Examination of actin and microtubule dependent APC localisations in living mammalian cells
© Langford et al; licensee BioMed Central Ltd. 2006
Received: 10 August 2005
Accepted: 19 January 2006
Published: 19 January 2006
Skip to main content
© Langford et al; licensee BioMed Central Ltd. 2006
Received: 10 August 2005
Accepted: 19 January 2006
Published: 19 January 2006
The trafficking of the adenomatous polyposis coli (APC) tumour suppressor protein in mammalian cells is a perennially controversial topic. Immunostaining evidence for an actin-associated APC localisation at intercellular junctions has been previously presented, though live imaging of mammalian junctional APC has not been documented.
Using live imaging of transfected COS-7 cells we observed intercellular junction-associated pools of GFP-APC in addition to previously documented microtubule-associated GFP-APC and a variety of minor localisations. Although both microtubule and junction-associated populations could co-exist within individual cells, they differed in their subcellular location, dynamic behaviour and sensitivity to cytoskeletal poisons. GFP-APC deletion mutant analysis indicated that a protein truncated immediately after the APC armadillo repeat domain retained the ability to localise to adhesive membranes in transfected cells. Supporting this, we also observed junctional APC immunostaining in cultures of human colorectal cancer cell line that express truncated forms of APC.
Our data indicate that APC can be found in two spatially separate populations at the cell periphery and these populations can co-exist in the same cell. The first localisation is highly dynamic and associated with microtubules near free edges and in cell vertices, while the second is comparatively static and is closely associated with actin at sites of cell-cell contact. Our imaging confirms that human GFP-APC possesses many of the localisations and behaviours previously seen by live imaging of Xenopus GFP-APC. However, we report the novel finding that GFP-APC puncta can remain associated with the ends of shrinking microtubules. Deletion analysis indicated that the N-terminal region of the APC protein mediated its junctional localisation, consistent with our observation that truncated APC proteins in colon cancer cell lines are still capable of localising to the cell cortex. This may have implications for the development of colorectal cancer.
The APC protein plays a pivotal role in WNT signal transduction, has been suggested to have important functions in cell migration and mitosis, and APC mutation is a crucial early event in the development of most colorectal cancers . The intracellular localization of APC has long been the subject of close scrutiny, with a number of distributions having been described in a variety of experimental systems (for a recent review see ). Two mammalian APC distributions have previously been described as populations found at peripheral cellular sites. The first of these to be identified and widely accepted consists of APC clusters that localize to specific cortical sites in a microtubule-dependent manner [3–5]. Support for this distribution has been presented in studies examining the behavior of Xenopus APC-GFP fusion proteins in living cells . In addition, an analogous localization to the plus-ends of microtubules close to the basolateral surface of highly polarized epithelial cells has been shown . Evidence for a second peripheral pool of APC in the form of an actin-dependent localization to membranes involved in cell-cell adhesion has been found by immunostaining studies [3, 5] although the validity of this has been questioned . Nevertheless, evidence that potentially supports a functional role for APC at cell junctions has recently been presented. In human cells, restoration of expression of full-length APC in a colorectal cancer cell line has been shown to promote cell-cell adhesion  while in Drosophila the APC homologue E-APC has been shown to associate with and play a role in maintaining the integrity of epithelial cell junctions, a localization mediated by its armadillo repeats .
To date the great majority of information about the distribution of full-length mammalian APC has been based upon immunostaining studies in fixed cells. Since recent work has raised questions about APC antibody specificity , confirmation of these different APC distributions by other means is clearly desirable. While some data on GFP-APC expression in COS-7 cells has been previously published , this work concentrated on a purely microtubule-associated GFP-APC localization in subconfluent populations of cells and the GFP-APC construct used in the study lacked the N-terminal region of the protein. At present the dynamic behavior of a junctional APC population in live mammalian cells, if such a population exists, has not been addressed.
