Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Original Contribution
Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma Siham Chaouche-Mazouni a,⁎, Arnaud Scherpereel b, Rima Zaamoum c, Adriana Mihalache d, Zine-Charaf Amir e, Nemcha Lebaïli a, Baptiste Delaire d, Pierre Gosset d a
Department of Biology, Kouba High School, Algiers, Algeria Department of Pulmonary and Thoracic Oncology, Lille University Hospital, Lille, France Department of Pneumophtysiology, Public Hospital of Rouiba, Algeria d Department of Pathology, Hospital of St Vincent, GHICL, Lille, France e Department of Pathology, Hospital of Mustapha Pacha, Algiers, Algeria b c
a r t i c l e
i n f o
Available online xxxx Keywords: Mesothelioma Lung adenocarcinoma Claudins Diagnosis Therapeutic target
a b s t r a c t Epithelioid malignant pleural mesothelioma (MPM) can easily be confused with lung adenocarcinomas (ACAs). In serous effusion, claudin (cldn) 3 is shown to be useful in the diagnosis of mesothelioma vs ACAs. Cldn15 is reported to be overexpressed in epithelioid mesothelioma and absent in human airway epithelium. The aim was to assess the value of cldn3 and cldn4 compared to that of BerEp4 and thyroid transcription factor–1 (TTF1) in differentiating lung ACA from epithelioid MPM and to examine the expression of cldn15 in these tumors. The expression of cldn3, cldn4, cldn15, BerEp4, and TTF1 was examined by immunohistochemistry in a total of 62 human specimen including 28 epithelioid MPMs and 34 ACAs of the lung. In lung ACA, cldn4 was strongly expressed in all 34 (100%) specimens followed by cldn3 in 33 (97%) of 34. BerEp4 was expressed in 32 (94.1%) of 34. TTF1 reacted for only 20 (58.82%) of 34 cases of lung ACA. In MPM specimens, the expression of cldn3 and4 as well as that of TTF1 was completely absent. In contrast, BerEp4 was focally expressed in 5 (17.85%) of 28 cases of epithelioid MPM. Cldn15 was strongly expressed in 53% pf epithelioid MPMs but also in 50% of lung ACAs. Its expression was moderate in normal pleura and limited in normal lung. Cldn3 and cldn4 appear to be the best performing carcinoma markers in discriminating lung ACA from mesothelioma compared with BerEp4 and TTF1. There is no differential expression of cldn15 between the 2 pathologies. However, the limited cldn15 expression in normal tissues and high expression in tumors make it an attractive candidate for cancer therapy. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Malignant pleural mesothelioma (MPM) is a highly aggressive tumor with difficult diagnosis. Differentiating between the most common, epithelioid form of mesothelioma and lung adenocarcinoma (ACA) is often problematic. Immunohistochemistry has greatly improved the diagnosis of mesothelioma. However, as there is no specific biomarker, the diagnosis is based in part on the negativity of markers that characterize other pathological pleural entities [1,2]. Emerging studies suggest the interest of claudin expression in cancer diagnosis and therapy [3]. Claudins are a family of 27 proteins that constitute the major components of tight junctions [4–7]. In the last decade, the role of some claudins in distinguishing mesothelioma from carcinomas has been reported. In 2002, Gordon et al [8] showed ⁎ Corresponding author at: Department of Biology, Kouba high School, Algiers, Algeria. Tel.: +213 664747387, +213 552115971. E-mail addresses: [email protected], [email protected] (S. Chaouche-Mazouni).
a highly differential gene expression of claudin 7 between mesothelioma and lung ACA. Some years later, Holloway et al [9] listed claudin 3 and claudin 7 among the highly differentially expressed genes between the 2 pathologies. Simultaneously, Soini et al [10] showed that cldn1, 3, 4, 5, and 7 protein expression could be used as an adjunct in the differential diagnosis between these tumors. Kleinberg et al [11] reported the usefulness of claudin1, 3, and 7 in serous effusion; and Facchetti et al [12,13] reported the usefulness of cldn4 in pleural and peritoneal biopsies and effusions. We previously found, in a limited number of specimens, the absence of cldn3 and 4 in MPM and their expression in 100% of lung ACAs [14]. Recently, Ordóñez [15] and Ohta et al [16] confirmed the diagnostic utility of cldn4 in distinguishing mesothelioma from ACAs. Taken together, these data highlight the interest of cldn3 and 4 in mesothelioma diagnosis. However, the number of published reports is limited; and the use of these proteins in routine practice still requires validation by further studies especially for claudin 3. In the present work, we compare the value of cldn3 and 4 with that of the 2 markers, BerEp4 and thyroid transcription factor–1 (TTF1), in 62 human
http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007 1092-9134/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
2
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Table 1 Antibodies used. Antibody Calretinin CK5/6 WT1 BerEp4 TTF1 Cldn3 Cldn4 Cldn15
Clone 5A5 D5/16B4 6F-H2 Ber-Ep4 8G7G3/1 Ab15102 Ab15104 NBP2-13842
Host and clonality Mouse monoclonal Mouse monoclonal Mouse monoclonal Mouse monoclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal
Source Leica Dako Roche Dako Dako Abcam Abcam Novus Bio
