Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 216 (12), 4351-4365

New Tools for "Hot-Wiring" Clathrin-Mediated Endocytosis With Temporal and Spatial Precision

Affiliations

New Tools for "Hot-Wiring" Clathrin-Mediated Endocytosis With Temporal and Spatial Precision

Laura A Wood et al. J Cell Biol.

Abstract

Clathrin-mediated endocytosis (CME) is the major route of receptor internalization at the plasma membrane. Analysis of constitutive CME is difficult because the initiation of endocytic events is unpredictable. When and where a clathrin-coated pit will form and what cargo it will contain are difficult to foresee. Here we describe a series of genetically encoded reporters that allow the initiation of CME on demand. A clathrin-binding protein fragment ("hook") is inducibly attached to an "anchor" protein at the plasma membrane, which triggers the formation of new clathrin-coated vesicles. Our design incorporates temporal and spatial control by the use of chemical and optogenetic methods for inducing hook-anchor attachment. Moreover, the cargo is defined. Because several steps in vesicle creation are bypassed, we term it "hot-wiring." We use hot-wired endocytosis to describe the functional interactions between clathrin and AP2. Two distinct sites on the β2 subunit, one on the hinge and the other on the appendage, are necessary and sufficient for functional clathrin engagement.

Figures

Figure 1.
Figure 1.
Chemically inducible endocytosis. (A) Illustration of chemically inducible endocytosis. Normally, clathrin recognizes the appendage and hinge of the β2 subunit of the AP2 complex after it has engaged cargo and membrane. In chemically induced endocytosis, cells coexpress a plasma membrane anchor (CD8–mCherry–FRB) and a clathrin hook (FKBP–β2–GFP). Rapamycin (200 nM) is added, which causes heterodimerization of FKBP and FRB domains. The clathrin hook is rerouted to the plasma membrane. Clathrin recognizes the clathrin hook, and the plasma membrane anchor is internalized. (B) Chemically inducible endocytosis in live cells. Cells expressing CD8–mCherry–FRB with either FKBP–β2–GFP or GFP–FKBP. Stills from live-cell confocal imaging experiments are shown: the frame before rerouting occurs (top) and 133 frames (665 s) later (bottom). See Videos 1 and 2. Insets show a 2× zoom of the boxed region; intense orange bar indicates rapamycin application. Bars: (main) 10 µm; (insets) 2 µm. (C) Example plots to show chemically inducible endocytosis. Colored traces indicate the total area of bright GFP puncta that form in cells after rerouting. Gray traces show the rerouting of the clathrin hook from the cytoplasm to the plasma membrane (see Materials and methods).
Figure 2.
Figure 2.
Chemically induced internalization is via CCVs. (A) Bright green spots formed by chemically induced endocytosis colocalize with clathrin and contain anti-CD8–Alexa 647. Cells coexpressing mCherry–LCa, CD8–dCherry–FRB, and either FKBP–β2–GFP or GFP–FKBP were incubated with anti-CD8–Alexa 647 to label the membrane anchor extracellularly, and rapamycin (200 nM) was applied. Insets show a 2× zoom of the boxed region; intense orange bar indicates rapamycin application. Bars: (main) 10 µm; (insets) 2 µm. (B) Live-cell confocal immunolabeling experiments of cells expressing CD8–dCherry–FRB (dark mCherry variant) with FKBP–β2–GFP. Cells were also transfected with siRNAs against GL2 (control) or CHC. Inhibition of uptake of transferrin–Alexa 647 (A647-Tf; blue) was used as a functional test of knockdown efficacy. Bar, 10 µm. (C–G) CLEM to study the uptake of immunolabeled CD8–dCherry–FRB in cells coexpressing either FKBP–β2–GFP or GFP–FKBP. Results from a single experiment are shown, although CLEM experiments were performed three times. (C) Stills from live-cell wide-field imaging showing cells sequentially labeled with anti-CD8 and Alexa 546 FluoroNanoGold-conjugated secondary antibody (FNG; red), before and after the addition of rapamycin (200 nM). The same cell was processed for electron microscopy and imaged. Bars: (main) 10 µm; (zoom) 2 µm. (D) Electron micrographs of the cells shown in A. Clear uptake of NanoGold into CCVs was seen in cells coexpressing FKBP–β2–GFP but not GFP–FKBP. Bars, 100 nm. (E) Segmentation of CCV profiles in rapamycin-treated FKBP–β2–GFP versus GFP–FKBP samples. Membrane (gray) and coat (purple) for multiple vesicles are shown overlaid. Note that CCVs in FKBP–β2–GFP contained NanoGold, whereas GFP–FKBP CCVs did not. (F) Scatterplot of CCV diameters in FKBP–β2–GFP versus GFP–FKBP samples. Diameter refers to the mean of the major and minor axes. (G) Scatterplot of ellipticity of CCV profiles in FKBP–β2–GFP versus GFP–FKBP samples. Ellipticity is the ratio of semimajor and semiminor axis. The p-values from Student’s t test with Welch’s correction are shown. Dots indicate CCVs and bars indicate mean ± SD.
Figure 3.
Figure 3.
CCVs generated by hot-wired endocytosis are trafficked and recycled similarly to normal endocytic cargo. (A–C) Stills from typical live-cell confocal imaging experiments. Vesicles in HeLa cells expressing CD8–dCherry–FRB, FKBP–β2–mRuby2 (top), or CD8–YAAL (bottom) together with the indicated GFP-tagged construct (A, Rab5; B, EEA1; C, Rab4). CD8 was labeled using extracellularly applied anti-CD8/Alexa 647. Endocytosis was triggered by rapamycin addition (200 nM) in the case of hot-wiring, and cells were imaged at 0.2 Hz. Bars, 1 µm. (D) Summary of imaging experiments to delineate the paths taken by hot-wired cargo or by CD8–YAAL. See also Fig. S4.
Figure 4.
Figure 4.
Hot-wired endocytosis can function independently of AP2. (A) Chemically induced endocytosis can operate independently of endogenous AP2. Live-cell confocal immunolabeling experiments of cells expressing CD8–dCherry–FRB (dark mCherry variant) with FKBP–β2–GFP. Cells were also transfected with siRNAs against GL2 (control) or μ2 subunit of AP2 complex. Antibody feeding (internalized anti-CD8/Alexa 568) is shown to assess internalization. Inhibition of uptake of transferrin–Alexa 647 was used as a functional test of knockdown efficacy. Bar, 10 µm. (B) Quantification of chemically induced endocytosis. Fraction of above-threshold puncta that were green (FKBP–β2–GFP) and red (A568-CD8) is shown for each cell (spots); bars indicate mean ± SD. Number of experiments = 3. ANOVA, P = 1.48 × 10−8. Tukey test for siGL2 versus siμ2, P = 0.5. (C) Western blot (WB) to assess depletion of μ2 by RNAi. Cells were prepared in parallel with the experiment shown in C. Blotting for μ2 or for GAPDH (as a loading control) is shown.
Figure 5.
Figure 5.
Hot-wired endocytosis does not inhibit normal CME. (A) Confocal images of HeLa cells expressing FKBP–β2–GFP alone or together with CD8–mCherry–FRB were tested for ability to internalize transferrin–Alexa 647 (A647-Tf) in the presence or absence of rapamycin (200 nM; filled orange bar). Bars: (main) 10 µm; (inset) 2 µm. (B) Quantification of transferrin uptake and number of GFP-positive puncta in cells. Scatterplot shows values for each cell, bars indicate mean ± SD. Number of experiments = 2. Tf, ANOVA, P = 0.1; GFP, ANOVA, P = 5.89 × 10−6.
Figure 6.
Figure 6.
Chemically induced endocytosis is inhibited during mitosis. Live-cell confocal imaging experiment, mitotic cell expressing CD8–mCherry–FRB with FKBP–β2–GFP. Rapamycin (200 nM) was applied during metaphase. Clear rerouting occurs, but no puncta form in the cytoplasm until after cytokinesis. This cell is shown in Video 5. Bar, 10 µm.
Figure 7.
Figure 7.
Chemically inducible internalization: comparison of clathrin hooks. (A) Comparison of four potential clathrin hooks for chemically inducible endocytosis. Cells expressed CD8–mCherry–FRB with FKBP–β1–GFP, FKBP–α–GFP, FKBP–epsin–GFP, or FKBP–β3–GFP. Stills from live-cell confocal imaging experiments are shown: (left) image shows the frame before rerouting occurs and (right) 133 frames (665 s) later. Inset shows a 2× zoom of the boxed region. Bars: (main) 10 µm; (insets) 2 µm. (B) Mean plots to show chemically inducible endocytosis. Colored traces indicate the total area of bright GFP puncta that form in cells after rerouting. Traces were time aligned and averaged. (C) Scatterplot to indicate variability in chemically induced endocytosis. The total normalized area above threshold at 5 min after rerouting for each cell analyzed is shown (spots). ANOVA, P = 1.142 × 10−5. Bars indicate mean ± SD. Number of cells = 8–15 and number of experiments = 3.
Figure 8.
Figure 8.
