Connexin 43 (Cx43) hemichannels are a subset of connexins that open in physiological and pathological conditions. Compared with other connexins, the biophysical properties of Cx43 hemichannels are not yet well characterised. We present a methodology for quantitatively selecting electrophysiology recordings suitable for model-based analysis of single Cx43 hemichannel activity. Human HeLa cells stably expressing a bicistronic Cx43IRESeGFP transcript were used for whole-cell patch clamp recordings. Single Cx43 hemichannel currents were evoked with voltage steps from 40 to 80 mV in 10 mV increments (30 s), in low extracellular Ca2+ solutions. Matlab was used to fit a Gaussian mixture model to the all-points histogram at each voltage step. The number of cells with 1/2/3 component Gaussians varied from 11/1/0 (n=12) at 40mV to 3/6/3 at 80mV (n=12). A component standard deviation (S.D.) much larger than the noise S.D indicates a low quality trace with excessive activity and/or macroscopic currents. The high quality traces at 80mV (n=6) had an average separation of 54.8 pS between distinct states. No distinct states were present in Cx43IRESeGFP HeLa cells treated with a generic connexin blocker LaCl3 (100 µM; n=3) or in non-transfected HeLa cells (n=2) across all voltages. This study develops a quantitative method for accepting whole-cell patch clamp recordings to be used for single Cx43 hemichannel analysis. Our findings will provide the necessary groundwork for a model-based analysis of connexin mimetic peptide mechanisms of action.
Supported by University of Auckland Doctoral Scholarship to YK, and the Department of Ophthalmology