Abstract
Background/Aim: Lung cancer is the most prevalent type of cancer globally and small cell lung cancer (SCLC) accounts for only 15% of all cases but exhibits a dismal prognosis. The standard of care of SCLC has not changed for decades and novel biomarkers and novel strategies for patient’s care are urgently needed. Materials and Methods: The expression of the two potential markers MUC1 and CD147 was evaluated in circulating tumor cells (CTCs) and CTC-derived SCLC cell lines using qRT PCR, western blotting, immunohistochemistry, and ELISA assays. Results: Both CTCs enriched from patient blood samples by Parsortix isolation technology and SCLC/CTC cell lines exhibited significant expression of MUC1 and CD147. Silencing of MUC1 increased chemosensitivity of an SCLC line to topotecan. Conclusion: Both markers, MUC1 and CD147, are highly expressed in patient-derived SCLC and SCLC CTC cell lines and show promise as potential biomarkers in SCLC.
Lung cancer is the leading cause of cancer death worldwide with an annual death rate that is higher than that of breast, colon and pancreatic cancer combined. In 2018, there have been 2.1 million new cases worldwide which corresponds to 11.6% of all diagnosed cancers. Small cell lung cancer (SCLC) accounts for 13 to 15% of all lung cancer cases, but due to its aggressiveness it shows poor overall survival rates (1, 2). In addition, over the past three decades progress in drug development has not resulted in superior survival rates of patients with SCLC compared to those with NSCLC, which has shown promising therapeutic changes using targeted agents and immunotherapy. Although SCLC patients respond to first line platinum-based chemotherapy, more than 80% relapse within one year (3). Treatment options are quite limited, as only few agents are approved as second line treatment such as topotecan in Europe and the U.S. and amrubicin in Japan (4, 5).
Hence, there is a big need for alternative treatment approaches in SCLC. In the past, CTCs have gained interest not only as a model of metastasis, but also for translational applications such as treatment stratification and monitoring (6, 7). Although CTCs are usually quite rare, SCLC patients show very high numbers of CTCs that made the establishment of several CTC lines ex vivo possible (8, 9). Most of these cell lines derive form large tumorospheres that contain quiescent and hypoxic cells due to limited perfusion. These spheres probably contribute also to radioresistance in patients as the hypoxic conditions hinder the formation of toxic oxygen radicals (9).
The CD147 glycoprotein plays an important role in the regulation of the tumor microenvironment and for the progression of tumors via different cellular pathways, in particular, through its effects on glycolysis and by its characteristic function in the induction of diverse proteinases triggering matrix degradation, tumor invasion, metastasis and neoangiogenesis (10). Furthermore, CD147 expression seems to represent an important tumor biomarker for diagnosis and prognosis, as well as a potential therapeutic target for cancer treatment. The enrichment of CTCs in clear cell renal cell carcinoma patients by a combination of antibodies directed to cell surface carbonic anhydrase 9 (CA9) and CD147 has shown high efficacy (11).
Mucin 1 (MUC1) is a cell membrane glycoprotein that exhibits typical alterations in glycosylation or expression in various cancer types including breast, ovarian and NSCLC (12, 13). MUC1 contributes to tumor progression by stimulating proliferation, facilitating aberrant glucose metabolism, driving metastasis by EMT induction and playing a role in chemoresistance (14). MUC1 also modulates a number of cancer signaling pathways including EGFR and NF-ĸB and, additionally, increased cell surface expression of MUC1 has been demonstrated to interfere with leucocyte interactions eventually causing immune system evasion (15). Normal cells express MUC1 in an orderly polarized manner on the cell membrane in contrast to cancer cells. This aberrant expression of MUC1 leads to the formation of a physical mucin barrier which masks the cell and hence impedes drug delivery. Another putative mechanism of drug resistance is the negative electric charge of most mucins that attracts electropositive drugs thus impairing their cellular uptake (16). In the past years, MUC1 has been investigated as a CTC marker in various cancer entities and has gained attention as a potential drug target due to its role in metastasis, growth and immunosuppression (17, 18). The altered expression of MUC1 has also been correlated to low disease free survival (DSF) and poor overall survival (OS) in NSCLC and other cancer entities (13, 16, 19). Recent studies showed that the use of MUC1 peptide for vaccination in cancer patients resulted in the generation of anti-MUC1 antibodies and a cytotoxic T-lymphocyte response (20). Other promising approaches are antibodies that target tumor surface-specific glycoepitopes of MUC1 and the usage of antibody-drug conjugates that target the human MUC1-C oncoprotein, which have both already shown successful tumor regression in human tumor cell xenografts (21-23).
