**Role of Claudin Proteins in Regulating Cancer Stem Cells and Chemoresistance-Potential Implication in Disease Prognosis and Therapy**

#### **Saiprasad Gowrikumar 1, Amar B. Singh 1,2,3 and Punita Dhawan 1,2,3,\***


Received: 27 October 2019; Accepted: 13 December 2019; Published: 20 December 2019

**Abstract:** Claudins are cell–cell adhesion proteins, which are expressed in tight junctions (TJs), the most common apical cell-cell adhesion. Claudin proteins help to regulate defense and barrier functions, as well as differentiation and polarity in epithelial and endothelial cells. A series of studies have now reported dysregulation of claudin proteins in cancers. However, the precise mechanisms are still not well understood. Nonetheless, studies have clearly demonstrated a causal role of multiple claudins in the regulation of epithelial to mesenchymal transition (EMT), a key feature in the acquisition of a cancer stem cell phenotype in cancer cells. In addition, claudin proteins are known to modulate therapy resistance in cancer cells, a feature associated with cancer stem cells. In this review, we have focused primarily on highlighting the causal link between claudins, cancer stem cells, and therapy resistance. We have also contemplated the significance of claudins as novel targets in improving the efficacy of cancer therapy. Overall, this review provides a much-needed understanding of the emerging role of claudin proteins in cancer malignancy and therapeutic management.

**Keywords:** claudins; cancer; stem cell; chemoresistance

#### **1. Introduction**

#### *1.1. Tight Junctions*

Tight junctions (TJs) are the sites where tissues interface directly with the external environment or internal compartments that are contiguous with the external environment and are lined by mucosal surfaces, where epithelial cells act insulation for the internal organ. These structures not only provide a protective layer but also act as a selective barrier between the body and the gut lumen that restricts free exchange across the paracellular space [1,2]. There are three main transport mechanisms across the epithelial layers, which include the trans-cellular pathway (passive diffusion), carrier dependent pathway (carrier or receptors), and the paracellular pathway (passage through spaces between cells). Among these transport mechanisms, the apical junctional complex, a crucial factor for the paracellular pathway, is composed of three junctions from apical to basal are known as the tight junction (zonula occludens), adherens junction (zonula adherens), and desmosome (macula adherens) [3]. The TJs are intercellular junctions, which act as permeability barriers in epithelial cells [1]. The tight junction proteins are diverse and include occludins (the first one to be found), claudins, tricellulin, cingulin, and junctional adhesion molecules (JAM). These proteins interact within themselves and with the cellular cytoskeleton to form a complex architecture [4–8]. Among these TJ proteins, claudins are key proteins, acting as both pores and barriers, aiding the paracellular pathway between epithelial cells [9,10].

#### *1.2. Claudins*

The functionalities of claudins are as follows: (1) Fence function, responsible for maintaining polarity by differentiating apical and basolateral cell domains; (2) Signaling molecule, involved in cell growth, survival, proliferation, and differentiation; (3) Barrier function, this gate function separates compartments with fluids to avoid intermixing [11]. Claudins were identified as a major integral membrane protein by Tsukita and his colleagues in 1998, before which the only known tight junction protein was occludin [12,13]. Studies conducted to overexpress claudins in fibroblasts, which do not have tight junctions, were able to reconstitute tight junction-like networks of strands, which shows the importance of claudins in tight junction assembly [14]. Several claudin isoforms have been identified in mammals. These have high sequence homology in the first and fourth transmembrane domains and extracellular loops. Further, the homologous classic claudins include claudins 1–10, 14, 15, 17, and 19, and non-classical claudins comprised of claudins 11–13, 16, 18, and 20–27, which are less homologous [15].

The structure of claudins is comprised of four transmembrane domains, the intracellular N and C termini, and the two extracellular loops (ECLs). The claudin structure encompasses N-termini (7 amino acids), C-termini (25–55 amino acids), and loops containing 25–55 aminoacids. The ECLs are involved in barrier and pore formations. There are two ECLs, ECL1 consists of ~50 amino acids with two conserved cysteines involved in the barrier function. Negative and positive charges in ECL1 contribute to pore formation. The schematic representation of the structure of claudins and its classification is depicted (Figure 1). The ECL2 is responsible for homo and heterotypic interactions and was recently shown to be involved in host cell binding and cytotoxicity for the *Clostridium perfringens* enterotoxin. The ECL2 usually has ~25 amino acids, but fewer in claudin-11 and more in claudin-18 [16]. Claudins interact with other TJ-associated proteins through carboxy-terminal tails, which contain a PDZ-domain binding motif [17].

**Figure 1.** Structural organization of claudin proteins (monomer), and its classification based on homologous sequences between them. Colour code: Green- transmembrane domains; Orange: Bilipid layer, Blue–Extracellular loops/N and C termini.

#### **2. Claudins as Oncogenic Signal Transducer**

The expression of claudins varies among different tissue types [18]. As an important structure in regulating paracellular permeability, claudin overexpression influences trans-epithelial resistance (TER) and ion permeability [19–22]. Aberrant expressions of claudins have been reported in various cancers. Some of the claudins known to be frequently dysregulated in cancers are claudin-1, -3, -4, and -7 [23]. A large body of evidence highlights claudins as pro and anti-tumorigenic factors [24–31]. The potential of claudins to act as proto-oncogene or tumor promotor in various cancers are summarized in Table 1. In addition, several recent studies have also demonstrated the importance of claudins as tumor suppressors [24–31]. A recent study by Chang et al. in 2019 provided evidence for intestinal hyperplasia and adenomas in claudin-7 knockdown mice [32]. Consistent with this, claudin-7 was downregulated in colon cancer patient samples as compared to normal tissue [33]. These effects of claudin-7 were achieved by inhibiting phosphorylation and nuclear localization of Akt. Conversely, claudin-7 association with Epithelial cell adhesion molecule (EPCAM) supports proliferation, upregulation of anti-apoptotic proteins, and drug resistance [33]. Claudin-18 knockout mice spontaneously developed lung adenocarcinomas, and its mRNA expression was decreased in lung adenocarcinomas. Claudin-18 inhibits Akt signaling through modulation of yes-associated protein/Taz (Yap/Taz) and insulin-like growth factor (IGF-1R) signaling in lung cancer [34]. Further, the depletion of claudin-3 induced tumor burden by enhancing β-catenin activity through (IL)-6/STAT3 signaling in colon cancer [35]. Yet another study by Che et al. in 2018 [36] identified claudin-3 as a suppressor of lung squamous cell carcinoma cells, in which overexpression of claudin-3 inhibited invasion, migration, and EMT of lung squamous cell carcinoma. Similarly, claudin-4 accelerates cell migration and invasion in ovarian tumor cell lines, in support of this, peptide-mediated silencing of claudin-4 in ovarian cancer cells exhibited lower tumor burden [37]. Claudin-6 was shown to be a tumor suppressor through genetic manipulation studies in cervical carcinoma cells wherein loss of claudin-6 exacerbated cell proliferation and tumor growth [38,39]. An array of articles from Dhawan et al., have proved a significant role of claudin-1 as a tumor promoter in colon cancer [40,41]. In one of their reports, increased claudin-1 expression was causally associated with metastasis [40]. In contrast to claudin-1, claudin-7 has an inverse role on EMT, wherein it causes mesenchymal to epithelial transformation (MET) in Rab25 dependent manner to combat colon cancer [42]. Similarly, claudin-2 is upregulated in colon cancer and is involved in cancer progression. Claudin-2 suppression in colon cancer cells has led to decreased cell proliferation through the modulation of EGF signaling [43]. Opposite colon cancer, claudin-1 is frequently down-regulated in invasive human breast cancer. Recently, mutations of claudin-1 have been reported in breast cancer, which has led to claudin-1 transcript variants shorter than classical claudin-1 transcript [44]. Taken together, it appears that the deregulated claudin composition in any given epithelial cells sheet may modify the signaling and associated changes in protein partnering to modulate oncogenesis.


**Table 1.** Claudins as tumour promotor/suppressor.

To glimpse how claudins can achieve its pro or anti-tumorigenic effect, understanding the regulation of claudins in normal and cancer cells is essential. Recently it has been demonstrated that claudins are not a static and rigid seal of the paracellular space; rather, they are dynamically capable of responding to various biochemical and mechanical stimuli through reshaping and remodeling [50,51]. Epigenetic regulation of claudins has recently gained significant importance. The claudin-3 promotor

is known to possess low DNA methylation and high histone H3 acetylation for its expression in ovarian cancer cells [52]. DNA hypomethylation of the claudin-4 promotor is an important factor for its high expressions in gastric cancer [53]. Downregulation of claudin 1 via DNA promoter methylation is reported in estrogen receptor-positive breast cancer [54]. Claudins are also regulated at the transcriptional level by different transcription factors. A study has reported novel post-transcriptional regulation of claudin-1 in colon cancer cells [55], the authors documented the role of histone deacetylase (HDAC)-dependent histone acetylation as a key post-transcriptional regulation over claudin-1 expression, as found through HDAC inhibitor studies. Studies demonstrate the interaction of Slug and Snail (transcriptional factors) with the E-box element in the claudin-1 promoter causes inhibition of claudin-1. Snail is known to act as a transcription factor causing repression of E-cadherin (E-CAD) and has a potential role in promoting tumorigenesis. Slug is also a pivotal transcription factor involved in cell migration during embryogenesis and in tumor cell invasion and migration [56]. Yet another transcriptional factor known to be associated with claudin-1 is Runt-related transcription factor 3 (RUNX3), which is a gastric tumor suppressor [47]. Caudal homeobox proteins (Cdx1 & Cdx2) and GATA binding protein 4, GATA4) are known activators of claudin-1 promoters in colon cancer [57]. Sp1 is a transcriptional factor known to regulate claudin-3 and claudin-4 promoter activity in ovarian cancer [52,58]. Apart from these transcriptional regulations, claudins are also known to be regulated by post-translational modifications involved in their protein localization, interaction with other proteins, and overall turnover [59,60]. The post-translational modification of claudins includes palmitoylation, O-glycosylation, and phosphorylation [61,62]. Phosphorylation is one of the key regulatory modifications for the regulation of intracellular localization and degradation of claudins.

