RhoA-ROCK and p38MAPK-MSK1 mediate vitamin D effects on gene expression, phenotype, and Wnt pathway in colon cancer cells

Abstract

The active vitamin D metabolite 1,25-dihydroxyvitamin D(3) (1,25(OH)(2)D(3)) inhibits proliferation and promotes differentiation of colon cancer cells through the activation of vitamin D receptor (VDR), a transcription factor of the nuclear receptor superfamily. Additionally, 1,25(OH)(2)D(3) has several nongenomic effects of uncertain relevance. We show that 1,25(OH)(2)D(3) induces a transcription-independent Ca(2+) influx and activation of RhoA-Rho-associated coiled kinase (ROCK). This requires VDR and is followed by activation of the p38 mitogen-activated protein kinase (p38MAPK) and mitogen- and stress-activated kinase 1 (MSK1). As shown by the use of chemical inhibitors, dominant-negative mutants and small interfering RNA, RhoA-ROCK, and p38MAPK-MSK1 activation is necessary for the induction of CDH1/E-cadherin, CYP24, and other genes and of an adhesive phenotype by 1,25(OH)(2)D(3). RhoA-ROCK and MSK1 are also required for the inhibition of Wnt-beta-catenin pathway and cell proliferation. Thus, the action of 1,25(OH)(2)D(3) on colon carcinoma cells depends on the dual action of VDR as a transcription factor and a nongenomic activator of RhoA-ROCK and p38MAPK-MSK1.

Figure 1.

1,25(OH)₂D₃ activates the RhoA–ROCK pathway. (A) SW480-ADH cells were treated with 1,25(OH)₂D₃ for the indicated times and RhoA activity was determined by GST pulldown. Normalized RhoAGTP levels are expressed as the mean ± SD (n = 3). (B) 1,25(OH)₂D₃ does not modulate Rac or Cdc42. Levels of active Rac (RacGTP) and Cdc42 (Cdc42GTP) were determined by GST pulldown in cells after 1,25(OH)₂D₃ addition. (C) Scheme of biochemical routes triggered by RhoAGTP and sites of inhibition by C3 exoenzyme and Y27632. (D) Cells were treated with 1,25(OH)₂D₃ for the indicated times and the level of phosphocofilin (p-cofilin) and phospho-PRK2 (p-PRK2) were determined by WB. Normalized p-cofilin levels are expressed as the mean ± SD (n = 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2.

1,25(OH)₂D₃ induces Ca²⁺ influx in SW480-ADH cells. (A) SW480-ADH cells were loaded with fura2/AM, perfused with external medium, and treated with 1,25(OH)₂D₃ (4 × 10⁻⁷ M) or vehicle at the times indicated, and the [Ca²⁺]_cyt was estimated by fluorescence imaging. Records are mean ± SEM of 19–27 cells representative of six independent experiments. Insets show fluorescence images coded in pseudocolor of fura2/AM-loaded SW480-ADH cells before and after stimulation with 1,25(OH)₂D₃. (B) Cells were incubated in normal or in Ca²⁺-free medium and treated with 1,25(OH)₂D₃ as indicated. Data of [Ca²⁺]_cyt are the mean ± SEM of 19 cells representative of three independent experiments. (C) SW480-ADH and SW480-R cells were incubated with 1,25(OH)₂D₃ or vehicle as indicated. The increase in [Ca²⁺]_cyt (right) corresponds to the maximum detected along the stimulation period for 211 and 169 individual cells studied in six independent experiments for each cell type. The mean increase in untreated cells during a similar period was subtracted. (D) IEC18 cells were loaded with fura2/AM and treated with vehicle or 1,25(OH)₂D₃ as indicated. Records are mean ± SEM of 33 and 28 cells, respectively, representative of two independent experiments. (E) SW480-ADH cells were incubated with 1,25(OH)₂D₃, lysophosphatidic acid (LPA), or the corresponding vehicle for 1 h in normal or in Ca²⁺-free medium. Normalized RhoAGTP levels are expressed as the mean ± SD (n = 3). (F) SW480-ADH cells were incubated with 1 μM nimodipine (left) or 20 μM LaCl₃ (right) and then with 1,25(OH)₂D₃ as indicated. Ca²⁺ measurements are mean ± SEM of 24 cells representative of two independent experiments. (G) Cells were preincubated with 2 μg/ml actinomycin D (ActD; left) or 75 μM DRB (right) for 30 min before 1,25(OH)₂D₃ treatment. Ca²⁺ measurements are mean ± SEM of 30 cells representative of three independent experiments. (H) Cells were treated with ActD or DRB as in G, and with 1,25(OH)₂D₃ or vehicle for 1 h. Normalized RhoAGTP levels are expressed as the mean ± SD (n = 3). **, P < 0.01; ***, P < 0.001.

