Puig et al. BMC Res Notes (2019) 12:68 https://doi.org/10.1186/s13104-019-4100-z BMC Research Notes RESEARCH NOTE Open Access Multiple immunofluorescence assay identifies upregulation of Active β‑catenin in prostate cancer Pere Puig1* , Nadina Erill1, Marta Terricabras1, Isaac Subirana1, Judit González‑García1, Adrià Asensi‑Puig1, Michael J. Donovan2, Lourdes Mengual3, M. Teresa Agulló‑Ortuño4, Mireia Olivan5, Antonio Alcaraz3, José A. López‑Martín4, Inés de Torres5, José Luis Rodríguez‑Peralto4, Alfredo Rodríguez‑Antolín4, Juan Morote5 and Víctor González‑Rumayor1 Abstract Objectives: To apply a systems pathology-based approach to the quantification of nuclear Active β-catenin and human leukocyte antigen class I, and assess the biomarker involvement in a cohort of prostate tumor patients. Results: The systems pathology approach applied allows a precise quantification of the marker expression in the different cell compartments as well as the determination of the areas that coexpress two markers. Our data shows that the accumulation of β-catenin in the nuclear compartment is significantly decreased in the adjacent normal areas when compared to tumor of the same patients (p < 0.001). In conclusion, the application of this novel multiple immunofluorescence assay demonstrates that the upregulation of Active β-catenin is a relatively common feature of prostate tumor development, and further supports the activation of the Wnt/β-catenin pathway in prostate cancer progression. Keywords: Active β-catenin, HLA class1, Wnt/β-catenin pathway, Prostate cancer, Systems pathology Introduction the expression of Active β-catenin was higher in tumor The origin of the data is a summary of single observa- areas than in adjacent normal tissue. In this research tions derived from a more comprehensive study encom- note we describe for the first time how a systems biology- passing seven protein biomarkers that were assembled based methodology yields a precise quantification of the in two multiplex assays and applied to a cohort of pros- nuclear Active β-catenin (effector fraction) and thus une- tate cancer patients (n = 505). Quantitative and cellular quivocally demonstrates the implication of this marker in attributes representing biomarkers from both multiplex prostate tumor formation. assays containing Active β-catenin (ABC), HLA-1 and Prostate cancer is the most common epithelial malig- CK18, among others, were correlated with clinical vari- nancy after skin cancer in men from western countries ables including biochemical recurrence (BCR). One of [1]. Systems pathology is a multidisciplinary approach the objectives of the comprehensive work (manuscript that integrates proteomic, molecular and imaging data in preparation) was to describe correlations between with the patients’ clinical history using machine learn- clinicopathological data and results from the quantifica- ing [2]. By applying multiplex immunofluorescence and tion and colocalization of individual marker expressions. image analysis, we have been able to construct quantita- When we started analyzing the data we observed that tive features for interrogating several biomarkers at the microscopic cellular level. This allows for investigations of co-localization and expression of cellular proteins, *Correspondence: ppuig@atryshealth.com 1 specifically during activation of specific signaling path- R&D Department, Atrys Health, Barcelona, Spain Full list of author information is available at the end of the article ways. Of paramount importance, the colocalization of a © The Author(s) 2019. This 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. Puig et al. BMC Res Notes (2019) 12:68 Page 2 of 5 given marker with DAPI can be applied in the quantifica- Table 1 Markers included in the multiplex reaction tion of the marker that is present within the nucleus [3]. Antibody Vendor Catalog # Dilution Isotype Fluorochrome HLA class I expression on the cell surface of cancer cells is critical for an appropriate anti-tumor immune HLA class 1 Abcam Ab52922 1:250 RIgG R488 response by presenting antigenic peptides to cytotoxic T Active Upstate 05-665 1:100 MIgG1 M555 β-catenin lymphocytes [4]. One of the most frequent escape mech- CK18 Abcam Ab82254 1:50 MIgG1 M647 anisms that tumor cells develop to avoid immune system recognition is the loss of HLA-1 expression, which corre- lates with properties