Claudins are phosphorylated by many different enzymes like protein-kinase A/C, protein phosphatase 2A and mitogen-activated protein kinase (MAPK) [63,64]. The localization or dissociation of claudins to TJs is regulated by phosphorylation. For phosphorylated claudin-1, -5, and -16 are localized in the TJs while in contrast, phosphorylated claudin-3 and -4 dissociate from TJs [64–66]. Furthermore, the rho family of small dimeric G proteins mediated phosphorylation of claudin-5 at T207 was recently reported [67]. The phosphorylation of claudin-1 at different serine sites (192, 205, 206, and T191) regulates its assembly at tight junctions [68]. The cAMP-dependent protein kinase (PKA) is known to phosphorylate of claudin-3 at amino acid 192 at the C terminus. Claudin-4 is phosphorylated by atypical PKC (aPKC) at serine195 [65]. Another important posttranslational modification playing a key role in claudin regulation is palmitoylation. Emerging articles have demonstrated the importance of palmitoylation in claudin localization into tight junctions. In claudin-5 self-assembly, palmitoylation restricts specific protein-protein conformations, as reported by Rajagopal et al. [61]. Claudin-7 interacts directly with EpCAM along the basal membrane. Palmitoylation regulates the ability of claudin-7 to interact with integrins, recruiting EpCAM, and concomitantly associate with the actin cytoskeleton [69].

#### **3. Claudins and Stem Cells**

Stem cells are crucial for the development and homeostasis of many different tissues. Stem cells are also involved in cell replacement therapies in the case of cell damage or degeneration [70]. Pluripotency of stem cells is defined as self-renewing and differentiating potential into all three germ layers. Human pluripotent stem cells are very promising in regenerative medicine. The stem cell further differentiates into a wide variety of cells under the influence of diverse signaling molecules, growth factors, and transcription factors [71]. Recent research is focused on understanding the signals, which maintain pluripotency or differentiation potential. Various intrinsic and extrinsic factors are involved in stem cell maintenance, self-renewal, and differentiation [71]. On the other hand, stem cells are also an important factor for many tumors. Dysregulated pluripotent stem cells in tumors are more aggressive and have the potential to reform the whole tumor [72]. Thus, it becomes important to selectively remove undifferentiated human pluripotent stem cells (hPSCs) from differentiated cultures. For achieving this, selective pluripotent-specific cell surface markers are needed, which can separate undifferentiating from the differentiated cells. While screening for a highly specific marker protein

specific for the undifferentiated hPSCs, Uri Ben-David et al. [73] found claudin-6 to be highly specific for undifferentiated hPSCs. The expression of claudin-6 was 90-fold higher in undifferentiated hPSCs than in differentiated cells. The proof for the involvement of claudins in epithelial differentiation from embryonic stem cells was first reported by Sugimoto et al. [74], where the potential of claudin-6 to trigger epithelial morphogenesis in mouse stem cells was reported. Also, claudin-6 regulated other tight-junction and microvillus molecules claudin-7, occludin, Zonula occludens (ZO-1α+), and ezrin/radixin/moesin-binding phosphoprotein50, which strongly proved the role of claudins in epithelial differentiation [74]. This was further supported by other studies, which also showed the expression of claudin-6 is an early marker in embryonic stem cells [75,76]. Differentiation of Human Embryonic Stem Cells to Hepatocyte-Like Cells resulted in a decrease in stem cell markers Oct3/4 and Nanog as expected. Along with stem cell markers, claudin-1 declined eventually, whereas claudin-4 increased and was highest at the end stage of differentiation [77].

A growing body of evidence focuses on cancer stem cells in cancer biology. The drawbacks of cancer treatment failures and drug resistance are proved to originate from cancer stem cells, which are a small subpopulation in tumors. Recently the factors regulating cancer stem cells have gained significant importance and opened new avenues for targeted therapies and thus decrease the chance of recurrence of the disease [78]. Cancer stem cells (CSCs) represent a small group of cells in typically heterogeneous tumors, which possess tumor-initiating and self-renewal properties, giving rise to non-tumorigenic progeny. CSCs are enriched after chemotherapy and lead to therapy failure and thus recurrence of cancer. The role of CSCs in tumor relapse, metastasis, and therapeutic refractoriness is well described [79]. The role of claudins in cancer stem cell (CSC) biology is gaining much attention. The WNT pathway is well known to provide the key signals for achieving this particular phenotype. It is also established that the *Wnt* signal transduction pathway is important in normal and malignant stem cells [80]. Recent articles have highlighted the link between claudin and the *Wnt*/β-catenin signaling pathway and the role of CSC in this cross-talk. Claudin-1 and claudin-2 transcription is regulated by WNT*t* signaling, and they are known to regulate the β-Catenin- T-cell factor/lymphoid enhancer-binding factor (TCF/LEF) signaling pathway to regulate CSC [81,82]. In contrast, other claudins negatively regulate WNTsignaling cascades, such as loss of claudin-3 inducing WNT/β-catenin activation, thus aiding in the promotion of colon cancer [35]. Darido et al. provided evidence for Tcf-4 and Sox-9 regulating the expression of claudin-7 [46]. In addition, studies by Prat et al. discovered a new claudin-low molecular subtype of breast cancer [83]. The key characteristics of this subtype are low expression of tight junction and junction adherens proteins (claudin-3, -4 and -7, and E-cadherin), and enriched in stem cell and EMT features. Patients having high-grade invasive ductal breast carcinoma in this subgroup had a poor prognosis, absence of luminal differentiation markers, enhanced EMT markers, expression of immune response genes, and most closely resembled mammary epithelial stem cells. This suggested that low claudin cells might emerge from more immature stem or progenitor cells and comprise cancer stem cells. Thus, identification of the low claudin subtype in breast cancer has shown the potential of claudins in regulating stem cells. In addition, claudin-3 is known to play an oncogenic role in non-small cell lung cancer (NSCLC). One of the major contributing factors for the role of claudin-3 is regulation of cancer stemness and chemoresistance in non-squamous NSCLC. The depletion of claudin-3 was able to combat the formation of spheres and tumor formation as well as increased sensitivity to cisplatin [84]. Further, claudin-3 inhibition by small-molecule inhibitors including withaferin A, estradiol and fulvestrant, suppressed cancer stemness and combated chemoresistance, giving strong evidence for the role of claudin-3 in inducing stemness. Another claudin playing an important role in stem cell regulation is claudin-18 in lung cancer [85], which has been reported to have a role in the aberrant proliferation of alveolar epithelial type II (AT2) cells, resulting in lung enlargement and parenchymal expansion by restrictions on stem/progenitor cell proliferation. Recently, claudin-2 was shown to be restricted in the stem/progenitor cell compartment of intestinal crypts. It enriches aldehyde dehydrogenase ALDHHigh cancer stem-like cells in heterogeneous colorectal cancer cell populations through the regulation of Yes-associated protein (YAP) activity and miR-222-3p

expression [86]. Overall, these studies give an overview of the potential role of claudins in stem cell biology. The role of claudins in the regulation of stem cells is summarized in Table 2. The claudin mediated enrichment of stem cells provides a new axis-of-evil for a preferential therapeutic target, which has potential clinical consequences.


**Table 2.** Claudins and stemness.

#### **4. Claudins in Chemoresistance**

Most cancer patients initially respond to chemotherapy. Eventually, cancer relapses due to chemoresistance resulting in treatment failure causing death. The mechanisms of chemoresistance in cancers are still largely unknown [87]. Since the role of claudins in the regulation of cancer stem cells is well documented, their correlation with drug resistance and distant metastasis is inevitable and obvious [49,88,89]. In brief, claudin-3 and -6 are correlated with lymph node metastasis in squamous cell lung carcinomas [90,91]. Claudin-4 is highly expressed in primary and metastatic prostate cancer [92] and gastric cancer [93,94]. Claudin-1 and -7 have proved to have an inverse role in colon cancer, wherein claudin-1 elevates the metastasis of colon cancer cells. On the other hand, suppression of claudin-7 leads to liver metastasis [40,42]. Epithelial to mesenchymal transition (EMT) is a piece of vital machinery responsible for invasiveness and initiation of metastasis and chemoresistance of cancer cells. Claudins are known inducers of EMT in cancers. Claudin-1 is known to induce EMT in colon, liver, nasopharyngeal carcinoma, and breast cancers [40,95,96]. At the same time, claudin-7 is reported to be involved in establishing MET in colon cancer [36,42]. Claudin-3 suppresses EMT in lung cancer cells [36]. Overall, the potential role of claudins in EMT, Metastasis and CSC enrichment provides the rationale for exploring them as a key factor in establishing drug resistance. Claudins as chemo-resistance modulators is an emerging field of research. In a recent article, the potential of claudin-6 in enhancing chemoresistance to Adriamycin in triple-negative breast cancer (TNBC) was documented [97]. This effect of claudin-6 was mediated through its regulation over the AF-6/extracellular signal–regulated kinases (AF-6/ERK signaling pathway and up-regulation of cancer stem cells. Claudin-3 is also identified as a molecule to combat cisplatin chemoresistance in non-squamous lung carcinoma [84]. Here, claudin-3 overexpressing lung cancer cells were insensitive to cisplatin treatment compared to control cells. Adding to this, knockdown of claudin-3 or claudin-4 in ovarian cancer cells induced resistance to cisplatin by the regulation of Cu transporter CTR1 [98]. Another study by a Japanese group of researchers reported a high expression of claudin-4 in the ovarian cancer tissues of platinum-resistant patients [99]. In lung cancer, claudin-1 is a key deciding factor for metastasis and a responsible factor for drug resistance towards cisplatin through the up-regulation of Unc-51 Like Autophagy Activating Kinase 1 (ULK1) phosphorylation [100,101]. It is also known to enhance drug resistance in liver cancer cells by modulating autophagy to achieve drug resistance. The

role of claudin-7 in drug resistance [102] has also been reported, wherein decreased drug resistance, increased apoptosis and diminished anti-apoptotic PI3K/Akt pathway was achieved by knocking down claudin-7, proving the potentiality in chemo-resistance [103]. It is well known that EpCAM associates with claudin-7 and is known to be involved in cancer metastasis. Florian et al. [69] have provided evidence for the increased migratory potential of pancreatic cancer cells upon EPCAM and claudin-7 association influencing cell-cell adhesion. Interestingly, the EPCAM and claudin-7 association seems to enhance drug resistance against cisplatin through enhancing MAPK and c-Jun N-terminal kinases (JNK) pathways. Altogether, these studies indicate the important role of claudins contributing to drug resistance in cancer cells. The pictorial representation of the role of claudins as a stem cell regulator and its impact in chemoresistance is shown in Figure 2.