specimens of lung ACA and epithelioid MPM, currently used in the panel for differentiating between the 2 pathologies. Moreover, claudin 15 has been listed among the highly expressed genes in epithelioid MPM [17] and in diffuse peritoneal malignant mesothelioma [18]. We previously found high levels of cldn15 protein in MPM cell lines [14]. The expression of cldn15 was detected in mesothelial cells of the pleura and peritoneum in mice [19] but was absent in bronchial human airways [20]. In this study, we examined the expression of cldn15 in human specimens of epithelioid mesothelioma and lung ACA to see whether it is differentially expressed between these tumors. 2. Material and methods Sixty-two (62) human specimens including epithelioid MPMs (n= 28) and lung ACAs (n=34), in addition to normal control tissues (colon, pleura, bronchi, and bronchioles), were used to examine the expression of cldn3, cldn4, cldn15, BerEp4, and TTF1 in addition to calretinin, cytokeratin 5/6, and WT1 to confirm the diagnosis of mesothelioma. Surgical resection specimens were provided by the CHU of Mustapha Pacha (Algiers, Algeria) and the Public Hospital of Blida (Algeria). No biopsy material was used because there was insufficient material for immunohistochemical analysis. The diagnosis of all specimens was validated by pathologists using both morphologic criteria and immunohistochemistry with a broad antibody panel that recognizes carcinoma or mesothelioma. In case of TTF1-negative cases, the diagnosis of primary lung carcinoma was based on clinicopathological
Dilution Prediluted 1:50 Prediluted 1:25 1:50 1:100 1:200 1:200
Incubation (min) 32 32 44 32 32 20 20 32
Pretreatment Citrate buffer pH 6.0 60 min EDTA, pH 8.4 60 mn EDTA, pH 8.4 60 min None Citrate buffer, pH 6.0 60 min Citrate buffer, pH 6.0 60 min EDTA, pH 8.4 36 min Citrate buffer, pH 6.0 90 min
correlation. Nevertheless, the primary vs secondary origin of ACA was less important for this study because it was aimed at differentiating between ACAs and epithelioid form of mesotheliomas. 2.1. Immunohistochemistry Formalin-fixed, paraffin-embedded samples were cut (4 μm–thick sections) and placed on silane-covered slides. Morphological assessment of the samples was obtained by examining sections stained with hematoxylin-eosin-saffron. Slides were stained with anti-cldn3, anticldn4, anti-cldn15, anti-BerEp4, anti-TTF1, anti-calretinin, anti-keratin 5/6, and anti-WT1 using a Ventana Ultraview DAB detection kit in a Ventana BenchMark XT processor (Ventana, Tucson, AZ). For the antibody dilutions, pretreatment, and sources, see Table 1. An external positive control section was included in each immunohistochemical analysis (1 section for each slide). The signal was interpreted as present (positive) or absent (negative) irrespective of the intensity of the signal. When present, there were at least several cells stained; and the intensity was at least moderate in all cases. 3. Results In lung ACAs, cldn4 (Fig. 1A) was strongly expressed along the cell membrane in all 34 specimens (100%). Cldn3 (Fig. 1B) was expressed along the cell membrane in 33 (97%) of 34 cases with less intensity of immunostaining than cld4. The immunostaining of BerEp4 (Fig. 1C) was as intense as that of cldn4 along the cell membrane in 32 (94.1%)
a
b
c
d
Fig. 1. Lung ACA immunostaining. (A) cldn-4 immunostaining (intense membrane positivity) ×400. (B) cldn-3 immunostaining (membrane positivity) ×400. (C) BerEp4 immunostaining (intense membrane positivity) ×200. (D) TTF1 immunostaining (nuclear positivity) ×400.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
3
a
b
c
d
Fig. 2. Epithelioid MPM immunostaining. (A) cldn-4 immunostaining (absence of expression) ×400. (B) cldn-3 immunostaining (absence of expression) ×400. (C) BerEp4 immunostaining (intense membrane positivity) ×400. (D) TTF1 immunostaining (absence of expression) ×400.
of 34 cases of lung ACA. TTF1 (Fig. 1D) was positive in only 20 (58.82%) of 34 cases of lung ACA, showing a nuclear staining. The 2 BerEp4negative cases of lung ACA were positive for TTF1. In MPM specimens, the expression of cldn4 (Fig. 2A) and cldn3 (Fig. 2B), as well as that of TTF1 (Fig. 2D), was completely absent, whereas BerEp4 (Fig. 2C) focally stained the cell membrane in 5 (17.85%) of 28 epithelioid MPM cases. In normal colon tissues used as control, cldn15 was expressed in the cytoplasm and along the cell membrane of epithelial cells (Fig. 3A). In tumors, cldn15 was highly expressed in the cytoplasm and cell membrane of 53% of epithelioid MPMs (Fig. 3B) but also in 50% of lung ACAs (Fig. 3C). In normal pleura (Fig. 3D), mesothelial cells reacted with cldn15 less intensely than in MPM. A weak cytoplasmic positivity
of cldn15 was observed in bronchioles (Fig. 3E) and bronchi (Fig. 3F) of normal airways. Results are summarized in Table 2. 4. Discussion Epithelioid MPM can easily be confused with lung ACA. Several markers are currently used in the diagnosis of MPM vs lung ACA. However, none of them is sufficiently discriminant. The most valuable tissue markers for mesothelial cells are calretinin (only the nuclear staining has diagnostic value), WT1 (nuclear staining), cytokeratin 5/6 (cytoplasmic staining), and epithelial membrane antigen (EMA; membranous staining) as positive markers and carcinoembryonic antigen
a
b
c
d
e
f
Fig. 3. Immunostaining of cldn 15. (A) Normal colon (intense membrane and cytoplasmic positivity) ×400. (B) MPM (intense cytoplasmic positivity) ×400. (C) Lung ACA (intense membrane and cytoplasmic positivity) ×400. (D) Normal pleura (membrane and cytoplasmic positivity) ×400. (E) Bronchioles (weak cytoplasmic positivity) ×400. (F) Bronchus (weak cytoplasmic positivity) ×400.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
4
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Table 2 Results of cldn 3, 4, and 15 immunostaining compared with BerEp4 and TTF1.
