
The human triple-negative breast cancer cell lines (MDA-MB-436 [Perou Lab, UNC], MDA-MB-468 [UNC Lineberger Tissue Culture Facility], and MDA-MB-231 (Otey Lab, UNC) were cultured in Dulbecco’s modified eagle medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (Gemini) and 1% MEM non-essential amino acids (Gibco), with the addition of 10 μg/ml insulin for MDA-MB-436 cells. The human normal breast epithelial cell line (ME16C, Perou lab) was cultured in MEBM (Lonza). All cells were maintained at 37 °C in a humidified atmosphere of 5% CO2.
Generation of docetaxel-resistant MDA-MB-436 TNBC cell line
Docetaxel-resistant MDA-MB-436 cells (MDA-436-DCXR) were established by culturing in growth media supplemented with increasing concentrations of docetaxel (5 nM to 50 nM) over a 9-month period. Cells at approximately 40–50% confluency were incubated with a selected concentration for 48 h followed by recovery in drug-free growth media. The concentration of docetaxel was incrementally increased until the cells were resistant, at a final concentration of 50 nM docetaxel (MDA-436-DCXR). Parental MDA-MB-436 cells (MDA-436-PR) were cultured and passaged in parallel in drug-free growth media to control for extended time in culture.
Docetaxel (Tocris), recombinant TRAIL (PeproTech), z-VAD-fmk (Enzo Life Sciences), TRAIL-R2 (DR5) neutralizing antibody (R&D system), and human control IgG (Jackson Immuno Research) were used to treat the cell lines.
CIB1 targeting via RNA interference in the absence and presence of docetaxel or TRAIL
Construction of control and CIB1 shRNA lentiviral plasmids has been described previously [8, 9]. Two different CIB1 targeting sequences (shCIB1-1 and shCIB1-2) were used to validate shRNA-induced CIB1 depletion. To knock down CIB1, cells of interest at ~ 30% confluence were infected with lentiviral shRNA containing 6 µg/ml polybrene (Sigma) for 16–18 h followed by replacement with fresh growth media. After 24 h, the shRNA-infected cells were treated with vehicle, docetaxel, or TRAIL for an additional 48 h. Treated cells were then harvested 4 days post-infection, during which time transduction efficiency exceeds 90% as determined by GFP fluorescence.
Trypan blue exclusion assay
Cell death was determined by trypan blue exclusion using automated fluorescently activated cell sorting (FACS) counting. Aliquots of 10 μl of both adherent and floating cell populations were collected and stained with 0.004% trypan blue (1:10 dilution). Stained dead cells were selected and counted using the Per-CP (APC) channel . Quantification of total live and dead cell populations was determined using an Accuri C6 Flow Cytometer (BD Biosciences) and data are presented as percent cell death.
TRAIL-R1 and -R2 fluorescence detection
To detect surface expression of TRAIL-R1 and -R2, 2 × 105 cells were detached using 2 mM EDTA, washed in PBS, and resuspended in PBS (0.1% BSA) containing antibodies for TRAIL-R1/DR4, TRAIL-R2/DR5 (eBiosciences) or IgG control. AlexaFluor 647 (Invitrogen) secondary antibody was used to detect the levels of TRAIL-R1/2, and mean fluorescence intensity was measured via Accuri C6 Flow Cytometer.
Cells were harvested and lysed with buffer containing 10 mM CHAPS as previously described . Equal amounts of protein based on total cell number were separated by SDS-PAGE, transferred to PVDF membrane, and incubated with primary antibodies overnight at 4 °C. Secondary HRP-conjugated antibodies against rabbit, mouse, and chicken were then used to visualize the immunoblots via enhanced chemiluminescence (ECL2, Pierce). The following antibodies were used in this study: PARP, GAPDH, cleaved caspase-8, caspase-8, cleaved caspase-9, caspase-9, TRAIL-1/DR4, TRAIL-R2/DR5, IGF-1R, phospho-JNK, and total JNK (Cell Signaling), CIB1 (chicken polyclonal antibody, ), vinculin (Sigma), Alix (Biolegend).
Mitochondrial membrane potential detection
To detect changes in mitochondrial membrane potential, cells were stained with JC-1 (Thermofisher), a cationic dye that accumulates in the mitochondria. Cells (1 × 106) were dissociated and incubated in PBS containing JC-1 (5 µM) for 30 min at 37 °C. JC-1 accumulates in functional mitochondria to form aggregates that fluoresce red. In dysfunctional mitochondria with decreased membrane potential, JC-1 will instead remain as monomers that fluoresce green. After washing with PBS, the emission of JC-1 fluorescence was analyzed using an Accuri C6 Flow Cytometer and associated software (BD Biosciences). Therefore, calculation of red:green JC-1 fluorescence ratio was used as the surrogate for changes in mitochondrial membrane potential.
