QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 21 October 2008
doi: 10.1242/jcs.033803

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JCS Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bonanomi, D. Articles by Valtorta, F. Search for Related Content PubMed PubMed Citation Articles by Bonanomi, D. Articles by Valtorta, F. Social Bookmarking                         What's this?

Identification of a developmentally regulated pathway of membrane retrieval in neuronal growth cones

Dario Bonanomi^1,*,, Eugenio F. Fornasiero^1,*, Gregorio Valdez^2,, Simon Halegoua², Fabio Benfenati^3,4,5, Andrea Menegon¹ and Flavia Valtorta^1,5,¶

¹ S. Raffaele Scientific Institute/Vita-Salute University and IIT Unit of Molecular Neuroscience, 20132 Milano, Italy
² Department of Neurobiology and Behavior, State University of New York at Stony Brook, Stony Brook, NY 11794, USA
³ Department of Experimental Medicine, University of Genova, 16132 Genova, Italy
⁴ Department of Neuroscience and Brain Technologies, IIT Central Laboratories, 16163 Genova, Italy
⁵ Istituto Nazionale di Neuroscienze, 10125 Torino, Italy

View larger version (65K): [in this window] [in a new window]   Fig. 1. Constitutive bulk membrane retrieval in axonal growth cones. (A) Basal uptake of FM4-64 (red in a' and b') in neurons at 2 DIV that were incubated with KRH containing the dye and imaged after a 2-minute washing. (a,a') A broad region of intense FM4-64 staining is associated with the growth cone (arrowhead), whereas few dispersed punctae appear in the cell body (asterisk). Note the lack of FM4-64 internalization along the axon shaft. (b,b') Higher-magnification views of a growth cone, showing internalization of FM4-64 in large organelles at the interface between the central and peripheral domains (T domain). Note the vacuolar organization of the FM4-64-positive compartment in b. (B) Basal uptake of FM4-64 (red) in neurons incubated with the dye for 1 minute, fixed and retrospectively stained for the axon-specific marker dephospho-Tau-1 (green). Intense FM4-64 uptake is visible in growth cones of Tau-1-positive axons (asterisks) but not in Tau-1-negative perspective dendrites (arrowheads). (C) Differential uptake of FM4-64 (red) in growth cones that are associated with distinct axonal branches of the same neuron. One of the growth cones (arrowhead) shows strong constitutive endocytosis, whereas the other (asterisk) is inactive. Scale bar: 13.5 µm (Aa,a'); 5 µm (Ab,b'); 25 µm (B); 10 µm (C).

View larger version (34K): [in this window] [in a new window]   Fig. 2. Bulk endocytosis correlates with growth-cone motility. (A) Time-lapse phase-contrast imaging of two representative growth cones (3 DIV) over a 5-minute period prior to exposure to FM4-64 for 1 minute and fixation. The dye is internalized only in the more motile growth cone (bottom panels). (B) Quantification of the area of growth cones that are either positive or negative for bulk FM4-64 endocytosis (mean ± s.e.m.; n>70 growth cones per condition). (C) Quantification of growth-cone motility in growth cones that are either positive or negative for bulk FM4-64 endocytosis. The motility index (µm2/minute) was calculated by measuring the variation in growth-cone area over the course of 6 minutes as described in the Materials and Methods (mean ± s.e.m.; n=25 growth cones from three independent preparations; *P<0.01, Student's t test, endocytosis-positive vs endocytosis-negative growth cones). Bulk endocytosis is preferentially associated with more-dynamic growth cones. Scale bar: 5 µm.