**Figure 2.** The central role of claudins in the regulation of epithelial to mesenchymal transition (EMT),

cancer stem cells, and chemoresistance in various cancers. - inhibition of claudin-7. The arrows indicate the upregulation and higher enrichment of the mentioned signaling molecules, colour is respective of each claudin.

#### **5. Claudins in Prognosis**

Emerging data defining mechanisms through which claudins augment cancer metastasis provides the rationale for exploring claudins as prognostic factors and therapeutic targets in cancer. The importance of claudins is established using cancer cell models, mouse models, and human patient samples. Target molecules for cancer surveillance in high-risk populations are desperately warranted. As a vital emerging modulator in molecular or cellular pathways related to cancers, claudins could be targeted or used as biomarkers for prognosis, diagnosis, and treatments. A number of recent studies have projected a role for claudins as key prognosis factors in cancers. In one of the study Lechpammer et al. [104] demonstrated the potential of claudins as a diagnostic and prognostic factor in renal cell carcinoma. Claudin-1, -3, -4, -7, and -8 were studied in human renal cell carcinomas and oncocytomas. The data from their research showed an inverse correlation between claudin-3 and -4 expression with overall survival in clear cell renal cell carcinomas, and these claudins could be considered for prognosis

in renal cell carcinomas. Claudin-7 and 8 can be implied as useful markers in the identification of renal cell carcinomas from oncocytomas [105].

Claudin-6 was reported as a prognosis factor in NSCLC patients. In this report, the patients with low claudin-6 had a lower survival rate than the patients with high claudin-6. [91] reported low claudin-6 as an independent indicator of prognosis in NSCLC patients. In this study, they documented low claudin-6 in 61 of 123 NSCLC tissue samples, and patients with low claudin-6 expression correlated with lower survival rates than those with high claudin-6 expression. The influence of claudin-3, claudin-7, and claudin-18 in gastric cancer patients were also studied [106]. Claudin-3 and claudin-7 were expressed in 25.4% and 29.9% of the gastric cancer tissues, respectively. However, 51.5% of gastric cancer tissues exhibited reduced expression of claudin-18. Claudin-7 expression correlated with shorter overall survival in gastric cancer patients, while the overall survival was increased in patients with claudin-18 expression. Recently, claudin-3 and -7 are also considered as novel prognostic factors in triple-negative breast cancer (TNBC) through its aberrant immunohistochemical expressions [107]. Claudin-3 cytoplasmic expression is an indicator of poor survival in triple-negative breast cancer. In addition, epigenetic modifications of claudins are reported to be a promising prognosis marker of various cancers. Zhenzhen et al. [106] recently demonstrated that the methylation of claudin-3 is a prognostic factor in gastric adenocarcinoma.

Further, the serum levels of claudin-7 among patients with colorectal cancer (CRC) was significantly reduced and correlated with high tumor stage and high carcinoembryonic antigen levels [108]. Claudin-7 was found to be downregulated in CRC, as reported by Bhat et al. [42], and associated with diminished EMT and tumor progression. These studies give a strong rationale to consider claudin-7 as a biomarker for predicting the development, proliferation, and prognosis of CRC. A claudin-low molecular subtype of breast cancer has been described with a concomitant upregulation of several EMT markers and an enrichment in stem cell features [109]. In an interesting article by Danzinger et al., the importance of claudin-3 in triple-negative breast cancer (TNBC) was documented. It was reported that claudin-3 expression was correlated with a Breast cancer type 1 (BRCA1) mutation [107]. This could help in guiding the decision for BRCA testing for triple-negative breast cancer (TNBC). Also, the expression of claudin-11 has been suggested as a biomarker for advanced-stage cutaneous squamous carcinoma, and reflects the distinct stages of tumor development and differentiation [110]. The clinical significance of claudin-11 was addressed in Laryngeal Squamous Cell Carcinoma (LSCC) by Nissinen et al. [110]. In this study, elevated promoter methylation of claudin-11 in tumor tissues was observed. Patients with lymph node metastasis with an advanced clinical stage showed more methylation in the claudin-11 promoter, which associated with poor overall survival of LSCC patients. In TNBCs, claudin-1, -3, -4, and -7 higher expression rates are more frequent than in other subtypes [111]. Claudin-4 high/claudin-1 low, claudin-4 high/claudin-7 low, and claudin-4 high/claudin-1 low/claudin-7 low types were also significantly correlated with lymph node metastasis, and showed worse survival. Apart from this, a recent article from Upadhaya et al. documented the therapeutic potential of claudin-1 in oral epithelial dysplasia and oral squamous cell carcinoma [112]. Overall the differential expression pattern of claudins may reflect the distinct stages of tumor development and differentiation and have been implied as prognostic factors for early determination of the tumor state.

#### **6. Claudins as Therapeutic Agents**

So far, over 100 monoclonal antibody (mAb) products are in clinical trials [113]. In an oncology setting, these monoclonal antibodies can mediate antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) against cancer cells [114]. There is a long-lasting history of antibody-mediated targeting of claudin-1 against hepatitis C virus (HCV) infections, and wherein many researchers have provided proof for the importance of claudins in HCV infections as viral entry point [115,116]. A study by Fofana et al. [117] designed monoclonal antibodies against claudin-1 to combat HCV entry. It was promising to see the antibodies raised against claudin-1 was able to block HCV entry. A recent study by Colpitts et al. has documented the humanization of a claudin-1-specific

monoclonal antibody and was investigated in a large panel of primary human hepatocytes, and was found to be very promising for clinical HCV prevention and cure [118]. These studies hold significance because these antibodies could prevent HCV infection after liver transplantation, and virus spread in chronically infected patients. These antibodies are now being tested in cancer models. Claudins, as a potential target in antibody-based therapies for carcinomas, was investigated by Offner et al. [119]. In this study, the antibodies were raised against the extracellular domains of claudin-1, -3, and -4. Recently Romani et al. engineered a fully human anti-claudin-3 IgG1 antibody (IgGH6) [120], which is specific to claudin-3 and no cross-reactivity with other claudins was observed. Recent work by Cherradi et al. [121] investigated the importance of claudin-1 in different colorectal cancer (CRC) molecular subtypes. There is a differential expression pattern of claudin-1 based on the subtype. A murine monoclonal antibody against the extracellular part of human claudin-1 (6 F6 mAb) was designed and generated, which was specifically able to pick claudin-1 positive CRC cell lines, and no other cross-reactivity was observed. Furthermore, 6 F6 mAb was able to combat colony formation, xenograft growth and metastasis of claudin-1 positive CRC cells suggesting its utility as a therapeutic. Fujiwara et al. recently targeted claudin-4 in CRC using an anti-claudin-4 extracellular domain antibody [122]. The efficacy of the anti-claudin-4 antibody is promising and observed to enhance the anti-tumorigenic potential of 5-fluorouracil (FU) and anti-EGFR antibodies. These works demonstrate the proof of concept for exploiting claudins as targets for monoclonal antibodies in therapies.

Some of the monoclonal antibodies against claudins, including anti-claudin-18.2 (IMAB362-claudin-18.2) and the anti-claudin-6 (IMAB027-claudin-6), have also found their way into clinical trials [123]. Claudin-18.2 is expressed on the outer cell membrane of gastric cancer cells and binds to monoclonal antibodies. The IMAB362 was proven to be clinically safe as the patients were devoid of any side effects. Also, IMAB027 is in an ongoing clinical trial for recurrent advanced ovarian cancer (NCT02054351), and patients have not demonstrated any adverse effects [123]. Clinical trials for claudiximab (claudin-18 targeting) in advanced gastroesophageal cancer patients are also underway [124]. Recently, claudiximab was reported to be a first-in-class chimeric monoclonal antibody for the treatment of gastric cancer targeting claudin-18, which is an important factor in gastric cancer metastasis. This is just the beginning of an exciting journey and more research is warranted to revolutionize claudins targeted monoclonal antibodies in cancer therapy.

Another avenue to exploit Claudins as a therapeutics is their ability to behave as receptors for microbes. *Clostridium perfringens* enterotoxin (CPE) has the potential to bind with claudin receptors. CPE binds to the C-terminal CPE domain at both the first and second extracellular loops (ECL-1 and ECL-2) of claudins [125]. The affinity of CPE to claudins causes a pore leading to calcium influx responsible for host cell death. The claudin–CPE interaction is gaining significance in receptor decoy therapeutics for potential applications in gastrointestinal disease, cancer therapy/diagnoses, and drug delivery [125]. Claudin-3 and claudin-4 have been widely demonstrated to function as CPE receptors [126,127]. The binding ability of CPE to claudins, especially claudin-3 and claudin-4, has raised a great opportunity to target cancers with dysregulated claudin-3 and -4 cancers, especially breast, ovarian, and pancreatic cancers. The binding of CPE to claudin-3 and -4 was documented to induce dose-dependent cytolysis in breast cancer cells expressing claudin-3 and -4 [128]. Recent studies have exploited the CPE mediated targeting of claudin-3 and 4 cancers to target therapy-resistant ovarian cancer, pancreatic, and breast cancer xenografts possessing increased expressions of claudin-3 and -4. In one of the studies, the possibility of CPE binding claudin-3 as a visualization tool for identifying of micrometastatic chemotherapy-resistant ovarian cancer has been demonstrated [129]. The applicability of CPE, claudin-3, and -4 interactions is exploited in gene therapy against colon cancer. Recombinant (recCPE) and optimized CPE expressing vector (optCPE) were demonstrated to have a cytotoxic effect in claudin overexpressing colon cancer cells [130,131]. Further, the recent identification of the crystal structure of claudin-9 revealed that human claudin-9 has high-affinity for the CPE receptor and treatment with CPE caused cell death in human claudin-9 expressing cells [132]. In continuation of these studies, an interesting approach of nanoparticle-based targeting of cancer

cells was documented by researchers, wherein the C-terminus of the CPE was conjugated to gold nanoparticles (AuNPs). This combination binds to claudin expressing tumor cells and kills the cells using gold nanoparticle-mediated laser perforation (GNOME-LP) technique [133,134]. Thus, the clinical relevance and functional importance of claudins in diverse cancers make them potential therapeutic targets.

