Inactivation of clathrin heavy chain inhibits synaptic recycling but allows bulk membrane uptake

Abstract

Synaptic vesicle reformation depends on clathrin, an abundant protein that polymerizes around newly forming vesicles. However, how clathrin is involved in synaptic recycling in vivo remains unresolved. We test clathrin function during synaptic endocytosis using clathrin heavy chain (chc) mutants combined with chc photoinactivation to circumvent early embryonic lethality associated with chc mutations in multicellular organisms. Acute inactivation of chc at stimulated synapses leads to substantial membrane internalization visualized by live dye uptake and electron microscopy. However, chc-inactivated membrane cannot recycle and participate in vesicle release, resulting in a dramatic defect in neurotransmission maintenance during intense synaptic activity. Furthermore, inactivation of chc in the context of other endocytic mutations results in membrane uptake. Our data not only indicate that chc is critical for synaptic vesicle recycling but they also show that in the absence of the protein, bulk retrieval mediates massive synaptic membrane internalization.

Figure 1.

Synaptic membrane internalization in chc hypomorphic mutants. (A–C) FM 1-43 labeling of endocytosed membrane at the NMJ of D. melanogaster third instar larvae. The motor neurons of w control (A), w chc⁴ (B), and w chc¹/w chc⁴ (C) female larvae were stimulated for 10 min with 90 mM KCl. Labeling is visible in NMJ boutons of both mutants and controls. However, in clathrin mutants, aberrant membranous structures not seen in controls are observed (arrows). Bar, 2.5 μm. (D and E) Quantification of the FM 1-43 labeling intensity (D) and membrane inclusion surface area (E), both normalized to total bouton surface area. The total amount of membrane internalized in chc mutants and controls is not significantly different (P = 0.55, ANOVA). However, membrane inclusions, which are not observed in controls, occupy 4–8% of the bouton surface area in chc mutants (P = 0.043, ANOVA). The number of animals tested is indicated in the bar graphs and error bars indicate SEM. t test: **, P < 0.01.

Figure 2.

Photoinactivation of chc protein reveals uncontrolled membrane uptake upon stimulation. (A) Creation of a genetic rescue construct encoding chc fused with an N-terminal FLAG-tetracysteine tag (4C), 4C-chc⁺. (a) Gap-repaired genomic rescue fragment including the chc gene. (b) PCR product containing left (L) and right (R) homology arms, the 4C tag, and LoxP site (gray triangles) flanked Kan marker. (c) A correct recombination event followed by Cre-mediated removal of the Kan marker results in a 4C-tagged chc, leaving an in-frame LoxP site as linker. (B and C) Expression of 4C-chc⁺. w control and w chc¹; 4C-chc⁺ third instar larval dissections were treated with the membrane-permeable dye FlAsH, and unbound FlAsH was washed away. Labeling in boutons and muscle was only detected in animals containing the 4C tag. Bar, 5 μm. (D–G) EJPs recorded from muscle 6 in HL3 with 0.25 mM calcium (D and E) and in 2 mM calcium (F and G) in both w control and w chc¹; 4C-chc⁺ larvae. EJPs from both genotypes are not different. (H–O) Photoinactivation of chc protein by FlAsH-FALI results in aberrant membrane internalizations. w control (H–K) and w chc¹; 4C-chc⁺ (L–O) animals were treated (+) or not treated (−) with FlAsH and/or illuminated with 500 nm epifluorescent light for 10 min (+) or not (−), as indicated at the bottom. After treatment, motor neurons were stimulated with 90 mM KCl for 10 min in the presence of FM 1-43. Excess dye was washed away and boutons were imaged. Note abnormal membranous structures in w chc¹; 4C-chc⁺ animals where chc was inactivated using FlAsH-FALI only. Bar, 2.5 μm.

Figure 3.

Both chemical inhibition and photoinactivation of chc protein show aberrant membrane inclusions that are quantitatively similar. (A–C) FM 1-43 dye uptake (10 min of 90 mM KCl) in synaptic boutons after chlorpromazine treatment on w control (A) and w chc¹; 4C-chc⁺ (B) animals, as well as on chlorpromazine-treated w chc¹; 4C-chc⁺ where chc was also inactivated using FlAsH-FALI (C). Aberrant FM 1-43–labeled membrane inclusions are clearly visible in all conditions. Bar, 2.5 μm. (D and E) Quantification of membrane inclusion surface normalized to total bouton surface (D) and relative FM 1-43 labeling intensity compared with w controls (E) in different conditions (FlAsH-treated and illuminated, chlorpromazine-treated, or both). Membrane inclusion phenotypes of double-treated animals are not significantly different than phenotypes in animals where chc was inactivated with either chlorpromazine or with FlAsH-FALI (P = 0.22, ANOVA). Furthermore, labeling intensity between the different conditions is not statistically different and is also not different from w controls that were not FlAsH treated or illuminated (w treated with FlAsH and illuminated, 100 ± 14%; w not FlAsH treated or illuminated, 105 ± 11%; Fig. 2 H; P = 0.9, ANOVA). The number of animals tested is indicated in the bars and error bars indicate SEM. ANOVA: **, P < 0.0001.

