Gastric cancer (GC) is the fifth most common cancer and the third cause of cancer-related death worldwide [1
]. Lauren’s classification, firstly reported in 1965, is currently used to distinguish two gastric adenocarcinoma subtypes based on histological and clinical features: intestinal-type gastric cancer (IGC) and diffuse-type gastric cancer (DGC) [2
]. It is widely accepted that IGC and DGC represent distinct disease entities with different epidemiology, etiology, carcinogenesis, progression and, to some extent, biological behaviors [3
gene encodes E-cadherin (Epithelial-cadherin), a Ca2+
-dependent transmembrane glycoprotein involved in cell−cell adhesion maintenance in epithelia [4
]. Accordingly, E-cadherin plays a crucial role in epithelium−mesenchymal transition (EMT): its underexpression reduces cell-cell cohesion, making it possible for tumor cells to dissociate from primary tissue, invade surrounding tissues and disseminate to other sites [6
]. Indeed, numerous studies have highlighted E-cadherin as a critical tumor suppressor in several carcinomas, including GC [9
is especially altered in hereditary diffuse GC (HDGC), where complete loss of protein expression often occurs due to a germline lesion and to a second hit, following Knudson’s theory of tumor suppressor gene inactivation [11
]. With regard to sporadic tumors, which account for 90% of GCs, it has been reported that epigenetic and structural alterations are as frequent in IGC as in DGC, suggesting histotype independence [12
]. Nonetheless, CDH1
status in IGC is not as extensively studied as in DGC.
In the last decade, the focus has been placed on alternative mechanisms capable of modifying E-cadherin expression, including epigenetic control exerted by noncoding transcripts. MicroRNAs (miRs) are small non-coding RNAs, which negatively regulate gene expression and orchestrate pathways involved in cell-cycle control, proliferation, apoptosis, angiogenesis, metastasis, and DNA-damage response in cancers, including GC [13
]. Of note, by downregulating the expression of target cancer-related genes, several miRs have been shown to be directly involved in carcinogenic processes, including EMT [20
]. In this scenario, CDH1
transcription is often susceptible to expression control of miRs and long noncoding RNAs [22
]. This occurs in both direct and indirect manners due to the number of pathways regulating CDH1
expression through the activity of Snail, Slug, ZEB1/2, EZH2, and Twist transcription factors [23
In this study, we aimed at analyzing the regulation of CDH1 expression in a case series of IGC and matched normal tissue samples, by focusing on the role of miRs that target CDH1 gene, either directly or indirectly.
E-cadherin is a transmembrane glycoprotein that plays a pivotal role in maintaining epithelial architecture and cell polarity [5
]. In the last few decades, E-cadherin tumor suppressor function has emerged in different epithelial tumors, including GC, and its downregulation has been observed during neoplastic progression and in association with tumor invasion [25
] and metastasis [8
]. In 1998, Karayiannakis et al. described aberrant or absent E-cadherin protein expression in both IGC and DGC [30
]. It has been reported that CDH1
alterations, both structural and epigenetic, occur in almost one-third of sporadic GCs, with slightly higher frequencies in DGC than in IGC [12
]. More recently, data mining of different repositories indicates that CDH1
is the second gene related to IGC [31
], raising the question of the mechanisms that contribute to its expression in this histotype. In recent years, miRs have emerged as promoters and suppressors of carcinogenesis and metastasis in many types of cancers [32
]. With regard to GC, a wide range of miRs have been associated with Helicobacter pylori
(HP)-related infection, a well-established event in IGC carcinogenesis [33
]. Interestingly, E-cadherin downregulation has been described in concomitance with HP infection-derived neutrophil infiltration [34
]. Moreover, it has been shown that several miRs are involved in epithelial-mesenchymal transition (EMT), modulating E-cadherin expression by directly targeting CDH1
or acting on one or more of its transcription factors, including EZH2, ZEB1, ZEB2, and Slug [21
]. Interestingly, these pathways emerged to have a role also in chemotherapeutic resistance in GC, CDH1
direct or indirect restoration may be a useful way to reduce it in such tumors [35
In this study, we aimed to characterize the CDH1
expression levels and its transcriptional regulation by investigating the impact of miRs on IGC carcinogenesis. We found that CDH1
is transcriptionally downregulated in 42.4% of IGCs, further confirming the importance of CDH1
downregulation in this gastric cancer histotype [36
Among the miRs filtered by in silico analysis as direct/indirect regulators of CDH1 expression, miR-34c, miR-506, miR-217, miR-199a, miR-153, and miR-544 were undetectable in both normal and tumor samples from a cohort of 17 IGCs. Conversely, among evaluable miRs, miR-101, miR-26b, and miR-200c proved to be significantly downregulated in IGC compared to control tissue both in this exploratory cohort and in our 33-patient case series.
miR-101 has been reported to act as a tumor suppressor by targeting CDH1
inhibitors, such as ZEB1/ZEB2
in different tumors [37
], including GC [39
]. Low levels of miR-101 in plasma have been reported to be associated with GC progression [41
] and HP-induced inflammation [42
]. Our data, in accordance with previous results from our group [36
], showed that miR-101 is significantly downregulated in IGC patients. However, likely due to the limited number of patients, no association between miR-101 expression and clinical pathological parameters, including HP infection, emerged from this study.
miR-26b is expressed at low levels in GC, and its downregulation is associated with a higher TNM classification and shorter survival [44
]. Several studies have shown that miR-26b, like miR-101, inhibits EZH2
expression leading to CDH1
downregulation in many tissues, including GC [45
]. In agreement with these findings, we observed a significant downregulation of miR-26b in tumor specimens compared to the normal counterparts.
miR-200 family members are known as transcriptional repressors of E-cadherin through the regulation of ZEB1
. In gastric cell lines, miR-200 family members are markedly downregulated during EMT with a concomitant decrease in E-cadherin, and their lower expression has been associated with poor prognosis in patients [46
]. There is evidence that miR-200 family members act as tumor suppressors in GC [49
]. Among these, miR-200c showed a significantly lower expression in neoplastic tissue than normal gastric mucosa in our study. Moreover, in line with previous findings [51
], its expression was significantly associated with tumor grade, being lower in poorly differentiated GC (G3) than in more differentiated tumors (G1 and G2). These observations reveal that miR-200c is potentially involved in IGC cell differentiation status.
Overall, although miR-101, miR-26b, and miR-200c were downregulated in more than half of our patients, only half of them showed concomitant CDH1 downregulation, indicating a more complex role in regulatory networks for these miRs.
We also found higher EZH2
expression levels in IGC specimens and we detected a statistically significant CDH1
downregulation in IGC patients showing EZH2
upregulation. In accordance with previous studies suggesting that miRs targeting EZH2
may be associated with the perturbation of E-cadherin expression [39
], we observed a statistically significant inverse association between EZH2
and miR-101/26b expression levels.
In conclusion, our results reinforced the emerging link between E-cadherin and intestinal-type GC and confirmed the role of EZH2 as a regulator of CDH1 expression. Furthermore, our findings highlighted the potential of some specific miRs to be exploited as molecular markers of tumorigenesis and aggressiveness in this specific cancer histotype.
However, the small number of patients involved was not sufficient for us to identify any significant correlation between the analyzed markers, making it necessary to perform these studies on larger cohorts in order to further refine the biomarkers selection and to identify new therapeutic targets in IGC.