Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Aug 7;216(8):2499-2513.
doi: 10.1083/jcb.201704068. Epub 2017 Jun 19.

Activity-dependent Trafficking of Lysosomes in Dendrites and Dendritic Spines

Affiliations
Free PMC article

Activity-dependent Trafficking of Lysosomes in Dendrites and Dendritic Spines

Marisa S Goo et al. J Cell Biol. .
Free PMC article

Abstract

In neurons, lysosomes, which degrade membrane and cytoplasmic components, are thought to primarily reside in somatic and axonal compartments, but there is little understanding of their distribution and function in dendrites. Here, we used conventional and two-photon imaging and electron microscopy to show that lysosomes traffic bidirectionally in dendrites and are present in dendritic spines. We find that lysosome inhibition alters their mobility and also decreases dendritic spine number. Furthermore, perturbing microtubule and actin cytoskeletal dynamics has an inverse relationship on the distribution and motility of lysosomes in dendrites. We also find trafficking of lysosomes is correlated with synaptic α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-type glutamate receptors. Strikingly, lysosomes traffic to dendritic spines in an activity-dependent manner and can be recruited to individual spines in response to local activation. These data indicate the position of lysosomes is regulated by synaptic activity and thus plays an instructive role in the turnover of synaptic membrane proteins.

Figures

Figure 1.
Figure 1.
Lysosomes are reliably labeled by LAMP1-GFP and are found in distal dendrites and dendritic spines. (A) Endogenous LAMP1 distribution is similar to LAMP1-GFP, as both labeled structures were found in somatic and dendritic compartments. Representative images of dissociated hippocampal neurons (DIV16) either stained for endogenous LAMP1 (left) or transfected with LAMP1-GFP (right) and allowed to express for ≤24 h. (B) Representative image of LysoTracker red–labeled acidic vesicles found in dendritic compartments (DIV16). Inset shows entire cell image of neuron expressing GFP (Sindbis virus). (C) LAMP1-GFP marks acidic vesicles. Representative image of a dendrite from a dissociated hippocampal neuron transfected with LAMP1-GFP and costained with LysoTracker red, which marks acidic compartments. Neurons were imaged live at DIV16. (D) LAMP1-GFP does not colocalize with early endosomes. Representative image of a dendrite from dissociated hippocampal neuron (DIV16) transfected with mCherry and LAMP1-GFP and stained for the endogenous early endosomal protein marker EEA1. (E and F) LAMP1-GFP–positive vesicles are found in distal dendrites. A straightened secondary distal dendrite from a dissociated hippocampal neuron (DIV16) with LAMP1-GFP present throughout the dendrite. Depicted are a representative whole-cell image (E) and straightened distal secondary dendrite (F). (G) LAMP1-GFP is present at the base, neck, and head of dendritic spines. Representative images of dendritic spines in cultured hippocampal neurons (DIV16) transfected with mCherry and LAMP1-GFP. (H) Quantification the percentage of spines with a lysosome either in the head, neck, or base of the spine. Data represent means ± SEM. (I) Representative two-photon image of a CA1 pyramidal neuron from a hippocampal organotypic (DIV8) slice biolistically transfected with LAMP1-GFP and dsRed. LAMP1-GFP–labeled lysosomes are found in dendritic spines.
Figure 2.
Figure 2.
Active lysosomes are found in distal dendrites. (A) Lysosome membrane disruption by GPN abolishes LysoTracker staining. Representative dissociated hippocampal neurons expressing GFP (Sindbis virus) and costained with LysoTracker. Images were taken before adding either vehicle (DMSO) or 40 µM GPN. Images were taken 5 min after either vehicle or GPN. Regions of interest (boxes) are magnified below. (B) Representative neuron transfected with mCherry and GCaMP3-TRPML1. Neuron was imaged live for 100 s in the 488 and 568 channel. (C) Straightened segment from B showing GCaMP3-TRPML1 signal 2 s after adding either vehicle or 40 µM GPN. (D) Detection of calcium release from lysosome stores by GCaMP3-TRPML1 in representative neuron transfected with mCherry and GCaMP3-TRPML1 from B and C. Vehicle was added 50 s after the start of the recording, and 40 µM GPN was added 80 s after the start of the recording. Asterisks show time where images from D were taken.
Figure 3.
Figure 3.
