Fig. 2: JEDI-1P has improved sensitivity, kinetics, brightness, and photostability under one-photon illumination in vitro.

a–c JEDI-1P has greater steady-state response amplitude to 1-s step depolarizations than ASAP2s and ASAP3. Fluorescence was acquired at ~1 kHz. n = 6 (ASAP2s), n = 7 (ASAP3) and n = 5 (JEDI-1P) HEK293A cells. a Mean response to voltage steps. b Quantification of (a). c Responses to 100-mV step voltages (−70 mV to +30 mV). p = 3.6 × 10⁻¹¹ (One-way ANOVA). d–f JEDI-1P shows larger and faster responses to short voltage transients compared with ASAP2s and ASAP3. Fluorescence was acquired at 1 kHz. n = 10 (ASAP2s), n = 8 (ASAP3), and n = 11 (JEDI-1P) HEK293A cells. d Left: mean response to spike waveforms simulating APs (2-ms full-width at half-maximum, +30 mV peak, baseline at –70 mV). Right: mean response to a 100-Hz spike train waveform. e JEDI-1P has a larger peak response amplitude to single AP waveforms than ASAP2s and ASAP3. p = 7.0 × 10⁻¹⁵ (One-way ANOVA). f JEDI-1P responds more quickly to single AP waveforms than ASAP2s and ASAP3, as shown by narrower optical responses. The width corresponds to the full-width at half-maximum. p = 8.4 × 10⁻⁸ (Welch’s One-way ANOVA). Pairwise tests with Dunnett’s T3 multiple comparisons test (two-tailed). g 10-min fluorescence trace of JEDI-1P, ASAP2s, and ASAP3 illuminated with 470/24-nm light at an irradiance of 3.2 mW/mm² at the sample plane, normalized to the brightness of ASAP2s in the first frame. n = 8 independent transfections per GEVI. h JEDI-1P is brighter than ASAP2s and ASAP3. Brightness was quantified as described in Fig. 1h. n = 4 independent transfections per GEVI. p = 1.7 × 10⁻⁷ (One-way ANOVA). i JEDI-1P displays larger relative photon budget (area under the curve of the normalized brightness, timecourses of 2 min) than ASAP2s and ASAP3 over a > 10-fold range of irradiance levels (3.2–46 mW/mm²). n = 12 (ASAP2s), n = 11 (ASAP3) and n = 12 (JEDI-1P) independent transfections per GEVI. j JEDI-1P has an excitation peak at 487 nm and an emission peak at 509 nm. n = 6 (EGFP) and 5 (JEDI-1P) independent transfections in HEK293-Kir2.1 cells. k JEDI-1P efficiently traffics to the plasma membrane in the soma and dendrites. Representative confocal image acquired from a DIV14 rat cortical neuron. Scale bars, 10 μm. l Single-trial recording of a representative dissociated cortical neuron (DIV12) expressing JEDI-1P-Kv. Upper: membrane voltage recording from whole-cell current clamp in response to a 100-pA current injection. Lower: voltage imaging of JEDI-1P. m–o JEDI-1P-Kv detects APs with larger response amplitudes and greater signal-to-noise ratio (SNR) than ASAP3-Kv in dissociated hippocampal neurons (DIV14-16). n = 5 neurons per variant. Experiments were done at 32–35 °C. m Mean fluorescence response (lower panel) to current-triggered APs (upper panel). Triggering was performed using 50–400 pA injected over 20 ms. n JEDI-1P-Kv produced larger fluorescence responses to APs than ASAP3-Kv. p = 0.0079 (Mann–Whitney U test, two-tailed). JEDI-1P-Kv’s responses exhibited a higher SNR for AP detection than ASAP3-Kv. p = 0.0159 (Mann–Whitney U test, two-tailed). All panels: ****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05; n.s. p > 0.05. Tukey’s HSD multiple comparison test (two-tailed) was used unless otherwise noted. Error bars or shading denote the 95% confidence interval (CI) of the mean. Experiments were done at room temperature unless otherwise noted.