Tissue-type plasminogen activator induces synaptic vesicle endocytosis in cerebral cortical neurons

Abstract

The release of the serine proteinase tissue-type plasminogen activator (tPA) from the presynaptic terminal of cerebral cortical neurons plays a central role in the development of synaptic plasticity, adaptation to metabolic stress and neuronal survival. Our earlier studies indicate that by inducing the recruitment of the cytoskeletal protein βII-spectrin and voltage-gated calcium channels to the active zone, tPA promotes Ca(2+)-dependent translocation of synaptic vesicles (SVs) to the synaptic release site where they release their load of neurotransmitters into the synaptic cleft. Here we used a combination of in vivo and in vitro experiments to investigate whether this effect leads to depletion of SVs in the presynaptic terminal. Our data indicate that tPA promotes SV endocytosis via a mechanism that does not require the conversion of plasminogen into plasmin. Instead, we show that tPA induces calcineurin-mediated dynamin I dephosphorylation, which is followed by dynamin I-induced recruitment of the actin-binding protein profilin II to the presynaptic membrane, and profilin II-induced F-actin formation. We report that this tPA-induced sequence of events leads to the association of newly formed SVs with F-actin clusters in the endocytic zone. In summary, the data presented here indicate that following the exocytotic release of neurotransmitters tPA activates the mechanism whereby SVs are retrieved from the presynaptic membrane and endocytosed to replenish the pool of vesicles available for a new cycle of exocytosis. Together, these results indicate that in murine cerebral cortical neurons tPA plays a central role coupling SVs exocytosis and endocytosis.

Keywords: endocytosis; plasmin; plasminogen; synaptic vesicles; tissue-type plasminogen activator.

Figure 1. tPA induces synaptic vesicle endocytosis

A. Representative micrographs of continuous confocal live-cell imaging monitoring of AM1-44 uptake by a presynaptic terminal of a Wt cerebral cortical neuron following 0, 1 and 2 minutes of treatment with 5 nM of tPA. Arrows in b & c point to an area (red) of AM1-44 uptake. Panel d corresponds to an electronic magnification of the area indicated in panel c. Magnification 60 X. B. Mean baseline uptake of AM1-44 in 375 presynaptic terminals located in the distal 50 μm of axons from 20 Wt cerebral cortical neurons. Lines denote SEM. *, **, and ***: p < 0.0001 compared to AM1-44 uptake after 0 and 5 minutes of incubation (two-tailed unpaired t-test). Ns: non significant. C. Mean AM1-44 uptake following 0 – 5 minutes of treatment with 5 nM of tPA in the 375 presynaptic terminals examined under baseline conditions in panel B. Values correspond to the difference between AM1-44 intensity in each examined presynaptic terminal at each time point minus background AM1-44 intensity in the same terminal before treatment with tPA. Lines denote SEM. *: p < 0.001 when intensity at each time-point is compared to its own background intensity (two-tailed unpaired t test). D. Mean AM1-44 loading into 300 presynaptic terminals from 15 Wt cerebral cortical neurons treated during 5 minutes with 5 nM of proteolytically active (atPA) or inactive tPA (itPA). Values correspond to the difference between AM1-44 intensity after 5 minutes of treatment with tPA minus the average background intensity before the beginning of the experiment. Lines denote SEM. *: p = 0.002 and **: p = 0.004 compared to baseline intensity for each group of cells (two-tailed unpaired t test). Ns: non-significant when mean AM1-44 loading by SVs of neurons treated with atPA was compared to loading by SVs of neurons treated with itPA. E. Quantification of synaptophysin (SYP) - positive puncta in the distal 50 μm of axons of Wt cerebral cortical neurons treated 0 – 60 seconds with 5 nM of tPA. n = 75 axons examined per time point. Ns: non-significant (one-way ANOVA). Lines denote SEM. F. Representative micrographs of SYP expression (red in b & d) and AM1-44 uptake (white in a & c and green/yellow in b & d) in a distal axon of Wt cerebral cortical neurons following 60 seconds of incubation with vehicle (control; a & b) or 5 nM of tPA (c & d). Magnification 40 X. G. Number of SYP/AM1-44- positive puncta in the distal 50 μm of axons from Wt cerebral cortical neurons treated 0 – 60 seconds with either 5 nM of tPA or a comparable volume of vehicle (control). * p = 0.004, ** p < 0.001 and *** p < 0.0001 compared to neurons incubated with vehicle (control) during 5, 15 or 60 seconds, respectively (one-way ANOVA). n = 75 axons examined per time-point. Lines denote SEM. H & I. Mean number of SY-AM1-44-positive puncta in the distal segment of 50 axons of Wt cerebral cortical neurons incubated during 60 seconds with 0 – 10 nM of tPA (H) or 60 seconds with either 5 nM ot tPA, or a combination of tPA and 100 nM of RAP, or with RAP alone (I). Lines denote SEM. Ns: non-significant.

