Morphological Evaluation Of Copper Effects On Cultured Human Intestinal Caco2 Cells

Several toxic stimuli resulting in an increase in tight-junction permeability in Caco-2 cells, including xenobiotics and natural toxins, have been shown to induce depolymerization of the F-actin cytoskeleton (5,27,53,59,60) and perturbation of tight-junction proteins (34,52,53,61). We have therefore investigated the morphological effect of copper in Caco-2 cells on the F-actin cytoskeleton and on junctional proteins by fluorescent microscopy and by confocal laser scanning microscopy (CLSM) and on the cell ultrastructure by transmission electron microscopy (TEM).

4.1. Fluorescent Labeling of F-Actin and Nuclei and Localization of Junctional Proteins by Immunofluorescence

4.1.1. Materials and Methods

For morphological studies, cells were seeded on cell culture chamber inserts fitted with transparent membranes (P.E.T. track-etched membrane, 25 mm diameter, 4.71 cm2 area, 0.4 pm pore diameter; Becton Dickinson Labware Europe, Meylan Cedex, France). Cells were treated from the AP compartment with 50-300 pM CuCl2 or with 200 pM Cu(His)2 in HBSS, pH 6.0 for 3 h at 37°C, and after rinsing with phosphate-buffered saline containing 1 mM CaCl2 and 1 mM MgCl2 (PBS+), they

(Continued from Fig. 6 caption)

triplicate filters from a representative experiment. (C) Caco-2 cells were treated with 30 pM CuCl2 or 30 pM Cu(Gly)2 for 3 h at pH 6.0, with or without 10 mM Gly-Gly or Gly-L-Leu in the AP compartment and the TEER was measured after treatment. Data, expressed as the percentage of their respective control, are the means ± SD of triplicate filters from a representative experiment.

Fig. 7. Fluorescence microscopy of F-actin localized by FITC-phalloidin binding (A, C, E) and of nuclei stained with propidium iodide (B, D, F) in Caco-2 cells either untreated (A, B) or treated from the AP side for 3 h at pH 6.0 with 50 ||M (C, D) or 300 ||M (E, F) CuCl2. Bars = 25 ||M.

were fixed with 2.5% paraformaldehyde in PBS+. Free aldehydes were quenched using 50 mM NH4Cl in PBS+. For immunofluorescent localization of junctional proteins, cells were permeabilized with 0.075% saponin in PBS+ and the cells were treated with primary antibody and secondary tetramethylrodamine isothiocyanate (TRITC)-conjugated antibody according to conventional techniques. The mouse monoclonal anti-E-cadherin was supplied by Zymed Laboratories Inc. (San Francisco, CA) and the secondary TRITC-conjugated affinity purified goat anti-mouse IgG was from Cappel, (Organon Tecknika Co., Durham, NC). For-F-actin localization cells were incubated for 30 min with 1.7 |ig/mL phalloidin (1.28 ||M) conjugated with fluorescein isothiocyanate (FITC) in PBS+ containing 0.075% saponin and 0.2% BSA. After rinsing, filters were mounted in Vectashield (Vector Laboratories, Burlingame, CA). When labeling of nuclei was required, before mounting, the filters were incubated for 30 min at 37°C in 20 |ig/mL RNAse A (from bovine pancreas [Boehringer Mannheim

Fig. 8. Immunofluorescence microscopy of the adherens junction protein E-cadherin (A, C) and double staining of F-actin by FITC-phalloidin binding (B, D) of control cells (A, B) and cells treated for 3 h with 50 ||M CuCl2 (C, D). In areas where F-actin staining was strongly decreased by copper treatment (D), E-cadherin distribution (C) appeared similar to that in control cells (A). Bars = 10 ||M.

Fig. 8. Immunofluorescence microscopy of the adherens junction protein E-cadherin (A, C) and double staining of F-actin by FITC-phalloidin binding (B, D) of control cells (A, B) and cells treated for 3 h with 50 ||M CuCl2 (C, D). In areas where F-actin staining was strongly decreased by copper treatment (D), E-cadherin distribution (C) appeared similar to that in control cells (A). Bars = 10 ||M.

Italia, Monza, Italy], previously boiled for 10 min to denature DNAse), rinsed in PBS+ and stained with 0.001% propidium iodide (Sigma-Aldrich Srl, Gallarate, Milan, Italy) in PBS+ for 10 s and rinsed extensively. The cells were viewed with a fluorescent microscope (Axioscope 2; C. Zeiss, Jena, Germany) or with a confocal laser microscope (CLSM Sarastro 2000; Molecular Dynamics, Sunnyvale, CA). Vertical and horizontal scans were reconstructed through Image Space software (Molecular Dynamics). In particular, images were collected either with the usual scanning method (i.e. laser scans of subsequent x-y-planes at various z-positions in the specimen) or with cross-sections through datasets along the x-z-plane (i.e., vertical scans), which provide a direct insight into the axial extension of the scanned structures.

4.1.2. Results and Discussion

Alterations in tight-junction permeability are often associated with changes at the level of the actin cytoskeleton. We therefore investigated the distribution of F-actin in control and in copper-treated cells by staining with fluorescent phalloidin. Figure 7A shows the organization of F-actin in control cells, whereas Fig. 7B shows the nuclei in the same microscopic field stained with propidium iodide. Treatment with 50 ||M CuCl2 for 3 h led to a marked reduction of F-actin staining in small areas of the cell monolayer (Fig. 7C); nuclear staining of the same microscopic field revealed that the cell monolayer was intact following copper treatment (Fig. 7D). Increasing the concentration of copper to 300 |M CuCl2 produced large areas of the cell monolayer in which the F-actin signal was either absent or strongly reduced and highly disorganized (Fig. 7E). Even in these extreme cases however, the nuclear staining showed lack of cellular loss (Fig. 7F).

Copper Toxicity in Intestinal Cells 409

wktfmm

0 0

Post a comment