Another promising marker for the identification of CTCs is CD147 (Basigin, extracellular matrix metalloproteinase inducer/EMMPRIN) which has been reported to be overexpressed on SCLC CTC cell lines. CD147 is an inducer of metalloproteinases involved in shaping the tumor microenvironment and in mechanisms of tumor progression such as glycolysis, tumor cell invasion, metastasis and angiogenesis (24). CD147 is involved in the transformation of normal fibroblasts to tumor-associated fibroblasts and stimulates adjacent fibroblasts to increase the secretion of metalloproteinase 2 (MMP2) which leads to remodeling of the tumor microenvironment. CD147 overexpression directly drives tumor angiogenesis by the stimulation of the production of VEGF via the PI3K/Akt pathway (25). By its interaction with monocarboxylate transporters (MCTs), CD147 influences the tumor cell microenvironment and leads to increased aerobic glycolysis by the tumor cells (26). CD147 induces drug resistance through interactions with the CD44-hyaloronan axis and the regulation of ABC transporters, leading to increased drug efflux that results in resistance to cisplatin and shorter survival in NSCLC (27-29). In various cancer entities, over-expression of CD147 has been associated with adverse tumor effects such as PFS, OS, and recurrence-free survival (30). The levels of soluble CD147 were demonstrated to be correlated with growth, metastatic potential and chemoresistance of tumors and CD147 has been suggested as predictive marker in breast and hepatocellular carcinoma (27, 31). In this present work, we examined the expression of these two markers in SCLC cells and their possible role as biomarkers for the identification of SCLC CTCs and as potential targets for the elimination of SCLC cells.
Materials and Methods
Blood samples. Blood samples were withdrawn from patients with SCLC at the Department of Respiratory Oncology at the Otto Wagner Hospital, Vienna, Austria. The acquisition of samples was approved by the Ethics Committee of the Medical University of Vienna, Austria (EK366/2003 and EK2266/2018) and all donors signed an informed consent form (9). Blood was collected in Vacuette EDTA tubes (Greiner Bio-One, Kremsmuenster, Austria) and processed on the same-day in accordance with a recently published protocol employing the label-free microfluidic Parsortix technology (Angle Plc., Surrey, UK) (32, 33). For SCLC, a separation cassette with a small step size of 6.5 μm was used.
Detection of SCLC markers by PCR. Total RNA was extracted from the cell lysates using the RNeasy Micro Kit (Qiagen, Hilden, Germany) without DNase-treatment. The RNA was transcribed into cDNA and qPCR was performed on the ViiA7 Real-Time-PCR System using the TaqMan® Universal Mastermix II and exon spanning TaqMan® assays specific for CHGA, SYP and ENO2 (Life Technologies, Carlsbad, CA, USA) as described recently (33).
