No abstract

Aim: Claudin-15 is mainly expressed in the small intestine and indirectly involved in glucose absorption. Similar to claudin-2 and -10b, claudin-15 is known to form a paracellular channel for small cations. Claudin-2, but not claudin-10b, also forms water channels. Here we experimentally tested whether claudin-15 also mediates water transport and if yes, whether water transport is Na + -coupled, as seen for claudin-2. Methods: MDCK C7 cells were stably transfected with claudin-15. Ion and water permeability were investigated in confluent monolayers of control and claudin-15expressing cells. Water flux was induced by an osmotic or ionic gradient. Results: Expression of claudin-15 in MDCK cells strongly increased cation permeability. The permeability ratios for monovalent cations indicated a passage of partially hydrated ions through the claudin-15 pore. Accordingly, its pore diameter was determined to be larger than that of claudin-2 and claudin-10b. Mannitol-induced water flux was elevated in claudin-15-expressing cells compared to control cells. In contrast to the Na + -coupled water flux of claudin-2 channels, claudin-15-mediated water flux was inhibited by Na + flux. Consequently, water flux was increased in Na + -free solution. Likewise, Na + flux was decreased after induction of water flux through claudin-15. Conclusion: Claudin-15, similar to claudin-2, forms a paracellular cation and water channel. In functional contrast to claudin-2, water and Na + fluxes through claudin-15 inhibit each other. Claudin-15 allows Na + to retain part of its hydration shell within the pore. This then reduces the simultaneous passage of additional water through the pore. K E Y W O R D Sclaudin-15 pore diameter, paracellular ion permeability, paracellular water permeability See Editorial Commentary: Alexander, R. T. 2020. Claudin-15 is not a drag! Acta Physiol. 228, e13397.

Claudins are integral components of tight junctions (TJs) in epithelia and endothelia. When expressed in cell lines devoid of TJs, claudins are able to form TJ-like strands at contacts between adjacent cells. According to a current model of TJ strand formation, claudin protomers assemble in an antiparallel double row within the plasma membrane of each cell (cis-interaction) while binding to corresponding double rows from the neighboring cells (trans-interaction). Cis-interaction was proposed to involve two interfaces of the protomers' first extracellular segment (extracellular loop (ECL)1). In the current study, three naturally occurring claudin-10 isoforms and two claudin-10 chimeras were used to investigate strand formation. All constructs were able to interact in cis (Förster/fluorescence resonance energy transfer (FRET)), to integrate into TJs of MDCK-C7 cells (confocal laser scanning microscopy), and to form TJ-like strands in HEK293 cells (freeze-fracture electron microscopy). Strand formation occurred despite the fact that isoform claudin-10a_i1 lacks both structural ECL1 elements reported to be crucial for cis-interaction. Furthermore, results from FRET experiments on claudin-10 chimeras indicated that identity of the first transmembrane region rather than ECL1 is decisive for claudin-10 cis-interaction. Therefore, in addition to the interaction interfaces suggested in the current model for TJ strand assembly, alternative interfaces must exist.

Claudins (Cldn) form the backbone of tight junction (TJ) strands and thereby regulate paracellular permeability for solutes and water. Polymeric strands are formed by homo- and heterophilic cis- and trans-interactions between claudin protomers. Crystal structures of some claudins have been resolved; however, the mechanism by which claudins assemble into TJ strands remains unclear. To elucidate strand architecture, TJ-like strands were reconstituted in HEK293 cells by claudin transfection. Determinants of prototypic, classic barrier-forming claudins (Cldn1, -3, and -5) involved in strand formation were analyzed by mutagenesis. The capability of claudin constructs to interact in trans and to form strands was investigated by cell contact-enrichment assays and freeze-fracture electron microscopy. Residues in extracellular loops 1 and 2 of the claudins affecting strand formation were identified. Using homology modeling and molecular docking, we tested working concepts for the arrangement of claudin protomers within TJ strands. We show that the charge of Lys65 in Cldn1 and Glu158 in Cldn3, but not of Arg30 or Asp145 in Cldn3, and the polarity of Gln56 and Gln62 in Cldn3 and of Gln57 in Cldn5 are necessary for TJ strand formation. These residues are all conserved among barrier-forming classic claudins. The results contribute to mechanistic understanding of claudin-based regulation of paracellular permeability.

Although functional and structural models for paracellular channels formed by claudins have been reported, mechanisms regulating charge and size selectivity of these channels are unknown in detail. Here, claudin-15 and claudin-10b cation channels showing high-sequence similarity but differing channel properties were analyzed. Mutants of pore-lining residues were expressed in MDCK-C7 cells. In claudin-15, proposed ion interaction sites (D55 and E64) conserved between both claudins were neutralized. D55N and E64Q substitutions decreased ion permeabilities, and D55N/E64Q had partly additive effects. D55N increased cation dehydration capability and decreased pore diameter. Additionally, residues differing between claudin-15 and -10b close to pore center were analyzed. Claudin-10b-mimicking W63K affected neither assembly nor function of claudin-15 channels. In contrast, in claudin-10b, corresponding (claudin-15b-mimicking) K64W and K64M substitutions disturbed integration into tight junction and slightly altered relative permeabilities for differently sized monovalent cations. Removal of claudin-10b-specific negative charge (D36A substitution) was without effect. The data suggest that a common tetra-aspartate ring (D55/D56) in pore center of claudin-15/-10b channels directly attracts cations, while E64/D65 may be at least partly shielded by W63/K64. Charge at position W63/K64 affects assembly and properties for claudin-10b but not for claudin-15 channels. Our findings add to the mechanistic understanding of the determinants of paracellular cation permeability.