Figure 3.

RhoA–ROCK activation is necessary for the induction of adhesive phenotype by 1,25(OH)₂D₃. (A) Phase-contrast microscopy of cells pretreated with 2 μg/ml of C3 exoenzyme for 2 h and incubated with 1,25(OH)₂D₃ or vehicle for an additional 24 h. (B) Phase-contrast microscopy of both SW480-ADH cells stably expressing exogenous mouse E-cadherin (SW480-ADH-E-cadherin) and SW480-R cells incubated with 1,25(OH)₂D₃ or vehicle in the presence or absence of 10 μM Y27632 for 48 h. (C) Phase-contrast micrographs of mock cells pretreated for 4 h with 10 μM Y27632 or vehicle and of N19-RhoA cells upon incubation with 1,25(OH)₂D₃ or vehicle for 48 h. (D) Confocal laser microscopy images of mock and N19-RhoA cells treated as in C. Costaining for the localization of RhoA (green) and F-actin (red). Merged images are also shown. All scanned, phase-contrast, and confocal microscopy images are representative of at least three independent experiments. Bars, 10 μm.

Figure 4.

RhoA–ROCK activation is required for the induction of E-cadherin expression by 1,25(OH)₂D₃. (A) SW480-ADH cells were pretreated with 2 μg/ml of C3 exoenzyme or vehicle for 2 h before addition of 1,25(OH)₂D₃ or vehicle (4 h), and the level of E-cadherin RNA was determined by qRT-PCR. (B) Mock and N19-RhoA cells were treated with 1,25(OH)₂D₃ as indicated and the level of E-cadherin RNA was determined as in A. (C) SW480-ADH cells were pretreated or not with 10 μM Y27632 for 4 h and then with 1,25(OH)₂D₃ or vehicle for an additional 4 h, and the level of E-cadherin RNA was determined as in A. The data in A–C are expressed as the mean ± SD (three independent experiments performed in triplicate). (D) SW480-ADH cells were pretreated with C3 exoenzyme (2 h) and then incubated with vehicle or 1,25(OH)₂D₃ for an additional 20 h, and the level of E-cadherin protein was assessed by WB. Mean ± SD (n = 3). (E) Mock and N19-RhoA cells were incubated with 1,25(OH)₂D₃ or vehicle (24 h) in the presence or absence of Y27632, and the expression of E-cadherin protein was assessed by WB. Mean ± SD (n = 3). (F) Mock and N19-RhoA cells were transiently transfected with the plasmid encoding a fragment of the human E-cadherin gene promoter. After overnight incubation they were treated with Y27632 (4 h) and then incubated with 1,25(OH)₂D₃ or vehicle (48 h). Mean ± SD (n = 3); r.l.u., relative luciferase units. (G) Confocal laser microscopy images showing the immunolocalization of E-cadherin in mock cells pretreated or not with Y27632 (4 h) and in N19-RhoA cells incubated with 1,25(OH)₂D₃ or vehicle (48 h). Bar, 10 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

RhoA–ROCK mediates the induction of CYP24 and other 1,25(OH)₂D₃ target genes. (A) SW480-ADH and -R cells were transiently cotransfected with a plasmid encoding VDR and with a construct containing a fragment of the human CYP24 promoter-luciferase construct. After overnight incubation the cells were treated with 2 μg/ml of C3 exoenzyme for 2 h as indicated before addition (24 h) of 1,25(OH)₂D₃ or vehicle. (B) Mock and N19-RhoA cells were transfected with the plasmid encoding a fragment of the human CYP24 promoter-luciferase construct. After overnight incubation they were treated with 1,25(OH)₂D₃ or vehicle for 6 or 9 h (left). Mock cells were similarly transfected and then incubated with 1,25(OH)₂D₃ or vehicle in the presence or absence of 10 μM Y27632 for 4 h (right). (C) Mock or N19-RhoA cells were incubated with 1,25(OH)₂D₃ or vehicle (4 h) and the level of CYP24 RNA was determined by qRT-PCR (left). Mock cells were incubated with 1,25(OH)₂D₃ or vehicle in the presence or absence of 10 μM Y27632 for 4 h and the level of CYP24 RNA was determined (right). (D) Mock or N19-RhoA SW480-ADH cells were incubated with 1,25(OH)₂D₃ or vehicle for the indicated times and the expression of OPN, OCN, and CYP3A RNA was determined by qRT-PCR. The data in A–D are expressed as the mean ± SD (three independent experiments performed in triplicate). (E) Mock or N19-RhoA SW480-ADH cells were incubated with 1,25(OH)₂D₃ or vehicle for 4 (p21^CIP1) or 48 h (all others), and the expression of integrin α3, ZO-1, DKK-1, paxillin, and p21^CIP1 proteins were determined by WB. Controls: β-tubulin and -actin. The numbers below tracks represent the fold increase values in 1,25(OH)₂D₃-treated versus vehicle-treated cells. (F) Confocal laser microscopy images of mock and N19-RhoA cells incubated with 1,25(OH)₂D₃ or vehicle (48 h) illustrating the localization of ZO-1 (green) and F-actin (red). Bar, 10 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 6.

