Microtubule plus-end-tracking proteins target gap junctions directly from the cell interior to adherens junctions

Abstract

Gap junctions are intercellular channels that connect the cytoplasms of adjacent cells. For gap junctions to properly control organ formation and electrical synchronization in the heart and the brain, connexin-based hemichannels must be correctly targeted to cell-cell borders. While it is generally accepted that gap junctions form via lateral diffusion of hemichannels following microtubule-mediated delivery to the plasma membrane, we provide evidence for direct targeting of hemichannels to cell-cell junctions through a pathway that is dependent on microtubules; through the adherens-junction proteins N-cadherin and beta-catenin; through the microtubule plus-end-tracking protein (+TIP) EB1; and through its interacting protein p150(Glued). Based on live cell microscopy that includes fluorescence recovery after photobleaching (FRAP), total internal reflection fluorescence (TIRF), deconvolution, and siRNA knockdown, we propose that preferential tethering of microtubule plus ends at the adherens junction promotes delivery of connexin hemichannels directly to the cell-cell border. These findings support an unanticipated mechanism for protein delivery to points of cell-cell contact.

Figure 1. Microtubules Are Associated with Cx43 Plaques at the Cell-Cell Border

(A) Confocal image (100×) of an isolated HeLa cell that was transfected with Cx43-YFP (green) and then fixed and stained with antibody to α-tubulin (red); this cell was selected for its low Cx43-YFP expression level for visualization of Cx43-YFP distribution on microtubules. (B) Deconvolution of live cell images of HeLa cells transfected with α-tubulin-GFP (red) and Cx43-RFP (green) is shown. (C) Enlargement of cell-cell border region from (B) is shown. (D) Manually traced composite cell-cell border region of a deconvolved 1.5 μm stack of images (consisting of 15 100 nm planes). Microtubules of right cell (red), microtubules of left cell (blue), and Cx43 plaques (green) are shown.

Figure 2. Cx43 Plaques Are Repopulated in a Microtubule-Dependent Manner

(A) FRAP procedure and plaque repopulation with Cx43-YFP in the presence and absence of nocodazole and Taxol are shown. (B) Recovery of fluorescence quantified and averaged for images viewed at either the XY or XZ plane is shown. Filled symbols indicate recovery within plaques, and open symbols indicate recovery in surrounding nonplaque regions. Data are represented as mean ± SEM. (n = 12 cell pairs, n = 8, and n = 9 for control, nocodazole, and Taxol conditions, respectively.)

Figure 3. Microtubule Disruptors Limit Plaque Formation

(A) HeLa cells transfected with Cx43-YFP (green), exposed to drugs, then fixed and immunostained with anti-α-tubulin (red) and a nuclear stain (blue) are shown. (B) Quantification of the effects of nocodazole and Taxol on Cx43 plaque formation at cell-cell borders (*p < 0.001) is shown. Data are represented as mean ± SEM. (n = 20 cell pairs for all three conditions.)

Figure 4. Cortical EB1 Dynamics Are Different at Cx43 Plaques

(A) Live HeLa cells cotransfected with both Cx43-RFP (green) and GFP-EB1 (black) are shown. A growing plaque is chosen (left side of bottom cell, Cx43 signal enhanced), and 2 μm radius semicircular regions are drawn at the plaque and at two regions of nonplaque cortical membrane (red semicircles, bottom panel). (B) EB1 dynamics within each semicircular region are shown. (C) Quantification of EB1 events at Cx43 plaques (n = 25 cells), at cell-cell borders without plaque (n = 22 cells), and at noncontacting cell edges (n = 24 cells) is shown. Last bar indicates EB1 events at cell-cell border of cells subject to p150(Glued)-siRNA knockdown (n = 10 cells; *p < 0.0001, **p < 0.01). Data are represented as mean ± SEM. (D) shows TIRF imaging of membrane and subcortical regions of HeLa cells transfected with EB1-GFP on coverslips that contain (top row) and do not contain (bottom row) the extracellular domain of cadherin. The presence of cadherin increases the frequency and duration of EB1 interaction with the cortical membrane (see quantification in Figure S1).

Figure 5. Cx43 Vesicle Delivery to +Cadherin Membrane

(A) Cx43-YFP in HeLa cells plated on coverslips containing the extracellular domain of N-cadherin (+cadherin) is shown as visualized with TIRF microscopy. In the three enlarged panels, the appearance and the paths of vesicles are tracked for 1 min of imaging. Scale bar is 4 μm. (B) Twenty seconds of vesicle movement at 5 s intervals, taken from the region in (A) are shown. (C) Cx43-YFP in HeLa cells plated on coverslips lacking N-cadherin (−cadherin) is shown. Scale bar is 4 μm. (D) Twenty seconds of vesicle movement at 5 s intervals, taken from the region in (C) are shown. (E) Average path length of vesicle movement (+cadherin, n = 78; −cadherin, n = 75; *p < 0.001) is shown. Data are represented as mean ± SEM. (F) Computed average of vesicle fusion events (incidence of vesicles becoming brighter and stationary) per minute per cell is shown. Data were normalized to average cell surface area. (+cadherin, n = 18; −cadherin, n = 17; **p = 0.006). Data are represented as mean ± SEM.

Figure 6. Plaque Formation Depends on p150(Glued) and β-catenin

(A) p150(Glued) (red) immunostaining in HeLa cells expressing Cx43-YFP (green) without and with siRNA knockdown is shown. (B) β-catenin (red) immunostaining in HeLa cells expressing Cx43-YFP (green) without and with siRNA knockdown is shown. (C) Quantification of Cx43 plaque (percent of cell-cell border) after siRNA knockdown of EB1 (n = 17), p150(Glued) (n = 19), and β-catenin (n = 20) is shown (*p < 0.0001). Data are represented as mean ± SEM. (D) β-catenin and Cx43 colocalization at the two intercalated discs (IDs) of a single binucleated (N) adult rat cardiomyocyte are shown. (E) p150(Glued) and Cx43 colocalization at the ID of adult rat cardiomyocyte are shown.

Figure 7. Gap Junction Plaques Depend on Homophilic Cadherin Interactions

(A) Immunostaining of Cx43-YFP-transfected (green) control cells (first row) and cells exposed to cadherin-blocking peptides (second row) is shown. Nuclei are shown in blue, and N-cadherin is shown in red. (B) Quantification of Cx43 plaques (percent at the cell-cell border) for control cell pairs (n = 27) and cell pairs exposed to N-cadherin-blocking peptide (n = 25) is shown (*p < 0.0001). The proposed paradigm for gap junction plaque formation at the adherens junctions (AJ) that exist at cell-cell contacts and provide an anchor for microtubule plus ends is also shown. Data are represented as mean ± SEM. (C) The presence of AJ greatly enhances EB1 interaction with the cortical membrane as revealed by TIRF microscopy. (D) Cortical capture of microtubules by AJ allows localized deposition of Cx43 hemichannels into the plasma membrane, whereas microtubules typically fail to reach the cell membrane without AJ. (E) From microtubule to cadherin, cortical capture and Cx43 delivery involves these intermediates: EB1, p150(Glued) of the dynein/dynactin complex, and β-catenin as revealed by siRNA knockdown of these proteins and by the coimmunoprecipitation of EB1 and Cx43. The result is focal delivery of Cx43 hemichannels directly to plaques at the cell-cell border.