Overexpression and Tyr421-phosphorylation of cortactin is induced by three-dimensional spheroid culturing and contributes to migration and invasion of pancreatic ductal adenocarcinoma (PDAC) cells
The nucleation-promoting factor cortactin is expressed and promotes tumor progression and metastasis in various cancers. However, little is known about the biological role of cortactin in the progression of pancreatic ductal adenocarcinoma (PDAC).
Cortactin and phosphorylated cortactin (Y421) were investigated immunohistochemically in 66 PDAC tumor specimens. To examine the functional role of cortactin in PDAC, we modulated cortactin expression by establishing two cortactin knockout cell lines (Panc-1 and BxPC-3) with CRISPR/Cas9 technique. Cortactin knockout was verified by immunoblotting and immunofluorescence microscopy and functional effects were determined by cell migration and invasion assays. A proteomic screening approach was performed to elucidate potential binding partners of cortactin.
Immunohistochemically, we observed higher cortactin expression and Tyr421-phosphorylation in PDAC metastases compared to primary tumor tissues. In PDAC cell lines Panc-1 and BxPC-3, knockdown of cortactin impaired migration and invasion, while cell proliferation was not affected. Three-dimensional spheroid culturing as a model for collective cell migration enhanced cortactin expression and Tyr421-phosphorylation. The activation of cortactin as well as the migratory capacity of PDAC cells could significantly be reduced by dasatinib, a Src family kinase inhibitor. Finally, we identified gelsolin as a novel protein interaction partner of cortactin in PDAC.
Our data provides evidence that cohesive cell migration induces cortactin expression and phosphorylation as a prerequisite for the gain of an invasive, pro-migratory phenotype in PDAC that can effectively be targeted with dasatinib.
KeywordsPancreatic ductal adenocarcinoma Cortactin Tumor cell migration and invasion
actin-related protein 2/3
Bovine serum albumin
Dulbecco’s Modified Eagles’ medium
Deutsche Sammlung von Mikroorganismen und Zellkulturen
fluorescent activated cell sorting
fetal calf serum
green fluorescent protein
phosphate buffered saline
pancreatic ductal adenocarcinoma
polyacrylamide gel electrophoresis
Roswell Park Memorial Institute medium 1640
sodium dodecyl sulphate
Src homology 3
small interfering RNA
Pancreatic ductal adenocarcinoma (PDAC) is one of the most common malignancies worldwide, and the fourth leading cause of cancer death in the world . By the time of diagnosis, in as many as 80% of patients the tumor has already metastasized, leading to a 5-year survival rate of less than 5% . In this context, understanding metastatic progression and the highly invasive potential of PDAC cells has become increasingly important. While single-cell invasion and metastatic dissemination upon epithelial-mesenchymal transition had long been held responsible for the spread of tumor cells throughout the body, alternative and more complex models of metastasis have recently been developed. Among those, three-dimensional reconstruction of histological slides as well as in vitro studies have provided evidence for the hypothesis that in fact many epithelial tumors migrate and invade in a collective manner, forming cohesive tumor cell clusters [3, 4].
Irrespective of a single-cell or cohesive cell mode of invasion, many of the steps involved in metastatic progression require the remodeling of the actin cytoskeleton. The altered expression of key regulatory proteins of the actin cytoskeleton, such as cortactin, contribute to tumorigenesis . Initially, cortactin was identified as a major substrate of v-Src kinase, and was named cortactin due to its localization at cortical actin structures . As a scaffolding protein and an actin-related protein 2/3 (Arp2/3) complex activator, cortactin is involved in various cell functions, including actin polymerization and stabilization of F-actin filaments, the formation of focal adhesions and cell motility structures such as podosomes and invadopodia, and extracellular matrix protein deposition [7, 8]. In a pathological context, these functions can lead to deregulated cell migration and invasion [9, 10].
Elevated cortactin expression has been linked to tumorigenesis, and has been shown to correlate with poor clinical outcome in breast cancer , head and neck squamous cell carcinoma , hepatocellular carcinoma , colorectal adenocarcinoma , and melanoma . In vitro, cortactin overexpression and activation by Src-mediated tyrosine phosphorylation leads to increased migration of fibroblasts and endothelial cells [16, 17]. Phosphorylation of cortactin occurring primarily at tyrosine 421 (Tyr421) , enhances the recruitment of SH2-domain proteins, the activation of Arp2/3 complex and focal adhesion stability and turnover [7, 19, 20]. Tyrosine phosphorylation of cortactin also correlates with invadopodia activity necessary for matrix degradation and cell invasion and is therefore often used as an invadopodial marker .