#### **7. Claudins as a Visualization Tool**

The use of monoclonal antibodies against claudins have also been utilized in imaging modalities. Recently, claudin-4 was studied as an imaging tool for x-ray computed tomography (CT) in the prognosis of pancreatic ductal adenocarcinoma (PDAC) [135]. Claudin-4 is a known biomarker in PDAC detection. In this study, researchers reported a novel radiolabeled anti-claudin-4 monoclonal antibody in detecting PDAC using single-photon emission computed tomography (SPECT) imaging. The results showed promising uptake of anti-claudin-4 monoclonal antibody by PDAC tumors and were helpful in early detection and characterization of PDAC malignancy. Also, the researcher later targeted the extracellular domain of claudin-4 (4D3) with monoclonal antibodies (4D3) in combating bladder and lung cancer [136].

Colonoscopic aided screening and polyp and tumor removal have led to the reduced incidence and mortality of colorectal cancer (CRC). However, the lack of specificity is a major pitfall in these approaches and makes them less effective. It is especially difficult to detect the regions of flat dysplasia or serrated polyps, which also possess malignant potential. Thus, a targeted approach for advanced endoscopic techniques is a cornerstone requirement. A promising approach was recently demonstrated for the real-time endoscopic imaging of colonic adenomas [137]. In this study, the researchers exploited claudin-1 as a potential target in endoscopic imaging of colonic adenomas. As claudin-1 is highly expressed in the early development of CRC, endoscopic imaging might be useful for detecting either polypoid or flat precancerous lesions that are difficult to visualize [138]. Peptide (peptide sequence—RTSPSSR), specific to claudin-1, was developed against the extracellular loop of claudin-1. This peptide showed greater intensity for human adenomas, hyperplastic polyps and sessile serrated adenomas thus proposing the possibility of using claudin-1 peptide aided endoscopic imaging for the future clinical translation to detect precancerous lesions. Recently another study by our group demonstrated the significance of claudin-1 as a useful target for near-infrared antibody-based imaging for visualization of colorectal tumors [138]. When animals injected with colon cancer cells subcutaneously were imaged using claudin-1 antibody conjugated LI-COR IR800DyeCW through a LI-COR Pearl Trilogy Fluorescence Imaging System, the system was able to target tumors specifically. These studies pave the way for using claudins as a tool for fluorescence-guided surgery, which will help in more specific targeting of the tumors in a stage-specific manner. A comprehensive representation encompassing the role of claudins and the monoclonal antibodies against claudins as therapeutic and detection tools is given in Figure 3, and the role of claudins as a therapeutic, prognostic, and detection agents is tabulated in Table 3.

**Figure 3.** Claudins as an employable platform for prognostic, diagnostic, and therapeutic targets. The upward arrow indicated upregulation and downwards arrow indicated downregulation of the mentioned signaling events.

**Table 3.** Claudins as prognostic, therapeutic and detection agents.




#### **8. Future Perspectives**

The quest for prognostic, diagnostic, and therapeutic markers for many cancers is of high importance. More reliable and earlier detection markers have implications for diagnostic and therapeutic targeting. As the role of claudins in the regulation and enrichment of cancer stem cells and chemo-resistance becomes obvious, targeting claudins for diminishing cancer stem cells, which are cancer-propagating subsets of malignant cells, would be very useful. The potential of the claudin–cancer stem cell axis provides great potential for combating invasive, metastatic, and drug resistance phenotypes of various cancers. Future studies focusing on the role of claudins in cancer stem cells will be warranted to specifically target these populations to curb down residual tumor cells left after standard therapies.

Claudins are gaining their importance as detection and therapeutic agents. Future engineering of more monoclonal antibodies against claudins will have potential applications in targeted therapy, and claudin assisted endoscopy, imaging of various tumors. Also, the antibody-based detections will provide ample opportunity for the early diagnosis of any inflammatory diseases before they reach cancer status. The ongoing clinical trials for monoclonal antibodies against claudins might lead to claudin directed immunotherapies. Recently, small molecules inhibitors have been gaining more attention in cancer biology, as they aid in targeted therapy. No known small molecule inhibitors are currently being researched for claudins. Thus, in the future, screening for more potent inhibitors against claudins is warranted. Overall, to strengthen the therapeutic window of claudins, a more translational view of claudins by researchers is warranted.

**Author Contributions:** Conceptualization, S.G., P.D., and A.B.S.; methodology, S.G. and P.D.; resources, S.G. and P.D.; writing—original draft preparation, S.G.; writing—review and editing, S.G., A.B.S., and P.D.; supervision, P.D.; project administration, P.D. and A.B.S.; funding acquisition, P.D. and A.B.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by grant numbers BX002086 (VA merit) and CA216746 (NIH/NCI). Further funding came from a pilot project award from the Fred and Pamela Buffet Cancer Center, which was funded by a National Cancer Institute Cancer Center Support Grant, under award number P30 CA036727 to P.D., BX002761 (VA merit) to A.B.S., and Nebraska research initiative (NRI to P.D and A.B.S).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Claudin-1, A Double-Edged Sword in Cancer**

**Ajaz A. Bhat 1, Najeeb Syed 1, Lubna Therachiyil 2,3, Sabah Nisar 1, Sheema Hashem 1, Muzafar A. Macha 4,5, Santosh K. Yadav 1, Roopesh Krishnankutty 2, Shanmugakonar Muralitharan 6, Hamda Al-Naemi 6, Puneet Bagga 7, Ravinder Reddy 7, Punita Dhawan 5, Anthony Akobeng 8, Shahab Uddin 2, Michael P. Frenneaux 9, Wael El-Rifai <sup>10</sup> and Mohammad Haris 1,6,\***


Received: 25 November 2019; Accepted: 13 January 2020; Published: 15 January 2020

**Abstract:** Claudins, a group of membrane proteins involved in the formation of tight junctions, are mainly found in endothelial or epithelial cells. These proteins have attracted much attention in recent years and have been implicated and studied in a multitude of diseases. Claudins not only regulate paracellular transepithelial/transendothelial transport but are also critical for cell growth and differentiation. Not only tissue-specific but the differential expression in malignant tumors is also the focus of claudin-related research. In addition to up- or down-regulation, claudin proteins also undergo delocalization, which plays a vital role in tumor invasion and aggressiveness. Claudin (CLDN)-1 is the most-studied claudin in cancers and to date, its role as either a tumor promoter or suppressor (or both) is not established. In some cancers, lower expression of CLDN-1 is shown to be associated with cancer progression and invasion, while in others, loss of CLDN-1 improves the patient survival. Another topic of discussion regarding the significance of CLDN-1 is its localization (nuclear or cytoplasmic vs perijunctional) in diseased states. This article reviews the evidence regarding CLDN-1 in cancers either as a tumor promoter or suppressor from the literature and we also review the literature regarding the pattern of CLDN-1 distribution in different cancers, focusing on whether this localization is associated with tumor aggressiveness. Furthermore, we utilized expression data from The Cancer Genome Atlas (TCGA) to investigate the association between CLDN-1 expression and overall survival (OS) in different cancer types. We also used TCGA data to compare CLDN-1 expression in normal and tumor tissues. Additionally, a pathway interaction analysis was performed to investigate the interaction of CLDN-1 with other proteins and as a future therapeutic target.

**Keywords:** claudin 1; tight junctions; tumor; metastasis; epithelial to mesenchymal transition

#### **1. Introduction**

Claudins and occludin, a group of cell junctional proteins, serve as the backbone of the tight junctions. Claudin family members perform dual roles; some have barrier activities, while others mediate the permeability of small molecules and ions. In addition to the localization pattern, the differential expression of claudins between normal and tumor tissue has drawn attention to these proteins as potential prime candidates for future cancer therapy. Another hot topic of discussion is the tumor-promoter or tumor-suppressor role of claudins. This opens a wide area of research in elucidating how the tissue-specific expression of claudins and their interaction with other molecules in the cell may result in these two opposing effects. Both defective tight junctions and the absence of tight junctions have shown to be associated with the development and progression of certain cancers. In this article, we include a brief introduction of tight junctions, the structure of claudins, and their role in various cancers. We also perform the bioinformatics analysis on TCGA data to supplement the literature review.

#### *1.1. Tight Junctions*

Adjacent epithelial cells are sealed into an epithelial barrier by the most apical intercellular junctions called tight junctions. Tight junctions, as a network of continuous strands, separate the plasma membrane into apical and basolateral domains [1,2]. Tight junctions between adjacent cells associate to form paired strands imparting mechanical strength to the cells [3,4] and serve as barriers to control the movement of small molecules and ions across the paracellular space [5–7]. Apart from their mechanical strength, maintaining polarity and paracellular movement, tight junction proteins are able to recruit signaling proteins for various cellular processes [4]. Alterations in the structure and function of tight junctions result in a multitude of diseases, especially adenocarcinoma of various organs [8–10]. The failure of tight junctiontight junctions or tight junction proteins is one of the many key factors that contribute to the progression of cancer, but this is not a universal phenomenon as there can be many other direct or indirect factors that contribute to the development of cancer. The second reason that loss of tight junctions or tight junction proteins is not a universal phenomenon to the development of cancers is that besides the epithelial cancers, there are also non-epithelial tumors such as small subset of laryngeal neoplasms [11], angiomas, lipomas and neuromas [12] which do not display failure of tight junctions but other contributing factors come into play.