Lung ADC (n = 34) Epithelioid MPM (n = 28)
Cldn3
Cldn4
Cldn15
BerEp4
TTF1
CK5/6
Calretinin
33 (97%) 0 (0%)
34 (100%) 0 (0%)
17 (50%) 15 (53%)
32 (94.1%) 5 (17.85%)
20 (58.82%) 0 (0%)
3 (8.8%) 15 (53%)
2 (5.88%) 20 (71.42%)
(cytoplasmic staining), BerEp4 (membranous staining), TTF1 (nuclear staining), and EMA (cytoplasmic staining) as negative markers. The International Mesothelioma Panel recommends that at least 2 mesothelial and 2 carcinoma markers be used in addition to a pancytokeratin for a reliable diagnosis of MPM [2,21]. Recently, cldn3 [10–12,14] and cldn4 [11,13–16] have been shown to be highly differentially expressed between carcinomas and mesotheliomas. However, whereas the cldn4 protein has been well evaluated as a useful marker in distinguishing between these tumors [12–16], only 1 study reported the usefulness of cldn3 in serous effusion [11]. Coordinate expression has been reported for cldn3 and cldn4 in several normal and neoplastic tissues [22]. In the present study, we evaluated the expression of cldn3 in human epithelioid MPMs and lung ACAs to see whether it performs similar to cldn4 in discriminating between the 2 pathologies compared with BerEp4 and TTF1. Cdn4 is reported to be the most interesting marker in distinguishing lung ACA from mesothelioma, with 100% specificity and sensitivity [12–16]. In this study, we found cldn4 to be strongly expressed in 100% of lung ACA specimens and completely absent in MPMs, confirming once again the results previously obtained and thus highlighting its usefulness in the differential diagnosis between these tumors. In the case of cldn3, Soini et al [10] reported a weak positivity in 6 (18%) of 33 epithelioid mesotheliomas. Kleinberg et al [11] found a limited cldn3 positivity in 1 (4%) of 25 serous effusions but no cldn3 expression in 13 solid tumors of mesothelioma. In contrast, Soini et al [10] showed that this protein was expressed in 21 (88%) of 24 ACAs of different origin, whereas Kleinberg et al found it positive in 7 (88%) of 8 serous effusions and 5 (100%) of 5 solid tumours of lung ACA. In the present study, we found cldn3 to be completely absent in all 32 MPMs and positive in 33 (97%) of 34 lung ACAs, showing its potential use in the differential diagnosis of the 2 pathologies. Among the immunohistochemical markers most frequently evaluated in the diagnosis of mesothelioma, BerEp4 was the second most commonly used marker after EMA, followed by carcinoembryonic antigen and calretinin [2,23]. BerEp4 is highly sensitive for ACA and currently is the most common carcinoma marker used to distinguish between lung ACA and epithelioid mesothelioma. However, it is reported that BerEp4 is not specific because 20% of MPMs are focally positive [21,24]. In this work, the results showed BerEp4 to be highly and strongly sensitive for lung ACA (94.1%); but at the same time, it reacted with 17.85% of epithelioid MPMs, highlighting its lack of specificity compared to that of cldn3 and cldn4. TTF1 is the most used marker for confirming lung origin of ACA. It is never expressed in mesothelioma, but its limited sensitivity (75%) [21] for lung ACA may decrease its usefulness. In our experience, TTF1 was actually absent in MPM specimens but was detected in only 58.82% of lung ACA. We note that the 2 BerEp4-negative cases in lung ACA were positive for TTF1. However, even though the combination of BerEp4 and TTF1 increased their sensitivity for lung ACA, the highest sensitivity of cldn4 and cldn3 and their 100% specificity make them more useful in discriminating between lung ACA and epithelioid MPM. Furthermore, previous studies [12,15] showed that the 2 proteins were expressed not only in lung ACA but in all metastatic carcinomas that metastasize to the serosal membranes. Claudin15 gene is found to be overexpressed in epithelioid MPM [17] and in diffuse peritoneal malignant mesothelioma [18]. We previously reported high levels of cldn15 protein in MPM cell lines [14]. On the other hand, Coyne et al [20] showed that cldn15 expression is absent in bronchial and bronchiolar epithelium in human airways. In human