In this study we investigated the intracellular distribution of APC using live imaging of human GFP-APC fusion proteins in mammalian cells. Our study confirms previous work showing that GFP-APC can be localized within the cell in a microtubule-dependent and highly dynamic manner. However, we show for the first time that human GFP-APC can remain associated with shrinking as well as growing microtubule tips. We also confirm that human GFP-APC is localized to sites of cell-cell adhesion in an actin-dependent way in confluent populations of cells. To our knowledge this is the first time that this actin associated GFP-APC population has been imaged in living mammalian cells.
However, we noted in COS-7 cells that peripheral GFP-APC puncta on microtubule ends frequently underwent retrograde linear movements with average velocities of 21.6 ± 2.4 μm/min (Figure 4, panels I-L (arrowhead); additional file 3). Typically this movement occurred over distances of less than 10 μM. In some cases portions of an original puncta were deposited within the cytoplasm while the remainder continued retrograde movement on the microtubule end (Figure 4, panels M-P arrowheads; additional file 4, examples also apparent in additional file 3). We also noted that the depolymerising microtubules tipped by GFP-APC puncta could undergo pause and then re-growth with the puncta of GFP-APC still attached (Figure 4, panels Q-T arrowhead; additional file 5). Interestingly, close examination of sequences such as additional file 5 indicates that separate GFP-APC puncta decorating the distal segment of a microtubule are swept up into a single tip-associated structure as the microtubule shrinks but re-separate into a string of beads-like distribution during microtubule regrowth, before finally beginning to re-coalesce into a single structure at the tip of paused microtubules near the cell membrane. The behaviour of GFP-APC at microtubule distal tips is clearly therefore very complex. Retrograde trafficking of GFP-APC was not documented in previous studies [6, 11]. Furthermore, these movements seem likely to represent puncta associated with shrinking microtubule tips and not with growing tips looping back from the cell edge or retrograde transport of GFP-APC puncta along microtubules since the microtubule distal segment could clearly be identified by GFP-APC labelling in many cases. We therefore conclude that mammalian APC can remain associated with shrinking microtubule tips at the cell periphery, a novel observation for this protein.
The junctional GFP-APC localisation was dominant in cells having extensive contacts with neighbouring cells, particularly those in densely confluent regions of an imaging dish. However, in regions of lower cell density we observed that both the microtubule and cell junction-associated GFP-APC distributions could be found in different regions of the same cell (Figure 6, panel H; additional file 9). The junctional population seen in these cells was unlikely to represent an overexpression artefact since it was present at low GFP-APC fluorescence intensities and no phenomena indicative of APC overexpression were seen (for example, GFP-APC decorated microtubule bundles). Observation of cells with both junctional- and microtubule-associated GFP-APC confirmed the very different dynamic behaviours of these protein populations. The microtubule-associated GFP-APC localisation was restricted to free cell edges and cell vertices and was highly motile whereas the junctional localisation was only present at sites of cell-cell contact and was essentially immobile. Dynamic GFP-APC microtubule-associated clusters were not seen in the vicinity of cell junctions, although when present the EB1-like localisation to growing microtubule distal tips was. We therefore suggest that the two GFP-APC populations might reflect the localised regulation of APC interactions in specific cellular regions.
In this study we have confirmed that APC can be found at sites of cell-cell contact in mammalian cells by live imaging of GFP-APC fusion proteins. Recently, other workers have presented an examination of GFP-APC dynamic behaviour in transfected COS-7 cells  but did not report findings similar to those presented here. The reasons for this are unclear. Although the GFP-APC construct used in the previous study contained a short N-terminal deletion our data indicates that this would not have precluded the observation of a junctional APC population (Figure 9, panel B). However, we note that the authors of this previous study were primarily focused upon defining the behaviour of GFP-APC decorated microtubules at peripheral sites rather than examining the possibility of a GFP-APC pool at cell junctions. They may not therefore have examined GFP-APC expressing cells in confluent cell cultures. Similarly, we are unable to conclusively explain why previous attempts at mapping the minimal region required for the GFP-APC junctional localisation were unsuccessful and led to the hypothesis that only full-length APC could achieve this distribution . We suspect, however, that difficulties similar to those we experienced in attempting to image GFP-APC in cell types other than COS-7 were a major contributory factor to this.