View larger version (50K): [in this window] [in a new window]   Fig. 3. Bulk endocytosis in the growth cone is linked to membrane ruffling and depends on actin dynamics, PI3-kinase activity and cholesterol levels. (A) Growth cones (2 DIV) that were either left untreated (KRH) or subjected to various treatments before incubation for 1 minute with KRH in the presence of FM4-64 are shown. Incubation of growth cones at 4°C during FM4-64 application prevents dye uptake (KRH 4°C). Incubation with either 2 mM EGTA-containing solution devoid of Ca2+ for 30 minutes (EGTA) or BFA (10 µg/ml) for 1 hour does not affect FM4-64 internalization. Treatments with CytD (10 µM, 15 minutes), LY294002 (50 µm, 30 minutes) or MβCD (5 mM, 3 minutes) inhibit FM4-64 uptake. (B) Time-lapse DIC imaging of a growth cone (2 DIV) incubated for 130 seconds in KRH prior to exposure for 1 minute to KRH containing FM4-64. The last panel on the right represents the overlay of the DIC image taken at 190 seconds with the FM4-64 staining (red). The dye is internalized in the region of the growth cone that is associated with active plasma-membrane ruffling (arrowhead). (C) Quantification of FM4-64 uptake in growth cones that were treated as shown in A. The amounts of FM4-64 loaded in 2-DIV growth cones during a 1-minute incubation in high K+ (KCl) or remaining after incubation in KRH for 30 minutes at 37°C (unloading) are also reported (mean ± s.e.m.; n=30-40 growth cones per treatment; *P<0.001, Dunnett's test for untreated growth cones vs growth cones subjected to various treatments). (D,E) Growth cones loaded with FM4-64 (KRH, 1 minute; red) and retrospectively stained with either FITC-phalloidin to visualize F-actin (D; green) or filipin to visualize cholesterol (E; blue). The overlay between the corresponding fluorescence and DIC images is shown in the right panels. The arrowhead in D points to a ruffling focus colocalizing with FM4-64 labeling. Scale bar: 10 µm.

View larger version (58K): [in this window] [in a new window]   Fig. 4. The high-capacity endocytic process is clathrin-independent and distinct from the SV and endosomal recycling pathways. (A) Basal uptake of FM4-64 (red in the merged image) in a growth cone incubated with the dye for 1 minute, fixed and retrospectively labeled for VAMP2 (blue in the merged image) and syntaxin 13 (green in the merged image). The overlay between the fluorescence and DIC images is shown in the merged image. The bulk of FM4-64 fluorescence colocalizes with neither VAMP2 nor syntaxin 13, although occasional punctae of overlap are observed in more proximal regions (arrowheads). (B) Constitutive endocytosis in growth cones probed by exposure to various tracers (green) for 1 minute in KRH: (a) Lucifer yellow (4 mg/ml), (b) 200-nm beads, (c) 20-nm beads. FM4-64 applied during a subsequent incubation is internalized in bead-containing compartments (right panel in c; red). The overlay between fluorescence and DIC images is shown. (C) Depletion of the clathrin heavy chain (CHC) in mouse hippocampal neurons by RNAi. (Left) Representative images of neurons that were co-transfected with YFP and either control Stealth (upper panels) or CHC Stealth (bottom panels). After basal FM4-64 uptake (red), 3-DIV mouse neurons were fixed and retrospectively immunostained to assess clathrin downregulation (green). (Right) Quantification of the effect of CHC RNAi on bulk plasma-membrane endocytosis. A fivefold reduction of CHC immunoreactivity (green bars; mean ± s.e.m.; n=20 growth cones per treatment) in individual growth cones is observed following RNAi. FM4-64 bulk endocytosis (red bars) is unchanged (*P<0.001, Student's t-test, treated vs control). (D) A neuron exposed to Alexa-Fluor-488-conjugated transferrin (green in the merged image) for 1 hour and subsequently to FM4-64 (red in the merged image) for 1 minute in KRH is shown. Transferrin, which is primarily internalized in the cell body (asterisk), does not overlap with FM4-64, which is endocytosed in the growth cone (arrowhead). (E) Growth cones that were loaded with FM4-64 (KRH, 1 minute; red) and retrospectively stained with anti-Rabankyryn-5 (green in the merged image). Scale bar: 10 µm (A,B); 15 µm (C); 29.5 µm (D); 3.8 µm (E).

View larger version (52K): [in this window] [in a new window]   Fig. 5. Exclusion of membrane proteins from bulk endocytosis. Growth cones (2 DIV) incubated in KRH in the presence of FM4-64 (red) for 1 minute, fixed and retrospectively stained with antibodies against integral membrane proteins that are known to undergo recycling in axons (Trk, L1, β1 integrin, APP; green). The overlay between the corresponding fluorescence and DIC images is shown in the right panels. Scale bar: 10 µm.