Immunofluorescence staining for MUC1. Immunofluorescence staining was performed using primary antibodies against EpCAM and MUC1, and DAPI dye (Sigma-Aldrich, St. Louis, MO, USA) after fixation with 4% formaldehyde and permeabilization with 0.5% Triton X-100 (Sigma-Aldrich) for 10 min each followed by treatment with Ultra-V Blocking Reagent (ThermoFisher, Carlsbad, CA, USA). Counter-staining for leukocytes was performed using a rat anti-CD45 antibody (Sigma-Aldrich). The slides were incubated with Dako REAL™ Antibody Diluent (Santa Clara, CA, USA) containing the appropriate dilution of primary antibodies overnight at 4°C. Visualization of antibody binding was performed by incubation with secondary antibodies labelled with Alexa Fluor® 488 (MUC1), Alexa Fluor® 555 (CD45) and Alexa Fluor® 647 (EpCAM) (ThermoFisher), each diluted with PBS containing 6% bovine serum albumin. After three washings, the nuclei were stained with DAPI. The specimens were sealed with Fluoromount-G mounting medium and a cover slip before evaluation for CTCs using a fluorescent microscope (Olympus BX50, Tokyo, Japan). Fluorescence was quantitated using a CellCelector™ platform (ALS, Jena, Germany).
Flow cytometry. Antibodies and isotype controls were from Biolegend (San Diego, CA, USA) and secondary reagents from Sigma-Aldrich. Dilutions of the antibodies were made according to the manufacturer’s instructions. The flow cytometric experiments were performed using the Cytomics FC500 cytometer and the data analysis and histogram overlays were performed employing the Kaluza flow analysis software (Beckman Coulter, Vienna, Austria) (9).
Chemosensitivity test. Aliquots of 1×104 cells in 200 μl medium were treated for four days with twofold dilutions of the test compounds in 96-well microtiter plates in quadruplicate (TTP, Trasadingen, Switzerland) and viability examined using a modified MTT assay (EZ4U, Biomedica, Vienna, Austria) as described (9).
Western blot array and ELISA. MUC1 expression was determined using data from the Proteome Profiler Human XL Oncology Array (R&D Systems, Minneapolis, MN, USA) and CD147 from the human EMMPRIN/CD147 Quantikine ELISA Kit (R&D systems) according to the manufacturer’s instructions. Data were analyzed and presented using Origin software (OriginLab, Northampton, MA, USA) (9).
Statistical analysis. Statistical analysis was performed by Student’s t-test and one-way ANOVA using Origin software (OriginLab, Northampton, MA, USA). Differences were considered significant when p<0.05.
Results
Staining of SCLC and SCLC CTC for EpCAM, MUC1 and CD147. SCLC tumor cells from peripheral blood were isolated by special Parsortix chips collecting cells larger and more rigid than the blood cells. For staining, cytospins were prepared for each patient and labeled with specific antibodies. Figure 1 shows an example for a positive MUC1 immunohistochemical staining of a pair of SCLC CTC cells. Figure 2 shows a representative MUC1 immunohistochemical staining of the SCLC cell line DMS53 as positive control.
Immunohistochemical staining of small cell lung cancer (SCLC) circulating tumor cells (CTCs) isolated from a patient blood sample. The paraffin-embedded SCLC CTC cells show positive staining for MUC1 (left). DAPI nuclear staining of all cells including leukocytes of the preparation (right). Cells stained positively for MUC1 are negative for the leukocyte marker CD45.
MUC1 staining of the DMS53 small cell lung cancer cell line (right) and DAPI staining of the cell nuclei (left).
The results of the detection of EpCAM, MUC1 and CD147 are summarized in Table I. With the exception of SCLC CTC cell line BHGc26 and the A549 cell lines, cells were found positive for the expression of EpCAM. MUC1 antigen expression was detected in all cell lines except in the SCLC CTCs UHGc5 cell line. Similar results were observed for CD147 expression that was found in all cells except in UHGc5 cells and A549 cells.
The results of immunohistochemical staining of paraffin-embedded SCLC/CTC cell lines and the A549 NSCLC line. Stainings are rated negative (0), weak (+), moderate (++) and strong (+++).
PCR analysis of the expression of SCLC markers. In order to confirm that the proprietary CTC cell lines originate from SCLC, the expression of SCLC markers chromogranin A (CHGA), enolase-2 (ENO2) and synaptophysin (SYP) was determined. All cell lines showed expression of at least two SCLC markers (Figure 3). In contrast, the NSCLC cell line IVIC-A, that represents a SCLC–NSCLC transformant, exhibited low expression of CHGA and SYP as well as lower expression of ENO.
