Epithelial tissues provide both a physical barrier and an antimicrobial barrier. Antimicrobial peptides of the human β-defensin (hBD) family are part of the innate immune responses that play a role in mucosal defense. hBDs are made in epithelia including oral epithelium where the bacterial load is particularly great. hBD-2 and hBD-3 are up-regulated in response to bacterial stimuli. Previous studies show that hBD-2 expression in human gingival epithelial cells (GEC) is stimulated by both nonpathogenic and pathogenic bacteria, including Porphyromonas gingivalis, a Gram-negative pathogen associated with periodontitis. Present evidence suggests that hBD-2 expression in GEC uses several signaling pathways, including an NF-κB-mediated pathway but without apparent LPS-TLR4 signaling. Protease-activated receptors (PAR) are G-protein-coupled receptors that mediate cellular responses to extracellular proteinases. P. gingivalis secretes multiple proteases that contribute to its virulence mechanisms. To determine whether PAR signaling is used in hBD-2 induction, GEC were stimulated with wild-type P. gingivalis or mutants lacking one or more proteases. hBD-2 mRNA expression was reduced in GEC stimulated with single protease mutants (11–67% compared with wild type), strongly reduced in double mutants (0.1–16%), and restored to wild-type levels (93%) in mutant with restored protease activity. Stimulation by wild type was partially blocked by inhibitors of phospholipase C, a main signaling pathway for PARs. Expression of hBD-3 was unaffected. Peptide agonist of PAR-2, but not PAR-1 activator, also induced hBD-2 in GEC. Thus, P. gingivalis proteases are directly involved in regulation of hBD-2 in cultured GEC, and this induction partially uses the PAR-2 receptor and signaling pathway.

DrrA and DrrB proteins confer resistance to the commonly used anticancer agents daunorubicin and doxorubicin in the producer organism Streptomyces peucetius. The drrAB locus has previously been cloned in Escherichia coli, and the proteins have been found to be functional in this host. DrrA, a soluble protein, belongs to the ABC family of proteins. It forms a complex with the integral membrane protein DrrB. Previous studies suggest that the function and stability of DrrA and DrrB are biochemically coupled. Thus, DrrA binds ATP only when it is in a complex with DrrB in the membrane. Further, DrrB is completely degraded if DrrA is absent. In the present study, we have characterized domains in DrrB that may be directly involved in interaction with DrrA. Several single-cysteine substitutions in DrrB were made. Interaction between DrrA and DrrB was studied by using a cysteine to amine chemical cross-linker that specifically cross-links a free sulfhydryl group in one protein (DrrB) to an amine in another (DrrA). We show here that DrrA cross-links with both the N- and the C-terminal ends of the DrrB protein, implying that they may be involved in interaction. Furthermore, this study identifies a motif within the N-terminal cytoplasmic tail of DrrB, which is similar to a motif recently shown by crystal structure analysis in BtuC and previously shown by sequence analysis to be also present in exporters, including MDR1. We propose that the motif present in DrrB and other exporters is actually a modified version of the EAA motif, which was originally believed to be present only in the importers of the ABC family. The present work is the first report where domains of interaction in the membrane component of an ABC drug exporter have been biochemically characterized.

DrrA and DrrB proteins form an ATP-dependent efflux pump for doxorubicin and daunorubicin in Streptomyces peucetius. DrrA, the catalytic subunit, forms a complex with the integral membrane protein DrrB. Previous studies have provided evidence for strong interaction between these two proteins, which was found to be critical for binding of ATP to DrrA and for stability of DrrB. Chemical cross-linking experiments carried out previously showed that in the resting state of the complex DrrA and DrrB are in contact with each other. Use of a cysteine-to-amine cross-linker then allowed identification of the N-terminal cytoplasmic tail of DrrB (residues 1-53) as the primary region of contact with DrrA. In this study, single-cysteine substitutions were introduced into different domains of DrrA in a strain already containing the S23C substitution in the N-terminal tail of DrrB. By using different arm-length disulfide cross-linkers, we found that a cysteine placed in the Q-loop region of DrrA traps DrrA in the dimeric state, thus indicating that in the closed conformation the Q-loops from opposing subunits are in the proximity of each other. Furthermore, the same region of DrrA was also found to interact with the N-terminus of DrrB, although the A-A interaction was much more prominent than the A-B interaction under these conditions. On the basis of additional data shown here, we propose that the interaction of the Q-loop with the N-terminal cytoplasmic tail of DrrB identifies an important step in the communication of conformational changes between DrrA and DrrB. The significance of these findings in the mechanism of the DrrAB complex is discussed, and a model based on analyses of different conformations of DrrA and DrrB is presented.

This paper describes the spontaneous formation of well-defined pores in planar lipid bilayers from the self-assembly of a small synthetic molecule that contains a benzothiazole aniline (BTA) group attached to a tetra-ethylene glycol (EG4) moiety. Macroscopic and single-channel current recordings suggest that these pores are formed by the assembly of four BTA-EG4 monomers with an open pore diameter that appears similar to the one of gramicidin pores (~0.4 nm). The single-channel conductance of these pores is modulated by the pH of the electrolyte and has a minimum at pH~3. Self-assembled pores from BTA-EG4 are selective for monovalent cations and have long open channel lifetimes on the order of seconds. BTA-EG4 monomers in these pores appear to be arranged symmetrically across both leaflets of the bilayer, and spectroscopy studies suggest that the fluorescent BTA group is localized inside the lipid bilayers. In terms of biological activity, BTA-EG4 molecules inhibited growth of gram-positive Bacillus subtilis bacteria (IC50~50 μM) and human neuroblastoma SH-SY5Y cells (IC50~60 μM), while they were not toxic to gram-negative Escherichia coli bacteria at a concentration up to 500 μM. Based on these properties, this drug-like, synthetic, pore-forming molecule with a molecular weight below 500 g mol(-1) might be appealing as a starting material for development of antibiotics or membrane-permeating moieties for drug delivery. From a biophysical point of view, long-lived, well-defined ion-selective pores from BTA-EG4 molecules offer an example of a self-assembled synthetic supramolecule with biological function.

This paper describes the formation of giant proteoliposomes containing P-glycoprotein (P-gp) from a solution of small proteoliposomes that had been deposited and partially dried on a film of agarose. This preparation method generated a significant fraction of giant proteoliposomes that were free of internalized vesicles, making it possible to determine the accessible liposome volume. Measuring the intensity of the fluorescent substrate rhodamine 123 (Rho123) inside and outside these giant proteoliposomes determined the concentration of transported substrates of P-gp. Fitting a kinetic model to the fluorescence data revealed the rate of passive diffusion as well as active transport by reconstituted P-gp in the membrane. This approach determined estimates for the membrane permeability coefficient (Ps) of passive diffusion and rate constants of active transport (kT) by P-gp as a result of different experimental conditions. The Ps value for Rho123 was larger in membranes containing P-gp under all assay conditions than in membranes without P-gp indicating increased leakiness in the presence of reconstituted transmembrane proteins. For P-gp liposomes, the kT value was significantly higher in the presence of ATP than in its absence or in the presence of ATP and the competitive inhibitor verapamil. This difference in kT values verified that P-gp was functionally active after reconstitution and quantified the rate of active transport. Lastly, patch clamp experiments on giant proteoliposomes showed ion channel activity consistent with a chloride ion channel protein that co-purified with P-gp. Together, these results demonstrate several advantages of using giant rather than small proteoliposomes to characterize transport properties of transport proteins and ion channels.