In PDAC, an invasive cancer with high metastatic potential, remodeling of the actin cytoskeleton, changes in cell–cell adhesions, and enhanced migratory potential all might be based on cortactin-mediated actin dynamics. Previously published data suggests an important role for cortactin in PDAC, indicating an association between high cortactin expression and decreased survival time . However, the specific cellular and molecular mechanisms involved in PDAC cell migration and invasion remain unclear. In our study, we analyzed the role of cortactin and its biological function in the tumor progression of PDAC.
Materials and methods
Patients, tumor specimens, and tissue microarray
Clinicopathological data of PDAC patients
Number of patients (%)
I + II
III + IV
Immunohistochemical staining of 4 µm thick sections (deparaffinized in xylene and rehydrated using standard procedures) was performed with a Bench-Mark ULTRA Autostainer (Ventana). After antigen retrieval and pretreatment with citrate buffer, the primary antibodies (cortactin, polyclonal rabbit, 1:50, #LS-B5768, phospho-cortactin, Tyr421, polyclonal rabbit, 1:50, #LS-C353975, LSBio) were incubated (30 min, RT). After washing, the sections were incubated with biotinylated secondary antibodies. Immunoreactions were visualized using the 3-amino-9-ethylcarbazole chromogen as a substrate (Ventana Optiview DAB IHC detection Kit, Ref: 760-700; Roche Diagnostics). Breast cancer tissue was used as a positive control for cortactin/phospho-cortactin (Tyr421) immunostaining, and omitting the primary antibody from the staining protocol served as negative control. Staining was graded as absent, low (weak staining signal and/or immunoreactivity in < 50% of tumor cells) or high (strong staining signal and/or immunoreactivity in ≥ 50% of tumor cells). Scoring was performed independently by two pathologists (K.S. and J.R.).
Cell lines and reagents
The PDAC cell lines BxPC-3, Capan-1, Capan-2, MIA PaCa-2, Panc-1 and MCF-7 were obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ). AsPC-1, Panc89 and PT45 were generously donated by Prof. Bence Sipos (Tübingen, Germany). All cell lines were maintained in Roswell Park Memorial Institute medium 1640 (RPMI) or Dulbecco’s Modified Eagles’ medium (DMEM), supplemented with 10% FCS (Life Technologies) under standard cell culture conditions (37 °C, 5% CO2) to 70% confluence. For spheroid culture, cells were seeded in six-well plates coated with soft agar. The cells were cultured for 6 days if not indicated otherwise, and the medium was supplemented with FCS every 2 days. Inhibitor treatment was performed with increasing concentrations of dasatinib (Santa Cruz Biotech) and DMSO as vehicle control.
Generation of cortactin knockout cells
Cortactin knockout (CTTN KO) and control cell lines were generated using human CRISPR/Cas9 Cortactin KO Plasmid and CRISPR/Cas9 Control Plasmid (Santa Cruz Biotech). Panc-1 and BxPC-3 cells were transfected with the CRISPR plasmids using lipofectamine 3000 reagent (Thermo Fisher Scientific) according to the manufacturer’s instructions. Twenty-four hours after transfection, transfected cells (GFP+) were isolated from untransfected cells (GFP−) by fluorescent activated cell sorting (FACS). Cells were cultured as single-cell clones, and cortactin KO was confirmed using immunoblotting and immunofluorescence analysis.
