Skip to main content
NCBI home page
EMBO Reports logoLink to EMBO Reports
. 2021 Feb 8;22(3):e50815. doi: 10.15252/embr.202050815

Mobile late endosomes modulate peripheral endoplasmic reticulum network architecture

Menno Spits 1, Iris T Heesterbeek 1, Lennard M Voortman 1, Jimmy J Akkermans 1, Ruud H Wijdeven 1, Birol Cabukusta 1,✉, Jacques Neefjes 1,✉
PMCID: PMC7926257  PMID: 33554435

Abstract

The endoplasmic reticulum (ER) is the largest organelle contacting virtually every other organelle for information exchange and control of processes such as transport, fusion, and fission. Here, we studied the role of the other organelles on ER network architecture in the cell periphery. We show that the co‐migration of the ER with other organelles, called ER hitchhiking facilitated by late endosomes and lysosomes is a major mechanism controlling ER network architecture. When hitchhiking occurs, emerging ER structures may fuse with the existing ER tubules to alter the local ER architecture. This couples late endosomal/lysosomal positioning and mobility to ER network architecture. Conditions restricting late endosomal movement—including cell starvation—or the depletion of tether proteins that link the ER to late endosomes reduce ER dynamics and limit the complexity of the peripheral ER network architecture. This indicates that among many factors, the ER is controlled by late endosomal movement resulting in an alteration of the ER network architecture.

Keywords: endoplasmic reticulum, late endosomes, membrane contact sites, organelle hitchhiking, starvation

Subject Categories: Membrane & Intracellular Transport,

The endoplasmic reticulum co‐migrates or “hitchhikes” with the RAB7‐positive late endosomes/lysosomes in a membrane contact site‐mediated manner to determine the architecture of the peripheral ER network.

graphic file with name EMBR-22-e50815-g013.jpg

Introduction

The endoplasmic reticulum (ER) is the largest intracellular membrane compartment with a complex architecture covering most cellular space outside the nucleus and interacting with every other organelle and the plasma membrane. The ER functions as a multifunctional platform mediating a plethora of biological processes, such as lipid synthesis, protein production and folding, controlling calcium homeostasis, and antigen presentation for immune surveillance (Elliott & Neefjes, 2006; Raffaello et al, 2016). The ER interacts with other organelles by forming molecular bridges and creating membrane contact sites (MCS) (Rowland et al, 2014; Hoyer et al, 2018). At MCS, the ER exchanges information and metabolites, such as cholesterol and calcium, between organelles. Overall, the ER controls many intracellular processes.

The ER forms a complex network structure with subdomains in the form of tubules and sheets, linked with junctions to generate its characteristic netted shape. This complex ER architecture is created and maintained through a wide range of proteins and lipids. Formation of membrane curvature and tubule formation requires proteins containing reticulon homology domains, which have a characteristic hairpin‐like transmembrane structure (Wang et al, 2016; Bhaskara et al, 2019). Meanwhile, three‐way ER junctions are formed by Atlastin‐1 homodimers on opposing ER membranes orchestrating homotypic fusion. Following membrane fusion, Lunapark localizes to the nascent junctions to stabilize the unique inward curvature of the junction (Goyal & Blackstone, 2013; Chen et al, 2015; Wang et al, 2016; Zhou et al, 2019).

The architecture of the ER is highly dynamic and undergoes constant remodeling (Guo et al, 2018; Zhou et al, 2019). Two mechanisms of ER movement have been described. In the first mechanism, the tip attachment complex (TAC)‐mediated ER movement, the ER‐resident protein STIM1 interacts with EB1 at the microtubule plus‐end and a new ER tubule travels alongside the polymerizing microtubule. Overexpression of STIM1 or inhibiting (de)polymerization of the microtubules results in an ER structure strongly resembling the shape of the microtubule network structure (Grigoriev et al, 2008; English et al, 2009). The second mechanism, named ER sliding occurs through the direct interaction between the motor protein Kinesin‐1 and the ER protein Kinectin allowing the ER to expand in the plus‐end direction of microtubules (Santama et al, 2004; Lin et al, 2012; Goyal & Blackstone, 2013). In parallel, a recently described third way of ER movement called ER hitchhiking demonstrated that ER tubules co‐migrate with an associated organelle (Salogiannis & Reck‐Peterson, 2017; Guo et al, 2018; Mogre et al, 2020). Here, ER tubules are being drawn forward by an associated organelle such as peroxisomes or lipid droplets, copying its mobility (Zajac et al, 2013; Guimaraes et al, 2015; Salogiannis et al, 2016). However, the frequency and nature of ER hitchhiking events and their relation to previously described ER movement mechanisms are unclear.

To investigate endosome‐related ER hitchhiking and to understand the relationship between ER movement and endosomal movement, we developed a method to quantify ER movement in association with other membrane structures. With this tool, we observed that ER hitchhiking occurs with various types of endosomes, but preferably RAB7‐marked late endosomes and lysosomes (LE/Ly). We show that late endosomal ER hitchhiking occurs independent of TAC‐mediated ER movement or ER sliding and contributes to half of all ER movements in COS7 cells. Reducing late endosomal transport by starvation or introducing factors driving LE/Ly to the microtubule minus‐end to remove them from the cell periphery also limits ER hitchhiking and results in a less complex and more open ER network in the cell periphery. Depletion of MCS proteins that tether the ER to LE/Ly, the ER‐resident VAP proteins, significantly reduced ER hitchhiking events, suggesting a novel function of ER‐LE/Ly contact sites. We propose that late endosomal mobility in the cell periphery is a major contributor to the formation and maintenance of the local ER network architecture.

Results

Mobile late endosomes/lysosomes and ER movement

To assess the contribution of endosomes on ER movement, we marked the ER by fusing eGFP to the N‐terminal cytosolic tail of the transmembrane region of the ER‐localized MOSPD2 or with mCherry‐KDEL (Di Mattia et al, 2018). COS7 cells stably expressing these constructs were used in our experiments. We visualized LE/Ly with the lysosome‐specific fluorophore SiR‐lysosome. The behavior of the ER relative to the LE/Ly compartment was imaged using parallel high‐speed acquisition by a spinning disk confocal microscope. We observed movements suggesting that LE/Ly co‐migrate with ER tubules. This was further illustrated by the accompanying kymograph (Fig 1A and B, Movie EV1) and is in line with previous observations (Guimaraes et al, 2015; Salogiannis et al, 2016; Guo et al, 2018). Notably, these movements often appeared to yield novel ER network junctions. Following the formation of new junctions, 78% of the observed vesicle‐ER tubule pairs resume with their co‐migration. To examine whether the LE/Ly co‐migration with ER tubules, so‐called ER hitchhiking events, involve the leading tip of dynamic ER tubules, we used CEPT1 and PIS, two previously described markers for these ER subdomains (English & Voeltz, 2013). Protein localizations during hitchhiking events revealed that PIS and CEPT1 are indeed enriched in 78% and 80% of the hitchhiking events, respectively (Fig EV1A and B, Movies EV4 and EV5). This suggests that ER hitchhiking events mainly take place at the leading edge of dynamic ER tubules.

Figure 1. Motility of late endosomes/lysosomes and the ER network.

Figure 1

Open in a new tab
  1. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing ER marker GFP‐TMD (green). The cells were loaded with SiR‐lysosome to visualize LE/Lys (orange). Zoom insets show a merge of the ER (GFP‐TMD) (green) and SiR‐lysosome labeled LE/Lys (orange). Zoom insets show a region of interest containing a mobile endosome as denoted by the white arrow. Scale bars: full size 10 µm, zoom inset 3 µm. Images related to Movie EV1.
  2. Kymograph of the representative image in panel (A).
  3. Schematic flow of membrane displacement analysis method.
  4. Representative time‐lapse spinning disk confocal images of n = 3 of COS7 cells stably expressing GFP‐TMD (ER) (green) and SiR‐lysosome (orange). Zoom insets highlight regions of interest containing a static (upper panels) or mobile endosome (lower panels) as denoted by the white arrows. First row shows ER (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two channels, and the fourth row shows a merge of ER (green) and LE/Lys (orange). Scale bars: full size 10 µm, zoom inset 3 µm. Images relate to Movies EV2 and EV3.
  5. Kymograph of the mobile vesicle in panel (D).
  6. MDA quantification of (D). Mobility of static and mobile endosomes was quantified and normalized to the static endosomes. Graph represents mean ± s.d. of analyses at multiple locations within the same cell(n = 20 cells per condition from 3 independent experiments). All images acquired for 90 frames at 4Hz. Significance two‐tailed Student t‐test, ****P < 0.0001.