In addition to the microtubule-associated GFP-APC localisations found by other workers we identified an association of GFP-APC puncta with shrinking microtubule tips in transfected COS-7 cells (Figure 4, panels I-L and panels M-P). APC is known to promote microtubule stability and assembly both in vitro and in vivo [20–22] and as such we might have expected it to be found only on polymerising microtubule tips. The presence of GFP-APC puncta on depolymerising tips would suggest that additional proteins involved in microtubule dynamics might regulate the effects of APC on microtubule plus-end dynamics. For example, the presence of a destabilizing factor on a microtubule tip, such as the kin I kinesin MCAK [23, 24], might override the normal stabilising ability of APC. However, if the destabilising factor is inactivated on the microtubule, any tip-associated APC could then re-promote microtubule stability and growth. That Xenopus APC has recently been shown to associate with MCAK raises the interesting possibility that APC might form part of a microtubule plus-end complex responsible for the general control of microtubule behaviour in specific regions of the cell periphery .
The mechanism by which APC associates with actin at cell junctions remains unclear. An interaction between APC and the mammalian homologue of the Drosophila discs large protein has been proposed to be responsible for the localization of APC to neuronal synapses , structures that can be regarded as a specialised form of cadherin-based cell-cell adhesion. However, we found that this interaction was not essential for localising APC to sites of cell-cell contact in COS-7 cells although an interaction between APC and DLG at the cortex once both proteins have been recruited there cannot be ruled out. Instead, our data indicated that a truncated APC containing only the heptad and armadillo repeat domains of the protein was capable of localizing to junctions.
We therefore propose that the N-terminal region of the APC protein mediates its localization to the cortex. We speculate that the armadillo repeat regions of APC mediate this localisation since we also noted that a larger construct lacking a number of the heptad repeats remained capable of localising to adhesive membranes (Figure 9, panel B). However, further work needs to be carried out to directly confirm this. This may be complicated by the observation that a number of potential binding partners exist for the N-terminal region of APC, particularly the armadillo repeat domain. These include the kinesin-associated protein KAP3, the Rac effector Asef and the Cdc42 activator IQGAP1 [13, 26, 27]. Any of these could plausibly mediate an APC junctional localisation, as could an interaction with an as yet unidentified binding partner for this region of APC. Notably, the Drosophila E-APC protein has also been shown to localise to sites of cell-cell adhesion via its armadillo repeats . Mislocalization of E-APC leads to the disruption or impairment of intercellular adhesion, implicating E-APC in the regulation of cell contacts in Drosophila. In the light of our study and recent observations by other investigators  it seems possible that APC might play a similar role in mammalian epithelial tissues.
Our observations of cells where both junctional and microtubule-associated GFP-APC populations were present indicated that these different pools did not spatially overlap. The question of how this might be achieved has parallels with a long-standing problem in APC biology: how is APC targeted to the tips of a subset of microtubule ends at specific sites at the cell periphery? Recent work from other investigators has indicated that an APC localisation to microtubule ends near free cell edges arises from a Cdc42-initiated signalling cascade that results in a local inhibition of GSK3β activity . Phosphorylation by GSK3β inhibits the microtubule-binding ability of APC . Therefore, local inhibition of the activity of this enzyme promotes the association of APC with microtubules in specific cellular regions where Cdc42 is active. Implicit in this model is the assumption that GSK3β activity elsewhere in the cell normally suppresses the association of APC with microtubules. If this model is combined with the observation that APC can localise to cell junctions when they are available then a plausible mechanism for generating the compartmentalisation seen in our study can be formulated.