PCR analysis of the typical small cell lung cancer (SCLC) markers for the SCLC and SCLC/circulating tumor cell lines. Data are shown as mean Cts±SD.
PCR analysis of MUC1 mRNA expression of SCLC CTC lines. MUC1 gene expression in the SCLC CTC lines was assessed relatively to normal lung tissue (Figure 4). Levels of MUC1 were comparatively equal to normal lung expression, with exception of a higher expression by BHGc16 and BHGc26, and a lower expression by UHGc5.
Relative MUC1 expression in the small cell lung cancer circulating tumor cell line in comparison to normal lung tissue (mean±SD). Asterisks indicate the significant differences of MUC1 RNA expression in the individual cell lines versus normal lung tissue and the differences between BHGc10 as single cells and in cell clusters.
MUC1 protein expression according to western blot analysis. In order to examine the overall protein expression of MUC1, Proteome Profiler Human XL Oncology Array (R&D Systems), which detects the expression levels of a range of human cancer-related proteins including MUC1 was carried out. Figure 5 shows the expression levels of MUC1 in the cells under investigation. The results of MUC 1 protein expression in three NSCLC (IVIC-A, B, D) and eight SCLC lines (NCI-H526, DMS153, SCLC 26A, S457, S597, S318, GLC14, GLC 16) compared to 10 SCLC CTC lines, BHGcx and UHGc5 showed that all lines were positive for MUC1 to various extent, whereas DMS153A displayed very high MUC1 expression.
Relative MUC1 protein expression levels in the small cell lung cancer and non-small cell lung cancer cell lines.
Cisplatin-based regimens are used as first line treatment for SCLC while topotecan or epirubicin-based combinations are used as second-line treatments. Therefore, the effects of these chemotherapeutics on MUC1 protein expression in SCLC (SCLC26A, GLC14 and GLC16) and SCLC CTC (BHGc10 and BHGc16) were examined using flow cytometric analysis of MUC1-labeled cells (Figure 6). The results showed that cisplatin and topotecan increased MUC1 expression in GLC14 and GLC16 cells, whereas these chemotherapeutic drugs exhibit minor effects on the two SCLC CTC lines. Epirubicin tends to lower the expression of the MUC1 antigen.
Alterations in MUC1 protein expression in response to preincubation with indicated concentrations of the chemotherapeutics for 3 days. Values represent mean fluorescence intensities (±SD).
Silencing of MUC1 expression in SCLC cell lines. MUC1 expression was silenced using the appropriate siRNAs as described and control and transfected SCLC cell lines were analyzed using flow cytometry (Figure 7). For SCLC26A primary SCLC cells, transfection resulted in a clear reduction in the expression of MUC1. The MUC1 relative fluorescence intensity in MUC1-siRNA transfected BHGc10 and BHGc16 cells compared to the control-siRNA transfected cells was statistically significantly reduced (Table II).
Overlay of the control histogram and the MUC1 antibody histogram. On the left side the results of a fluorescence activated cell sorting (FACS) analysis of SCLC26A cells transfected with the control-siRNA. On the right side the results of a FACS analysis of SCLC26A cells transfected with MUC1-siRNA. The blue represents the fluorescence intensity of the FACS-control and the red represents the fluorescence intensity of MUC1.
Fluorescence activated cell sorting analysis of SCLC26A, BHGc10 and BHGc16 cells transfected with MUC1-siRNA (data represent mean fluorescence intensities±SD).
Effects of silencing MUC1 expression on chemosensitivity of the SCLC26A cell line. MUC1 was silenced using siRNA in SCLC and SCLC CTC cell lines followed by MTT cell viability assays. siRNA silencing of MUC1 led to altered cell viability upon chemotherapeutic treatment. The IC50 value of topotecan decreased in the SCLC26 cell line, whereas no effect was observed in the two SCLC CTC cell lines (Figure 8).
MTT assay of MUC1-siRNA or control siRNA transfected SCLC26Acells treated with different concentrations of topotecan (0.01 μM to 5 μM as ten 2-fold dilutions) for four days.
MUC1 siRNA silencing induced adhesion of the SCLC26 cell line, which is usually in suspension when kept in culture. After two weeks of siRNA transfection the cell line showed adherent cells as seen in Figure 9. Similar effects were observed in the CTC cell line BHGc16.
SCLC26A growing as single cell suspension (A) acquire an adherent phenotype (B) upon silencing of MUC1.
CD147 staining of cell lines and CD147 ELISA. SCLC CTC cell lines showed positive CD147 staining. Figure 10 shows a representative immunohistochemical staining of the cell line DMS 53 with positive CD147 staining. In order to examine the overall protein expression of CD147, supernatants of the SCLC/CTC cell lines were analyzed using a specific CD147 ELISA test (Figure 11). All cell lines tested exhibited significant CD147 concentrations exceeding 100 pg/ml.
Representative DAPI stain (top) and immunohistochemical staining of cell line DMS 53 which exhibits positive staining for CD147(bottom).
ELISA results of the determination of CD147 in the medium supernatants of the small cell lung cancer/circulating tumor cell lines (mean values±SD).
Discussion
The significance of CTCs as predictive marker has been convincingly demonstrated in a number of cancers; however, their importance in diagnosis and monitoring of treatment regimens remains to be established (32). The clinical role of CTCs is currently studied in breast, lung, colorectal, and prostate cancer that were shown to exhibit significant cell counts, especially in the advanced state. Investigations are based on the assessment of CTC counts or on molecular features of CTCs such as cell surface markers and gene expression partially in order to replace invasive biopsies by easily performed blood tests. The FDA-approved CellSearch system (Janssen Diagnostics, Raritan, NJ, USA) enriches EpCAM+ circulating epithelial cells employing antibody-coated magnetic particles and verification of the CTCs by immunofluorescence staining for CK 8, 18, 19 and proof of the absence of CD45 expression. The sensitivity of CellSearch CTC detection is 27%, 32%, 70% for metastatic breast, colorectal and prostate cancer, respectively (34). A marker-independent technique, the “Isolation by Size of Tumor cells” (ISET) method showed a poor predictive power for the development of malignant lesions based on CTC counts.
The quantification of MUC1 (defined as antigens CA 15-3, KL-6 or BM7) in patients is in clinical use as a response marker and as a prognostic indicator for survival. The CA 15-3 antigen is the most widely used serum antigen in breast cancer, with CA 15-3 levels correlating with tumor grade, lymph node positivity, and presence of metastases (35). In this study, we examined the expression of MUC1 and CD147 in blood samples, SCLC cell lines and CTC-derived SCLC cell lines. Aberrant expression of both markers has been previously described as a biomarker in various cancer entities but have yet to be defined in SCLC. SCLC CTCs were enriched from blood samples of patients using the Parsortix technology and most CTC cytospins were MUC1 positive (33, 36). Most of the SCLC CTC cell lines examined in this study were MUC1 positive. The origin of the SCLC CTC cell lines was verified by positive staining for the neuroendocrine markers CHGA, ENO2, and SYP (37). The MUC1 protein expression in all SCLC CTC cell lines was further demonstrated using western blot arrays. Furthermore, gene expression of MUC1 was found in all SCLC and SCLC CTC cell lines with several lines exhibiting higher expression than normal lung tissue.
Expression of MUC1 has been suggested to protect cells from chemotherapy and, therefore, we investigated the effect of chemotherapeutics on MUC1 in cancer cells, following pretreatment with cisplatin, epirubicin or topotecan. Cisplatin and topotecan showed only minor effects on MUC1 expression levels, but pretreatment with epirubicin resulted in reduced MUC1 expression in SCLC cell lines. Overall, the expression of MUC1 was heterogenous in the SCLC CTC cell lines after chemotherapeutic treatment. The potential effect of MUC1 on the chemosensitivity of SCLC cells was investigated using gene silencing experiments and chemosensitivity tests in SCLC tumor cells. Successful siRNA silencing of MUC1 was confirmed by fluorescence activated cell sorting analysis of SCLC26A cells. The silencing of MUC1 in SCLC26A cells resulted in a chemosensitizing effect in case of topotecan treatment, with a two-fold increased cell death rate in accordance with a protective effect of MUC1 against chemotherapeutics.
Overall, SCLC-derived cell lines, CTC cell lines, and CTCs isolated from patient blood express MUC1. Expression levels were quite heterogenous and ranged from very low to very high. However, MUC1 is not expressed in leucocytes in human blood and is, hence, a promising biomarker for CTCs isolated from human blood samples. CD147 was another highly expressed marker on our established SCLC cell lines. We could show that most paraffin embedded cell lines express CD147 and this antigen was detectable in western blot arrays. MUC1 is overexpressed in NSCLC tissues and lung cancer A549 paclitaxel resistant cells (38). Silencing MUC1 inhibited proliferation and induced apoptosis of A549/PR cells through the upregulation of proapoptotic Bax and Caspase 3 and reduction of Bcl 2, indicating that modulation of MUC1 in combination with chemotherapy enhances antitumor efficacy.
The increase in MUC1 expression accompanying the transformation of premalignant lung lesions to invasive carcinomas indicates an important role of MUC1 (39). In particular, MUC1, is highly expressed in exosomes compared to membrane proteins and the level of exosomal MUC1 is elevated in cancer patients compared to normal controls (17). MUC1-C over-expression leads to disruption of adherent-junctions and cytoskeleton rearrangement and enables basement membrane invasion (40). Over-expression of MUC1 predicts a poor prognosis in breast, pancreatic, and colon cancer patients (41, 42). Tumor-associated MUC1 binds to cell adhesion molecules (CAMs) likely mediating the adherence of MUC1-expressing CTCs to blood vessels and extravasation/metastasis (43, 44). In a study involving advanced breast cancer, responses to therapy were indicated by a significant reduction of MUC1 mRNA in cells selected by EpCAM, cytokeratin-19, and MUC1 (45-47). MUC1 has been found to be expressed in a high percentage of CTCs from metastatic breast, NSCLC, pancreatic, and colon cancer patients, among others, and our results demonstrate that the use of MUC1 as a CTC marker can be extended to SCLC (46, 48).
CD147 constitutes a glycoprotein regulating the tumor microenvironment and cancer progression, mainly via modulation of glycolysis and induction of proteinases such as matrix metalloproteinases (MMPs) and serine proteases (uPA, plasmin), triggering degradation of extracellular matrix and tumor cell dissemination (49). Therefore, increased expression of CD147 marks tumor progression and predicts a poor prognosis (50). For example, expression of both carbonic anhydrase 9 and CD147 was prevalent in renal cancer patients (51). CD147 has been shown to be present at high levels in primary tumors and in micrometastatic cells, corroborating its essential role in tumor progression and early metastasis (52). In addition, CD147–integrin interactions were described that stimulate cancer invasiveness via induction of MMP synthesis through the focal adhesion kinase (FAK)-PI3K signaling pathway (50, 53). Furthermore, CD147 promotes tumor angiogenesis by increasing VEGF and MMP expression in both tumor and stromal compartments (25, 54). In conclusion, MUC1 and CD147 cooperate to stimulate tumor invasion and, thus, can function in blood-borne tumor dissemination.
Footnotes
Authors’ Contributions
B.S., M.H. C.S. and G.H. designed and performed experiments and wrote the article; B.R., E.O. and S.S. provided technical support and conceptual advice. Writing, review, and/or revision of the manuscript: B.S., B.R. and G.H.
Conflicts of Interest
The Authors declare that they have no competing interests in relation to this study.
- Received August 3, 2021.
- Revision received October 29, 2021.
- Accepted November 4, 2021.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.