You are here

  • Home
  • Online First
  • Immunohistochemical analysis of the epithelial to mesenchymal transition in uterine carcinosarcoma

Article Text

Article menu

other Versions

Download PDFPDF

Original Article
Immunohistochemical analysis of the epithelial to mesenchymal transition in uterine carcinosarcoma
  1. Mitsumasa Osakabe1,
  2. Daisuke Fukagawa1,
  3. Chie Sato1,
  4. Ryo Sugimoto1,
  5. Noriyuki Uesugi1,
  6. Kazuyuki Ishida1,
  7. Hiroaki Itamochi2,
  8. Toru Sugiyama2 and
  9. Tamotsu Sugai1
  1. 1 Department of Molecular Diagnostic Pathology, School of Medicine, Iwate Medical University, Morioka, Japan
  2. 2 Department of Obstetrics and Gynecology, School of Medicine, Iwate Medical University, Morioka, Japan
  1. Correspondence to Tamotsu Sugai, Department of Obstetrics and Gynecology, School of Medicine, Iwate Medical University, Morioka, Japan; tsugai{at}iwate-med.ac.jp

Abstract

Objective Uterine carcinosarcoma (UCS) is a highly aggressive neoplasm that is composed of an intricate admixture of carcinomatous and sarcomatous elements. The relationship between UCS and the epithelial to mesenchymal transition (EMT) has been reported. In this study, we examined how expression of E-cadherin was associated with the expression of EMT-related proteins in UCS.

Methods UCS samples were histologically divided into three components: carcinomatous, transitional, and sarcomatous regions. Next, we examined the expression of E-cadherin and EMT-related proteins, including SNAI2, ZEB1, and TWIST1, in each component of the UCS using immunohistochemistry. The expression score was determined by combining the staining intensity and staining area of the target cells.

Results The expression score of E-cadherin was significantly lower in transitional and sarcomatous components than in the carcinomatous component. In addition, a significant difference in the low expression score of E-cadherin between transitional and sarcomatous components (transitional > sarcomatous components) was found. There were significant differences between the expression scores of ZEB1 in the three components (sarcomatous > transitional > carcinomatous components). However, no difference in the expression of TWIST1 between the components was found. Conversely, the expression level of SNAI2 was higher in sarcomatous or transitional components than in the carcinomatous component. However, a significant difference between the transitional and sarcomatous components was not detected.

Conclusion These results suggest that the EMT plays an essential role in the pathogenesis of UCS.

  • uterine carcinosarcoma
  • immunohistochemistry
  • epithelial to mesenchymal transition
  • SNAI2
  • ZEB1

https://doi.org/10.1136/ijgc-2018-000038

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    HIGHLIGHTS

    • SNAI2 and ZEB1 upregulate in transitional and sarcomatous areas in uterine carcinosarcoma.

    • The epithelial to mesenchymal transition plays an essential role in the pathogenesis of uterine carcinosarcoma.

    • Uterine carcinosarcoma may be representative of an ‘epithelial to mesenchymal transition’-related tumor.

    Introduction

    In female cancer patients, 4% of all diagnoses are for uterine cancer. As such, it is the sixth most common cancer in women around the world.1 Uterine carcinosarcoma (UCS) is a rare subtype of uterine cancer that constitutes less than 5% of all uterine cancers.2 3 UCS is highly aggressive and shows a high prevalence of chemo-resistance and a very poor prognosis compared with other types of uterine cancer.2 3 Therefore, understanding the underlying molecular alterations of UCS will contribute to advances in medical diagnosis and treatment of UCS.

    UCS is a biphasic neoplasm composed of both malignant epithelial and mesenchymal elements.2 The mesenchymal element of UCS originates from the malignant epithelial component based on findings that a transitional area between the malignant epithelial and mesenchymal elements is frequently found within the UCS.4–12 Recent studies have shown that the epithelial-mesenchymal transition (EMT), the histological change from epithelial elements to mesenchymal elements, plays an important role in the progression and metastasis of various cancers.13 14 Carcinosarcoma, which is composed of epithelial and mesenchymal components, may be a prototype of EMT-related neoplasia. The hallmark of the EMT is the loss of epithelial surface markers, most notably E-cadherin, and the acquisition of mesenchymal markers including vimentin and N-cadherin.15–17 The downregulation of E-cadherin during the EMT can be mediated by its transcriptional repression due to the binding of EMT transcription factors (SNAI1, SNAI2, ZEB1, and TWIST1) to E-boxes present in the E-cadherin promoter. However, the association of uterine carcinosarcoma with reduced expression of EMT-related proteins is not fully understood.18–24 The aim of this study was to identify the role of EMT-related proteins, including E-cadherin, SNAI2, ZEB1, and TWIST1, in UCS.