Figure EV1. PIS and CEPT1 bare enriched in domains following LE/Ly‐mediated ER hitchhiking.

Figure EV1

Open in a new tab
  1. Representative stills of spinning disk confocal images of n = 2 of COS7 cells stably expressing GFP‐TMD (green), RFP‐PIS (orange) (left panels), or RFP‐CEPT1 (right panels) (orange) and containing lysosomal marked SiR‐lysosome (magenta). Zoom insets show region of interest depicting a lysosomes pulling an ER tubule containing an RFP‐PIS or RFP‐CEPT1 enriched domain. All images acquired for 120 frames at 1Hz. Image scale bars: 20 µm and zoom insets: 2 µm.
  2. Percentage of SiR‐lysosomes marked vesicles followed by ER containing a PIS or CEPT1 enriched domain. Graph represents n = 20 vesicles from 2 independent experiments.
  3. Graph representing the percentage of SiR‐lysosome marked vesicles that either or not continued pulling an ER tubule following junction formation, as indicated. Graph represents n = 20 vesicles from 2 independent experiments.
  4. Graph shows mean ± s.d. of validation of siRNA treated COS7 cell depletion efficiencies expressed as fraction of the mRNA of Kinectin‐1 (siKTN1) relative to siRNA control (siControl) by qPCR. n = 3 independent experiments. Normalized to control as set at 1. Significance Student two‐tailed t‐test. ***P < 0.001.
  5. Western blot analyses of COS7 cells transfected with siKinectin1 or siControl and stained with rabbit anti‐human KTN1. Loading control is stained for actin.

To quantify ER hitchhiking processes, a method was developed for automated image analysis. The available analyses methods—such as particle tracking and network analysis—are unable to accurately quantify the membrane displacement of heterogeneously shaped membranes such as the ER. Therefore, we developed a membrane displacement analysis (MDA) method based on optical flow analysis (Horne, 1981) (Fig 1C). This tool allowed us to accurately assess the movement fraction of any fluorescently labeled membrane and captured any detectable motion without requiring an a priori accurate and reliable localization of points of interest.

Using MDA, we analyzed image sequences acquired at 4Hz at multiple areas of 36 µm2 per cell, in a range of cells, as an accurate representation of membrane structures moving in relation to each other. We discerned differences in ER movement surrounding static late endosomes and mobile late endosomes (Fig 1D–F, Movies EV2 and EV3). This revealed a 1.8‐fold increase of ER movement surrounding mobile late endosome (Fig 1F). Thus, MDA showed a significant increase in ER movement when LE/Ly moved through the focused area and suggests that LE/Ly are involved in ER dynamics. Furthermore, we observed that 80% of the hitchhiking events continue to interact with the ER following apparent ER network formation (Fig EV1C), suggesting the absence of a specific hitchhiking termination following these events.

ER hitchhiking occurs independently of TAC‐mediated ER movement and ER sliding

Two mechanisms of ER mobility have been reported: TAC‐mediated ER movement and ER sliding (Fig 2A). We assessed whether these mechanisms could explain ER hitchhiking with LE/Ly. To study the role of TAC‐mediated ER movement, we evaluated ER hitchhiking occurring along polymerizing microtubules in cells expressing EB3‐GFP, which selectively marks growing microtubules (van de Willige et al, 2016). An ER tubule following a late endosome was often detected in the absence of any nearby EB3‐GFP signal (Fig 2B, Movie EV6). To further determine the contribution of growing microtubules in ER hitchhiking, we exposed cells to paclitaxel, which blocks microtubule polymerization therefore abolishes TAC‐mediated ER movement (Horwitz, 1994). As expected, paclitaxel treatment reduced the total ER mobility significantly; however, it failed to affect local ER movements connected to mobile LE/Ly (Fig 2C–E, Movie EV7 and EV8). Alternatively, ER sliding is responsible for the ER movement observed during ER hitchhiking events. ER sliding is facilitated by the ER‐resident protein Kinectin interacting with the microtubule motor Kinesin‐1 (Santama et al, 2004; Friedman et al, 2010; Lin et al, 2012; Goyal & Blackstone, 2013; Chen et al, 2015). To assess this possibility, cells were depleted of Kinectin and analyzed by time‐lapse microscopy (Fig EV1D and E). MDA confirmed that Kinectin‐depleted cells demonstrated reduced total ER movement. But the ER movement surrounding LE/Ly hitchhiking events remained unaffected (Fig 2F–H, Movies EV9 and EV10). These data confirm that TAC‐mediated ER movement and ER sliding contribute significantly to the total ER movement, but also suggest that ER hitchhiking with LE/Ly occurs independently of these two mechanisms.

Figure 2. ER hitchhiking and other processes of ER movement.

Figure 2

Open in a new tab
  1. Schematic representation of TAC‐mediated (I) ER movement and ER sliding (II).
  2. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing mCherry‐KDEL (green), transiently expressing EB3‐GFP (yellow), and SiR‐lysosome‐stained LE/Lys (magenta). Zoom insets show regions of interest containing a mobile LE/Lys as denoted by the white arrow. First row shows ER (mCherry‐KDEL) (green), the second row shows SiR‐lysosome‐stained LE/Lys (magenta), the third row shows EB3‐GFP (yellow) and the fourth row shows the merge of the three preceding channels. Scale bar: 30 µm and zoom insets 3 µm. All images acquired for 90 frames at 1Hz. Images related to Movie EV6.
  3. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing GFP‐TMD (green) marked ER and SiR‐lysosome‐stained LE/Lys (orange). Zoom insets highlight regions of interest containing a mobile endosome as denoted by the white arrow. First row represents ER (GFP‐TMD) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels, and the fourth row shows a merge of ER (green) and LE/Lys (orange). Samples were treated with either DMSO or 2 µM paclitaxel. Scale bar: 30 µm and zoom insets 3 µm. All images acquired for 90 frames at 1Hz. Images related to Movie EV7 and EV8.
  4. Kymograph of panel (C).
  5. MDA quantification of panel (C) normalized to control cells, showing the average ER movement surrounding LE/Lys under control or treated conditions and the overall average ER mobility under control or treatment conditions. Graph represents mean ± s.d. of analyses at multiple locations within the same cell (n = 20 cells per condition from 3 independent experiments). Significance two‐tailed Student t‐test ****P < 0.0001; ns = not significant.
  6. Representative images of n = 3 of COS7 cells stably expressing GFP‐TMD (green) to mark the ER and SiR‐lysosome stained LE/Lys (orange) merged. Zoom insets highlight region of interest containing a mobile endosome as denoted by the white arrow. First row represents ER (GFP‐TMD) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels, and the fourth row shows a merge of ER (green) and LE/Lys (orange). Samples were treated with either siControl or siKTN1. Scale bar: 30 µm and zoom insets 3 µm. All images acquired for 90 frames at 1Hz. Images related to Movies EV9 and EV10.
  7. Kymograph of panel (F).
  8. MDA quantification of panel (F) normalized to control cells, showing the average ER movement surrounding LE/Lys under control or treated conditions and the overall average ER mobility under control or treatment conditions, respectively. Graph represents mean ± s.d. of analyses at multiple locations within the same cell (n = 20 cells per condition from 3 independent experiments). Significance two‐tailed Student t‐test. ****P < 0.0001; ns = not significant.

ER hitchhiking mainly involves peripheral mobile RAB7‐marked late endosomes that are manipulated by TMEM55B and starvation

We wondered whether all endosomal structures contribute to ER hitchhiking or whether RAB7‐marked LE/Ly are the main contributors. For this, various endosomes were marked by GFP‐labeled RAB GTPases: RAB5 for early endosomes, RAB11 and RAB14 for recycling endosomes, and RAB7 for LE/Ly (Ullrich et al, 1996; Jordens et al, 2001; Junutula et al, 2004; Rocha et al, 2009; Friedman et al, 2013; Wijdeven et al, 2016; Bakker et al, 2017; Langemeyer et al, 2018; Zulkefli et al, 2019). We used MDA to quantify ER movement associated with the movement of each endosomal population. Mobile RAB5, RAB11, and RAB14 mobility were all associated to an average of 1.2‐fold increase in ER mobility in the analyzed areas compared to their static counterparts (Fig 3C, Fig EV2A–F, Movies EV13–EV18). Mobile RAB7 demonstrated a 2‐fold increase in ER mobility compared to its static counterparts (Fig 3A–C, Fig EV2G, Movies EV11 and EV12). Meanwhile, overexpression of RABs did not influence the overall mobility of ER or vesicles, or caused aberrant changes in the peripheral ER architecture (Fig EV2H–K). This suggests that RAB7‐positive LE/Ly contribute to ER hitchhiking significantly more than the other endosomal compartments tested.

