Carcinogenesis is a complex biological process involving cellular proliferation, invasion, and alterations in adhesion, with the cytoskeleton playing a pivotal role in maintaining cell shape, polarity, and mechanical stability. Among the proteins that anchor cytoskeletal elements ‒ such as microfilaments, intermediate filaments, and microtubules ‒ to cell junctions, the plakin family has emerged as essential for both tissue integrity and tumor biology. Members of this family, including Desmoplakin (DSP), Envoplakin (EVPL), Periplakin (PPL), Plectin (PLEC), and Bullous Pemphigoid Antigen-1 (BPAG1), provide structural links between cytoskeletal filaments and junctional complexes such as desmosomes and hemidesmosomes.1 Alterations in their expression or localization have been associated with various cancers, and some plakins have been proposed as potential biomarkers. For example, elevated serum anti-BPAG1 autoantibodies have been reported in melanoma patients compared with healthy individuals, although these findings have not been consistently replicated across studies.2,3 Furthermore, studies in breast cancer have shown that decreased levels of plakins accompany early cytoskeletal disorganization. Their reduced expression leads to centrosome mispositioning and weakening of intercellular junctions ‒ changes that reflect loss of epithelial polarity and are characteristic of Epithelial-Mesenchymal Transition (EMT).4

Epiplakin (EPPK1) represents a less-characterized member of the plakin family, distinguished by its unique arrangement of tandem plakin repeat domains. Unlike other plakins, EPPK1 lacks spectrin repeats and a clear actin-binding domain, suggesting a divergent structural and functional role. While it is expressed in multiple epithelial tissues, its contribution to tumorigenesis remains incompletely understood, and studies on its involvement in cancer are still limited.1 Data from GEPIA2 indicate that EPPK1 mRNA levels are reduced in cutaneous melanoma, whereas UALCAN-based proteomic datasets show that protein-level alterations remain unclear.5,6 Given that melanocytes and keratinocytes are interconnected through cadherin-based junctions, plakin proteins such as EPPK1 may contribute to the stabilization of these adhesion complexes. Clarifying whether EPPK1 expression varies between benign and malignant cutaneous lesions may offer diagnostic value.7 Despite this potential relevance, systematic studies examining EPPK1 expression across diverse cutaneous malignancies ‒ such as Squamous Cell Carcinoma (SCC) and Basal Cell Carcinoma (BCC) ‒ are currently lacking, and its expression profile in benign lesions like intradermal nevi has not been characterized either.

Previous studies have linked EPPK1 to the development of several cancer types, including hepatocellular, cervical, colorectal, bladder urothelial, and esophageal squamous cell carcinomas.8–12 Given that plakins contribute to cytoskeletal anchorage and junctional stability, altered EPPK1 expression may influence adhesion-related pathways that are also reflected in cadherin dynamics. Because the balance between E-cadherin and N-cadherin is an important indicator of adhesion loss and invasive potential, evaluating their relationship with EPPK1 expression may help clarify whether this plakin participates in adhesion remodeling in skin tumors. The present study aims to assess the immunohistochemical expression of EPPK1 alongside E-cadherin and N-cadherin in SCC, BCC, intradermal nevi, and normal skin specimens. Expression levels are compared across lesion types, and correlations with Breslow thickness are analyzed to explore associations with tumor invasiveness. Normal skin specimens serve as controls. Findings from this study may provide new insights into the diagnostic significance of EPPK1 and its potential interplay with cadherin-mediated adhesion in cutaneous lesions. In addition to the immunohistochemical evaluation, this study also aims to investigate the molecular interaction network of plakin family proteins and to identify their overlap with SCC-associated protein networks. Furthermore, GO and KEGG enrichment analyses are conducted to identify significantly enriched biological processes and signaling pathways associated with the overlapping proteins.

