Ion modulation of membrane permeability: Effect of cations on intact cells and on cells and phospholipid bilayers treated with pore-forming agents

Abstract

Leakage of ions (Na+, K+) and phosphorylated metabolites (phosphorylcholine, 2-deoxyglucose 6-phosphate) through membrane lesions in intact cells or in cells modified by 'pore-forming' agent has been studied. Leakage from intact cells is induced by protons and by divalent cations such as Cu2+, Cd2+ or Zn2+. Leakage from agent-modified cells--or across phospholipid bilayers modified by agent--is prevented by low concentrations of the same cations and by higher concentrations of Ca2+, Mn2+ or Ba2+; Mg2+, dimethonium, spermine, or spermidine are virtually ineffective. The relative efficacy of a particular cation (e.g. Ca2+) depends more on cell type than on the nature of the pore-forming agent. The predominant effect is on binding of cation to specific sites, not on surface charge. Surface charge, on the other hand, does affect leakage from agent-modified cells in that suspension in nonionic media reduces leakage, which can be restored by increasing the ionic strength: univalent (Na+, K+, Rb+, NH4+) and divalent (Mg2+, dimethonium) cations are equally effective; addition of protons or divalent cations such as Zn2+ to this system inhibits leakage. From this and other evidence here presented it is concluded that leakage across membranes is modulated by the presence of endogenous anionic components: when these are in the ionized state, leakage is favored; when unionized (as a result of protonation) or chelated (by binding to divalent cation), leakage is prevented. It is suggested that such groups are exposed at the extracellular face of the plasma membrane.

Action of diphtheria toxin does not depend on the induction of large, stable pores across biological membranes

“…Rapid leakage of Na + and K + does occur ( Fig. 2; Papinini et al, 1988;Sandvig & Olsnes, 1988), which indicates that failure to observe metabolite or anion leakage is not merely the result of pores being shut at the low pH (Bashford et al, 1986(Bashford et al, , 1988a. Hence there is no evidence for a stable channel of > 1.8 nm in diameter being formed in cells.…”

“…The efficacy of Cd 2+ over Ca 2+ or Zn 2+ as a protectant is not related to a special property of Vero cells. Vero cells treated with S. aureus a toxin, for example, are protected against Na+/K + leakage (and against phosphoryl[3H]choline leakage) by Zn 2+ > Cd 2+ > Ca 2+, just as are Lettre cells and erythrocytes (Bashford et al, 1988a).…”

“…After 15 rain at 37~ toxin was removed and replaced by low pH medium (150 mM NaCI, 5 mM KC1, 10 mM MES, pH 4.5; MES-BS) and incubated at 37~ for 5 rain; under these conditions, bound diphtheria toxin enters the cells (Sandvig & Olsnes, 1980 (see below) were washed three times with 150 mM choline chloride and extracted with 0.1% triton X-100 in 0.1 M LiNO3 in H2804, and Na + and K" content measured by atomic absorption. Leakage of K § from damaged cells is accompanied by leakage of Na + into cells, with little change in the total cellular content of Na-plus K + (Bashford et ai., 1985;1988a). A convenient index of cation leakage is the relative Na + content of the cells, namely Na+/(Na + + K +) which has a value close to 0.13 for control Veto cells (untreated with diphtheria toxin) and which can rise to a maximum value of 0.97 when the intracellular and extracellular pools of Na + and K + have equilibrated.…”

“…Conductivity was measured as described by Bashford et al (1988a). Diphtheria toxin was added to the bathing solution on one side (cis side); that compartment was connected to the virtual ground and voltage signs are referred to it.…”

“…The sonicate, which contained mainly small, unilamellar vesicles, was applied to a column of Sephadex G50 and eluted with 160 mM NaC1, 10 mM HEPES, pH 7.0 (NaOH); the fractions used for leakage experiments contained >80% entrapped calcein. Planar lipid bilayers of asolectin were prepared as previously described (Bashford et al, 1988a).…”