Synaptic activity and activity-dependent competition regulates axon arbor maturation, growth arrest, and territory in the retinotectal projection

Abstract

In the retinotectal projection, synapses guide retinal ganglion cell (RGC) axon arbor growth by promoting branch formation and by selectively stabilizing branches. To ask whether presynaptic function is required for this dual role of synapses, we have suppressed presynaptic function in single RGCs using targeted expression of tetanus toxin light-chain fused to enhanced green fluorescent protein (TeNT-Lc:EGFP). Time-lapse imaging of singly silenced axons as they arborize in the tectum of zebrafish larvae shows that presynaptic function is not required for stabilizing branches or for generating an arbor of appropriate complexity. However, synaptic activity does regulate two distinct aspects of arbor development. First, single silenced axons fail to arrest formation of highly dynamic but short-lived filopodia that are a feature of immature axons. Second, single silenced axons fail to arrest growth of established branches and so occupy significantly larger territories in the tectum than active axons. However, if activity-suppressed axons had neighbors that were also silent, axonal arbors appeared normal in size. A similar reversal in phenotype was observed when single TeNT-Lc:EGFP axons are grown in the presence of the NMDA receptor antagonist MK801 [(+)-5-methyl-10,11- dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate]. Although expansion of arbor territory is prevented when neighbors are silent, formation of transient filopodia is not. These results suggest that synaptic activity by itself regulates filopodia formation regardless of activity in neighboring cells but that the ability to arrest growth and focusing of axonal arbors in the target is an activity-dependent, competitive process.

Figure 1.

Expression of TeNT-Lc fused to fluorescent proteins blocks neurotransmitter release in hippocampal neurons. A, Images of axons expressing sypHy (green) and TeNT-Lc:TdT (red). B, Changes in sypHy fluorescence (ΔF/F) for control (i) and TeNT-Lc-expressing (ii) neurons in response to 40 action potentials at 20 Hz (black bar). Fluorescence profiles of individual synapses (gray traces) are shown together with the average change in fluorescence (black trace). C, Mean response profile for control neurons (black; n = 5) and neurons expressing tetanus toxin (blue; n = 4). D, Spontaneous increases in sypHy fluorescence after addition of bafilomycin (black bar) for control (i) and TeNT-Lc:TdT-expressing (ii) neurons. Fluorescence profiles of individual synapses (gray traces) are shown together with the average change in fluorescence (black trace). Note that the fluorescence is normalized to the maximum value for each trace. E, Plot showing an overlay of the increase in fluorescence after bafilomycin treatment for control (black) and TeNT-Lc-expressing (blue) neurons. Bafilomycin was added at time-point zero. Data are the same as that shown in D. Curves were fit with a single-exponential function (red), from which we established the time constant for each condition (control τ = 4.8 min; tetanus-toxin τ = 29.5 min). F, Dissociated rat hippocampal neurons transfected with an inactive mutant form of TeNT (TeNT-Lc mut:EGFP; top row) and the functional form of TeNT-Lc fused to EGFP (TeNT-Lc:EGFP; bottom row). Presynaptic function assayed by FM4-64 labeling is suppressed in neurons expressing TeNT-Lc:EGFP compared with TeNT-Lc mut:EGFP (middle column). The EGFP and FM4-64 images are merged in the right column. Scale bar, 5 μm. G, Quantification of FM4-64 data. Mean values ± SEM of 209 synapses in 13 TeNT-Lc mut:EGFP-expressing neurons and 113 synapses in 7 TeNT-Lc:EGFP-expressing neurons are shown. Statistical analysis was performed using a Mann–Whitney test (**p < 0.01). A.U., Arbitrary units.

Figure 2.

TeNT-Lc:EGFP suppresses synaptic function in vivo. A, TeNT-Lc:EGFP expression inhibits visually evoked background adaptation. In bright light, wild-type (WT) larvae (n = 20; right side) contract black pigmented melanophores in the skin. TeNT-Lc:EGFP-expressing larvae (n = 37; left side) fail to contract melanophores in bright light. B, TeNT-Lc:EGFP expression also suppresses the visually driven optokinetic response (n = 34 wild type and n = 45 TeNT-Lc:EGFP larvae), locomotor activity (n = 14 wild-type and n = 50 TeNT-Lc:EGFP larvae), and stimulus-evoked calcium transients in tectal cells (n = 8 wild-type and n = 11 TeNT-Lc:EGFP larvae). C, Optic tectum labeled with Oregon Green 488 BAPTA-1 AM. Tectal cell bodies are in the bottom right corner. Representative traces of calcium responses from wild-type (top) and TeNT-Lc:EGFP-expressing (bottom) larvae. Responses from individual cells are shown in gray, and the mean response is shown in black. D, Summary of calcium imaging data. Each data point shows the mean response from each larvae. The horizontal line indicates the median for each group.