#### *1.2. Claudins*

Claudins are integral to the structure and function of tight junctions with four membrane-spanning regions, which include two extracellular loops, N- and C-terminal cytoplasmic domains. The extracellular loops are highly conserved, and the C-terminal domain is important for localizing the claudins to tight junctions (Figure 1). Being a part of a multigene family, there are about 27 members of claudins that are unique in their tissue-specific expression and their molecular weight ranges from 20–34 kDa [3]. Claudins play an important role in regulating transepithelial permeability by regulating the epithelium's paracellular permeability to small molecules and ions [5,7,13]. Post-translational modifications such as phosphorylation alter the paracellular functions of claudins, which in turn modulate diverse signal mechanisms [14–17].

**Figure 1.** A schematic presentation of tight junction complex involving claudins and other major components. Claudins contain four transmembrane domains (TMD-1, TMD-2, TMD-3, and TMD-4) and two extracellular (ECL) loops. The PDZ-binding domain of the –COOH terminal of claudin undergoes post-transcriptional modification and has been implicated in signal transduction.

#### *1.3. Claudins and Cancer*

One of the important factors in cellular transformation and tumorigenesis is the loss of cell-to-cell adhesion [1]. Accordingly, the claudin family of proteins is significantly involved in the progression and growth of several cancers [9,10,18]. Tumor progression is characterized by migration, invasion, and metastasis of cancer cells. Claudins are believed to play a significant role in these processes as their loss contributes to the loss of cell junctions in a tissue-dependent manner [18,19]. Claudins have been also reported to play a vital role in the epithelial–mesenchymal transition (EMT) (Figure 2), a process that favors the spread of carcinomas, generation of cancer stem cells (CSCs) or tumor-initiating cells (TICs), and chemo-resistance [20–23]. The loss of claudins in epithelial cells results in disrupted tight junction function responsible for impaired cell polarity and epithelial integrity [6,7]. Several studies have reported the mislocalization and altered expression of claudins in various cancers [19,24]. The CLDN-1 and CLDN-7 members of the claudin family are primarily found to be downregulated in several invasive cancers including breast, esophageal, and prostate cancers [9,19,25–27]. However, in contrast, overexpression of CLDN-1 has been observed in colon, nasopharyngeal, ovarian and oral squamous cell cancers [9,10], while CLDN-3 and -4 are highly overexpressed in ovarian cancer and upregulated in breast, gastric, pancreatic, prostate and uterine cancers [28–30]. Human carcinomas such as those of the breast, liver, ovary, prostate, colon, liver and stomach are found to exhibit altered expression of claudins [19]. The expression and localization patterns of some of the claudins serve as an important prognostic predictor in many cancers [30,31]. The consensus of whether claudin expression increases or decreases during tumorigenesis is still a debatable topic and open to more research.

**Figure 2.** Schematic model of role and regulation of CLDN-1 in a normal or diseased state. In normal physiological conditions, CLDN-1 expression/ integrity is regulated by transcription factors, growth factors and cytokines, which in turn maintain the normal gate function and barrier function of tight junctions. Dysregulation of CLDN-1 expression can result in the compromise of membrane barrier functions and gate functions, which subsequently can lead to the upregulation of the expression of pro-inflammatory markers such as IFN-γ and TNF-α. In cancer, the loss of CLDN-1 facilitates the malignant transformation of cancer cells and epithelial-mesenchymal transition (EMT).

#### **2. Claudin-1 and Cancer; Tumor Promoter or Suppressor**

CLDN-1 is a membrane protein that, along with occludin and other claudins form the backbone of the tight junctions and is essential for epithelial barrier functions [32]. It was the first member of the claudin family to be identified with a molecular weight of 22 kDa and is strongly expressed in the intestine, spleen, brain, liver, kidney, and testis [19,33]. Studies have shown the direct involvement of CLDN-1 in the development and progression of several cancers, such as colon cancers [34], oral squamous cell carcinomas [35], breast cancers [36], melanomas [37,38], and in many other cancers as discussed in this review. In some cancers, CLDN-1 has the opposite role where the decreased expression of CLDN-1 is associated with cancer progression, invasion and development of the metastatic phenotype [37,39]. The expression of CLDN-1 in different types of cancer and cancer subtypes is summarized in Table 1. Based on the literature, CLDN-1 is one of the most deregulated claudins in human cancer and can function as a tumor promoter or suppressor depending on the type of cancer (Figure 3) (Table 2). The role of CLDN-1 as a tumor promoter is mostly through its effect on the invasion or motility of cancer cells. Considering the importance of claudins in cancer, targeting claudin expression appears to have promise in the treatment of cancer. The specific role of CLDN-1 in various cancers is discussed in the following sections.


#### **Table 1.** Expression of CLDN-1 in different types of cancer.

"-" no subtypes.



**Figure 3.** CLDN-1 expression in different subtypes of breast cancer as characterized by the presence or absence of estrogen receptor (ER). Luminal A, and Luminal B subtypes of human invasive breast cancer (ER-positive) exhibit low levels of CLDN-1, which suggests the suppressor role of CLDN-1 in these tumors. However, aggressive forms (ER-negative) exhibit overall high levels of CLDN-1 expression, which signifies CLDN-1 role as a tumor promoter.

#### *2.1. Claudin-1 and Breast Cancer*

Breast cancer is the second major cause of death in women, and its heterogeneous molecular nature is a significant obstacle in treatment planning [65]. It has several subtypes, such as human epidermal growth factor receptor 2 (HER2), triple-negative or basal-like, Luminal A and Luminal B type depending on the presence or absence of several hormone receptors like HER2, estrogen, and progesterone [66–68]. Recently, another subtype of breast cancer known as the claudin-low subtype has been reported [67]. Each subtype of breast cancer exhibits unique prognostic features and different molecular markers [69].

The *CLDN-1* gene has been found to be upregulated during the early involution of the mammary gland [70]. The differential expression of CLDN-1 observed in different cancers outlines the complexity of the potential role that it plays in the cancer process. The CLDN-1 expression level in breast cancer differs depending on the cancer subtypes [71]. Studies have shown a correlation between increased malignancy, invasiveness and recurrence of breast cancer with total or partial loss of CLDN-1 expression [36,70]. In most of the invasive human breast cancers such as ER+ luminal A and luminal B, CLDN-1 expression is found to be downregulated, while an increased expression and cytoplasmic delocalization of CLDN-1 has been observed in some of the aggressive ER- basal-like breast cancer (BLBC) subtypes [40,72,73]. CLDN-1 is also found to be downregulated in HER2 enriched and claudin low breast cancer subtypes [41]. CLDN-1 acts as a tumor suppressor in ER+ and as a tumor promoter in ER- cancer subtypes [25]. In hereditary and sporadic breast cancer, CLDN-1 is found to be involved in tumorigenesis by suppressing the proliferation of mammary epithelial cells [74]. Further, CLDN-1 overexpression in MDA-MB 361 breast cancer cells resulted in increased apoptosis [75,76]. While one study reported that the activation of CLDN-1 was repressed by the binding of E-cadherin to CLDN-1 promoter [77], knockdown of CLDN-1 has been found to be associated with decreased cell migration and induction of EMT in breast cancer cells [76]. Another study showed a unique pattern of expression for CLDN-1 in ER-ve and ER+ve tumors. The authors showed that the protein expressions of CLDN-1 were significantly higher in the basal-like subtype of breast cancers (ER-ve, Her-2-ve, EGFR+ve, CK5/6+ve, a subtype largely linked to poor outcome [40]. CLDN-1 expression has also been observed in a small percentage of invasive human breast cancers that exhibit different pathological lesions leading to complexity in CLDN-1 expression [78]. CLDN-1 also possesses tumor-promoting effects by increasing cell migration and by exhibiting anti-apoptotic effects in some breast cancer cell lines like MCF-7 [76,79].

Several proteins interact with CLDN-1 to fuel the progression of breast cancer, including the following: Ephrin B1, ESCRT, CD9 and EpCAM [80–83]. CLDN-1 mediates the tyrosine phosphorylation of Ephrin B1, a transmembrane protein, in a receptor independent manner which provides the evidence that ephrin-B1 inhibits the formation of the tight cell–cell adhesion in a wide variety of epithelial and cancer cells regardless of the existence of cognate Eph receptors [80]. Endosomal sorting complexes required for transport (ESCRT) machinery are a set of proteins present in the cytosol that are involved in the maintenance of cell polarity and the regulation of membrane-bound proteins [81]. When the function of ESCRT is inhibited, CLDN-1 accumulates in the cytoplasm causing the tight junctions to disassemble and lose cell polarity [25]. The loss of ESCRT function is also linked with increased proliferation and less stable tissue structure in the cancer cells. CLDN-1 is also found to interact with CD9, a transmembrane protein that regulates cell migration, proliferation, differentiation and fusion [82]. CD9 prevents the association between CLDN-1 and tight junctions that could cause the progression of the tumor. The subcellular co-localization of CLDN-1 and CD9 supports their interaction, and this was confirmed in many cell clines including different human breast cancer cell lines [82]. EpCAM (also known as epithelial cell adhesion molecule), another surface transmembrane glycoprotein known to be expressed in some invasive carcinomas is involved in cell proliferation and metastasis and has been shown to protect CLDN-1 from degradation. [83]. This could be a cause for the cytoplasmic accumulation of CLDN-1 in some breast cancer cell lines [76,83]. Several transcript variants for CLDN-1 were found in human invasive breast cancer as a result of splicing and mis splicing events suggesting that through alternative splicing CLDN-1 is downregulated in invasive type of breast cancers [72].

