Altered paracellular cation permeability due to a rare CLDN10B variant causes anhidrosis and kidney damage

Abstract

Claudins constitute the major component of tight junctions and regulate paracellular permeability of epithelia. Claudin-10 occurs in two major isoforms that form paracellular channels with ion selectivity. We report on two families segregating an autosomal recessive disorder characterized by generalized anhidrosis, severe heat intolerance and mild kidney failure. All affected individuals carry a rare homozygous missense mutation c.144C>G, p.(N48K) specific for the claudin-10b isoform. Immunostaining of sweat glands from patients suggested that the disease is associated with reduced levels of claudin-10b in the plasma membranes and in canaliculi of the secretory portion. Expression of claudin-10b N48K in a 3D cell model of sweat secretion indicated perturbed paracellular Na+ transport. Analysis of paracellular permeability revealed that claudin-10b N48K maintained cation over anion selectivity but with a reduced general ion conductance. Furthermore, freeze fracture electron microscopy showed that claudin-10b N48K was associated with impaired tight junction strand formation and altered cis-oligomer formation. These data suggest that claudin-10b N48K causes anhidrosis and our findings are consistent with a combined effect from perturbed TJ function and increased degradation of claudin-10b N48K in the sweat glands. Furthermore, affected individuals present with Mg2+ retention, secondary hyperparathyroidism and mild kidney failure that suggest a disturbed reabsorption of cations in the kidneys. These renal-derived features recapitulate several phenotypic aspects detected in mice with kidney specific loss of both claudin-10 isoforms. Our study adds to the spectrum of phenotypes caused by tight junction proteins and demonstrates a pivotal role for claudin-10b in maintaining paracellular Na+ permeability for sweat production and kidney function.

Fig 1. The CLDN10b specific gene variant segregates anhidrosis and mild kidney failure.

(A) Pedigree of the consanguineous Pakistani family segregating anhidrosis and heat intolerance, alacrima, xerostomia, mild kidney failure and Mg²⁺ retention (black filled symbols). All affected individuals are homozygous for chromosome 13q32 marker alleles flanking the CLDN10 gene with the missense variant c.144C>G (black bars). (B) Body temperature measurements at rest over 25 min when exposed to 45°C and 45% humidity in affected (n = 2; black line) and age-matched controls (n = 3; grey line). The patients interrupted the study after 20 min due to distress. (C) The N48K variant is located within the 1^st extracellular loop (ECL1) and results in the introduction of the positively charged Lysine in the claudin consensus sequence W- G/NLW-C-C (filled amino acids). The first 73 amino acids (black circles) are specific for the claudin-10b isoform encoded by exon 1b. Grey circles correspond to amino acids shared with the claudin-10a isoform. +/- indicate charged residues in the extracellular loops. ECL–extracellular loop. TM–Transmembrane segment. (D-E) Immunostaining of the secretory portion of sweat glands from a healthy control (D) and from the affected ind. 15 (E). In the control, claudin-10 (green) is predominantly found in the canaliculi, the cell peripheries and in membranes lining the lumen. Strong occludin staining (red) is found lining the main lumen of the sweat gland. In the patient, claudin-10 signal (green) is observed in membranes facing the lumen but shows a more even intracellular distribution without accumulation in canaliculi but in spots resembling vesicles. A strong occludin signal (red) is observed both in the canaliculi and in membranes lining the main lumen. Scale bars: 20μm.

Fig 2. Reduced size of cysts formed by MDCK-C7 cells transfected by claudin-10b N48K.

(A) Histogram of the size distribution of cyst lumen diameter (grey, untransfected control, n = 151 cysts [10 different z-stacks]; red, claudin-10b WT transfected cells: clone #3, n = 140 [9], clone #39, n = 187 [11]; blue claudin-10b N48K transfected cells: clone #5, n = 167 [9]; clone # 21, n = 113 [7]). (B) Normal distribution of the data shown in (A) to visualize the shift in lumen size in the presence of claudin-10b. (C) Representative images of cysts (red, actin-staining by phalloidin-Alexa 594; blue, nuclei stained with DAPI), (D) TJ localization of N48K and WT claudin-10b in 2D culture. (E) Presence of WT and N48K claudin-10b (green) in 3D culture (co-staining as in D).

Fig 3. Claudin-10b N48K alters electrophysiological properties in MDCK-C7 cell layers.

(A) Dilution potentials (and thus the permeability ratio for Na⁺ and Cl^-; P_Na/P_Cl) of cell layers expressing claudin-10b WT (blue) and claudin-10b N48K (red) are significantly higher than those of control cell layers (black; empty vector). Cell layers transfected with claudin-10b WT maintain these properties over several weeks whereas cell layers transfected with claudin-10b N48K lose these properties over a time-course of three weeks. n = 1–28, means +/- SEM. (B) Expression of claudin-10b WT (blue symbols) cause reduced TER (x-axis) and increased dilution potentials (y-axis) compared to vector transfected controls (black symbols). The effects when expressing claudin-10b N48K (red symbols) were less pronounced. n = 8–40, means +/- SEM. (C) Permeability for monovalent cations (P_X) relative to P_Na. Expression of claudin-10b WT altered Eisenman-sequence for monovalent cations from Eisenman-sequence IV in control layers to Eisenman-sequence VIII to X, depending on transfection strength. Expression of claudin-10b N48K resulted in Eisenman-sequences between III and X. n = 2–13, means +/- SEM.