To assess morphological changes induced by CIB1 depletion in the absence and presence of docetaxel or TRAIL, cells were imaged in 6-well plates at the conclusion of each experiment. Differential interference contrast (DIC) images were captured using a Nikon TE300 microscope equipped with an Andor Zyla sCMOS camera (20x objective).
The statistically significant differences were determined using either a one-way ANOVA or student’s two-tailed t-test. More details on p-values are indicated in the figure legends.
CIB1 depletion selectively enhances docetaxel-induced TNBC cell death
We previously reported that CIB1 depletion leads to TNBC cell death while sparing normal cells , suggesting that CIB1 may be a viable, safe target. To address the need to enhance tumor cell death and maintain tolerability by normal cells, we tested the targeting of CIB1 via RNA interference in combination with the commonly used chemotherapeutic agent, docetaxel, in both TNBC and a normal breast epithelial cell line. Two different CIB1 shRNA sequences (shCIB1-1 and shCIB1-2) were used to validate CIB1 depletion (Additional file 2: Figure S1A). CIB1 depletion-induced cell death quantification, cell death signaling, and cellular morphology were also validated using both shRNA sequences (Additional file 2: Figure S1B–H) and Western blots were quantified in Additional file 1: Table S1. Since similar results were observed with both shRNA sequences, we chose CIB1 shRNA-1 for the remainder of the experiments.
To elucidate the mechanism underlying the enhanced TNBC cell death induced by CIB1 depletion plus docetaxel (Fig. 1a–c), we analyzed cell lysates for cleaved PARP, a marker of apoptosis. In agreement with the observed enhanced cell death, the combined treatment-induced PARP cleavage was only observed in TNBC (Fig. 1a–c) but not normal breast epithelial ME16C cells (Fig. 1d). CIB1 knockdown was confirmed by Western blotting (Fig. 1a–d). Collectively, our results indicate that the novel combination of CIB1 depletion and docetaxel selectively enhances TNBC cell death via increased apoptotic signaling.
CIB1 depletion with docetaxel induces death receptor-mediated apoptosis
CIB1 depletion sensitizes TNBC, but not normal cells to TRAIL
Interestingly, we also observed a similar increase in caspase-8 activation and cell death upon addition of TRAIL to CIB1-depleted MDA-468 (Additional file 5: Figure S4A) but not MDA-231 TNBC cells (Additional file 5: Figure S4B). This suggests that the cell death induced by the combination of CIB1 depletion and TRAIL, unlike that induced by docetaxel, is cell-type specific based on sensitivity to CIB1 depletion. Thus, adding TRAIL to CIB1 depletion-sensitive cells is a potent combination for inducing caspase-8 activation and cell death.
TRAIL-R2 upregulation by CIB1 depletion sensitizes TNBC cells to TRAIL
CIB1 depletion plus docetaxel or TRAIL induces caspase-independent cell death
Docetaxel-resistant TNBC cells remain sensitive to CIB1 depletion alone or in combination with docetaxel or TRAIL
Since another strategy to circumvent chemo-resistance is to induce non-apoptotic, caspase-independent cell death , we asked whether paraptosis also occurs in docetaxel-resistant cells. Indeed, we found that either CIB1 depletion alone or in combination with docetaxel or TRAIL decreased Alix levels in MDA-436-DCXR cells (Fig. 6c). Moreover, paraptotic morphologies such as swelling and intracellular vacuole formation were observed in resistant cells in response to CIB1 depletion alone or in combination with docetaxel/TRAIL (Fig. 6d). The morphology of CIB1-depleted cells mimicked the effects of docetaxel (Fig. 6d), a parental compound to paclitaxel, which is known to cause paraptotic morphology at high doses . Although the combination treatments did not activate the JNK pathway, CIB1 depletion alone led to upregulation of IGF-1R, an upstream effector of paraptosis, in docetaxel-resistant TNBC cells similar to that observed in parental cells (Additional file 7: Figure S6E). These results collectively show that the combination treatments both restore apoptosis and induce paraptosis as potential mechanisms to overcome chemo-resistance.
The standard of care for TNBC patients, chemotherapy and/or surgery, often fails due to development of resistance and toxicity [3, 28]. To address this need, clinical trials have centered on combining targeted therapies with chemotherapeutics, yet resistance and toxicity persist to limit the overall efficacy [2, 4, 5, 6]. Our previous work indicated that CIB1 may be a potentially safe target due to its selectivity for killing TNBC but not normal cells when depleted . Therefore, we tested a novel combination treatment with CIB1 depletion and a commonly used chemotherapeutic, docetaxel to enhance TNBC-selective cell death.