Figure 3. RAB7‐positive endosomes and ER mobility.

Figure 3

Open in a new tab
  1. Representative time‐lapse spinning disk confocal images of n = 3 live COS7 cells stably expressing mCherry‐KDEL (green) and transiently expressing GFP‐RAB7 (orange). Zoom insets show a region of interest containing a mobile RAB7‐positive endosome as denoted by the white arrows. First row represents the ER (mCherry‐KDEL) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels, and the fourth row shows a merge of ER (green) and GFP‐RAB7 (orange). Scale bar: 10 µm, zoom insets 3 µm. Images related to Movie EV11. Images were acquired for 90 frames at 1Hz.
  2. Kymograph of panel (A).
  3. MDA quantification of panel (A) and Fig EV2A–G normalized to 1 showing average ER movement surrounding static endosomes and mobile endosomes. Graph represents mean ± s.d. of analyses at multiple locations within the same cell (n = 15 cells per condition from 3 independent experiments). All images acquired for 90 frames at 1Hz. Significance two‐tailed Student t‐test. *P < 0.05, **P < 0.01, ****P < 0.0001 ns = not significant.
  4. Violin plots of fractional distance analysis showing the average distribution of endosomes in cells from panel (E) and Fig EV3A. Graph represents mean ± s.d. of n = 15 cells per condition from 3 independent experiments. Dashed lines represent median and solid lines represent interquartile range. Cartoons representing the overexpression and control LE/Ly phenotype. Significance two‐tailed Student t‐test. ****P < 0.0001.
  5. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing GFP‐TMD (green) and in lower panels transiently expressing mScarlet‐RILP (magenta) and SiR‐lysosome‐stained LE/Lys (orange). Zoom insets shows region of interest containing ER (GFP‐TMD). First row represents ER (GFP‐TMD) (green), the second row shows MDA‐generated ER movement (magenta), and third row shows a merge of the two preceding channels. Scale bar: 10 µm, zoom insets 3 µm. Images related to Movie EV19 and Movie EV20. Images were acquired for 90 frames at 1Hz.
  6. MDA quantification of panel (E) and Fig EV3A normalized to 1 showing total average ER movement. Graph represents mean ± s.d. of analyses at multiple locations within the same cell (n = 15 cells per condition from 3 independent experiments). All images acquired 90 frames at 1Hz. Significance two‐tailed Student t‐test. ****P < 0.0001 ns = not significant.

Figure EV2. ER movement related to various static and mobile endosomal populations.

Figure EV2

Open in a new tab
  • A–G
    Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing mCherry‐KDEL (green) and transiently expressing GFP‐RAB5 (A and B), GFP‐RAB11 (C and D), GFP‐RAB14 (E and F), or GFP‐Rab7 (G) (orange). Zoom insets shows region of interest containing a mobile (A, C, and E) or static (B, D, F, and G) endosome as denoted by the white arrow. First row represents ER (mCherry‐KDEL) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels, and the fourth row shows a merge of ER (green) and marked endosomal population (orange). Panels A and B; C and D; and E and F are representatives of mobile and static RAB‐positive structures within the same cells. Scale bar: 10 µm, zoom insets 3 µm. Images related to Movies EV14–EV20. Images were acquired for 90 frames at 1Hz.
  • H
    Normalized MDA quantification of total ER movement of COS7 cells not transfected or overexpressing GFP‐RAB5, GFP‐RAB7, GFP‐RAB11, or GFP‐RAB14. Graph represents mean ± s.d. of analyses at multiple locations within one cell of n = 15 cells from 3 independent experiments. Normalized to control set at 1. Significance two‐tailed Student t‐test. ns = not significant
  • I
    Normalized MDA quantification of total movement of vesicles marked with GFP‐RAB5, GFP‐RAB7, GFP‐RAB11, or GFP‐RAB14 in COS7 cells. Graph represents mean ± s.d. of n = 15 cells from 3 independent experiment. Normalized to control set at 1. Significance two‐tailed Student t‐test. Ns = not significant.
  • J
    MDA quantification of total vesicle movement and total ER movement of COS7 cells treated with SiR‐Lysosome or transfected with GFP‐RAB7. All images acquired for 90 frames at 1Hz. Graph represents mean ± s.d. from n = 15 (two left bars) and the mean ± s.d. of analyses at multiple locations within one cell of n = 15 (two right bars) cells from 3 independent experiments. Significance two‐tailed Student t‐test. Ns = not significant.
  • K
    Still images of n = 3 of COS7 cells transfected with RFP‐SEC61B (magenta) and GFP‐RAB5, GFP‐RAB7, GFP‐RAB11, or GFP‐RAB14 (green). Zoom insets show regions of interest of peripheral ER morphology. Scalebar 5 µm zoom inset 2 µm.

To assess the overall contribution of LE/Ly transport to ER movement, we blocked their mobility by relocating these structures to the microtubule minus‐end close to the microtubule‐organizing center (MTOC) and thus avert their presence in the cell periphery. This was accomplished by overexpressing either the RAB7 effector RAB‐interacting lysosomal protein (RILP) or the phosphatidylinositol‐4,5‐biphosphate 4‐phosphatase (TMEM55B). RILP recruits p150glued and hereby the dynein motor to LE/Ly. TMEM55B localizes in LE/Ly and binds to JIP4 which then recruits the dynein motor. Recruitment of dynein to LE/Ly by RILP or TMEM55B leads to minus‐end transport and their clustering in the perinuclear region (Jordens et al, 2001; Rocha et al, 2009; Wijdeven et al, 2016; Willett et al, 2017; Cabukusta & Neefjes, 2018). Indeed, overexpression of RILP or TMEM55B resulted in a characteristic LE/Ly clustering phenotype (Fig 3D). We performed MDA on ER movement in these cells in fields of 36 µm2 at various locations in order to capture the average motility of the ER alone. This yielded an average reduction of 50% in overall ER movement in cells overexpressing RILP and 60% in cells overexpressing TMEM55B relative to control samples (Fig 3E–F, Fig EV3A, Movies EV19–EV21). Of note, overexpression of RILP or TMEM55B was selective for LE/Ly and did not reduce the movement of mitochondria, peroxisomes or RAB5‐positive vesicles (Fig EV3B–D, Movies EV22–EV29). It appears that constraining LE/Ly transport significantly reduces total ER mobility.

Figure EV3. TMEM55B‐ and starvation‐ induced endosomal sequestration limits ER movement and peripheral ER junctions.

