The Journal of Membrane Biology

Ion-induced chemical shifts in the carbonyl carbon resonances of synthesized and verified (1-13C)d-Val8 gramicidin A and (1-13C)d-Leu14 gramicidin A are utilized in combination with the previously determined location of the ion binding sites of the gramicidin A channel (using the carbonyls ofl-residues) to determine that the helix sense of the gramicidin A channel is left-handed. Having resolved the handedness issue, the location of the ion binding sites (which are fundamental to understanding the mechanism of ion transport) are further delineated with the results indicating two sites separated by just over 20 Å. Furthermore, the demonstration that the divalent barium ion interacts at the binding site while not being transported through the channel is used to argue that the mechanism of monovalentvs. divalent cation selectivity is due to the positive image force contribution to the central barrier.

CACO-2 BBE was used to determine the response of a gastrointestinal epithelium to tumor necrosis factor-α (TNF). Incubation of CACO-2 BBE with TNF did not produce any effect on transepithelial resistance (TER) within the first 6 hr but resulted in a 40–50% reduction in TER and a 30% decrease in I sc (short circuit current) relative to time-matched control at 24 hr. The decrease in TER was sustained up to 1 week following treatment with TNF and was not associated with a significant increase in the transepithelial flux of [14C]-d-mannitol or the penetration of ruthenium red into the lateral intercellular space. Dilution potential and transepithelial 22Na+ flux studies demonstrated that TNF-treatment of CACO-2 BBE cell sheets increased the paracellular permeability of the epithelium to Na+ and Cl−. The increased transepithelial permeability did not associate with an increase in the incidence of apoptosis. However, there was a TNF-dependent increase in [3H]-thymidine labeling that was not accompanied by a change in DNA content of the cell sheet. The increase in transepithelial permeability was concluded to be across the tight junction because: (i) 1 mm apical amiloride reduced the basolateral to apical flux of 22Na+, and (ii) dilution potential studies revealed a bidirectionally increased permeability to both Na+ and Cl−. These data suggest that the increase in transepithelial permeability across TNF-treated CACO-2 BBE cell sheets arises from an alteration in the charge selectivity of the paracellular conductive pathway that is not accompanied by a change in its size selectivity.

Effects of divalent cations on oscillations of membrane potentials (i.e., spontaneous repetitive hyperpolarizing responses) and on hyperpolarizing responses induced by electrical stimuli as well as on resting potentials were studied in large nondividing L cells. Deprivation of Ca2+ from the external medium inhibited these hyperpolarizing responses accompanying slight depolarization of the resting potential. Sr2+ or Mn2+ applied to the external medium in place of Ca2+ was able to substitute for Ca2+ in the generation of hyperpolarizing responses, while Mg2+, Ba2+ or La3+ suppressed hyperpolarizing responses. The addition of A23187 to the bathing medium or intracellular injection of Ca2+, Sr2+, Mn2+ or La3+ induced membrane hyperpolarization. When the external Ca2+, Sr2+ or Mn2+ concentration was increased, the resting potential also hyperpolarized, in a saturating manner. The amplitude of maximum hyperpolarization produced by high external Ca2+ was of the same order of magnitude as those of hyperpolarizing responses and was dependent on the external K+ concentration. In the light of these experimental observations, it was deduced that the K+ conductance increase associated with the hyperpolarizing excitation is the result of an increase in the intracellular concentration of free Ca2+ mainly derived from the external solution.

α1 and β1 subunits of human GABAA receptors were expressed in Sf9 cells using the Sf9-baculovirus system. Better expression was obtained by manipulating the system. Cell growth phase at the time of infection determined the practical range of virus titre, the period postinfection during which cells were useful for signal detection and the maximal current obtained. Cells in the early exponential phase were relatively insensitive to multiplicity of infection (MOI) whereas cells in the midto late-exponential phase were highly dependent on MOI and they responded with the largest Cl−} current generated by GABA. Channels activated by GABA were chloride-selective. Half the maximum peak whole-cell current was obtained with 11 μm GABA. The time course of Cl−} currents activated by saturating GABA concentrations in cells infected with α1β1recombinant viruses was examined employing a rapid perfusion system which allowed whole-cell solution exchange in less than 1 msec. The current decay could be fitted by 3 to 4 exponentials for the first 8 sec. The initial fast current decrease had a time constant of about 23 msec. No voltage dependence of time constants was detected but the whole-cell IV relation showed outward rectification. Currents were depressed by bicuculline, penicillin and picrotoxin and potentiated by pentobarbitone.