MPM specimens, to our knowledge, cldn15 expression has not been reported yet. In this study, we examined cldn15 protein expression to see whether it is differentially expressed between MPM and lung ACA. Our results showed that immunohostochemically cldn15 is strongly expressed in 53% of epithelioid MPMs. However, 50% of lung ACAs also reacted with cldn15. When compared with normal tissues, we found cldn15 to be moderately expressed in normal pleura and weakly expressed in bronchi and bronchioles. These findings demonstrated little value of the polyclonal rabbit anti-claudin 15 in the differential diagnosis between lung ACA and MPM on paraffin section. However, because the overexpression of cldn15 has been linked to tumor progression [25], the limited cldn15 expression in normal lung and pleura and high expression in tumors make it an attractive candidate for cancer therapy. 5. Conclusion Among all carcinoma markers currently used, the high specificity and sensitivity of cldn3 and cldn4 make them useful markers for discriminating between lung ADC and MPM. These two markers should be included in the panel for the diagnosis of epithelioid mesothelioma. The polyclonal anti-cldn15 antibody used in this study showed no diagnostic utility in differentiating between the 2 pathologies. However, the limited cldn15 expression in normal tissues and overexpression in tumors could have an interest as a potential therapeutic target. Conflict of interest None to declare. References [1] Scherpereel A, et al. Guidelines of the European Respiratory Society and the European Society of Thoracic Surgeons for the management of malignant pleural mesothelioma. Eur Respir J 2010;35:479–95. [2] Astoul P, Roca E, Galateau Sallé F, Scherpereel A. Malignant pleural mesothelioma: from the bench to the bedside. Respiration 2012;83:481–93. [3] Morin JP. Claudin proteins in human cancer: promising new targets for diagnosis and cancer therapy. Cancer Res 2005;65:9603–6. [4] Morita K, Furuse M, Fujimoto K, Tsukita S. Claudin multigene family encoding fourtransmembrane domain protein components of tight junction strands. Proc Natl Acad Sci 1999;96:511–6. [5] Tsukita S, Furuse M. The structure and function of claudins, cell adhesion molecules at tight junctions. Ann N Y Acad Sci 2000;915:129–35. [6] Tsukita S, Furuse M, Itoh M. Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol 2001;2:285–93. [7] Angelow S, Ahlstrom R, Yu AS. Biology of claudins. Am J Physiol Renal Physiol 2008; 295:867–76. [8] Gavin Gordon J, Jensen RV, Hsiao Li-Li, Gullans SR, Blumenstock JE, Sridhar R, et al. Translation of microarray data into clinically relevant cancer diagnostic tests using gene expression ratios in lung cancer and mesothelioma. Cancer Res 2002;62:4963–7. [9] Holloway AJ, Diyagama DS, Opeskin K, Creaney J, Robinson BWS, Lak RA, et al. A molecular diagnostic test for distinguishing lung adenocarcinoma from malignant mesothelioma using cells collected from pleural effusions. Clin Cancer Res 2006(17):12. [10] Soini Y, Kinnula V, Kahlos K, Pääkkö P. Claudins in differential diagnosis between mesothelioma and metastatic adenocarcinoma of the pleura. J Clin Pathol 2006;59:250–4. [11] Kleinberg L, Holth A, Fridman E, Shwartz I, Davidson B. The diagnosis role of claudins in serous effusion. Am J Clin Pathol 2007;127:928–37. [12] Facchetti F, Lonardi S, Gentili F, Murer B. Claudin 4 identifies a wide spectrum of epithelial neoplasm and represents a very useful marker for carcinoma versus mesothelioma diagnosis in pleural and peritoneal biopsies and effusion. Virchows Arch 2007;451:669–80. [13] Lonardi S, Manera C, Marucci R, Santoro A, Lorenzi L, Facchetti F. Usefulness of claudin 4 in the cytological diagnosis of serosal effusion. Diagn Cytopathol 2011; 39(5):313–7. [14] Chaouche-Mazouni S, Copin MC, Lassalle P, Lebaili N, Cortot A, Scherpereel A. M14K and M38K malignant pleural mesothelioma cell lines preserve the same claudinbased phenotype in vivo. In Vivo 2013;27:227–32.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx [15] Ordóñez NG. Value of claudin-4 immunostaining in the diagnosis of mesothelioma. Am J Clin Pathol 2013;139(5):611–9. [16] Ohta Y, Sasaki Y, Saito M, Kushima M, Takimoto M, Shiokawa A, et al. Claudin-4 as a marker for distinguishing malignant mesothelioma from lung carcinoma and serous adenocarcinoma. Int J Surg Pathol 2013;21(5):493–501. [17] Gordon GJ, Rockwell GN, Jensen RV, Rheinwald JG, Glickman JN, Aronson JP, et al. Identification of novel candidate oncogenes and tumor suppressors in malignant pleural mesothelioma using large-scale transcriptional profiling. Am J Pathol 2005;166(6): 1827–40. [18] Davidson B, Zhang Z, Kleinberg L, Li M, Florenes VA, Wang TL, et al. Gene expression signatures differentiate ovarian/peritoneal serous carcinoma from diffuse malignant peritoneal mesothelioma. Clin Cancer Res 2006;12:5944–50. [19] Inai T, Sengoku A, Guan X, Hirose E, Lida H, Shibata Y. Heterogeneity in expression and subcellular localisation of tight junction proteins, claudin-10 and -15, examined by RT-PCR and immunofluorescence microscopy. Arch Histol Cytol 2005;68(5):349–60.
