Effect of claudins 6 and 9 on paracellular permeability in MDCK II cells

Abstract

The neonatal proximal tubule has a lower permeability to chloride, higher resistance, and higher relative sodium-to-chloride permeability (P(Na)/P(Cl)) than the adult tubule, which may be due to maturational changes in the tight junction. Claudins are tight-junction proteins between epithelial cells that determine paracellular permeability characteristics of epithelia. We have previously described the presence of two claudin isoforms, claudins 6 and 9, in the neonatal proximal tubule and subsequent reduction of these claudins during postnatal maturation. The question is whether changes in claudin expression are related to changes in functional characteristics in the neonatal tubule. We transfected claudins 6 and 9 into Madin-Darby canine kidney II (MDCK II) cells and performed electrophysiological studies to determine the resultant changes in physiological characteristics of the cells. Expression of claudins 6 and 9 resulted in an increased transepithelial resistance, decreased chloride permeability, and decreased P(Na)/P(Cl) and P(HCO3)/P(Cl). These findings constitute the first characterization of the permeability characteristics of claudins 6 and 9 in a cell model and may explain why the neonatal proximal tubule has lower permeability to chloride and higher resistance than the adult proximal tubule.

Fig. 1.

Effect of claudins 6 or 9 transfection on expression of endogenous claudins in Madin-Darby canine kidney II (MDCK II) cells. Representative immunoblots of whole cell lysate (60 μg per lane) from empty vector-transfected and claudins 6- or 9-transfected MDCK II cells (n = 4). Cldn, claudin.

Fig. 2.

Localization of transfected claudins 6 and 9 in MDCK II cells. Three days posttransfection, filters with confluent cell monolayers were labeled with claudin 6 or 9 antibody and the appropriate secondary antibody. Control mock-transfected cells are shown in the bottom 2 panels. Claudin 6 and 9 protein abundance quantified using densitometry on immunoblot. This graph represents relative protein abundance of claudins 6 and 9 compared with β-actin. Bars and error bars represent means and SD (n = 4 for each claudin, *P < 0.05).

Fig. 3.

Transepithelial resistance (TER) in MDCK II cells transfected with claudin 6 or 9. TER was measured across cell monolayers in a modified Ussing chamber. Results expressed as TER in Ω·cm² (n = 8 for each sample). Bars and error bars represent means ± SE. *P < 0.05.

Fig. 4.

Representative voltage tracing showing transepithelial potential differences across a MDCK II cell monolayer using asymmetric solutions with a bracketed method. NaCl and HCO₃⁻ concentrations were changed in apical or basal chambers bathing MDCK II cell monolayers. Black solid line represents empty vector-transfected cells; dashed line represents claudin 6-transfected cells. NaCl and HCO₃⁻ concentrations for each segment are indicated at the top of the figure.

Fig. 5.

Sodium permeability-to-chloride permeability ratio (P_Na/P_Cl) in MDCK II cells transfected with claudin 6 or 9. The relative ionic permeability was calculated using the Goldman-Hodgkin-Katz equation based on the transepithelial potential difference across each cell monolayer with each replacement solution (n = 8 for each sample). Bars and error bars represent means ± SE. *P < 0.05.

Fig. 6.

Bicarbonate permeability-to-chloride permeability (P_HCO3/P_Cl) ratio in MDCK II cells transfected with claudin 6 or 9. The relative ionic permeability was calculated using the Goldman-Hodgkin-Katz equation based on the transepithelial potential difference across each cell monolayer with each replacement solution (n = 8 for each sample). Bars and error bars represent means ± SE. *P < 0.05.

Fig. 7.

Chloride permeability after application of concentration gradient across MDCK II cells transfected with claudin 6 or 9. A chloride concentration gradient was applied with 111 mmol chloride solution on the apical side and chloride-free solution on the basal side at 4°C. Samples were taken from the basal side at time 0 and 1 h, and the chloride concentration was measured. Permeability was calculated and is represented as mmol/M·h·cm² on this graph (n = 8 for each sample). Bars and error bars represent means ± SE. *P < 0.05 by ANOVA.