A few days after transfection, we observed red fluorescent somata in CA3 and bright fluorescent puncta in CA1 (Figure 1B). Ratio-sypHy was effectively targeted to boutons, which were 23-fold brighter than the axonal shafts (Figure S1A available online). The probe showed a variable level of surface expression (median fsurf = 0.20, quartile coefficient of variation, QCV: 0.59, n = 922 boutons, 25 slices), which was correlated with its expression
level (R2 = 0.43, p = 0.0024, n = 19 cells; Figures S1C and S1D). We see more stimulated individual ratio-sypHy expressing CA3 pyramidal cells and their axons by current injection in the whole-cell patch-clamp configuration. To release a sizable fraction of the total recycling pool at both high and low release probability (Pr) synapses, we evoked trains of 200 APs at 30 Hz by brief somatic current injections. This stimulation led to a robust increase in green but not in red fluorescence at individual boutons along single CA3 axons Dasatinib (Figure 1C). Differential bleaching of the red fluorescence by the imaging laser meant that only one ratiometric measurement per bouton could be performed. However, the fixed stoichiometry of green and red fluorescent proteins permitted sequential measurements to be calibrated on a separate set
of boutons. To activate green fluorescence in all vesicles for calibration, we neutralized the pH of intracellular compartments
by NH4Cl or protonophores (see below) at the end of every experiment ( Figures 1A and S1B). We corrected for photobleaching ( Figures S1E and S1F), surface-stranded protein (see the Experimental Procedures), and calculated the released fraction (RF), that is, the PLEKHM2 number of released vesicles divided by the total number of vesicles present at the synapse ( Figure 1D). This method of separating release measurements and calibration allowed us to quantify vesicle cycle parameters at a large number of consecutive boutons along individual axons irrespective of their depth in tissue. The large distance between optically recorded boutons and their somata (973 ± 109 μm, n = 12 cells) minimized potential artifacts from whole-cell dialysis. The median RF at mature SC boutons (DIV 14–28) in response to 200 APs was 28.6% ± 3.3% (average of 12 cells). In all cells tested, RFs were variable between individual boutons (average QCV 0.30 ± 0.04; n = 12 cells, 5–90 boutons each) with a distribution skewed toward smaller RFs (Figures 1E and 1F). This high variability was not due to unreliable stimulation or measurement noise because response amplitudes were reproducible for repeated trials (trial-to-trial QCV: 0.12 ± 0.01, n = 12 cells, 3 trials per bouton, 1–5 boutons each). In dissociated culture, neighboring boutons show similar release parameters (Branco et al., 2008; Murthy et al., 1997; Peng et al.