5
[20] Coyne CB, Gambling TM, Boucher RC, Carson JL, Johnson LG. Role of claudin interactions in airway tight junctional permeability. Am J Physiol Lung Cell Mol Physiol 2003;285:1166–78. [21] Husain AN, Colby T, Ordonez NG, et al. Guidelines for pathologic diagnosis of malignant mesothelioma. Arch Pathol Lab Med 2012;137. [22] Hewitt KJ, Agarwall R, Morin PJ. The claudin gene family: expression in normal and neoplastic tissues. BMC Cancer 2006;6(186):1471–24070. [23] Bij S, Schaake, Koffijberg H, Burgers JA, Moons KGM. Markers for the non-inavasive diagnosis of mesothelioma. Br J Cancer 2011:1–9. [24] Ordonez NG. The immunohistochemical diagnosis of mesothelioma: a comparative study of Epithelioid mesothelioma and lung adenocarcinoma. Am J Surg Pathol 2003;27(8):1031–51. [25] Takehara M, Nishimura T, Mima S, Hoshino T, Mizushima T. Effect of claudin expression on para-cellular permeability, migration and invasion of colonic cancer cells. Biol Pharm Bull 2009;32(5):825–31.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Original Contribution
Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma Siham Chaouche-Mazouni a,⁎, Arnaud Scherpereel b, Rima Zaamoum c, Adriana Mihalache d, Zine-Charaf Amir e, Nemcha Lebaïli a, Baptiste Delaire d, Pierre Gosset d a
Department of Biology, Kouba High School, Algiers, Algeria Department of Pulmonary and Thoracic Oncology, Lille University Hospital, Lille, France Department of Pneumophtysiology, Public Hospital of Rouiba, Algeria d Department of Pathology, Hospital of St Vincent, GHICL, Lille, France e Department of Pathology, Hospital of Mustapha Pacha, Algiers, Algeria b c
a r t i c l e
i n f o
Available online xxxx Keywords: Mesothelioma Lung adenocarcinoma Claudins Diagnosis Therapeutic target
a b s t r a c t Epithelioid malignant pleural mesothelioma (MPM) can easily be confused with lung adenocarcinomas (ACAs). In serous effusion, claudin (cldn) 3 is shown to be useful in the diagnosis of mesothelioma vs ACAs. Cldn15 is reported to be overexpressed in epithelioid mesothelioma and absent in human airway epithelium. The aim was to assess the value of cldn3 and cldn4 compared to that of BerEp4 and thyroid transcription factor–1 (TTF1) in differentiating lung ACA from epithelioid MPM and to examine the expression of cldn15 in these tumors. The expression of cldn3, cldn4, cldn15, BerEp4, and TTF1 was examined by immunohistochemistry in a total of 62 human specimen including 28 epithelioid MPMs and 34 ACAs of the lung. In lung ACA, cldn4 was strongly expressed in all 34 (100%) specimens followed by cldn3 in 33 (97%) of 34. BerEp4 was expressed in 32 (94.1%) of 34. TTF1 reacted for only 20 (58.82%) of 34 cases of lung ACA. In MPM specimens, the expression of cldn3 and4 as well as that of TTF1 was completely absent. In contrast, BerEp4 was focally expressed in 5 (17.85%) of 28 cases of epithelioid MPM. Cldn15 was strongly expressed in 53% pf epithelioid MPMs but also in 50% of lung ACAs. Its expression was moderate in normal pleura and limited in normal lung. Cldn3 and cldn4 appear to be the best performing carcinoma markers in discriminating lung ACA from mesothelioma compared with BerEp4 and TTF1. There is no differential expression of cldn15 between the 2 pathologies. However, the limited cldn15 expression in normal tissues and high expression in tumors make it an attractive candidate for cancer therapy. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Malignant pleural mesothelioma (MPM) is a highly aggressive tumor with difficult diagnosis. Differentiating between the most common, epithelioid form of mesothelioma and lung adenocarcinoma (ACA) is often problematic. Immunohistochemistry has greatly improved the diagnosis of mesothelioma. However, as there is no specific biomarker, the diagnosis is based in part on the negativity of markers that characterize other pathological pleural entities [1,2]. Emerging studies suggest the interest of claudin expression in cancer diagnosis and therapy [3]. Claudins are a family of 27 proteins that constitute the major components of tight junctions [4–7]. In the last decade, the role of some claudins in distinguishing mesothelioma from carcinomas has been reported. In 2002, Gordon et al [8] showed ⁎ Corresponding author at: Department of Biology, Kouba high School, Algiers, Algeria. Tel.: +213 664747387, +213 552115971. E-mail addresses: [email protected], [email protected] (S. Chaouche-Mazouni).
a highly differential gene expression of claudin 7 between mesothelioma and lung ACA. Some years later, Holloway et al [9] listed claudin 3 and claudin 7 among the highly differentially expressed genes between the 2 pathologies. Simultaneously, Soini et al [10] showed that cldn1, 3, 4, 5, and 7 protein expression could be used as an adjunct in the differential diagnosis between these tumors. Kleinberg et al [11] reported the usefulness of claudin1, 3, and 7 in serous effusion; and Facchetti et al [12,13] reported the usefulness of cldn4 in pleural and peritoneal biopsies and effusions. We previously found, in a limited number of specimens, the absence of cldn3 and 4 in MPM and their expression in 100% of lung ACAs [14]. Recently, Ordóñez [15] and Ohta et al [16] confirmed the diagnostic utility of cldn4 in distinguishing mesothelioma from ACAs. Taken together, these data highlight the interest of cldn3 and 4 in mesothelioma diagnosis. However, the number of published reports is limited; and the use of these proteins in routine practice still requires validation by further studies especially for claudin 3. In the present work, we compare the value of cldn3 and 4 with that of the 2 markers, BerEp4 and thyroid transcription factor–1 (TTF1), in 62 human
http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007 1092-9134/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
2
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Table 1 Antibodies used. Antibody Calretinin CK5/6 WT1 BerEp4 TTF1 Cldn3 Cldn4 Cldn15
Clone 5A5 D5/16B4 6F-H2 Ber-Ep4 8G7G3/1 Ab15102 Ab15104 NBP2-13842