The cytoskeleton and its associated junctional complexes play a central role in maintaining epithelial tissue integrity by regulating cell adhesion, polarity, and mechanotransduction ‒ processes that are profoundly altered during carcinogenesis. Plakin family proteins serve as critical structural bridges linking intermediate filaments to desmosomes and hemidesmosomes, thereby stabilizing epithelial architecture under physiological conditions.1 Accumulating evidence indicates that disruption of plakin-mediated cytoskeletal anchoring enhances cellular plasticity, promotes migratory capacity, and contributes to tumor progression across various malignancies. For instance, studies in ovarian cancer have shown that decreased PLEC and PPL expression in high-grade tumors coincides with reduced structural stability and a shift toward a more permissive state for invasion, even in the absence of overt EMT marker changes. These observations suggest that plakin loss may represent an early structural destabilization, priming tumor cells for migration by weakening epithelial anchorage.21 PPI network analysis of plakin family members in SCC revealed a highly interconnected network comprising 246 nodes and 8258 edges, with an average node degree of 67.1 and a clustering coefficient of 0.692 (PPI enrichment p < 1.0e16), highlighting a biologically meaningful organization. KEGG pathway enrichment of the overlapping proteins highlighted significant associations with multiple cancer-related pathways, including proteoglycans (FDR < 0.05). These results suggest that plakins function not only as structural components but may also participate in SCC signaling cascades that regulate tumor cell behavior, adhesion, and migration. Experimental evidence from HeLa cells shows that EPPK1 knockdown accelerates keratinocyte motility and induces cytoskeletal rearrangements, whereas overexpression suppresses motility.22

Among plakins, EPPK1 is distinctive in that it lacks spectrin repeats and a clear actin-binding domain, and its role in cancer remains poorly understood.1,23 Research on EPPK1 is still limited, and its involvement in cutaneous malignancies has yet to be fully elucidated. Previous studies in esophageal SCC have reported markedly elevated EPPK1 expression compared with normal controls, whereas in colorectal adenocarcinomas, EPPK1 expression was decreased, suggesting a context-dependent role for this protein.10,12 A recent review on head and neck cancers highlighted that desmosomal components, including plakin proteins, can paradoxically act, as their expression, localization, and interactions dynamically change during cancer progression.24 Likewise, although plectin is frequently upregulated in many tumor types, its downregulation has been reported in certain malignancies, emphasizing a dual expression pattern and underscoring the pivotal role of plakins in regulating tumor progression depending on cellular and tissue context.25 These observations suggest that the functional consequences of EPPK1 expression may vary according to tissue context and tumor type, highlighting the need for further investigation, particularly in cutaneous tumors. In cutaneous cancers, elevated serum anti-BPAG1 autoantibodies have been reported in melanoma patients compared with healthy individuals, indicating a potential, yet still uncertain, role of plakins as biomarkers.2,3 In the present study, the authors aimed to systematically evaluate the immunohistochemical expression of EPPK1 in benign intradermal nevi, BCC, and SCC to clarify its potential role in skin tumor biology. The present findings demonstrated significantly higher EPPK1 expression in cutaneous SCC (392.66 ± 55.24) compared with BCC (70.39 ± 15.32) and benign intradermal nevi (27.08 ± 3.82) (p < 0.05). This variation likely reflects intrinsic cell type-dependent differences, as keratinocytic SCC cells possess distinct cytoskeletal architecture and keratinization-related features compared with BCC and benign melanocytic lesions, potentially contributing to their higher EPPK1 expression. Given this differential expression pattern, EPPK1 may have potential diagnostic utility in cutaneous malignancies. Routinely used markers, such as E-cadherin, p53, and certain metalloproteinases, have been suggested to aid in differentiating SCC from pseudocarcinomatous hyperplasia. However, a limitation of these studies was that marker expression was not analyzed according to tumor grade, and the differentiation of well-differentiated SCC from pseudocarcinomatous hyperplasia remained challenging.26 In this study, EPPK1 demonstrated a distinct expression pattern in low- to mid-grade SCC, suggesting that it may provide additional diagnostic value. Notably, in well-differentiated SCC regions, E-cadherin immunoreactivity remained largely at intercellular junctions, while EPPK1 showed intense cytoplasmic expression. In areas where some tumor cells exhibited increased cytoplasmic E-cadherin, EPPK1 expression was already widespread, indicating that changes in EPPK1 may precede detectable alterations in E-cadherin localization.