Fig 4. Perturbed formation of tight junctions and inhibition of claudin-10b trans-interactions by the N48K mutation.

(A-D) Freeze fracture electron microscopy of HEK293 cells expressing YFP-claudin-10b. (A) Stable expression of wt claudin-10b led to formation of complex mesh-works of branched tight junction strands. Numerous continuous strands (arrow) were detected on the protoplasmic face (P-face, PF) and as particle-free grooves (arrowhead) on the exoplasmic face (E-face, EF) of the plasma membrane. (B) Stable expression of claudin-10b N48K resulted in few tight junction strands and less complex meshworks. Strands were detected on the P-face as rows of rather separated intramembranous particles (black arrow) and two- dimensional particle arrays (white arrow). (C) Similar to the stable expression, transient expression of wt claudin-10b led to mesh-works of continuous tight junction strands detected on the P-face (arrow) and E-face (arrowhead). (D) For transient expression of claudin-10b N48K, tight junction mesh-works were infrequent. Particle-type strands were detected as noncontinuous rows of separated, beaded intramembranous particles (black arrow) on the P-face. Scale bars, 200 nm. (E-H) Inhibition of YFP-claudin-10b trans-interactions by the p.N48K mutation. (E) Transiently expressed claudin-10b WT (green) is enriched at contacts (arrows) between claudin-expressing HEK293 cells suggesting trans-interaction. Arrowheads indicate plasma membrane outside of contacts between two claudin-expressing cells. (F) Transient expression of claudin-10b N48K (green) in HEK293 cells does not show strong enrichment at contacts (arrows) between cells suggesting lack of trans-interaction. (G) For transient expression in HEK293 cells, the enrichment factor (Enr. F) was significantly lower for claudin-10b N48K compared to claudin-10b WT. n = 27–44; *, p < 0.01. (H) For stable expression in HEK293 cells, the product of enrichment factor and length of enrichment (Enr. F x Length) was significantly lower for claudin-10b N48K compared to claudin-10b WT. n = 27–42; *, p < 0.01. Means +/- SEM. Size bars: 5μm.

Fig 5. The claudin-10b N48K increases FRET efficiency in cis.

(A, B) FRET efficiencies for YFP-claudin-10b WT/CFP-claudin-10b WT interaction (blue), YFP-claudin-10b N48K/CFP-claudin-10b N48K (red) and YFP-claudin-10b N48K/CFP-claudin-10b WT (black) constructs overexpressed in HEK293 cells (A) and MDCK-C7 cells (B) plotted against the YFP/CFP ratio. Only values above the critical YFP/CFP ratio as defined by Milatz et al. (2015) are shown. (C) In HEK293 (** p<0.01) and in MDCK-C7 (# p<0.05) cells average FRET efficiency is significantly higher for YFP-claudin-10b N48K/CFP-claudin-10b N48K interaction than for YFP-claudin-10b WT/CFP-claudin-10b WT interaction. Interaction between claudin-10b N48K/claudin-10b WT is significantly lower than for both claudin-10b N48K/claudin-10b N48K and claudin-10b WT/claudin-10b WT (** p<0.01, only tested in HEK cells). Statistical analysis: MDCK-C7, t-test, n = 19 (WT) and 21(N48K); HEK293, t-test + Bonferroni-Holm, n = 83 (WT), 52 (N48K) and 15 (WT-N48K). Error bars are presented with ± standard error (SEM).

Fig 6. Homology modeling predicts structural alterations of claudin-10b p.N48K.

(A) Homology model of the tertiary structure of claudin-10b based on crystal structure of murine claudin-15 (PBD ID: 47P9) as template. Backbone is shown as cartoon in green and residues of the claudin consensus motif (W-G/NLW-C-C), T27, D28 and K51 are shown as sticks in violet. (B) Region of claudin-10b model around N48 (green stick) in detail highlighting likely electrostatic interactions (dotted lines) between N48 and backbone of surrounding residues (T27, L49, W50) as well as potential electrostatic interaction between residues K51 and D28. (C) In the claudin-15 structure (cyan), the corresponding residue N47 participates in similar interactions (dotted lines) whereas residues Y50 (corresponding to K51) and S27 (corresponding to D28) do not interact. (D) The p.N48K substitution (blue) prevents the interactions mediated by N48 and may also disturb a D28-K51 interaction. O atoms, red; N atoms, dark blue; S yellow.