View larger version (50K): [in this window] [in a new window]   Fig. 6. The pathway of bulk membrane retrieval is distinct from the pathway of cholera-toxin internalization. A growth cone (2 DIV) that was exposed simultaneously to FM4-64 (red in the merged images) and Alexa-Fluor-488-conjugated cholera toxin subunit B (CtxB; green in the merged images) for 1 minute in KRH, washed and imaged during the following 30 minutes at 37°C. Immediately after loading (t=0), FM4-64 is internalized in large compartments, whereas CtxB is still associated with the plasma membrane and is subsequently endocytosed in punctae that do not overlap with FM4-64 labeling. Scale bar: 10 µm.

View larger version (59K): [in this window] [in a new window]   Fig. 7. Rac1 activity is required for bulk plasma-membrane endocytosis in the growth cones. (A) Growth cones of 3-DIV neurons coexpressing YFP (green in the merged images) and either the dominant-negative mutant Rac1-N17 (top panels) or the constitutively active mutant Rac1-V12 (middle panels), or expressing only YFP (bottom panels), incubated in KRH containing FM4-64 (red in the merged images) for 1 minute and fixed. The Rac1 mutants are detected with an anti-FLAG antibody (blue in the merged images). Expression of Rac1-N17 inhibits bulk FM4-64 internalization whereas endocytosis via smaller vesicles is unaffected (arrowheads). (B) Quantification of FM4-64 uptake in Rac1-N17- or Rac1-V12-expressing growth cones treated as shown in A (mean ± s.e.m.; n=27 growth cones per condition; *P<0.001, Dunnett's test for neurons expressing either Rac1-N17 or Rac1-V12 vs control, i.e. growth cones expressing YFP only). The total growth-cone area was reduced by Rac1-N17 expression (mean ± s.e.m.: control, 44±2.5 µm2; Rac1-N17, 35±2.6 µm2; P<0.05, Dunnett's test vs control), but not by Rac1-V12 expression (mean ± s.e.m.: 42±2.7 µm2). Scale bar: 10 µm.

View larger version (50K): [in this window] [in a new window]   Fig. 8. The marker of neuronal macroendocytosis, Pincher, regulates constitutive plasma-membrane retrieval at the growth cone. Growth cones of neurons at 3 DIV that were infected with adenoviruses driving the simultaneous expression of GFP and either HA-Pincher or HA–Pincher-G68E are shown. (A) Growth cones expressing low levels of HA-Pincher or HA–Pincher-G68E that were stained with anti-HA antibody (green) after basal FM4-64 (red) uptake (1 minute) to reveal the distribution of the exogenous proteins. The overlay between the fluorescence and DIC images is shown in the right panels. (B) Growth cone expressing HA-Pincher at low levels exposed to 20-nm beads (red in the merged image) for 1 minute in KRH, fixed and stained with anti-HA antibody (green in the merged image). The images are maximal projections of deconvoluted z-stacks. Pincher overlaps with the compartments of bulk endocytosis. (C,D) Basal FM4-64 uptake (1 minute) in growth cones overexpressing GFP and either HA-Pincher or HA–Pincher-G68E, or infected with adenoviruses expressing only GFP (control). (C) Shows the overlay between the fluorescence and DIC images of a HA-Pincher/GFP-expressing growth cone (arrowhead) and of an uninfected growth cone (asterisk) located in the same field of view. Overexpression of either Pincher or the dominant-negative mutant Pincher G68E inhibits constitutive bulk plasma-membrane uptake, whereas endocytosis via smaller vesicles is unaffected (arrowheads in D). (E) Quantification of FM4-64 uptake in HA-Pincher- or HA–Pincher-G68E-expressing growth cones (mean ± s.e.m.; n=50 growth cones per condition; *P<0.001, Dunnett's test vs control, i.e. growth cones expressing only GFP). Expression of either mutant does not affect the total growth-cone area (P>0.05, Dunnett's test vs control). (F) Quantification of axon length in 2-DIV neurons that were nucleofected with GFP alone (control) or together with either HA-Pincher or HA–Pincher-G68E. The length of the longest neurite, corresponding to the axonal process, was measured for each transfected neuron (mean ± s.e.m.; n=100-150 growth cones per condition from three independent experiments; **P<0.001, Tukey's test Pincher G68E vs control; *P<0.05, Pincher wild-type vs control; P<0.05, Pincher wild-type vs Pincher G68E). Scale bar: 10 µm (A,C,D); 3.3 µm (B).