RhoA–ROCK activation mediates the redistribution of α-, β-, and p120-catenins and the inhibition of β-catenin–TCF transcriptional activity by 1,25(OH)₂D₃. (A) Mock and N19-RhoA cells were incubated with 1,25(OH)₂D₃ for the indicated times and the formation of complexes between E-cadherin and β-catenin was assessed by coimmunoprecipitation (IP). Cell extracts were first immunoprecipitated using an anti–β-catenin antibody and the presence of E-cadherin and β-catenin (as control) in the immunoprecipitates was determined by WB. The IP was repeated four times with similar results. (B) Y27632 and N19-RhoA inhibit the redistribution toward the plasma membrane compartment of α-, β-, and p120-catenins induced by 1,25(OH)₂D₃. Confocal laser microscopy images of mock (pretreated or not with 10 μM Y27632) and N19-RhoA cells incubated with 1,25(OH)₂D₃ or vehicle for 48 h. α-, β-, and p120-catenins were stained by immunofluorescence using specific antibodies and F-actin by phalloidin-rhodamine labeling, respectively. The images are representative of at least three independent experiments. Bar, 10 μm. (C) Mock and N19-RhoA SW480-ADH cells were transiently transfected with wild-type (TOP) or mutant (FOP) luciferase-base reporter plasmids for the transcriptional activity of β-catenin–TCF complexes. After overnight incubation, the cells were treated with 1,25(OH)₂D₃ or vehicle for 48 h in the presence or absence of 10 μM Y27632 and the intracellular luciferase activity was determined. β-catenin–TCF transcriptional activity (TOP/FOP ratio) is expressed as the mean ± SD (n = 3). (D) Mock and N19-RhoA cells were incubated with 1,25(OH)₂D₃ or vehicle for the indicated times and the number of cells in the cultures was determined. The data represent one out of the four independent experiments performed in triplicate. ***, P < 0.001.

Figure 7.

p38MAPK and MSK1 mediate the induction of E-cadherin and CYP24 by 1,25(OH)₂D₃. (A) SW480-ADH cells were pretreated with either vehicle, 3.5 μM GF109203X, or 2 μM Ro318220 for 2 h before incubation with 1,25(OH)₂D₃ or vehicle for 20 h, and the level of E-cadherin protein and phosphoprotein kinase D (p-PKD) were determined by WB. Mean ± SD (n = 3). (B) Cells were pretreated with either vehicle, 10 μM H89, or 20 μM Rp-cAMP for 2 h before incubation with 1,25(OH)₂D₃ for 20 h and E-cadherin protein was determined by WB. Quantification was as in A. (C) Cells were pretreated with either vehicle, 20 μM U0126, or 30 μM SB203580 for 2 h before incubation with 1,25(OH)₂D₃ or vehicle for 20 h and E-cadherin protein and total and p-ERK1/2 were determined by WB. Quantification was as in A. (D) Cells were incubated (4 h) with 1,25(OH)₂D₃ or vehicle in the presence or absence of the indicated kinase inhibitor (same doses as in A–C) and E-cadherin RNA was determined by qRT-PCR (three independent experiments performed in triplicate). (E) SW480-ADH cells were transfected with the plasmid encoding a fragment of the human E-cadherin gene promoter. After overnight incubation they were treated (48 h) with 1,25(OH)₂D₃ or vehicle in the presence or absence of 1 μM Ro318220. Mean ± SD (n = 3). (F) SW480-ADH cells were incubated and treated as in D and the level of CYP24 RNA was determined and quantificated as in D. (G) Cells were treated with 1,25(OH)₂D₃ or vehicle and the expression of total and p-p38MAPK and p-MSK1 was determined by WB. Mean ± SD (n = 3). (H) SW480-ADH cells were transfected with MSK1 or MSK2 siRNA oligonucleotides or with scrambled oligos as control (C). 1 d later cells were treated with 1,25(OH)₂D₃ or vehicle for an additional 6 h. E-cadherin, MSK1, and MSK2 protein levels were determined by WB. Mean ± SD (n = 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 8.

