SEARCH

SEARCH BY CITATION

Keywords:

  • spectrin;
  • actin;
  • enteropathogenic Escherichia coli

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

Enteropathogenic Escherichia coli (EPEC) manipulate the cytoskeleton of host intestinal epithelial cells, producing membrane protrusions termed pedestals that the bacteria reside on throughout the course of their infections. By definition pedestals are actin-based structures, however recent work has identified the spectrin cytoskeleton as a necessary component of EPEC pedestals. Here, we investigated the detailed arrangement of the spectrin and actin cytoskeletons within these structures. Immunofluorescent imaging revealed that the spectrin network forms a peripheral cage around actin at the membranous regions of pedestals. Myosin S1 fragment decorated actin filaments examined by electron microscopy demonstrated that actin filaments orientate with their fast-growing barbed ends toward the lateral membranes of EPEC pedestals. These findings provide a detailed descriptive analysis, which further illustrate the spectrin cytoskeletal organization within these structures. Anat Rec, 2012. © 2011 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

The spectrin cytoskeleton lies beneath the cytoplasmic surface of the plasma membrane of eukaryotic cells and is composed of a meshwork of proteins providing structural support and a high degree of organization to the plasma membrane. The framework of the spectrin scaffold is built on dimers of α/β spectrin monomers, which assemble into heterotetramers (Bennett and Baines,2001). Spectrin heterotetramers link to adjacent heterotetramers by simultaneously binding short actin filaments (14–16 monomers), forming a spectrin–actin complex (Bennett,1989). These complexes can be a source of exposed actin, enabling the addition of actin monomers to polymerize off of the spectrin network (DiNubile,1999). The spectrin cytoskeleton is suggested to play a key role in cell migratory processes by concentrating near the back of lamellipodia and thereby providing support for the growing actin network at the leading edge (Bournier et al.,2006). Disruption of the spectrin meshwork impedes cell migration (Fukata et al.,1999).

The spectrin cytoskeleton functions in large part through interactions with a selection of binding partners that most often include adducin and protein 4.1 (p4.1). There are a large number of spectrin, adducin, and p4.1 isoforms that have diverse functions within cells and tissues. Over seven established isoforms of spectrin have been identified in humans; some that are only expressed in specific cell types. Similarly, there are multiple adducin and p4.1 isoforms, each exhibiting unique expression patterns and functions (Bennett and Baines,2001). Adducin associates with the fast-growing ends of actin filaments, helping to recruit spectrin to those regions, while also acting as an actin-capping protein (Pinder et al.,1984; Kuhlman and Fowler,1997; Li et al.,1998). P4.1 enhances spectrin–actin interactions by providing a 109 increase in the association constant of the spectrin–actin complex in vitro (Ohanian et al.,1984). Both adducin and p4.1 can also bind to a broad range of integral membrane proteins, providing links between the spectrin/actin cytoskeletons and the plasma membrane (Butler et al.,2008; Baines,2010). Thus, the spectrin meshwork is heavily involved in actin remodeling during cell migration, actin-capping activities, and linking the cortical actin network to the plasma membrane.

The human intestinal pathogen enteropathogenic Escherichia coli (EPEC) dramatically manipulates the spectrin and actin cytoskeletal networks of host cells (Sanger et al.,1996; Ruetz et al.,2011). On attachment to enterocytes, EPEC utilizes a type-three secretion system to translocate bacterial proteins (effectors) into the host cytosol (Jarvis et al.,1995). These effectors commandeer and rearrange the subcellular machinery of the host cell for the benefit of the microbe (Dean and Kenny,2009). Effector actions culminate in the polymerization of actin filaments beneath sites of EPEC attachment, ultimately causing the bacteria to rise off the natural surface of the host cell on actin-rich structures referred to as “pedestals” (Shaner et al.,2005). Pedestals are dynamic structures that allow the bacteria to “surf” along the top of the host cells in an actin dependant manner (Sanger et al.,1996,2005). The use of actin inhibitory drugs, such as cytochalasin, abolishes this motility (Sanger et al.,1996). Recently, spectrin and its associated proteins have been identified in EPEC pedestals (Ruetz et al.,2011); however, the precise location of these components has not yet been determined.

