Connexin 43 plays an important role in the transformation of cholangiocytes with Clonochis sinensis excretory-secretory protein and N-nitrosodimethylamine

Background Clonorchis sinensis is a group I bio-carcinogen responsible for cholangiocarcinoma (CHCA) in humans. However, the mechanism by which C. sinensis promotes carcinogenesis is unclear. Methodology Using the human cholangiocyte line H69, we investigated cell proliferation and gap junction protein expression after stimulation with the hepatotoxin N-nitrosodimethylamine (NDMA) and/or excretory-secretory products (ESP) of C. sinensis, which induce inflammation. NDMA and ESP treatment increased proliferation by 146% and the proportion of cells in the G2/M phase by 37%. Moreover, the expression of the cell proliferation-related proteins E2F1, Ki-67, and cancer related protein cytokeratin 19 and Cox-2 increased in response to combined treatment with NDMA and ESP. The gap-junction proteins connexin (Cx) 43 and Cx26 increased. In contrast, Cx32 expression decreased in cells treated with NDMA and ESP. Silencing of Cx43 reduced cell proliferation and significantly suppressed Cx26 and Cox-2 expression. Conclusions These results suggest that Cx43 is an important factor in CHCA induced by C. sinensis ESP and NDMA and further investigations targeting this pathway may allow prevention of this deadly disease.


Introduction
Clonorchis sinensis is a human liver fluke that induces cholangiocarcinoma (CHCA) in humans [1]. Clonorchiasis has been endemic in Asia for a long time, especially among residents who live along rivers and consume raw freshwater fish [2].
The mechanism by which C. sinensis induces CHCA is not well understood, but chronic hepatobiliary damage, a precursor to CHCA, in clonorchiasis is a multi-factorial outcome of the mechanical and biochemical irritation of the biliary epithelium by flukes via their suckers, metabolites, and excretory-secretory products (ESP) [3]. Local inflammation and the systemic immune response in the host [4][5][6][7] produce reactive oxygen species and reactive nitrogen compounds, which may cause DNA damage, leading to neoplasms [8,9].
N-nitrosodimethylamine (NDMA) is a potent hepatotoxin that can cause fibrosis and tumors in the liver of rats via the activation of CYP450 enzymes [10] and hamsters infected with C. sinensis are at a greater risk of developing NDMA-induced or inflammation-mediated CHCA than uninfected hamsters [11,12]. Previously, we reported that exposure to NDMA and the ESP of C. sinensis increases HEK293T cell proliferation and the proportion of cells in the G2/M phase [3].
CHCA is potentially caused by increased levels of proinflammatory cytokines and nuclear factor kappa B (NFκB), which regulate the activities of cyclooxygenase-2 (Cox-2) and inducible nitric oxide synthase, and disturb the homeostasis of oxidants/anti-oxidants and DNA repair enzymes [13]. ESPs of C. sinensis induce pro-inflammatory responses in vitro by the upregulation of TLR4 and its downstream transduction signals, including MyD88-dependent IκB-α degradation and NFκB activation [14,15]. NFκB may also influence the production of Cx43, a gap-junction protein, in liver cirrhosis [16,17].
Gap junctions are clusters of transmembrane channels on the cell membrane that permit direct intercellular exchange of ions, secondary messengers, and small signaling molecules influencing cell growth, differentiation, and cancerous changes [17][18][19][20][21]. Among gap junction proteins, Cx43 is involved in almost all steps of the inflammatory response of cells, cytokine production, and inflammatory cell migration [17,20,22]. The substantial role of Cx43 in carcinogenesis is highlighted by the fact that high levels of Cx-43 expression increase the invasion of breast tumor cells and promote tumors in melanoma [22].
Alterations of Cx expression have been reported in cancer [21,23]. In hepatocellular carcinoma (HCC), for instance, reduced Cx32 expression is accompanied by increased expression of Cx43, which promotes HCC via cell-to-cell communication [16,20,23]. Fujimoto et al. [23] have shown that Cx32 has a suppressive effect in metastatic renal cell carcinoma. However, the role of connexins in cancer is still controversial [23], and the influence of gap junctions in CHCA caused by C. sinensis has not yet been examined.
In this study, to determine the mechanisms underlying the carcinogenic effects of ESP from C. sinensis, we investigated changes in cell proliferation, proinflammatory molecules, and connexin production in cholangiocytes (H69 cell line) exposed to ESP from C. sinensis and the carcinogen NDMA. We found that the silencing of Cx43 reduced ESP-and NDMA-induced cell proliferation and the expression of Cox-2.