Figure EV3

Open in a new tab
  1. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing GFP‐TMD (green) and transiently expressing mScarlet‐TMEM55B (magenta) and SiR‐lysosome stained LE/Lys (orange). Zoom insets show region of interest containing ER (GFP‐TMD). First row represents ER (GFP‐TMD) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels. Scale bar: 10 µm, zoom insets 3 µm. Images related to Movie EV21. Images were acquired for 90 frames at 1Hz.
  2. MDA quantification of MitoTracker‐labeled mitochondria in cells overexpressing GFP‐RILP or GFP‐TMEM55B. The scatter plot represents n = 15 cells from 2 independent experiments. All images acquired for 90 frames at 1Hz.
  3. MDA quantification of peroxisomes in cells overexpressing GFP‐RILP or GFP‐TMEM55B. The scatter plots represent n = 15 cells from 2 independent experiments. All images acquired for 90 frames at 1Hz.
  4. MDA quantification of RAB5‐labeled early endosomes in cells overexpressing RFP‐RILP or RFP‐TMEM55B. The scatter plot represents n = 15 cells from 2 independent experiments. All images acquired for 90 frames at 1Hz.
  5. Representative time‐lapse spinning disk confocal images of n = 3 of live COS7 cells stably expressing GFP‐TMD (green) to mark the ER and SiR‐lysosome labeled LE/Lys (orange) merged under control or starvation conditions. Zoom insets highlight region of interest containing a mobile endosome as denoted by the white arrow. First row represents ER (GFP‐TMD) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels. Scale bar: 10 µm, zoom insets: 3 µm. Images related to Movies EV22 and EV23. Images were acquired for 90 frames at 1Hz.
  6. MDA quantification of average total ER of panel (E) normalized to 1 for control cells showing total average ER movement surrounding static endosomes or mobile endosomes. Graph represents mean ± s.d. of analyses at multiple locations within one cell of n = 15 cells per condition from 3 independent experiments. Normalized to 1 to control conditions (fed). All images acquired for 90 frames at 1Hz. Significance two‐tailed Student t‐test. ***P < 0.001,
  7. MDA quantification of SiR‐lysosome stained LE/Lys of panel (E) normalized to 1 for control cells showing total average endosomal movement. Graph represents n = 12 ‐ 15 cells per condition from 3 independent experiments. Normalized to 1 to control conditions (fed). Significance two‐tailed Student t‐test. *P < 0.05
  8. MDA quantification of total vesicle movement of COS7 under control or starved conditions overexpressing either GFP‐RAB5, GRP‐RAB7, GFP‐RAB11 or GFP‐RAB14. Graph represents mean ± s.d. of n = 14 ‐ 15 cells from 3 independent experiments. All images acquired for 90 frames at 1Hz. Normalized to control conditions (fed) counterparts set at 1. Significance two‐tailed Student t‐test. **P < 0.01 ****P < 0.001, ns = not significant
  9. MDA quantification of total local ER movement of COS7 under fed or starved conditions marking the ER and overexpressing either GFP‐RAB5, GRP‐RAB7, GFP‐RAB11 or GFP‐RAB14. Graph represents mean ± s.d. of analyses at multiple locations within one cell of n = 15 from 3 independent experiments. Normalized to control conditions set at 1 (fed). Significance two‐tailed Student t‐test. ****P < 0.0001, ns = not significant
  10. Representative stills of n = 3 of spinning disk confocal images of COS7 cells stably expressing TMD‐GFP (green) and SiR‐lysosome stained endosomes/lysosomes (orange). COS7 cells under control and starvation conditions. Zoom insets show regions of interest containing ER (TMD‐GFP) (green) and junction analysis of skeletonized ER where number of ER junctions representing the quantified average. Scale bars: 15 µm. zoom insets: 3 µm.
  11. Quantification of number of ER junctions per µm2 from (J) as resulted from ER junction analysis. Graph represents mean ± s.d. of analyses at multiple locations within one cell of n = 15 from 3 independent experiments. Significance two‐tailed Student t‐test. ****P < 0.0001.

Interestingly, TMEM55B connects late endosomal transport to the metabolic state of cells: Cells undergoing starvation upregulate TMEM55B expression to sequester the LE/Ly in the perinuclear region around the MTOC (Willett et al, 2017). This results in LE/Ly clustering therefore restricting their motility. We thus wondered whether starvation of cells also results in reduced ER movement similar to TMEM55B overexpression. Starved cells displayed a 22% reduction in endosomal mobility and a 33% reduction in overall ER dynamics (Fig EV3E–G, Movies EV30 and EV31). Next, we wondered whether the depletion of LE/Ly from the cell periphery as a result of starvation somehow causes non‐lysosomal endosomes to be more prevalent in ER hitchhiking. To this end, we examined the contribution of RAB5, RAB11, and RAB14, which previously appeared auxiliary to ER hitchhiking, to ER hitchhiking events under starvation conditions. Under starvation, the RAB7‐positive endosomes, LE/Ly, displayed the most significant reduction in total vesicular movement and ER hitchhiking (Fig EV3H–I, Movies EV32–EV39). Meanwhile, the RAB5‐, RAB11‐ or RAB14‐marked endosomes did not show any additional contribution to ER hitchhiking under starvation. If anything, starvation reduced the ability of each vesicle type to generate ER movement. Collectively, our data suggest that RAB7‐positive LE/Ly provide the majority of peripheral ER hitchhiking events.

Mobile late endosomes/lysosomes design the peripheral ER network architecture as modulated by starvation

Next, we tested whether the LE/Ly‐mediated ER movement had an effect on the overall architecture of the peripheral ER network. To address this, we analyzed the ER architecture by quantifying the number of ER junctions per µm2. This was accomplished by developing a semi‐automatic analysis method (Fig 4A). We observed that restricting LE/Ly mobility by overexpression of RILP or TMEM55B also decreased the number of peripheral ER junctions per µm2, resulting in a more open ER structure (Fig 4B and C). Similar to ER mobility, the complexity of the ER network architecture was also reduced upon starvation (Fig EV3J and K). Overall, these data suggest that LE/Ly movement and ER hitchhiking contribute significantly to the overall ER architecture and explains how starvation of cells translates in a “more open” peripheral ER network.

Figure 4. Endosomal clustering reduces peripheral ER junctions.

Figure 4

Open in a new tab
  1. Schematic flow overview of the process for ER junction quantification.
  2. Representative stills of spinning disk confocal images of n = 3 of COS7 cells stably expressing TMD‐GFP (green) and SiR‐lysosome‐stained endosomes/lysosomes (orange). COS7 cells in the middle panels overexpress mCherry‐RILP (magenta). The COS7 cells in the lower panels overexpress mCherry‐TMEM55B (magenta). Zoom insets show region of interest containing the ER (TMD‐GFP) (green) and junction analysis of the skeletonized ER where the number of ER junctions representing the quantified average. Scale bars: 15 µm. zoom insets: 3 µm.
  3. Quantification of the number of ER junctions per µm2 from panel (B) as resulted from ER junction analysis. Graph represents mean ± s.d. of analyses at multiple locations within one cell (n = 15 from 3 independent experiments). Significance two‐tailed Student t‐test. ***P < 0.001.

ER–late endosome/lysosome membrane contact site proteins facilitate ER hitchhiking

Our data suggest that LE/Ly contribute to ER mobility and ER network architecture. As previously reported, ER hitchhiking is defined as the ER tethered to another organelle propelled by a motor protein (Salogiannis & Reck‐Peterson, 2017; Guo et al, 2018; Mogre et al, 2020). Accordingly, disrupting MCS between the ER and LE/Ly could diminish ER hitchhiking events with LE/Ly and consequently reduce the complexity of the peripheral ER network. To this end, we focused on the ER‐resident VAP proteins, VAPA, VAPB, and MOSPD2. These proteins participate in the formation of MCS between the ER and other organelles, including LE/Ly (Murphy & Levine, 2016; Di Mattia et al, 2018). To test whether MCS proteins can be involved in ER hitchhiking, we depleted cells of these proteins (Fig EV4A–C). While the depletion of individual VAP proteins yielded a mild reduction in ER mobility, combined depletion of all three VAP proteins demonstrated the most significant decrease in ER mobility (Fig 5A, C and D, Fig EV4D–I, Movies EV40–EV47). This also corroborates a previous report where co‐depletion of all three tethering proteins leads to the most significant reduction in ER‐LE/Ly contact sites, most likely due to possible redundancy among VAPA, VAPB, and MOSPD2 that have an overlapping substrate specificity (Di Mattia et al, 2018). Silencing the VAP proteins did not affect total endosomal mobility or overall general ER morphology (Fig EV4J), suggesting that MCS between the ER and LE/Ly is required for ER hitchhiking with LE/Ly but not for LE/Ly trafficking (Figs 5D and Fig EV4K).

Figure EV4. ER tethers limit ER movement.

Figure EV4

Open in a new tab
  • A
    Graph represents mean ± s.d. validation of siRNA treated depletion efficiencies expressed as fraction of the mRNA of VAPA (siVAPA), VAPB (siVAPB) or MOSPD2 (siMOSPD2) relative to siRNA control (siControl) by qPCR, n = 3 independent experiments.
  • B
    Western blot analysis of COS7 cells transfected with siControl, siVAPA, siVAPB or siMOSPD2 and stained with rabbit anti‐human VAPA or VAPB. The loading control is stained for actin.
  • C
    Western blot analysis of COS7 cells transfected with a combination of siControl, siVAPA, siVAPB and siMOSPD2 and stained with rabbit anti‐human VAPA or VAPB. Loading control is stained for actin.
  • D–I
    Representative images of n = 3 of time‐lapse spinning disk confocal images of COS7 cells stably expressing GFP‐TMD (green) and SiR‐lysosome stained LE/Lys (orange). Cells shown were transfected with siRNA against VAPA (siVAPA) (D), VAPB (siVAPB) (E), MOSPD2 (siMOSPD2) (F), siVAPA and siVAPB (G), siVAPA and siMOSPD2 (H) or siVAPB and siMOSPD2 (I). Zoom insets show regions of interest containing mobile LE/Lys as denoted by the white arrow. First row represents ER (TMD‐GFP) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels and the fourth row shows a merge of ER (green) and SiR‐lysosome stained LE/Lys (orange). Scale bar: 5 µm, zoom insets: 3 µm. Images related to Movies EV24–EV29. Images were acquired for 90 frames at 1Hz.
  • J
    Still confocal images of n = 3 COS7 cells transfected with RFP‐SEC61B and either siControl or a combination of siVAPA, siVAPB and siMOSPD2. Zoom insets show regions of interest depicting ER morphology. Scale bars: 5 µm and zoom insets: 2 µm.
  • K
    Normalized MDA quantification of total ER movement of COS7 cells transfected with either siControl, siVAPA, siVAPB, siMOSPD2 or a combination thereof. Graph represents n = 15 cells from 3 independent experiments. Significance two‐tailed Student t‐test. *P < 0.05, **P < 0.01, ns = not significant.