In this study, we tested the hypothesis that the spectin cytoskeleton is situated peripheral to the actin network, supporting the idea that the spectrin network may provide a link between actin filaments and the plasma membrane. We chose to investigate isoforms of the spectrin network that have broad expression patterns, which are also expressed in epithelial cells (βII-spectrin, α-adducin, and p4.1G). Using a combination of immunofluorecent and electron microscopy, we demonstrate that spectrin cytoskeletal proteins cage the actin filaments within EPEC pedestals. Spectrin proteins are recruited not only at the base of the pedestals, toward the minus ends of the actin filaments (Ruetz et al.,2011), but also along the plasma membrane of the pedestals. By decorating the actin filaments within pedestals with the S1 fragment of myosin, we demonstrate that actin filaments within these structures are oriented with their fast-growing (barbed) ends toward the lateral membranes of the pedestals, while also confirming the long standing evidence that actin filament barbed ends are directed toward sites of EPEC attachment (Sanger et al.,2005). These findings reveal a unique architectural arrangement of the spectrin and actin cytoskeletons within EPEC pedestals and provide further descriptive evidence of the spectrin network within these structures.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

Antibodies and Reagents

Primary antibodies consisted of mouse monoclonal anti-β-Spectrin II (used at 2.5 μg/mL or 0.25 μg/mL for westerns; Becton Dickinson), rabbit anti-α-adducin (used at 2 μg/mL or 0.2 μg/mL for westerns; Santa Cruz Biotechnology), and rabbit anti-EPB41 (p4.1; used at 1.7 μg/mL or 0.17 μg/mL for westerns; Sigma Aldrich; Product # HPA006642). The p4.1 antibody was raised by the supplier against the following sequence: ESQSSLRRQKR EKETSESRGISRFIPP WLKKQKSYTLVVAKDGGDKKE PTQAVVEEQVLDKEEPLPEEQRQAKGDAEEMAQKKQ EIKVEVKEEKPSVSKEEKPSVSKVEMQPTELVSKEREE KVKETQEDKLEGGA. Antibody specificity was confirmed using western blot analysis (Supporting Information Fig. S1). Secondary antibodies included goat anti-mouse or anti-rabbit antibodies conjugated to AlexaFluor 568 or 594 (used at 0.02 μg/mL; Invitrogen). For F-actin staining AlexaFluor 488 conjugated phalloidin (Invitrogen) was used according to the manufacturer's instructions. Bacterial and host cell DNA was visualized using the mounting medium Prolong Gold with DAPI (Invitrogen).

Cells, Bacteria and Growth Conditions

HeLa cells were grown in Dulbecco's Modified Eagles Medium supplemented with 10% fetal bovine serum (BSA). Bacterial strains consisted of wild-type EPEC [strain E2348/69] and EPEC strain JPN15. The JPN15 strain does not encode the bundle forming pilus (bfp) gene; thus, they do not form microcolonies and form larger independent pedestals. Bacteria were grown in standard Luria broth (LB), in a standing culture at 37°C.

Infections

HeLa cells were seeded on cover slips or 0.2 μm transwells (for electron microscopy studies) 2 days before infection and grown to 70% confluency. Bacteria were grown in LB overnight (16 hr) in standing cultures at 37°C, before infection. Bacteria were then added to HeLa cells at an MOI of 10, for a 6-hr infection period. Cells were then quickly washed three times with PBS buffer (Hyclone) and processed for immunofluorescent or electron microscopy.

