Background
Triple-negative breast cancer (TNBC) represents approximately 15 % of all breast cancer and is a subtype for which therapeutic options are more limited due to lack of effective targeted therapies. Triple-negative breast cancer is defined by the lack of expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (Her2), but it is genetically and genomically diverse. Due to the lack of targeted therapies and intrinsic biological aggressiveness, novel therapeutic strategies, beyond standard-of-care chemotherapy, are desperately needed.
We first identified protein tyrosine kinase C theta isoform (PRKCQ)/PKCθ as a candidate regulator of anchorage-independent survival of breast cancer cells in a functional kinome screen [
1]. PRKCQ/PKCθ is a member of the novel protein kinase C (PKC) family characterized by a unique protein domain structure consisting of diacylglycerol binding sites, but lacking calcium (Ca+) binding sites typical of classical PKCs. PRKCQ maps to Chromosome 10p15, a region frequently mutated in T cell leukemia, lymphoma and T cell immunodeficiency (as reviewed in [
2]). PRKCQ/PKCθ is widely expressed throughout the hematopoietic system, primarily in T cells, mast cells, natural killer (NK) cells and platelets, and in skeletal muscle, liver, thymus, and the nervous system [
3‐
6].
Much of the known isoform-specific functions of PRKCQ/PKCθ are in the context of immune function; mice deficient in PRKCQ expression exhibit defects in T cell activation due to impaired Ca+ signaling and nuclear factor of activated T cells (NFAT) activation [
7,
8]. PRKCQ/PKCθ also regulates the survival of T cells by regulating the expression of pro-apoptotic and anti-apoptotic B cell lymphoma 2 (Bcl2) family members [
2,
8,
9]. More recent evidence supports a relative specific role for PRKCQ in immune response; it is mostly dispensable for immunity against viral and bacterial pathogens [
2,
10‐
12]. In contrast, PRKCQ appears to be required for immune responses associated with autoimmune disease and allograft rejection, perhaps due to a specific requirement for PRKCQ in the maturation of T helper (Th)17 cells, a subset of CD4+ T cells [
13‐
16].
In contrast, the functions of PRKCQ in non-hematopoietic tissue and in cancer have not yet been fully elucidated. PRKCQ is reported to be expressed in solid tumors including gastrointestinal stromal tumors (GIST) and more recently, in breast cancer, specifically ER-negative tumors [
17‐
20]. In fact, PRKCQ expression directly suppresses the expression of ERα in breast cancer cells and is required for c-rel-induced mammary tumorigenesis [
20]. PRKCQ also stimulates breast cancer cell migration by stabilizing the expression of Fra-1 in TBNC cells [
21]. Furthermore, PRKCQ may contribute to the formation or maintenance of a breast cancer stem cell population by promoting the expression of genes associated with epithelial-mesenchymal transition (EMT) through direct chromatin interactions [
22,
23].
The role of PRKCQ/PKCθ in the proliferation and survival of breast cancer cells and the responsible mechanisms, including dependency on kinase activity, remain to be clarified. Here we show that PRKCQ is sufficient to drive growth-factor-independent proliferation, migration and anoikis resistance of breast epithelial cells (MCF-10A). PRKCQ promotes proliferation by activating extracellular signal-related kinase (Erk)/mitogen-activated protein kinase (MAPK) activity in a kinase-activity-dependent manner. PRKCQ is not only sufficient to promote these phenotypes in MCF-10A cells, but, we showed for the first time that it is required for in vitro and in vivo growth of a subset of TNBC cells. These studies support PRKCQ/PKCθ as an attractive candidate therapeutic target in TNBC.