Figure 5. ER tethers support ER hitchhiking.

Figure 5

Open in a new tab
  1. Representative images of time‐lapse spinning disk confocal images of n = 3 of COS7 cells stably expressing GFP‐TMD (green) and SiR‐lysosome‐stained LE/Lys (orange) treated with either siControl or co‐treated with siVAPA, siVAPB, and siMOSPD2. Zoom insets show regions of interest containing mobile LE/Lys as denoted by the white arrow. First row represents ER (TMD‐GFP) (green), the second row shows MDA‐generated ER movement (magenta), third row shows a merge of the two preceding channels, and the fourth row shows a merge of ER (green) and SiR‐lysosome stained LE/Lys (orange). Scale bar: 5 µm, zoom insets 3 µm. Images related to Movies EV40 and EV47. Images were acquired for 90 frames at 1Hz.
  2. Representative stills of spinning disk confocal images of n = 3 of COS7 cells stably expressing TMD‐GFP (green) and SiR‐lysosome‐stained endosomes/lysosomes (orange) treated with siControl or with siVAPA, siVAPB, and siMOSPD2. Zoom insets show region of interest containing ER (TMD‐GFP) (green) and junction analysis of skeletonized ER where number of ER junctions representing the quantified average. Scale bars: 10 µm. zoom insets: 3 µm.
  3. MDA quantification of ER of panels (A, B) and Fig EV4B–G normalized to control cells showing average ER movement surrounding mobile endosomes. Graph represents mean ± s.d. of analyses at multiple locations within the same cell (n = 30 cells per condition from 3 independent experiments). All images acquired for 90 frames at 1Hz. Significance two‐tailed Student t‐test. **P < 0.01, ****P < 0.0001, ns = not significant.
  4. MDA quantification of SiR‐lysosome‐stained LE/Lys of (A, B) and Fig EV4B–G normalized to 1 showing total average endosomal movement. Graph represents mean ± s.d. of n = 15 cells per condition from 3 independent experiments. Significance two‐tailed Student t‐test. ns = not significant.
  5. Quantification of number of ER junctions per µm2 from (B) and fig. H – M as resulted from ER junction analysis. Graph represents mean ± s.d. of analyses at multiple locations within one cell of n = 15 from 3 independent experiments. Significance two‐tailed Student t‐test. *P < 0.05, **P < 0.01 ***P < 0.001, ****P < 0.0001, ns = not significant.

Next, we examined whether the VAP‐mediated late endosomal ER hitchhiking events affect the peripheral ER network architecture. We observed that the depletion of individual VAP proteins reduces the number of peripheral ER junctions. However, triple depletion of VAP proteins showed the largest decrease (Figs 5B and E, and Fig EV5A–F), further substantiating the role of hitchhiking events in the formation of the peripheral ER architecture. Overall, our results suggest that the tether proteins facilitating ER‐LE/Ly contact sites contribute to LE/Ly‐mediated ER hitchhiking, which ultimately designs the peripheral ER network architecture.

Figure EV5. ER tethers limit peripheral ER junctions.

Figure EV5

Open in a new tab
  • A–F
    Representative stills of n = 3 of spinning disk confocal images of COS7 cells stably expressing TMD‐GFP (green) and SiR‐lysosome stained endosomes/lysosomes (orange). Cells shown were transfected with siRNA against VAPA (siVAPA) (A), VAPB (siVAPB) (B), MOSPD2 (siMOSPD2) (C), siVAPA and siVAPB (D), siVAPA and siMOSPD2 (E) or siVAPB and siMOSPD2 (F). Zoom insets show regions of interest containing ER (TMD‐GFP) (green) and junction analysis of skeletonized ER with number of ER junctions representing the quantified average. Scale bars: 15 µm. zoom insets: 3 µm.

Discussion

The ER is the most extensive membranous organelle spanning the entire cytoplasm and interacting with every other compartment to exchange information and metabolites. The ER is consisting of sub‐compartments such as the nuclear envelope, the perinuclear region, and the peripheral region, but the exact nature of these regions is still discussed. While the perinuclear ER is dense and difficult to distinguish by light microscopy, this is not the case for the more peripheral ER. The peripheral ER is composed of dynamic tubular structures. New tubules arise from the existing ER to build the complex interconnected ER network. This has been visualized by high‐end microscopy in a study that also visualized endosomes and peroxisomes leading the formation of new ER tubules (Guo et al, 2018). Here, we studied the contribution of the endosomal compartments in designing the architecture of the ER (Fig 6). We used image analysis methods to quantify this contribution and show that especially RAB7‐marked LE/Ly support the formation of new ER tubules through a mechanism called hitchhiking. LE/Ly then drag along associated ER tubules in their transport process, a process involving ER‐located VAP proteins. Reducing LE/Ly in the peripheral regions of cells by overexpression of RILP or TMEM55B or by starvation of cells directly affects this way of ER architecture formation. As the ER affects many LE/Ly processes such as transport and fusion, LE/Ly also contributes to the ER network design (Rocha et al, 2009; van der Kant et al, 2013).

Figure 6. Model for ER hitchhiking induced ER structuring.

Figure 6

Open in a new tab
  1. (1) A tethered LE/Lys initiates movement. (2) Upon movement, the tether remains intact and the ER is able to hitchhike along with the LE/Lys. (3) When the tethered pair encounter an ER tubule, junction formation may be initiated. (4) After formation the ER may remain tethered.
  2. ER hitchhiking is likely facilitated by a range of ER‐ LE/Lys tethers yet will require motor protein recruitment to drive hitchhiking.

Two independent mechanisms of microtubule‐mediated ER movement have been reported before: ER sliding and TAC‐mediated ER movement (Grigoriev et al, 2008; Bola & Allan, 2009; Friedman et al, 2010). Here, we followed a third mechanism, ER hitchhiking with LE/Ly. Our image analyses tool allowed us to quantify the relative contribution of each process in COS7 cells. This suggested that about half of peripheral ER movements occurs by ER hitchhiking, while TAC and ER sliding each contribute around 25%. While the numbers are based on perturbations that could be incomplete and variable under different conditions, they still define that at least three systems are contributing to the total ER tubule formation and architecture. The relative contribution of each process is likely dictated by the type of cell and physiological conditions. We observed that following an ER hitchhiking event, a protruding ER tubule dragged along by LE/Ly is capable of fusing with existing distal ER tubules to generate a novel ER junction. This suggests that the regulation of ER architecture is not an autonomous process build by LE/Ly movement. Accordingly, reducing late endosomal trafficking by RILP or TMEM55B overexpression or by starvation results in a significant reduction in LE/Ly transport in the cell periphery and a concomitantly reduced complexity of the ER architecture.

While we tested several endosomal compartments labeled by their characteristic RAB GTPases, we observed that ER movement was predominantly linked to mobile LE/Ly. This is in line with previous observations in which LE/Ly were found to be the dominant type of endosomes interacting with the ER (Friedman et al, 2013). Multiple protein complexes localize at MCS between the ER and LE/Ly: late endosomal ORP1L, STARD3, and STARD3NL each form molecular bridges with VAP proteins (Rocha et al, 2009; Di Mattia et al, 2018). The best characterized contacts are between the ER‐localized VAPA, VAPB, and MOSPD2 (Jordens et al, 2001; Rocha et al, 2009; Wijdeven et al, 2016; Zhao et al, 2018). We found that VAPA, VAPB and MOSPD contribute to the peripheral ER hitchhiking events, as depletion of these three proteins yielded a 33% reduction in ER mobility. The remainder could be accounted for by other non‐VAP‐mediated contact sites, such as RNF26 and TOLLIP or Protrudin and RAB7 (Jongsma et al, 2016; Raiborg et al, 2016).