Immunofluorescent Localizations

Cells were treated at room temperature with 3% paraformaldehyde for 15 min, permeabilized using 0.1% Triton X-100 in PBS (without calcium or magnesium) for 5 min, then washed 3 times (10 min each) with PBS (Hyclone). Samples were blocked in 5% normal goat serum in TPBS/0.1% BSA (0.05% Tween-20 and 0.1% BSA in PBS) for 20 min. Antibodies were then incubated on the cover slips overnight at 4°C. The next day, the cover slips were washed three times (10 min each) with TPBS/0.1% BSA. After the final wash, secondary antibodies were applied for 1 hr at 37°C. This was followed by three additional washes (10 min each) with TPBS/0.1% BSA. The cover slips were then mounted on slides using Prolong Gold either with or without DAPI (Invitrogen). Samples were all visualized using an inverted Leica DMI4000b inverted fluorescent microscope, fitted with phase and fluorescent optics and illumination as well as a motorized Ludl Z-stage. Widefield images were taken with a Hamamatsu Orca R2 CCD camera (Hamamatsu, Japan) and used a Leica HCX PL APO 100×/1.40 numerical aperture, Oil immersion PH3 CS Objective.

Myosin Subfragment-1 Isolation

Myosin was obtained using the method of Margossian and Lowey (1982), and subfragment-1 (S1) was isolated using the protocol described by (Pollard,1982). Starting material was skeletal muscle from four adult male Sprauge Dawley rats originally obtained from Charles River Laboratories. The animals, all weighing more than 500 g, were euthanized by CO2 inhalation and then muscle was removed from the hind limbs until a total of 125 g was obtained. Once the S1 was isolated, aliquots were frozen and stored in liquid nitrogen until used.

Myosin S1 Decoration of Actin Filaments

For S1 decoration of actin filaments in pedestals, HeLa cells were grown on 0.2 μm membranes in Transwells (Millipore) and infected with EPEC [strain E2348/69] at an MOI of 10 for 6 hr. The media in the top and bottom of the chambers was replaced with suspension buffer (10 mM sodium phosphate buffer, 150 mM NaCl, 5 mM MgCl2, 2 mM EDTA, and protease inhibitors [Roche] pH 7.0) containing 50% glycerol. After 1–2 min of extraction, the cells were washed once with suspension buffer containing 0.1% BSA and then incubated for 15 min at room temperature with suspension buffer containing 2 mg/mL S1. The cells were washed once with suspension buffer and then fixed for 1 h at room temperature in a solution containing 1.0% gluteraldehyde, 0.2% tannic acid, and 0.1 M sodium phosphate buffer (pH 7.0; Begg et al.,1978). After fixation, the membranes were cut from the transwells, transferred to scintillation vials containing 0.1 M sodium phosphate buffer (pH 7.0), and then processed further for electron microscopy.

Electron Microscopy

The membranes were washed three times (10 min each wash) with 0.1 M sodium phosphate buffer (pH 7.0) and then postfixed 1 hr on ice with 1% OsO4 in 0.1 M sodium phosphate buffer (pH 6). They were washed three times (10 min each wash) with ddH20, “en bloc” stained 1 hr with 1% aqueous uranyl acetate, washed again three times with ddH20, and then dehydrated through a graded series of ETOH concentrations. The membranes were infiltrated through propylene oxide into Embed-812 and then embedded. Thin sections were stained both with uranyl acetate and with lead citrate and imaged on a Philips 300 electron microscope operated at 60 kV.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

Spectrin, P4.1, and Adducin Encase Actin Filaments at EPEC Pedestals

To investigate the detailed localization of spectrin and p4.1 within EPEC pedestals, we infected HeLa cells with the JPN15 strain of EPEC. EPEC JPN15 is an EAF cured strain of the prototypic EPEC E2348/69 strain, which does not encode the bfp gene but contains identical pedestal host components as the E2348/69 strain (Badea et al.,2003). Without bfp, EPEC JPN15 is unable to produce the characteristic microcolonies on the surface of host cells (Hicks et al.,1998), but rather they generate large individual pedestals, enabling more precise imaging of the pedestal structure. We immunolocalized spectrin, p4.1 and filamentous actin at EPEC JPN15 pedestals and found that, in agreement with our previous work on EPEC E2348/69 (Ruetz et al.,2011), spectrin was highly concentrated at the base of the pedestal, partially colocalizing with filamentous actin (Fig. 1a,b). When examined in more detail, we found that spectrin also localized to the sides and apical tips of pedestals (Fig. 1b). On the sides of pedestals, spectrin appeared peripheral to the actin network. P4.1 also localized near the plasma membrane peripheral to actin and was colocalized with spectrin (Fig. 1b,c). However, unlike spectrin, p4.1 did not form a concentrated mass near the base of the pedestals (Fig. 1b,c). Rather, p4.1 appeared to surround the entire actin–spectrin network (Fig. 1b,c). When we examined pedestals from a top–down (enface) view, both spectrin and p4.1 localized peripheral to actin (Fig. 1d [spectrin] and 1e [p4.1]).

thumbnail image

Figure 1. Immunofluorescent images detailing the location of spectrin and p4.1 within EPEC pedestals. HeLa cells were infected with EPEC JPN15 for 6 hr, then fixed and stained for spectrin, p4.1, and probed for actin or DNA (DAPI). (a) Spectrin localizes to the base of the pedestal and along the pedestal periphery. P4.1 lines the pedestal membranous region, and actin localizes to the top half of the pedestal. (b) Overlay of spectrin (green) and p4.1 (red) within the pedestal. (c) Overlay of p4.1 (red) and actin (green) within the pedestal. We also show a top–down (enface) view of a pedestal showing (d) spectrin (red) peripheral to actin (green) and (e) p4.1 (red) peripheral to actin (green). Scale bars are 5 μm (a–c) and 2.5 μm (d and e).

Download figure to PowerPoint

In addition to spectrin and p4.1, we examined the organization of the spectrin-associated protein adducin. Adducin was localized to peripheral plasma membrane regions of pedestals but was not found at pedestal tips. It was also present at the base of pedestals, but not to the areas known to concentrate spectrin directly beneath the polymerized actin (Fig. 2a). The enface view confirmed that adducin localized peripheral to actin (Fig. 2b).

thumbnail image

Figure 2. Adducin localizes to the periphery of EPEC pedestals, but not the pedestal tip. HeLa cells were infected with EPEC JPN15 for 6 hr, then fixed and stained for immunfluorescent microscopy. (a) Adducin (red) localizes to the peripheral membranous region of the pedestal, forming a cup around the filamentous actin (blue). (b) The enface view identifies adducin (red) localizing peripheral to actin (green) beneath EPEC (blue). Scale bars are 2.5 μm.

Download figure to PowerPoint

The Orientation of Actin Filaments in Relation to Spectrin Within EPEC Pedestals

Because we observed spectrin and its associated components, peripheral to the actin filaments within pedestals, we sought to examine how actin filaments were oriented at the lateral membranous regions of the pedestals. To accomplish this we infected HeLa cells with EPEC, labeled the actin within the pedestals with purified S1 fragments of the myosin II motor protein and visualized the results using electron microscopy. The S1 labeling technique not only identifies actin filaments but also indicates filament polarity by generating a characteristic arrowhead appearance (Small et al.,1978). Arrowheads point toward the slow-growing minus end (− end) of the filament, whereas the barbed end indicates the fast-growing plus end (+ end) of actin filaments (Woodrum et al.,1975). In agreement with previous work (Shaner et al.,2005), we observed actin filaments positioned with the fast-growing + (barbed) ends near sites of EPEC attachment toward the apices of pedestals (Fig. 3). In addition, actin filaments were observed with their + ends directed toward the lateral membranes of the pedestals. In virtually all observed instances, the − (pointed) ends of the filaments pointed toward the pedestal base (Fig. 3).

thumbnail image

Figure 3. S1 decorated actin filaments identify the orientation of actin within EPEC pedestals. HeLa cells were infected with EPEC (E2348/69) for 6 hr then incubated with S1 fragments of myosin II, before processing for electron microscopy. The bacteria are at the top of images. Arrows indicate the direction of the − ends of the actin filaments, with the barbed + ends in the opposite direction. Actin filament barbed ends are observed orientated toward the apex of the pedestal as well as toward the lateral membranes of the pedestals. Scale bar is 0.5 μm.

Download figure to PowerPoint

Overall Actin Distribution in Pedestals

The enface views of pedestals highlight the actin arrangement within the pedestal and indicate that the actin distribution is not uniform throughout the pedestal. Actin appears dense around the periphery of the cylindrical-shaped pedestals and more diffuse at the centre (Figs. 1d,e, and 2b).

