- Email: [email protected]
Distinctive immunostaining of claudin-4 in spiradenomas
Annals of Diagnostic Pathology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Original Contributions
Distinctive immunostaining of claudin-4 in spiradenomas☆,☆☆ Nuri Yiğit, MD a,⁎, Ertuğrul Çelik, MD b, İbrahim Yavan, MD a, Armağan Günal, MD a, Bülent Kurt, MD a, Yıldırım Karslıoğlu, MD a, Önder Öngürü, MD a, Ayhan Özcan, MD a a b
Department of Pathology, Gülhane Military Medical Academy and School of Medicine, Ankara, Turkey Department of Pathology, İzmir Military Hospital, İzmir, Turkey
a r t i c l e
i n f o
Keywords: Claudin-4 Spiradenoma Intercellular bridge proteins Eccrine tumors Skin tumors
a b s t r a c t The intercellular bridges are essential structures in maintaining the histologic organization of the epithelium, while providing a very efficient way to exchange molecules between cells and transduction of the cell-to-cell and matrix-to-cell signals. Derangement in those important structures' physical integrity and/or function, which can be assessed by the presence or absence of several intercellular bridge proteins including claudin-4, E-cadherin, and β-catenin, was found to be related to several phenomena in the path to the neoplastic transformation. However, these proteins have not been studied in the wide variety of the skin neoplasms, in detail. Herein, we immunohistochemically assessed the expression patterns of these 3 intercellular bridge proteins on a total of 86 epidermal and eccrine adnexal tumors including basal cell carcinoma, squamous cell carcinoma, poroma, spiradenoma, syringoma, and hidradenoma. We observed a selective and distinct claudin-4 expression in the ductal-type cells of all cases of spiradenomas. Similarly, in the poromas, syringomas, and hidradenomas, claudin-4 was only positive in the luminal cells of microcystic structures, although not as conspicuous as in the spiradenomas. On the other hand, E-cadherin and β-catenin were positive in almost all types of the tumors, in a way which was not contributory to differentiate from each other. In conclusion, we think that claudin-4 can be helpful at least in making a reliable differential diagnosis of spiradenoma when overlapping morphologic features do not allow to further subclassification in the overwhelming variety of the adnexal tumors. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Normal epithelia maintain their intricate and vulnerable organizations mostly by intercellular bridges which perpetuate the arrangement of the cells in spatial order as well as the transportation of crucial molecules between the adjacent cells. Intercellular bridges, which comprise the highly modified portions of the neighboring cells' membranes, are complex structures harboring some important proteins which play crucial roles in the functions of desmosomes, gap junctions, zonula adherens, and tight junctions [1,2]. The claudins, occludins, cadherins, and catenins are some of the well-known constituents of the membrane and membrane-associated cytoskeletal proteins at the site of intercellular bridges [3,4]. Up to date, in many tumors, dysregulation of those junctional proteins has been shown that they might have some promoter effects ☆ Conflict of interest statement: The authors declare no conflict of interest. ☆☆ Disclosure statement: The authors declare no relevant financial or nonfinancial relationship to disclose. ⁎ Corresponding author at: Department of Pathology, Gülhane Military Medical Academy and School of Medicine, Ankara 06010, Turkey. Tel.: +90 312 304 3734; fax: +90 312 304 3700. E-mail addresses: [email protected] (N. Yiğit), [email protected] (E. Çelik), [email protected] (İ. Yavan), [email protected] (A. Günal), [email protected] (B. Kurt), [email protected] (Y. Karslıoğlu), [email protected] (Ö. Öngürü), [email protected] (A. Özcan).
throughout the event in the tumorigenesis [5]. Overexpression or underexpression of such molecules was found to be resulted in loss of cell-cell cohesion, increase in motility, invasiveness, and metastasis [6,7]. Despite the expanding resources and the ancillary techniques, diagnosis of the most skin tumors is still based on the morphologic findings and pattern. Only few cases having overlapping histologic features cause diagnostic difficulties and warrant to be studied with further workups. In routine pathology practice, the immunohistochemical markers frequently provide useful data to make correct diagnosis. However, the contribution of immunohistochemistry in differential diagnosis has been shown limited for many skin neoplasms in contrast to the nonskin tumors. Hence, this inability led to the requirement of finding further ancillary techniques when needed. Herein, we investigated immunohistochemical characteristics of 3 intercellular bridge proteins including claudin-4, E-cadherin, and β-catenin on a total of 86 cases of different types of frequently encountered malignant and benign skin tumors, to analyze their expression patterns. 2. Materials and methods 2.1. Case selection We retrospectively selected 86 cases in total, including 20 basal cell carcinomas (BCCs), 20 squamous cell carcinomas (SCCs), 20 poromas, 10 spiradenomas, 10 syringomas, and 6 hidradenomas among the files of the Department of Pathology of Gulhane Military Medical Academy
http://dx.doi.org/10.1016/j.anndiagpath.2015.10.005 1092-9134/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
2
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
in Ankara. All of them were excisional biopsy specimens. Archival slides of all cases were reevaluated to confirm their diagnosis. Five uninvolved normal skin samples were also used as a control group in the study. The specimens were fixed in 10% neutral-buffered formalin, embedded in paraffin, and processed routinely, and the sections were stained with hematoxylin and eosin (H&E). The design of this study was approved by the local ethical committee of Gulhane Military Medical Academy in Ankara. 2.2. Immunohistochemistry All tissue sections were immunostained against claudin-4 (rabbit polyclonal, 1:150; Neomarkers, Fremont, CA), E-cadherin (rabbit monoclonal, 1:200; Cell Marque, Rocklin, CA), and β-catenin (mouse monoclonal, 1:200; Dako, Carpinteria, CA) by using an autostainer (Ventana Benchmark XT; Ventana Medical Systems, Inc, Tucson, Arizona), following the manufacturer's protocol. Appropriate staining was confirmed using external positive controls for each antibody. The intensity of the staining was scored as follows: 0, negative; +, weak; ++, moderate; or +++, strong. All stained slides have independently been evaluated by 2 blinded pathologists (N.Y. and E.Ç.). 3. Results Claudin-4 exhibited membranous staining without any cytoplasmic or nuclear marking in all groups. In the control group, claudin-4