Whole cell protein lysates were extracted using cell lysis buffer (Cell Signaling Technology) following the manufacturer’s instructions and protein concentration was determined using the Bradford Assay Kit (Sigma-Aldrich). Immunoblotting was performed according to standard methods. The following primary antibodies were used according to the manufacturer’s instructions: cortactin (#3502), phospho-cortactin Tyr421 (#4569), ERK1/2 (#4695), phoshpho-ERK1/2 Thr202/Tyr204 (#4370), GAPDH clone D16H11 (#5174), Src (#2109) and phospho-Src Tyr416 (#2101), gelsolin D9W8Y (#12953) (all rabbit host species and obtained from Cell Signaling Technology) and E-cadherin NCH-38 (#MA5-12547, mouse, Thermo Fisher Scientific). After Secondary HRP-labeled antibody (rabbit #1706515, mouse #1706516, both obtained from Bio-Rad Laboratories) incubation (1 h, RT) immunolabeling was detected using an enhanced Chemiluminescence Detection Kit (SignalFire ECL Reagent; Cell Signaling Technology) and the Molecular Imager ChemiDoc system (Image Lab Software; Bio-Rad Laboratories).
Cells were seeded in two-well chambers (Sarstedt) with 5–8 × 104 cells per well. After a washing step with cold PBS, cells were immediately fixed with cold PFA (4%) for 15 min. After washing, cells were permeabilized with triton X-100 (0.1%) in PBS for 5 min and blocked with 5% BSA in PBS for 30 min. The primary antibodies (cortactin H222 (#3503, Cell Signaling Technology), phospho-cortactin Tyr421 (LSBio) and gelsolin ERP1942 (#ab109014, Abcam)) were incubated at 4 °C overnight. AlexaFlour-488- or AlexaFlour-568-conjugated secondary antibodies (Invitrogen) were incubated for 2 h at RT. Phalloidin (AlexaFlour-568 labeled, Invitrogen) was added for the last 30 min. The slides were mounted with Vectashield mounting medium with DAPI (Vector Laboratories Inc.).
Cell proliferation was monitored in real-time using the xCELLigence RTCA DP instrument (ACEA Biosciences Inc.). Cells were seeded in duplicates or triplicates in 16-well E-view plates (104 cells/well) and monitored for 90 h. Real-time changes in electrical impedance were expressed as “cell index”, a dimensionless parameter representing the cell number.
Cell migration assay using the xCELLigence system
For migration assays triplicates of 4 × 104 cells were seeded in 130 µl serum-free medium in the upper compartment of the CIM plate of the xCELLigence RTCA DP instrument for real-time cell migration analysis. The lower compartment contained 2.5% FCS to stimulate migration or 0% FCS as a negative control. The cell index representing the amount of migrated cells was calculated using RTCA software from ACEA Biosciences from the change in impedance of the microelectrode sensors when cells adhere to the bottom surface of the microporous membrane. For calculation of the relative cell migration, the cell index after 8 h was normalized to controls and depicted as a bar chart (GraphPad Prism 7, GraphPad Software). Experiments were repeated at least three times. Differences between data points were assessed for statistical significance using two-tailed Student’s t-test comparisons and considered significant at p < 0.05.
In addition, cell migration was measured by performing scratch assays. Cells were grown to full confluence in growth medium. A scratch was made in the cell monolayer using a sterile pipette tip, and cells were washed twice with PBS and cultured in medium supplemented with 2% FCS for up to 18–24 h. Migration was monitored by taking images at the beginning (0 h) and at the predefined cell line-dependent endpoint (BxPC-3, 18 h; Panc-1, 24 h). The percentage of the closure of the gap was analyzed using NIH ImageJ software. Experiments were repeated at least three times and minimum 12 scratches were recorded per experiment.
Collagen invasion assay
Cell invasion was analyzed measuring the spreading of spheroids embedded in collagen after 24 h. Prior to seeding BxPC-3 cells as spheroids (5 × 103 cells/spheroid), cells were stained by DiO (Invitrogen) for 20 min at 37 °C and washed twice with PBS. After overnight culturing, each cell spheroid was harvested and embedded in 40 µl of a solution of 1% methyl cellulose in RPMI medium. Spheroids were then mixed with 60 µl of 3 mg/ml collagen matrix (Rat Tail Collagen Type I, Corning) and seeded in 96-well plates. After incubation for 30 min at 37 °C, the gelled collagen was detached from the well wall with a pipette tip to ensure proper media supply, and wells were filled with 100 µl of complete media. Images were taken at the beginning and after 12 and 24 h using an inverted fluorescent microscope (Olympus). The cross-sectional areas of the spheroid and the invasion area were analyzed with the ImageJ software. The percentage of cell invasion was calculated by normalizing the cross-sectional area at 12 and 24 h to that at 0 h. At least three independent experiments were performed (each in quintuplicates).
Matrix degradation assay
A QCM Gelatin Invadopodia Assay (EMD Millipore) was performed according to the manufacturer’s instructions. Two-well glass chamber slides (Nunc Lab Tek II, Thermo Fisher) were coated with Cy3 (red) gelatin. 5 × 104 cells in RPMI 1640 supplemented with 10% FCS were seeded onto the fluorescent gelatin matrix and incubated for 24 h. Cells were then fixed and stained with phalloidin. Matrix degradation was quantified as the proportion of cells associated with the degraded gelatin matrix, as defined by the formation of at least one degradation patch underneath the cell, regardless of its size, and the total number of cells counted. Experiments were repeated three times and data in 10 fields at a magnification of 20× were used for the calculation.
Immunoprecipitations (IP) were performed using the µMACS Protein A/G MicroBeads Kit (Miltenyi Biotec) according to the manufacturer’s instructions. Whole cell lysate of 5 × 106 cells was mixed with cortactin 4F11 mouse monoclonal antibody (EMD Millipore; 1 µg per 500 µg protein lysate) and 50 µl Protein G MicroBeads, and incubated for 1 h on ice. For magnetic IP, the cell lysate was passed over a separation column placed in the magnetic field of a µMACS separator and washed four times with 200 µl cell lysis buffer. To conduct SDS-PAGE and immunoblotting analysis, the immune complex was eluted from the column using heated 1× SDS sample buffer. Rabbit antibodies (see Immunoblot analysis protocol) were used for immunoblotting to avoid the detection of any interfering IgG bands from the mouse antibody used during the IP protocol. Protein–protein interactions were depicted by immunoblotting, using antibodies specific to the putative co-immunoprecipitated targets. For mass spectrometric analysis, aliquots of the IP eluates were analyzed by Coomassie blue staining (Brilliant Blue R Concentrate; #B8647; Sigma) following SDS-PAGE (40 µl). Isolated bands were subjected to mass spectrometry (Core Unit Proteomics, Interdisciplinary Center for Clinical Research University of Münster, Germany).
Enhanced cortactin expression level and phosphorylation correlates with tumor metastasis
In cryo-conserved PDAC samples of matched non-neoplastic pancreatic and tumor tissue of 11 patients analyzed via qRT-PCR (Additional file 2: Materials and methods) no significant dysregulation of cortactin at the mRNA level was observed (Additional file 1: Figure S2).
Cortactin expression and cortactin KO in PDAC cell lines
To further assess the functional role of cortactin in PDAC cell lines, we applied CRISPR/Cas9 gene editing for depletion of cortactin (CTTN KO) in BxPC-3 and Panc-1 cell lines. Immunoblots of BxPC-3 and Panc-1 CTTN KO and control cells are shown in Fig. 2d. In immunofluorescence, CTTN KO cells showed a slight, collapsed fluorescent signal around the nucleus. However, no signal was observed at the cell periphery or cell–cell contacts (Additional file 1: Figure S3), indicating that cortactin was functionally inactive.
Knockout of cortactin decreases cell motility
Scratch assays performed with BxPC-3 and Panc-1 CTTN KO cells showed a significant decrease (30–50%) in migratory capacity (Fig. 3b, d) compared to controls. Cell migration measured with the xCELLigence technique showed a significant reduction in cell migration of up to 50% in Panc-1 CTTN KO cells compared to controls (Fig. 3e). Since cortactin is also associated with matrix-metalloproteinase secretion [24, 25], we performed gelatin degradation assays. Here, a slight but not significant tendency towards diminished matrix degradation potential was observed after cortactin KO (Fig. 3f). In collagen invasion assays BxPC-3 CTTN KO cells showed significantly impaired cell invasion into the collagen matrix compared to control cells (Fig. 3g). Together, these results indicate a pivotal role for cortactin in PDAC cell invasion in vitro.
