A role for the age-dependent loss of α(E)-catenin in regulation of N-cadherin expression and cell migration

Abstract

The aging kidney has a decreased ability to repair following acute kidney injury. Previous studies from our laboratory have demonstrated a loss in α-catenin expression in the aging rat kidney. We hypothesize that loss of α-catenin expression in tubular epithelial cells may induce changes that result in a decreased repair capacity. In these studies, we demonstrate that decreased α-catenin protein expression is detectable as early as 20 months of age in male Fischer 344 rats. Protein loss is also observed in aged nonhuman primate kidneys, suggesting that this is not a species-specific response. In an effort to elucidate alterations due to the loss of α-catenin, we generated NRK-52E cell lines with stable knockdown of α(E)-catenin (C2 cells). Interestingly, C2 cells had decreased expression of N-cadherin, decreased cell-cell adhesion, and increased monolayer permeability. C2 had deficits in wound repair, due to alterations in cell migration. Analysis of gene expression in the migrating control cells indicated that expression of N-cadherin and N-CAM was increased during repair. In migrating C2 cells, expression of N-CAM was also increased, but the expression of N-cadherin was not upregulated. Importantly, a blocking antibody against N-cadherin inhibited repair in NRK-52E cells, suggesting an important role in repair. Taken together, these data suggest that loss of α-catenin, and the subsequent downregulation of N-cadherin expression, is a mechanism underlying the decreased migration of tubular epithelial cells that contributes to the inability of the aging kidney to repair following injury.

Keywords: Aging; N‐cadherin; migration; wound repair; α‐catenin.

Figure 1.

The age‐dependent loss of α‐catenin. (A) Protein expression of α‐catenin was determined by western blot analysis of cortical lysates from male Fischer 344 rats at 4, 20, and 24 months; each lane represents a sample from an individual animal. (B) Densitometric analysis of α‐catenin expression, normalized to β‐actin, is shown; n = 3–7 animals, *indicates a significant difference from 4 months. (C) Age‐dependent loss of α‐catenin is also seen in the aging nonhuman primate kidney, as shown by western blot of kidney lysates.

Figure 2.

Characterization of stable knockdown of α(E)‐catenin in NRK‐52E cells. Several shRNA constructs targeting α(E)‐catenin were designed and cell lines were commercially generated that demonstrated varying levels of α(E)‐catenin knockdown at the gene (A) and protein level (B). (C) Loss of α(E)‐catenin gene expression in C2 cells is seen as compared to the vector control cell line (NT3). Neither NT3 or C2 cells expressed α(N)‐catenin, and the expression of α(T)‐catenin or α‐catulin was not significantly increased in C2 cells. (D) A corresponding loss of α‐catenin protein expression is seen in C2 cells via western blot analysis. The C2 cells also have decreased β‐catenin and P‐cadherin expression, and almost complete loss of N‐cadherin expression. (E) Analysis of cell aggregation over time demonstrated that a decrease in the number of large cell–cell aggregates (>51 cells/aggregate) in C2 cells as compared to NT3 cells; *indicates a significant difference as compared to NT3 cells. (F) C2 cell monolayers also had increased permeability of FITC‐labeled albumin; *indicates a significant difference in fluorescence as compared to NRK‐52E cells (n = 6). Cell proliferation was assessed in serum (G) and serum‐free (H) conditions; each data point represents the fold‐increase over the T0 time point; n = 18 from three independent experiments; *indicates a significant difference from NT3.

Figure 3.

Loss of α‐catenin expression decreases N‐cadherin expression. (A) A loss of N‐cadherin protein expression in cortical lysates from male Fischer 344 rats at 20, or 24 months is shown; each lane represents a sample from an individual animal. (B) Densitometric analysis of N‐cadherin expression, normalized to β‐actin; n = 3–7 animals, *indicates a significant difference from 4 months. (C) The age‐dependent loss of N‐cadherin is also seen in the aging nonhuman primate kidney, as shown by western blot of kidney lysates in the bottom panel. (D) The expression of Twist1 is decreased in C2 cells as compared to NT3 controls; similar results were seen in independent experiments.

Figure 4.

Loss of α(E)‐catenin reduces wound healing. Repair of a scratch in monolayers of C2 cells was inhibited as compared to NT3 cells in serum‐free (SF) conditions. (A) Representative images of NT3 and C2 cells 24 h following injury in serum‐free conditions; the inset in each image is the T0 time point. (B) Quantitative assessment of wound healing at 24 h is shown in both serum and serum‐free conditions; *indicates a statistically significant difference between C2 and NT3, or C2 SF and NT3 SF (n = 6). (C) Gene expression of E‐cadherin, N‐cadherin and N‐CAM in control and migrating NT3 and C2 cells. A similar pattern is seen for E‐cadherin and N‐CAM in migrating NT3 and C2 cells, that is, a decrease in E‐cadherin and increase in N‐CAM in both cells. However, migrating NT3 cells exhibit an increase in N‐cadherin gene expression that is not seen in C2 cells. (D) Inhibition of N‐cadherin via a blocking antibody (GC‐4) inhibits repair at 24 h in 5% FBS or serum‐free media.