exclusively and faintly highlighted upper layer of epidermis as was documented previously (Fig. 1a) [8]. The inner root sheath of hair follicle and some attached sebaceous cells showed weak membranous labeling (Fig. 1b). Also, both secretory and ductal cells of the eccrine sweat glands demonstrated strong and membranous marking (Fig. 1c). Interestingly, in all spiradenomas, claudin-4 displayed strong and membranous staining in the cells, which were morphologically corresponded to the ductal structures on H&E-stained slides, and spared the other components of the glandular cells of small dark and larger pale cells (Fig. 1d-f). Appearance of this distinct ductal staining on the sections gave the impression of an easily detectable lace-like network, which exposed the typical ductal part of the spiradenomas. Subsequently, to be sure whether these stained cells in spiradenomas are ductal nature indeed, we used further immunohistochemical stain against carcinoembryonic antigen (CEA) (rabbit polyclonal, 1:400, Cell Marque, Rocklin, CA) labeling of which was reported to be more specific for ducts than for the secretory portion, on a few selected cases of the spiradenomas [9]. We observed moderate staining in the luminal surface of the many ductal cells with a minute population of secretory cells, confirming that the claudin-4 selectively highlighted the ductal parts of the spiradenomas (Fig. 1g). When we compared the immunostaining intensity of claudin-4 and CEA in these ductal cells, claudin-4 immunoexpression was definitely stronger and extensive than CEA. On the other hand, claudin-4 was somewhat positive in the luminal cell layer of microcysts in the poromas, syringomas, and hidradenomas, in a way of noncontributory to their differential diagnosis (Fig. 2a-c).
Fig. 1. In the uninvolved skin sample, claudin-4 immunostaining was weak in the granular layer of epidermis (a), moderate in the cells of inner root sheath of hair follicle with expression in some sebaceous cells (b), and strong in the eccrine sweat glands (c) (immunoperoxidase, original magnification ×400). (d) Spiradenoma containing the small dark and larger pale cells with scattered ductal structures (H&E, ×100). (e and f) Claudin-4 expression in the tumor selectively reveals the ductal component, with a lace-like appearance, on a background of the purely negative glandular cells (immunoperoxidase, ×400 and ×200, respectively). (g) CEA highlighted the luminal surfaces of some ductal cells with a weaker intensity compared with the claudin-4 (immunoperoxidase, ×400). (h and i) E-cadherin and β-catenin showing diffuse and strong membranous staining in all cell types of spiradenomas (immunoperoxidase, both ×400).
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
In the cases of SCCs, claudin-4 displayed moderate staining in apparently more mature and keratinized squamous cells, which were only a minor population and mostly located in the central portions of the tumoral nests (Fig. 2d). The BCCs cases were all negative against claudin-4. Its expression profile in the peritumoral skin did not differ from the control group. E-cadherin indicated diffuse, moderate to strong membranous staining in epidermis, with an exception of granular layer, and dermal adnexal structures of the control cases. It showed diffuse and strong staining in all cells of the poromas, syringomas, hidradenomas, and BCCs. E-cadherin was positive in the well- and moderately differentiated squamous cells in SCCs, but not in the cells showing high degree of atypia. E-cadherin strongly highlighted glandular and ductal cells of the spiradenomas (Fig. 1h). β-Catenin showed diffuse but weak membranous staining in all epidermal and dermal appendages' cells. In the cases of BCC, it was positive in the peripherally located atypical cells of the nests with a membranous and cytoplasmic manner and usually negative in the center of the tumor nests. The remaining groups of poroma, syringoma, hidradenoma, and SCC demonstrated diffuse and weak to moderate staining with β-catenin. All epithelial cells of the spiradenomas were strongly highlighted with β-catenin (Fig. 1i). To summarize, the staining characteristics of both E-cadherin and β-catenin were not specific for any entity and not seemed to be useful to discriminate from each other. 4. Discussion Tight junctions are located in the most apical part of the epithelial cells to maintain the polarity, which is extremely critical to the normal structure and physiological function of the epithelium, as a whole [10]. The best-characterized tight junction components are the families of claudins, occludin, and zonula occludens proteins [11,12]. The claudins are a group of major transmembrane proteins and play essential roles in tight junction formation, integrity, and function [13]. A total of 27 different subtypes of claudin family in human tissues have been described to date [14]. They are variably expressed in normal cells depending on
3
the tissue types. Aberrant expression of the claudins, resulting in dysregulation of tight junctional function, is a known phenomenon in the development of tumorigenesis, in particular for carcinomas. Several published studies reported that claudin subtypes of 1, 2, 4, 5, 6, 7, 10, 12, and 18 were especially found to be more altered and associated in the development of different tumors [15–17]. It has been found that claudin-4 was overexpressed in carcinomas of the pancreas, breast, kidney, and ovary [15,18,19]. Conversely, its loss of expression has been reported in colorectal carcinomas and mesotheliomas [20,21]. Its overexpression was also shown to be a poor prognostic parameter in clear cell renal cell carcinomas and nasopharyngeal carcinomas, whereas it has been regarded as a good prognostic factor in hepatocellular and breast carcinomas [22–25]. To the best of our knowledge, although there are many studies investigating the immunoexpression profile of claudin-4 in nonskin neoplasms, it has not been assessed in skin tumors, except SCC and actinic keratosis [26]. In that study, Hintsala et al [26] showed that claudin-4 displayed a weaker overall expression in the cases of SCC and actinic keratosis compared with the normal skin. The eccrine sweat glands consist of secretory (glandular) and ductal portions. The secretory portion is composed of inner layer of secretory epithelial cells, also known as pyramidal cells, and outer layer of myoepithelial cells. The secretory cells are arranged a mixture of small, dark basaloid cells with hyperchromatic nuclei and larger cells with a clear cytoplasm and paler nucleus. Although the dark cells tend to line the luminal surface, the pale cells are situated peripherally. Besides, the eccrine ducts are lined by a double row of small cuboidal epithelial cells. The spiradenomas are benign tumors and consisted of 2 distinctive cell components similar to those of normal eccrine glands: (1) glandular cells, mostly centrally placed small dark and peripheral large pale cells, both of which occur aggregates in sheets, cords, or clusters and correspond to the secretory portion of the eccrine gland, and (2) ductal cells, which form easily recognizable ductlike structures throughout the glandular cell nests. The tumor may also harbor lymphocytes, variable amount of squamous cells, and microvessels.