Interestingly, there appears to be few actin filaments in host cytoplasmic regions at the base of pedestals (Figs. 1a,c, 2a, and 3). These findings, in conjunction with the observed spectrin immunolocalization in these regions (in the absence of actin), supports our previous finding that spectrin is a major player in producing the pedestal structure (Ruetz et al.,2011).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

We have previously reported that the spectrin cytoskeleton is recruited to EPEC pedestals and is crucial for pedestal formation (Ruetz et al.,2011). Due to the importance of this cytoskeletal system in these infections, in this study, we qualitatively examined the precise organization of βII-spectrin, α-adducin, and p4.1G within EPEC pedestals. We demonstrate that the spatial arrangements of the spectrin and actin networks are distributed differently than previously thought.

Previous work identified spectrin lining the base of the pedestal (Ruetz et al.,2011). Here, we show that the spectrin network in EPEC pedestals is not only at the pedestal base, but also peripheral to the actin filaments. These findings suggest that spectrin lies between the actin filaments and the plasma membrane at the lateral and apical pedestal boundaries. It has previously been shown that actin filaments are positioned with their fast-growing (barbed) ends at the apex of pedestals near the site of EPEC attachment (Sanger et al.,1996; Gruenheid et al.,2001; Shaner et al.,2005). In this study, we confirm those findings, but we also reveal that actin filaments occur with their barbed ends aimed at the lateral membranes of the pedestals. In all cases, actin filaments are oriented with their barbed ends toward the spectrin cytoskeleton at the plasma membrane.

The spectrin cytoskeleton provides two potential functions at the interface between actin and the plasma membrane: (1) capping the fast-growing ends of actin filaments and (2) providing a link to the plasma membrane. The spectrin-associated protein adducin is capable of capping the fast-growing ends of actin filaments, as the binding of adducin completely blocks elongation and depolymerization of actin filaments at the barbed ends (Kuhlman et al.,1996). The orientation of F-actin barbed ends toward the lateral membranes of pedestals, together with the indication that actin monomers are not polymerizing on the filaments at these peripheral regions (Shaner et al.,2005) and the observed concentration of adducin in this location, suggests adducin could be capping actin filaments in those areas. Furthermore, the absence of adducin at the pedestal apex could explain how actin monomers are able to polymerize on the barbed ends of filaments in this zone (Shaner et al.,2005).

A second possible function of the spectrin network at the membranes of the pedestal could be to provide a link between actin and the plasma membrane. Extensive work has shown that spectrin, adducin, and p4.1 collectively enable the binding of a subset of membrane lipids and integral membrane proteins (Lombardo et al.,1994; Kizhatil et al.,2007; Abdi and Bennett,2008; Lowe et al.,2008). Consequently, the spectrin cytoskeleton is thought to provide a link between the actin cytoskeleton and the plasma membrane in other systems (Baines,2009) and could provide such functions within EPEC pedestals. Additionally, this cytoskeletal system could provide structural support to the pedestal as previous work has shown that the absence of any spectrin cytoskeletal proteins completely inhibits pedestal formation (Ruetz et al.,2011).

Our demonstration that spectrin localizes beneath actin at EPEC pedestals is similar to previous observations of spectrin at sites of membrane ruffling in motile cells. Spectrin localizes with the actin remodeling protein EVL, which is known to regulate actin dynamics during many cellular processes including cell migration (reviewed in Krause et al.,2003). EVL, together with spectrin, trails the leading edge of filopodia and lamellipodia (Bournier et al.,2006) and when spectrin cytoskeletal components are disrupted, cell migration is inhibited (Fukata et al.,1999). Consequently, membrane remodeling and the actin network are directly influenced and dependent on the spectrin cytoskeleton during actin-based membrane reorganization. In a similar manner as filopodia and lamelipodia, EPEC pedestals are dynamic structures that enable the bacteria to “surf” atop the surface of the host cells and require the recruitment of actin reorganization machinery as well as the polymerization of actin filaments at the plasma membrane to accomplish this function (Sanger et al.,1996; Kalman et al.,1999; Gruenheid et al.,2001; Shaner et al.,2005; Cantarelli et al.,2006; Brown et al.,2008). Taken together, the demonstration that spectrin cytoskeletal components concentrate basal to the actin network in pedestals, coupled with the requirement of this network in the formation of these structures (Ruetz et al.,2011), further shows that spectrin is a necessary component of actin-based membrane remodeling events.