Animals
The animal experimental protocol was approved and reviewed by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University Health System, Seoul, Korea (approval no. SNU-060511-1) and followed the National Institutes of Health (NIH) guideline for the care and use of laboratory animals (NIH publication no. 85- 23, 1985, revised 1996). The facility is accredited by the Ministry of Food and Drug Administration and by the Ministry of Education, Science and Technology (LNL08-402) as an animal experiment facility. Male Sprague-Dawley rats at 6 weeks of age were purchased from Koatech Co. (Seoul, Korea) and housed in an ABL-2 animal facility at Seoul National University College of Medicine. All rats were bred in filter cages under positive pressure according to institutionally approved guidelines.

Recovery of metacercariae of C. sinensis
Pseudorasbora parva, the second intermediate host of C. sinensis, which was naturally infected with C. sinensis, was purchased from Sancheong-gun, Gyeongsangnam-do, Republic of Korea. This is an endemic area for clonorchiasis, which we have previously described [2]. Metacercariae of C. sinensis were collected by digestion of fish with pepsin-HCl (0.6%) artificial gastric juice for 1 h at 37˚C, then filtering the digestion solution through a 200㎛ stainless steel filter. Finally, and C. sinensis metacercariae were isolated under stereomicroscopic identification.

Infection of experimental animals with C. sinensis and collection of ESP
Sprague-Dawley rats were individually infected orally with 50 metacercariae of C. sinensis. Eight weeks post-infection, adult worms were collected from bile ducts and washed several times with phosphate-buffered saline (PBS). The freshly isolated worms were then incubated in sterile PBS containing antibiotics for 24 h in an atmosphere of 5% CO 2 at 37˚C. After incubation, the medium was centrifuged for 10 min at 800 rpm to remove the worms and debris. The supernatant was then further centrifuged for 10 min at 3000 rpm and filtered with a syringe-driven 0.45-μm filter unit. The amount of protein in each extract was measured using the Bradford assay (Thermo, Rockford, IL, USA). The concentration of endotoxin (LPS) was measured using an LAL QCL-1000 Kit (LONZA, Switzerland), and as a result, the LPS contained in 10 μg/mL of ESP was confirmed to be less than 0.001 (EU). Therefore, there is no effect of LPS on the results of this paper.

Cell proliferation assay
The PrestoBlue cell viability reagent was utilized to evaluate cell proliferation. For each assay, cells were seeded at a density of 5 × 10 3 cells/well on 96-well plates. After 24 h of incubation, the medium was replaced with 2% FBS-DMEM without phenol red. The cells were then incubated in the presence of PBS (vehicle) or with 100 ng/mL NDMA with or without 10 μg/mL ESP for another 72 h. The PrestoBlue cell viability reagent (1 mg/mL) was dissolved in warm medium, and 1.25 mM phenazine methosulfate (PMS) was prepared in PBS. Following incubation for the indicated periods, 50 μL of the PrestoBlue cell viability reagent was added to each well. The plates were then incubated for 1 h. The conversion of PrestoBlue cell viability reagent was quantified by measuring the absorbance at 570 and 600 nm using a microtiter plate reader.

Cell cycle analysis
For the cell cycle analysis, H69 cells were plated in six-well culture plates at 2 × 10 5 cells/well in 2 mL of DMEM containing 10% FBS. They were then treated with 100 ng/mL NDMA with or without 10 μg/mL ESP for 72 h and stained with propidium iodide (PI). The PI-stained cells were analyzed using a FACSCalibur multicolor flow cytometer (Becton-Dickinson, Franklin lakes, NJ, USA), and data were analyzed using CellQuest software (Becton-Dickinson).

Confocal microscopy
Cells were washed with cold PBS three times and fixed with 2% paraformaldehyde in PBS for 30 min. Permeabilization was performed by treating the cells with 0.2% (w/v) Triton X-100 in PBS for 5 min and then blocking with 0.5% BSA in PBS for 1 h. After blocking, the cells were incubated with primary antibodies against connexin 26, 32, and 43 (Invitrogen) diluted in BSA-PBS at 25˚C for 2 h and then incubated in secondary antibodies diluted in BSA-PBS at room temperature for 30 min. After washing with 1× PBS, the cells were stained with DAPI and observed under a confocal laser scanning microscope (LSM PASCAL; Carl Zeiss, Jena, Germany).

Real-time PCR
RNA samples from each cell line were column-purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Reverse transcription and real-time PCR (RT-PCR) were performed to determine the mRNA expression levels of Cx26, Cx32, Cx43, and Cox-2, using GAPDH as a  control (Applied Biosystems, Santa Clara, CA, USA). The levels of RNA expression were calculated using the 2ΔΔCt method and individual expression data was normalized to GAPDH. The thermal cycling parameters for reverse transcription were modified according to the Applied Biosystems manual. Hexamer incubation at 25˚C for 10 min and reverse transcription at 42˚C for 30 min was followed by reverse transcriptase inactivation at 95˚C for 5 min. cDNA (20 ng) from the previous step was subjected to qRT-PCR using specific sets of primers ( Table 2 in a total reaction volume of 25 μL (Applied Biosystems). qRT-PCR was performed in an optical 96-well plate using an ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems) and TaqMan probe detection chemistry. The running protocol was as follows: initial denaturation at 95˚C for 10 min, and 40 cycles of amplification at 95˚C for 15 s and 60˚C for 1 min. After PCR, a dissociation curve was constructed by increasing the temperature from 65˚C to 95˚C to evaluate the PCR amplification specificity. The cycle threshold (Ct) value was recorded for each sample.

Statistical analysis
Statistical significance was analyzed using Student's t-tests. Differences were considered statistically significant at a � P < 0.05, �� P < 0.01 and ��� P < 0.001 versus control. Data are presented as the mean ± standard error of the mean (SEM) from at least three independent experiments.

ESP and NDMA synergistically increase H69 cell proliferation
The roles of NDMA and ESP in the proliferation of H69 cells, as determined by cell viability, were investigated. Proliferation was higher in H69 cells treated with NDMA and ESP of C. sinensis than in control cells. Compared to control cells, the average increase for cells treated  with NDMA was 112%, for cells treated with ESP was 120% (P < 0.05), and for cells treated with NDMA + ESP was 146% (P < 0.001). NDMA and ESP had synergistic effects on cell proliferation ( Fig 1A).

Cell cycle distribution upon NDMA and/or ESP treatment in H69 cells
Cell cycle progression was monitored using propidium iodide (PI) staining (Fig 1B). In H69 cells treated with NDMA, ESP, or NDMA + ESP for 72 h, the number of cells in SubG1 phase did not change significantly, whereas cell numbers in the G0/G1 phase significantly decreased, but cells in the G2/M phase increased significantly compared to cell counts in the control group ( Fig 1B). Fewer S-phase cells were identified in the ESP-and NDMA + ESP-treated cells than in control cells. There is no significant increase in G2/M phase in cells treated with NDMA compared to control group (Fig 1B). The proportions of G2/M-phase cells in each condition were as follows: control, 17%; NDMA, 20%; ESP, 26%; NDMA + ESP, 37% (P = 0.007).

Upregulation of cell proliferation-And cancer-related proteins by NDMA and ESP in H69 cells
Immunoblotting was utilized to detect the regulation of cell-cycle-related proteins in each group using calnexin as a loading control. The expression levels of several cell proliferation- including E2F1, Ki-67 and Cy-19, and Cox-2 (an essential regulator of the G2/M transition) as cancer-related makers, were upregulated, especially in NDMA + ESP-treated cells (Fig 2).

Gap junction proteins in H69 cells
The expression levels of Cx26 and Cx43 were high in NDMA + ESP-treated cells, as confirmed by immunoblotting (Fig 3). The intracellular concentrations of Cx26 and Cx43 in each condition were observed under a confocal microscope. Western blotting indicated increases in Cx26 and Cx43 expression in cells treated with ESP and NDMA + ESP (Fig 3). The expression of Cx32 (blue) was markedly lower in cells treated with NDMA + ESP than in other cells. Immunofluorescence confocal microscopy indicated that the intracellular concentrations of Cx26 (green) and Cx43 (green) increased in cells treated with NDMA + ESP (Fig 4).

Reduced cell proliferation upon NDMA and ESP treatment in H69 cells by Cx43 silencing compare to silencing negative control
Downregulation of Cx43 by Cx43-specific small interfering RNA resulted in a slight increase in cell proliferation, but the differences among groups were not significant, even though ESP and NDMA synergistically stimulated cell proliferation in silencing negative control. The average increases compared to the control were as follows: NDMA, 104.5%; ESP, 105.1%; and NDMA + ESP, 107.6% (Fig 5A).

Downregulation of Cx43 by Cx43-specific small interfering RNA reduced the expression of Cx26 and Cox-2 in H69 cells
To evaluate the effect of Cx43 downregulation on the expression of other gap junction proteins and Cox-2, H69 cells were harvested after treatment with NDMA and ESP of C. sinensis for 72 h. Transfection with Cx43 siRNA resulted in a reduction in Cx43 expression of greater than 70% when compared with that of the liposome-only control (Fig 5A). Real-time PCR showed that Cx43 silencing resulted in the downregulation of Cx26 and Cox-2 (Fig 5C and 5D, respectively). However, Cx32 expression was not affected in H69 cells transfected with Cx43-specific siRNA.

Discussion
Our results provided the first evidence for the involvement of gap junction proteins in the pathogenesis of CHCA by C. sinensis. NDMA and ESP of C. sinensis increased Cx43 expression substantially in H69 human cholangiocytes. In addition, in the presence of ESP and NDMA stimulation, Cx43 knockdown inhibited Cox-2 expression in H69 cells. Yan et al. [15] reported that iNOS is highly expressed in Kupffer cells, sinusoidal endothelial cells, and biliary epithelial cells in BALB/c mice after C. sinensis infection. Nitric oxide (NO) formation and nitrosation may contribute to the development of C. sinensis-associated carcinogenesis [11,12,14,15]. The induction of iNOS under inflammatory conditions suggests that NO is involved in the upregulation of Cx43 [16,19]. Therefore, we also expect the involvement of iNOS in the elevation of Cx43 expression under inflammatory conditions in the present C sinensis model. Further studies are warranted to test this hypothesis. Our results indicated that ESP of C. sinensis and NDMA had a synergistic effect on the proliferation of human cholangiocytes (Fig 1). In addition to the observed increase in cell proliferation and alteration of the cell cycle, the expression of the gap junction proteins Cx43 and  [17][18][19][20][21][22]. When used in combination with NDMA, ESP of C. sinensis induced cell proliferation and increased the expression of E2F1, and Ki-67 as cell proliferation proteins. Additionally, as shown in Fig 1B, treatment with NDMA + ESP maximized the proportion of cells in the G2/M phase, implying that NDMA and ESP synergistically stimulate cell cycle progression. We analyzed the expression of cell proliferation-, cell cycle-, and cancer-related proteins (Fig 2), including E2F1, Ki-67, and Cy19, Cox-2 [24][25][26]. E2F1 is able to induce cell cycle progression, resulting in cell proliferation [3,24]. Consistent with these previous findings, we observed that increased expression of E2F1 stimulated cell proliferation. C-Met is involved in early events of carcinogenesis, and Ki-67 is involved in the formation of invasive carcinoma [25][26][27][28]. Biliary epithelial cells retain Cy19 expression after neoplastic transformation in almost all cases [26,28]. In our study, Cy19 and Ki-67 were upregulated in response to the stimulation of H69 cell proliferation.
Cyclooxygenase 2 (Cox-2), an enzyme involved in the production of prostaglandins, was over-expressed when cells were stimulated by NDMA and ESP. Cox-2 over-expression has been observed in various inflammatory diseases and in bile duct carcinoma cells, mainly in the cytoplasm [29][30]. Importantly, bile duct epithelial cells in primary sclerosing cholangitis show very strong Cox-2 expression, comparable to that in carcinoma cells. In contrast, epithelial cells in primary biliary cirrhosis show moderate levels of Cox-2 expression [29,30]. In this context, the over-expression of Cy19, Ki-67, and Cox-2 may result in the transformation of normal H69 cells to cancer-like cells by stimulation with NDMA and ESP of C. sinensis. In the After treatment with NDMA and ESP for 72 h, the concentration of intracellular connexin 26, 32, and 43 in H69 cells was measured by laser scanning microscopy (×4,000). A-C: Immunofluorescence confocal microscopy indicated that the intracellular concentrations of Cx26 (green)/DAPI (blue), Cx32 (blue)/ DAPI (red), and Cx43 (green)/ DAPI (red) increased in cells treated with NDMA + ESP. Scale bar = 25 μm. Data represent the mean ± SE of five independent experiments. � P < 0.05, �� P < 0.01 and ��� P < 0.001 versus control.
https://doi.org/10.1371/journal.pntd.0006843.g004 present study, Cx43 and Cx26 expression levels were increased in H69 cells upon stimulation with NDMA and ESP of C. sinensis. In contrast, Cx32 was significantly downregulated. Increased expression of hepatic Cx43 was noted in cirrhosis and in a mouse model of acuteon-chronic liver failure in response to LPS, and this effect was related to the severity of inflammation [19]. This increased Cx43 expression was likely an adaptive protective response of the liver to enable better cell-to-cell communication [16,20,21]. The expression levels of Cx26 and Cx32, major connexins in the liver, are extremely low in several HCC lines, but Cx43, a minor connexin in the liver, is highly expressed in metastatic cancer [17,[20][21][22]. Faniku et al. [30] reported that the inhibition of Cx43 signaling plays a more significant role in regulating  Table 3. Full spectrum of connexins expressed in rodent and human livers.

Connexins
Localization References
In general, cells contain several known connexins, classified according to their intracellular location (Table 3).
Intercellular communication via gap junctions is inhibited by increased Cox-2 expression, as is frequently observed in several forms of human malignancies [29,33]. Recently, several reports have suggested that the carcinogenic mechanisms of hydrogen peroxide, TPA, and quinones may be involved in the inhibition of GJIC by Cx43 phosphorylation via ERK1/2 activation in rat liver epithelial cells [13,17]. Furthermore, increased expression of Cx43 is positively correlated with NFκB activation in muscular arteries of patients undergoing coronary artery bypass graft surgery [34]. NFκB plays a central role in general inflammatory and immune responses. The 5 0 -flanking region of the Cox-2 promoter contains an NFκB binding site, and NFκB is a critical regulator of Cox-2 expression in many cell lines [13,34]. Taken together, the present findings suggest that Cx43 expression induces Cox2 over-expression via NFκB activation. However, until now, the link between NFκB activation, Cx43 expression, and Cox2 overexpression has not been clearly established. In the future, it will be interesting to examine the relationship between the GJIC and NFκB activation, Cox-2 by ESP of C. sinensis and NDMA.
In conclusion, our results suggest that Cx43 plays a key role in cell proliferation, potentially leading to CHCA development upon stimulation by ESP of C. sinensis and NDMA.