#### *2.2. Claudin-1 and Thyroid Cancer*

Thyroid cancer is the most commonly occurring endocrine malignancy [84,85]. A study by Nemeth et al. performed independent microarray expression analyses of two types of thyroid carcinomas, namely papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC) [42]. The study showed that high expression of CLDN-1 is specific for the regional lymph node metastasis associated with PTC [42] and found increased expression of *CLDN-1* gene in PTC [86,87]. Sobel et al. reported high levels of CLDN-1 expression in serous papillary endometrial carcinoma [88]. Another study about the role of CLDN-1 in follicular-cell derived thyroid carcinoma cell lines (FTC-133 and FTC-238) found higher expression of CLDN-1 in the nuclei of FTC-238 cells as compared to the FTC-133 cells [43]. The same study demonstrated the increased pathogenic character of FTC-133 cells by RASV12 transfection was associated with high expression of CLDN-1 and enhanced cell proliferation and migration [43]. Conversely, the downregulation of CLDN-1 by siRNA caused decreased cell invasion and migration accompanied by decreased phospho-PKC expression in the FTC-238 cells, suggesting that the aggressiveness of follicular thyroid carcinoma associated with high CLDN-1 expression can be influenced by PKC activity [43]. Another study described the reduced expression of CLDN-1 in follicular carcinomas vs adenomas, specifically in the poorly-differentiated and undifferentiated types of human thyroid carcinomas [89]. The expression of CLDN-1 was significantly different between malignant and benign thyroid neoplasms, and between follicular and papillary carcinomas [90]. Similarly, papillary carcinomas showed significantly higher positive CLDN-1 expression. While negative CLDN-1 expression was observed in the tissue samples of normal thyroid and solitary-follicular-patterned-nodules [91].

#### *2.3. Claudin-1 and Colorectal Cancer*

Colorectal cancer (CRC) is the fourth leading cause of cancer-related deaths and the third most frequently diagnosed malignancy worldwide [92]. Increased expression of CLDN-1 is associated with the progression and metastasis of colon carcinoma [34,93]. In mouse xenograft studies, tumor growth and metastasis is regulated by genetic modulation of CLDN-1 [94]. The nucleus and cytoplasm of colon carcinoma cells and metastatic lesions showed intensified CLDN-1 expression [34]. Many studies suggest that the genes encoding tight junction proteins (TGPs) in CRC are differentially expressed and involved in the process of invasion and cellular transformation [95]. Several studies reported up-regulation of CLDN-1 in CRC [34,44,96,97]. A similar study showed that CLDN-1 overexpression induced a highly invasive and metastatic potential in CRC cells [34]. Noncancerous cells with normal CLDN-1 expression were found to form a monolayer, whereas cells that overexpressed CLDN-1 grew as aggregates. CLDN-1 regulates cellular morphology and behavior in the colonic epithelium [34,44]. The possible involvement of CLDN-1 in the tumorigenesis of ulcerative colitis (UC)-associated CRC has also been demonstrated [44]. Another study demonstrated that the prognostic factor for CRC is the independent expression of CLDN-1 [98]. Delocalization of CLDN-1 from the membrane to cytoplasm and nuclei of cancer cells supports cancer growth and malignancy [34]. In colon cancer cells, CLDN-1 decreases the expression of E-cadherin by upregulating ZEB-1 repressor resulting in invasion and reduction of anoikis [59].

The level of CLDN-1 mRNA was found to be higher in the distal site of the colon as compared to the proximal site and demonstrated significant effects on xenografted tumors growth in athymic mice by changes in the expression of CLDN-1, showing its role in CRC tumorigenesis [34]. Both mRNA and protein levels of CLDN-1 were found to be upregulated in sporadic human CRC compared to the normal mucosa [34]. Dhawan and colleagues reported that T84 cell transfection with CLDN -1 resulted in aggregation and multilayer formation in transfected T84 cells as compared to the T84 parent cells. The interactions between claudin family members are both homophilic and heterophilic and are considered to play a significant role in the progression of CRC and several other cancers [99]. The progression of colon cancer has been linked with the dysregulation of the CLDN-1 expression causing disorganization of the tight junction fibrils leading to increased paracellular permeability [100]. Increased potential for invasion and metastasis has been demonstrated in xenografts that express CLDN-1 [101]. CLDN-1 positively correlates with CRC cell proliferation and influences the growth and evolution of the tumor. Its expression was also found to be associated with accelerated serrated lesions of CRC and was related to anoikis resistance and cellular dis-cohesion [101]. Moreover, serrated polyps with over-expressed CLDN-1 were found to have a higher potential for the development and progression into higher-grade lesions [101].

Activation of the Wnt signaling pathway is strongly implicated in the development of colorectal cancer [102]. Wnt signaling is activated by the loss of the adenomatous polyposis coli (APC) protein or by the activation of β-catenin mutations [103,104]. CLDN-1, one of the target genes in the Wnt signaling pathway, has two β-catenin binding sites (TCF/LEF) in its promoter region for the activation of the transcription process [105,106]. A study showed that the expression of CLDN-1 was found to be elevated in the intestinal adenomas of the APC in mice as compared to normal tissue [94]. Additionally, high expressions of CLDN-1 are seen in the dysplastic areas of the colon in patients with chronic inflammatory disease [94]. CLDN-1 was overexpressed in metastatic colorectal cancer (mCRC) samples as compared to normal mucosa with differential expression in other CRC subtypes. Consensus molecular subtype CMS2, transit-amplifying and C5 subtypes of the mCRC exhibited higher expression of CLDN-1 [107].

#### *2.4. Claudin-1 and Gastric Cancer*

CLDN-1 is highly expressed in gastric cancers [108,109]. High expression of CLDN-1 was reported in intestinal type gastric cancer that correlated with lymph node metastasis, advanced TNM (classification of malignant tumors) stage, recruitment, and activation of MMP-2 and MMP-9, which are all responsible for enhanced cell invasion and metastasis [60,109]. The invasion of gastric adenocarcinoma cells is associated with the levels of CLDN-1 expression as CLDN-1 is found to be upregulated in gastric carcinoma and participates in the metastatic behavior of these cancer cells [45].

One study demonstrated that the localization and correlation of CLDN-1 expression are linked with anoikis resistance in gastric cancer through mediating membrane β-catenin expression and by inducing cell aggregation and inhibiting apoptosis cascade [110]. The authors also observed that the levels of CLDN-1 expression in gastric cancer tissues decreased from well to moderate to poorly differentiated tumors, suggesting that reduced CLDN-1 expression is an adverse prognostic factor predicting a lower survival rate [110]. However, another study showed that in comparison to CLDN-4, the expression of CLDN-1 was higher in well-to-moderately differentiated gastric adenocarcinomas [111].

#### *2.5. Claudin-1 and Hypopharyngeal Squamous Cell Carcinoma*

Head and neck squamous cell carcinoma (HNSCC) is the sixth most frequent tumor worldwide [112]. Tissue microarray and immunohistochemistry assays of surgical samples suggested that CLDN-1 expression is increased in squamous cell cancer [46,113]. It has been shown previously that CLDN-1 induces the generation of tumor lymphatic vessels and increases the lymph node metastasis [47]. Additionally, a study demonstrated that CLDN-1 expression in squamous cancers differs in an organ-specific manner [113]. CLDN-1 was found to be upregulated in hypopharyngeal squamous cell carcinoma (HSCC). This study showed a positive association of CLDN-1 expression with the degree of tumor differentiation and lymph node metastasis [61].

#### *2.6. Claudin-1 and Hepatocellular Carcinoma*

Hepatocellular carcinoma (HCC) is the third major cause of death due to cancer and the fifth most common cancer malignancy worldwide [114]. Studies have reported the involvement of CLDN-1in the tumorigenesis and metastasis of HCC [48,63]. HCC cell line experiments demonstrated the role of CLDN-1 in the process of cancer cell invasion [115]. Primary HCC samples were found to be positive for CLDN-1, suggesting its significant role in the formation of metastasis and hepato-carcinogenesis [115]. Another study demonstrated the involvement of CLDN-1 in the epithelial to mesenchymal transition in HCC and hepato-carcinogenesis [62]. One study established that overexpression of CLDN-1 induces MMP-2 in SNU-354, -423 and -449 HCC cells resulting in increased invasion and migration of the cancer cells compared to the normal liver cells and other CLDN-1 expressing HCC cells such as SNU-398 and SNU-475 [63]. However in another study reduced expression of CLDN-1 was reported to be a marker for a poor prognosis in HCC [49], and a further study showed that reduced expression of CLDN-1 reinforced the invasive and cancer stem cell (CSC) like properties of HCC cell lines (Huh7 and Hep3B) in vitro, while the forced expression of CLDN-1 diminished the CSC-like properties of HCC cells [116].

#### *2.7. Claudin-1 and Lung Adenocarcinoma*

Lung cancer is one of the leading causes of death worldwide. Several studies have shown that CLDN-1 has a significant role in the pathogenesis of lung cancer [117–119]. In lung cancer, CLDN-1 acts as a cancer invasion/metastasis suppressor [57]. CLDN-1 was found to be associated with increased expressions of cancer metastasis suppressors such as connective tissue growth factor (CTGF), thrombospondin 1 (THBS1), deleted in liver cancer 1 (DLC1), occludin (OCLN), zona occludens 1 (ZO-1) and reduced expressions of cancer metastasis enhancers such as secreted phosphoprotein 1 (SPP1), cut-like homeobox 1 (CUTL1), transforming growth factor-alpha (TGF-α), solute carrier family 2 (facilitated glucose transporter) member 3 (SLC2A3) and placental growth factor (PGF) in lung adenocarcinoma [57]. For patients with lung adenocarcinoma, CLDN-1 is a potential drug treatment target and a useful predictor of prognosis. Studies showed that the invasive ability of HOP62 lung adenocarcinoma cells is increased by knockdown of endogenous expression of CLDN-1 [57]. Immunohistochemistry and RT-PCR analysis showed that CLDN-1 is either reduced or undetected in adenocarcinomas [57]. The authors demonstrated that CLDN-1 overexpression inhibited adenocarcinoma cell dissociation in wound-healing time-lapse images [57].