Host and clonality Mouse monoclonal Mouse monoclonal Mouse monoclonal Mouse monoclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal
Source Leica Dako Roche Dako Dako Abcam Abcam Novus Bio
specimens of lung ACA and epithelioid MPM, currently used in the panel for differentiating between the 2 pathologies. Moreover, claudin 15 has been listed among the highly expressed genes in epithelioid MPM [17] and in diffuse peritoneal malignant mesothelioma [18]. We previously found high levels of cldn15 protein in MPM cell lines [14]. The expression of cldn15 was detected in mesothelial cells of the pleura and peritoneum in mice [19] but was absent in bronchial human airways [20]. In this study, we examined the expression of cldn15 in human specimens of epithelioid mesothelioma and lung ACA to see whether it is differentially expressed between these tumors. 2. Material and methods Sixty-two (62) human specimens including epithelioid MPMs (n= 28) and lung ACAs (n=34), in addition to normal control tissues (colon, pleura, bronchi, and bronchioles), were used to examine the expression of cldn3, cldn4, cldn15, BerEp4, and TTF1 in addition to calretinin, cytokeratin 5/6, and WT1 to confirm the diagnosis of mesothelioma. Surgical resection specimens were provided by the CHU of Mustapha Pacha (Algiers, Algeria) and the Public Hospital of Blida (Algeria). No biopsy material was used because there was insufficient material for immunohistochemical analysis. The diagnosis of all specimens was validated by pathologists using both morphologic criteria and immunohistochemistry with a broad antibody panel that recognizes carcinoma or mesothelioma. In case of TTF1-negative cases, the diagnosis of primary lung carcinoma was based on clinicopathological
Dilution Prediluted 1:50 Prediluted 1:25 1:50 1:100 1:200 1:200
Incubation (min) 32 32 44 32 32 20 20 32
Pretreatment Citrate buffer pH 6.0 60 min EDTA, pH 8.4 60 mn EDTA, pH 8.4 60 min None Citrate buffer, pH 6.0 60 min Citrate buffer, pH 6.0 60 min EDTA, pH 8.4 36 min Citrate buffer, pH 6.0 90 min
correlation. Nevertheless, the primary vs secondary origin of ACA was less important for this study because it was aimed at differentiating between ACAs and epithelioid form of mesotheliomas. 2.1. Immunohistochemistry Formalin-fixed, paraffin-embedded samples were cut (4 μm–thick sections) and placed on silane-covered slides. Morphological assessment of the samples was obtained by examining sections stained with hematoxylin-eosin-saffron. Slides were stained with anti-cldn3, anticldn4, anti-cldn15, anti-BerEp4, anti-TTF1, anti-calretinin, anti-keratin 5/6, and anti-WT1 using a Ventana Ultraview DAB detection kit in a Ventana BenchMark XT processor (Ventana, Tucson, AZ). For the antibody dilutions, pretreatment, and sources, see Table 1. An external positive control section was included in each immunohistochemical analysis (1 section for each slide). The signal was interpreted as present (positive) or absent (negative) irrespective of the intensity of the signal. When present, there were at least several cells stained; and the intensity was at least moderate in all cases. 3. Results In lung ACAs, cldn4 (Fig. 1A) was strongly expressed along the cell membrane in all 34 specimens (100%). Cldn3 (Fig. 1B) was expressed along the cell membrane in 33 (97%) of 34 cases with less intensity of immunostaining than cld4. The immunostaining of BerEp4 (Fig. 1C) was as intense as that of cldn4 along the cell membrane in 32 (94.1%)
a
b
c
d
Fig. 1. Lung ACA immunostaining. (A) cldn-4 immunostaining (intense membrane positivity) ×400. (B) cldn-3 immunostaining (membrane positivity) ×400. (C) BerEp4 immunostaining (intense membrane positivity) ×200. (D) TTF1 immunostaining (nuclear positivity) ×400.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
3
a
b
c
d
Fig. 2. Epithelioid MPM immunostaining. (A) cldn-4 immunostaining (absence of expression) ×400. (B) cldn-3 immunostaining (absence of expression) ×400. (C) BerEp4 immunostaining (intense membrane positivity) ×400. (D) TTF1 immunostaining (absence of expression) ×400.
of 34 cases of lung ACA. TTF1 (Fig. 1D) was positive in only 20 (58.82%) of 34 cases of lung ACA, showing a nuclear staining. The 2 BerEp4negative cases of lung ACA were positive for TTF1. In MPM specimens, the expression of cldn4 (Fig. 2A) and cldn3 (Fig. 2B), as well as that of TTF1 (Fig. 2D), was completely absent, whereas BerEp4 (Fig. 2C) focally stained the cell membrane in 5 (17.85%) of 28 epithelioid MPM cases. In normal colon tissues used as control, cldn15 was expressed in the cytoplasm and along the cell membrane of epithelial cells (Fig. 3A). In tumors, cldn15 was highly expressed in the cytoplasm and cell membrane of 53% of epithelioid MPMs (Fig. 3B) but also in 50% of lung ACAs (Fig. 3C). In normal pleura (Fig. 3D), mesothelial cells reacted with cldn15 less intensely than in MPM. A weak cytoplasmic positivity
of cldn15 was observed in bronchioles (Fig. 3E) and bronchi (Fig. 3F) of normal airways. Results are summarized in Table 2. 4. Discussion Epithelioid MPM can easily be confused with lung ACA. Several markers are currently used in the diagnosis of MPM vs lung ACA. However, none of them is sufficiently discriminant. The most valuable tissue markers for mesothelial cells are calretinin (only the nuclear staining has diagnostic value), WT1 (nuclear staining), cytokeratin 5/6 (cytoplasmic staining), and epithelial membrane antigen (EMA; membranous staining) as positive markers and carcinoembryonic antigen
a
b
c
d
e
f
Fig. 3. Immunostaining of cldn 15. (A) Normal colon (intense membrane and cytoplasmic positivity) ×400. (B) MPM (intense cytoplasmic positivity) ×400. (C) Lung ACA (intense membrane and cytoplasmic positivity) ×400. (D) Normal pleura (membrane and cytoplasmic positivity) ×400. (E) Bronchioles (weak cytoplasmic positivity) ×400. (F) Bronchus (weak cytoplasmic positivity) ×400.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
4
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Table 2 Results of cldn 3, 4, and 15 immunostaining compared with BerEp4 and TTF1.