Figure 3.

Single RGCs silenced by TeNT-Lc:EGFP fail to arrest growth. A, Single RGC axon expressing EGFP (top row) or TeNT-Lc:EGFP (bottom row) imaged at 3, 5, and 7 dpf in the optic tectum of live zebrafish larvae. Scale bar, 30 μm. B, Quantification of summed axon branch length over time. TeNT-Lc:EGFP-expressing axons fail to arrest growth at 5 dpf, resulting in significant increase in total branch length at 7 dpf compared with EGFP- and TeNT mut:EGFP-expressing axons. C, Silencing presynaptic activity with TeNT-Lc:EGFP also results in increased axonal arbor coverage area in the tectum relative to controls. D, Suppressing presynaptic activity does not significantly alter axon branch number. All graphs show mean values ± SEM from 21 EGFP-, 18 TeNT-Lc:EGFP-, and 13 TeNT-Lc mut:EGFP-expressing axons. For statistical analysis of total branch length over time, a repeated-measures ANOVA with Bonferroni's post hoc test was performed. For arbor area and branch number, statistical analysis was performed using a parametric one-way ANOVA with Bonferroni's post hoc test (ns, not significant; p > 0.05). ***p < 0.001.

Figure 4.

Time-lapse analysis of RGC axon arbor dynamics. A, Series of still images taken from 10 h time-lapse movies between 6 and 7 dpf, at imaging interval of 10 min. Axons silenced by TeNT-Lc:EGFP expression (left column) show significant net growth of some axonal branches (yellow arrowheads) and numerous transient branches (blue arrowheads). Axons expressing TeNT-Lc mut:EGFP (right column) show little or no net growth over the imaging period and far fewer transient branches. Time in minutes is indicated in the top right corner. Scale bar, 30 μm. B, Quantification of net change in total axon arbor length over the 10 h imaging period. Silenced axons grow significantly more than control TeNT-Lc mut:EGFP axons. C, Number of new branch formation events during a 10 h time lapse. Silenced axons form significantly more new branches than controls. D, Number of branches that were present in the first frame of a time lapse that persisted until the final frame, 10 h later. Branch stability is unaffected by TeNT-Lc:EGFP expression. E, Histogram of branch lifetime. Silenced axons produce many more of the shortest-lived branches, but the number of longer-lived branches is unaffected by suppressing synaptic function. B, C, and E also demonstrate that there is no significant difference in the amount of net growth or new branches formed by TeNT-Lc:EGFP-expressing axons during a 10 h period between 5 and 7 dpf and immature EGFP axons imaged for 10 h between 3 and 5 dpf (n = 4). For statistical analysis of branch lifetime, a two-way ANOVA with Bonferroni's post hoc test was performed (***p < 0.0001). For all other analyses, a nonparametric one-way ANOVA, followed by a Dunn's post hoc test was used (ns, not significant; p > 0.05). *p < 0.05; **p < 0.01. All graphs show mean ± SEM values of four EGFP-labeled axons, six TeNT-Lc:EGFP-expressing axons, and six axons expressing TeNT-Lc mut:EGFP.

Figure 5.

Axonal arbor morphology is normal when activity is blocked in all RGCs. A, Injection of transposase mRNA and a plasmid encoding TeNT-Lc:EGFP flanked by Tol2 sites into the Isl2b:GAL4 transgenic line of zebrafish results in widespread expression of TeNT-Lc:EGFP specifically in RGCs (left). Coinjection of a plasmid encoding TdT (without Tol2 sites) results in mosaic labeling of isolated RGCs (middle). The red is displayed as magenta. The red and the green channels are merged in the far right panel. Autofluorescence from the skin is indicated by the arrowhead in the left panel. B, Single axon arbor morphology visualized in the tectum of TeNT-Lc:EGFP Tol2 fish. EGFP fluorescence from widespread expression of TeNT-Lc:EGFP in RGC axons is displayed in the left panel. The limit of the tectal neuropil is marked with a dashed line. Sparse labeling in tectal cells marked with * is occasionally observed. Skin autofluorescence is marked with an arrowhead. In the example shown, a single axon is also brightly labeled with TeNT-Lc:EGFP against a background of less strongly labeled axons. A single axon labeled with TdT, displayed as magenta, is displayed in the middle panel. EGFP and TdT images are merged in the right panel. C, Branch number is unaffected by suppressing synaptic activity in a single cell or globally. D, Quantification of summed axon branch lengths at 7 dpf. Axons grown under conditions of widespread of synaptic transmission (TeNT-Lc:EGFP Tol2 and TeNT-Lc mRNA) are not statistically different from control (EGFP) axons grown under normal conditions. For reference, the value for singly silenced axons (TeNT-Lc:EGFP) is also shown. E, Arbor area is also reduced to control levels when synaptic activity is suppressed globally. All graphs show mean ± SEM values of 21 EGFP-labeled axons in nonsilenced background, 18 single silenced axons (TeNT-Lc:EGFP), 13 axons in TeNT-LC mRNA-injected zebrafish, and 12 TeNT-Lc:EGFP Tol2 axons. For statistical analysis of total branch length over time, a parametric two-way ANOVA with Bonferroni's post hoc test was performed. For arbor area and branch number, statistical analysis was performed using a parametric one-way ANOVA with Bonferroni's post hoc test (ns, not significant; p > 0.05). **p < 0.01; ***p < 0.001.

