The role of the desmosomal protein desmocollin 2 in tumour progression in triple negative breast cancer patients

Patient cohorts

All patients, from whom the tissue samples were derived, were treated at the University Medical Centre Hamburg-Eppendorf, Germany, Department of Gynaecology between 1991 and 2002. All patients gave written approval for the utilisation of their tissue samples and the reviewing of their medical records according to our investigational review board and ethics committee guidelines (Ethik-Kommission der Ärztekammer Hamburg, #OB/V/03). Further cohort details and patient characteristics are listed in the Additional file 2: Table S1. Microarray analyses of DSC2 mRNA levels in patients with and without distant metastases were performed on 197 mRNA extracts from primary breast cancer tissue samples. For the western blot, a total of 111 samples were collected based on tissue availability from the same patient cohort. Slides of four tissue microarrays, constructed under permission of the Ethikkommission Beider Basel (EKBB # 395/11), kindly provided within collaborative efforts in the frame of the Pathobiology study group of the EORTC by Dr. Serenella Eppenberger-Castori from the Biobank at Institute of Medical Genetics and Pathology at the University of Basel, Switzerland, were used for immunhistochemical analysis. Patient characteristics are supplied in Additional file 3: Table S2.

Microarray data

We analysed DSC2 mRNA levels using microarray data (Affymetrix HG-U133A) from the aforementioned cohort. Here, two probe sets (204750_s_at and 204751_x_at) corresponding to DSC2 were available and analysed independently. Additionally, the mean expression value of the 2 probe sets was calculated and also included in further analyses. According to the DSC2 mRNA values of each probe set and the mean value, the cohort was divided into quartiles of similar size, representing low, moderate-low, moderate-high, and high DSC2 levels. Correlations between DSC2 mRNA levels (quartiles) and clinicopathological factors, such as histological grading, stage, lymph node involvement, oestrogen, and progesterone receptor status (ER, PR) were statistically examined by χ2-tests. Overall survival was analysed by Kaplan–Meier analysis and log-rank tests. Additionally, the correlation between DSC2 mRNA levels (continuous data) and disease-free and overall survival was calculated using Cox regression analyses. Multivariate Cox regression analyses including the clinical stage, nodal involvement and molecular subtype were performed for all probe sets and the DSC2 mean value. Here, a backwards analysis with stepwise removal of insignificant terms was used. Probability values less than 0.05 were regarded as statistically significant. All statistical analyses were conducted using SPSS software Version 26 (SPSS Inc., Chicago, IL, USA). For validation purpose we used an independent Affymetrix microarray dataset consisting of 572 breast cancer samples from Gene Expression Omnibus (GSE2603, GSE2034, GSE12276) for which detailed information on metastatic localization was available [32].

Protein lysate preparation and western blot analysis

Tissue samples were obtained intraoperatively and immediately stored in liquid nitrogen as fresh frozen samples. The histological characteristics of each sample were evaluated on cryo-cut and haematoxylin–eosin-stained sections. The tissue was tailored, where necessary, to obtain at least 70% tumour cells in the sample used for protein extraction. Approximately 100 mg of tissue was excised and pulverised using a micro-dismembrator (Braun-Melsungen, Melsungen, Germany) for 2 min and 45 s at 200 r.p.m. Proteins were lysed in ice-cold sample buffer (50 mM Tris pH 6.8, 1% sodium dodecyl sulphate (SDS)), 10% sucrose and 10 μl/ml protease inhibitor cocktail (Sigma, Taufkirchen, Germany). For western blot analyses, volumes of tumour lysates containing 20 μg of protein were loaded per well. The following antibodies were utilised in the western blot detection process: mouse monoclonal anti-DSC2 IgG (Millipore, MABT411) dilution 1:1000, mouse monoclonal anti-β-Actin (C4) (Santa Cruz Biotechnology, sc-47778) dilution 1:2000 and goat anti-mouse IgG-HRP (Santa Cruz Biotechnology, sc-2055) dilution 1:8000. Antibodies were visualised using a chemiluminescent reagent (SuperSignal ® West Pico chemiluminescent Substrate, Thermo Scientific, Rockford, USA). Protein band intensities were quantified using a calibrated densitometer (GS-800 Imaging Densitometer, Bio-Rad, Munich, Germany). The primary breast cancer protein lysate UPA497 was used as a positive control for DSC2, with its DSC2 expression being defined as 100% for the purpose of standardisation. Protein expression values in all detected bands were also normalised using the loading control β-Actin. For the statistical analyses, these values were divided into four equal groups (quartiles), representing very low, low-moderate, moderate and high protein expression.