APEX2 technology for EM shows ectopic expression of LAMP1 is specific to lysosomes. (A) Schematic for how APEX2 is used to label lysosomes by EM. APEX2 was cloned on the cytoplasmic side of LAMP1 (LAMP1-APEX2) which allowed for the labeling (with minimal spread of the DAB reaction) of the outside perimeter of intact lysosomes. (B and C) Representative transmission EM images of cultured hippocampal neurons (DIV16) transfected with LAMP1-APEX2. APEX2-stained lysosomes are present in the cell body and in dendrites of neurons. (D) LAMP1-APEX2–labeled lysosome found near dendritic spines. EM image of lysosomes near a base of a dendritic spine. Arrows point to LAMP1-APEX2–positive structures.
Figure 4.
Figure 4.
Lysosomal inhibition alters lysosome trafficking and decreases dendritic spine density. (A) Lysosomal inhibition decreases lysosome trafficking in hippocampal dendrites. Representative live images of secondary dendrites from cultured hippocampal neurons (DIV16) transfected with mCherry and LAMP1-GFP with the corresponding kymograph below. Cultures were treated with 200 µM leupeptin for 3 h. Images represent the first image in the time-lapse sequence. Live images were taken every second for 100 s. Vertical lines in kymographs represent stationary structures. (B) Quantification of LAMP1-GFP movement in dendrites from kymographs represented in A. Movement was manually counted in a blinded fashion. 306 (control) and 245 (leupeptin) vesicles from n = 23 and 19 dendrites for control and leupeptin, respectively. *, P < 0.05 between stationary groups; *, P < 0.05 between mobile, unpaired Student’s t test. Data represent mean ± SEM. Experimenter was blinded to condition upon analysis. (C–F) Lysosomal inhibition decreases mEPSC frequency and dendritic spine density. (C) Representative traces of mEPSCs recorded from control and 200 µM leupeptin (2–4 h)–treated cultured hippocampal neurons (DIV 18–24); n = 25 and 31 cells for control and leupeptin, respectively; mean mEPSC amplitude (D); cumulative probability distribution of interevent interval (IEI) of all mEPSCs record from control and leupeptin-treated neurons n = 3679 and 4552 events, respectively (E); inter-event interval (*, P < 0.05 unpaired Student’s t test). (F). Data represent mean ± SEM. Bars: 200 ms; (traces) 20 pA. Experimenter was blinded to condition upon analysis. (G) Representative straightened dendrites after control and 200 µM leupeptin (3 h)–treated cultured hippocampal neurons (DIV15 to DIV16) expressing GFP via Sindbis viral transduction (16 h). (H) Quantification of spine density from conditions displayed in G. ****, P < 0.0001 unpaired Student’s t test. Data represent mean ± SEM with ≥49 dendrites quantified per treatment. Experimenter was blinded to condition upon analysis.
Figure 5.
Figure 5.
Perturbations in microtubule and actin dynamics alter trafficking of lysosome. (A and B) Disruption of microtubule and actin dynamics inversely affects lysosome trafficking. (A) Representative images of cultured neurons under control conditions, after treatment with 10 µg/ml nocodazole (1 h) or 20 µM latrunculin A (10 min) with respective kymographs. Straightened dendrites represent the first image in the time-lapse sequence. Live images were taken every second for 100 s. (B) Quantification of LAMP1-GFP movement after nocodazole or latrunculin A treatment from Fig. 4 A. Approximately 435 vesicles from 36, 15, and 24 dendrites in control, nocodazole-, and latrunculin A–treated cells, respectively, were analyzed from four independent experiments. ****, P < 0.0001; **, P < 0.01, unpaired Student’s t test. Data represent mean ± SEM. Experimenter was blinded to condition upon analysis. (C) Representative images of dissociated hippocampal cell expressing LAMP1-GFP and Lifeact-RFP. Arrows point to LAMP1-GFP juxtaposed to Lifeact-RFP. (D–G) Disruption of microtubule dynamics with nocodazole increases lysosomes in dendritic spines. (D) Representative image of a secondary dendrite from a cultured hippocampal neuron (DIV16) transfected with mCherry and LAMP1-GFP under control or 10 µg/ml nocodazole treatment (1 h). Arrows point to LAMP1-GFP in a dendritic spine. (E–G) Quantification of the percentage of spines that have LAMP1-GFP in the head of a spine (E). Number of spines per micrometer (F) and signal intensity of LAMP1-GFP (G) showed no significant difference between treatments. 690 and 512 spines for nocodazole from >25 dendrites for control and nocodazole treated cells, respectively, were analyzed from two independent experiments. *, P < 0.05, unpaired Student’s t test. Data represent mean ± SEM. Experimenter was blinded to condition upon analysis.
Figure 6.