Fig. 2. (a-c) Claudin-4 marking the luminal cells of microcysts in poroma, syringoma, and hidradenoma, respectively (immunoperoxidase, ×100, ×400, and ×200, respectively). (d) Claudin-4 showing mature keratinizing squamous cells localized in the central parts of the tumoral nests in SCC (immunoperoxidase, ×200).
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
4
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
In the present study, claudin-4 was found to highlight ductal component of the spiradenomas and completely spared the secretory cells. Although claudin-4 stains both secretory and ductal cells of the normal eccrine glands, it was expressed in only ductal cells of spiradenomas. This finding somehow might reflect the ontogenetic differences between those cell types. Besides, the poromas, syringomas, and hidradenomas are all well-known tumors having eccrine sweat gland differentiation. In our study, we observed that claudin-4 marks all cell types of the normal eccrine glands, which are composed of both secretory and ductal cells. However, it selectively stains the cells which form duct-like structures or the cells lining microcystic spaces in spiradenomas, poromas, syringomas, and hidradenomas. Interestingly, the distribution and the intensity of claudin-4 staining were so specific and strong to the ductal cells in spiradenoma cases that one can observe all the lace-like interconnected networks of duct-like structures, which are relatively a minor component when compared with the overwhelming glandular cells conspicuously throughout the tumor. Thus, it might be hypothesized that claudin-4 protein might be somehow down-regulated in the glandular cells and spared in ductal cells during the tumorigenesis of the spiradenomas. The same phenomenon may be true for other eccrine gland derived tumors. For the practical point of view, the claudin-4 seems to be a very specific marker for the ductal cells, either normal or neoplastic nature in all types of eccrine gland tumors. Our study showed that E-cadherin and β-catenin demonstrated unremarkable and not too useful staining profiles in differential diagnosis. The postulated origin of the spiradenomas has been still debatable. The spiradenomas are commonly believed to be originated from the straight intradermal portion of eccrine ducts [27]. Alternatively, a recent study claimed that the spiradenomas are actually derived from the hair follicle bulge, the folliculosebaceous-apocrine unit [28]. This assertion was reached based on the identical immunophenotypical findings, between hair follicle bulges and tumors, with the use of follicular stem cell markers of CD200, CK15, CK19, and PHLDA1. In our study, we observed that claudin-4 stains the inner root sheath, but we did not detect any positivity in follicule bulge both in control and in tumors. Therefore, we concluded that the spiradenomas are most likely be derived from eccrine sweat glands as is widely considered. As a note for the spiradenomas harboring a high degree of vascularity, the vascular-rich histomorphology can mimic a cellular hemangioma at the first glance and may cause a diagnostic pitfall [29]. In these cases, expanding the immunohistochemical panels only with vascular markers such as CD31 and CD34, highlights the already conspicuous blood vessels in the lesion and may pave the way to the misdiagnosis of a hemangioma, in exchange for overlooking of the spiradenoma. To prevent this, the basic immunohistochemistry panel with epithelial membrane antigen and CEA, which are 2 frequently used markers to exclude the possibility of other neoplasms including the soft tissue tumors, should be further extended. However, it should be kept in mind that negative or focal positive staining with epithelial membrane antigen and CEA can be seen in some spiradenomas, as was well documented previously, and these do not always provide evidence indicating the epithelial nature of the lesion. Thus, it would be a good idea to add claudin-4 into the immunohistochemistry panel, to minimize the risk of overlooking an eccrine gland tumor with microvessel-rich or not very cellular ones. 5. Conclusion We think that claudin-4 immunostaining can be a useful tool for the diagnosis of spiradenoma when histomorphologic and clinical findings are not distinctive. Claudin-4 immunohistochemistry effectively discriminates spiradenomas from the other possibilities in differential diagnosis, particularly against the possibilities of poroma, syringoma, hidradenoma, or cellular hemangioma, with a distinct staining on the ductal cells reminiscent of a lacy network. Our findings also support that the spiradenomas seem to have an eccrine gland origin rather than hair follicle bulge.