    Methods

    Patients

    Patient consent was obtained, and the study was approved by the Iwate Medical University Institutional Review Board. We used 22 UCS samples that were surgically resected in the Department of Obstetrics and Gynecology at Iwate Medical University Hospital, Japan, from January 2009 to April 2017. Patients who had undergone preoperative chemotherapy were excluded. The clinicopathological findings are summarized in Table 1.

    View this table:
    Table 1

    Clinicopathological findings for uterine carcinosarcomas in this study

    The research protocol was approved by the ethics committee of Iwate Medical University (No. H29-6).

    Sampling of the lesion

    Resected specimens were fixed in 10% buffered formalin and cut into 5 mm slices parallel to the sagittal section of the uterus. Paraffin-embedded blocks were made from slices including the deepest part of the infiltration.

    Histological examination

    Histological criteria and stage were determined according to the rules of the WHO Classification of Tumors of Female Reproductive Organs.25 For this study, archival slides of all cases were reviewed by two pathologists (MO and KI). In this study, we evaluated each of the three components.

    Immunohistochemical procedures

    All specimens were fixed in 10% formalin and embedded in paraffin wax. For this study, sections (3 µm) were cut and mounted on poly-L-lysine-coated glass slides (Matsunami, Tokyo, Japan). An automatic staining machine (DAKO Envision+system, DakoCytomation, Glostrup, Denmark) was used for the immunohistochemical procedure.26 The slides were counterstained in hematoxylin, dehydrated and mounted. The antibodies used in this study are shown in Table 2.

    View this table:
    Table 2

    List of primary antibodies

    Immunohistochemical assessment

    Two pathologists (MO and KI) performed independent semi-quantitative evaluations of the staining. Immunohistochemical positivity in the carcinomatous component, transitional area and the sarcomatous component were evaluated in a field measuring 0.950 mm2 through an objective lens (WHK 22×ocular lens; Olympus, Tokyo, Japan). The immunostaining intensity for target cells was classified into four categories according to staining intensity as follows: negative, weak, moderate, and strong. The immunostaining area of target cells was semi-quantified as follows: 0%, 0%–33%, 33%–66%, 66%–100%. The combination of intensity and area was scored. The immunohistochemical positivity score (IPS) was given by the product of staining intensity score (IS) and the staining proportion score (PS), namely “Embedded Image ”. Membrane immunostaining of E-cadherin was considered positive. Moreover, intra-nuclear staining of EMT-related proteins (SNAI2, ZEB1, and TWIST1) was also regarded as positive. When disagreements arose, slides were reviewed by two pathologists together and a consensus view was obtained.

    Statistical analysis

    Statistical analysis of data was performed using the statistical software JMP Pro 13 (SAS Institute Japan Inc., Tokyo, Japan) and PRISM6 (GraphPad Software, La Jolla, CA). A χ2 test and Mann-Whitney U test was performed for comparing two groups. A value of p<0.05 was considered a significant difference.

    Results

    The tumor tissues showed various histological differences in the carcinomatous area. The well to moderately differentiated component was selected to evaluate the targeted lesion. Determination of the transitional component was sometimes difficult, but intermediate histology between carcinomatous and sarcomatous components could be selected. Finally, a hot spot was defined in most stained areas and was selected to examine immunohistochemical scores of target cells. Representative illustrations of carcinomatous, transitional and sarcomatous components are shown in Figures 1–3, respectively.