Hydrochlorothiazide (HCTZ) was shown to inhibit the transepithelial NaCl transport and the apical Na+-Cl− symport and to depolarize the apical membrane potential in the rabbit gallbladder epithelium. The depolarization was likely related to the opening of a Cl− conductance. To better understand whether an apical Cl− leak is involved in the mechanism of action of HCTZ, the transapical Cl− backflux was measured radiochemically by the washout technique. The gallbladder wall, pretreated with pronase on the serosal side to homogenize the subepithelium, was loaded with 36Cl− on the luminal side; mucosal and serosal 36Cl− effluxes (J m , J s ) were then measured every 2 min. The pretreatment with pronase did not alter the membrane potentials and the selectivity of the epithelium. Under control conditions and the tissue in steady-state, J m and J s time courses were each described by two exponential decays (A,B); the rate constants, k A and k B , were 0.71 ±0.03 and 0.16±0.01 min−1, respectively, and correspondingly the half-times (t 1 2 , t 1 2 ) were 1.01±0.05 and 5.00±0.44 min (n=10); these parameters were not significantly different for J m and J s time courses. J s was always greater than J m (J s /J m =2.02±0.22 and 1.43 ±0.17 for A and B decays). Under SCN− treatment in steady-state conditions, both J m and J s time courses were described by only one exponential decay, the component B being abolished. Moreover t 1 2 was similar to that predictable for the subepithelium. It follows that it is the component B which exits the epithelial compartment. Based on the intracellular specific activity and 36Cl− J at 0 min time of the washout experiment, the cell-lumen Cl− backflux in steady-state was calculated to be equal to about 2 μmol cm−2hr−1, in agreement with the value indirectly computable by other techniques. The experimental model was well responsive to different external challenges (increases in media osmolalities; luminal treatment with nystatin). HCTZ (2.5 · 10−4 m) largely increased 36Cl− J . The increase was abolished by luminal treatment with 10−4 m SITS, which not only brought back the efflux time courses to the ones observed under control conditions but even increased J s /J m of the cellular component, an indication of a reduced J . It is concluded that HCTZ opens an apical, SITS-sensitive Cl− leak, which contributes to dissipate the intracellular Cl− accumulation and to inhibit the NaCl transepithelial transport. Moreover, the drug is likely to reduce the basal electroneutral Cl− backflux supported by Na+-Cl− cotransport, in agreement with the inhibition of the cotransport itself.

Bilayer membranes formed from lipids dissolved in decane were exposed to glycophorin, a sialoglycoprotein which had been extracted from human red cell membranes. The interaction with the bilayer produced an increase in the steady state electrical conductance of the membrane proportional to the amount added. Fluctuations in membrane current when the electrical potential difference was constant were observed concommitantly with this increase in membrane conductance. The minimum size of the fluctuations corresponds to a conductance of 10−10 mho. The increase in conductance as well as the current fluctuations persisted after extensive washout of the chamber containing the protein (cisside). Subsequent addition of lectins (wheat germ agglutinin and phytohemoagglutinin) to the cis-side produced rupture of the membranes, whilst these hemoagglutinins added to the trans-side failed to produce an effect. Measurements of changes in surface potential using K+ nonactin as a probe indicated that glycophorin induces a negative surface charge. At high protein concentrations, the magnitude of the induced surface potential became independent of glycophorin concentration. The maximum number of charges introduced onto the membrane under these conditions was 1.4×105/μm2. Cis (but not trans)-side addition of neuraminidase abolished these charges, indicating that they can be ascribed to the sialic acid residues that the protein bears. These results suggest that glycophorin incorporates into bilayer membranes with its N-terminal end (where the sialic acid and carbohydrates are located) facing the cis-side. Spectrin reversibly lowered the glycophorin-induced membrane conductance when added to the trans-side. Cis-side additions failed to produce an effect. Trypsin present on the trans-side irreversibly lowered the membrane conductance. These results indicate that parts of the glycophorin molecule, probably the C-terminal end, are accessible to reagents in the solution bathing the trans-side of the membrane. Thus glycophorin spans the planar bilayer in much the same way as it spans the red cell membrane.