Lung ADC (n = 34) Epithelioid MPM (n = 28)
Cldn3
Cldn4
Cldn15
BerEp4
TTF1
CK5/6
Calretinin
33 (97%) 0 (0%)
34 (100%) 0 (0%)
17 (50%) 15 (53%)
32 (94.1%) 5 (17.85%)
20 (58.82%) 0 (0%)
3 (8.8%) 15 (53%)
2 (5.88%) 20 (71.42%)
(cytoplasmic staining), BerEp4 (membranous staining), TTF1 (nuclear staining), and EMA (cytoplasmic staining) as negative markers. The International Mesothelioma Panel recommends that at least 2 mesothelial and 2 carcinoma markers be used in addition to a pancytokeratin for a reliable diagnosis of MPM [2,21]. Recently, cldn3 [10–12,14] and cldn4 [11,13–16] have been shown to be highly differentially expressed between carcinomas and mesotheliomas. However, whereas the cldn4 protein has been well evaluated as a useful marker in distinguishing between these tumors [12–16], only 1 study reported the usefulness of cldn3 in serous effusion [11]. Coordinate expression has been reported for cldn3 and cldn4 in several normal and neoplastic tissues [22]. In the present study, we evaluated the expression of cldn3 in human epithelioid MPMs and lung ACAs to see whether it performs similar to cldn4 in discriminating between the 2 pathologies compared with BerEp4 and TTF1. Cdn4 is reported to be the most interesting marker in distinguishing lung ACA from mesothelioma, with 100% specificity and sensitivity [12–16]. In this study, we found cldn4 to be strongly expressed in 100% of lung ACA specimens and completely absent in MPMs, confirming once again the results previously obtained and thus highlighting its usefulness in the differential diagnosis between these tumors. In the case of cldn3, Soini et al [10] reported a weak positivity in 6 (18%) of 33 epithelioid mesotheliomas. Kleinberg et al [11] found a limited cldn3 positivity in 1 (4%) of 25 serous effusions but no cldn3 expression in 13 solid tumors of mesothelioma. In contrast, Soini et al [10] showed that this protein was expressed in 21 (88%) of 24 ACAs of different origin, whereas Kleinberg et al found it positive in 7 (88%) of 8 serous effusions and 5 (100%) of 5 solid tumours of lung ACA. In the present study, we found cldn3 to be completely absent in all 32 MPMs and positive in 33 (97%) of 34 lung ACAs, showing its potential use in the differential diagnosis of the 2 pathologies. Among the immunohistochemical markers most frequently evaluated in the diagnosis of mesothelioma, BerEp4 was the second most commonly used marker after EMA, followed by carcinoembryonic antigen and calretinin [2,23]. BerEp4 is highly sensitive for ACA and currently is the most common carcinoma marker used to distinguish between lung ACA and epithelioid mesothelioma. However, it is reported that BerEp4 is not specific because 20% of MPMs are focally positive [21,24]. In this work, the results showed BerEp4 to be highly and strongly sensitive for lung ACA (94.1%); but at the same time, it reacted with 17.85% of epithelioid MPMs, highlighting its lack of specificity compared to that of cldn3 and cldn4. TTF1 is the most used marker for confirming lung origin of ACA. It is never expressed in mesothelioma, but its limited sensitivity (75%) [21] for lung ACA may decrease its usefulness. In our experience, TTF1 was actually absent in MPM specimens but was detected in only 58.82% of lung ACA. We note that the 2 BerEp4-negative cases in lung ACA were positive for TTF1. However, even though the combination of BerEp4 and TTF1 increased their sensitivity for lung ACA, the highest sensitivity of cldn4 and cldn3 and their 100% specificity make them more useful in discriminating between lung ACA and epithelioid MPM. Furthermore, previous studies [12,15] showed that the 2 proteins were expressed not only in lung ACA but in all metastatic carcinomas that metastasize to the serosal membranes. Claudin15 gene is found to be overexpressed in epithelioid MPM [17] and in diffuse peritoneal malignant mesothelioma [18]. We previously reported high levels of cldn15 protein in MPM cell lines [14]. On the other hand, Coyne et al [20] showed that cldn15 expression is absent in bronchial and bronchiolar epithelium in human airways. In human
MPM specimens, to our knowledge, cldn15 expression has not been reported yet. In this study, we examined cldn15 protein expression to see whether it is differentially expressed between MPM and lung ACA. Our results showed that immunohostochemically cldn15 is strongly expressed in 53% of epithelioid MPMs. However, 50% of lung ACAs also reacted with cldn15. When compared with normal tissues, we found cldn15 to be moderately expressed in normal pleura and weakly expressed in bronchi and bronchioles. These findings demonstrated little value of the polyclonal rabbit anti-claudin 15 in the differential diagnosis between lung ACA and MPM on paraffin section. However, because the overexpression of cldn15 has been linked to tumor progression [25], the limited cldn15 expression in normal lung and pleura and high expression in tumors make it an attractive candidate for cancer therapy. 