Acknowledgments We would like to thank Emrah Özenç for his excellent technical assistance in immunohistochemistry. References [1] Evans WH, Martin PE. Gap junctions: structure and function (review). Mol Membr Biol 2002;19:121–36. [2] Paris L, Tonutti L, Vannini C, Bazzoni G. Structural organization of the tight junctions. Biochim Biophys Acta 1778;2008:646–59. [3] Piepenhagen PA, Nelson WJ. Defining E-cadherin–associated protein complexes in epithelial cells: plakoglobin, beta- and gamma-catenin are distinct components. J Cell Sci 1993;104:751–62. [4] Langbein L, Pape UF, Grund C, Kuhn C, Praetzel S, Moll I, et al. Tight junction-related structures in the absence of a lumen: occludin, claudins and tight junction plaque proteins in densely packed cell formations of stratified epithelia and squamous cell carcinomas. Eur J Cell Biol 2003;82:385–400. [5] Mullin JM. Epithelial barriers, compartmentation, and cancer. Sci STKE 2004;2004 [pe2]. [6] Itoh M, Bissell MJ. The organization of tight junctions in epithelia: implications for mammary gland biology and breast tumorigenesis. J Mammary Gland Biol Neoplasia 2003;8:449–62. [7] Iwatsuki M, Mimori K, Yokobori T, Ishi H, Beppu T, Nakamori S, et al. Epithelialmesenchymal transition in cancer development and its clinical significance. Cancer Sci 2010;101:293–9. [8] Kubo T, Sugimoto K, Kojima T, Sawada N, Sato N, Ichimiya S. Tight junction protein claudin-4 is modulated via ΔNp63 in human keratinocytes. Biochem Biophys Res Commun 2014;455:205–11. [9] Saga K. Histochemical and immunohistochemical markers for human eccrine and apocrine sweat glands: an aid for histopathologic differentiation of sweat gland tumors. J Investig Dermatol Symp Proc 2001;6:49–53. [10] Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 2001;16:126–30. [11] Tsukita S, Katsuno T, Yamazaki Y, Umeda K, Tamura A, Tsukita S. Roles of ZO-1 and ZO-2 in establishment of the belt-like adherens and tight junctions with paracellular permselective barrier function. Ann N Y Acad Sci 2009;1165:44–52. [12] Günzel D, Yu AS. Claudins and the modulation of tight junction permeability. Physiol Rev 2013;93:525–69. [13] Lal-Nag M, Morin PJ. The claudins. Genome Biol 2009;10:235. [14] Mineta K, Yamamoto Y, Yamazaki Y, Tanaka H, Tada Y, Saito K, et al. Predicted expansion of the claudin multigene family. FEBS Lett 2011;585:606–12. [15] Rangel LB, Agarwal R, D'Souza T, Pizer ES, Alò PL, Lancaster WD, et al. Tight junction proteins claudin-3 and claudin-4 are frequently overexpressed in ovarian cancer but not in ovarian cystadenomas. Clin Cancer Res 2003;9:2567–75. [16] Merikallio H, Pääkkö P, Harju T, Soini Y. Claudins 10 and 18 are predominantly expressed in lung adenocarcinomas and in tumors of nonsmokers. Int J Clin Exp Pathol 2011;4:667–73. [17] Tokuhara Y, Morinishi T, Matsunaga T, Ohsaki H, Kushida Y, Haba R, et al. Claudin-1, but not claudin-4, exhibits differential expression patterns between well- to moderatelydifferentiated and poorly-differentiated gastric adenocarcinoma. Oncol Lett 2015;10:93–8. [18] Tsutsumi K, Sato N, Tanabe R, Mizumoto K, Morimatsu K, Kayashima T, et al. Claudin-4 expression predicts survival in pancreatic ductal adenocarcinoma. Ann Surg Oncol 2012;19(Suppl. 3):S491–9. [19] Jiwa LS, van Diest PJ, Hoefnagel LD, Wesseling J, Wesseling P, Dutch Distant Breast Cancer Metastases Consortium, Moelans CB. Upregulation of claudin-4, CAIX and GLUT-1 in distant breast cancer metastases. BMC Cancer 2014;14:864. [20] 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. [21] Süren D, Yıldırım M, Kaya V, Alikanoğlu AS, Bülbüller N, Yıldız M, et al. Loss of tight junction proteins (claudin 1, 4, and 7) correlates with aggressive behavior in colorectal carcinoma. Med Sci Monit 2014;20:1255–62. [22] Lechpammer M, Resnick MB, Sabo E, Yakirevich E, Greaves WO, Sciandra KT, et al. The diagnostic and prognostic utility of claudin expression in renal cell neoplasms. Mod Pathol 2008;21:1320–9. [23] Szasz AM, Nemeth Z, Gyorffy B, Micsinai M, Krenacs T, Baranyai Z, et al. Identification of a claudin-4 and E-cadherin score to predict prognosis in breast cancer. Cancer Sci 2011;102:2248–54. [24] Bouchagier KA, Assimakopoulos SF, Karavias DD, Maroulis I, Tzelepi V, Kalofonos H, et al. Expression of claudins-1, -4, -5, -7 and occludin in hepatocellular carcinoma and their relation with classic clinicopathological features and patients' survival. In Vivo 2014;28:315–26. [25] Suren D, Yildirim M, Kaya V, Elal R, Selcuk OT, Osma U, et al. Expression patterns of claudins 1, 4, and 7 and their prognostic significance in nasopharyngeal carcinoma. J BUON 2015;20:212–7. [26] Hintsala HR, Siponen M, Haapasaari KM, Karihtala P, Soini Y. Claudins 1, 2, 3, 4, 5 and 7 in solar keratosis and squamocellular carcinoma of the skin. Int J Clin Exp Pathol 2013;6:2855–63. [27] Mambo NC. Eccrine spiradenoma: clinical and pathologic study of 49 tumors. J Cutan Pathol 1983;10:312–20. [28] Sellheyer K. Spiradenoma and cylindroma originate from the hair follicle bulge and not from the eccrine sweat gland: an immunohistochemical study with CD200 and other stem cell markers. J Cutan Pathol 2015;42:90–101. [29] Jorquera Barquero E, Lara Bohórquez C, de Alba Rioja I. Giant vascular eccrine spiradenoma. Actas Dermosifiliogr 2015. http://dx.doi.org/10.1016/j.ad.2015.03. 019 [Epub ahead of print].