Although an outwardly rectifying K+-conductance has been described in murine peritoneal macrophages and a murine macrophage cell line, the expression of this conductance in human monocyte-derived macrophages (HMDMs) is rare. Whole-cell current recordings in this study were obtained from HMDMs differentiated in adherent culture for varying periods of time following isolation and compared to currents obtained in human alveolar macrophages (HAMs) obtained from bronchoalveolar lavage. These studies were undertaken to compare ionic current expression in the in vitro differentiated macrophage to that of a human tissue macrophage. HAMs are the major population of immune and inflammatory cells in the normal lung and are the most readily available source of human tissue macrophages. Of the 974 HMDMs in the study obtained from a total of 36 donors, we were able to observe the presence of the inactivating outward current (I A ) which exhibited voltage-dependent availability in only 49 (or 5%) of the cells. In contrast, whole-cell current recordings from HAMs, revealed a significantly higher frequency ofI A expression (50% in a total of 160 cells from 26 donors). In the alveolar cell, there was no correlation observed between cell size and peakI A amplitude, nor was there a relationship between peakI A amplitude and time in culture. The current in both cell types was K+ selective and 4-aminopyridine (4-AP) sensitive.I A in both cell types inactivated with a time course which was weakly voltage-dependent and which exhibited a time constant of recovery from inactivation of approximately 30 sec. The time course of current inactivation was dependent upon the external K+ concentration. An increase in the time constant describing current decay was observed in elevated K+. Current activation was half-maximal at approximately −18 mV in normal bathing solution. Steady-state inactivation was half-maximal at approximately −44 mV. The presence of the outwardly rectifying K+ conductance may alter the potential of the mononuclear phagocyte to respond to extracellular signals mediating chemotaxis, phagocytosis, and tumoricidal functions.

The action of antifungal drug, amphotericin B (AmB), on solvent-containing planar lipid bilayers made of sterols (cholesterol, ergosterol) and synthetic C14–C18 tail phospholipids (PCs) or egg PC has been investigated in a voltage-clamp mode. Within the range of PCs tested, a similar increase was achieved in the lifetime of one-sided AmB channels in cholesterol- and ergosterol-containing membranes with the C16 tail PC, DPhPC at sterol/DPhPC molar ratio ≤1. The AmB channel lifetimes decreased only at sterol/DPhPC molar ratio >1 that occurred with sterol/PC molar ratio of target cell membranes at a pathological state. These data obtained on bilayer membranes two times thicker than one-sided AmB channel length are consistent with the accepted AmB pore-forming mechanism, which is associated with membrane thinning around AmB–sterol complex in the lipid rafts. Our results show that AmB can create cytotoxic (long open) channels in cholesterol membrane with C14–C16 tail PCs and nontoxic (short open) channels with C17–C18 tail PCs as the lifetime of one-sided AmB channel depends on ~2–5 Å difference in the thickness of sterol-containing C16 and C18 tail PC membranes. The reduction in toxic AmB channels efficacy can be required at the drug administration because C16 tails in native membrane PCs occur almost as often as C18 tails. The comparative analysis of AmB channel blocking by tetraethylammonium chloride, tetramethylammonium chloride and thiazole derivative of vitamin B1, 3-decyloxycarbonylmethyl-4-methyl-5-(2-hydroxyethyl) thiazole chloride (DMHT), has proved that DMHT is a comparable substitute for both tetraalkylammonia that exhibits a much higher affinity.

The cardiac L-type Ca2+ channel current (I Ca,L) plays an important role in controlling both cardiac excitability and excitation–contraction coupling and is involved in the electrical remodeling during postnatal heart development and cardiac hypertrophy. However, the possible role of endothelin-1 (ET-1) in the electrical remodeling of postnatal and diseased hearts remains unclear. Therefore, the present study was designed to investigate the transcriptional regulation of I Ca,L mediated by ET-1 in neonatal rat ventricular myocytes using the whole-cell patch-clamp technique, quantitative RT-PCR and Western blotting. Furthermore, we determined whether the extracellular signal–regulated kinase 1/2 (ERK1/2) pathway is involved. ET-1 increased I Ca,L density without altering its voltage dependence of activation and inactivation. In line with the absence of functional changes, ET-1 increased L-type Ca2+ channel pore-forming α1C-subunit mRNA and protein levels without affecting the mRNA expression of auxiliary β- and α2/δ-subunits. Furthermore, an actinomycin D chase experiment revealed that ET-1 did not alter α1C-subunit mRNA stability. These effects of ET-1 were inhibited by the ETA receptor antagonist BQ-123 but not the ETB receptor antagonist BQ-788. Moreover, the effects of ET-1 on I Ca,L and α1C-subunit expression were abolished by the ERK1/2 inhibitor (PD98059) but not by the p38 MAPK inhibitor (SB203580) or the c-Jun N-terminal kinase inhibitor (SP600125). These findings indicate that ET-1 increased the transcription of L-type Ca2+ channel in cardiomyocytes via activation of ERK1/2 through the ETA receptor, which may contribute to the electrical remodeling of heart during postnatal development and cardiac hypertrophy.