In normal skin, EPPK1 expression was predominantly observed in the suprabasal layers of the epidermis, with the most intense immunoreactivity localized near the granular layer, suggesting a role in late keratinocyte differentiation and epidermal barrier organization. In SCC samples, EPPK1 expression was particularly prominent in tumor nests and was most intense in squamous cells surrounding keratin pearls. This distinct spatial distribution may indicate that EPPK1 expression in SCC is associated with areas of keratinization and squamous differentiation. Such spatial patterns align with findings from epithelial models showing that keratin intermediate filaments and plakin family cytolinkers cooperatively stabilize keratin-rich surface structures, supporting the idea that plakins may influence keratinization dynamics in human epidermal lesions.27 Beyond malignancy, EPPK1 has also been implicated in epithelial barrier regulation. In psoriasis, EPPK1 is specifically downregulated in an interferon-γ–dependent manner, and its deficiency has been associated with impaired epithelial adhesion and barrier-related gene expression, supporting a role for EPPK1 in maintaining epithelial stability.28 Consistent with this concept, in the non-melanoma skin cancer group, EPPK1 expression showed a significant positive correlation with E-cadherin, suggesting that EPPK1 may be functionally linked to cadherin-mediated adhesion dynamics in cutaneous epithelial lesions.

Plakin family members can exert distinct and sometimes opposing effects in tumor biology. Notably, DSP displays a consistent tumor-suppressive pattern, with marked reductions observed across oral and lung carcinomas. This loss has been linked to poorer clinical outcomes, and experimental data demonstrate that DSP depletion enhances keratinocyte proliferation and activates prosurvival ERK/Akt signaling, whereas its overexpression suppresses lung cancer cell growth via modulation of Wnt pathway mediators. In contrast, the plakin-related protein MACF1 exhibits a more oncogenic profile, being highly expressed in glioblastoma; its knockdown reduces proliferation and migration while downregulating Wnt pathway components.23 Regarding EPPK1, previous studies in esophageal SCC reported functional knockdown experiments revealing reductions in cell proliferation, colony formation, migration, and invasion.10 Similarly, in adenocarcinomas such as colorectal cancer, EPPK1 expression was positively correlated with Ki67, suggesting a role in cellular proliferation.12 In line with these findings, the GO functional enrichment analyses demonstrated that the shared protein network formed by plakin family members in SCC was predominantly associated with cell proliferation–related processes, including “regulation of epithelial cell proliferation”, “positive regulation of cell population proliferation”, and “regulation of fibroblast proliferation”, all enriched with highly significant FDR values (FDR < 0.05). Beyond proliferative signaling, the network also showed strong associations with cytoskeletal organization, especially “intermediate filament cytoskeleton organization”, consistent with the canonical structural roles of plakins. Interestingly, reflecting the known dual and sometimes opposing functions of plakin proteins, the GO analysis revealed that the overlapping proteins were enriched in both positive and negative regulation of cell adhesion, highlighting their context-dependent contribution to junctional stability or loosening. Furthermore, enrichment in “positive regulation of cell migration” and processes linked to apoptotic signaling underscores a potential involvement of plakins in pathways that facilitate tumor cell motility and survival during carcinogenesis. In the context of the present study, these functional signatures guided the investigation of EPPK1 expression in cutaneous malignancies, prompting us to examine its associations with EMT markers and Breslow thickness to better understand its potential contribution to tumor progression.

EMT involves the loss of cell–cell junctions and epithelial polarity, including adherens junctions. Cadherins, as adhesion molecules, can also act as signaling mediators, influencing cellular behaviors such as migration, proliferation, apoptosis, and differentiation.29 Pogorzelska-Dyrbuś et al. reported significantly higher E-cadherin and N-cadherin expressions in SCC, with N-cadherin levels being significantly elevated compared with BCC, which was associated with a relatively higher metastatic potential.30 Similarly, Kim et al. observed increased vimentin expression in SCC.31 In the present study, although SCC samples exhibited higher N-cadherin levels (216.37 ± 59.83), the difference compared with BCC (139.61 ± 45.47) was not statistically significant (p > 0.05), likely due to the predominance of low- and mid-grade SCCs in the studied cohort. Suiqing et al. reported that E-cadherin expression is markedly lower in poorly differentiated SCC compared with well-differentiated tumors.32 Consistently, in the present study, E-cadherin showed intense cytoplasmic immunoreactivity in SCC cases (389.94 ± 70.56, p < 0.05). In the literature, this shift of cadherins from the cell membrane to the cytoplasm is often interpreted as a functional loss of these adhesion molecules, potentially compromising cell integrity and facilitating malignant transformation and metastasis during EMT.29 Interestingly, EPPK1, which exhibited a positive correlation with E-cadherin in the non-melanoma skin cancer samples, may similarly be involved in EMT-related or other cellular processes, including proliferation. Its immunoreactive localization with enhanced cytoplasmic expression was observed particularly in nests of squamous cells surrounding keratin pearls. The positive moderate correlation between EPPK1 expression and Breslow thickness in SCC may partly reflect increased keratinization and the more prominent formation of keratin pearls in thicker tumors, given the structural association of EPPK1 with the cytoskeletal network.