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Original Contributions
Distinctive immunostaining of claudin-4 in spiradenomas☆,☆☆ Nuri Yiğit, MD a,⁎, Ertuğrul Çelik, MD b, İbrahim Yavan, MD a, Armağan Günal, MD a, Bülent Kurt, MD a, Yıldırım Karslıoğlu, MD a, Önder Öngürü, MD a, Ayhan Özcan, MD a a b
Department of Pathology, Gülhane Military Medical Academy and School of Medicine, Ankara, Turkey Department of Pathology, İzmir Military Hospital, İzmir, Turkey
a r t i c l e
i n f o
Keywords: Claudin-4 Spiradenoma Intercellular bridge proteins Eccrine tumors Skin tumors
a b s t r a c t The intercellular bridges are essential structures in maintaining the histologic organization of the epithelium, while providing a very efficient way to exchange molecules between cells and transduction of the cell-to-cell and matrix-to-cell signals. Derangement in those important structures' physical integrity and/or function, which can be assessed by the presence or absence of several intercellular bridge proteins including claudin-4, E-cadherin, and β-catenin, was found to be related to several phenomena in the path to the neoplastic transformation. However, these proteins have not been studied in the wide variety of the skin neoplasms, in detail. Herein, we immunohistochemically assessed the expression patterns of these 3 intercellular bridge proteins on a total of 86 epidermal and eccrine adnexal tumors including basal cell carcinoma, squamous cell carcinoma, poroma, spiradenoma, syringoma, and hidradenoma. We observed a selective and distinct claudin-4 expression in the ductal-type cells of all cases of spiradenomas. Similarly, in the poromas, syringomas, and hidradenomas, claudin-4 was only positive in the luminal cells of microcystic structures, although not as conspicuous as in the spiradenomas. On the other hand, E-cadherin and β-catenin were positive in almost all types of the tumors, in a way which was not contributory to differentiate from each other. In conclusion, we think that claudin-4 can be helpful at least in making a reliable differential diagnosis of spiradenoma when overlapping morphologic features do not allow to further subclassification in the overwhelming variety of the adnexal tumors. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Normal epithelia maintain their intricate and vulnerable organizations mostly by intercellular bridges which perpetuate the arrangement of the cells in spatial order as well as the transportation of crucial molecules between the adjacent cells. Intercellular bridges, which comprise the highly modified portions of the neighboring cells' membranes, are complex structures harboring some important proteins which play crucial roles in the functions of desmosomes, gap junctions, zonula adherens, and tight junctions [1,2]. The claudins, occludins, cadherins, and catenins are some of the well-known constituents of the membrane and membrane-associated cytoskeletal proteins at the site of intercellular bridges [3,4]. Up to date, in many tumors, dysregulation of those junctional proteins has been shown that they might have some promoter effects ☆ Conflict of interest statement: The authors declare no conflict of interest. ☆☆ Disclosure statement: The authors declare no relevant financial or nonfinancial relationship to disclose. ⁎ Corresponding author at: Department of Pathology, Gülhane Military Medical Academy and School of Medicine, Ankara 06010, Turkey. Tel.: +90 312 304 3734; fax: +90 312 304 3700. E-mail addresses: [email protected] (N. Yiğit), [email protected] (E. Çelik), [email protected] (İ. Yavan), [email protected] (A. Günal), [email protected] (B. Kurt), [email protected] (Y. Karslıoğlu), [email protected] (Ö. Öngürü), [email protected] (A. Özcan).
throughout the event in the tumorigenesis [5]. Overexpression or underexpression of such molecules was found to be resulted in loss of cell-cell cohesion, increase in motility, invasiveness, and metastasis [6,7]. Despite the expanding resources and the ancillary techniques, diagnosis of the most skin tumors is still based on the morphologic findings and pattern. Only few cases having overlapping histologic features cause diagnostic difficulties and warrant to be studied with further workups. In routine pathology practice, the immunohistochemical markers frequently provide useful data to make correct diagnosis. However, the contribution of immunohistochemistry in differential diagnosis has been shown limited for many skin neoplasms in contrast to the nonskin tumors. Hence, this inability led to the requirement of finding further ancillary techniques when needed. Herein, we investigated immunohistochemical characteristics of 3 intercellular bridge proteins including claudin-4, E-cadherin, and β-catenin on a total of 86 cases of different types of frequently encountered malignant and benign skin tumors, to analyze their expression patterns. 2. Materials and methods 2.1. Case selection We retrospectively selected 86 cases in total, including 20 basal cell carcinomas (BCCs), 20 squamous cell carcinomas (SCCs), 20 poromas, 10 spiradenomas, 10 syringomas, and 6 hidradenomas among the files of the Department of Pathology of Gulhane Military Medical Academy
http://dx.doi.org/10.1016/j.anndiagpath.2015.10.005 1092-9134/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
2
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
in Ankara. All of them were excisional biopsy specimens. Archival slides of all cases were reevaluated to confirm their diagnosis. Five uninvolved normal skin samples were also used as a control group in the study. The specimens were fixed in 10% neutral-buffered formalin, embedded in paraffin, and processed routinely, and the sections were stained with hematoxylin and eosin (H&E). The design of this study was approved by the local ethical committee of Gulhane Military Medical Academy in Ankara. 2.2. Immunohistochemistry All tissue sections were immunostained against claudin-4 (rabbit polyclonal, 1:150; Neomarkers, Fremont, CA), E-cadherin (rabbit monoclonal, 1:200; Cell Marque, Rocklin, CA), and β-catenin (mouse monoclonal, 1:200; Dako, Carpinteria, CA) by using an autostainer (Ventana Benchmark XT; Ventana Medical Systems, Inc, Tucson, Arizona), following the manufacturer's protocol. Appropriate staining was confirmed using external positive controls for each antibody. The intensity of the staining was scored as follows: 0, negative; +, weak; ++, moderate; or +++, strong. All stained slides have independently been evaluated by 2 blinded pathologists (N.Y. and E.Ç.). 3. Results Claudin-4 exhibited membranous staining without any cytoplasmic or nuclear marking in all groups. In the control group, claudin-4
