Inhibitors of Myosin Light Chain Kinase Block Synaptic Vesicle Pool Mobilization during Action Potential Firing

FM1-43 visualization of vesicle turnover block by MLCK or myosin inhibition. a, A Nomarski image of hippocampal culture showing a typical axodendritic network. Scale bar, 3 μm. b, Fluorescence of same field after a 900 AP dye load. c, Same as in b, after a 900 AP unload. d, Same as in b after a second 900 AP load. e, Same as in c after unloading with 900 AP in the presence of 30 μm ML-9. A large fraction of the fluorescence remains in the boutons.f, The fluorescence retained in e is now released with 900 additional AP applied after ML-9 washout.

Inhibition of vesicle pool mobilization by MLCK or myosin inhibitors. The degree of inhibition from a typical experiment as in Figure 1 was measured over a population of 44 synaptic boutons. To control for possible rundown, I performed an additional control run of loading and unloading. Boutons were selected for measurement on the basis of their appearance in both control runs. A, Histogram of ML-9 inhibition in a single experiment. Inhibition is defined as (1 − ΔF^ML-9₀/ΔF^control₀), where ΔF^control₀ is the average fluorescence signal of the two control runs. B, Percentage of vesicle pool remaining after 900 AP in 30 μm ML-9, 15 μm ML-7, 25 mm BDM, or control saline. All three inhibitory conditions reduce the number of vesicles turned over during prolonged AP stimuli. Slightly longer wash-in and washout times (4 min) were used for experiments with ML-7, a more hydrophobic compound than ML-9. Each concentration was measured in at least two experiments over 40–120 boutons. C, Dose–response relationship of ML-9 inhibition shown as the fraction of the recycling vesicle pool remaining after 900 AP unloading stimuli. The maximal block by ML-9 saturates for [ML-9] >25 μmand shows a half-maximal inhibition of ∼12.5 μm. The inhibition at each concentration was measured in at least three separate experiments over a total of 75–95 boutons.

ML-9 and BDM do not block endocytosis.Schematic, The protocol used to measure the time course of endocytosis is shown. FM1-43 uptake at Δt = 25 sec after the beginning of a 100 AP stimulus train (10 Hz) normalized to uptake at Δt = 0 sec is shown for three conditions: 30 μm ML-9 (n = 38; two experiments), 25 mm BDM (n = 66; two experiments), and control saline (n = 44; two experiments). FM1-43 was applied for 1 min and rinsed for 10 min before the uptake was measured by using a long unloading train of AP as in Figure 1.

Kinetics of vesicle pool turnover during MLCK inhibition. Synaptic boutons were loaded with FM1-43, using long AP stimuli, before measuring the kinetics of release of the dye in the presence or absence of varying concentrations of ML-9 (amplitude, ΔF₀ in schematic of Fig. 1). A second round of stimulation released the remaining fluorescence after inhibitor washout (amplitude, ΔF₁ in schematic of Fig. 1). For each bouton the fluorescence was normalized to ΔF_T, the total releasable fluorescence signal from both rounds of unloading. Shown here are population averages of dye release kinetics from populations of boutons of individual experiments in control saline, 10 μm ML-9, 15 μm ML-9, and 30 μm ML-9. The 15 μm and control experiments were performed sequentially on the same boutons, whereas the data for each of the other concentrations were obtained from different experiments. The accompanying control runs for [ML-9] = 10 μm and [ML-9] = 30 μmwere identical to that for the 15 μm run (data not shown). Fluorescence intensity kinetics for each boutoni were fit by using the equationF_i(t) = ΔF₀e^{−(t−t₀)/α_i}+ ΔF₁, wheret₀ marks the start of the stimulus and ΔF₀ and ΔF₁were fixed as the total fluorescence loss during each run. The following values were obtained for the decay constants: control saline, <α> = 17.7 ± 0.4 sec (n = 39); [ML-9] = 10 μm, <α> = 21.3 ± 1.2 sec (n = 39); [ML-9] = 15 μm, <α> = 20.1 ± 1.3 sec (n = 40); [ML-9] = 30 μm, <α> = 22.8 ± 1.6 sec (n= 25), where the brackets indicate the average value obtained over all α_i, and n is the number of boutons measured for each experiment. At [ML-9] = 30 μm, certain boutons release too little fluorescence to extract kinetic parameters (some boutons are near complete inhibition at this concentration; see Fig. 3A). These were excluded from the kinetic analysis performed here. The average inhibition in these data is 22 ± 1.8% ([ML-9] = 10 μm), 35 ± 1.2% ([ML-9]= 15 μm), and 58 ± 2.4% ([ML-9] = 30 μm). The solid line showsF(t), using the average values of α obtained in each case.

MLCK inhibition increases with the number of stimuli. Dye loading assays were used to measure the impact of ML-9 inhibition for low numbers of APs. Synaptic boutons were loaded with varying numbers of AP, n, and unloaded with 900 AP to determine ΔF_n. A second run that used 900 AP loading and unloading was used to determine the fraction of the vesicle pool turned over by n AP as ΔF_n/ΔF₉₀₀.A, Fractional vesicle pool turnover in control conditions measured over many experiments at various ndisplays two kinetic regimes. Each regimen, 0–20 AP and 20–75 AP, was fit with a linear dependence on n (dashed lines). The total number of boutons for each nvaried between 26 and 148 obtained from at least two experiments for each n. B, Inhibition produced by 30 μm ML-9 for different n. The degree of inhibition was determined by measuring the amount of dye taken up byn AP in the absence (ΔF_n^control) or presence (ΔF_n^ML-9) of ML-9 sequentially in the same boutons. Inhibition is calculated as (1 − ΔF_n^ML-9/ΔF_n^control). Shown here is the average over a population that varied between 30 and 61 boutons measured in at least two experiments at eachn. Inhibition is near zero for n ≤ 20 AP and gradually increases with the increasing number of AP. A simple model to explain the use dependence of inhibition whereby vesicle pool turnover is 85% inhibited for n > 20 AP, and not inhibited for n ≤ 20 AP, is shown (thick dashed line).