Most pedestal proteins have been described based on their requirement for the formation of these structures. Proteins such as the endocytic components clathrin, epsin, and Eps15 are needed for pedestal formation, but their precise functions beyond that remain elusive (Veiga et al.,2007; Lin et al.,2011). Others such as ZO-1 (Hanajima-Ozawa et al.,2007) have been demonstrated to be involved in proper pedestal morphology, but again the mechanisms remain unresolved. Finally, there are actin-related proteins such as IQGAP1, that are at pedestals and may be part of a signaling complex involved in forming these structures (Brown et al.,2008) along with the actin-associated proteins N-WASp and the Arp2/3 complex. Although extensive work has examined N-WASP and Arp2/3, any interaction of spectrin with those proteins has not been demonstrated.

Spectrin occupies nearly half of the pedestal and overlaps with actin in the middle ∼1/3 of these structures. This indicates that the actin cytoskeleton may not be acting alone, but rather in concert with the spectrin cytoskeleton to produce the protruding pedestal structures (Ruetz et al.,2011; Fig. 4). The concentration of spectrin at the base of EPEC pedestals suggests that this cytoskeletal network may also provide a support platform off of which the actin network can generate the membranous protrusion. A similar spectrin substratum exists at the base of intestinal epithelial microvilli. The microvillar actin core bundles are cross-linked and supported by the spectrin network at the terminal web of the brush border intestinal epithelium (Hirokawa et al.,1982). Hence, it is not surprising that we see a similar architecture at EPEC pedestals, with spectrin and associated proteins at the base of these structures.

thumbnail image

Figure 4. Model depicting the spectrin–actin cytoskeletal networks within EPEC pedestals. (a) In this schematic, we show the spectrin cytoskeleton (spectrin, p4.1 and adducin) lining the membranous region of the pedestal, surrounding the actin cytoskeleton. The actin core is a network of actin filaments, with their barbed ends orientated toward the site of EPEC attachment, as well as the lateral sides of the pedestal. The spectrin framework, specifically the actin-capping protein adducin, may be involved in modulating actin dynamics at the lateral sides of the pedestal. P4.1, together with adducin, is capable of interacting with a number of membrane proteins and may provide a link between the actin–spectrin networks and the plasma membrane. The bottom third of the pedestal is composed of spectrin, lined by p4.1 and adducin, in the absence of actin. (b) The top–down cross-sectional view depicts the spectrin network localized peripheral to actin, near the membrane of the pedestal. The actin network within the pedestal is most dense along the outer periphery, with decreasing intensity toward the center of the structure.

Download figure to PowerPoint

Contrasting previous definitions of pedestals as “actin-based structures” (Goosney et al.,2000; Brown et al.,2008; Croxen and Finlay,2010), our findings suggest that pedestals are actin–spectrin based membrane protrusions and that the spectrin cytoskeleton appears to encapsulate the actin network within the pedestals. This spectrin-based cage may function to retain actin filaments within the boundaries of the pedestals, maintaining structural integrity throughout pathogenesis. Consequently, our studies support a novel pedestal architecture, requiring a redesigned model of the structure (Fig. 4). Future research should focus on addressing the mechanisms of spectrin cytoskeletal targeting during EPEC pathogenesis, identifying the precise role that this cytoskeletal system provides in pedestal formation and understanding the relationship of the spectrin components with other known pedestal proteins.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

We would like to thank Fern Ness for the artistic rendering of the EPEC actin/spectrin pedestal model.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
AR_21544_sm_SuppFig1.tif1040KSupporting Information Fig. S1. Western blot confirming antibody specificity. Uninfected HeLa cells were analyzed by western blot to show antibody specificity. Bands using βII-spectrin, α-adducin, and p4.1G antibodies are evident at the predicted molecular weights of 280 kDa, 120 kDa, and 170-180 kDa respectively.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.