exclusively and faintly highlighted upper layer of epidermis as was documented previously (Fig. 1a) [8]. The inner root sheath of hair follicle and some attached sebaceous cells showed weak membranous labeling (Fig. 1b). Also, both secretory and ductal cells of the eccrine sweat glands demonstrated strong and membranous marking (Fig. 1c). Interestingly, in all spiradenomas, claudin-4 displayed strong and membranous staining in the cells, which were morphologically corresponded to the ductal structures on H&E-stained slides, and spared the other components of the glandular cells of small dark and larger pale cells (Fig. 1d-f). Appearance of this distinct ductal staining on the sections gave the impression of an easily detectable lace-like network, which exposed the typical ductal part of the spiradenomas. Subsequently, to be sure whether these stained cells in spiradenomas are ductal nature indeed, we used further immunohistochemical stain against carcinoembryonic antigen (CEA) (rabbit polyclonal, 1:400, Cell Marque, Rocklin, CA) labeling of which was reported to be more specific for ducts than for the secretory portion, on a few selected cases of the spiradenomas [9]. We observed moderate staining in the luminal surface of the many ductal cells with a minute population of secretory cells, confirming that the claudin-4 selectively highlighted the ductal parts of the spiradenomas (Fig. 1g). When we compared the immunostaining intensity of claudin-4 and CEA in these ductal cells, claudin-4 immunoexpression was definitely stronger and extensive than CEA. On the other hand, claudin-4 was somewhat positive in the luminal cell layer of microcysts in the poromas, syringomas, and hidradenomas, in a way of noncontributory to their differential diagnosis (Fig. 2a-c).
Fig. 1. In the uninvolved skin sample, claudin-4 immunostaining was weak in the granular layer of epidermis (a), moderate in the cells of inner root sheath of hair follicle with expression in some sebaceous cells (b), and strong in the eccrine sweat glands (c) (immunoperoxidase, original magnification ×400). (d) Spiradenoma containing the small dark and larger pale cells with scattered ductal structures (H&E, ×100). (e and f) Claudin-4 expression in the tumor selectively reveals the ductal component, with a lace-like appearance, on a background of the purely negative glandular cells (immunoperoxidase, ×400 and ×200, respectively). (g) CEA highlighted the luminal surfaces of some ductal cells with a weaker intensity compared with the claudin-4 (immunoperoxidase, ×400). (h and i) E-cadherin and β-catenin showing diffuse and strong membranous staining in all cell types of spiradenomas (immunoperoxidase, both ×400).
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
In the cases of SCCs, claudin-4 displayed moderate staining in apparently more mature and keratinized squamous cells, which were only a minor population and mostly located in the central portions of the tumoral nests (Fig. 2d). The BCCs cases were all negative against claudin-4. Its expression profile in the peritumoral skin did not differ from the control group. E-cadherin indicated diffuse, moderate to strong membranous staining in epidermis, with an exception of granular layer, and dermal adnexal structures of the control cases. It showed diffuse and strong staining in all cells of the poromas, syringomas, hidradenomas, and BCCs. E-cadherin was positive in the well- and moderately differentiated squamous cells in SCCs, but not in the cells showing high degree of atypia. E-cadherin strongly highlighted glandular and ductal cells of the spiradenomas (Fig. 1h). β-Catenin showed diffuse but weak membranous staining in all epidermal and dermal appendages' cells. In the cases of BCC, it was positive in the peripherally located atypical cells of the nests with a membranous and cytoplasmic manner and usually negative in the center of the tumor nests. The remaining groups of poroma, syringoma, hidradenoma, and SCC demonstrated diffuse and weak to moderate staining with β-catenin. All epithelial cells of the spiradenomas were strongly highlighted with β-catenin (Fig. 1i). To summarize, the staining characteristics of both E-cadherin and β-catenin were not specific for any entity and not seemed to be useful to discriminate from each other. 4. Discussion Tight junctions are located in the most apical part of the epithelial cells to maintain the polarity, which is extremely critical to the normal structure and physiological function of the epithelium, as a whole [10]. The best-characterized tight junction components are the families of claudins, occludin, and zonula occludens proteins [11,12]. The claudins are a group of major transmembrane proteins and play essential roles in tight junction formation, integrity, and function [13]. A total of 27 different subtypes of claudin family in human tissues have been described to date [14]. They are variably expressed in normal cells depending on
3
the tissue types. Aberrant expression of the claudins, resulting in dysregulation of tight junctional function, is a known phenomenon in the development of tumorigenesis, in particular for carcinomas. Several published studies reported that claudin subtypes of 1, 2, 4, 5, 6, 7, 10, 12, and 18 were especially found to be more altered and associated in the development of different tumors [15–17]. It has been found that claudin-4 was overexpressed in carcinomas of the pancreas, breast, kidney, and ovary [15,18,19]. Conversely, its loss of expression has been reported in colorectal carcinomas and mesotheliomas [20,21]. Its overexpression was also shown to be a poor prognostic parameter in clear cell renal cell carcinomas and nasopharyngeal carcinomas, whereas it has been regarded as a good prognostic factor in hepatocellular and breast carcinomas [22–25]. To the best of our knowledge, although there are many studies investigating the immunoexpression profile of claudin-4 in nonskin neoplasms, it has not been assessed in skin tumors, except SCC and actinic keratosis [26]. In that study, Hintsala et al [26] showed that claudin-4 displayed a weaker overall expression in the cases of SCC and actinic keratosis compared with the normal skin. The eccrine sweat glands consist of secretory (glandular) and ductal portions. The secretory portion is composed of inner layer of secretory epithelial cells, also known as pyramidal cells, and outer layer of myoepithelial cells. The secretory cells are arranged a mixture of small, dark basaloid cells with hyperchromatic nuclei and larger cells with a clear cytoplasm and paler nucleus. Although the dark cells tend to line the luminal surface, the pale cells are situated peripherally. Besides, the eccrine ducts are lined by a double row of small cuboidal epithelial cells. The spiradenomas are benign tumors and consisted of 2 distinctive cell components similar to those of normal eccrine glands: (1) glandular cells, mostly centrally placed small dark and peripheral large pale cells, both of which occur aggregates in sheets, cords, or clusters and correspond to the secretory portion of the eccrine gland, and (2) ductal cells, which form easily recognizable ductlike structures throughout the glandular cell nests. The tumor may also harbor lymphocytes, variable amount of squamous cells, and microvessels.
Fig. 2. (a-c) Claudin-4 marking the luminal cells of microcysts in poroma, syringoma, and hidradenoma, respectively (immunoperoxidase, ×100, ×400, and ×200, respectively). (d) Claudin-4 showing mature keratinizing squamous cells localized in the central parts of the tumoral nests in SCC (immunoperoxidase, ×200).
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005
4
N. Yiğit et al. / Annals of Diagnostic Pathology xxx (2015) xxx–xxx
In the present study, claudin-4 was found to highlight ductal component of the spiradenomas and completely spared the secretory cells. Although claudin-4 stains both secretory and ductal cells of the normal eccrine glands, it was expressed in only ductal cells of spiradenomas. This finding somehow might reflect the ontogenetic differences between those cell types. Besides, the poromas, syringomas, and hidradenomas are all well-known tumors having eccrine sweat gland differentiation. In our study, we observed that claudin-4 marks all cell types of the normal eccrine glands, which are composed of both secretory and ductal cells. However, it selectively stains the cells which form duct-like structures or the cells lining microcystic spaces in spiradenomas, poromas, syringomas, and hidradenomas. Interestingly, the distribution and the intensity of claudin-4 staining were so specific and strong to the ductal cells in spiradenoma cases that one can observe all the lace-like interconnected networks of duct-like structures, which are relatively a minor component when compared with the overwhelming glandular cells conspicuously throughout the tumor. Thus, it might be hypothesized that claudin-4 protein might be somehow down-regulated in the glandular cells and spared in ductal cells during the tumorigenesis of the spiradenomas. The same phenomenon may be true for other eccrine gland derived tumors. For the practical point of view, the claudin-4 seems to be a very specific marker for the ductal cells, either normal or neoplastic nature in all types of eccrine gland tumors. Our study showed that E-cadherin and β-catenin demonstrated unremarkable and not too useful staining profiles in differential diagnosis. The postulated origin of the spiradenomas has been still debatable. The spiradenomas are commonly believed to be originated from the straight intradermal portion of eccrine ducts [27]. Alternatively, a recent study claimed that the spiradenomas are actually derived from the hair follicle bulge, the folliculosebaceous-apocrine unit [28]. This assertion was reached based on the identical immunophenotypical findings, between hair follicle bulges and tumors, with the use of follicular stem cell markers of CD200, CK15, CK19, and PHLDA1. In our study, we observed that claudin-4 stains the inner root sheath, but we did not detect any positivity in follicule bulge both in control and in tumors. Therefore, we concluded that the spiradenomas are most likely be derived from eccrine sweat glands as is widely considered. As a note for the spiradenomas harboring a high degree of vascularity, the vascular-rich histomorphology can mimic a cellular hemangioma at the first glance and may cause a diagnostic pitfall [29]. In these cases, expanding the immunohistochemical panels only with vascular markers such as CD31 and CD34, highlights the already conspicuous blood vessels in the lesion and may pave the way to the misdiagnosis of a hemangioma, in exchange for overlooking of the spiradenoma. To prevent this, the basic immunohistochemistry panel with epithelial membrane antigen and CEA, which are 2 frequently used markers to exclude the possibility of other neoplasms including the soft tissue tumors, should be further extended. However, it should be kept in mind that negative or focal positive staining with epithelial membrane antigen and CEA can be seen in some spiradenomas, as was well documented previously, and these do not always provide evidence indicating the epithelial nature of the lesion. Thus, it would be a good idea to add claudin-4 into the immunohistochemistry panel, to minimize the risk of overlooking an eccrine gland tumor with microvessel-rich or not very cellular ones. 5. Conclusion We think that claudin-4 immunostaining can be a useful tool for the diagnosis of spiradenoma when histomorphologic and clinical findings are not distinctive. Claudin-4 immunohistochemistry effectively discriminates spiradenomas from the other possibilities in differential diagnosis, particularly against the possibilities of poroma, syringoma, hidradenoma, or cellular hemangioma, with a distinct staining on the ductal cells reminiscent of a lacy network. Our findings also support that the spiradenomas seem to have an eccrine gland origin rather than hair follicle bulge.
