Toward a unifying model of malaria-induced channel activity

Abstract

Infection of RBC by the malaria parasite Plasmodium falciparum activates, at the trophozoite stage, a membrane current 100- to 150-fold larger than in uninfected RBC. This current is carried by small anion channels initially described in supraphysiological ion concentrations (1.115 M Cl(-)) and named plasmodial surface anion channels (PSAC), suggesting their plasmodial origin. Our results obtained with physiological ion concentrations (0.145 M Cl(-)) support the notion that the parasite-induced channels represent enhanced activity versions of anion channels already present in uninfected RBCs. Among them, an 18-pS inwardly rectifying anion channel (IRC) and a 4- to 5-pS small conductance anion channel (SCC) were present in most single-channel recordings of infected membranes. The aim of this study was to clarify disparities in the reported electrophysiological data and to investigate possible technical reasons why these discrepancies have arisen. We demonstrate that PSAC is the supraphysiological correlate of the SCC and is inhibited by Zn(2+), suggesting that it is a ClC-2 channel. We show that in physiological solutions 80% of the membrane conductance in infected cells can be accounted for by IRC and 20% can be accounted for by SCC whereas in supraphysiological conditions the membrane conductance is almost exclusively carried by SCC (PSAC) because the IRC is functionally turned off.

Fig. 1.

Whole-cell patch-clamp of P. falciparum-infected RBCs. Whole-cell I–V relationships (I–V curves) were obtained by evoking a series of test potentials from −100 to +100 mV in 10-mV steps for 500 ms from a holding potential of 0 mV. Recordings A, B, and C were obtained with solutions A (150 mM NaCl plus 5 mM KCl), B (155 mM NMDG-Cl), and C (155 mM choline-Cl) in bath and pipettes. They display no significant difference and demonstrate the absence of specific effects of NMDG and choline ions in our usual experimental conditions. I–V curves plotted in D were obtained at different concentrations of chloride ions in bath and pipette corresponding to 0.115 M NaCl plus 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 M choline chloride. They are mean values of 13, 10, 11, 13, 13, 12, and seven experiments, respectively. SEM bars are not shown for clarity. (E) The corresponding cord conductances ± SEM calculated between −100 mV and −50 mV for each chloride concentration.

Fig. 2.

Inhibition of whole-cell currents by NPPB (A), furosemide (B), and Zn²⁺ (C). Inhibitors were added to the bathing solutions to obtain final nominal concentrations 1, 10, and 100 μM for NPPB, 20, 200, 500, and 1,000 μM for furosemide, and 1 mM for Zn²⁺. The percentage of inhibition was calculated on the cord conductance between −100 mV and −50 mV for each chloride concentration. Vertical bars correspond to SEM, and each calculated value was a mean of six or more I–V curves obtained by evoking a series of test potentials from −100 to +100 mV in 10-mV steps for 300 or 500 ms from a holding potential of 0 mV. Typical examples are given in C (a, b, and c).

Fig. 3.

Cell-attached single-channel recordings of SCC activity in infected RBCs at different chloride concentrations (0.115 M NaCl plus 0, 200, 400, 600, 800, or 1,000 mM choline chloride). The I–V curves presented in A were established in the typical range of −100 mV to +100 mV by steps of 10 mV. I–V curves were constructed from mean ± SEM values obtained from 12, six, six, 12, six, and 12, respectively, complete curves. For clarity's sake 0.115 M and 1.115 M curves are presented in full. Intermediary curves are presented by the linear regression dashed lines. B shows the corresponding cord conductances (g) ± SEM calculated between −100 mV and −50 mV for each chloride concentration. Typical recordings obtained at −100 mV (C) show bursts of channel openings separated by brief closings. D and E indicate the mean ± SEM number of channels (n) present on membrane patches and the open probability (Po) calculated from 1-min recordings at 0.115 M (n = 12), 0.715 M (n = 12), and 1.115 M (n = 12) chloride concentrations.

Fig. 4.

Cell-attached single-channel recordings of IRC activity in infected RBCs at different chloride concentrations (0.115 M NaCl plus 0, 200, 400, 600, 800, or 1,000 mM choline chloride). The I–V curves presented in A were established in the typical range of −100 mV to +100 mV by steps of 10 mV. I–V curves were constructed from mean ± SEM values obtained from 15, seven, six, eight, six, and six, respectively, complete curves. For clarity's sake 0.115 M and 0.715 M curves are presented in full. Other curves are presented by the linear regression dashed lines. B shows the corresponding cord conductances (g) ± SEM calculated between −100 mV and −50 mV for each chloride concentration. Typical recordings obtained at −100 mV (C) show bursts of channel openings separated by brief closings and similar current amplitude at 0.715 and 1.115 M. D and E indicate the mean ± SEM number of channels (n) present on membrane patches and the open probability (Po) calculated from 1-min recordings at 0.115 M (n = 12), 0.715 M (n = 8), and 1.115 M (n = 6) chloride concentrations.

Fig. 5.

Estimation of the relative contribution of SCC and IRC to the whole-cell global membrane conductance (G, measured between −100 and −50 mV). This graph was obtained by plotting, for each chloride concentration, the mean single-channel conductance (g) multiplied by the mean number of channels present on the patch (n) and by the corresponding open probability (Po) for SCC (doted line) and IRC (dashed line) and the sum of both (solid line). It shows the increasing contribution of SCC when chloride concentration increases and the opposite evolution of IRC.