5. Conclusion Among all carcinoma markers currently used, the high specificity and sensitivity of cldn3 and cldn4 make them useful markers for discriminating between lung ADC and MPM. These two markers should be included in the panel for the diagnosis of epithelioid mesothelioma. The polyclonal anti-cldn15 antibody used in this study showed no diagnostic utility in differentiating between the 2 pathologies. However, the limited cldn15 expression in normal tissues and overexpression in tumors could have an interest as a potential therapeutic target. Conflict of interest None to declare. References [1] Scherpereel A, et al. Guidelines of the European Respiratory Society and the European Society of Thoracic Surgeons for the management of malignant pleural mesothelioma. Eur Respir J 2010;35:479–95. [2] Astoul P, Roca E, Galateau Sallé F, Scherpereel A. Malignant pleural mesothelioma: from the bench to the bedside. Respiration 2012;83:481–93. [3] Morin JP. Claudin proteins in human cancer: promising new targets for diagnosis and cancer therapy. Cancer Res 2005;65:9603–6. [4] Morita K, Furuse M, Fujimoto K, Tsukita S. Claudin multigene family encoding fourtransmembrane domain protein components of tight junction strands. Proc Natl Acad Sci 1999;96:511–6. [5] Tsukita S, Furuse M. The structure and function of claudins, cell adhesion molecules at tight junctions. Ann N Y Acad Sci 2000;915:129–35. [6] Tsukita S, Furuse M, Itoh M. Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol 2001;2:285–93. [7] Angelow S, Ahlstrom R, Yu AS. Biology of claudins. Am J Physiol Renal Physiol 2008; 295:867–76. [8] Gavin Gordon J, Jensen RV, Hsiao Li-Li, Gullans SR, Blumenstock JE, Sridhar R, et al. Translation of microarray data into clinically relevant cancer diagnostic tests using gene expression ratios in lung cancer and mesothelioma. Cancer Res 2002;62:4963–7. [9] Holloway AJ, Diyagama DS, Opeskin K, Creaney J, Robinson BWS, Lak RA, et al. A molecular diagnostic test for distinguishing lung adenocarcinoma from malignant mesothelioma using cells collected from pleural effusions. Clin Cancer Res 2006(17):12. [10] Soini Y, Kinnula V, Kahlos K, Pääkkö P. Claudins in differential diagnosis between mesothelioma and metastatic adenocarcinoma of the pleura. J Clin Pathol 2006;59:250–4. [11] Kleinberg L, Holth A, Fridman E, Shwartz I, Davidson B. The diagnosis role of claudins in serous effusion. Am J Clin Pathol 2007;127:928–37. [12] Facchetti F, Lonardi S, Gentili F, Murer B. Claudin 4 identifies a wide spectrum of epithelial neoplasm and represents a very useful marker for carcinoma versus mesothelioma diagnosis in pleural and peritoneal biopsies and effusion. Virchows Arch 2007;451:669–80. [13] Lonardi S, Manera C, Marucci R, Santoro A, Lorenzi L, Facchetti F. Usefulness of claudin 4 in the cytological diagnosis of serosal effusion. Diagn Cytopathol 2011; 39(5):313–7. [14] Chaouche-Mazouni S, Copin MC, Lassalle P, Lebaili N, Cortot A, Scherpereel A. M14K and M38K malignant pleural mesothelioma cell lines preserve the same claudinbased phenotype in vivo. In Vivo 2013;27:227–32.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
S. Chaouche-Mazouni et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx [15] Ordóñez NG. Value of claudin-4 immunostaining in the diagnosis of mesothelioma. Am J Clin Pathol 2013;139(5):611–9. [16] Ohta Y, Sasaki Y, Saito M, Kushima M, Takimoto M, Shiokawa A, et al. Claudin-4 as a marker for distinguishing malignant mesothelioma from lung carcinoma and serous adenocarcinoma. Int J Surg Pathol 2013;21(5):493–501. [17] Gordon GJ, Rockwell GN, Jensen RV, Rheinwald JG, Glickman JN, Aronson JP, et al. Identification of novel candidate oncogenes and tumor suppressors in malignant pleural mesothelioma using large-scale transcriptional profiling. Am J Pathol 2005;166(6): 1827–40. [18] Davidson B, Zhang Z, Kleinberg L, Li M, Florenes VA, Wang TL, et al. Gene expression signatures differentiate ovarian/peritoneal serous carcinoma from diffuse malignant peritoneal mesothelioma. Clin Cancer Res 2006;12:5944–50. [19] Inai T, Sengoku A, Guan X, Hirose E, Lida H, Shibata Y. Heterogeneity in expression and subcellular localisation of tight junction proteins, claudin-10 and -15, examined by RT-PCR and immunofluorescence microscopy. Arch Histol Cytol 2005;68(5):349–60.
5
[20] Coyne CB, Gambling TM, Boucher RC, Carson JL, Johnson LG. Role of claudin interactions in airway tight junctional permeability. Am J Physiol Lung Cell Mol Physiol 2003;285:1166–78. [21] Husain AN, Colby T, Ordonez NG, et al. Guidelines for pathologic diagnosis of malignant mesothelioma. Arch Pathol Lab Med 2012;137. [22] Hewitt KJ, Agarwall R, Morin PJ. The claudin gene family: expression in normal and neoplastic tissues. BMC Cancer 2006;6(186):1471–24070. [23] Bij S, Schaake, Koffijberg H, Burgers JA, Moons KGM. Markers for the non-inavasive diagnosis of mesothelioma. Br J Cancer 2011:1–9. [24] Ordonez NG. The immunohistochemical diagnosis of mesothelioma: a comparative study of Epithelioid mesothelioma and lung adenocarcinoma. Am J Surg Pathol 2003;27(8):1031–51. [25] Takehara M, Nishimura T, Mima S, Hoshino T, Mizushima T. Effect of claudin expression on para-cellular permeability, migration and invasion of colonic cancer cells. Biol Pharm Bull 2009;32(5):825–31.
Please cite this article as: Chaouche-Mazouni S, et al, Claudin 3, 4, and 15 expression in solid tumors of lung adenocarcinoma versus malignant pleural mesothelioma, Ann Diagn Pathol (2015), http://dx.doi.org/10.1016/j.anndiagpath.2015.03.007