Figure 6.
Distribution and trafficking of lysosomes is highly correlated with internalized membrane proteins and synaptic AMPARs. (A) Model of bulk surface membrane internalization assay. (B) Representative images of dissociated hippocampal neurons expressing LAMP1-GFP (green) and internalized membrane proteins (red). Outlines of dendritic spines were generated by outlining the mCherry signal. (C) Representative immunofluorescent images of dissociated hippocampal neurons (16 DIV) expressing GFP and GFP-GluA1. Surface GFP-GluA1 was labeled with Alexa Fluor 647 and subsequently stained for Bassoon, a presynaptic marker. Surface GluA1 is juxtaposed to Bassoon. (D) Representative images of dissociated hippocampal cultures (DIV16) transfected with GFP, GFP-GluA1, and LAMP1-RFP. GFP-GluA1 can colocalize with a lysosome in and at the base of a dendritic spine. Cultures were treated with 100 µg/ml leupeptin and live-labeled with GFP antibody conjugated to Alexa Fluor 647. After washout, cultures were treated with 100 µM AMPA for 10 min to promote endocytosis. Outlines of dendritic spines were generated by outlining the GFP signal. Surface GFP-GluA1 labeled with anti-GFP Alexa Fluor 647 was false-colored green to show colocalization with LAMP1-RFP. (E) Live imaging of labeled surface GluA1 with LAMP1-RFP after 100 µM AMPA treatment. LAMP1-RFP labeled lysosomes persist at a location of surface-labeled AMPARs, rapidly move bidirectionally between two sites of surface-labeled AMPARs, and cotraffic with surface-labeled AMPARs likely destined for degradation.
Figure 7.
Figure 7.
Activity-dependent trafficking of lysosomes to dendritic spines. (A) Representative image of secondary dendrites from dissociated hippocampal neurons (DIV 16) transfected with mCherry and LAMP1-GFP under control conditions or after 200 µM AMPA treatment (2 min). Arrows point to LAMP1-GFP in a dendritic spine. (B–D) Quantification of the percentage of spines that have LAMP1-GFP-labeled lysosomes in the head of a spine (B). Number of spines per micrometer (C) and signal intensity of LAMP1-GFP (D) showed no significant difference between treatments. 830 (control) and 833 (AMPA) dendritic spines were analyzed (n = 41 for control and AMPA) over three independent experiments, *, P < 0.05, unpaired Student’s t test. Experimenter was blinded to condition upon analysis. (E) Representative image of secondary dendrites (DIV16) transfected with mCherry and LAMP1-GFP under control conditions or after 200 µM glycine (10 min), 200 µM glycine (10 min) with 50 µM AP5 (60-min treatment pretreatment), or 50 µM AP5 alone (60 min) in HBS containing 0 mM Mg+2. Arrows point to LAMP1-GFP in a dendritic spine. (F–H) Quantification of the percentage of spines that have LAMP1-GFP-labeled lysosomes in the head of a spine (F). Number of spines per micrometer (G) Signal intensity of LAMP1-GFP (H) showed no significant difference between treatments. 211 dendritic spines for control, 246 for glycine, 356 for glycine/AP5, and 573 for AP5 were analyzed (n = 14 for control, 14 for glycine, 18 for glycine/AP5, and 28 for AP5) over three independent experiments. ***, P < 0.001; **, P < 0.01; *, P < 0.05, one-way analysis of variance with Tukey’s post hoc analysis. Experimenter was blinded to condition upon analysis. All data represent mean ± SEM.
Figure 8.
Figure 8.
Activation of a single spine recruits a lysosome to the base of the spine. (A) Experimental timeline of two-photon imaging and MNI-glutamate uncaging in hippocampal organotypic cultures. Individual spines on secondary dendrites of CA1 pyramidal neurons were visualized with a laser tuned to 900 nm. A 720-nm laser was used to stimulate individual spines, using 0.5-ms pulses at 1 Hz for 1 min. Stimulation was done with MNI-glutamate (2.5 mM MNI-glutamate) or without (“mock uncaging”) as a control in ACSF containing 0 mM Mg2+. Arrows signal the time points at which the representative images from B and C were taken. (B and C) Representative images of a secondary dendrite in a CA1 pyramidal neuron in a rat organotypic hippocampal slice. Time points are 5 and 2 min before uncaging and 2 min and 5 min after uncaging. (D) Mean dwell time for a lysosome at the base of a spine in either control or MNI-glutamate conditions. *, P ≤ 0.05, Mann–Whitney U test. Data represent mean ± SEM. (E) Cumulative probability distribution for each condition with and without (mock uncaging) MNI-glutamate. n = 12 for control, 10 for uncaging. P ≤ 0.01, Kolmogorov–Smirnov test. Experimenter was blinded to condition upon analysis.