In this study we show that APC is capable to localising to both microtubules and to junctions within the cell, depending on cellular context. The junctional localisation is likely to be mediated by the N-terminal region of the APC protein. Consistent with this we find that colon cancer cell lines expressing truncated APC proteins are also capable of localising to the cortex in cells having the necessary cell-cell contacts. As well as losing function within the WNT signalling pathway, it seems possible that truncated APC proteins might act as dominant-negative mutants in cells that retain a normal copy of the APC protein, with the mutant copy interfering with the action of full-length APC molecules at intercellular junctions. This may have direct implications for the development of colorectal cancer.
COS-7 cells were cultured as described previously . Caco-2 and SW480 cells were obtained from the Cancer Research UK Cell Line Service at the London Research Institute, Lincoln's Inn Fields, UK and cultured according to the instructions supplied. Incubations with cytoskeletal poisons were performed as described previously . Nocodazole and Cytochalasin D were obtained from Sigma.
The rabbit polyclonal anti M-APC antibody  was a kind gift from Dr Inke Näthke, (University of Dundee), and was used at a 1/500 dilution for immunostaining. The monoclonal APC antibody ALI 12-28 used in this study was obtained from Cancer Research UK Antibody Service at the London Research Institute and was used at a 1/500 dilution for immunostaining. It has also been made commercially available by Abcam and this antibody was used for Westerns at 1/5000 dilution. Rabbit polyclonal and mouse monoclonal anti-GFP antibodies were obtained from Clontech and used at a 1/1000 dilution for immunostaining and a 1/2000 dilution for Western Blotting. Rabbit polyclonal α-catenin, mouse monoclonal β-catenin, rabbit polyclonal pan-cadherin and mouse monoclonal β-actin antibodies were all obtained from Sigma. All secondary antibodies were Alexa 488 and 594 conjugates obtained from Molecular Probes, as was Alexa 594 conjugated phalloidin.
For Western blotting, cells were resuspended in RIPA buffer (50 mM Tris pH7.5, 150 mM NaCl, 1% (v/v) Igepal, 0.5% (v/v) Sodium Deoxycholate, 1 mM EDTA, 0.1% (v/v) SDS) buffer containing EDTA free complete protease inhibitors (Roche) and phosphastase inhibitors cocktail II (Sigma). Protein concentrations for the cell extracts was determined using a protein assay kit (Pierce). NuPage™ Loading Buffer (Invitrogen) was then added to 10–20 μg of protein extract following the manufacturers instructions. Proteins were then separated by SDS-PAGE using a 3–8% Tris-Acetate NuPage™ gradient gel system (Invitrogen). Proteins were transferred onto nitrocellulose membranes using NuPage™ transfer buffer as per manufacturers instructions (Invitrogen). After transfer the nitrocellulose membranes were incubated in 3% (w/v) BSA/PBS/0.1% (w/v) Tween 20 for 1 hour at room temperature. Membranes were then incubated overnight at 4°C with the specific antibody diluted in 3% (w/v) BSA/PBS/0.1% (w/v) Tween 20. After washes in PBS/0.1% Tween 20, membranes were incubated with an HRP-conjugated secondary antibody for 2 h before further extensive washing. Immunoreactivity was detected using the SuperSignal® West Pico Chemiluminescent Substrate Kit (Pierce).
For immunoprecipitation, COS-7 cells were transfected as described below (see Live Imaging) then resuspended in modified RIPA buffer (50 mM Tris pH7.4, 80 mM KCl, 10 mM EDTA, 1% Triton X-100 containing EDTA-free complete protease inhibitors (Roche) and phosphastase inhibitors cocktail II (Sigma)). Extracts were then immunoprecipitated with 2 μg of polyclonal anti-GFP antibody (Clontech) overnight at 4°C. The following day 40 μl of an 80% protein-G sepharose/PBS solution was added to each extract and incubated for a further 3 hours at 4°C. Extracts were then washed ×3 in modified RIPA buffer and the pellets resuspended in 30 μl PBS. Extracts were then processed for SDS-PAGE and Western blotting as described previously.
Cells were cultured on sterile coverglasses, processed for immunocytochemistry using cold methanol  or paraformaldehyde fixation  and imaged using the CCD camera based system described below in conjunction with excitation/emission filtersets for DAPI, FITC and TRITC.
Cells were grown and transfected in 35 mm glass-bottomed culture dishes (Iwaki brand; Asahi Techno Glass Corporation, Japan) obtained from Bibby Sterilin. Transfections were performed using GeneJuice (Novagen) according to the manufacturers instructions. 12–18 h after transfection the standard cell culture medium was replaced by 2 ml of pre-warmed medium supplemented with 20 mM HEPES. The cells were transferred to a Zeiss Axiovert 200 inverted microscope with a heated chamber enclosing the microscope stage (Solent Scientific, UK) allowing the temperature to be maintained at 37°C throughout imaging. Cells were imaged by fluorescence microscopy using a Zeiss Plan Apo 63X/1.4NA oil immersion lens. Time-lapse images were captured at 3–10s intervals for durations of 5–10 min using Ludl shutters and a Hamamatsu Orca II ER camera. Images were obtained using 2 × 2 binning with exposure times of less than 350 ms/frame. An excitation/emission filterset optimised for EGFP imaging was used (Chroma Technology Corp., Brattleboro, USA; filterset ID 86007). Microscope, camera, filterwheels and shutters were controlled by Kinetic Imaging AQM 6 software (Kinetic Imaging, Nottingham, UK). Cells expressing as low a level of fusion protein as could be successfully imaged using our CCD camera were used in this work. Brightly fluorescent cells, typically those clearly visible without the use of the camera, often displayed evidence of extensive microtubule bundling and were excluded from the study. Time-lapse image series were saved as uncompressed AVI files then cropped, compressed and converted into Quicktime movies using Adobe ImageReady CS. Tracking analyses were performed on unprocessed data files using Motion Analysis software from Kinetic Imaging.
The pEGFP-APC plasmid used in this study was a kind gift from Dr. J. Victor Small (Salzburg, Austria). It directs the expression of full length human APC (2843aa) N-terminally tagged with the fluorescent protein eGFP. This construct was subjected to a range of restriction digests and sequencing to confirm its identity. This indicated that the plasmid directs the expression of a full-length GFP-APC protein (Figure 1, panel A) similar to that previously described  rather than a N-terminally truncated protein as recently used by other workers . A further four constructs were produced by restriction enzyme digestion of this plasmid. The first of these, pEGFP-APCΔC, was obtained by digestion of pEGFP-APC with BspE I and Avr II, resulting in an APC fragment lacking its final 54 amino acids. This fragment was ligated into pEGFP-C1 (Clontech) digested with Kpn I and Xba I (Figure 1, panel B). A second construct, pEGFP-APCΔNΔC, was obtained by digestion of pEGFP-APC with Kpn I and Avr II. This results in the removal of the first 206 amino acids from the N-terminus of APC in addition to the removal of the last 54 amino acids. This product was cloned into pEGFP-C1 digested with BspE I and Xba I (Figure 1, panel C). The third construct, pEGFP-APC-C, was created by PCR amplification of a C-terminal APC fragment using the following primers to obtain the last 170 amino acids of APC:
GFPC1For - 5'CCTAGATCTTCCGGATCTCCCACAG3'
The forward primer contains a BspE I restriction site and the reverse primer a Kpn I restriction site to allow subcloning of the PCR product into pEGFP-C1 digested with the same enzymes (Figure 1, panel D). A fourth construct, pEGFP-APC-N, was obtained by digesting pEGFP-APC with BspE I and Hind IIII and cloning the resulting product into pEGFP-C1. The resulting plasmid directs the expression of the first 746 amino acids of APC N-terminally fused to EGFP (Figure 1, panel E).
Cancer Research UK supported this work.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.