Breast cancer cell lines, cell culture and stable transfections

The human TNBC cell line MDA-MB231 and its brain seeking subline MDA-MB231-BR were provided by Dr Takara (University of Texas). Cells of both lines were cultivated in Dulbecco’s Modified Eagle’s Medium (DMEM, ThermoFisher Scientific, Waltham, MA, USA) supplemented with 10% fetal calf serum (FCS) under standard cell culture conditions. Cells were authenticated before usage. Two different DSC2-knockdown MDA-MB231-BR cell lines were generated by lentiviral transduction using vectors containing shRNA-sequences targeting specific regions of the DSC2 mRNA sequence (MISSION shRNA III and V, Sigma-Aldrich, GmbH). Similarly, a control cell line was established using a scramble shRNA sequence (Addgene, plasmid ID1864). The full DSC2 cDNA sequence obtained from a commercially available vector (Des476-Desmocollin 2-myc Plasmid; Addgene Plasmid ID: 32233) was cloned into LeGO-iC2-Puro + Plasmid (kindlykindly provided by AG Fehse, Center for Oncology, Department of Stem Cell Transplantation, UKE, Hamburg, Germany) using BamHI and EcoRI restriction enzymes. After lentiviral production in HEK293T cells, MDA-MB231 cells were transduced. The corresponding empty vector was taken as a negative control. After selection with puromycin (2ug/mL), the level of DSC2 mRNA and protein was detected using real time quantitative polymerase chain reaction RT-PCR and western blot analysis, respectively.

RNA isolation and real-time quantitative polymerase chain reaction

RNA was isolated using the RNeasy Mini Kit (Qiagen, Hilden, Germany) and a QIAshredder (Qiagen, Hilden, Germany), and was subsequently reverse transcribed using qScriber cDNA Synthesis Kit (HighQu, Kraichtal, Germany) according to the manufacturer’s instructions. RT-PCR was carried out with ORA™ qPCR Green ROX H Mix (HighQu, Kraichtal, Germany) using the StepOnePlus System (Applied Biosystems, Thermo Fisher Scientific Inc.). The data analysis was performed using the ΔΔCt method. The following primers were used for DSC2: forward primer, 5’-GCCCATCTTCTTCTTGTCGTT-3’; reverse primer, 5’-CCCGTCTTGGTGAAAAAGTGT-3’. Primer sequences for the housekeeping gene were as follows: forward primer, 5’-GTCAGTGGTGGACCTGACCT- 3’; reverse primer, 5’ -TGCTGTAGCCAAATTCGTTG-3’.

Immunofluorescence

Cells (1 × 105) were seeded on coverslips, cultured for 48 h, and fixed with 3.7% formaldehyde for 20 min at room temperature. After blocking with 1% bovine serum albumin (BSA) in phosphate buffered saline (PBS) (Mg+/Ca+) for 1 h at room temperature, cells were incubated with a polyclonal DSC2 antibody (1:50 in 1% BSA/PBS; Sigma-Aldrich, Hamburg, Germany) over night at 4 °C. Subsequent to washing, cells were incubated with a second antibody solution (mouse anti-rabbit IgG Alexa Fluor® 488; Jackson ImmunoResearch, Ely, UK; 1:500 in 1% BSA/ PBS) for 1 h at room temperature. Coverslips were carefully placed on slides using mounting medium and DAPI (Vectashield). Images were acquired using a fluorescence microscope BZ9000 and the software BZII Viewer (Keyence, Germany).

Proliferation assay

For cell proliferation analyses, the Cell Proliferation Kit II (XTT) (Roche Applied Science, Mannheim, Germany) was used according to the manufacturer’s instructions. Briefly, cells were seeded in a final volume of 100 µl medium per well in a 96-well plate (1.5 × 103 MDA-MB231-BR cells per well, 2 × 103 MDA-MB231 cells per well). After 24 h, 48 h and 72 h, cell viability was determined by adding XTT labelling mixture and by measuring the absorbance after 6 h at 490 nm using a microplate reader (DIAS Max002, Dynex Technologies, Chantilly, USA). Each experiment was performed with 12 replicates (wells) per condition (n = 12). Images shown are representative of three independently performed experiments.

Cytotoxicity analysis

For the MDA-MB231-BR cell line and the DSC2-knock down sublines, an Annexin-V/PI staining was performed after cisplatin treatment in order to quantify the extent of apoptotic and necrotic cells. Briefly, cells were seeded into 6-well plates at a density of 2.5 × 105 cells per well, incubated for 24 h and treated with cisplatin (Accord Healthcare Limited, North Harrow, United Kingdom) in three different concentrations (10 µM, 25 µM, 50 µM) for 48 h using serum-reduced DMEM medium. Subsequently, cells within the supernatant, as well as adherent cells, which were carefully detached using AccuMax (eBioscience, San Diego, CA, USA), were stained with an APC-labelled Annexin-V antibody (Annexin-V-APC, AnxA100, MabTag GmbH, Oldenburg, Germany) for 30 min at 4 °C in the dark. After washing with PBS (+/+), cells were resuspended in 1% BSA in PBS and stained with PI (BD Pharmingen, San Diego, CA, USA). FACS analysis was performed using the FACS Calibur (BD Biosciences, Heidelberg, Germany) and all data were analysed using FlowJo Software.

DSC2 overexpressing and control MDA-MB231 cells display a strong fluorescence, due to the transduction with the previously mentioned LeGO-iC2-Puro + Plasmid, which includes the mCherry-coding sequence. For these cell lines, the cisplatin-induced cytotoxicity was assessed using XTT, as described above. Briefly, MDA-MB231 cells were seeded into 96-well plates with 2 × 104 cells in 100 µl per well. After incubating for 24 h, cells were treated with cisplatin in three different concentrations (10 µM, 25 µM, 50 µM) for 48 h and the Cell Proliferation Kit (Roche) was used as described in the previous section. Each experiment was performed in duplicates. Images shown are representative of three independently performed experiments.

Migration assay

To investigate cell migration, the Oris™ Universal Cell Migration Kit (Platypus Technologies, Madison, WI) was used according to the manufacturer’s protocol. Briefly, cells were seeded (5 × 104 cells in 200 µl per well) in a 96-well plate fitted with sterile silicon stoppers using serum-reduced medium (5% FCS). After 24 h incubation, stoppers were gently removed allowing cells to migrate into the central cell-free detection zone. Migration potential was assessed by analysing the cell-free area of each well at four different time points (0 h, 24 h, 48 h and 72 h post removal of stoppers) with the ImageJ Wound Healing Tool (Wayne Rasband, National Institute of Health). Each experiment was performed with 12 replicates (wells) per cell line. Images shown are representative of three independently performed experiments.

Invasion assay

Matrigel Growth Factor Reduced (BD Biosciences, Heidelberg, Germany) was diluted to a concentration of 3.5 mg/ml with serum-free medium. Afterwards, 96-well plates were coated with a 1:1 mixture of Matrigel and serum-reduced medium (5% FCS) and incubated for 30 min at 37 °C. Hereafter, Oris™ Universal Cell Migration Kit (Platypus Technologies, Madison, WI) was used according to the manufacturer’s protocol as described above. After a 24 h incubation period, the stoppers and medium were carefully removed and 40 μl of newly prepared Matrigel coating solution was added. Plates were incubated again for 30 min at 37 °C. Finally, serum-reduced medium was added to all wells. For determining cell invasion potential, analyses were performed as described above. Each experiment was performed with 12 replicates (wells) per cell line. Images shown are representative of three independently performed experiments.

Cell spheroid formation

Cells were seeded at 5 × 103 cells in 200 μl medium per well on 2% agarose-coated (UltraPure™ Agarose, Invitrogen, Carlsbad, CA, USA, dissolved in PBS) 96-well plates. To assess and observe spheroid formation of cells, spheroids were examined and documented every second day using light microscopy and a camera (Axiovert 40 C, Carl Zeiss AG, Leica DFC320, Wetzlar, Germany). For further investigation of compactness, spheroids were dissociated by pipetting each spheroid up and down five times and by comparing formation immediately afterwards. Each cell line was seeded in quadruplicates. Images shown are representative of three independently performed experiments.

Cytotoxicity in a 3D Model

Cisplatin-induced cell cytotoxicity on 3D-structures was assessed using immunocytochemical detection of phosphorylated gamma H2AX (γH2AX)—an established marker for DNA double-strand breaks—on tumour cell aggregates grown on polyHEMA (Sigma-Aldrich) coated flasks. Here, cell lines were seeded at a density of 2 × 106 cells in 12 ml per polyHEMA coated T75 culture flask and cultured as cell aggregates for 72 h. Subsequently, cells were incubated with cisplatin in a final concentration of 50 μM for 8 h and 24 h, and were subsequently fixed in formalin, embedded into 2% agar (Agar NOBEL, Difco Laboratories, Detroit, MI, USA) and then embedded in paraffin. Slides from FFPE-cells were pre-treated in a steamer (citrate buffer pH 6.0) at 125 °C for 4 min and S1699 (pH 6.0, DAKO) at 121° C for 10 min, respectively) and incubated with an anti-DSC2 (1:25 HPA011911, Atlas Antibodies, Sigma-Aldrich, Hamburg, Germany) or an anti-γH2A.X antibody (1:10,000; ab81299, abcam, Berlin, Germany) for 1 h at room temperature. Incubation with biotin-labelled swine anti-rabbit secondary antibody occurred for 30 min at room temperature (1:200 dilution in TBS, E0353, DAKO, Glostrup, Denmark). For detection, sections were incubated with Vectastain® ABC-AP Kit (Vector Laboratories, Burlingame, CA USA) for 30 min and stained with Permanent Red (K0640, DAKO, Glostrup, Denmark). Rabbit immunoglobulin normal fraction (X0903, Agilent, Santa Clara, CA, USA) was used as a negative control for the anti-yH2A.x primary antibody. A rabbit polyclonal IgG (ab37415, abcam, Berlin, Germany) was used as a negative control for anti-DSC2 primary antibody. All slides were slightly counterstained with haematoxylin. Stained slides were scanned using the Axio Scan.Z1 (Zeiss, Jena, Germany) and images were acquired using netScope Viewer Pro software version 1.0.7079.25167 (NetBase Software GmbH, Freiburg, Germany). For quantifying the γH2A.X staining, manual counts of positive stained cells (4 × 100 cells in 4 different areas of each slide) were performed using the assistant electronic memory counter Counter AC-15 (Karl Hecht Assistant, Altnau; Switzerland).

Intracardiac metastasis mouse model

The intracardiac mouse model was conducted as previously described [33]. Briefly, female 8 to 9-week-old SCID mice (CB17/lcr-Prkdcscid/lcrlcoCrl) were anesthetized and 1 × 106 tumour cells were injected intracardially into the left ventricle of the heart (n = 15 per group). Tumour cells were previously transduced with luciferase-bearing plasmid and bioluminescence signals were tested before injection via bioluminescence imaging (BLI). After intracardiac injection, tumour cell dissemination was monitored weekly under BLI. Assessment of subsequent metastases was monitored in vivo weekly by imaging for up to 3 weeks. Mice showing termination criteria were immediately sacrificed. At the endpoint (21 days), animals were anesthetized and blood was collected from the left ventricle by cardiac puncture immediately before the final killing was executed by cervical dislocation. Ex vivo bioluminescence imaging was conducted from the lungs and brain. Lungs and brain were equally divided and frozen down for DNA isolation and subsequent ALU-PCR or paraffin-embedded for further analysis (H&E and luciferin staining) as previously described [33, 34]. The animal experiments were approved by the Authority for Social Affairs, Family, Health, and Consumer Protection of the Free and Hanseatic City of Hamburg through application N005/2020.

