Synaptic vesicle pools are a major hidden resting metabolic burden of nerve terminals

Abstract

The brain is a metabolically fragile organ as compromises in fuel availability rapidly degrade cognitive function. Nerve terminals are likely loci of this vulnerability as they do not store sufficient ATP molecules, needing to synthesize them during activity or suffer acute degradation in performance. The ability of on-demand ATP synthesis to satisfy activity-driven ATP hydrolysis will depend additionally on the magnitude of local resting metabolic processes. We show here that synaptic vesicle (SV) pools are a major source of presynaptic basal energy consumption. This basal metabolic processes arises from SV-resident V-ATPases compensating for a hidden resting H⁺ efflux from the SV lumen. We show that this steady-state H⁺ efflux (i) is mediated by vesicular neurotransmitter transporters, (ii) is independent of the SV cycle, (iii) accounts for up to 44% of the resting synaptic energy consumption, and (iv) contributes substantially to nerve terminal intolerance of fuel deprivation.

Fig. 1.. SV V-ATPase, not the plasma membrane Na⁺/K⁺-ATPase, is a primary energy burden in resting synapses.

(A) Syn-ATP fluorescence (F) (top) and luminescence (L) (bottom) images acquired from primary hippocampal neurons in glucose (left) and 25 min after replacing glucose with 2DG (right) in the presence of TTX (five frames average taken at 20 to 25 min). Scale bar, 10 μm. (B and D) Average ATP_presyn (L/F intensity ratio) time course normalized to baseline measured in glucose. (B) Ensemble average time course from neurons incubated in 2DG (n = 13; gray trace) or incubated in 2DG + ouabain [1 mM] (n = 14; blue trace). (C) Average of ΔATP_presyn (ΔL/F) values after 25 min in 2DG (2DG_25min) in the presence of ouabain (blue dots) normalized to control (gray dots): mean ± SEM: 0.97 ± 0.065 versus 1 ± 0.037. (D) Ensemble average time course from neurons incubated in 2DG (n = 19; gray trace) or incubated in 2DG + bafilomycin [1 μM] (n = 21; red trace). (E) Average ΔATP_presyn (ΔL/F) values after 25 min in 2DG (2DG_25min) in the presence of bafilomycin (red dots) normalized to control (gray dots): mean ± SEM: 0.56 ± 0.093 versus 1 ± 0.1. Error bars indicate SEM. **P < 0.01, Wilcoxon-Mann-Whitney test. n.s., not statistically significant.

Fig. 2.. Resting SVs have a constant H⁺ efflux.

H⁺ flux from SVs measured in hippocampal neurons expressing vG-pH. (A) Average vG-pH traces in response to 100 APs (10 Hz) in control neurons (Ctrl, gray trace, n = 13) or in neurons where exocytosis is genetically suppressed by either ablating expression of Munc13 (Munc13-KD, red trace, n = 6) or expressing TeNT (blue trace, n = 6). ΔF values are normalized to maximal ΔF from NH₄Cl treatment (ΔF/ΔF_NH4Cl). (B) vG-pH average traces of Ctrl, Munc13-KD and TeNT neurons measured in the presence of TTX before and after perfusion with bafilomycin. Fluorescence expressed as a percentage of the fluorescence expected for pH 6.9 (% F_pH6.9) based on perfusion with NH₄Cl (see Methods). ±SEM intervals are indicated by shaded colored areas. (C) Bafilomycin application causes SV vG-pH fluorescence to increase at the same rate in all three conditions and thus is not related to exocytosis. Average rates of alkalization for Ctrl, Munc13-KD, and TeNT neurons measured over the first 6 min in bafilomycin, respectively, as 0.117% F_pH6.9/s ± 0.01, 0.124% F_pH6.9/s ± 0.018, and 0.119% F_pH6.9/s ± 0.016.

Fig. 3.. SV H⁺ efflux is not correlated with individual synapse release properties.

(A) Representative images showing variation in synaptic release in an individual neuron expressing vG-pH. Perfusion with NH₄Cl perfusion [(a) pseudo-colored green]reveals expression pattern of vG-pH across synapses. vG-pH responses to electrical activity [100 APs (10 Hz); (b) pseudo-colored red] shows that a portion of the nerve terminals are silent [(c) overlay between left and middle images] indicated by circles. (B) Average vG-pH response to stimulation from active (black), silent (red), and total synapses (gray) per neuron (n = 13); where silent synapse is defined as showing responses less than 1.2 SD of the prestimulus baseline. (C) Percentage of nerve terminals in each category on a neuron-by-neuron basis (gray dashed lines) shows that ~80% of nerve terminals show robust responses and 20% are silent. (D) Ensemble average of individual neuron responses of SV H⁺ efflux kinetics measured in TTX and bafilomycin for active and silent synapses across 13 neurons (active: 0.12% F_pH6.9/s ± 0.013 and silent: 0.13% F_pH6.9/s ± 0.021; P = 0.24). (E) Show that the SV H⁺ efflux is unrelated to individual bouton exocytic properties [active synapses: mean ± SD: 0.12% F_pH6.9/s ± 0.12 (n = 1035) and silent synapses: 0.12% F_pH6.9/s ± 0.14 (n = 239)].