    Figure 1
    Figure 1

    Histological and immunohistochemical findings of the carcinomatous component of carcinosarcoma. (A) Hematoxylin and eosin stain. (B) Membrane staining for E-cadherin (staining intensity score (IS) 3; staining proportion score (PS) 3; immunohistochemical positivity score (IPS) 9). (C). Negative nuclear staining for SNAI2 (IS 0; PS 0; IPS 0). (D) Negative nuclear staining for ZEB1 (IS 0; PS 0; IPS 0). (E) Negative nuclear staining for TWIST1 (IS 0; PS 0; IPS 0).

    Figure 2
    Figure 2

    Histological and immunohistochemical analysis of the transitional component of carcinosarcoma. (A) Hematoxylin and eosin stain. (B) Membranous staining for E-cadherin (staining intensity score (IS) 1; staining proportion score (PS) 1; immunohistochemical positivity score (IPS) 1). (C) Nuclear staining for SNAI2 (IS 1; PS 3; IPS 3). (D) Nuclear staining for ZEB1 (IS 1; PS 1; IPS 1). (E) Negative nuclear staining for TWIST1 (IS 0; PS 0; IPS 0).

    Figure 3
    Figure 3

    Histological and immunohistochemical analysis of the sarcomatous component of carcinosarcoma. (A) Hematoxylin and eosin stain. (B) Negative membrane staining for E-cadherin (staining intensity score (IS) 0; staining proportion score (PS) 0; immunohistochemical positivity score (IPS) 0). (C) Nuclear staining for SNAI2 (IS 3; PS 1; IPS 3). (D). Nuclear staining for ZEB1 (IS 3; PS 3; IPS 9). (E) Negative nuclear staining for TWIST1 (IS 0; PS 0; IPS 0).

    Immunohistochemical positivity of E-cadherin

    The immunohistochemical scores for E-cadherin differed significantly between the carcinomatous, transitional, and sarcomatous components (p<0.0001). In addition, statistically significant differences in the immunohistochemical score of E-cadherin between transitional and sarcomatous components were also found (p<0.0001) (Figure 4A).

    Figure 4
    Figure 4

    The immunohistochemical positivity scores in the carcinoma, transitional, and sarcomatous components. Each dot represents the immunohistochemical positivity score in each case. The horizontal line represents the median value. (A) Immunohistochemical score of E-cadherin. (B) Immunohistochemical score of SNAI2. (C) Immunohistochemical score of ZEB1. (D) Immunohistochemical score of TWIST1.

    Immunohistochemical positivity of EMT transcription factors

    Immunohistochemical positivity of SNAI2

    The immunohistochemical score of SNAI2 was significantly higher in sarcomatous components than in carcinomatous components (p<0.0001). Moreover, there were also significant differences in the immunohistochemical scores of SNAI2 in transitional vs carcinomatous components (p=0.0125) (Figure 4B).

    Immunohistochemical positivity of ZEB1

    There were significant differences in the immunohistochemical scores of ZEB1 between sarcomatous, carcinomatous, and transitional components (p<0.0001). In addition, the immunohistochemical score of ZEB1 was also significantly higher in sarcomatous components than in transitional components (p<0.0001) (Figure 4C).

    Immunohistochemical positivity of TWIST1

    There were no differences in the immunohistochemical scores of TWIST1 between the three components (Figure 4D).

    Discussion

    UCS consists of a group of biphenotypic tumors that simultaneously show both epithelial and mesenchymal elements. UCS is widely accepted as a metaplastic carcinoma.27 UCS is a highly aggressive tumor compared with non-endometrioid carcinoma.2 The malignant mesenchymal component of UCS is thought to originate from the malignant epithelial component.4–12 Epithelial elements undergo an EMT, changing to mesenchymal elements, an alteration that plays an important role in cancer invasion and metastasis.13 14 This study attempted to identify the association of the EMT with UCS carcinogenesis. In the present study, the EMT played a major role in the pathogenesis of UCS.

    The EMT is defined as a morphological change of epithelial elements to mesenchymal elements.13 14 Loss of E-cadherin is an important characteristic finding for the EMT.15–17 In the present study, the expression of E-cadherin decreased in the following order: the carcinomatous component, the transitional component, and the sarcomatous component. Reduced E-cadherin expression causes morphological changes in epithelial cells, altering their phenotype from epithelial-like features to more fibroblastic.28 29 The morphological changes from the carcinomatous component to the sarcomatous component in UCS may be caused by decreased expression of E-cadherin. In the present study, the reduction of E-cadherin expression was closely associated with a change from a carcinomatous histology to a sarcomatous one. This finding suggests that loss of E-cadherin expression plays a major role in the pathogenesis of UCS.