Ligand binding and pathway-specific activation of G protein-coupled receptors is currently being studied with great effort. Individual answers may depend on the nature of the ligands and the effector pathway. Recently, we have presented a detailed model of neuropeptide Y bound to the YR. Accordingly, the C-terminal part of the peptide binds deeply in the transmembrane bundle and brings the side chain of the most essential Y in close proximity to W Here, we investigate the role of this interaction for ligand binding and activation of this receptor. BRET sensors were used for detailed investigation of effector coupling and led to the identification of preassembly of the YR-G complex. It further confirmed ligand-dependent recruitment of arrestin3. Using equally sensitive readouts for G activation and arrestin recruitment as well as quantification with operational models of agonism allowed us to identify a strong inherent bias for G activation over arrestin3 recruitment for the wild-type receptor. By systematic mutagenesis, we found that W does not contribute to the binding affinity, but acts as an allosteric connector to couple ligand binding to G activation and arrestin3 recruitment. However, even mutagenesis to a small threonine did not lead to a complete loss of signaling. Interestingly, signaling was restored to wild-type levels by ligands that contain a naphthylalanine as the C-terminal residue instead of Y Steric and polar contributions of W for the activation of the receptor are discussed in the context of different mechanisms of G protein coupling and arrestin recruitment.

No abstract

Background:Spondyloarthritides (SpA) are characterized by frequent extra-musculoskeletal manifestations (EMM) among them acute anterior uveitis (AAU) and Crohn’s disease (CD). Vice versa, about 50% of AAU and 20% of CD patients have concomitant SpA. SpA patients show gut dysbiosis together with frequent subclinical gut inflammation. Biomarkers reflecting disturbed gut barrier (intestinal-fatty acid binding protein (iFABP), lipopolysaccharide binding protein (LBP) and zonulin) were previously found to be elevated in patients with radiographic axial SpA (r-axSpA) (1).Objectives:To evaluate whether biomarkers reflecting leaky gut are altered in patients with AAU, CD and SpA compared to healthy controls and whether they differ between patients with EMM with and without concomitant SpA.Methods:A total of 100 patients from the German Spondyloarthritis Inception Cohort (GESPIC) were included – among them 20 patients with r-axSpA without EMM, 40 patients with CD and 40 patients with non-infectious AAU – out of which 19 and 20 patients, respectively, had concomitant SpA (11/8 and 20/0 axial/peripheral SpA, respectively). The GESPIC patients were compared to 20 age- and sex-matched healthy donors (HD). The following five serum biomarkers were analyzed with ELISA: calprotectin, iFABP, LBP, soluble CD14 (sCD14) and zonulin.Results:Patient characteristics are shown in Table 1. Serum levels of calprotectin, LBP, sCD14 and zonulin differed significantly between patients with r-axSpA, AAU and CD with and without concomitant SpA and HD (Figure 1). When comparing patients with EMM with and without underlying SpA, calprotectin serum levels were significantly elevated in CD patients with SpA (8.6µg/ml (SD 5.5µg/ml)) compared to CD patients without SpA (5.7µg/ml (SD 4.1µg/ml); Mann-Whitney U Test, p=0.031). Serum levels of the analyzed biomarkers did not differ between AAU patients with and without axSpA. Spearman rank correlation revealed a significant association between CRP and calprotectin (correlation coefficient r=0.230; p=0.012), LBP (r=0.596; p<0.0001), sCD14 (r=0.428; p<0.0001) and zonulin (r=0.221; p=0.016), respectively. Furthermore, LBP and zonulin serum levels correlated positively (r=0.208; p=0.023); as well as LBP and sCD14 levels (r=0.418; p<0.0001).Table 1.Patient characteristics. Mean values (standard deviation) or absolute numbers are shown.CD + SpACDr-axSpAAAU + axSpAAAUHDN192120202020Age39.1 (11.3)38.7 (14.4)38.4 (10.3)39.6 (12.0)39.2 (12.5)38.6 (12.9)Male (%)9 (47%)9 (43%)9 (45%)9 (45%)9 (45%)9 (45%)HLA-B27 positive (%)5 (26%)3 (14%)17 (85%)17 (85%)13 (65%)2 (10%)CRP in mg/l14.3 (25.6)18.0 (41.8)9.1 (11.3)6.7 (9.9)2.3 (3.4)0.6 (0.7)ASDAS2.8 (1.1)3.2 (0.6)2.2 (1.0)BASDAI4.1 (2.2)5.4 (1.2)3.2 (2.4)Figure 1.Biomarkers reflecting disturbed gut barrier show distinct signatures in patients with acute anterior uveitis, Crohn’s disease and axial Spondyloarthritis. Kruskal Wallis Test; p values shown. Dunn-Bonferroni Post-Hoc analyses, significant pairwise differences are marked; * p<0.05; **p<0.01; ***p<0.0001Conclusion:We found substantial differences in biomarkers reflecting disturbed gut barrier between with SpA, CD, AAU and healthy controls. The presence of SpA was associated with higher calprotectin serum levels in CD as compared to CD without SpA.References:[1]Ciccia F, et al. Dysbiosis and zonulin upregulation alter gut epithelial and vascular barriers in patients with ankylosing spondylitis. 2017.Acknowledgements:Dr. Judith Rademacher is participant in the BIH-Charité Clinician Scientist Program funded by the Charité –Universitätsmedizin Berlin and the Berlin Institute of Health.Disclosure of Interests:None declared