Similar articles

See all similar articles

Cited by 29 articles

See all "Cited by" articles

References

    1. Alvarez-Castelao B., and Schuman E.M. 2015. The regulation of synaptic protein turnover. J. Biol. Chem. 290:28623–28630. 10.1074/jbc.R115.657130 - DOI - PMC - PubMed
    1. Arancibia-Cárcamo I.L., Yuen E.Y., Muir J., Lumb M.J., Michels G., Saliba R.S., Smart T.G., Yan Z., Kittler J.T., and Moss S.J. 2009. Ubiquitin-dependent lysosomal targeting of GABA(A) receptors regulates neuronal inhibition. Proc. Natl. Acad. Sci. USA. 106:17552–17557. 10.1073/pnas.0905502106 - DOI - PMC - PubMed
    1. Bananis E., Nath S., Gordon K., Satir P., Stockert R.J., Murray J.W., and Wolkoff A.W. 2004. Microtubule-dependent movement of late endocytic vesicles in vitro: Requirements for dynein and kinesin. Mol. Biol. Cell. 15:3688–3697. 10.1091/mbc.E04-04-0278 - DOI - PMC - PubMed
    1. Bingol B., and Schuman E.M. 2006. Activity-dependent dynamics and sequestration of proteasomes in dendritic spines. Nature. 441:1144–1148. 10.1038/nature04769 - DOI - PubMed
    1. Bingol B., Wang C.F., Arnott D., Cheng D., Peng J., and Sheng M. 2010. Autophosphorylated CaMKIIalpha acts as a scaffold to recruit proteasomes to dendritic spines. Cell. 140:567–578. 10.1016/j.cell.2010.01.024 - DOI - PubMed

Publication types

MeSH terms

Substances

Feedback