Identification of Human TRIAC Transmembrane Transporters.

Background: 3,5,3'-Triiodothyroacetic acid (TRIAC) is a T3-receptor agonist pharmacologically used in patients to mitigate T3 resistance. It is additionally explored to treat some symptoms of patients with inactivating mutations in the thyroid hormone (TH) transporter monocarboxylate transporter 8 (MCT8, SLC16A2). MCT8 is expressed along the blood-brain barrier, on neurons, astrocytes, and oligodendrocytes. Hence, pathogenic variants in MCT8 limit the access of TH into and their functions within the brain. TRIAC was shown to enter the brain independently of MCT8 and to modulate expression of TH-dependent genes. The aim of the study was to identify transporters that facilitate TRIAC uptake into cells. Methods: We performed a whole-genome RNAi screen in HepG2 cells stably expressing a T3-receptor-dependent luciferase reporter gene. Validation of hits from the primary and confirmatory secondary screen involved a counter screen with siRNAs and compared the cellular response to TRIAC to the effect of T3, in order to exclude siRNAs targeting the gene expression machinery. MDCK1 cells were stably transfected with cDNA encoding C-terminally myc-tagged versions of the identified TRIAC-preferring transporters. Several individual clones were selected after immunocytochemical characterization for biochemical characterization of their 125I-TRIAC transport activities. Results: We identified SLC22A9 and SLC29A2 as transporters mediating cellular uptake of TRIAC. SLC22A9 encodes the organic anion transporter 7 (OAT7), a sodium-independent organic anion transporter expressed in the plasma membrane in brain, pituitary, liver, and other organs. Competition with the SLC22A9/OAT7 substrate estrone-3-sulfate reduced 125I-TRIAC uptake. SLC29A2 encodes the equilibrative nucleoside transporter 2 (ENT2), which is ubiquitously expressed, including pituitary and brain. Coincubation with the SLC29A2/ENT2 inhibitor nitrobenzyl-6-thioinosine reduced 125I-TRIAC uptake. Moreover, ABCD1, an ATP-dependent peroxisomal pump, was identified as a 125I-TRIAC exporter in transfected MDCK1 cells. Conclusions: Knowledge of TRIAC transporter expression patterns, also during brain development, may thus in the future help to interpret observations on TRIAC effects, as well as understand why TRIAC may not show a desirable effect on cells or organs not expressing appropriate transporters. The identification of ABCD1 highlights the sensitivity of our established screening assay, but it may not hold significant relevance for patients undergoing TRIAC treatment.

Spatially resolved protein map of intact human cytomegalovirus virions.

Herpesviruses assemble large enveloped particles that are difficult to characterize structurally due to their size, fragility and complex multilayered proteome with partially amorphous nature. Here we used crosslinking mass spectrometry and quantitative proteomics to derive a spatially resolved interactome map of intact human cytomegalovirus virions. This enabled the de novo allocation of 32 viral proteins into four spatially resolved virion layers, each organized by a dominant viral scaffold protein. The viral protein UL32 engages with all layers in an N-to-C-terminal radial orientation, bridging nucleocapsid to viral envelope. We observed the layer-specific incorporation of 82 host proteins, of which 39 are selectively recruited. We uncovered how UL32, by recruitment of PP-1 phosphatase, antagonizes binding to 14-3-3 proteins. This mechanism assures effective viral biogenesis, suggesting a perturbing role of UL32-14-3-3 interaction. Finally, we integrated these data into a coarse-grained model to provide global insights into the native configuration of virus and host protein interactions inside herpesvirions.

Mechanism of threonine ADP-ribosylation of F-actin by a Tc toxin.

Tc toxins deliver toxic enzymes into host cells by a unique injection mechanism. One of these enzymes is the actin ADP-ribosyltransferase TccC3, whose activity leads to the clustering of the cellular cytoskeleton and ultimately cell death. Here, we show in atomic detail how TccC3 modifies actin. We find that the ADP-ribosyltransferase does not bind to G-actin but interacts with two consecutive actin subunits of F-actin. The binding of TccC3 to F-actin occurs via an induced-fit mechanism that facilitates access of NAD+ to the nucleotide binding pocket. The following nucleophilic substitution reaction results in the transfer of ADP-ribose to threonine-148 of F-actin. We demonstrate that this site-specific modification of F-actin prevents its interaction with depolymerization factors, such as cofilin, which impairs actin network turnover and leads to steady actin polymerization. Our findings reveal in atomic detail a mechanism of action of a bacterial toxin through specific targeting and modification of F-actin.

Claudin-Targeted Suicide Gene Therapy for Claudin-Overexpressing Tumor Cells by Using Modified Clostridium perfringens Enterotoxin (CPE).

Bacterial toxins gain growing attention as potential cancer treatment due to their potent cytotoxic effects. Among the very different toxins with diverse modes of action, the Clostridium perfringens enterotoxin (CPE) is in focus to treat solid cancers. This toxin targets the tight junction proteins claudin-3 and -4 (Cldn-3/4), which are frequently overexpressed in solid cancers. Binding to these claudins induces pore formation in the host cell plasma membrane leading to rapid oncoleaking cell death of tumor cells. Based on this, extending the targeting of CPE beyond Cldn-3/4 is of interest, since other claudins, such as claudin-1 or -5 are often overexpressed in various cancer entities such as non-small-cell lung cancer (NSCLC) or papillary thyroid carcinoma. In this chapter we describe the modification of a CPE-encoding vector by structure-directed mutagenesis to either preferentially target Cldn-1 and -5 or to expand targeting to Cldn1-9 for improved broadened cytotoxic targeting of claudin-overexpressing tumors such as but not limited to lung cancer via CPE gene transfer.

Molecular architecture and assembly of the tight junction backbone.

The functional and structural concept of tight junctions has developed after discovery of claudin and TAMP proteins. Many of these proteins contribute to epi- and endothelial barrier but some, in contrast, form paracellular channels. Claudins form the backbone of tight junction (TJ) strands whereas other proteins regulate TJ dynamics. The current joined double-row model of TJ strands and channels is crucially based on the linear alignment of claudin-15 in the crystal. Molecular dynamics simulations, protein docking, mutagenesis, cellular TJ reconstitution, and electron microscopy studies largely support stability and functionality of the model. Here, we summarize in silico and in vitro data about TJ strand assembly including comparison of claudin crystal structures and alternative models. Sequence comparisons, experimental and structural data substantiate differentiation of classic and non-classic claudins differing in motifs related to strand assembly. Classic claudins seem to share a similar mechanism of strand formation. Interface variations likely contribute to TJ strand flexibility. Combined in vitro/in silico studies are expected to elucidate mechanistic keys determining TJ regulation.

Aryl Hydrocarbon Receptor Modulation by Tuberculosis Drugs Impairs Host Defense and Treatment Outcomes.

Antimicrobial resistance in tuberculosis (TB) is a public health threat of global dimension, worsened by increasing drug resistance. Host-directed therapy (HDT) is an emerging concept currently explored as an adjunct therapeutic strategy for TB. One potential host target is the ligand-activated transcription factor aryl hydrocarbon receptor (AhR), which binds TB virulence factors and controls antibacterial responses. Here, we demonstrate that in the context of therapy, the AhR binds several TB drugs, including front line drugs rifampicin (RIF) and rifabutin (RFB), resulting in altered host defense and drug metabolism. AhR sensing of TB drugs modulates host defense mechanisms, notably impairs phagocytosis, and increases TB drug metabolism. Targeting AhR in vivo with a small-molecule inhibitor increases RFB-treatment efficacy. Thus, the AhR markedly impacts TB outcome by affecting both host defense and drug metabolism. As a corollary, we propose the AhR as a potential target for HDT in TB in adjunct to canonical chemotherapy.

Targeting claudin-overexpressing thyroid and lung cancer by modified Clostridium perfringens enterotoxin.

Clostridium perfringens enterotoxin (CPE) can be used to eliminate carcinoma cells that overexpress on their cell surface CPE receptors - a subset of claudins (e.g., Cldn3 and Cldn4). However, CPE cannot target tumors expressing solely CPE-insensitive claudins (such as Cldn1 and Cldn5). To overcome this limitation, structure-guided modifications were used to generate CPE variants that can strongly bind to Cldn1, Cldn2 and/or Cldn5, while maintaining the ability to bind Cldn3 and Cldn4. This enabled (a) targeting of the most frequent endocrine malignancy, namely, Cldn1-overexpressing thyroid cancer, and (b) improved targeting of the most common cancer type worldwide, non-small-cell lung cancer (NSCLC), which is characterized by high expression of several claudins, including Cldn1 and Cldn5. Different CPE variants, including the novel mutant CPE-Mut3 (S231R/S313H), were applied on thyroid cancer (K1 cells) and NSCLC (PC-9 cells) models. In vitro, CPE-Mut3, but not CPEwt, showed Cldn1-dependent binding and cytotoxicity toward K1 cells. For PC-9 cells, CPE-Mut3 improved claudin-dependent cytotoxic targeting, when compared to CPEwt. In vivo, intratumoral injection of CPE-Mut3 in xenograft models bearing K1 or PC-9 tumors induced necrosis and reduced the growth of both tumor types. Thus, directed modification of CPE enables eradication of tumor entities that cannot be targeted by CPEwt, for instance, Cldn1-overexpressing thyroid cancer by using the novel CPE-Mut3.

Host monitoring of quorum sensing during Pseudomonas aeruginosa infection.

Pseudomonas aeruginosa rapidly adapts to altered conditions by quorum sensing (QS), a communication system that it uses to collectively modify its behavior through the production, release, and detection of signaling molecules. QS molecules can also be sensed by hosts, although the respective receptors and signaling pathways are poorly understood. We describe a pattern of regulation in the host by the aryl hydrocarbon receptor (AhR) that is critically dependent on qualitative and quantitative sensing of P. aeruginosa quorum. QS molecules bind to AhR and distinctly modulate its activity. This is mirrored upon infection with P. aeruginosa collected from diverse growth stages and with QS mutants. We propose that by spying on bacterial quorum, AhR acts as a major sensor of infection dynamics, capable of orchestrating host defense according to the status quo of infection.

Use of Modified Clostridium perfringens Enterotoxin Fragments for Claudin Targeting in Liver and Skin Cells.

Claudins regulate paracellular permeability in different tissues. The claudin-binding domain of Clostridium perfringens enterotoxin (cCPE) is a known modulator of a claudin subset. However, it does not efficiently bind to claudin-1 (Cldn1). Cldn1 is a pharmacological target since it is (i) an essential co-receptor for hepatitis C virus (HCV) infections and (ii) a key element of the epidermal barrier limiting drug delivery. In this study, we investigated the potential of a Cldn1-binding cCPE mutant (i) to inhibit HCV entry into hepatocytes and (ii) to open the epidermal barrier. Inhibition of HCV infection by blocking of Cldn1 with cCPE variants was analyzed in the Huh7.5 hepatoma cell line. A model of reconstructed human epidermis was used to investigate modulation of the epidermal barrier by cCPE variants. In contrast to cCPEwt, the Cldn1-binding cCPE-S305P/S307R/S313H inhibited infection of Huh7.5 cells with HCV in a dose-dependent manner. In addition, TJ modulation by cCPE variant-mediated targeting of Cldn1 and Cldn4 opened the epidermal barrier in reconstructed human epidermis. cCPE variants are potent claudin modulators. They can be applied for mechanistic in vitro studies and might also be used as biologics for therapeutic claudin targeting including HCV treatment (host-targeting antivirals) and improvement of drug delivery.

Reversible opening of the blood-brain barrier by claudin-5-binding variants of Clostridium perfringens enterotoxin's claudin-binding domain.

The blood-brain barrier (BBB) prevents entry of neurotoxic substances but also that of drugs into the brain. Here, the paracellular barrier is formed by tight junctions (TJs) with claudin-5 (Cldn5) being the main sealing constituent. Transient BBB opening by targeting Cldn5 could improve paracellular drug delivery. The non-toxic C-terminal domain of Clostridium perfringens enterotoxin (cCPE) binds to a subset of claudins, e.g., Cldn3, -4. Structure-based mutagenesis was used to generate Cldn5-binding variants (cCPE-Y306W/S313H and cCPE-N218Q/Y306W/S313H). These cCPE-variants were tested for transient TJ opening using multiple in vitro BBB models: Primary porcine brain endothelial cells, coculture of primary rat brain endothelial cells with astrocytes and mouse cerebEND cells. cCPE-Y306W/S313H and cCPE-N218Q/Y306W/S313H but neither cCPE-wt nor cCPE-Y306A/L315A (not binding to claudins) decreased transendothelial electrical resistance in a concentration-dependent and reversible manner. Furthermore, permeability of carboxyfluorescein (with size of CNS drugs) was increased. cCPE-Y306W/S313H but neither cCPE-wt nor cCPE-Y306A/L315A bound to Cldn5-expressing brain endothelial cells. However, freeze-fracture EM showed that cCPE-Y306W/S313H did not cause drastic TJ breakdown. In sum, Cldn5-binding cCPE-variants enabled mild and transient opening of brain endothelial TJs. Using reliable in vitro BBB models, the results demonstrate that cCPE-based biologics designed to bind Cldn5 improve paracellular drug delivery across the BBB.

A cCPE-based xenon biosensor for magnetic resonance imaging of claudin-expressing cells.

The majority of malignant tumors originate from epithelial cells, and many of them are characterized by an overexpression of claudins (Cldns) and their mislocalization out of tight junctions. We utilized the C-terminal claudin-binding domain of Clostridium perfringens enterotoxin (cCPE), with its high affinity to specific members of the claudin family, as the targeting unit for a claudin-sensitive cancer biosensor. To overcome the poor sensitivity of conventional relaxivity-based magnetic resonance imaging (MRI) contrast agents, we utilized the superior sensitivity of xenon Hyper-CEST biosensors. We labeled cCPE for both xenon MRI and fluorescence detection. As one readout module, we employed a cryptophane (CrA) monoacid and, as the second, a fluorescein molecule. Both were conjugated separately to a biotin molecule via a polyethyleneglycol chemical spacer and later via avidin linked to GST-cCPE. Nontransfected HEK293 cells and HEK293 cells stably expressing Cldn4-FLAG were incubated with the cCPE-based biosensor. Fluorescence-based flow cytometry and xenon MRI demonstrated binding of the biosensor specifically to Cldn4-expressing cells. This study provides proof of concept for the use of cCPE as a carrier for diagnostic contrast agents, a novel approach for potential detection of Cldn3/-4-overexpressing tumors for noninvasive early cancer detection.

Structural determinants of a conserved enantiomer-selective carvone binding pocket in the human odorant receptor OR1A1.

Chirality is a common phenomenon within odorants. Most pairs of enantiomers show only moderate differences in odor quality. One example for enantiomers that are easily discriminated by their odor quality is the carvones: humans significantly distinguish between the spearmint-like (R)-(-)-carvone and caraway-like (S)-(+)-carvone enantiomers. Moreover, for the (R)-(-)-carvone, an anosmia is observed in about 8% of the population, suggesting enantioselective odorant receptors (ORs). With only about 15% de-orphaned human ORs, the lack of OR crystal structures, and few comprehensive studies combining in silico and experimental approaches to elucidate structure-function relations of ORs, knowledge on cognate odorant/OR interactions is still sparse. An adjusted homology modeling approach considering OR-specific proline-caused conformations, odorant docking studies, single-nucleotide polymorphism (SNP) analysis, site-directed mutagenesis, and subsequent functional studies with recombinant ORs in a cell-based, real-time luminescence assay revealed 11 amino acid positions to constitute an enantioselective binding pocket necessary for a carvone function in human OR1A1 and murine Olfr43, respectively. Here, we identified enantioselective molecular determinants in both ORs that discriminate between minty and caraway odor. Comparison with orthologs from 36 mammalian species demonstrated a hominid-specific carvone binding pocket with about 100% conservation. Moreover, we identified loss-of-function SNPs associated with the carvone binding pocket of OR1A1. Given carvone enantiomer-specific receptor activation patterns including OR1A1, our data suggest OR1A1 as a candidate receptor for constituting a carvone enantioselective phenotype, which may help to explain mechanisms underlying a (R)-(-)-carvone-specific anosmia in humans.

Polar and charged extracellular residues conserved among barrier-forming claudins contribute to tight junction strand formation.

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.

Membrane-traversing mechanism of thyroid hormone transport by monocarboxylate transporter 8.

Monocarboxylate transporter 8 (MCT8) mediates thyroid hormone (TH) transport across the plasma membrane in many cell types. In order to better understand its mechanism, we have generated three new MCT8 homology models based on sugar transporters XylE in the intracellular opened (PDB ID: 4aj4) and the extracellular partly occluded (PDB ID: 4gby) conformations as well as FucP (PDB ID: 3o7q) and GLUT3 (PDB ID: 4zwc) in the fully extracellular opened conformation. T3-docking studies from both sides revealed interactions with His192, His415, Arg445 and Asp498 as previously identified. Selected mutations revealed further transport-sensitive positions mainly at the discontinuous transmembrane helices TMH7 and 10. Lys418 is potentially involved in neutralising the charge of the TH substrate because it can be replaced by charged, but not by uncharged, amino acids. The side chain of Thr503 was hypothesised to stabilise a helix break at TMH10 that undergoes a prominent local shift during the transport cycle. A T503V mutation accordingly affected transport. The aromatic Tyr419, the polar Ser313 and Ser314 as well as the charged Glu422 and Glu423 lining the transport channel have been studied. Based on related sugar transporters, we suggest an alternating access mechanism for MCT8 involving a series of amino acid positions previously and newly identified as critical for transport.

Targeting and alteration of tight junctions by bacteria and their virulence factors such as Clostridium perfringens enterotoxin.

The integrity of tight junctions, which regulate paracellular permeability, is challenged by many bacterial pathogens. This is caused by inflammatory responses triggered by pathogens and direct interaction of bacteria or their toxins with host epithelial cells. In some cases, tight junction proteins represent receptors for cell surface proteins or toxins of the pathogen, such as Clostridium perfringens enterotoxin (CPE). CPE causes diarrhea and cramps-the symptoms of a common foodborne illness, caused by C. perfringens type A. It uses a subgroup of the claudin family of tight junction proteins as receptors and forms pores in the membrane of intestinal epithelial cells. Ca2+ influx through these pores finally triggers cell damage. In this review, we summarize tight junction targeting and alteration by a multitude of different microorganisms such as C. perfringens, Escherichia coli, Helicobacter pylori, Salmonella typhimurium, Shigella flexneri, Vibrio cholerae, Yersinia enterocolitica, protozoan parasites, and their proteins. A focus is drawn towards CPE, the interaction with its receptors, cellular, and pathophysiological consequences for the intestinal epithelium. In addition, we portend to the use of CPE-based claudin modulators for drug delivery as well as diagnosis and therapy of cancer.

Defining a conformational consensus motif in cotransin-sensitive signal sequences: a proteomic and site-directed mutagenesis study.

The cyclodepsipeptide cotransin was described to inhibit the biosynthesis of a small subset of proteins by a signal sequence-discriminatory mechanism at the Sec61 protein-conducting channel. However, it was not clear how selective cotransin is, i.e. how many proteins are sensitive. Moreover, a consensus motif in signal sequences mediating cotransin sensitivity has yet not been described. To address these questions, we performed a proteomic study using cotransin-treated human hepatocellular carcinoma cells and the stable isotope labelling by amino acids in cell culture technique in combination with quantitative mass spectrometry. We used a saturating concentration of cotransin (30 micromolar) to identify also less-sensitive proteins and to discriminate the latter from completely resistant proteins. We found that the biosynthesis of almost all secreted proteins was cotransin-sensitive under these conditions. In contrast, biosynthesis of the majority of the integral membrane proteins was cotransin-resistant. Cotransin sensitivity of signal sequences was neither related to their length nor to their hydrophobicity. Instead, in the case of signal anchor sequences, we identified for the first time a conformational consensus motif mediating cotransin sensitivity.

Directed structural modification of Clostridium perfringens enterotoxin to enhance binding to claudin-5.

Clostridium perfringens enterotoxin (CPE) binds to distinct claudins (Clds), which regulate paracellular barrier functions in endo- and epithelia. The C-terminal domain (cCPE) has the potential for selective claudin modulation, since it only binds to a subset of claudins, e.g., Cld3 and Cld4 (cCPE receptors). Cld5 (non-CPE receptor) is a main constituent in tight junctions (TJ) of the blood-brain barrier. We aimed to reveal claudin recognition mechanisms of cCPE and to create a basis for a Cld5-binder. By utilizing structure-based interaction models, mutagenesis and assays of cCPE-binding to the TJ-free cell line HEK293, transfected with human Cld1 and murine Cld5, we showed how cCPE-binding to Cld1 and Cld5 is prevented by two residues in extracellular loop 2 of Cld1 (Asn(150) and Thr(153)) and Cld5 (Asp(149) and Thr(151)). Binding to Cld5 is especially attenuated by the lack of a bulky hydrophobic residue like leucine at position 151. By downsizing the binding pocket and compensating for the lack of this leucine residue, we created a novel cCPE-variant; cCPEY306W/S313H binds Cld5 with nanomolar affinity (K d 33 ± 10 nM). Finally, the effective binding to endogenously Cld5-expressing blood-brain barrier model cells (murine microvascular endothelial cEND cell line) suggests cCPEY306W/S313H as basis for Cld5-specific modulation to improve paracellular drug delivery, or to target claudin overexpressing tumors.

Molecular and structural transmembrane determinants critical for embedding claudin-5 into tight junctions reveal a distinct four-helix bundle arrangement.

The mechanism of TJ (tight junction) assembly and the structure of TJ strand-forming Cldns (claudins) are unclear. To identify determinants of assembly of blood-brain barrier-related Cldn3 and Cldn5, chimaeric mutants were analysed by cellular reconstitution of TJ strands and live-cell imaging. On the basis of the rescue of mutants deficient for strand formation, we identified Cldn5 residues (Cys128, Ala132, Ile142, Ala163, Ile166 and Leu174) involved in Cldn folding and assembly. Experimental results were combined with structural bioinformatics approaches. Initially the experimentally validated previous model of the ECL2 (extracellular loop 2) of Cldn5 was extended to the flanking transmembrane segments (TM3/TM4). A coiled-coil interface probably caused by alternating small and large residues is supported by concomitant knob-into-hole interactions including Cldn5-specific residues identified in the present paper. To address arrangement of the TMs in a four-helix bundle, data from evolutionary sequence couplings and comparative modelling of intramolecular interfaces in the transmembrane region of Cldns led to a complete Cldn5 model. Our suggested Cldn subtype-specific intramolecular interfaces that are formed by conserved coiled-coil motifs and non-conserved residues in distinct TM positions were confirmed by the recently released crystal structure of Cldn15. The identified molecular and structural determinants essentially contribute to assembly of Cldns into TJ strands.

Claudin-3 and claudin-5 protein folding and assembly into the tight junction are controlled by non-conserved residues in the transmembrane 3 (TM3) and extracellular loop 2 (ECL2) segments.

The mechanism of tight junction (TJ) assembly and the structure of claudins (Cldn) that form the TJ strands are unclear. This limits the molecular understanding of paracellular barriers and strategies for drug delivery across tissue barriers. Cldn3 and Cldn5 are both common in the blood-brain barrier but form TJ strands with different ultrastructures. To identify the molecular determinants of folding and assembly of these classic claudins, Cldn3/Cldn5 chimeric mutants were generated and analyzed by cellular reconstitution of TJ strands, live cell confocal imaging, and freeze-fracture electron microscopy. A comprehensive screening was performed on the basis of the rescue of mutants deficient for strand formation. Cldn3/Cldn5 residues in transmembrane segment 3, TM3 (Ala-127/Cys-128, Ser-136/Cys-137, Ser-138/Phe-139), and the transition of TM3 to extracellular loop 2, ECL2 (Thr-141/Ile-142) and ECL2 (Asn-148/Asp-149, Leu-150/Thr-151, Arg-157/Tyr-158), were identified to be involved in claudin folding and/or assembly. Blue native PAGE and FRET assays revealed 1% n-dodecyl ß-d-maltoside-resistant cis-dimerization for Cldn5 but not for Cldn3. This homophilic interaction was found to be stabilized by residues in TM3. The resulting subtype-specific cis-dimer is suggested to be a subunit of polymeric TJ strands and contributes to the specific ultrastructure of the TJ detected by freeze-fracture electron microscopy. In particular, the Cldn5-like exoplasmic face-associated and particle-type strands were found to be related to cis-dimerization. These results provide new insight into the mechanisms of paracellular barrier formation by demonstrating that defined non-conserved residues in TM3 and ECL2 of classic claudins contribute to the formation of TJ strands with differing ultrastructures.

Determinants contributing to claudin ion channel formation.

Pore-forming properties of claudins (Cld) are likely defined by residues of their first extracellular loop (ECL1). Detailed mechanisms are unclear. MDCK cells overexpressing FLAG-Cld-1 wild-type and mutants were characterized by transepithelial resistance (TER) and ion permeability measurements. Replacing ECL1 residues of sealing Cld-1 by corresponding Cld-2 residues we aimed to identify new determinants responsible for sealing and/or pore formation. We found that E48K and S53E substitutions in human Cld-1 strongly reduced TER and increased permeability for Na(+) and Cl(-) . In contrast, K65D, D68S, and other single substitutions showed no significant change of TER and permeability for Na(+) and Cl(-) . Double substitution S53E/K65D did not change TER and ion permeability, whereas S53E/D68S decreased TER, albeit weaker than S53E. Ratio of permeabilities for Na(+) and Cl(-) revealed no clear charge specificity of the pore induced by S53E or S53E/D68S in Cld-1, suggesting that primarily S53 and potentially D68 in Cld-1 are involved in sealing of the paracellular cleft and that charge-unselective pores may be induced by substituting S53E.