ER hitchhiking places ER architecture under the control of LE/Ly trafficking, implying multiple physiological consequences. For instance, access to the calcium ion stores in the ER may become subject to late endosomal/lysosomal transport regulation (van der Kant & Neefjes, 2014). This could possibly take place, during protrusion formation, where LE/Ly are directed to the sites of membrane expansion to provide membrane materials. At the same time, the ER is required here for calcium supply and ER hitchhiking could help to couple both processes (Goyal & Blackstone, 2013; Raffaello et al, 2016). Also, any process affecting LE/Ly trafficking should affect the ER architecture. These include starvation and cholesterol accumulation.

Here, we have shown that LE/Ly movements, ER dynamics, and the ER network architecture are intimately linked cell biological processes. When LE/Ly are tethered to the ER and proceed to move, the ER can remain tethered and hitchhike along with the moving LE/Ly (Fig 6). When proximity criteria are met, the hitchhiking ER tubule may fuse to another ER tubule to form a novel ER junction. Therefore, ER dynamics and structure are partially controlled by late endosomal transport, a process that in itself is regulated by various physiological conditions. Overall, we define ER hitchhiking with LE/Lys as a significant contributor to peripheral ER mobility and architecture.

Materials and Methods

Cell culture and constructs

COS7 cells (CRL‐1651, ATCC, USA) were cultured in DMEM (Thermo Scientific, USA) supplemented with 10% FCS (Biowest, France). The cell line was regularly tested for mycoplasma contamination using MycoAlert™ Mycoplasma Detection Kit (Lonza, cat LT07‐318).

A stable cell line of COS7 cells was generated using antibiotic pressure containing the transmembrane region of MOSPD2 with GFP fused to the N‐terminus.

mScarlet ‐C1 and HA‐C1 plasmids were described previously (Jongsma et al, 2020). GFP‐ and Scarlet‐tagged MOSPD2 (IMAGE clone 4821861) transmembrane region was cloned into EGFP‐C1 and mScarlet‐C1 plasmids using 5’‐CCCAGAATTCCGCAGAGAATCACGCCCA‐3’ as forward primer and 5’‐CCCAGGATCCTTACTGCTGGAACCAGATAC‐3’ as reverse primers and BamHI and HindIII restriction sites. TMEM55B (IMAGE clone 3940519) using 5’‐ CCCATGTACAAGATGGCGGCGGCAGATGGA‐3’ as forward primer and 5’‐CCCACAATTGTCAGGAGAAGTTCTGGACAGGG‐3’ as reverse primer in EGFP‐C1, mScarlet ‐C1 and 2xHA‐C1 using BsrgI and MunI restriction sites. EGFP‐Peroxisomes‐0 and mCherry‐KDEL were kind gifts from Michael Davidson (Addgene plasmid #55041, 54501; Rizzo et al, 2009). GFP‐RILP was previously described and was cloned into mScarlet‐C1 vector using HindIII and EcoRI restriction sites (Jordens et al, 2001). EB3‐mGFP was a kind gift from Dr. Anna Akhmanova. GFP‐RAB5, GFP‐RAB7, GFP‐RAB11, and GFP‐RAB14 were produced as previously described (Kuijl et al, 2007; Kuijl et al, 2013; Sapmaz et al, 2019). mCherry‐SEC61B, mCherry‐CEPT1, and mCherry‐PIS were gifts from Gia Voeltz (Addgene plasmid # 119078, 119079, 49155) (Zurek et al, 2011; English & Voeltz, 2013). GFP‐ SEC61B was a gift from Christine Mayr (Addgene plasmid # 121159).(Ma & Mayr, 2018). pmScarlet_peroxisome_C1 was a gift from Dorus Gadella (Addgene plasmid # 85063 (Bindels et al, 2017).

Antibodies and reagents

Following antibodies were used in this study: Rabbit anti‐VAPA (Proteintech Cat#15275‐1‐AP), Rabbit anti‐VAPB (Proteintech Cat# 14477‐1‐AP), Rabbit anti‐KTN1 (Proteintech Cat#z19841‐1‐AP), Actin (Sigma‐Aldrich Cat# A5441), Anti‐Rabbit‐HRP (Thermo Fisher Cat#G‐21234), and Anti‐Mouse‐HRP (Thermo Fisher Cat#G‐21040). To inhibit microtubule (de)polymerization, tissue culture medium was supplemented with 2 µM of paclitaxel for two hours (Merck, USA) in DMSO. The late endosomal/lysosomal compartment was visualized using SiR‐lysosome at a concentration of 1 µM (Stein am Rhein, Switzerland). To ensure adequate retention of SiR‐lysosome, cells were treated with 10 µM Verapamil as per manufacturer’s recommendation. Mitochondria were labeled using Mitotracker Deep Red FM (Thermo Fischer, USA) at a concentration of 500 nM, as per manufacturers recommendation.

Transfection

COS7 cells were transfected using Effectene (Qiagen, Germany) according to the manufacturer’s instructions. For siRNA‐mediated silencing, cells were reverse transfected with DharmaFECT transfection reagent #1 (Dharmacon, USA) and 50 nM siRNA and samples were analyzed 72 hours later by confocal microscopy or qPCR. siRNAs were purchased from Dharmacon (USA) and were as follows: siVAPA (MQ‐021382‐01‐0010), siVAPB (MQ‐017795‐0010), siMOSPD2 (MQ‐017039‐010), and siKinectin (MQ‐010605‐01‐0005).

cDNA synthesis and qPCR

RNA isolation, complementary DNA synthesis, and quantitative PCR with reverse transcription were performed according to the manufacturer’s instructions (Roche, Switzerland). Signal was normalized to GAPDH and calculated using the pfaffl formula. Primers used for detection of GAPDH, VAPA, VAPB, and MOSPD2 were as follows: GAPDH forward: 5’‐TGTTGCCATCAATGACCCCTT −3’ and reverse 5’‐CTCCACGACGTACTCAGCG −3’. VAPA forward 5’ – AATGCTACAGCCCTTTGACTATG – 3’ and reverse 5’ – TCATCAGGTTTTGCCTCTTTCC −3’. VAPB forward 5’ – AGATGGACTGCGGATGAGGAA −3’ and reverse 5’ – CAGTTGGGGCTAGCGCTGAAA −3’ and MOSPD2 forward 5’ – ATGCTCGATGAGAGTTTTCAGTG −3’ and reverse 5’‐CCAACCAGAATGCTATGAGCTTC‐3’. Kinectin‐1 forward 5’ – AAATGTCTTCGTAGACGAACCCC −3’ and reverse 5’ – TTTGCAGTTTCAGTCTTCAGTT −3’.

Microscopy

Cells were seeded in 35 mm glass bottom tissue culture dishes (MatTek, USA) or in 35 mm glass bottom tissue culture dish with 4 compartments (Greiner, Austria). Image acquisition was performed by an Andor Dragonfly 505 spinning disk confocal on a Leica DMi8 microscope with an Andor iXon Life 888 EMCCD camera and Andor Zyla 4.2 + sCMOS camera using a 565 nm long pass dichroic image splitter. (Oxford, UK). The microscopy set‐up allowed for the simultaneous acquisition of two channels. For live cell imaging, the microscopes were equipped with a humidified climate control system at 37˚C supplemented with 5% CO2. Specific frame rates are noted at the relevant experiments.

For fixed cell imaging, cells were seeded on glass coverslips and transfected. The day after transfections, cells were washed with PBS and fixed in 3.7% (w/v) formaldehyde in PBS solution for 10min at RT. Next, cells were washed and mounted. Samples were imaged using a Leica TCS SP8 microscope (Leica Microsystems, Wetzlar, Germany) equipped with 405/488/561/647 nm laser lines. Images were analyzed using ImageJ Software.