    The expression of E-cadherin during the EMT is regulated by EMT transcription factors, including SNAI1, SNAI2, ZEB1, and TWIST1, at E-boxes present in the E-cadherin promoter.15–17 Several reports have shown that the expression of SNAI2, ZEB1, and TWIST1 is higher in sarcomatous components than in carcinoma components of UCS.19 21 23 30 These reports compared the carcinomatous component with the sarcomatous component using molecular biological techniques by cell culture. In this study, however, we used immunohistochemistry to identify the role of EMT-related proteins in this disease. The reason why our results using immunohistochemistry differ from previous results may be due to the use of different methods.

    The development of sarcomatous components in UCS is enhanced through transcriptional upregulation of the SNAI2, ZEB1, and TWIST1 genes.19 21 23 30 In the present study, although SNAI2 expression was significantly higher in transitional components than in carcinomatous components, the differences in expression between the sarcomatous components and the transitional areas were not significant. Moreover, there were statistically significant differences in the expression of ZEB1 between the three components. This result indicates that expression of SNAI2 and ZEB1 greatly contributes to the development of the mesenchymal transition of UCS, suggesting this tumor may be an EMT-related tumor. Finally, we showed that expression of ZEB1 may be an enhanced factor in the pathogenesis of UCS.

    In conclusion, we suggest that the biphenotypic character of UCS is the result of the EMT caused by upregulation of SNAI2 and ZEB1. In addition, this is the first report describing the association of uterine carcinosarcoma with an EMT-like phenomenon. From these results, we propose that UCS may be representative of an EMT-related tumor.

    Acknowledgments

    The authors gratefully acknowledge the support and technical assistance of members of the Department of Molecular Diagnostic Pathology, Iwate Medical University.

