Fast Optical Investigation of Cardiac Electrophysiology by Parallel Detection in Multiwell Plates

Figures

• 

  FIGURE 1. MULTIPLE optomechanical system design. (A) Optical scheme of MULTIPLE platform. The excitation system combines a red LED followed by a band-pass filter (625/50 nm) with a blue LED using a large-area dichroic beam mounted on a kinematic fluorescence cube (DFM2/M, Thorlabs). A divergent lens combined with an optical diffuser is used to homogenize the light intensity irradiating the multiwell plate. The fluorescence signal is filtered with a long-pass filter and focused into a sCMOS camera sensor through a camera lens. (B) 3D mechanical scheme of MULTIPLE characterized by a final footprint of 450 × 450 × 900 mm. The illumination, detection and sample holder blocks are vertically aligned on a construction rail through drop on rail carriages.

• 

  FIGURE 2. Illumination and detection performances of MULTIPLE. Light intensity distribution across the whole 24-well plate for (A) blue and (B) red LED light, respectively, using maximum LED power. The intensity of both LEDs radially decreases moving away from the optical axis causing a more than 50% reduction for the outermost wells. (C) Image of the 24-well plate reporting for each well the accessible area exploitable for ROI measurements. Detection losses registered at the peripheral wells are introduced by the non-telecentric camera adopted in the optical scheme.

• 

  FIGURE 3. Optical recording of optogenetically induced APs in channelrhodopsin-expressing HL-1 cells. (A) Representative confocal microscope images of HL-1 cells expressing blue light-activable ion channels after transduction with a commercial AAV serotype capsid 9-pseudotyped AAVV encoding mCherry-tagged ChR2(H134R) (red channel) and a custom-made vesicular stomatitis G protein-pseudotyped LV encoding eGFP-tagged CheRiff expression (green channel). Cell nuclei were stained with DAPI (blue channel). The higher eGFP signal attests to the more efficient transduction of the HL-1 cells with the LV than with the AAVV. Scale bar = 25 μm. (B) Representative traces of optically induced APs registered in AAVV- and LV-transduced HL-1 cell layers after loading of the cells with VSD and stimulating with blue light pulses at 1 Hz (blue lines in the graph). Consistent with the higher transduction efficiency of the HL-1 cells, APs were registered only in the LV-transduced HL-1 cells. CTRL, untransduced HL-1 cells.

• 

  FIGURE 4. Signal-to-noise ratio versus detection area. Representative optical recording of optogenetically induced APs collected from different-sized surface areas (2 Hz stimulation frequency with blue light pulses duration of 30 ms). Reduction of the detection area from the whole well area (≈190 mm2) to 0.17 mm2, reduced the signal-to-noise ratio from ≈ 600 to ≈ 2. The estimated number of cells present in the different-sized detected area is indicated in the right column.

• 

  FIGURE 5. AP kinetics versus optogenetic pulse duration. APA and AP duration (i.e., APD50 and APD90) values registered from LV-transduced HL-1 cells seeded in 24-well plates and irradiated at maximum light intensities and for the pulse durations indicated at the top of the figure (1 Hz stimulation frequency). Average ± standard deviation of APAs and AP durations were calculated for all 24 wells in the plate and for each experimental condition. Representative traces taken from wells placed at different distances from the optical axis are shown for each stimulation time. By increasing pulses duration from 10 to 50 ms, interwell differences decreased for all the AP parameters considered, thus overcoming the spatial heterogeneity induced by the simplicity of the implemented optical scheme.

• 

  FIGURE 6. Interwell variability of optogenetically induced APs. Interwell variability of APA and AP duration as a function of the light pulse duration. Upon increasing the optical stimulus from 10 to 50 ms, decreasing trends in the interwell variability of APA and AP duration (i.e., APD50 and APD90) were observed, confirming that differences in AP triggering due to non-uniform illumination can be overcome by stimulating with pulses ≥ 30 ms.

• 

  FIGURE 7. Patch-clamp recording of AP in HL-1 cells. (A) Representative AP recording on a HL-1 cells using single cell patch-clamp (whole-cell configuration). APs were elicited at 0.2 Hz in current-clamp mode. (B) Table reporting average ± standard error of maximal diastolic potential (MDP), APA, APD50, and APD90 measured in patch-clamp cells (number of cells = 15). APA, APD50, and APD90 measured with MULTIPLE (pulses duration of 30, 40, and 50 ms) at the same stimulation frequency (0.2 Hz) are reported for direct comparison.

• 

  FIGURE 8. MULTIPLE detection of AP duration with E-4031. (A) E-4031, a selective Kv11.1 channel blocker, was applied at 7 different concentrations ranging from 0 to 10 μM to CheRiff-expressing HL-1 cells in a 24-well plate with each dose given to six wells distributed across the whole plate. (B) Resulting APD90 values were quantified, which attested that interwell variability was not a critical issue since similar values were registered from wells with cells exposed to the same E-4031 concentration. (C,D) Average AP traces of CheRiff-expressing HL-1 cells from 10 optical recordings (1 Hz stimulation frequency). For comparison datasets (two for each concentration) were taken from wells randomly positioned. AP duration increased in a drug dose-dependent manner, with clear APD prolongation at E-4031 concentrations higher than 1 μM.

• 

  FIGURE 9. Dose–dependent effects of E-4031 on AP kinetics. (A) APA, (B) APD30, (C) APD50, and (D) APD90 as a function of E-4031 dose.

• 

  FIGURE 10. Effect of ivabradine and carbachol on spontaneously beating hiPSC-CMs. (A) Representative fluorescence imaging of spontaneous AP propagation in a hiPSC-CM monolayer during a 520 ms recording. (B,C) Spontaneous AP traces recorded in hiPSC-CM monolayer in the absence (black lines, CTR) and presence (red lines) of ivabradine (Iva, 10 μM) or carbachol (CCh, 20 nM). Histograms represent AP number/s, expressed as mean ± standard error of the mean, before and after Iva or CCh treatment. ****p < 0.0001 CTRL vs. Iva (10 μM); ***p < 0.001 CTRL vs. CCh (20 nM).