associated with ‘stemness’ resulting in treatment resistance and a poor prognosis [5]. responsible for the regulation of the immune system in The Wnt/β-catenin signaling pathway is an evolution- humans (human leukocyte antigen class I). HLA-1 down- arily conserved developmental pathway involved in sev- regulation is associated with stemness in multiple cancer eral physiological processes [6]. Aberrant activation of types [9]. β-catenin Active forms were detected with the the Wnt/β-catenin pathway is implicated in many tumor mouse monoclonal clone 8E7. Prior to multiplex assem- types. β-catenin, a key protein in the Wnt pathway, can bly, individual IF conditions were tested, with selection of trigger the transcription of important regulators of cell the optimal antibody-fluorochrome for each biomarker. cycle progression, cell proliferation and cell stemness The construction of the multiplex required a common upon activation and nuclear translocation [7]. β-catenin heat-induced epitope retrieval (HIER) strategy that expression has been shown to be associated with more incorporated the use of a pH9 solution in a decloaking malignant, high Gleason score tumors. Both β-catenin chamber at 120 °C for 3 min. After HIER, antibodies and and wnt-1 expressions are located in the advancing Zenons (invitrogen) were added in two addition steps, edge of the tumor [8]. The correlation between higher first HLA-1 combined with ABC, and then CK18. Image malignancy related to castrate resistant prostate cancer acquisition was performed within 48 h. The immunofluo- (CRPC) and β-catenin suggests Wnt/β-catenin signaling rescent multiplex reaction was first analyzed using appro- pathway upregulation in CRPC [6]. Here we demonstrate priate cell-line and tissue positive controls. A series of the association between nuclear localization of Active reproducibility experiments were carried out prior to the β-catenin and prostate tumor progression. final conditions applied to the TMA. These setup experi- ments were used to select the optimal conditions for the image acquisition, as the best filter combination and Main text exposition time for each of the filters. The imaging system Materials and methods included a multispectral camera (Perkin Elmer) attached Patient selection and TMA production ­ etamorph® to a 90i Nikon epifluorescent microscope with an autom- Prostate cancer samples were obtained from 505 patients atized stage controlled by the compatible m (age range 40–75, median 63 years) with localized software (Molecular Devices). Multispectral images were and locally advanced disease that underwent a radical acquired at 20× magnification with the Nuance imaging prostatectomy (Hospital de la Vall d’Hebron, Hospital ­ uance® software to system using the selected filter combination and expo- Clínic de Barcelona and Hospital 12 de Octubre). The Gleason breakdown of the tumors was as follows: 81 sition time, and analyzed with the N tumors Gleason < 6, 130 tumors Gleason 6, 293 tumors unmix and quantify individual biomarker expressions, as Gleason > 6. Whole tissue sections from each patient well as determining biomarker co-localization in selected were first stained and three tumor cores and a normal regions within each core. The individual expression of core previously selected were included in tissue microar- each marker is obtained as the result of the quantification of positive pixels for the marker that are included in the rays (TMAs). Blank 3.5 µM thick consecutive sections of tumor area. All expression and coexpression results were the TMA were used for IF analysis. obtained as an average of up to three cores from each patient in the case of tumor samples, and directly from Multiplex setup a single core of their normal counterpart. Statistical tests An Immunofluorescent (IF) multiplex reaction was were performed after all values were obtained from the assembled for the study of the following biomarkers: multispectral images (see “Statistical tests” subsection). Active β-catenin (ABC), HLA-1 and CK18, described in Table 1, along with the nuclear dye, DAPI (4′,6-diamid- Nuclear Active β‑catenin assessment ino-2-phenylindole, dilactate). CK18 was included The antibody for Active β-catenin (ABC) is able to detect to help identify epithelial cells and obtain colocaliza- nuclear, cytoplasmic and membranous ABC. The val- tion values with the other