Acknowledgments We would like to thank Emrah Özenç for his excellent technical assistance in immunohistochemistry. References [1] Evans WH, Martin PE. Gap junctions: structure and function (review). Mol Membr Biol 2002;19:121–36. [2] Paris L, Tonutti L, Vannini C, Bazzoni G. Structural organization of the tight junctions. Biochim Biophys Acta 1778;2008:646–59. [3] Piepenhagen PA, Nelson WJ. Defining E-cadherin–associated protein complexes in epithelial cells: plakoglobin, beta- and gamma-catenin are distinct components. J Cell Sci 1993;104:751–62. [4] Langbein L, Pape UF, Grund C, Kuhn C, Praetzel S, Moll I, et al. Tight junction-related structures in the absence of a lumen: occludin, claudins and tight junction plaque proteins in densely packed cell formations of stratified epithelia and squamous cell carcinomas. Eur J Cell Biol 2003;82:385–400. [5] Mullin JM. Epithelial barriers, compartmentation, and cancer. Sci STKE 2004;2004 [pe2]. [6] Itoh M, Bissell MJ. The organization of tight junctions in epithelia: implications for mammary gland biology and breast tumorigenesis. J Mammary Gland Biol Neoplasia 2003;8:449–62. [7] Iwatsuki M, Mimori K, Yokobori T, Ishi H, Beppu T, Nakamori S, et al. Epithelialmesenchymal transition in cancer development and its clinical significance. Cancer Sci 2010;101:293–9. [8] Kubo T, Sugimoto K, Kojima T, Sawada N, Sato N, Ichimiya S. Tight junction protein claudin-4 is modulated via ΔNp63 in human keratinocytes. Biochem Biophys Res Commun 2014;455:205–11. [9] Saga K. Histochemical and immunohistochemical markers for human eccrine and apocrine sweat glands: an aid for histopathologic differentiation of sweat gland tumors. J Investig Dermatol Symp Proc 2001;6:49–53. [10] Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 2001;16:126–30. [11] Tsukita S, Katsuno T, Yamazaki Y, Umeda K, Tamura A, Tsukita S. Roles of ZO-1 and ZO-2 in establishment of the belt-like adherens and tight junctions with paracellular permselective barrier function. Ann N Y Acad Sci 2009;1165:44–52. [12] Günzel D, Yu AS. Claudins and the modulation of tight junction permeability. Physiol Rev 2013;93:525–69. [13] Lal-Nag M, Morin PJ. The claudins. Genome Biol 2009;10:235. [14] Mineta K, Yamamoto Y, Yamazaki Y, Tanaka H, Tada Y, Saito K, et al. Predicted expansion of the claudin multigene family. FEBS Lett 2011;585:606–12. [15] Rangel LB, Agarwal R, D'Souza T, Pizer ES, Alò PL, Lancaster WD, et al. Tight junction proteins claudin-3 and claudin-4 are frequently overexpressed in ovarian cancer but not in ovarian cystadenomas. Clin Cancer Res 2003;9:2567–75. [16] Merikallio H, Pääkkö P, Harju T, Soini Y. Claudins 10 and 18 are predominantly expressed in lung adenocarcinomas and in tumors of nonsmokers. Int J Clin Exp Pathol 2011;4:667–73. [17] Tokuhara Y, Morinishi T, Matsunaga T, Ohsaki H, Kushida Y, Haba R, et al. Claudin-1, but not claudin-4, exhibits differential expression patterns between well- to moderatelydifferentiated and poorly-differentiated gastric adenocarcinoma. Oncol Lett 2015;10:93–8. [18] Tsutsumi K, Sato N, Tanabe R, Mizumoto K, Morimatsu K, Kayashima T, et al. Claudin-4 expression predicts survival in pancreatic ductal adenocarcinoma. Ann Surg Oncol 2012;19(Suppl. 3):S491–9. [19] Jiwa LS, van Diest PJ, Hoefnagel LD, Wesseling J, Wesseling P, Dutch Distant Breast Cancer Metastases Consortium, Moelans CB. Upregulation of claudin-4, CAIX and GLUT-1 in distant breast cancer metastases. BMC Cancer 2014;14:864. [20] 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. [21] Süren D, Yıldırım M, Kaya V, Alikanoğlu AS, Bülbüller N, Yıldız M, et al. Loss of tight junction proteins (claudin 1, 4, and 7) correlates with aggressive behavior in colorectal carcinoma. Med Sci Monit 2014;20:1255–62. [22] Lechpammer M, Resnick MB, Sabo E, Yakirevich E, Greaves WO, Sciandra KT, et al. The diagnostic and prognostic utility of claudin expression in renal cell neoplasms. Mod Pathol 2008;21:1320–9. [23] Szasz AM, Nemeth Z, Gyorffy B, Micsinai M, Krenacs T, Baranyai Z, et al. Identification of a claudin-4 and E-cadherin score to predict prognosis in breast cancer. Cancer Sci 2011;102:2248–54. [24] Bouchagier KA, Assimakopoulos SF, Karavias DD, Maroulis I, Tzelepi V, Kalofonos H, et al. Expression of claudins-1, -4, -5, -7 and occludin in hepatocellular carcinoma and their relation with classic clinicopathological features and patients' survival. In Vivo 2014;28:315–26. [25] Suren D, Yildirim M, Kaya V, Elal R, Selcuk OT, Osma U, et al. Expression patterns of claudins 1, 4, and 7 and their prognostic significance in nasopharyngeal carcinoma. J BUON 2015;20:212–7. [26] Hintsala HR, Siponen M, Haapasaari KM, Karihtala P, Soini Y. Claudins 1, 2, 3, 4, 5 and 7 in solar keratosis and squamocellular carcinoma of the skin. Int J Clin Exp Pathol 2013;6:2855–63. [27] Mambo NC. Eccrine spiradenoma: clinical and pathologic study of 49 tumors. J Cutan Pathol 1983;10:312–20. [28] Sellheyer K. Spiradenoma and cylindroma originate from the hair follicle bulge and not from the eccrine sweat gland: an immunohistochemical study with CD200 and other stem cell markers. J Cutan Pathol 2015;42:90–101. [29] Jorquera Barquero E, Lara Bohórquez C, de Alba Rioja I. Giant vascular eccrine spiradenoma. Actas Dermosifiliogr 2015. http://dx.doi.org/10.1016/j.ad.2015.03. 019 [Epub ahead of print].
Please cite this article as: Yiğit N, et al, Distinctive immunostaining of claudin-4 in spiradenomas, Ann Diagn Pathol (2015), http://dx.doi.org/ 10.1016/j.anndiagpath.2015.10.005