Inflammatory mediators, such as TNFα, plays a significant role in the process of tumorigenesis [120]. Studies have shown that CLDN-1 is involved in the mediation of inflammatory responses initiated by TNFα in different cancers [64,121,122]. An experiment in human lung cancer cell lines, observed that TNFα induced the expression of CLDN-1, and knockdown of CLDN-1 blocked 75% of TNFα-induced gene expression. In CL1-5 lung cancer cells, cell migration activity was inhibited by over-expression of CLDN-1 and restored by CLDN-1 knockdown in addition to cell invasion ability. The above findings signify a signal mediator role of CLDN-1 in TNFα induced gene expression and cell migration [57]. One

study demonstrated that CLDN-1 expression correlated with Ras and epidermal-growth-factor-receptor (EGFR) expression suggesting the involvement of the latter two signaling pathways in the regulation of CLDN-1 in lung adenocarcinoma [119]. The results of this study demonstrated an association between CLDN-1 and Ras/EFGR in the development of lung cancer and the combination of both has strong clinical significance [119]. Surprisingly, as compared to the previous studies that showed that overexpression of CLDN-1 suppressed metastatic abilities of lung adenocarcinoma cells [50,57], this study showed that patients with positive expressions of both CLDN-1 and Ras/EGFR were found to have poor prognosis as compared to CLDN-1(+) Ras/EGFR(−), CLDN-1(−) Ras/EGFR(+), and CLDN-1(−) and Ras/EGFR(−) patients [119]. Another study found that enhanced cell migration by tumor necrosis factor and a similar morphology like fibroblast was found to be reduced by small CLDN-1 interfering RNA in the cells of lung cancer [121].

#### *2.8. Claudin-1 and Pancreatic Cancer*

Pancreatic cancer is the fourth major cause of deaths caused by cancer worldwide, with a strong capacity for metastasis and recurrence [123]. In pancreatic cancer (PC), increased expression of CLDN-1 was found to be associated with epithelial-mesenchymal transition. CLDN-1 is expressed by ductal pancreatic adenocarcinomas as well as intra-ductal papillary mucinous pancreatic tumors. One study demonstrated that 58% positive CLDN-1 immunostaining in ductal pancreatic adenocarcinomas and intraductal papillary pancreatic tumors [51]. Another study showed that through activation of mitogen-activated protein kinase 2 (MEK2), CLDN-1 was involved in cell dissociation of PC cells [124]. A further study observed the role of CLDN-1 in the progression of human PC using the PANC-1 cell line [64]. Increased expression of PARP [poly-(ADP-ribose) polymerase], an apoptosis marker, and decreased PANC-1 proliferation was observed after treatment with TNF-α [64]. Increased proliferation in PANC-1 cells was observed after treatment with TNF-α and CLDN-1 siRNA against CLDN-1, suggesting the cells were resistant to TNF-α-induced apoptosis when transfected with CLDN-1 siRNA. These findings clearly demonstrated that the CLDN-1 expression plays a role in the proliferation of PC cells [64].

#### *2.9. Claudin-1 and Ovarian Cancer*

Ovarian cancer affected a significant number of women worldwide and is the seventh most frequent cause of deaths due to cancer in women [125]. The upregulation of the *CLDN-1* gene is found to be associated with ovarian cancer [52]. Studies have shown that overexpression of CLDN-1 caused reduced cell differentiation and a high invasive growth rate [126]. The role of CLDN-1 has been widely studied in two different types of ovarian cancers, namely, ovarian serous and ovarian endometroid carcinoma [127]. The expression of CLDN-1 was shown to be negatively regulated by microRNA-155 (miR-155) which results in reduced proliferation and invasion of human ovarian cancer-initiating cells [52]. Reports have shown that elevated level of CLDN-11, 4, and 7 promotes the growth of both benign and malignant epithelial ovarian cancers [52]. Extensive studies conducted and analyzed for the association of CLDN-1 with survival and anatomical site showed 85% elevation of CLDN-1 expression [58]. Recently, the level of CLDN-1 expression was also investigated in borderline tumors of the ovary (BOT) [53]. A significantly higher level of CLDN-1 expression was associated with the peritoneal implants and micropapillary patterns that are specifically seen only in serous BOT [53].

#### *2.10. Claudin-1 and Oral Squamous Cell Carcinoma*

Almost 90% of all oral carcinomas are oral squamous cell carcinomas (OSCC) [128]. It has been shown previously that the invasive activity of OSCC cells is enhanced by CLDN-1 through activation of MMP-1 and 2, resulting in increased cleavage of Laminin-5 γ2 chains. The authors of the study further revealed elevated expression of CLDN-1 in OSC-4 and NOS-2 cell lines which are highly invasive [129]. One study demonstrated the association of high CLDN-1 expression with aggressive histopathologic features such as perineural and vascular invasion and suggested that CLDN-1 might

be directly or indirectly involved in the progression of OSCC [130]. Another study found that the absence of CLDN-1 was associated with poorly differentiated tumors [131]. Immunohistochemical analysis revealed that the presence of CLDN-1 in the invasive front of tumor islands was associated with neck mode metastasis. The results obtained from this study further suggested that the expression of CLDN-1 is linked with the recurrence of OSCC [132].

#### *2.11. Claudin-1 and Melanoma*

Melanoma, which arises from melanocytes, causes 75% of deaths related to skin cancers [133]. CLDN-1 was found to be upregulated in melanoma tissues [134]. In melanoma, CLDN-1 is abnormally/aberrantly expressed in the cytoplasm of malignant cells and not in the cell membrane. This may be related to its influence on protein kinase C (PKC) activity [38]. PKC activation caused an increase in transcription and protein expression of CLDN-1 and thus, cell motility [38]. When melanoma cells transfected with CLDN-1, it increased the secretions of matrix metalloproteinase- 2 (MMP-2) reflecting its contribution to the cell invasion process. The data from French, et al. supported the hypothesis that the invasive capacity of melanoma cells is increased by cytoplasmic expression of CLDN-1 and not by the elevated nuclear expression of CLDN-1 [56]. In melanoma patients with brain metastases, the expression of CLDN-1 was downregulated, and the introduction of CLDN-1 retrovirus reduced the tumor aggressiveness and tumor migration ability and diminished micro-metastasis in the brain. This shows that reduction in CLDN-1 supports tumor progression and metastasis and that CLDN-1 can be used as a prognostic predictor for melanoma patients with increased risk of brain metastasis [135].

#### *2.12. Claudin-1 and Prostate Cancer*

Prostate cancer is the second most diagnosed malignancy and fifth-most leading cause of cancer deaths in men [136]. The Gleason grading score system is the most commonly used method to evaluate the aggressiveness of prostate cancers, whereas, the changes in the glandular architecture indicate the tumor grades [54]. The typical glandular architecture is supported by cellular polarity and cell-to-cell contact, and thus the alterations and dysregulations of proteins mediating normal cellular connection may impact the histology and Gleason grade [54]. It is suggested that the loss of tight junction protein, CLDN-1, is associated with cancer invasion, progression and the transformation into metastatic phenotype in prostate cancers. A study reported that the lower expression of CLDN-1 correlated with higher prostate-specific antigen in prostate cancer [54]. Further studies are needed to thoroughly investigate the association between prognostic factors and claudins expression in prostate cancer.

#### **3. Claudin-1 and Tight Junction Barrier Function**

CLDN-1 is an integral membrane protein that in conjunction with other claudins forms the tight junctions and together plays an essential role in epithelial barrier functions. CLDN-1 has a significant role in epithelial differentiation and loss of CLDN-1 can impair the functioning of tight junctions [137]. Numerous studies have reported the involvement of CLDN-1 in transepithelial electrical resistance (TER) [138,139] and paracellular permeability [140] showing its importance in tight junction barrier functions. The study showed that CLDN-1 increased the TER and reduced paracellular flux in Madin-Darby Canine Kidney (MCDK) cells [138]. While another study reported that induced CLDN-1-myc in MDCK cells resulted in the formation of aberrant tight junction strands independently without the participation of ZO-1 and occludin [139]. CLDN-1 controls the flux of solutes by localizing at tight junctions and modulating the paracellular permeability. Any defect in the expression of CLDN-1 can result in tight junction dysfunction causing increased paracellular permeability leading to various pathologies such as in Neonatal ichthyosis-sclerosing cholangitis (NISCH) syndrome [140]. This study showed that the silencing of CLDN-1 leads to increased hepatic paracellular permeability [140]. Mostly, an increase in the tight junction proteins should lead to an increase in the tight junction integrity but this is not a universal phenomenon as one study reported that high expression of CLDN-1 resulted in decreased TER and increased permeability causing loss of barrier function in intestinal epithelial cells (IEC-18) treated with TNF-α [141]. The authors explained the reason for this contradictory finding is that the loss of barrier function was due to the reduced expression of occludin protein and not CLDN-1 [141]. Besides, the other study reported that high expression of CLDN-1 resulted in blood–brain-barrier (BBB) leakiness during post-stroke recovery and targeting of CLDN-1 by a CLDN-1 peptide improved the permeability of brain endothelial barrier [142]. So, these studies suggest that the upregulation of CLDN-1 cannot necessarily be universal to the increased barrier function and there might be other contributing factors that regulate these functions, and also we should not rule out the tissue specific expression of claudins as other possibility to their dichotomous roles.

#### **4. Gene Expression, Survival and Pathway Interaction Analysis of claudin-1 across di**ff**erent Cancers**

Survival Analysis was performed using online survival analysis tools to assess the influence of CLDN-1 expression on survival in different types of cancer using the Cancer Genome Atlas (TCGA) datasets. This analysis was to investigate whether an alteration in gene expression correlates with poor survival or with tumor recurrence. The data showed that the gene expression of CLDN-1 did not affect survival significantly in cancers like breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) and pancreatic adenocarcinoma (PAAD) (*p* > 0.05), but did significantly associate with survival for cancers like thyroid carcinoma (THCA), adrenocortical carcinoma (ACC), rectum adenocarcinoma (READ) (\* *p* < 0.05) (Figure 4). All survival analysis was performed using online survival analysis tools.

**Figure 4.** Survival analysis of CLDN-1 in various cancers. The red line denotes higher expression, and the blue line indicates lower expression. (**A**) Breast invasive carcinoma (BRCA); (**B**) rectum adenocarcinoma (READ); (**C**) cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC); (**D**) thyroid carcinoma (THCA); (**E**) adrenocortical carcinoma (ACC); (**F**) pancreatic adenocarcinoma (PAAD).

We further compared the expression of CLDN-1 between tumor and normal tissues for ACC, CESC, PAAD, READ, THCA and BRCA from TCGA datasets. We found that most of the cancers showed significant expression differences of CLDN-1 between tumor and normal type. We observed

that for most of the cancers, tumor tissues have higher expression than normal tissues but for cancer like ACC, CLDN-1 expression in normal tissue was found to be greater than tumor tissues (Figure 5). We did not find any significant difference in CLDN-1 expression between tumor and normal tissue of BRCA.

**Figure 5.** Boxplot showing the distribution of CLDN-1 expression in tumors and normal tissues for different types of cancers. Significant differences are shown with an asterisk (\*). Boxplots were generated using GEPIA1 webserver and *p*-value < 0.01 was considered as significant.

To determine the interaction of CLDN-1 with other genes, we performed gene interaction analysis using the Gene MANIA prediction server. We found that CLDN-1 significantly interacted with several key genes that play an important role in normal cell physiology. Any disturbance in CLDN-1 expression or its partners may result in the manifestation of various diseases including cancers. The interaction of CLDN-1 with other key molecules can be individually analyzed in different cancers. This may highlight the key pathways which can be therapeutically targeted to suppress cancer growth or metastasis (Figure 6).

**Figure 6.** CLDN-1 interaction network using the Gene MANIA prediction server.

#### **5. Claudin-1 as a Drug Target**

The involvement of CLDN-1 in various pathological conditions has provided new perceptions into drug development targeting CLDN-1. The approach of targeting CLDN-1 either by monoclonal antibodies or chimeric antibodies has great potential but needs more research to reach the level of clinical trials. The initial studies have laid an important foundation towards the new strategies that could be employed and further modified towards the potential usefulness of CLDN-1 as a therapeutic target. The localization of CLDN-1 as a transmembrane protein makes it a perfect target for the enhanced drug absorption for preventing infection and treating cancer. One of the studies observed that the human hepatocytes treated with mouse anti-CLDN-1 monoclonal antibodies (mAbs), showed improved drug absorption and prevented hepatitis C virus (HCV) infection [143]. A human-mouse chimeric CLDN-1 mAb (clone 3A2) demonstrated cellular cytotoxicity against CLDN-1 expressing cancer cells [144]. The other aspect of claudins that is being exploited for therapeutic targeting is their role in regulating paracellular permeability in different tissues. *Clostridium perfringens* enterotoxin (cCPE) binds with claudin through its claudin binding domain and inhibits the claudin function. It was observed that blocking CLDN-1 with cCPE variants in the Huh7.5 hepatoma cell line inhibited infection of Huh7.5 cells with HCV in a dose-dependent manner and this also opened the epidermal barrier in the reconstructed human epidermis [145]. To eliminate the possible limitation facing CLDN-1 targeted therapies due to genotype-dependent escape via CLDN-6 and CLDN-9 and to improve anti-HCV activity, humanized anti- CLDN-1 monoclonal antibody (mAb) could be an alternative. One study developed functional mAB against extracellular domains of CLDN-1 and found that these antibodies have a very high affinity for intact CLDN-1, efficiently inhibited HCV infections both in vitro and in vivo, further demonstrating that anti-CLDN1 mAbs could be useful in inhibiting HCV infections [146]. In a very recent study, CLDN-1 was successfully targeted with anti-CLDN1 near-infrared fluorophore to track the colorectal cancer cells, and it may provide a novel way for fluorescence-guided surgery of tumor [147].

The main concern with claudin-targeted therapies is the presence of claudins in both normal epithelial cells and cancer cells that makes the targeting difficult. However, it has been observed that claudins are localized at the tight junctions in normal tissues, while in malignant tissues, there is a dysregulation of claudins localization from the tight junctions to the cell surface [24,148]. Claudins with aberrant localization in malignant tumors can be recognized by utilizing the C-terminal claudin-binding domain of cCPE fused with protein synthesis inhibitory factor (C-CPE-PSIF), causing less cytotoxicity to normal cells and a study has shown how CLDN-4 can be used as a target for tumor therapy by fusion of cCPE with PSIF. The results of the study showed that C-CPE-PSIF was cytotoxic to cells with undeveloped tight junctions (preconfluent cultures of Caco-2) and was not cytotoxic to cells with developed tight junctions (postconfluent cultures of Caco-2) [149]. Several other studies have used CLDN-4 as a target for tumor therapy and showed the accumulation of anti-CLDN-4 mAbs specifically in the tumors and reduced the growth of human colorectal and gastric tumors in mice [150]. One study detected CLDN-4 upregulation non-invasively in mice pancreatic ductal adenocarcinoma xenografts by using MRI and 18FDG-PET [151] (for detailed reviews see [152]). In the future, a similar approach can be applied to CLDN-1 by preparation of CLDN-1 targeting molecule and can be tested for cytotoxicity to normal cells.

#### **6. Claudins and Autosomal Recessive Disorders**

As explained in the previous sections, the claudin family of proteins is an integral part of tight junctions that determine paracellular selectivity and permeability to small ions by acting as pores or barriers in polarized epithelia. We also discussed how overexpression or reduction of claudins could both promote and limit cancer progression, revealing complex dichotomous roles for claudins depending on cellular context. Besides the fact that the abnormal or deregulated expression of claudins has been associated with different human diseases like cancer, there are also other human disorders such as autosomal recessive disorders that have been reported due to clearly defined mutations in the corresponding claudin genes. Such disorders are mostly observed in skin, liver, kidney, the inner ear, and the eye. The first evidence that showed a mutation in the claudin family of tight junction proteins causes human disorders were from the group of Lifton [153]. In their study, they reported that the mutations in the human gene, paracellin-1 (PCLN-1)/CLDN-16 causes an autosomal recessive disorder called Familial hypomagnesemia with hypercalcinuria and nephrocalcinosis (FHHNC) characterized with renal Mg2<sup>+</sup> and Ca<sup>+</sup> wasting. Later the same group revealed additional evidence that loss of function mutations in paracellin-1 PCLN-1/CLDN-16, are causative of FHHNC [154]. PCLN-1 is related to the claudin family of tight junction proteins and is in tight junctions of the thick ascending limb of Henle (TAL). CLDN-16 is a cattle ortholog of PCLN-1 with ∼ 90% sequence homology, and PCLN-1/CLDN16 mutations have been shown to be strongly associated with bovine chronic interstitial nephritis with diffuse zonal fibrosis (CINF) [155]. Although both renal disorders FHHN and CINF are caused by PCLN-1/CLDN16 mutations, but the clinical features of both diseases are quite different which may be due to specific mutations/deletions in the same gene or through species specificity. Since the first report [153], several other tight junction disorders have been shown to cause human diseases including mutations in claudin proteins such as CLDN-1 [156], CLDN-9 [157], CLDN-10 [158], CLDN-14 [159], CLDN-16 [160], CLDN-19 [161] (for detailed reviews see [162]).

#### **7. Conclusions**

It is clear from the literature that among all the tight junction proteins, the claudin family of proteins is particularly important in regulating normal cell physiology. Among the claudin family, CLDN-1 is the most extensively studied protein and has been shown to be involved directly or indirectly in the development and progression of cancer, and also has a suppressive role in some cancers. CLDN-1 acts alone or in combination with other molecules to exert its tumor promoting or suppressing effect. Likewise, the shuttling of CLDN-1 between the cell membrane, cytoplasm and nucleus is a deciding factor in the development and progression of cancers. Another important aspect is the involvement of CLDN-1 in many signaling pathways, especially in Wnt and Notch signaling. The association of CLDN-1 with patient survival or recurrence in many cancers suggests its importance as a prognostic marker and as a potential therapeutic target. Also, pathway interaction analysis revealed CLDN-1 interacting partners, which can be further explored as potential drug targets. Based on the complexity of the topic, there is no one statement we can make for CLDN-1 role in cancer or barrier function since it is more intricate (Claudins are upregulated or downregulated in cancer and may or may not play a role in barrier function). In other words, universal statements concerning CLDN-1 and cancer or CLDN-1 and barrier function are dangerous oversimplifications.

**Author Contributions:** Writing—original draft preparation, A.A.B., L.T., S.N., S.U., M.H.; writing—review and editing, S.H., M.A.M., P.D., W.E.-R., M.P.F., N.S., R.K.; visualization, S.M., H.A.-N., A.A.; supervision, A.A.B., M.H., W.E.-R., S.U., P.B., R.R.; project administration, A.A.B., N.S., S.K.Y., M.H., S.U.; funding acquisition, A.A.B., M.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Sidra Medicine to Mohammad Haris (50610110002) and Ajaz A. Bhat (5011041002). Muzafar A. Macha is a recipient of Ramanujan Fellowship from Science and Engineering Research Board (SERB), Department of Science and Technology, Govt. of India, New Delhi.

**Acknowledgments:** The authors would like to express their gratitude to Vineeta Tanwar (Research Scientist, Ohio State University, Ohio, Columbus, USA) for help in English editing and valuable suggestions to improve the quality of the manuscript. The publication of this article was funded by the Qatar National Library.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**


#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*International Journal of Molecular Sciences* Editorial Office E-mail: ijms@mdpi.com www.mdpi.com/journal/ijms

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18