A new electro-optical approach for conductance measurement: an assay for the study of drugs acting on ligand-gated ion channels

Ligand gated ion channels are involved in many pathophysiological processes and represent a relevant, although challenging, target for drug discovery. We propose an innovative electro-optical approach to their analysis able to derive membrane conductance values from the local membrane potential changes imposed by test current pulses and measured by fast voltage-sensitive fluorescent dyes. We exploited the potential of this proprietary method by developing a drug testing system called “ionChannel Optical High-content Microscope” (ionChannelΩ). This automated platform was validated by testing the responses of reference drugs on cells expressing different ligand-gated ion channels. Furthermore, a double-blind comparison with FLIPR and automated patch-clamp was performed on molecules designed to act as antagonists of the P2RX7 receptor. ionChannelΩ proved highly reliable in all tests, resulting faster and more cost-effective than electrophysiological techniques. Overall, ionChannelΩ is amenable to the study of ligand gated ion channels that are receiving less attention due to limitations in current assays.


Supplementary-figure 1
Cell membrane  The upper panel in a shows the equivalent circuit model of a plasma membrane with the pathways of the currents flowing through a cell exposed to an electric field. A square current pulse (I), generated by parallel plate electrodes, induces capacitive currents that hyperpolarize one side of the membrane and depolarize the opposite one. As membrane potential becomes more negative or positive, ionic currents are driven to re-establish the electrochemical equilibrium. The overall consequence of this process is that the change in membrane potential, during the electrical stimulation, reaches a new stable value that depends on the membrane resistance (see V simulated traces). In fact, if the membrane resistance (control resistance, R c ) is reduced by effect of a drug (R d ), the membrane potential variation is more effectively counteracted by ionic currents and a lower plateau is observed for both hyperpolarization and depolarization (compare green with purple V traces). These changes in membrane potential can be monitored by di-4-ANEPPS, a fast VSD that increases its fluorescence upon hyperpolarization and decreases it in the presence of depolarization (see the F simulated traces). The images of fluorescence are acquired at two specific times, highlighted by the light blue intervals (F R ): at rest (F RC or F RD , i.e. in the absence, or presence of a drug, respectively) and during the plateau phase of the stimulus (F SC or F SD ). Accordingly, changes in fluorescence can be measured as the difference (ΔF) between these couple of values: ΔF c or ΔF d , whether in the absence or in the presence of a drug affecting resistance.
Panel b demonstrates the capability of di-4-ANEPPS, a fast VSD, to follow the kinetics of membrane potential changes imposed by a square current pulse in CHO cells. Fluorescence intensity values (dots on the right graph) were collected at high rate (2000 frame/s) and measured within the square highlighted in the image on the left. The response of the membrane potential to the square pulse was as expected from a typical resistor-capacitor (RC) circuit (red line).

Supplementary-figure 2
Frequency Panel a shows the simulation of the frequency distributions of positive and negative ΔF/F R values, before (blue line) and after (red line) administration of a test solution (KRH, in "control experiments" and a drug able to determine an increase in membrane permeability, in "drug testing experiments" (left and right panels respectively).
In controls, the ΔF/F R values recorded in the reference passage are reproduced also in the test passage with a frequency distribution that is given by the superposition of two bell-shaped patterns crossing to zero (a, left panel). If a drug acting on an ion channel is added at the test passage, both positive and negative ΔF/F R values are reduced, giving rise to a single narrow peak around zero (a, right panel).
In a scatter plot, the correlation values for pairs of ΔF/F R (obtained from individual pixels, first at the reference and then at the test passage) are, ideally, laying on a line, whose slope provides a measure of the change in membrane resistance (b). In control experiments, reference and test ΔF c /F Rc values (blue dashed arrow lines) are, ideally, identical and the slope value is 1 (blue line).
In a drug testing experiment, in which membrane resistance is changed, the ΔF d /F Rd test values are reduced with respect to the ΔF c /F Rc reference values (red dashed arrow line); as a consequence, the lower is the resistance, the closer to zero is the slope value (red line). Changes in ΔF c /F Rc values are expressed in percent. Note that when channels have a ohmic behaviour, slope values are the same in both depolarization and hyperpolarization. Analysis: The raw conductance in voltage clamp was obtained by the relation I (+40mV) / ΔV where I (+40mV) is the current amplitude measured at +40 mV, upon application of the protocol 1, and ΔV is (+40 -E rev. ) (reversal potential of the whole-cell currents).
In current clamp mode, the absolute Vm value was averaged in the last 30 ms of the injection step and then subtracted from the basal Vm measured at the beginning of the protocol. Membrane conductance was calculated as the ratio between ΔVm in vehicle and ΔVm in the presence of increasing concentrations of capsaicin.