Figure 4.

Photoinactivation of chc causes massive membrane invaginations and a dramatic reduction in vesicle density. (A–D) Electron micrographs of synaptic bouton cross sections (muscles 6 and 7). (A) w chc¹; 4C-chc⁺ control bouton stimulated with 90 mM KCl for 10 min but not incubated in FlAsH. (B–D) Images from stimulated w chc¹; 4C-chc⁺ boutons where chc was inactivated using FlAsH-FALI. Note massive membrane invaginations, cisternae, and larger vesicles in boutons lacking functional chc (arrows) not observed in controls. Dense bodies (arrowheads and inset) in synapses where chc was inactivated consistently show clustered vesicles. m, mitochondria. A and C, conventional EM; B, inset, and D, high voltage EM. Bars: (A–D) 0.6 μm; (inset) 0.1 μm. (E) Histograms presenting the vesicle diameter in control (top) and synapses with photoinactivated chc (bottom). Cisternae and larger vesicles are readily observed when chc is inactivated. (F) Vesicle density in boutons of controls and with photoinactivated chc. Round or oval-shaped vesicles were included for quantification. Error bars indicate SEM. n, at least seven boutons per condition from different animals. t test: **, P < 0.001.

Figure 5.

Acute loss of chc function by photoinactivation inhibits synaptic vesicle recycling but not neurotransmitter release. (A–D) FM 1-43 loading (10 min of 90 mM KCl; A and C) and unloading (10 min of 90 mM KCl; B and D) at the third instar NMJ in w control (A and B) and w chc¹; 4C-chc⁺, where chc was inactivated using FlAsH-FALI (C and D). While unloading of FM 1-43–labeled vesicles using KCl stimulation in controls is efficient, labeled membrane in synapses where chc was photoinactivated is largely retained and cannot be released upon stimulation. Bar, 2.5 μm. (E) Quantification of FM 1-43 labeling intensity after loading and unloading of FM 1-43 (A–D). The number of animals tested is indicated in the bars and error bars indicate SEM. ANOVA: P = 0.67; **, P < 0.0001. (F and G) Sample EJPs recorded in 0.5 mM of extracellular calcium in w chc¹; 4C-chc⁺ animals incubated in FlAsH, without illumination (F) and with illumination to inactivate chc (G). (H) Quantification of EJP amplitudes recorded in 0.25, 0.5, 1, and 2 mM of extracellular calcium in w chc¹; 4C-chc⁺ animals incubated in FlAsH, without illumination and with illumination to photoinactivate chc. No difference in EJP amplitude before and after illumination was observed for each of the tested calcium concentrations. The number of animals tested is indicated in the bars and error bars indicate SEM (t test). (I) Relative EJP amplitude measured during 10 min of 10-Hz nerve stimulation in w control and w chc¹; 4C-chc⁺ animals incubated in FlAsH reagent. Control data is pooled from w, w incubated in FlAsH, and not treated w chc¹; 4C-chc⁺ animals (at least three animals each). All genotypes and conditions were first stimulated for 2 min while recording EJPs before illumination. EJP amplitudes were binned per 30 s and normalized to the mean amplitude of the first 10 EJPs. Note a reduction in relative EJP amplitude in w chc¹; 4C-chc⁺, where chc is acutely inactivated by FlAsH-FALI. Error bars indicate SEM.

Figure 6.

Inactivation of chc in other endocytic mutants linked to dynamin function. (A–D) KCl-induced FM 1-43 labeling of synaptic boutons of w control (A) and several endocytic mutants (synj¹ [B], dap160^Δ1/dap160^Δ2 [C], and shi^ts1 [D]), not treated (left) and treated (right) with chlorpromazine to inactivate chc. Animals were stimulated with 90 mM KCl in FM 1-43 for 10 min. Experiments with shi^ts1 and CS controls (not depicted) were done at 32°C. Bar, 2.5 μm. (E and F) FlAsH-FALI of chc in +; 4C-chc⁺ dominantly inactivates chc. FM 1-43 labeling of +/Y; 4C-chc⁺/+ (obtained from crossing 4C-chc⁺ males to CS virgins) using 3 min of 90-mM KCl stimulation either without (E) or with (F) FlAsH-FALI of chc. Note the membrane inclusions in animals where chc was inactivated using FlAsH-FALI. (G–I) Acute double mutant shi and chc NMJs show aberrant membrane internalizations. (G and H) FM 1-43 dye uptake (3 min of 90 mM KCl) in synaptic boutons of control shi^ts1/Y; 4C-chc⁺/+ animals, labeled at 20°C (G) or at 32°C (H) without FlAsH-FALI of chc. Note that the shi mutation at the restrictive temperature blocks membrane internalization. However, FM 1-43 labeling of shi^ts1/Y; 4C-chc⁺/+ animals where dynamin is inactivated at high temperature (32°C) and chc is photoinactivated using FlAsH-FALI reveals clear membrane internalization (I). Bar, 2.5 μm.