Methods
Reagents, cells, and cell culture
MDA-MB-231-luc-D3H2LN cells were obtained from Perkin Elmer. Cells were cultured in MEM with Earle’s Salts supplemented with non-essential amino acids, GlutaMAX™, sodium pyruvate, penicillin/streptomycin (P/S; Life Technologies) and 10 % heat-inactivated fetal bovine serum (FBS) (Life Technologies) at 37 °C in 5 % CO2. All other cell lines were purchased from ATCC (Manassas, Virginia). MDA-MB-436 cells were cultured in Leibovitz’s L-15 media supplemented with 10 ug/mL insulin, 16 ug/mL glutathione, P/S, and 10 % FBS at 37 °C under atmospheric conditions. MCF10A cells were cultured in 1:1 DMEM/F12 supplemented with 5 % horse serum, 20 ng/mL EGF (Peprotech), 500 ug/mL hydrocortisone (Sigma-Aldrich), 100 ng/mL cholera toxin (Sigma-Aldrich), 10 ug/mL insulin (Sigma-Aldrich), and penicillin/streptomycin (Life Technologies). HCC1806 and HCC38 cells were cultured in RPMI-1640 media supplemented with 10 % FBS and P/S. AEB071 was purchased from Selleckchem.
Antibodies directed against the following proteins were obtained from the indicated suppliers: AbCam: rabbit monoclonal protein kinase C (PKC-θ) (EPR1487(2)); Cell Signaling Technologies: phospho-retinoblastoma (Rb) (S807/811) and total Rb, phospho-p44/p42 MAPK (ERK1/2) (T202/Y204) and total p44/42 MAPK (ERK1/2), phospho-Akt (S473) and total (pan-) Akt, PKC isoform antibody sampler kit (for PKCδ, PKCα, PKCμ, PKCζ) phospho-PKCθ (T538), phospho-PKCα/βII (T638/641), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and beta-tubulin. Additional GAPDH antibody from Santa Cruz Biotechnology was used.
Constructs, viral production, and stable cell line generation
PRKCQ complementary DNA (cDNA) encoding retroviral vector was obtained from Joshua LaBaer (Harvard Institute of Proteomics). Mutagenesis to create PRKCQ mutants was performed using QuikChange Lightning Site-Directed Mutagenesis Kit (Stratagene) and the following primers: 5′ GCATCAGCGCCGGGGTGAAATCAAGCAGGCAAAGGTC-3′ for A148E (constitutively active) and 5′ CCAATCAATTTTTCGCAATAAGGGCCTTAAAGAAAGATGTGG-3′ for K409R (catalytically inactive) [
24]. Retrovirus was generated using 293-GPG cells (gift of R. Mulligan) and according to protocols described previously [
25].
Constructs encoding short hairpin RNA (shRNA) sequences targeting PRKCQ (TRCN0000001790, TRCN0000199654 and TRCN0000197216, referred to as 90, 54 and 16, respectively) were purchased from Open Biosystems/Thermo Scientific. Viral packaging 293 T cells were transfected according to standard protocols to produce lentiviral particles, as described previously [
25]. Viral supernatant was collected 24, 48, 72, and 96 hours post-transfection, pooled, and concentrated using Amicon Ultra Centrifugal Filter Units (Millipore). MDA-231-luc or MDA-436 cells were spin-infected in the presence of 2 ug/mL polybrene (Sigma-Aldrich) and concentrated viral supernatant at 2250 rpm for 30 minutes. Cells were exposed to viral supernatant overnight before changing to complete media and were allowed to recover for 24 hours.
Anoikis and transwell migration assays
Anoikis (anchorage-independent viability) assay was performed by culturing cells in suspension on polyhema-coated plates for the indicated amount of time, and cell death was assessed using the Cell Death ELISA
PLUS Kit (Roche) according to manufacturer’s instructions. Transwell migration assay was performed as described previously [
25].
Proliferation assay/growth curves
Cells were plated at a density of 1 × 10^4 cells per well in triplicate in a 24-well plate. Cells were harvested and counted using a hemacytometer at the indicated number of days after seeding.
3-D culture
Three-dimensional (3-D) Matrigel™ (BD Biosciences) cultures were performed as described previously using 3000–5000 cells/well [
25].