View larger version (46K): [in this window] [in a new window]   Fig. 9. Plasma-membrane endocytosis in growth cones at early developmental stages is insensitive to depolarizing stimuli. (A) Uptake of FM4-64 (red in the merged image, arrowhead) in a growth cone of a 2-DIV neuron incubated with the dye for 3 minutes in KRH (left), followed by incubation with high K+ for 1 minute (middle). After depolarization, the growth cone was fixed and retrospectively stained for VAMP2 (green). In the right panel the overlay between the fluorescence and DIC images is shown. (B) Growth cone of a 2-DIV neuron expressing ECFP-VAMP2 (green in the merged image) incubated with high K+ for 1 minute in the presence of FM4-64 (red in the merged image). VAMP2-positive SVs do not internalize FM4-64 upon depolarization. Scale bar: 10 µm (A); 5.5 µm (B).

View larger version (59K): [in this window] [in a new window]   Fig. 10. Developmental control of endocytic activities in the growth cone. (A) Growth cone (DIC, left) of a neuron at 5 DIV expressing ECFP-VAMP2 (green) that was subjected to two 15-minute-spaced applications of FM4-64 (red) for 1 minute in either KRH (first application, middle) or high K+ (second application, right). The depolarizing step induces FM4-64 loading in VAMP2-positive SVs (arrowheads) with a pattern that is markedly different from that of constitutive endocytic compartments. (B) Neurons at 7 DIV that were incubated with FM4-64 for 1 minute in KRH. At this stage, constitutive FM4-64 uptake is not associated with growth cones (asterisk), whereas it is present in punctate endocytic compartments along the neurite network. (C) Growth cone of a 7-DIV neuron expressing ECFP-VAMP2 (green) incubated with high K+ for 1 minute in the presence of FM4-64 (red). Arrowheads point to VAMP2-positive SVs that internalized the dye upon depolarization. (D) Quantification of basal FM4-64 uptake in growth cones of neurons at 3, 5 and 7 DIV incubated with FM4-64/KRH for 1 minute, rapidly washed and then exposed to high K+ for 1 minute. Bulk endocytosis is suppressed at later developmental stages (mean ± s.e.m.; n=30-40 growth cones per treatment). At all time points, FM4-64 uptake was significantly different (P<0.001, Tukey's test). Scale bar: 7 µm (A,C); 12.5 µm (B, left); 7.5 µm (B, right).

View larger version (30K): [in this window] [in a new window]   Fig. 11. Developmental changes in the endocytic properties of the growth cone. Top: at early stages of differentiation (3 DIV), constitutive high-capacity plasma-membrane recycling occurs via a bulk endocytic pathway mediated by large compartments (red) in the T domain of the growth cone. SV precursors (white circles), which, at these stages, are not prone to either constitutive or evoked recycling, do not participate in the bulk retrieval process. Middle: as neuronal differentiation proceeds (4-5 DIV), the intensity of the bulk recycling process decreases (orange) concomitantly with the emergence of activity-dependent SV recycling (green circles) in both the growth cone and nascent synaptic structures. Bottom: bulk plasma-membrane retrieval ceases at later stages of neuronal differentiation (7 DIV), when frequent interactions of the growth cone with targets result in the establishment of synaptic contacts associated with robust activity-dependent SV recycling. Exo-endocytosis of SVs is maintained also in the growth cone. The dissociation of bulk endocytosis from the SV recycling pathway at the onset of synaptogenesis might preserve the molecular identity of the forming pool of SVs and the stability of nascent synaptic structures.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter