Comparison of the cytotoxic effects of cadmium (Cd²⁺) in high and low resistance strains of MDCK cells that express different levels of E-Cadherin

Abstract

Previous studies from our laboratory have shown that cadmium (Cd²⁺) can disrupt the adhering and occluding junctions between MDCK cells. Recently, we have obtained evidence to suggest that Cd²⁺ produces this effect by interacting with E-cadherin, a Ca²⁺-dependent cell adhesion molecule that is localized at the adhering junctions of epithelial cells. The objective of the present study was to examine the junctional and cytotoxic effects of Cd²⁺ in subcloned strains of MDCK cells that express different levels of E-cadherin. One strain (MDCK I) expresses high levels of E-cadherin and develops a transepithelial electrical resistance of more than 800 Ω·cm², whereas the other strain (MDCK II) expresses much lower levels of E-cadherin and develops a transepithelial resistance of less than 100 Ω·cm². The results showed that exposure to 20 μm Cd²⁺ for 2–4 hours caused a pronounced loss of E-cadherin from the cell borders in both strains of cells. In the MDCK I cells, the loss of E-cadherin coincided with a decrease in the transepithelial electrical resistance and the loss of the tight junction-associated proteins, ZO-1 and occludin, from the cell borders. By contrast, the MDCK II cells first exhibited a significant increase in the transepithelial electrical resistance that did not begin to decline until the cells had been exposed for 4–6 hours, a time that coincided with the loss of ZO-1 and occludin from the cell borders. Additional results showed that the MDCK I cells were slightly more sensitive to the lethal effects of Cd²⁺ than were the MDCK II cells. These findings indicate that E-cadherin may be an early target for Cd²⁺ toxicity in both high and low resistance strains of MDCK cells. However, they also suggest that the disruption of E-cadherin-dependent cell–cell junctions may trigger somewhat different responses in the two cell lines.

Introduction

Cadmium (Cd²⁺) is an important industrial and environmental pollutant that has been shown to cause severe damage to a variety of organs, including the lung, liver, kidney, testis and placenta (for reviews see Foulkes, 1986; Friberg et al., 1986; Morselt, 1991). In addition, Cd²⁺ has been shown to have teratogenic and carcinogenic activities (Elinder and Kjellstrom, 1986). Although these general toxic effects of Cd²⁺ have been fairly well characterized, the specific mechanisms underlying many of these effects have yet to be elucidated. In this regard, the recent findings from our laboratory showing that Cd²⁺ can disrupt the Ca²⁺-dependent junctions between the renal epithelial cells in culture may be particularly important.

Cells of transporting epithelia and vascular endothelia are attached to each other by specialized junctional complexes that are necessary for the restriction of permeability and the normal transport of materials across the cell monolayer (for reviews see Alberts et al., 1994; Boyer and Thiery, 1989; Woods and Bryant, 1993). These junctional complexes include the zonulae occludens (tight or occluding junctions), the zonula adherens (adhering junctions or belt-like desmosomes), the macula adherens (spot-like desmosomes) and gap junctions. The complexes are composed of specific junction-associated proteins such as vinculin, the catenins, the cadherins, the connexins, etc., and are closely associated with actin filaments and other cytoskeletal elements of the individual cells (Alberts et al., 1994; Boyer and Thiery, 1989; Geiger, 1983, Madara et al., 1988; Meza et al., 1980, Meza et al., 1983).

Several years ago, we noticed that many of the effects of Cd²⁺ in vivo appeared to involve an increase in the permeability of various endothelial and epithelial surfaces (Prozialeck and Niewenhuis, 1991a). To further explore this issue, we began a series of studies examining the effects of Cd²⁺ on the barrier characteristics of several epithelial cell lines in culture. The model systems that were employed for most of these studies included LLC-PK₁ cells, an established renal epithelial cell line derived from pig kidney, and MDCK cells, an immortal cell line derived from canine kidney (Gstraunthaler et al., 1990) . Results of these studies showed that Cd²⁺ can selectively damage the Ca²⁺-dependent junctions between the cells. Exposure to micromolar concentrations of Cd²⁺ for 1–4 hr caused the cells to separate from each other without killing them (Prozialeck and Niewenhuis, 1991a, Prozialeck and Niewenhuis, 1991b; Prozialeck and Lamar, 1997), or altering levels of ATP (Prozialeck and Niewenhuis, 1991a) or glutathione (Prozialeck and Lamar, 1995). This effect coincided with a drop in the transepithelial electrical resistance, a reorganization of the actin cytoskeleton, and changes in the structure of the adhering and occluding junctional complexes (Prozialeck and Niewenhuis, 1991a; Niewenhuis et al., 1997).

Evidence from these previous studies suggests that Cd²⁺ produces its junctional effects by interacting with a Ca²⁺-sensitive site on the basolateral cell surface (Prozialeck and Niewenhuis, 1991b; Prozialeck and Lamar, 1993, Prozialeck and Lamar, 1997). In considering potential candidates for this site, we feel that a likely possibility is the Ca²⁺-dependent cell adhesion molecule, E-cadherin. E-cadherin is an integral, transmembrane, Ca²⁺-binding glycoprotein that belongs to the cadherins family of Ca²⁺-dependent cell adhesion molecules (Grunwald, 1996; Takeichi, 1990). In epithelial cells, E-cadherin is primarily localized at the adhesion belts of the adhering junctional complexes (zonulae adherens) where it plays a key role in homophilic, Ca²⁺-dependent cell–cell adhesion (Boller et al., 1985; Grunwald, 1996; Nelson et al., 1990; Ringwald et al., 1987; Yap et al., 1997). The protein contains an intracellular domain that is linked to the actin cytoskeleton through a group of molecules called catenins, a transmembrane domain, and an extracellular domain that contains the putative Ca²⁺-binding sites, as well as the adhesive regions of the molecule (Nelson et al., 1990; Overduin et al., 1995; Ozawa et al., 1990; Shirayoshi et al., 1986; Takeichi, 1990).

The evidence that E-cadherin is the molecular target on which Cd²⁺ acts to disrupt epithelial cell–cell junctions may be summarized as follows. First, the initial junction-perturbing effects of Cd²⁺ appear to involve the adhering junctions (Niewenhuis et al., 1997; Prozialeck and Niewenhuis, 1991a). As was noted previously, E-cadherin is closely associated with adhering junctions of epithelial cells (Boller et al., 1985). Secondly, the disruption of intercellular junctions by Cd²⁺ is much more pronounced when Cd²⁺ is added to the basolateral cell surface than when it is added to the apical surface (Prozialeck and Lamar, 1997; Prozialeck and Niewenhuis, 1991b), indicating that Cd²⁺ is acting at a site that is located on the basolateral side of the occluding junctions. Thirdly, the severity of the disruption of intercellular junctions by Cd²⁺ depends on the concentration of Ca²⁺ in the incubation medium. The effects of Cd²⁺ are more pronounced when Ca²⁺ is present at low concentrations and are greatly attenuated when Ca²⁺ is present at high concentrations (Prozialeck and Lamar, 1997; Prozialeck and Niewenhuis, 1991b). The interaction between Ca²⁺ and Cd²⁺ appears to be competitive, suggesting that Cd²⁺ may be competing with Ca²⁺ for binding sites, such as those that are present on the E-cadherin molecule. Fourthly, exposure to sublethal concentrations of Cd²⁺ causes a pronounced decrease in the amount of E-cadherin that is associated with the contacts between LLC-PK₁ (Prozialeck and Niewenhuis, 1991b) and MDCK (Prozialeck and Lamar, 1997) cells, an effect that is qualitatively similar to that observed when the cells are incubated in Ca²⁺-free media. Fifthly, we have recently found that Cd²⁺ can bind to E-CAD1, a recombinant, 145-residue polypeptide that corresponds to one of the extracellular Ca²⁺-binding domains of mouse E-cadherin (Prozialeck et al., 1996).

In the light of evidence that E-cadherin may be a direct molecular target for Cd²⁺ toxicity in epithelial cells, we felt that it would be of interest to examine the cytotoxic effects of Cd²⁺ in cell lines that express different levels of E-cadherin. We were particularly interested in determining whether cells that express different levels of E-cadherin might exhibit different cytotoxic responses to Cd²⁺. In addition, we wanted to determine whether the disruption of E-cadherin-dependent cell–cell junctions occurs independently of other cytotoxic actions of Cd²⁺, or whether it is part of a cascade of events leading to more severe cellular injury and cell death. To address these issues, we have examined the junctional and cytotoxic effects of Cd²⁺ in subcloned strains of MDCK cells (Richardson et al., 1981; Stevenson et al., 1988) that express different levels of E-cadherin. One strain, termed MDCK I, expresses high levels of E-cadherin and develops a transepithelial electrical resistance of more than 800 Ω·cm², whereas the other strain termed MDCK II, expresses much lower levels of E-cadherin and develops a TER of less than 100 Ω·cm² (Collares-Buzato et al., 1994; Stevenson et al., 1988).