Resting state Orai1 diffuses as homotetramer in the plasma membrane of live mammalian cells

Abstract

Store-operated calcium entry is essential for many signaling processes in nonexcitable cells. The best studied store-operated calcium current is the calcium release-activated calcium (CRAC) current in T-cells and mast cells, with Orai1 representing the essential pore forming subunit. Although it is known that functional CRAC channels in store-depleted cells are composed of four Orai1 subunits, the stoichiometric composition in quiescent cells is still discussed controversially: both a tetrameric and a dimeric stoichiometry of resting state Orai1 have been reported. We obtained here robust and similar FRET values on labeled tandem repeat constructs of Orai1 before and after store depletion, suggesting an unchanged tetrameric stoichiometry. Moreover, we directly visualized the stoichiometry of mobile Orai1 channels in live cells using a new single molecule recording modality that combines single molecule tracking and brightness analysis. By alternating imaging and photobleaching pulses, we recorded trajectories of single, fluorescently labeled Orai1 channels, with each trajectory consisting of bright and dim segments, corresponding to higher and lower numbers of colocalized active GFP label. The according brightness values were used for global fitting and statistical analysis, yielding a tetrameric subunit composition of mobile Orai1 channels in resting cells.

FIGURE 1.

Orai1-Orai1 tandem repeat constructs in HEK-293 cells. A and B, whole-cell patch clamp experiments on HEK-293 cells expressing STIM1 and either the monomeric Orai1 subunit (black) or the Orai1-Orai1 tandem construct (red). Stores were depleted passively by using 10 mm EGTA in the pipette solution, yielding comparable currents for both constructs. A, time course of mean currents recorded at −74 mV during voltage ramps (n = 6 for Orai1-Orai1 dimer, n = 9 for Orai1). B, I-V curve recorded 150 s after break in. C, HEK-293 cells expressing STIM1 and the two tandem repeat constructs CFP-Orai1-Orai1 and YFP-Orai1-Orai1 showed comparable FRET before and after store depletion via 2 μm thapsigargin (upper and lower panels, respectively). D, time course of FRET for HEK-293 cells expressing CFP-and YFP-labeled Orai1-Orai1 constructs with (blue) and without (red) coexpression of STIM1. Thapsigargin was added after 2 min. All error bars denote standard errors of the mean.

FIGURE 2.

Characterization of the Orai1 functional state and mobility in T24 cells. A, passive store depletion via BHQ in cells expressing both CFP-STIM1 and YFP-Orai1 led to the formation of Orai1 puncta in the plasma membrane (top row). In our standard T24 cell system stably transfected only with Orai1-mGFP, we did not observe this effect (bottom row). Images were recorded directly before and 9 min after addition of 30 μm BHQ. B, whole-cell patch clamp experiments of wild-type T24 cells (white) and T24 cells stable expressing Orai1-mGFP (gray) yielded similar results. Extracellular and intracellular solution contained 10 and 100 nm Ca²⁺, respectively. Bars show mean values and S.E. (n = 4 for both cell lines). Currents measured directly after break in (t = 0 s), and also 100 s before (t = 200 s) and after (t = 400 s) the addition of 10 μm La³⁺. C, mobility analysis of Orai1-mGFP in T24 cells. The mean square displacement increases linearly with time lag, indicating free diffusion with a diffusion coefficient of D = 0.13 μm²/s. FRAP experiments (inset) revealed a mobile fraction of Orai1-mGFP of 91 ± 9% (n = 11 cells).

FIGURE 3.

Experimental strategy. A, excitation protocol used for stoichiometry measurements. Imaging is performed at low excitation power (pulses i, ii, iii … ). High power laser pulses are used for photobleaching of Orai1-mGFP. Pulses α₁ and α₂ completely destroy all active fluorophores within the photobleached area. The pulse β is shorter and results in only partial bleaching. Recording of images for the brightness analysis starts with readout pulse iii. B, scheme of the TOCCSL method. Overexpression of Orai1-mGFP results in a high density of labeled ion channels (i). Bleach pulse α₁ destroys all active fluorophores in a sharply defined area in a short time (ii). During the recovery time, nonbleached Orai1-mGFP pores enter the region of interest by diffusion (iii). In the center of the photobleached area molecules can be resolved and tracked. C, raw data were recorded as described in B. Intensity was downscaled by a factor of 3 for image i. D, the single channel brightness was determined on each image by integrating the photon counts in a 3 × 3 pixel region (dotted line). The background level was estimated from the surrounding 5 × 5 pixel ring. E, exemplary brightness trajectory of a single Orai1-mGFP spot. The brightness fluctuated at high levels (frames 1–11). The bleach pulse β reduced the fluorescence signal significantly. F, the illumination protocol allowed for grouping the recorded brightness values in three different data sets (see text for details).

FIGURE 4.

Subunit stoichiometry of Orai1 pores. A, brightness distributions of the three different data sets shown as blue (1), green (2), and red (3) dotted curves. Continuous lines show the fitted curves using the global fitting algorithm and a 4-Gaussian model. B, mean values (asterisks) and width (circles) of the individual Gaussians obtained using the tetramer model. The mean values show a linear increase, the widths increase with the square root (red fit curves). C, weights of the four individual Gaussians used to model data set 1 (blue). The distribution was well reproduced by a binomial distribution (red), yielding a probability of 0.88, which can be interpreted as the mGFP maturation degree. D, correlation plot of brightness values within individual trajectories. Each brightness value was plotted versus all preceding values of the same trajectory (data set 2), revealing strong correlation (Pearson's correlation coefficient of 0.65). All data were recorded at 37 °C.