    References

    1. 1.↵
      2. Torre LA ,
      3. Bray F ,
      4. Siegel RL , et al
      . Global cancer statistics, 2012. CA Cancer J Clin 2015;65:87–108.doi:10.3322/caac.21262
      OpenUrlCrossRefPubMed
    2. 2.↵
      2. Amant F ,
      3. Cadron I ,
      4. Fuso L , et al
      . Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high-risk epithelial endometrial cancer. Gynecol Oncol 2005;98:274–80.doi:10.1016/j.ygyno.2005.04.027
      OpenUrlCrossRefPubMed
    3. 3.↵
      2. Nam JH ,
      3. Park JY
      . Update on treatment of uterine sarcoma. Curr Opin Obstet Gynecol 2010;22:36–42.doi:10.1097/GCO.0b013e328334d90f
      OpenUrlPubMed
    4. 4.↵
      2. Silverberg SG
      . Malignant mixed mesodermal tumor of the uterus: an ultrastructural study. Am J Obstet Gynecol 1971;110:702–12.doi:10.1016/0002-9378(71)90258-4
      OpenUrlPubMed
    5. 5.↵
      2. Gorai I ,
      3. Yanagibashi T ,
      4. Taki A , et al
      . Uterine carcinosarcoma is derived from a single stem cell: an in vitro study. Int J Cancer 1997;72:821–7.doi:10.1002/(SICI)1097-0215(19970904)72:5&lt;821::AID-IJC19&gt;3.0.CO;2-B
      OpenUrlCrossRefPubMedWeb of Science
    6. 6.↵
      2. Auerbach HE ,
      3. LiVolsi VA ,
      4. Merino MJ
      . Malignant mixed Müllerian tumors of the uterus. An immunohistochemical study. Int J Gynecol Pathol 1988;7:123–30.doi:10.1097/00004347-198805000-00003
      OpenUrlPubMed
    7. 7.↵
      2. George E ,
      3. Manivel JC ,
      4. Dehner LP , et al
      . Malignant mixed Müllerian tumors: an immunohistochemical study of 47 cases, with histogenetic considerations and clinical correlation. Hum Pathol 1991;22:215–23.doi:10.1016/0046-8177(91)90153-G
      OpenUrlCrossRefPubMedWeb of Science
    8. 8.↵
      2. Szukala SA ,
      3. Marks JR ,
      4. Burchette JL , et al
      . Co-expression of p53 by epithelial and stromal elements in carcinosarcoma of the female genital tract: an immunohistochemical study of 19 cases. Int J Gynecol Cancer 1999;9:131–6.doi:10.1046/j.1525-1438.1999.09905.x
      OpenUrlCrossRefPubMedWeb of Science
    9. 9.↵
      2. Abeln EC ,
      3. Smit VT ,
      4. Wessels JW , et al
      . Molecular genetic evidence for the conversion hypothesis of the origin of malignant mixed Müllerian tumours. J Pathol 1997;183:424–31.doi:10.1002/(SICI)1096-9896(199712)183:4&lt;424::AID-PATH949&gt;3.0.CO;2-L
      OpenUrlCrossRefPubMedWeb of Science
    10. 10.↵
      2. Wada H ,
      3. Enomoto T ,
      4. Fujita M , et al
      . Molecular evidence that most but not all carcinosarcomas of the uterus are combination tumors. Cancer Res 1997;57:5379–85.
      OpenUrlAbstract/FREE Full Text
    11. 11.↵
      2. Kounelis S ,
      3. Jones MW ,
      4. Papadaki H , et al
      . Carcinosarcomas (malignant mixed mullerian tumors) of the female genital tract: comparative molecular analysis of epithelial and mesenchymal components. Hum Pathol 1998;29:82–7.doi:10.1016/S0046-8177(98)90394-X
      OpenUrlCrossRefPubMedWeb of Science
    12. 12.↵
      2. Fujii H ,
      3. Yoshida M ,
      4. Gong ZX , et al
      . Frequent genetic heterogeneity in the clonal evolution of gynecological carcinosarcoma and its influence on phenotypic diversity. Cancer Res 2000;60:114–20.
      OpenUrlAbstract/FREE Full Text
    13. 13.↵
      2. Thiery JP ,
      3. Acloque H ,
      4. Huang RY , et al
      . Epithelial-mesenchymal transitions in development and disease. Cell 2009;139:871–90.doi:10.1016/j.cell.2009.11.007
      OpenUrlCrossRefPubMedWeb of Science
    14. 14.↵
      2. Polyak K ,
      3. Weinberg RA
      . Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer 2009;9:265–73.doi:10.1038/nrc2620
      OpenUrlCrossRefPubMedWeb of Science
    15. 15.↵
      2. Peinado H ,
      3. Olmeda D ,
      4. Cano A
      . Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev Cancer 2007;7:415–28.doi:10.1038/nrc2131
      OpenUrlCrossRefPubMedWeb of Science
    16. 16.↵
      2. Wang J ,
      3. Lee S ,
      4. Teh CE , et al
      . The transcription repressor, ZEB1, cooperates with CtBP2 and HDAC1 to suppress IL-2 gene activation in T cells. Int Immunol 2009;21:227–35.doi:10.1093/intimm/dxn143
      OpenUrlCrossRefPubMed
    17. 17.↵
      2. Nieto MA
      . The ins and outs of the epithelial to mesenchymal transition in health and disease. Annu Rev Cell Dev Biol 2011;27:347–76.doi:10.1146/annurev-cellbio-092910-154036