markers. HLA-1 is a member ues for nuclear β-catenin, accounting for the effector of the HLA system, a cell surface protein family that is Puig et al. BMC Res Notes (2019) 12:68 Page 3 of 5 fraction of the transcription factor, were deduced from is used to both select the glandular area and normalize the co-localization between DAPI and ABC in cells that the expression of the other two markers to CK18 expres- were expressing CK18 (see Additional file 1: Figure S1 sion. With this normalization, the expression of a marker for details on subcellular localization of ABC). ABC can inferred from the percentage of positive pixels in the be membranous, cytoplasmic and nuclear in cells with fluorescent picture is referred to the amount of glandu- activation of the wnt/β-catenin pathway. To avoid any lar compartment. Thus, the glandular normalization con- potential overestimation of nuclear ABC due to some tributes to a fair comparison between two images with background issue related to the cytoplasmic ABC expres- discordant amounts of tumor or normal glandular areas. sion a correction value was added. The correction fac- tor, the co-localization value between CK18 and DAPI, is based on the fact that CK18 is only expressed both in HLA Class 1 in tumor formation membrane and in cytoplasm and should not be found in HLA class 1 is a critical component of the immune co-localization with DAPI. Thus, the correction factor response and a candidate marker to show differences in should be zero in samples that are not affected by nuclear expression between normal and tumor areas. HLA class ABC overestimation. 1 downregulation in prostate cancer has been described in previous articles [10, 11]. HLA class 1 expression data Statistical tests was obtained from paired samples using the value for A Wilcoxon signed-rank test was applied to elucidate CK18 as a glandular normalization tool, as described in differences in the marker expression measures between the previous section. Following these criteria 383 out of tumor and adjacent normal areas. 505 patients of those included in the cohort were ana- lyzed with a Wilcoxon signed-rank test (Table 2). Our Results and discussion data shows a tendency for HLA class 1 downregula- Marker expression data tion in prostate tumor samples although the p value for The immunofluorescent images were acquired and the difference between normal and tumor expressions unmixed with the multispectral camera software. After is not as significant (p = 0.025) as expected according to unmixing, the individual marker expression and the coex- the results described in previous papers [10, 11]. A more pression values were obtained from images as depicted extensive cohort may be needed to shed more light on in Fig. 1. DAPI, HLA class I, Active β-catenin and CK18 the relation between HLA class 1 downregulation and are shown in discrete software assigned colors. DAPI was prostate cancer. used as a nuclear counterstain, and its coexpression with a given value applied to the assessment of this marker nuclear fraction. The other markers show a glandular Active β‑catenin in tumor formation pattern. CK18 is very important in this assay because it Nuclear Active β-catenin expression was found asso- ciated with prostate tumorigenesis according to the Wilcoxon signed-rank test calculation (p < 0.001). This correlation was obtained comparing the median values of normalized nuclear ABC in the tumor cores with the val- ues of the same normalized ABC fraction in normal cores (n = 383 tumor-normal paired samples). The same level of significance (p < 0.001) was obtained when total ABC expression was assessed and compared in tumor cores vs. normal cores (Table 2). Table 2 Wilcoxon signed-rank test for paired samples (n = 383) Marker Normal expression Tumor p-value Median [Q1; Q3] expression Median [Q1; Q3] Nuclear Active 2.05 [0.00; 7.12] 4.96 [0.46; 10.9] < 0.001 Fig. 1 Individual markers with DAPI and multiplex assembling. β-catenin Multiple fluorochrome labeled antibody hybridization on a prostate Total Active 0.28 [0.13; 0.58] 0.49 [0.20; 0.89] < 0.001 sample. Marker and DAPI expressions are shown in different colors; β-catenin HLA class I in green, CK18 in red, ABC in yellow and DAPI in blue HLA class I 0.29 [0.10; 0.70] 0.28 [0.16; 0.50] 0.025 Puig et al. BMC Res Notes (2019) 12:68 Page 4 of 5 In the current study, we present nuclear accumulation Abbreviations ABC: Active β-catenin; BCR: biochemical recurrence; CRPC: castrate resistant of Active β-catenin in prostate malignancy development. prostate cancer; TMA: tissue microarrays; DAPI: 4′,6-diamidino-2-phenylindole, This finding is in agreement with the published litera- dilactate; IF: immunofluorescent; HIER: heat-induced epitope retrieval. ture [7, 8, 12–14]. β-catenin, a dual effector protein that Authors’ contributions was first described in the regulation of intracellular adhe- Project design and supervision: PP, NE, MJD, and VGR. Data collection: NE, LM, sion, is also a key nuclear signaling protein in the activa- MTAO, MO, AA, ARA, JALM, IT, and JM. Pathological analysis at the hospitals: tion of the Wnt/β-catenin pathway. Downstream targets of AA, JLRP, IT, and JM. Pathological analysis in Atryshealth: MT. Statistical analysis: IS. Methodology Setup: PP, JGG, and MJD. Writing—original draft: PP. Writing β-catenin including c-Myc, Cyclin D1 and CD44, among review and editing: PP, NE, MT, AAP, MJD, LM, MTAO, MO, IT, and VGR. All others, are proliferation agents that are involved in onco- authors read and approved the final manuscript. genesis [15]. β-catenin signaling may play important roles Author details in prostate cancer progression [7] and in the acquisition of 1 R&D Department, Atrys Health, Barcelona, Spain. 2 Icahn School of Medicine tumor malignant phenotypes and the capacity for invasion at Mount Sinai, MSSM, New York City, USA. 3 Department and Laboratory through the induction of AR activity [13]. of Urology, Hospital Clínic, Institut d’Investigacions Biomèdiques August Pi iSunyer (IDIBAPS), Universitat de Barcelona, Barcelona, Spain. 4 Pathology Interestingly, the correlation between tumor progres- Department, Urology Department and Laboratory of Translational Oncology, sion and nuclear Active β-catenin is also found with total Instituto de Investigación Hospital, 12 de Octubre (i + 12), Madrid, Spain. ABC (nuclear, cytoplasmic and membranous). However, 5 Group of Biomedical Research in Urology, Vall d’Hebron Research Institute (VHIR) and Universitat Autònoma de Barcelona (UAB), Barcelona, Spain. when average expression values are compared, the differ- ence between tumor cores and normal cores is bigger for Acknowledgements nuclear Active β-catenin than for total Active β-catenin. We would like to thank the technical support received from the technician staff of the involved centers and the accurate language edition of Mr. Anthony Our data further supports a role for the Wnt/β-catenin Brady. signaling pathway in prostate cancer formation and as a potential therapeutic target. Furthermore, applications Competing interests The authors declare that they have no competing interests. which interrogate biomarkers at the intact tissue-cellular level will continue to advance our understanding of basic Availability of data and materials tumor biology. The datasets generated and/or analyzed during the current study are not publicly available due to the personal data they contain, but are available from the corresponding author on reasonable request. Conclusions Consent for publication To summarize, our multiplex-based systems pathology Not applicable. approach is a novel tool for the precise quantification of Ethics approval and consent to participate CK18 epithelial nuclear Active β-catenin through colo- The study was approved by the ethics committee of the hospitals, and all calization of DAPI and ABC. The additional evaluation enrolled patients provided signed informed consent. Hospital de la Vall d’Hebron: Clinical Research Ethics Committee (CEIC), PR_AG_313-2012. Hospi‑ of HLA provides insight into the biology underpinning tal Clínic de Barcelona: Clinical Research Ethics Committee (CEIC), 2012-8040. prostate cancer progression. Thus, the use of the mul- Hospital 12 de Octubre: Clinical Research Ethics Committee (CEIC), 13/165. tiplex approach is essential for a more comprehensive Funding analysis of various markers, including DAPI, HLA class I, Funding support has been received from The Spanish Ministery of Economy Active β-catenin and CK18. Our results show a trend for and Competitivity and the European Regional Development Funds (ERDF/ HLA Class I downregulation, and strongly support the FEDER). Grant # IPT-2012-1311-300000, cofunded by the ERDF and the Min‑ istery of Economy and Competitivity in the frame of the Programa Nacional implication of Active β-catenin, both nuclear and total, in de cooperación Público-Privada, Subprograma INNPACTO, within the línea prostate cancer development. Instrumental de Articulación e Internacionalización del Sistema, Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica 2008-2011. Limitations Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ The study needs an expanded cohort of patients and a lished maps and institutional affiliations. more robust assessment of HLA and nuclear ABC to fur- ther characterize such discrete cell populations. Received: 23 November 2018 Accepted: 19 January 2019 Additional file References 1. Siegel R, Miller K, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. Additional file 1: Figure S1. Subcellular localization of ABC. Magnified 2017;67:7–30. images demonstrating subcellular localization of ABC. 2. Saidi O, Cordon-Cardo C, Costa J. Technology insight: will systems pathol‑ ogy replace the pathologist? Nat Clin Pract Urol. 2007;4:39–45. Puig et al. BMC Res Notes (2019) 12:68 Page 5 of 5 3. Donovan MJ, Hamann S, Clayton M, Khan FM, Sapir M, Bayer-Zubek and genetically modified HLA-class-I-knockdown human embryonic V, Fernandez G, Mesa-Tejada R, Teverovskiy M, Reuter VE, Scardino PT, stem cells. Cell Sci. 2011;124:3029–37. Cordon-Cardo C. Systems pathology approach for the prediction of 10. Kitamura H, Torigoe T, Asanuma H, Honma I, Sato N, Tsukamoto T. prostate cancer progression after radical prostatectomy. J Clin Oncol. Down-regulation of HLA class I antigens in prostate cancer tissues and 2008;26(24):3923–9. up-regulation by histone deacetylase inhibition. J Urol. 2007;178:692–6. 4. Garrido C, Algarra I, Maleno I, Stefanski J, Collado A, Garrido F, Garcia-Lora 11. Carretero FJ, Del Campo AB, Flores-Martín JF, Mendez R, García-Lopez C, AM. Alterations of HLA class I expression in human melanoma xenografts Cozar JM, Adams V, Ward S, Cabrera T, Ruiz-Cabello F, Garrido F, Aptsiauri in immunodeficient mice occur frequently and are associated with N. Frequent HLA class I alterations in human prostate cancer: molecu‑ higher tumorigenicity. Cancer Immunol Immunother. 2010;59(1):13–26. lar mechanisms and clinical relevance. Cancer Immunol Immunother. 5. Nijland M, Veenstra RN, Visser L, Xu C, Kushekhar K, van Imhoff GW, Kluin 2016;65(1):47–59. https​://doi.org/10.1007/s0026​2-015-1774-5. PM, van den Berg A, Diepstra A. HLA dependent immune escape mecha‑ 12. Francis JC, Thomsen MK, Taketo MM, Swain A. β-catenin is required for nisms in B-cell lymphomas: implications for immune checkpoint inhibitor prostate development and cooperates with Pten loss to drive invasive therapy? Oncoimmunology. 2017;6(4):e1295202. carcinoma. PLoS Genet. 2013;9(1):e1003180. 6. Yokoyama NN, Shao S, Hoang BH, Mercola D, Zi X. Wnt signaling in 13. Lee SH, Luong R, Johnson DT, Cunha GR, Rivina L, Gonzalgo ML, Sun castration-resistant prostate cancer: implications for therapy. Am J Clin Z. Androgen signaling is a confounding factor for β-catenin-mediated Exp Urol. 2014;2:27–44. prostate tumorigenesis. Oncogene. 2016;35(6):702–14. 7. Yu X, Wang Y, Jiang M, Bierie B, Roy-Burman P, Shen MM, Taketo MM, 14. Wan X, Liu J, Lu JF, Tzelepi V, Yang J, Starbuck MW, Diao L, Wang J, Wills M, Matusik RJ. Activation of beta-Catenin in mouse prostate causes Efstathiou E, Vazquez ES, Troncoso P, Maity SN, Navone NM. Activation of HGPIN and continuous prostate growth after castration. Prostate. β-catenin signaling in androgen receptor-negative prostate cancer cells. 2009;69(3):249–62. Clin Cancer Res. 2012;18:726–36. 8. Chen G, Shukeir N, Potti A, Sircar K, Aprikian A, Goltzman D, Rabbani SA. 15. Mohammed MK, Shao C, Wang J, W Q, Wang X, Collier Z, Tang S, Liu H, Up-regulation of Wnt-1 and beta-catenin production in patients with Zhang F, Huang J, Guo D, Minpeng L, Liu F, Liu J, Ma C, Shi L, Athiviraham advanced metastatic prostate carcinoma: potential pathogenetic and A, He TC, Lee Michael J. Wnt/β-catenin signaling plays an ever-expanding prognostic implications. Cancer. 2004;101:1345–56. role in stem cell self-renewal, tumorigenesis and cancer chemoresistance. 9. Deuse T, Seifert M, Tyan D, Tsao PS, Hua X, Velden J, Eiermann T, Volk HD, Genes Dis. 2016;3(1):11–40. Reichenspurner H, Robbins RC, Schrepfer SJ. Immunobiology of naïve Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions