Cadmium induces Wnt signaling to upregulate proliferation and survival genes in sub-confluent kidney proximal tubule cells

Abstract

Background: The class 1 carcinogen cadmium (Cd2+) disrupts the E-cadherin/beta-catenin complex of epithelial adherens junctions (AJs) and causes renal cancer. Deregulation of E-cadherin adhesion and changes in Wnt/beta-catenin signaling are known to contribute to carcinogenesis.

Results: We investigated Wnt signaling after Cd2+-induced E-cadherin disruption in sub-confluent cultured kidney proximal tubule cells (PTC). Cd2+ (25 microM, 3-9 h) caused nuclear translocation of beta-catenin and triggered a Wnt response measured by TOPflash reporter assays. Cd2+ reduced the interaction of beta-catenin with AJ components (E-cadherin, alpha-catenin) and increased binding to the transcription factor TCF4 of the Wnt pathway, which was upregulated and translocated to the nucleus. While Wnt target genes (c-Myc, cyclin D1 and ABCB1) were up-regulated by Cd2+, electromobility shift assays showed increased TCF4 binding to cyclin D1 and ABCB1 promoter sequences with Cd2+. Overexpression of wild-type and mutant TCF4 confirmed Cd2+-induced Wnt signaling. Wnt signaling elicited by Cd2+ was not observed in confluent non-proliferating cells, which showed increased E-cadherin expression. Overexpression of E-cadherin reduced Wnt signaling, PTC proliferation and Cd2+ toxicity. Cd2+ also induced reactive oxygen species dependent expression of the pro-apoptotic ER stress marker and Wnt suppressor CHOP/GADD153 which, however, did not abolish Wnt response and cell viability.

Conclusions: Cd2+ induces Wnt signaling in PTC. Hence, Cd2+ may facilitate carcinogenesis of PTC by promoting Wnt pathway-mediated proliferation and survival of pre-neoplastic cells.

Figure 1

Cd²⁺ disrupts the adherens junction complex and redistributes β-catenin from the periphery to cytosol and nuclei of kidney PTC. (A) Time dependent increase of β-catenin protein distribution in cytoplasmic and nuclear fractions of WKPT-0293 Cl.2 cells without (ctl) or with Cd²⁺. Cytosolic and nuclear fractions were immunoblotted with β-catenin antiserum. For loading controls, the same membranes were reprobed with antibodies to β-actin and γ-tubulin for cytoplasmic and nuclear fractions, respectively. β-catenin signals were normalized to loading markers and compared to controls, which were set to 100%. Means ± SEM of 3 experiments are shown. One-way ANOVA was used assuming equality of variance with Levene's test and Tukey post hoc test for pair-wise comparison between control and Cd²⁺ exposed cells. (B) Immunofluorescence staining patterns of β-catenin in WKPT-0293 Cl.2 cells ± Cd²⁺. Nuclei were stained with propidium iodide (PI). Note the peripheral β-catenin labeling in control cells whereas in Cd²⁺ exposed cells β-catenin is found diffusely distributed in cytosol and nuclei. Bars = 20 μm. (C) Distribution of α-catenin in cytoplasmic and nuclear fractions increases with Cd²⁺ exposure time, as shown by immunoblotting. (D) Cd²⁺ reduces association of β-catenin with E-cadherin and α-catenin as a function of time. Cell lysates were immunoprecipitated (IP) with β-catenin antiserum and immunoblotted (IB) with E-cadherin or α-catenin antisera.

Figure 2

Cd²⁺ triggers nuclear translocation of TCF4 and activates TCF4/β-catenin mediated transcription of cell proliferation and survival genes in kidney PTC. (A) Distribution of TCF4 protein in cytoplasmic and nuclear fractions of WKPT-0293 Cl.2 cells ± Cd²⁺ was determined by immunoblotting. (B) Increased association of immunoprecipitated β-catenin with TCF4 by Cd²⁺ in WKPT-0293 Cl.2 cells. (C) Cd²⁺ induces TCF transcriptional activity. Luciferase activity of TOPflash or FOPflash transfected cells treated with 25 μM Cd²⁺ for 3-12 h. Data were corrected for protein and normalized to controls at each time point. Means ± SEM (n = 3-8) are shown. Student's unpaired t-test compares Cd²⁺ treated cells to respective controls. (D) Effect of Cd²⁺ exposure on mRNA of Wnt pathway target genes. GAPDH was used as housekeeping gene. (E) EMSA of TCF4 binding to the ABCB1 promoter region. Extracts from controls (ctl), Cd²⁺ (25 μM) treated cells, cells exposed to 3 h Cd²⁺ followed by incubation with a 200-fold excess of competing unlabeled oligonucleotides (comp.), and lysate-free sample (blank) were loaded. Binding of oligonucleotides to TCF4 was enhanced upon Cd²⁺ exposure. (F) Effect of TCF4 overexpression on Cd²⁺ induced c-Myc mRNA expression. WKPT-0293 Cl.2 cells transiently transfected with full length human TCF4 (TCF4), human TCF4 lacking the interaction domain for β-catenin (ΔN-TCF4), or empty vector were treated with ± 25 μM Cd²⁺. TCF4 overexpression enhanced basal and Cd²⁺-induced c-Myc expression, whereas ΔN-TCF4 had no effect.