      OpenUrlCrossRefPubMed
    18. 18.↵
      2. Cherniack AD ,
      3. Shen H ,
      4. Walter V , et al
      . Integrated molecular characterization of uterine carcinosarcoma. Cancer Cell 2017;31:411–23.doi:10.1016/j.ccell.2017.02.010
      OpenUrl
    19. 19.↵
      2. Inoue H ,
      3. Hashimura M ,
      4. Akiya M , et al
      . Functional role of ALK-related signal cascades on modulation of epithelial-mesenchymal transition and apoptosis in uterine carcinosarcoma. Mol Cancer 2017;16:37.doi:10.1186/s12943-017-0609-8
    20. 20.↵
      2. Zhao S ,
      3. Bellone S ,
      4. Lopez S , et al
      . Mutational landscape of uterine and ovarian carcinosarcomas implicates histone genes in epithelial-mesenchymal transition. Proc Natl Acad Sci U S A 2016;113:12238–43.doi:10.1073/pnas.1614120113
      OpenUrlAbstract/FREE Full Text
    21. 21.↵
      2. Inoue H ,
      3. Takahashi H ,
      4. Hashimura M , et al
      . Cooperation of Sox4 with β-catenin/p300 complex in transcriptional regulation of the Slug gene during divergent sarcomatous differentiation in uterine carcinosarcoma. BMC Cancer 2016;16:53.doi:10.1186/s12885-016-2090-y
    22. 22.↵
      2. Dwivedi SK ,
      3. McMeekin SD ,
      4. Slaughter K , et al
      . Role of TGF-β signaling in uterine carcinosarcoma. Oncotarget 2015;6:14646–55.doi:10.18632/oncotarget.3711
      OpenUrl
    23. 23.↵
      2. Díaz-Martín J ,
      3. Díaz-López A ,
      4. Moreno-Bueno G , et al
      . A core microRNA signature associated with inducers of the epithelial-to-mesenchymal transition. J Pathol 2014;232:319–29.doi:10.1002/path.4289
      OpenUrlCrossRefPubMed
    24. 24.↵
      2. Chiyoda T ,
      3. Tsuda H ,
      4. Tanaka H , et al
      . Expression profiles of carcinosarcoma of the uterine corpus - are these similar to carcinoma or sarcoma? Genes Chromosomes Cancer 2012;51:229–39.doi:10.1002/gcc.20947
      OpenUrlPubMed
    25. 25.↵
      2. Wells M ,
      3. Palacios J ,
      4. Oliva E
      . Mixed epithelial and mesenchymal tumours. In: Kurman RJ , Carcangiu ML , Herrington CS , eds. World Health Organization Classification of Tumours of Female Reproductive Organs 2014. Lyon: IARC Press, 2014: 148–50.
    26. 26.↵
      2. Sugai T ,
      3. Habano W ,
      4. Uesugi N , et al
      . Three independent genetic profiles based on mucin expression in early differentiated-type gastric cancers - a new concept of genetic carcinogenesis of early differentiated-type adenocarcinomas. Mod Pathol 2004;17:1223–34.doi:10.1038/modpathol.3800170
      OpenUrlPubMed
    27. 27.↵
      2. McCluggage WG
      . Malignant biphasic uterine tumours: carcinosarcomas or metaplastic carcinomas? J Clin Pathol 2002;55:321–5.doi:10.1136/jcp.55.5.321
      OpenUrlAbstract/FREE Full Text
    28. 28.↵
      2. Mayer B ,
      3. Johnson JP ,
      4. Leitl F , et al
      . E-cadherin expression in primary and metastatic gastric cancer: down-regulation correlates with cellular dedifferentiation and glandular disintegration. Cancer Res 1993;53:1690–5.
      OpenUrlAbstract/FREE Full Text
    29. 29.↵
      2. Saito T ,
      3. Oda Y ,
      4. Sugimachi K , et al
      . E-cadherin gene mutations frequently occur in synovial sarcoma as a determinant of histological features. Am J Pathol 2001;159:2117–24.doi:10.1016/S0002-9440(10)63063-5
      OpenUrlPubMed
    30. 30.↵
      2. Castilla MÁ ,
      3. Moreno-Bueno G ,
      4. Romero-Pérez L , et al
      . Micro-RNA signature of the epithelial-mesenchymal transition in endometrial carcinosarcoma. J Pathol 2011;223:72–80.doi:10.1002/path.2802
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors MO, the first author, contributed to the preparation of the manuscript, including all aspects of data collection and analysis. TS, the corresponding author, contributed to writing the manuscript. DF and CS constructed the figures and tables. RS and NU supported statistical analysis. KI provided input during the preparation of the manuscript. HI and TS assisted with clinical data and experiments.

    • Patient consent Obtained.

    • Ethics approval All procedures were performed in accordance with the ethical standards of the Iwate Medical University and with the Declaration of Helsinki.

    • Provenance and peer review Not commissioned, externally peer reviewed.