VDR knockdown abrogates the induction of nongenomic signaling and gene expression by 1,25(OH)₂D₃. (A) Phase-contrast images of SW480-ADH cells stably expressing control or VDR shRNA treated with 1,25(OH)₂D₃ or vehicle for 48 h as indicated. Bar, 10 μm. (B) Expression of VDR and E-cadherin proteins (left) and CYP24 promoter activity (right) in control and VDR shRNA cells that were treated or not with 1,25(OH)₂D₃ for 24 h. Protein levels were determined by WB using β-actin as control and CYP24 promoter activity was analyzed as in legend to Fig. 5. (C) Effect of VDR knockdown on the increase in [Ca²⁺]_cyt induced by 1,25(OH)₂D₃. The three cell types were loaded with fura2/AM and assessed for responsiveness to 1,25(OH)₂D₃. Traces are mean ± SEM values of control cells (black) or the two VDR shRNA lines (red and blue traces). Data are representative of at least two independent experiments for each cell line. (D) Control and VDR-4 shRNA cells were treated with 1,25(OH)₂D₃ or vehicle for 1 h and RhoA activity was determined by GST pulldown. Normalized RhoAGTP levels are expressed as the mean ± SD (n = 3). (E) Control and VDR-4 shRNA cells were treated with 1,25(OH)₂D₃ or vehicle for 2 h and the level of phospho- and total p38MAPK was determined by WB (mean ± SD, n = 3). (F) Control and VDR-4 shRNA cells were treated with 1,25(OH)₂D₃ or vehicle for 2 h and the level of phospho-MSK1 (p-MSK1) and phospho-CREB (p-CREB) in nuclear (N) and cytosolic (C) fractions was determined by WB. Total CREB and lamin B and β-tubulin were used as respective controls. *, P < 0.05; ***, P < 0.001.

Figure 9.

p38MAPK-MSK1 activation by 1,25(OH)₂D₃ depends on RhoA–ROCK and is necessary for the interference of the Wnt–β-catenin pathway. (A) SW480-ADH cells were incubated with 1,25(OH)₂D₃ in the presence of 2 μM Ro318220 or vehicle. The [Ca²⁺]_cyt was determined after 1,25(OH)₂D₃ addition. Mean ± SEM of 28–30 cells (three independent experiments). (B) Mock and N19-RhoA cells were incubated with 1,25(OH)₂D₃ or vehicle (2 h) and the levels of total and phospho-p38MAPK, -MSK1, -CREB, and -ATF1 were determined by WB. Mean ± SD (n = 3). (C) Mock and N19-RhoA cells were pretreated with vehicle, 30 μM Ro318220, or 30 μM SB203580 (2 h) and then incubated with 1,25(OH)₂D₃ for additional 4 h, and the level of E-cadherin RNA was determined by qRT-PCR. Mean ± SD (three independent experiments performed in triplicate). (D) Mock and N19-RhoA cells were incubated (4 h) with 1,25(OH)₂D₃ in the presence of vehicle, Ro318220, or SB203580, and the level of CYP24 RNA and quantification was determined as in C. (E) SW480-ADH cells were transfected with wild-type (TOP) or mutant (FOP) reporter plasmids for β-catenin–TCF activity. After overnight incubation, the cells were treated (48 h) with either vehicle or 1,25(OH)₂D₃ in the presence or absence of 1 μM Ro318220. Mean ± SD (n = 3). (F) Scheme of the mechanism of action of 1,25(OH)₂D₃ in human SW480-ADH cells. A rapid, VDR-dependent nongenomic signaling pathway that starts with Ca²⁺ influx and continues with the sequential activation of RhoA GTPase and the kinases ROCK, p38MAPK, and MSK1 converges in the nucleus with ligand-activated VDR to regulate gene expression and interfere with the Wnt–β-catenin pathway leading to proliferation arrest and epithelial differentiation. MSK1 may target VDR and/or its coregulators and/or downstream transcription factors. *, P < 0.05; **, P < 0.01; ***, P < 0.001.