Figure 6.

Time-lapse imaging of single axonal arbors under conditions of global suppression of synaptic activity. A, Series of still images taken from 10 h time-lapse movies between 6 and 7 dpf, with imaging interval of 10 min. Axons grown under conditions of global suppression of synaptic activity show numerous transient branches (white arrowheads) but little net growth over the 10 h imaging period. B, Quantification of net arbor length change over the 10 h imaging period. Similar to control axons imaged in nonsilent backgrounds between 5 and 7 dpf (TeNT-Lc mut:EGFP at 5–7 dpf), axons grown under conditions of global suppression of synaptic activity (TeNT-Lc mRNA at 5–7dpf) show very little net growth. For comparison, data for singly silenced axons between 5 and 7 dpf (TeNT-Lc:EGFP at 5–7 dpf) are also shown. C, Cells grown under conditions of global silencing produce as many new transient branches as single silenced axons. D, Histogram showing that the distribution of new branch lifetimes is the same under conditions of single-cell and global suppression of synaptic activity. For comparison, values for axons expressing TeNT-LC mut:EGFP between 5 and 7 dpf are also shown. For statistical analysis of branch lifetime, a two-way ANOVA with Bonferroni's post hoc test was performed (***p < 0.0001). All other tests were performed using a nonparametric one-way ANOVA, followed by a Dunn's post hoc test (ns, not significant; p > 0.05). *p < 0.05. All graphs show mean ± SEM values of five axons imaged under conditions of global silencing (TeNT-Lc mRNA at 5–7 dpf), six single silenced axons (TeNT-Lc:EGFP at 5–7 dpf), and six control axons under nonsilenced conditions (TeNT-Lc mut:EGFP at 5–7 dpf).

Figure 7.

Morphology of single cells silenced by TeNT-Lc:EGFP is rescued by bath application of the NMDAR antagonist MK-801. A, B, Treatment with MK-801 did not significantly alter the growth of control (EGFP)-expressing axons (p > 0.05), but MK-801 treatment reversed the increase in total branch length induced by TeNT-Lc:EGFP expression to control levels (p > 0.05). C, Treatment with MK-801 also reverses the increase in arbor area induced by single-cell expression of TeNT-Lc:EGFP. Note that the area of control axons is unaffected by MK-801 treatment. D, Branch number of control and silenced axons is unaffected by MK-801 treatment. Graphs shows mean ± SEM values of 21 EGFP-labeled axons minus MK-801, 16 EGFP-labeled axons plus MK-801, 18 single silenced axons (TeNT-Lc:EGFP) minus MK-801, and 15 TeNT-Lc:EGFP-expressing axons plus MK-801. For statistical analysis, a parametric two-way ANOVA with Bonferroni's post hoc test was performed (ns, not significant; p > 0.05). **p < 0.01.

Figure 8.

Global suppression of synaptic activity disrupts convergence of RGC axons in the tectum. A, DiI labeling of the retinotectal projection of nasodorsal RGCs in wild-type (control, top row), TeNT-Lc:EGFP mRNA-injected (middle row), and TeNT-Lc:EGFP Tol2 (bottom row) zebrafish larvae at 6 dpf. Dorsal views of labeled RGCs in the tectum are shown in the right images, and the corresponding injection site in the retina is shown in the far left images. The area measured in the retina is indicated by the dashed line. Areas of fluorescence outside this area correspond to DiI labeling of the skin (arrowheads). To better illustrate variations in fluorescence labeling intensity, a thermal lookup table was applied to images of the RGC projection fields (middle images). Note the more dispersed projection field and absence of a central bright region of fluorescence in the silenced larvae. Scale bars, 50 μm. B, Scatter plot of tectal projection area as a function of retinal injection area. Wt, Wild type.