Doctor of Philosophy / Department of Biochemistry / John M. Tomich / Channel replacement therapy represents a new treatment modality that could augment existing therapies against cystic fibrosis. It is based on designing synthetic channel-forming peptides (CFPs) with desirable selectivity, high ion transport rates and overall ability to supersede defective endogenous chloride channels. We derived synthetic CFPs from a peptide initially reconstituted from the second transmembrane segment of the α-subunit of Glycine receptor (M2GlyR). Our best candidate peptide NK4-M2GlyR T19R, S22W (p22-T19R, S22W) is soluble in aqueous solutions, has the ability to deliver itself to the epithelial cell membranes without the use of a delivery system, is non-immunogenic, but when assembled into a pore, lacks the structural properties for anion selectivity. Previous findings suggested that threonine residues at positions 13, 17 and 20 line the pore of assembled p22-T19R, S22W and recent studies indicated that an introduction of positively charged 2, 3-diaminopropionic acid (Dap) at either T13 or T17 in the sequence increases transepithelial ion transport rates across the apical membranes of Madin-Darby canine kidney (MDCK) epithelial cells. This study focused on further structural modifications of the pore-lining interface of p22-T19R, S22W assembled pore. It was hypothesized that singly, doubly or triply introduced Dap residues modify the pore geometry and that their positively charged side chains impact discrimination for anions. Dap-substituted p22-T19R, S22W peptides retain the α-helical secondary structure characteristic for their parent p22-T19R, S22W. The sequences containing multiple Dap-substituted residues induce higher short circuit current across the epithelial MDCK cells compared to peptides with single Dap-substitutions or no Dap-substitutions. Whole-cell voltage clamp recordings using Xenopus oocytes indicate that Dap-substituted peptide assemblies induce higher levels of voltage-dependent but non-selective ion current relative to p22-T19R, S22W. Studies using the D-enantiomer of p22-T19R, S22W and shorter truncated sequences of a full length L-p22-T19R, S22W and L-Dap-substituted peptides provided evidence that peptide-induced ion transport rates can be attributed to formation of de novo pathways. Results of preliminary computer modeling studies indicate that Dap residues affect the pore geometry but not ion selectivity. Future studies focusing on modifying the existing electrostatic environment towards anion selectivity will focus on staggering the charged residues of Dap at various locations inside synthetic pores.

Proton channels are essential for many of the processes of life. The influenza A viral protein M2 is responsible for sensing the conditions necessary for viral RNA release. The proton-translocating FoF1 ATPase (ATP synthase) uses a proton gradient to drive adenosine triphosphate (ATP) synthesis. We have directly measured proton uptake in vesicles containing reconstituted M2 or FO by monitoring external pH after addition of valinomycin to vesicles with 100-fold diluted external [K+]. This proton flux assay was utilized to quantify proton flux through single M2 and Fo channels. Contrary to previous reports, proton uptake by M2 was not significantly altered by acidification of the extravesicular pH. We conclude that pH only weakly affects proton flux through M2 in the pH range of 5.4 - 7.0. Theoretical analysis utilized for such vesicle uptake assays illuminates the appropriate time scale of the initial slope and an important limitation that must be placed on inferences about channel ion selectivity. The rise in pH over 10 seconds after ionophore addition yielded time-averaged single channel conductances of 0.35±0.2 aS and 0.72±0.4 aS at pH 5.4 and 7.0 respectively. Such a low time-average conductance implies that M2 is only conductive 10^-6 to 10^-4 of the time. M2 selectivity for hydrogen over potassium is ~10^7. Fo translocates protons across membranes, converting electrochemical energy to rotational inertia. Previous experiments have been partially confounded by a contaminating channel, CL, which co-purifies with Fo and leaks cations. CL activity is shown to not decrease following deletion of the previously uncharacterized yraM open reading frame of E. coli. Fo purified from a deletion strain lacking yraM is just as active as Fo purified from the wild-type strain. Using Fo from the deletion strain, the single-hit hypothesis of DCCD inhibition of passive proton flux through Fo was examined. A DCCD-induced reduction in ATP synthase activity correlates with a reduction in the total initial slope, the number of functional Fo per µg protein, and the single channel proton flux. At least 2 DCCD per Fo are required to totally inactivate passive proton flux. M2 and Fo have similar single channel conductances but different open probabilities.