Figure 3

Cd²⁺-induced nuclear translocation of β-catenin and Wnt signaling are abolished in confluent kidney PTC, which show increased E-cadherin expression. (A) Immunoblots with β-catenin or α-catenin antisera in ~50% and 100% confluent WKPT-0293 Cl.2 cells (25 μM Cd²⁺, 6 h). For statistical analysis of immunoblots β-catenin signals were normalized to the loading markers and compared to controls, which were set to 100%. Means ± SEM (n = 3) are shown. Student's unpaired t-test compares Cd²⁺ treated cells to respective controls. (B) Cd²⁺-induced TCF transcriptional activity depends on the confluence of kidney PTC. WKPT-0293 Cl.2 cells transfected with TOPflash, FOPflash or ΔN-β-catenin were treated with 25 μM Cd²⁺ for 6 h. Data presented are means ± SEM of 4-5 experiments. Student's unpaired t-test compares Cd²⁺ treated or ΔN-β-catenin-transfected cells to respective controls. (C) To determine association of TCF4 with β-catenin in ~50% and 100% confluent cells, nuclear fractions were immunoprecipitated with an antibody against β-catenin and immunoblotted for TCF4. (D) Expression of E-cadherin in whole cell lysates of ~50% and 100% confluent cells by immunoblotting.

Figure 4

E-cadherin overexpression in kidney PTC abolishes β-catenin redistribution, decreases proliferation and protects against Cd²⁺-induced epithelial barrier disruption and cytotoxicity. (A)β-catenin immunoblots of WKPT-0293 Cl.2 cells transiently transfected with full length E-cadherin expression plasmid (E-cadherin) or empty vector (vector) and incubated for 6 h ± 25 μM Cd²⁺. β-catenin signals were normalized to loading markers and the ratio of β-catenin in Cd²⁺ and respective control samples was determined. Means ± SEM (n = 7) are shown. Student's unpaired t-test compares Cd²⁺ treated cells to respective controls. (B) Vector or E-cadherin PTC were reseeded in SCM in ECIS 8WE10 arrays. C_{40 kHz} reflects cell attachment, spreading and proliferation (see Methods). Initial values of the cell-free electrode were set to 100% (vector, 63.5 ± 1.9 nF; E-cadherin, 65.2 ± 3.3 nF; means ± SEM of 3 experiments). Data show means ± SEM of 3 experiments. (C) At confluence PTC were exposed to SFM ± 20 μM Cd²⁺. Values of the confluent cell-covered electrode were set to 100% (vector, 3670 ± 83 ohm; E-cadherin, 2994 ± 300 ohm; means ± SEM of 3 different experiments). A drop in R_{400 Hz} indicates disruption of epithelial barrier and cell detachment. (D) Vector or E-cadherin cells were exposed to 25 μM Cd²⁺ for 6 h and cell viability was determined by MTT assay. Graph depicts means ± SEM of 9-10 experiments. Student's unpaired t-test compares Cd²⁺ treated cells to respective controls as well as Cd²⁺-exposed vector to Cd²⁺-exposed E-cadherin cells.

Figure 5

Cd²⁺-induced reactive oxygen species (ROS) formation in kidney PTC induces up-regulation of CHOP, but has no effect on Wnt signaling. (A) When used, the ROS scavenger α-tocopherol was pre-incubated for 1 h. Cells were loaded with 20 μM carboxy-H₂DCFDA. Cell suspensions were incubated with Cd²⁺ ± α-tocopherol and fluorescent signals were measured immediately at 500/535 nm. Graph shows means ± SEM of 4-7 experiments. (B) β-catenin immunoblot of WKPT-0293 Cl.2 cells ± α-tocopherol (TOCO) (100 μM) ± 25 μM Cd²⁺ for 6 h. (C) Expression of CHOP mRNA in WKPT-0293 Cl.2 cells without (ctl) or with 25 μM Cd²⁺ by RT-PCR. (D) Prevention of Cd²⁺-mediated increase in CHOP mRNA expression by α-tocopherol. (E) Association of CHOP with TCF4 is increased in WKPT-0293 Cl.2 cells incubated with 25 μM Cd²⁺ for 6 h when compared to controls (ctl). (F) Effect of CHOP overexpression on Cd²⁺-induced expression of c-Myc by RT-PCR. PTC transiently transfected with full length human CHOP (CHOP) or empty vector (vector) were exposed to 25 μM Cd²⁺.

Figure 6

Cd²⁺ triggers TCF4 up-regulation in kidney PTC. (A) TCF4 protein expression was increased by Cd²⁺ in whole cell lysates from PTC. (B) RT-PCR of mRNA was performed with primers specific for rat TCF4 in WKPT-0293 Cl.2 cells without (ctl) or with Cd²⁺. Cd²⁺ increased TCF4 gene expression. (C) TCF4 expression in whole cell lysates of ~50% confluent and 100% confluent cells ± Cd²⁺ (25 μM for 6 h) by immunoblotting confirmed TCF4 up-regulation induced by Cd²⁺ in subconfluent cells. (D) Actinomycin D abolished Cd²⁺-induced TCF4 up-regulation. PTC were treated with or without Cd²⁺ (25 μM) ± actinomycin D (ACT) (10 μg/ml, 1 h pre-incubation), a transcriptional inhibitor. Expression of TCF4 protein was determined in whole cell lysates of WKPT-0293 Cl.2 cells by immunoblotting.

Figure 7

Model for the effects of Cd²⁺ on the adherens junction complex and Wnt signaling in WKPT-0293 Cl.2 kidney PTC. For further details, see discussion.