Histology and immunohistochemistry

The whole brain and the right lung of the mice were fixed in 4% buffered formalin and processed for wax histology. 4 µm sections were cut from brain for immunohistochemistry and 10 sections from the middle of the block were stained with hematoxylin and eosin (H.E.). The lungs were fixed en block and subsequently cut into 1 mm thick slices and embedded in 2% agar. Afterwards, the lung slices were paraffin-embedded and cut into 4 µm thick sections. Ten sections of each paraffin wax block were H.E. stained and metastases were counted at a 200-fold magnification using Zeiss Axiophot photomicroscope (Zeiss, Jena, Germany). Additionally, two series of serial sections out of the middle of each paraffin wax block were preserved for further immunohistochemical analyses. The immunohistochemical staining was performed on 4 µm sections. Sections were deparaffinized in descending ethanol concentrations and pre-treated with citrate buffer solution (pH 6.1) in a steamer for 4 min. After incubation for 1 h at room temperature with the primary antibody DSC2 (Atlas, HPA011911), samples were washed twice with TBS-T (TBS + 0.1% TWEEN-20) and once with TBS for 5 min. After incubation with anti-rabbit secondary antibody (LS-Bio, LS-C350860) for 30 min at room temperature, antibody binding was visualized using the Vectastain ABC-AP Kit (VectorLabs., Burlingame, CA, USA) and Permanent Red Solution (Dako) according to the manufacturer’s instruction. The nuclei were counterstained in Mayer’s hemalum solution.

Circulating tumour cell detection

Mouse blood samples (200–500 µl) were obtained via cardiac puncture and collected into EDTA KE/1.3 tubes (Sarstedt, Germany). To perform cardiac puncture, mice were deeply anaesthetized under isoflurane, and a 21-gauge needle coated with heparin was inserted into the heart. Mice were euthanized immediately following the cardiac puncture. Blood samples were processed on the label-independent, microfluidic system Parsortix® (ANGLE plc., United Kingdom), a device designed for the size-based capture of rare cells from whole blood [35]. The isolated cells were harvested and spun onto a glass slide (190 g, 7 min). Slides were dried overnight at room temperature and stored at − 80 °C until further analysis.

Tumour cells isolated with the Parsortix® system were identified via immunocytochemistry. Briefly, dried cytospin slides were brought to room temperature and fixed with 2% PFA (Sigma Aldrich, Germany) for 10 min. The samples were washed with 0.5 mL of 1x-PBS prior to permeabilization with 0.1% Triton X 100/PBS (Sigma Aldrich, Germany) for 10 min. Following two additional wash steps, 10% AB-serum/PBS (BioRad, Germany) was applied for blocking (60 min). Standard detection of CTCs is usually achieved with epithelial antibodies [36], however, TNBC cells lack epithelial markers and are successfully detected with CD298 [37]. Subsequently, directly anti-human PE labelled CD298 (clone LNH-94, Biolegend, USA) and anti-mouse Alexa Fluor 488 conjugated CD45 (clone HI30, Biolegend, USA) antibodies were incubated for 60 min, followed by 5 min of DAPI-incubation (1 µg/mL). Cytospins were covered with Prolong Gold Antifade Reagent (Thermo Fisher Scientific, Dreieich, Germany), sealed with a cover slip and examined by fluorescence microscopy (Axio Observer 7, Zeiss). CD298-positive, DAPI-positive, CD45-negative cells with intact morphology were defined as tumour cells. Clusters were defined when 2 or more cells were found together.

Statistical analyses

All statistical analyses were performed using SPSS Statistics version 24 for Windows (IBM, Armonk, NY, USA). Correlations between mRNA and protein expression values were assessed using two-sided Pearson tests. Chi-square tests were used to correlate both mRNA expression (Microarray data, cohort A) and protein expression (WB data, cohort B) with the following clinical and pathological parameters; histological grading (G1/G2/G3), molecular subtype (Luminal/HER2 positive/TNBC), ER and PR Status (positive/negative) and the presence of metastases (loco-regional/bone/lung/visceral/brain). Kaplan–Meier estimates and the log-rank test were carried out to ascertain and compare disease-free and overall survival. The associated hazard ratios for the multivariate analyses were determined by Cox regression. Proliferation assays and cytotoxicity assays measured with XTT were statistically analysed using GraphPad Prism 5 (GraphPad, La Jolla, CA, USA). Each in vitro assay was performed at least three times. Statistical significance was assessed using unpaired two-tailed Student’s t-test. The assumption of homogeneity of variances was checked via Levene’s Test of Equality of Variances (p > 0.05). Results are given as mean ± s.d. or s.e. Probability values (p-value) ≤ 0.05 were considered to be statistically significant.

留言 (0)

沒有登入
gif