Membrane displacement analysis

To quantify membrane displacements of heterogeneously shaped membranes, we developed a membrane displacement analysis (MDA) method. This analysis method is for a large part based on optical flow analysis, which identifies corresponding areas between two frames and generates a vector flow map(Horne, 1981). This type of map describes frame‐to‐frame movement of local image areas. The vector flow map is visualized by color‐coding the image frame by the magnitude of each flow vector (Fig 1B, panel II). In a parallel step, also (noisy) background pixels are matched from frame to frame. To mitigate these erroneous flow vectors, the flow vectors are based on the labeled membranes (Fig 1B, panel III). The masked vector flow maps reflect the end‐point degree of displacement of labeled membranes between two images (Fig 1B, panel IV). By overlaying this with the original image acquisition, sequences are generated where the extent of movement of a membrane acquired was reflected by color intensity. The higher the color intensity, the further that pixel will have moved in relation to the previous image (Fig 1B, panel V). These data are then quantified in terms of fractions of membrane mobility. Using this quantification, the fraction of membrane movement can be determined (Fig 1B, panel VI). MDA has implemented the described membrane displacement analysis (MDA) as an ImageJ/Fiji plugin. Core functionality relies on the Gaussian Window MSE plugin by Stephan Saalfeld and Pavel Tomancak. Masking was performed using an automatic threshold of the image intensities after median filtering. The MDA plugin is available upon request with m.spits@lumc.nl or l.m.voortman@lumc.nl.

ER junction analysis

ER junction analysis was performed by generating a mask of the ER using the auto local thresholding algorithm Phansalkar (Fig 4A, panel I). Next, a skeleton in generated from the ER mask (Fig 4A, panel II). This skeleton representation of the ER is then analyzed for the number of junctions (fig 4A, panel III) (Arganda‐Carreras et al, 2010). The number of junctions is then divided by the analyzed surface area in µm2. The ER junction analysis plugin for ImageJ is available upon request with m.spits@lumc.nl or l.m.voortman@lumc.nl

Western blotting

siRNA‐transfected COS7 cells were lysed in ice‐cold lysis buffer (0.5% (v/v) NP‐40, 50 mM Tris–HCl 8.0, 150 mM NaCl, 0.1% SDS and complete™ Protease Inhibitor Cocktail) and lysates were mixed with 2xSDS sample buffer (20% glycerol, 2% SDS, 125 mM Tris–HCl pH 6.8 and 0.002%(w/v) Bromophenol Blue). Samples were incubated at 95ºC for 10 min. Proteins were separated by SDS–PAGE and transferred to nitrocellulose membranes. Membranes were blocked overnight with 5% (w/v) milk in PBST (PBS containing 0.05% (v/v) Tween 20), incubated with primary antibody solution in PBST, washed thrice with PBST, incubated with secondary antibody solution in PBST, washed thrice with PBST and twice with PBS before imaging. For detection, membranes were incubated with the substrate and imaged using Amersham Imager 600 (GE Healthcare).

Statistics

All error bars correspond to standard deviation. Statistical evaluations report on Student’s t‐test as described in corresponding figure legends. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns = not significant.

Author contributions

MS designed the study with input from JN. MS performed and analyzed nearly all experiments with help from ITH. BC and JJA provided technical support. MS and LMV designed the macro to analyze ER movement. MS designed the macro to analyze ER junctions. RHW performed and analyzed all qPCR experiments. MS wrote the manuscript with input from JN and BC. JN and MS coordinated the study.

Conflict of interest

The authors declare that they have no conflict of interest.

Supporting information

Expanded View Figures PDF

Click here for additional data file. (6.3MB, pdf)

MoviesEV1–EV47

Click here for additional data file. (38.2MB, zip)

Review Process File

Click here for additional data file. (889.7KB, pdf)

Acknowledgements

We would like to thank the microscopy facility of the LUMC for their support and members of the Neefjes group for critical discussions. This work was supported by the Dutch Research Council (NWO) Building Blocks of Life (737016002) and the European Research Council (ERC) Advanced Grant ERCOPE (ERC‐2015‐AdV) awarded to J.N.

EMBO reports (2021) 22: e50815

Contributor Information

Birol Cabukusta, Email: b.cabukusta@lumc.nl.

Jacques Neefjes, Email: j.j.c.neefjes@lumc.nl.

Data availability

This study includes no data deposited in external repositories.