Figure 2. Calcineurin mediates tPA-induced synaptic vesicle endocytosis

A. Representative micrographs of synaptophysin (SYP) expression (red in b, d, f & h) and AM1-44 uptake (white in a, c, e & g; green/yellow in b, d, f & h) in the distal axon of Wt cerebral cortical neurons treated during 60 seconds with either vehicle (control; a & b), or 5 nM of tPA (c & d), or a combination of 5 nM of tPA and 0.08 μM of cyclosporine A (CsA; e & f), or 0.08 μM of CsA alone (g & h). B. Quantification of SYP/AM1-44 - positive puncta in the distal 50 μm of axons from Wt cerebral neurons treated 60 seconds with 5 nM of tPA, alone or in combination with 0.08 μM of CsA. Lines denote SEM. *: p < 0.001 compared to untreated cells. ** p < 0.001 compared to neurons treated with tPA in absence of CsA (one-way ANOVA). n = 250 presynaptic terminals from 20 different neurons examined per condition.

Figure 3. TPA induces endocytosis of synaptic vesicles via dynamin I dephosphorylation

A. Representative micrographs of dynamin I phosphorylated at Ser 774 (p-Dyn) in the distal axon of Wt cerebral cortical neurons treated with vehicle (control) or 5 nM of tPA. B – E. Representative Western blot analysis of p-Dyn in cytosolic extracts from synaptoneurosomes from Wt cerebral cortical neurons treated 0 – 60 seconds with 5 nM of tPA (B & C), or 60 seconds with tPA either alone or in combination 0.08 μM of cyclosporine A (CsA; D & E). * in C: p < 0.001 compared to untreated cells. * in E p < 0.001 compared to cells incubated in the absence of tPA and CsA. F. Quantification of SYP/AM1-44 - positive puncta in the distal 50 μm of axons from Wt cerebral cortical neurons treated during 60 seconds with 5 nM of tPA, alone (n = 265 presynaptic terminals) or in combination with 25 μM of the dynamin inhibitor dynasore (n = 250 presynaptic terminals), or with dynasore alone (n = 230 presynaptic terminals). Lines denote SEM. * p < 0.001 compared to untreated cells and with cells treated with tPA in the presence of dynasore (one-way ANOVA).

Figure 4. Effect of tPA on the cytoskeleton of the endocytic zone

A & B. Representative Western blot analysis of proflin II expression in membrane extracts from synaptoneurosomes prepared from Wt cerebral cortical neurons treated 0 – 60 seconds with 5 nM of tPA, alone or in combination with 25 μM of the dynamin inhibitor dynasore. * p < 0.0001 compared to other experimental groups (one-way ANOVA). C. Representative heat maps of F-actin fluorescence in the distal axons of Wt cerebral cortical neurons treated 60 seconds with vehicle (control; a) or 5 nM of tPA (c). b & d correspond to electronic magnifications of the areas denoted by the arrows in a & c, respectively. D & E. Mean intensity of phalloidin fluorescence measured every 0.02 μm in optically magnified and straightened F-actin clusters from Wt cerebral cortical neurons treated during 60 seconds with 5 nM of tPA (red; n = 443 clusters) or vehicle (control; blue; n = 373 clusters). Data were pooled together to obtain an average F-Actin fluorescence intensity in each experimental group (D). Ns: non- significant. Lines denote SEM. F. Mean length of the F-actin clusters examined in C. A sub-group of neurons was treated with tPA in combination with 25 μM of the dynamin inhibitor dynasore (n = 280 clusters examined). Lines denote SEM. * p = 0.01 compared to untreated cells and 0.003 compared to neurons treated with a combination of tPA and dynasore (one-way ANOVA).

Figure 5. TPA promotes the association of SVs with F-actin clusters

A. Electronic magnification of a representative micrographs of an actin cluster located in the distal axon of a Wt cerebral cortical neuron treated 60 seconds with either vehicle-control (C; panels a – c) or 5 nM of tPA (tPA; panels d – f) in the presence of AM1-44. Black in a & d and red in c & f correspond to F-actin. Black in b & e and green-white in c & f depict AIM1-44-containing SVs. B & C. Mean AM1-44 fluorescence intensity measured in the optically straightened F-actin clusters examined in Figure 4C. Data were pooled together to obtain mean AM1-44 fluorescence intensity (panel C). Lines denote SEM. * p < 0.001 compared to control neurons (two-tailed unpaired t test). D. Mean number of AM1-44-positive puncta in the F-actin clusters examined in B & C. * p = 0.002 compared to vehicle (control)-treated neurons. Statistical analysis was performed with two-tailed unpaired t-test. E. Mean ratio AM1-44-positive puncta/length for each actin cluster examined in panels B & D (* p = 0.0001; two-tailed unpaired t test).