In lung adenocarcinoma, EPPK1 knockdown led to increased E-cadherin expression and a concomitant decrease in vimentin levels, suggesting a role in modulating epithelial–mesenchymal characteristics.33 In corneal epithelial wound healing models, EPPK1 deficiency was associated with decreased expression of E-cadherin, keratin-6, and vimentin, indicating a role in cytoskeletal regulation and potentially facilitating cell migration during tissue repair.34 In the present study, EPPK1 expression in non-melanoma skin cancers showed a strong positive correlation with E-cadherin (r = 0.565, p < 0.001) and a modest positive correlation with N-cadherin (r = 0.329, p < 0.05). Lopes et al. reported a positive correlation between N-cadherin and E-cadherin in melanoma, interpreted as a partial EMT state with a hybrid cadherin expression profile.35 Venza et al. found that reduced E-cadherin in cutaneous melanoma did not significantly correlate with clinical stage or Breslow thickness, and they suggested that its downregulation may be more closely associated with regulating melanoma cell proliferation.36 In the non-melanoma skin cancer samples, the low correlation between EPPK1 and N-cadherin, along with the absence of a significant correlation with Breslow thickness, indicates that EPPK1 alone may not serve as a reliable invasion marker, although it could play a role in the early stages of EMT.

Epidemiologically, intradermal nevi are predominantly observed in female patients, with 80.46% of cases reported in women.37 A retrospective analysis of cases diagnosed between 2010 and 2018 similarly confirmed this female predominance, with 1973 women and 667 men affected.38 In contrast, BCC and SCC primarily affect older adults, with the most frequent age of onset around 70–85 years.39 Specifically, cutaneous SCC generally presents around 70-years of age, with over 80% of cases occurring in individuals aged 60 or older.40 Consistently, in the present study, SCC and BCC groups exhibited a higher proportion of males and an increased mean age compared with the intradermal nevus group, reflecting known epidemiological trends.

The present study represents the first systematic evaluation of EPPK1 expression in the context of cutaneous malignancy progression. Although limited by its retrospective design, the findings provide valuable insights into the potential role of EPPK1, particularly in relation to EMT processes. Notably, EPPK1 expression was significantly higher in well-differentiated SCC compared with BCC and benign intradermal nevi, and it demonstrated a positive correlation with E-cadherin, suggesting a role in EMT dynamics and potentially other early tumor-related processes.

In conclusion, EPPK1 appears to act as a context-dependent molecule associated with EMT, showing elevated expression in keratinocytic SCC. Its differential expression pattern and correlation with key adhesion markers highlight its potential utility as a diagnostic biomarker for cutaneous malignancies. Future studies with larger cohorts, including high-grade SCCs, are warranted to further clarify the prognostic significance of EPPK1.

Özlem Türelik: 0000-0001-6057-9171

Ethical approval was obtained from the Ethics Committee of Bilecik University (approval number: 2025/7-12, Date: August 04, 2025). All methods were conducted in accordance with the ethical standards of the institutional research committee and with the Declaration of Helsinki.

The entire dataset supporting the results of this study was published in this article.

None declared.

Damla Gül Fındık: Study conception and planning; data collection, analysis and interpretation; statistical analysis; preparation and writing of the manuscript; approval of the final version of the manuscript.

Özlem Türelik: Data collection, analysis and interpretation; preparation and writing of the manuscript; approval of the final version of the manuscript.

None declared.