In this study, we focused on detecting and understanding the mechanisms of cell death to assess the combination of CIB1 depletion and chemotherapeutics. Many previous pre-clinical cell culture-based TNBC studies that combine targeted agents with chemotherapeutics have focused primarily on slowing cell proliferation as opposed to cell death [29, 30, 31, 32]. Moreover, studies that have quantified cell death were often limited to studying classical caspase-dependent apoptosis [33, 34, 35, 36]. In contrast, we investigated both apoptotic and non-apoptotic TNBC cell death in response to CIB1 depletion alone or in combination with docetaxel. In addition, we demonstrate that induction of non-apoptotic cell death is an effective strategy to circumvent resistance, which is often associated with dysfunctional apoptotic signaling [16, 37, 38]. Thus, we propose that quantification of cell death in tumor versus normal cells may identify combinations that eradicate tumors while sparing normal cells, which may be a better indicator for a safer, more efficacious therapy.
Initially, we found that combining CIB1 depletion with docetaxel significantly enhances TNBC cell death while sparing normal breast epithelial cells. The enhanced cell death correlated with increased death receptor-mediated (caspase-8) apoptotic signaling, selectively in TNBC relative to normal cells. Subsequently, we found that the combination of CIB1 depletion and the death receptor ligand TRAIL is potent in selectively killing TNBC cells via increased caspase-8 activation. While TNBC cells are inherently resistant to TRAIL alone, CIB1 depletion appears to sensitize these cells to TRAIL by upregulating TRAIL receptor-2. Once believed to be a safe, tumor-specific targeting therapy, TRAIL as a mono-therapy failed due to resistance driven by mutations in TRAIL receptors and dysfunctional signaling complexes at the receptor intracellular domain [10, 11, 39]. Subsequent studies have focused on combination approaches using chemotherapeutics, natural compounds and targeted agents to sensitize otherwise resistant breast cancer cells to TRAIL by upregulating TRAIL receptors and restoring TRAIL receptors’ intracellular signaling complex activity [14, 40, 41, 42, 43, 44, 45]. In comparison, our approach of targeting CIB1 allowed us to use five to tenfold lower concentrations of TRAIL to induce as much or greater cytotoxicity in cultured TNBC cells. The selectivity of TRAIL-based combination treatments for tumor cells is consistent between our results and the published results of other studies [14, 43, 45]. However, those studies provide limited understanding as to why normal cells are unaffected by TRAIL-based combination treatments. We speculate that CIB1 depletion may upregulate decoy receptors, which inhibit intracellular TRAIL receptor signaling and initiate pro-survival NF-κB signaling [46, 47] (data not shown), thereby protecting normal breast epithelial cells from TRAIL-induced cell death. Another mechanism may involve the internalization of TRAIL receptors and expression of intracellular signaling complex modulators such as c-FLIP, XIAP, and IAP in normal cells, to explain TNBC-selective cell death. We believe that a better understanding of cell death mechanisms underlying tumor selectivity versus normal cells could provide further rationale to test a TRAIL-based combination treatment with CIB1 targeting.
The induction of non-apoptotic death could prove to be an effective therapeutic strategy to circumvent resistance [16, 38, 48, 49]. In addition to previously identifying GAPDH nuclear translocation , we found paraptosis as another mode of non-apoptotic cell death associated with CIB1 depletion. To our knowledge, we are the first to report paraptosis as a mechanism for overcoming chemoresistance in TNBC cells. While autophagy is another mode of non-apoptotic cell death that shares phenotypic characteristics similar to paraptosis, we did not observe activation of LC-III, a marker of autophagy. Necroptosis, like apoptosis, can be activated downstream of death receptors to cause a caspase-independent paraptotic-like, necrotic phenotype . However, we found no effect on the necrotic marker p-RIPK or the necroptosis inhibitor necrostatin-1 and therefore concluded that CIB1 depletion alone or in combination with either docetaxel or TRAIL does not induce necroptosis. Because pre-treatment with cycloheximide, a protein synthesis inhibitor, rescued CIB1 depleted TNBC cells from cell death, protein synthesis may be investigated to further understand non-apoptotic cell death in relation to CIB1. Future studies examining non-apoptotic modes of cell death associated with CIB1 targeting will provide additional insight into mechanisms to circumvent chemo-resistance.
AHC: study conception and design and data acquisition and interpretation and manuscript writing; TML: study conception and design and manuscript review; GJD, MMB, ALK: data acquisition and interpretation; LVP: study supervision and manuscript review. All authors read and approved the final manuscirpt.
We thank Dr. Wolfgang Bergmeier for sharing his flow cytometer (Accuri C6, BD Biosciences) and the software to perform fluorescence cell death analysis. Dr. Perou, Dr. Otey, and UNC Lineberger Tissue Culture Facility kindly provided the TNBC cell lines tested. We also thank Karel Alcedo for establishing the automated trypan blue exclusion assay via flow cytometry.
TM Leisner and LV Parise are co-founders of Reveris Therapeutics, LLC. The other authors declare no competing interests.
Availability of data and materials
All data including additional information generated or analyzed during this study are included in this article. All original data are available upon request.
Consent for publication
Ethics approval and consent to participate
This work was supported by Grant 2017BIG6515 from the North Carolina Biotechnology Center, R41CA200189 from the NIH, and a fellowship from the Graduate School at UNC Chapel Hill.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis 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.