Cortactin levels are enhanced after three-dimensional cell culturing and suppressed by dasatinib
The expression of cortactin at cell–cell contacts (Fig. 2b), the morphological switch and diminished expression of phosphorylated cortactin at the cell membrane upon dasatinib treatment (Fig. 4b), and the observation that cortactin phosphorylation is enhanced with increasing cell confluency (Additional file 1: Figure S7), prompted us to investigate if cortactin expression and/or phosphorylation was inducible by culturing the cells on soft agar enabling the formation of three-dimensional cell–cell contacts. Immunoblot analysis of three-dimensionally versus adherently cultured cells revealed increased expression or phosphorylation of cortactin after 6 days in BxPC-3, Capan-1 and Panc-1 cells (Fig. 4d). E-cadherin, a cell–cell contact molecule that has been shown to be upregulated or re-expressed in hepatocytes after spheroid formation , was also elevated in all three cell lines when they were grown as spheroids.
Immunohistochemical staining of agarose-embedded BxPC-3 cells confirmed the high expression of phosphorylated cortactin and E-cadherin at cell–cell contacts when cells were cultured as spheroids (Fig. 4e). Treating spheroid-cultured cells with dasatinib showed a decrease in the spheroids’ stability (data not shown), and immunoblotting of spheroid-cultured BxPC-3 and Panc-1 cells showed reduced E-cadherin and cortactin expression as compared to control cells (Fig. 4f).
Gelsolin is a possible interaction partner of cortactin
PDAC is one of the most aggressive human neoplasms, marked by extensive local (perineural) spread, and early metastasis to lymph nodes or the liver. Due to late symptoms and challenging tumor imaging, most cases are either too advanced for surgery, and/or the tumor has already formed metastases. The presence of metastases is an adverse prognostic factor for PDAC patients . Even after radical surgery, there is a high level of tumor recurrence. Therefore, an understanding of the cellular and molecular mechanisms underlying PDAC cell migration and invasion is crucial for the development of better therapeutic modalities for PDAC patients.
In various cancers, cortactin overexpression is associated with a dismal prognosis and increased metastasis [11, 12, 13, 14, 15]. At a mechanistic level, cortactin overexpression may promote invasion and metastasis by facilitating the formation of invasive structures such as lamellipodia and invadopodia, as cortactin has been shown to be required for the formation and proper function of those protrusive structures formed by migrating and invading cells [24, 29]. However, little is known about the biological function of cortactin and its role in cellular and molecular mechanisms involved in the tumor progression of PDAC. In our cohort of tumor specimens from patients with histologically confirmed PDAC, we observed a significant upregulation of cortactin and phosphorylated (i.e. activated) cortactin in metastases of PDAC patients as compared to primary tumors. Patients with low cortactin expression in the primary tumor appeared to have a survival advantage in the first 2 to 3 years after diagnosis (Additional file 1: Figure S1), which is in line with a study by Tsai et al., who demonstrated that high cortactin expression in tumorous tissue significantly correlates with a lower survival rate in a cohort of 50 PDAC patients . However, the correlation of cortactin expression levels with overall survival rates was of no statistical significance in our study, which may be due to the size of the group of patients with low cortactin expression and without metastases. In contrast to primary breast cancers and head and neck squamous cell carcinoma cell lines, for which a correlation of cortactin gene amplification with overexpression at the mRNA level was reported [11, 23, 30], we were unable to detect a significant upregulation of cortactin at the mRNA level when comparing cortactin mRNA expression in fresh frozen samples from PDAC tumor tissue and non-neoplastic pancreatic tissue. This indicates that cortactin in PDAC might either be regulated by post-transcriptional factors, or that the phosphorylation/activation status in addition to the expression level of cortactin might be crucial for its biological relevance in PDAC. Other mechanisms involved in elevated cortactin expression in PDAC might be based on post-transcriptional modification enhancing protein stability. It has been shown that ERK-induced serine phosphorylation of cortactin leads to polyubiquitination and degradation within the proteasome . Deregulated proteolysis might increase cortactin protein stability in cancer with high cortactin expression but without gene amplification.
Since cortactin is phosphorylated by oncogenes (e.g. Src , ERK , Arg ), which are known to be upregulated in PDAC [34, 35], we examined the activation of cortactin via phosphorylation of Tyr421 in our cohort. Cortactin showed significantly greater phosphorylation in metastases compared to primary tumors. Elevated cortactin activation by phosphorylation has been revealed to promote cancer progression in breast cancer and malignant melanoma by, e.g. enhancing the interaction of tumor cells with endothelial cells and elevating the tumor cell’s capacity for cell invasion into tissues [36, 37]. Using cortactin mutants, Li et al. reported that cortactin phosphorylation promoted metastasis in breast cancer cells; cortactin that was deficient in tyrosine phosphorylation induced more than 70% fewer bone metastases than wild type cortactin-expressing cells . Interestingly, it has also been reported that hyperphosphorylated cortactin might suppress cell migration by a negative feedback mechanism, as knockdown of cortactin in gastric cancer cells with high cortactin phosphorylation levels and high migratory potential further enhanced cell migration .