References

  1. Arganda‐Carreras I, Fernandez‐Gonzalez R, Munoz‐Barrutia A, Ortiz‐De‐Solorzano C (2010) 3D reconstruction of histological sections: Application to mammary gland tissue. Microsc Res Tech 73: 1019–1029 [DOI] [PubMed] [Google Scholar]
  2. Bakker J, Spits M, Neefjes J, Berlin I (2017) The EGFR odyssey ‐ from activation to destruction in space and time. J Cell Sci 130: 4087–4096 [DOI] [PubMed] [Google Scholar]
  3. Bhaskara RM, Grumati P, Garcia‐Pardo J, Kalayil S, Covarrubias‐Pinto A, Chen W, Kudryashev M, Dikic I, Hummer G (2019) Curvature induction and membrane remodeling by FAM134B reticulon homology domain assist selective ER‐phagy. Nat Commun 10: 2370 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bindels DS, Haarbosch L, van Weeren L, Postma M, Wiese KE, Mastop M, Aumonier S, Gotthard G, Royant A, Hink MA et al (2017) mScarlet: a bright monomeric red fluorescent protein for cellular imaging. Nat Methods 14: 53–56 [DOI] [PubMed] [Google Scholar]
  5. Bola B, Allan V (2009) How and why does the endoplasmic reticulum move? Biochem Soc Trans 37: 961–965 [DOI] [PubMed] [Google Scholar]
  6. Cabukusta B, Neefjes J (2018) Mechanisms of lysosomal positioning and movement. Traffic 19: 761–769 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen S, Desai T, McNew JA, Gerard P, Novick PJ, Ferro‐Novick S (2015) Lunapark stabilizes nascent three‐way junctions in the endoplasmic reticulum. Proc Natl Acad Sci U S A 112: 418–423 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Di Mattia T, Wilhelm LP, Ikhlef S, Wendling C, Spehner D, Nomine Y, Giordano F, Mathelin C, Drin G, Tomasetto C et al (2018) Identification of MOSPD2, a novel scaffold for endoplasmic reticulum membrane contact sites. EMBO Rep 19: e45453 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Elliott T, Neefjes J (2006) The complex route to MHC class I‐peptide complexes. Cell 127: 249–251 [DOI] [PubMed] [Google Scholar]
  10. English AR, Voeltz GK (2013) Rab10 GTPase regulates ER dynamics and morphology. Nat Cell Biol 15: 169–178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. English AR, Zurek N, Voeltz GK (2009) Peripheral ER structure and function. Curr Opin Cell Biol 21: 596–602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Friedman JR, Dibenedetto JR, West M, Rowland AA, Voeltz GK (2013) Endoplasmic reticulum‐endosome contact increases as endosomes traffic and mature. Mol Biol Cell 24: 1030–1040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Friedman JR, Webster BM, Mastronarde DN, Verhey KJ, Voeltz GK (2010) ER sliding dynamics and ER‐mitochondrial contacts occur on acetylated microtubules. J Cell Biol 190: 363–375 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Goyal U, Blackstone C (2013) Untangling the web: mechanisms underlying ER network formation. Biochim Biophys Acta 1833: 2492–2498 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grigoriev I, Gouveia SM, van der Vaart B, Demmers J, Smyth JT, Honnappa S, Splinter D, Steinmetz MO, Putney JW Jr, Hoogenraad CC et al (2008) STIM1 is a MT‐plus‐end‐tracking protein involved in remodeling of the ER. Curr Biol 18: 177–182 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Guimaraes SC, Schuster M, Bielska E, Dagdas G, Kilaru S, Meadows BR, Schrader M, Steinberg G (2015) Peroxisomes, lipid droplets, and endoplasmic reticulum "hitchhike" on motile early endosomes. J Cell Biol 211: 945–954 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Guo Y, Li D, Zhang S, Yang Y, Liu JJ, Wang X, Liu C, Milkie DE, Moore RP, Tulu US et al (2018) Visualizing Intracellular Organelle and Cytoskeletal Interactions at Nanoscale Resolution on Millisecond Timescales. Cell 175: 1430–1442 [DOI] [PubMed] [Google Scholar]
  18. Horne DF (1981) Optical scales, reticles, gratings, masks, and standards. Appl Opt 20: 4000–4008 [DOI] [PubMed] [Google Scholar]
  19. Horwitz SB (1994) Taxol (paclitaxel): mechanisms of action. Ann Oncol 5(Suppl 6): S3–S6 [PubMed] [Google Scholar]
  20. Hoyer MJ, Chitwood PJ, Ebmeier CC, Striepen JF, Qi RZ, Old WM, Voeltz GK (2018) A Novel Class of ER Membrane Proteins Regulates ER‐Associated Endosome Fission. Cell 175: 254–265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jongsma ML, Bakker J, Cabukusta B, Liv N, van Elsland D, Fermie J, Akkermans JL, Kuijl C, van der Zanden SY, Janssen L et al (2020) SKIP‐HOPS recruits TBC1D15 for a Rab7‐to‐Arl8b identity switch to control late endosome transport. EMBO J 39: e102301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jongsma ML, Berlin I, Wijdeven RH, Janssen L, Janssen GM, Garstka MA, Janssen H, Mensink M, van Veelen PA, Spaapen RM et al (2016) An ER‐Associated Pathway Defines Endosomal Architecture for Controlled Cargo Transport. Cell 166: 152–166 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jordens I, Fernandez‐Borja M, Marsman M, Dusseljee S, Janssen L, Calafat J, Janssen H, Wubbolts R, Neefjes J (2001) The Rab7 effector protein RILP controls lysosomal transport by inducing the recruitment of dynein‐dynactin motors. Curr Biol 11: 1680–1685 [DOI] [PubMed] [Google Scholar]
  24. Junutula JR, De Maziere AM, Peden AA, Ervin KE, Advani RJ, van Dijk SM, Klumperman J, Scheller RH (2004) Rab14 is involved in membrane trafficking between the Golgi complex and endosomes. Mol Biol Cell 15: 2218–2229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. van der Kant R, Fish A, Janssen L, Janssen H, Krom S, Ho N, Brummelkamp T, Carette J, Rocha N, Neefjes J (2013) Late endosomal transport and tethering are coupled processes controlled by RILP and the cholesterol sensor ORP1L. J Cell Sci 126: 3462–3474 [DOI] [PubMed] [Google Scholar]
  26. van der Kant R, Neefjes J (2014) Small regulators, major consequences ‐ Ca(2)(+) and cholesterol at the endosome‐ER interface. J Cell Sci 127: 929–938 [DOI] [PubMed] [Google Scholar]
  27. Kuijl C, Pilli M, Alahari SK, Janssen H, Khoo PS, Ervin KE, Calero M, Jonnalagadda S, Scheller RH, Neefjes J et al (2013) Rac and Rab GTPases dual effector Nischarin regulates vesicle maturation to facilitate survival of intracellular bacteria. EMBO J 32: 713–727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kuijl C, Savage ND, Marsman M, Tuin AW, Janssen L, Egan DA, Ketema M, van den Nieuwendijk R, van den Eeden SJ, Geluk A et al (2007) Intracellular bacterial growth is controlled by a kinase network around PKB/AKT1. Nature 450: 725–730 [DOI] [PubMed] [Google Scholar]
  29. Langemeyer L, Frohlich F, Ungermann C (2018) Rab GTPase Function in Endosome and Lysosome Biogenesis. Trends Cell Biol 28: 957–970 [DOI] [PubMed] [Google Scholar]
  30. Lin S, Sun S, Hu J (2012) Molecular basis for sculpting the endoplasmic reticulum membrane. Int J Biochem Cell Biol 44: 1436–1443 [DOI] [PubMed] [Google Scholar]
  31. Ma W, Mayr C (2018) A Membraneless Organelle Associated with the Endoplasmic Reticulum Enables 3'UTR‐Mediated Protein‐Protein Interactions. Cell 175: 1492–1506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mogre SS, Christensen JR, Niman CS, Reck‐Peterson SL, Koslover EF (2020) Hitching a ride: mechanics of transport initiation through linker‐mediated hitchhiking. Biophys J 118: 1357–1369 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Murphy SE, Levine TP (2016) VAP, a Versatile Access Point for the Endoplasmic Reticulum: Review and analysis of FFAT‐like motifs in the VAPome. Biochim Biophys Acta 1861: 952–961 [DOI] [PubMed] [Google Scholar]
  34. Raffaello A, Mammucari C, Gherardi G, Rizzuto R (2016) Calcium at the Center of Cell Signaling: Interplay between Endoplasmic Reticulum, Mitochondria, and Lysosomes. Trends Biochem Sci 41: 1035–1049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Raiborg C, Wenzel EM, Pedersen NM, Stenmark H (2016) ER‐endosome contact sites in endosome positioning and protrusion outgrowth. Biochem Soc Trans 44: 441–446 [DOI] [PubMed] [Google Scholar]
  36. Rizzo MA, Davidson MW, Piston DW (2009) Fluorescent protein tracking and detection: applications using fluorescent proteins in living cells. Cold Spring Harb Protoc 2009: pdb.top64. [DOI] [PubMed] [Google Scholar]
  37. Rocha N, Kuijl C, van der Kant R, Janssen L, Houben D, Janssen H, Zwart W, Neefjes J (2009) Cholesterol sensor ORP1L contacts the ER protein VAP to control Rab7‐RILP‐p150 Glued and late endosome positioning. J Cell Biol 185: 1209–1225 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rowland AA, Chitwood PJ, Phillips MJ, Voeltz GK (2014) ER contact sites define the position and timing of endosome fission. Cell 159: 1027–1041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Salogiannis J, Egan MJ, Reck‐Peterson SL (2016) Peroxisomes move by hitchhiking on early endosomes using the novel linker protein PxdA. J Cell Biol 212: 289–296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Salogiannis J, Reck‐Peterson SL (2017) Hitchhiking: A Non‐Canonical Mode of Microtubule‐Based Transport. Trends Cell Biol 27: 141–150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Santama N, Er CP, Ong LL, Yu H (2004) Distribution and functions of kinectin isoforms. J Cell Sci 117: 4537–4549 [DOI] [PubMed] [Google Scholar]
  42. Sapmaz A, Berlin I, Bos E, Wijdeven RH, Janssen H, Konietzny R, Akkermans JJ, Erson‐Bensan AE, Koning RI, Kessler BM et al (2019) USP32 regulates late endosomal transport and recycling through deubiquitylation of Rab7. Nat Commun 10: 1454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ullrich O, Reinsch S, Urbe S, Zerial M, Parton RG (1996) Rab11 regulates recycling through the pericentriolar recycling endosome. J Cell Biol 135: 913–924 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wang S, Tukachinsky H, Romano FB, Rapoport TA (2016) Cooperation of the ER‐shaping proteins atlastin, lunapark, and reticulons to generate a tubular membrane network. Elife 5: e18605 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wijdeven RH, Janssen H, Nahidiazar L, Janssen L, Jalink K, Berlin I, Neefjes J (2016) Cholesterol and ORP1L‐mediated ER contact sites control autophagosome transport and fusion with the endocytic pathway. Nat Commun 7: 11808 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Willett R, Martina JA, Zewe JP, Wills R, Hammond GRV, Puertollano R (2017) TFEB regulates lysosomal positioning by modulating TMEM55B expression and JIP4 recruitment to lysosomes. Nat Commun 8: 1580 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. van de Willige D, Hoogenraad CC, Akhmanova A (2016) Microtubule plus‐end tracking proteins in neuronal development. Cell Mol Life Sci 73: 2053–2077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Zajac AL, Goldman YE, Holzbaur EL, Ostap EM (2013) Local cytoskeletal and organelle interactions impact molecular‐motor‐ driven early endosomal trafficking. Curr Biol 23: 1173–1180 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zhao YG, Liu N, Miao G, Chen Y, Zhao H, Zhang H (2018) The ER Contact Proteins VAPA/B Interact with Multiple Autophagy Proteins to Modulate Autophagosome Biogenesis. Curr Biol 28(1234–1245): e1234 [DOI] [PubMed] [Google Scholar]
  50. Zhou X, He Y, Huang X, Guo Y, Li D, Hu J (2019) Reciprocal regulation between lunapark and atlastin facilitates ER three‐way junction formation. Protein Cell 10: 510–525 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zulkefli KL, Houghton FJ, Gosavi P, Gleeson PA (2019) A role for Rab11 in the homeostasis of the endosome‐lysosomal pathway. Exp Cell Res 380: 55–68 [DOI] [PubMed] [Google Scholar]
  52. Zurek N, Sparks L, Voeltz G (2011) Reticulon short hairpin transmembrane domains are used to shape ER tubules. Traffic 12: 28–41 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Expanded View Figures PDF

Click here for additional data file. (6.3MB, pdf)

MoviesEV1–EV47

Click here for additional data file. (38.2MB, zip)

Review Process File

Click here for additional data file. (889.7KB, pdf)

Data Availability Statement

This study includes no data deposited in external repositories.

Articles from EMBO Reports are provided here courtesy of Nature Publishing Group

RESOURCES