Chemically inducible internalization: using a clathrin hook of nonendocytic origin. (A) Colocalization of clathrin with bright green puncta formed by chemically induced internalization with a GTSE1 clathrin hook. Cells were coexpressing CD8–dCherry–FRB, mCherry–LCa, and FKBP–GTSE1–GFP. (B) Scatterplot to indicate variability in chemically induced endocytosis. The total normalized area above threshold at 10 min after rerouting for each cell analyzed is shown (spots). Bars: (main) 10 µm; (insets) 2 µm. Bars indicate mean ± SD. Number of cells = 35 and number of experiments = 10. (C) Antibody uptake after hot-wiring using a GTSE1 clathrin hook. Cells coexpressing CD8–mCherry–FRB and FKBP–GTSE1–GFP were incubated with anti-CD8–Alexa 647 to label the membrane anchor extracellularly, and rapamycin (200 nM) was applied. In A and C, stills from live-cell confocal imaging experiments are shown: (top) the frame before rerouting occurs and (bottom) 180 frames (15 min) later; empty and filled orange bars indicate before and after rapamycin addition. Bars: (main) 10 µm; (insets) 2 µm. (D) Table to show the rate of puncta appearance for GTSE1 compared with other clathrin hooks. Slope is the coefficient of a line fit to the averaged puncta data, postrapamycin; units are puncta per second.
Figure 9.
Figure 9.
Clathrin functionally engages AP2 at the plasma membrane using two distinct interactions. (A) Schematic illustration of the putative interactions between β2 and clathrin. Inset shows mutations to delete the CBM (ΔCBM) or mutation of tyrosine residue Y815 in the appendage (Y-A). Note that Y815 in earlier papers refers to Y829 in our longer isoform. (B) Analysis of clathrin–β2 interaction in vitro. Binding experiments using GST–β2(616–951) WT or mutants to pull down MBP–CHC(1–1074) as indicated. Interaction was assessed by Western blotting (i), and a Coomassie-stained gel was run in parallel to check for equivalent capture on beads. Three independent experiments were performed and analyzed by densitometry (ii). Bar indicates mean. (C) Quantification of live-cell confocal imaging experiments to show extent of endocytosis triggered by FKBP–β2–GFP or related mutants (see inset). Colored traces indicate the total area of bright GFP puncta that form in cells after rerouting. Traces were time aligned and averaged. Mean traces are shown (i), and a scatterplot (ii) of values at 5 min after rerouting for each cell analyzed is shown (spots). Bars indicate mean ± SD. Number of cells = 11–15 and number of experiments = 3. ANOVA, P = 7.92 × 10−5. Note that the values for FKBP–β2–GFP (WT) and GFP–FKBP (GFP) are also shown in Fig. 8. Example live-cell movies are shown in Videos 6, 7, and 8. (D) Representative confocal images (i) of an immunolabeling experiment to assess endocytosis of CD8 by triggered endocytosis. Bars: (main) 10 µm; (insets) 2 µm. Scatterplot (ii) to indicate variability in chemically induced endocytosis. The fraction of green puncta that were also red for each cell analyzed is shown (spots). Bars indicate the mean ± SD. Number of cells = 8–14 and number of experiments = 3. ANOVA, P = 1.66 × 10−6.
Figure 10.
Figure 10.
Optically inducible endocytosis. (A) Illustration of optically induced endocytosis. Cells coexpress a plasma membrane anchored LOVpep (CD8–TagRFP657–LOVpep) and a clathrin hook with a PDZb1 tag (PDZb1–β2–mCherry). Upon illumination with blue light, an epitope is exposed to which the PDZ domain can bind, and the clathrin hook is rerouted to the plasma membrane so that endocytosis can occur at that site. (B) Rapid and reversible ePDZb1–mCherry recruitment to CD8–TagRFP657–LOVpep(T406A,T407A,I532A) at the plasma membrane. Brief (<1 s) patterned illumination of plasma membrane areas followed by rapid imaging. Mean pixel density of an ROI at the membrane (purple) compared with one in the cytoplasm (gray) is shown. (C) Endocytosis is activated in a spatiotemporal manner by sustained, patterned light activation within a defined ROI (blue dashed box). For analysis, the number of vesicles formed within the ROI (purple box) is compared with a similar ROI outside the activated region (gray box). (D) Stills from a typical optogenetic hot-wiring experiment. HeLa cell expressing CD8–TagRFP657–LOVpep(T406A,T407A,I532A) and ePDZb1–β2–mCherry. Only the mCherry confocal channel is shown. Time 0 indicates when patterned illumination began. Bar, 10 µm. A movie of this panel is available as Video 9. (E) Plots of mean ± SEM vesicle formation in cells expressing CD8–TagRFP657–LOVpep(T406A,T407A,I532A) and either ePDZb1–β2–mCherry (number of cells = 13 and number of experiments = 4) or ePDZb–β2–mCherry (number of cells = 8 and number of experiments = 3) as indicated. Purple and gray indicate vesicles formed inside and outside the photo-activation zone, respectively.

Similar articles

See all similar articles

Cited by 6 articles

See all "Cited by" articles

References

    1. Aguet F., Antonescu C.N., Mettlen M., Schmid S.L., and Danuser G. 2013. Advances in analysis of low signal-to-noise images link dynamin and AP2 to the functions of an endocytic checkpoint. Dev. Cell. 26:279–291. 10.1016/j.devcel.2013.06.019 - DOI - PMC - PubMed
    1. Borner G.H., Antrobus R., Hirst J., Bhumbra G.S., Kozik P., Jackson L.P., Sahlender D.A., and Robinson M.S. 2012. Multivariate proteomic profiling identifies novel accessory proteins of coated vesicles. J. Cell Biol. 197:141–160. 10.1083/jcb.201111049 - DOI - PMC - PubMed
    1. Cheeseman L.P., Harry E.F., McAinsh A.D., Prior I.A., and Royle S.J. 2013. Specific removal of TACC3-ch-TOG-clathrin at metaphase deregulates kinetochore fiber tension. J. Cell Sci. 126:2102–2113. 10.1242/jcs.124834 - DOI - PMC - PubMed
    1. Dannhauser P.N., and Ungewickell E.J. 2012. Reconstitution of clathrin-coated bud and vesicle formation with minimal components. Nat. Cell Biol. 14:634–639. 10.1038/ncb2478 - DOI - PubMed
    1. Dell’Angelica E.C., Klumperman J., Stoorvogel W., and Bonifacino J.S. 1998. Association of the AP-3 adaptor complex with clathrin. Science. 280:431–434. 10.1126/science.280.5362.431 - DOI - PubMed

Publication types

MeSH terms

Feedback