Cell cycle analysis/flow cytometry
Cells were plated at subconfluent densities, harvested using 0.05 % Trypsin/EDTA (Life Technologies), washed with PBS, resuspended as single cells in cold PBS and fixed with ice-cold ethanol at a final concentration of 70 %. After fixation cells were stored at 4 °C until staining. Cells were stained using propidium iodide (PI)/RNase Staining Buffer (BD Biosciences) at 37 °C for 30 minutes before sample processing. DNA content was assessed by PI staining using BD Canto and FACSDiva. Cell cycle profiles were analyzed using Flow Jo (version 9.0). Percentages were calculated using the Watson pragmatic algorithm. Significance was calculated using the standard Student t test.
In vivo tumor xenograft models
Female nude mice (nu-/-) were obtained from Jackson Laboratories. At age 6–8 weeks, 5 × 10^5 MDA-231-luc cells per mouse were injected subcutaneously in a total volume of 100 uL of complete media 48 hours after infection with PRKCQ shRNA lentiviral particles. Tumor dimensions were measured with calipers and the volume was calculated as (L x W2)/2. Stastical significance was calculated using the Whitney-Mann-Wilcoxon rank sum test. All procedures and studies with mice were performed in accordance with protocols pre-approved by the Institutional Animal Care and Use Committee of Mount Sinai.
PRKCQ transcript expression analysis in breast tumors
The Cancer Genome Atlas (TCGA) dataset
Level-3 expression IlluminaHiSeq-RNASeqV2 expression data were downloaded from the TCGA data portal [
26] and processed for quality control as follows: log(x + 1) transformation was performed to rescale the expression data, followed by quantile-normalization, using normalize.quantiles() from R package “preprocessCore”. The quantile-normalized data were split for tumor and normal tissue samples. Correction for batch effects was performed using batch ID, tissue source site ID, center ID and plate ID, where batch ID was obtained from TCGA biospecimen files, and other IDs were obtained from TCGA barcode. Batch and age corrections were performed using the linear regression (lm()) function in the statistical computing software R, for each gene expression profile, thereby removing discrepancy between different batch IDs, and preserving the overall mean across all samples. Expression of PRKCQ was then extracted and patients were classified as receptor positive (ER, PR, or Her2 positive,
n = 731) or TNBC (ER, PR, and Her2 negative,
n = 86). The significance of difference in log expression was tested using the one-sided Student
t test.
METABRIC-normalized Illumina HT12v3 data were downloaded from the European Bioinformatics Institute, quantile-normalized, and corrected for age [
27]. Samples were stratified as TNBC or receptor-positive as follows: samples with negative expression of ER, PR, and Her2, as reported by Curtis et al. [
27] in the columns “ER.Expr,” “PR.Expr”, and “Her2.Expr,” respectively, and not classified as luminal A, luminal B, or Her2 by PAM50 subtyping, also reported by Curtis et al. [
27], were labeled TNBC (n = 276); all other samples were labeled receptor-positive (n = 1698). PRKCQ expression was extracted and log expression was compared in the TNBC and receptor-positive samples using the one-sided Student
t test.
Consent statement
We confirm that this study does not involve human patients and no consent was necessary.
Discussion
The functions of PKC family members are varied and diverse. PKC expression or activity is detected in cancer cells and for this reason, inhibitors of the PKC family have been developed and evaluated as potential cancer therapeutic agents. However, their use in cancer therapy has been limited. One potential reason may be the lack of biomarkers and the potential isoform-specific functions of PKC enzymes. While many of the PKC enzymes are associated with pro-tumorigenic activities, there is also evidence that supports possible tumor-suppressive roles for some PKC family members. For example, mutations detected in PKC enzymes in patient tumors have been reported to have growth-inhibitory properties in vitro [
30]. Therefore, the precise functions of specific PKC isoforms in human cancer need to be clearly elucidated for optimal development of PKC isoform-targeted therapeutic agents.
PRKCQ/PKCθ has been extensively studied in the context of hematopoiesis and immunity, and is required for the development and maturation of specific T cell subsets. PRKCQ also plays a role in Notch-driven T cell leukemia [
31]. However, its function in epithelial cancer has been relatively underexplored. PRKCQ is expressed or activated in a subset of GIST, uveal melanoma, and breast cancers [
17,
19,
20,
32]. Here, we show that PRKCQ is sufficient to promote oncogenic activities such as anoikis resistance and growth-factor-independent migration and proliferation. Our studies are also the first to demonstrate the isoform-specific requirement for endogenous PRKCQ in the in vitro and in vivo growth of a subset of human TNBC cell lines. PRKCQ/PKCθ expression is relatively greater in human triple-negative tumors compared to other subtypes, further supporting PRKCQ/PKCθ as a candidate therapeutic target for this subtype.
Our studies highlight the critical role of PKCθ kinase activity in tumorigenic phenotypes; kinase activity is required to enhance growth-factor-independent cell-cycle progression of MCF-10A cells via Erk/MAPK activity and Rb phosphorylation, and inhibition of PKCθ kinase activity impairs the growth and invasiveness of TNBC cells. These results support the possibility of targeting PRKCQ kinase activity as a therapeutic approach for a subset of TNBC.
Although our studies are the first to show a critical role for endogenous PRKCQ in the growth and survival of a subset of human TNBC cells, our data complement previous work that suggested a role for PRKCQ in ER-negative breast cancers. Belguise et al. report that PRKCQ suppresses ER transcription via Akt activation and Forkhead inhibition, and PRKCQ is required for c-Rel-driven, ER-negative development of a mouse mammary tumor [
20]. PRKCQ also promotes the stabilization of Fra-1, a member of the Fos transcription factor family in ER-negative breast cancer [
21].
Interestingly, in addition to its role in regulating signaling in ER-negative breast cancer, PRKCQ has been shown to localize to the nucleus where it directly interacts with chromatin complexes to induce expression of genes associated with an epithelial-to-mesenchymal transition, and markers associated with a cancer stem cell phenotype [
23]. Kinase activity of PKCθ was required for this induction of mesenchymal gene expression. These data and our own results showing the kinase-activity-dependent activation of Erk/MAPK signaling by PRKCQ highlight the potentially multiple ways by which PRKCQ kinase activity may promote oncogenic activity in TNBC cells via signaling and transcriptional mechanisms in specific subcellular compartments.
The recent synthesis of PRKCQ/PKCθ small molecule kinase inhibitors raises the attractive possibility of using these to therapeutically target PRKCQ in TNBC. These inhibitors are highly specific for PRKCQ, relative to other PKC isoforms, and have been developed for the treatment of autoimmune or inflammatory disease [
33]. The kinase dependency of PRKCQ oncogenic activity supports the evaluation of these inhibitors in the context of breast cancer models. Given the expression of PRKCQ in the cells of the immune system, the effect of PRKCQ inhibition on host immunity must be considered. However, as PRKCQ-/- mice appear to maintain normal immune responses to most bacterial and viral pathogens, it is unlikely that PRKCQ inhibition will significantly compromise the host defense mechanisms reviewed in [
2]. Interestingly, PRKCQ is specifically required for the maturation and development of Th17 cells, which have been implicated in autoimmune diseases, such as multiple sclerosis and arthritis [
14,
15]. Although the role of Th17 cells in cancer development or progression is controversial, IL-17 and other cytokines secreted by Th17 T cells have been shown to promote tumorigenesis [
34]. The use of these isoform-specific PRKCQ inhibitors could provide unique insights into the interaction between triple-negative tumor cells, their immune microenvironment and systemic immune responses.
Conclusions
Our studies support PRKCQ/PKCθ as a therapeutic target for TNBC. Enhanced PRKCQ expression is sufficient to promote anoikis resistance, migration, and EGF-independent growth via kinase-activity-dependent stimulation of Erk/MAPK. Inhibition of PRKCQ, either by downregulation of expression or inhibition of kinase activity, suppresses TNBC growth, highlighting the potential of PRKCQ small molecule inhibitor treatment as a therapeutic strategy.
Acknowledgements
HYI was supported in part by a Susan G Komen for the Cure Career Catalyst Award and the Breast Cancer Research Foundation.
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.