Fig. 4.. Vesicular neurotransmitter transporters (i.e., vGlut1 and vMAT2) mediate a large fraction of the resting SV H⁺ efflux.

(A and B) SV H⁺ efflux was measured in the presence of TTX and bafilomycin in hippocampal neurons expressing either Syphy-pH (control, gray trace), Syphy-pH and an shRNA suppressing expression of vGlut1 (vGlut1-KD, green trace) or Syphy-pH, vGlut1-KD and a human variant of vGlut1, resistant to the ShRNA (vGlut1-rsc, pink trace). (A) Control, vGlut1-KD, and vGlut1-rsc show similar exocytic responses to electrical stimulation (inset), but vGlut1-KD neurons show much lower SV H⁺ efflux. (B) SV H⁺ efflux rates in Ctrl: 0.12% F_pH6.9/s ± 0.016 (n = 12) versus vGlut1-KD: 0.052% F_pH6.9/s ± 0.006 (n = 11), ***P < 0.001. SV H⁺ efflux is fully recovered in neurons where vGlut1 was rescued [0.12% F_pH6.9/s ± 0.022 (n = 6) versus vGlut1-KD ***P < 0.001]. (C) Hippocampal neuronal expression of the exogenous transporter vMAT2, either coupled to pHluorin (vMAT2-pH, blue trace, n = 7) or coexpressed with Syphy-pH (vMAT2, violet trace, n = 17) led to a faster SV H⁺ efflux in the presence of bafilomycin, compared with control neurons expressing native vGlut1 (Syphy-pH, gray trace, n = 12). This increase in H⁺ flux is completely abolished by application of 100 nM reserpine (red trace, n = 8). (D) Bar plot of the SV H⁺ efflux rates of vMAT2-pH (0.34% F_pH6.9/s ± 0.043), vMAT2 (0.28% F_pH6.9/s ± 0.033), reserpine (0.124% F_pH6.9/s ± 0.029) and control (0.12% F_pH6.9/s ± 0.016). vMAT2-pH versus: Ctrl, ***P < 0.001 and reserpine, **P < 0.01. vMAT2 versus: Ctrl, ***P < 0.001 and reserpine, **P < 0.01. Error bars indicate SEM. **P < 0.01 and ***P < 0.001, Wilcoxon-Mann-Whitney test.

Fig. 5.. SV H⁺ efflux mediated by vGlut1 is compensated by the action of the v-ATPase.

(A) Resting ATP_presyn consumption measured using Syn-ATP in TTX and 2DG expressed as normalized L/F in vGlut1-KD (green trace; n = 12), vGlut1-KD + vGlut1-rsc (pink trace; n = 7), and control neurons (gray trace; n = 19). (B) Suppression of vGlut1 expression lowers the resting ATP consumption by 33% (measured after 25 min in 2DG, 2DG_25min) in vGlut1-KD neurons (green dots) normalized to control neurons (gray dots): mean ± SEM: 0.69 ± 0.056 versus 1 ± 0.1 (*P < 0.02). Resting ATP_presyn consumption is fully rescued compared to Ctrl levels (vGlut1-rsc: mean ± SEM: 1.05 ± 0.052, P = 0.6). Error bars indicate SEM. *P < 0.02 and ***P < 0.001, Wilcoxon-Mann-Whitney test.

Fig. 6.. Basal SV energy consumption affects synaptic function in hypometabolic conditions.

(A and B) Synaptic function measured in hippocampal neurons expressing Syphy-pH in the absence of glucose (0 mM glucose). Color-coded traces represent responses to one stimulus round (50 APs, 10 Hz) applied every minute (from red = first round to violet = ninth round). Responses are normalized to the peak. Vertical dashed line represents three times τ measured at the first round. Horizontal dashed line represents 50% retrieval of the exocytic signal. (A) Traces to six stimulus rounds in a control neuron. (B) Traces to nine stimulus rounds in a neuron suppressing expression of vGlut1 (vGlut1-KD). (C) Percentage of endocytic block from the examples shown in (A) and (B), measured each round at 3τ time point. Suppression of vGlut1 expression significantly prolongs the ability to recycle SVs in restricted fuel conditions. (D) Number of rounds of stimulation before endocytic block exceeds 50% is smaller in neurons expressing vGlut1 (gray dots, n = 12) than in vGlut1-KD neurons (red dots, n = 13): mean ± SEM: 5.92 rounds ± 0.52 versus 8.31 rounds ± 0.67. Error bars indicate SEM. **P < 0.01, Wilcoxon-Mann-Whitney test.