In accordance with previous reports , in our experiments stimulation with growth factors showed re-localization of cortactin to lamellipodia, indicating functional activity of cortactin in PDAC cell lines. In functional migration and invasion assays as well as proliferation assays with the two PDAC cell lines BxPC-3 and Panc-1, we were able to demonstrate that a CRISPR/Cas9-mediated cortactin knockout resulted in significant reduction in migration and invasion, while proliferation of the PDAC cells was unaltered. It has to be pointed out that, employing of a serum gradient, the xCELLigence assay is unable to clearly distinguish between chemotaxis and migration; however, the effects of CTTN knockout were also observed in scratch assays in the absence of a serum gradient. Comparable results using RNAi knockdown of cortactin were demonstrated by Miyazawa et al., showing that suppression of cortactin decreased CDCP1-PKCδ-mediated cell migration and invasion of BxPC-3 cells . In epithelial cells and fibroblasts, overexpression of cortactin resulted in more motile and invasive cells, while none of the cell types showed altered growth rates or morphologies due to cortactin modulation [16, 21]. Therefore, our results are in line with the well-established role of cortactin in cell motility.
Activated cortactin accumulates at the cell periphery . In three-dimensional cell cultures, which have been shown to reflect an in vivo situation more accurately than monolayer cultures  as they allow cells to interact with surrounding cells and to form multi-dimensional structures, cortactin was enhanced and/or hyperphosphorylated. Immunohistochemical staining confirmed the accumulation of phosphorylated cortactin at sites of cell–cell contacts together with the cell–cell contact molecule E-cadherin, which was re-expressed/upregulated in cells cultured in spheroids. E-cadherin has been demonstrated to be important for cell–cell attachment and spheroid formation . In spheroid-cultures as well as in adherent cell cultures, we were able to demonstrate that cortactin activation as assessed with regard to its localization and/or phosphorylation on Tyr421 is pharmacologically targetable by dasatinib, an inhibitor that targets Src family kinases . After treatment, phosphorylation of cortactin decreased and cortactin delocalized from the cell membrane to the cytoplasm. Additionally, dasatinib treatment led to decreased cell–cell contacts and decreased E-cadherin expression in three-dimensional spheroid culture (data not shown). Interestingly, other researchers have observed an increased stability of E-cadherin expression after dasatinib treatment with enhanced integrity of cell–cell adhesions [40, 41], while in our experiments treatment with dasatinib reversed the effects induced by three-dimensional cell culturing. It is therefore well conceivable that three-dimensional cell growth induces cortactin expression and Tyr421-phosphorylation, contributing to a pro-metastatic, migratory phenotype in PDAC.
It has been shown that the modulation of molecular mechanisms in which cortactin is involved, such as F-actin cross-linking and cofilin-cortactin interaction, is pH-dependent [42, 43]. Tumor cells growing in a three-dimensional spheroid may develop an acidic microenvironment when compared to two-dimensional tumor models which might contribute to the phosphorylation of cortactin in vitro as well as in the in vivo situation.
Consistent with the results from others, treatment with the Src family kinase inhibitor dasatinib reduced cell migration in PDAC cell lines . It is possible that the dephosphorylation of cortactin as well as its relocalization that we observed after dasatinib treatment contribute to the inhibitory effect of dasatinib on cell migration. However, the precise mechanism by which cortactin could be utilized to target cancer metastasis must be further investigated. MacGrath and Koleske have provided an excellent review of cortactin as a promising target molecule in cancer therapy approaches , showing the importance of clarifying the role of cortactin in PDAC progression as well as of identifying and understanding its interactions with other proteins that contribute to cancer migration and invasion in an effort to combine various strategies in multimodal therapy approaches and thereby prolong recurrence-free survival of PDAC patients.
In this study, we performed co-immunoprecipitation assays to identify novel proteins interacting with cortactin in PDAC, and detected gelsolin as a cortactin-binding partner using mass spectrometry. This finding was confirmed by further co-immunoprecipitation experiments from Panc-1 and BxPC-3 PDAC cell lysates followed by immunofluorescence showing a co-localization of cortactin and gelsolin at the cell periphery. The actin binding protein gelsolin regulates actin dynamics in a Ca2+-dependent manner by severing and capping actin filaments and has been shown to modulate several signaling pathways, altering cytoskeleton architecture and cell motility . In an in vitro approach using RNAi, Thompson et al. reported that a reduction of gelsolin significantly impaired pancreatic cancer cell motility . Tanaka et al. described an EMT switch after gelsolin knockdown by siRNA with a conversion of E-cadherin to N-cadherin in mammary epithelial cells . Here, in cells cultured as spheroids with three-dimensional cell–cell contacts, we observed an elevated expression of gelsolin along with E-cadherin re-expression and cortactin induction. Further experiments will be necessary to analyze the functional role of the interaction between cortactin and gelsolin in tumor progression of PDAC.
It has to be stated that low or loss of E-cadherin expression has been postulated to induce EMT and to be a prerequisite for metastatic disease progression . However, carcinomas have shown to be remarkably heterogeneous and to be able to adopt some mesenchymal features while retaining characteristics of well-differentiated epithelial cells like E-cadherin expression . Moreover, most experimental data on cancer cell migration and invasion are derived from two-dimensional cell culture models, while it has been shown that in the in vivo situation, epithelial tumors have the ability to migrate and invade collectively in three-dimensional cell clusters . Data from Gagliano et al. revealed an upregulation of N-cadherin, vimentin and podoplanin in three-dimensionally cultured PDAC cells compared to adherent cell culture . PDAC cells cultured as spheroids or during collective cell migration might therefore gain a ‘partial EMT’ status with upregulated E-cadherin, N-cadherin and vimentin. Cortactin expression and Tyr421-phosphorylation would then contribute to a pro-metastatic, migratory phenotype in PDAC.
In conclusion, our results identify cortactin and its phosphorylation on Tyr421 as an important molecular player in PDAC progression. Our data provide evidence that upregulation and/or activation of cortactin in a three-dimensional tumor microenvironment might promote migration and invasion in this fatal disease. A short graphical summary of a proposed model based on our results is depicted in Fig. 5e. This may provide an opportunity for cancer treatment by inhibition of kinases that regulate cortactin or by disruption of protein interactions with the SH3 domain of cortactin. The identification of novel binding partners of cortactin may further elucidate the mechanisms how cortactin contributes to the aggressiveness of PDAC cells.
KS: conception and design of the study, performed experiments, evaluated and interpreted data, drafted the manuscript, RB: performed experiments, JHM: acquisition and interpretation of data, AH: acquisition and interpretation of data, JR: evaluation of immunohistochemistry and critical revision of the manuscript for important intellectual content, MT: critical revision of the manuscript for important intellectual content, EW: critical revision of the manuscript for important intellectual content, WH: critical revision of the manuscript for important intellectual content, JS: conception and design of the study, drafted the manuscript, evaluated and interpreted data, critical revision of the manuscript for important intellectual content, KS: conception and design of the study, drafted the manuscript, evaluated and interpreted data, critical revision of the manuscript for important intellectual content. All authors read and approved the final manuscript.
The authors would like to thank Simone König for performing proteomic analysis and Inka Buchroth, Christin Fehmer and Petra Fischer for excellent technical assistance.
The authors declare that they have no competing interests.
Availability of data and materials
Consent for publication
The manuscript has been read and approved in the present form for submission by all named authors.
Ethics approval and consent to participate
The study was approved by the Ethics Committee of the University of Münster (Approval No. 2015-102-f-S).
This work has been supported by the IMF program (I-SP111504) and a research fund for laboratory equipment (0367_Steinestel) of the Medical Faculty, University of Münster, Germany and by the Deutsche Forschungsgemeinschaft (STE 2467/1-1).
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.Zhang Q, Zeng L, Chen Y, Lian G, Qian C, Chen S, et al. Pancreatic cancer epidemiology, detection, and management. Gastroenterol Res Pract. 2016;2016:1–10.Google Scholar
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.