EDInfo Biomedical Sciences Maciver Lab. Home ABP  A-Z Encyclopaedia Amoebae Protist Links Cytoskeleton Links Site Index

The Invasion of Eukaryotic cells by Bacteria

 (the role of the cytoskeleton)


A disturbingly large number of bacterial species have seemingly targeted our bodies for colonisation.  As in all such relationships, it is in the long-term interest of both to co-exist as happily as possible (it is not a good idea to kill your host!), and some bacteria (e.g. commensal Salmonella) enjoy a "special relationship" (Xavier & Podolsky, 2000; Neish et al, 2000).  Mutations and recombinations in bacteria cause upsets in the balance.  These modified bacteria emerge as new and intimidating diseases (e.g. E.coli O157).  New and old disease causing organisms are increasingly likely to be resistant to the antibiotics that we have used to upset this balance in our favour, and so it is vital that we discover the mechanisms that bacteria use to colonise our bodies so that we may develop new antibiotics to maintain the upper hand.  Many pathogenic bacteria have evolved strategies to evade the host immune response by subverting the cytoskeleton through various toxins to persuade host cells to take the bacteria into the cell, or in the case mainly of “professional phagocytes” to prevent their uptake into the phagolysosome.

Bacteria secrete toxins
The pathogenesis of bacteria is in part due to secreted proteins that alter the host cell function to facilitate the pathogenic/commensal relationship. Many of these proteins turn out to work to the bacteria's advantage through the manipulation of the host cell cytoskeleton.


Figure 1 Various types of bacterial secretion used by  gram negative bacteria. Bacteria secrete toxins through three types of apparatus.  All types require an ATPase activity (represented by the ‘A’ in the circle). Type I and Type III are similar, both result in proteins being transferred across both membranes in one step, the difference is in the complexity.  Type II secretion, proteins are placed in the periplasm, where a signal is cleaved off, then the mature protein is secreted .  Note that the Type III machinery has many similarities to the bacterial flagella from which this system is thought to have evolved!

Pathogenic Escherichia coli

E.coli though not quite as abundant as many assume are none the less important natural partners in the animal gut, accounting for 0.1% of the total bacteria present (~50% are Gram-negative Bacteroides spp - B. vulgatus, B. fragilis and B. thetaiotaomicron. Distinct populations inhabit the mucosal surface (attached) and the lumen (~faeces). Gram-positive anaerobes: Eubacterium spp, Bifidibacterium spp, Lactobacilli and Clostridium spp make up the other 50%).  There are many E.coli strains that infect humans, the most common are listed in table 1 below in order of increasing severity.

Abbrev. Full name  Common name and features    inocolum  Source
ETEC    Enterotoxin E.coli Montizuma’s revenge, traveler’s tummy  108     Faecal contamination
EIEC     Enteroinvasive E.coli  Invades, Shigella pathogenicity island   high Food & waterborne
EPEC    Enteropathogenic E.coli   Pedestal formation, infant diarrhoea  108 - 10            Nosocomial, community
EHEC   Enterohaemorrhagic E.coli   (O157)“Hamburger disease” Shiga toxin   3  Cattle faeces, meat
Table 1 Pathogenic E. coli

Figure 2 Interaction of various pathogenic E.coli with the epithelial cells of the gut.   A.ETEC bind loosely via fimbriae, secrete toxins (like Cholera toxins) into the gut that then gain entry into the cell without disruption of cytoskeleton. B. EPEC destroys the brush border microvilli, and becomes firmly attached through a pedestal consisting of actin and actin binding proteins. C. EIEC, gains entry into the cell, escaping from the immune system by digesting the phagolyosome.  EIEC can grow and divide in the cell cytoplasm and gain entry to neighbouring cells by bursting through and digesting membranes. D. EHEC, operates like EPEC, but in addition Shiga toxins are liberated that the epithelial cells take up in coated pits and taken to the Golgi.  The toxins then travel from Golgi to the E.R. where they destroy ribosomes by the removal of a single adenine residue from the 28SrRNA.  This results in the death of the cell.

EHEC targets cells in the gut that will be around the longest by binding to cell surface nucleolin a protein expressed in dividing cells. In other cell type (mast cells), caveolae have been suggested of being involved in binding and uptake of pathogenic E.coli (Shin et al, 2000).






ETEC is responsible for most “upset stomachs” encountered when travelling to warmer countries.  The toxin is of a similar type produced by Vibrio cholerae the causitive agent of cholera.  This toxin ADP-ribosylates the a subunit of the Gs-protein controlling adenyl cyclase, resulting in elevated cAMP levels and chloride ions being pumped from the cell.  This leads to horrendous diarrhoea and dehydration.  ETEC does not involve the cytoskeleton so will not be discussed further. 

EPEC forms a “pedestal” by the rearrangement of the host actin based cytoskeleton and may provide a useful and tractable system with which to explore signal transduction pathways that result in actin modulation (Goosney et al, 1999).  This bacteria is responsible for many deaths in “developing” countries not because it cannot be cured, but through a lack of antibiotics.

Figure 3. 

EIEC are highly invasive bacteria, avoiding the immune system and gaining nutrition  While EPEC are known to invade cultured cells, they are not thought to be naturally invasive.  It has been discovered (Boudeau et al, 1999) that a new type of invasive E.coli (AIEC) isolated from ileal lesions of Crohn’s disease sufferers, suspected of involvement in the disease.  These E.coli are known to have received a “pathogenicity island” from Shigella bacteria (see below) that is an invasive bacteria.  This apparently has allowed EIEC to become invasive.  Many bacteria (and viruses) have been suggested to cause Crohn’s disease, it is possible/probable that this is not one disease at all but many.

EHEC is very similar in its mode of action as EPEC.  The most common EHEC subtype is E.coli O157, the famous bacteria responsible for the Lanarkshire outbreak that killed around 20 people and infected some 500 others in 1996 (Pennington, 1997).  This is a highly infectious bacterium, as few as three bacteria may produce an infection! What makes E.coli O157 so infective and pathogenic to humans is the vero/shiga toxins that it contains.  Apparently, this toxin has very recently been transferred from Shigella dysenteriae to E.coli via bacteriophage causing the EHEC phenotype.  This event may have taken place within the last 30 or so years and this strain is now turning up all over the world due to “globalisation” of commerce and human travel.  A scary thought!!  0157 probably does not affect cattle as they lack the toxin receptor globotriaosylceramide (Gb3) (Pruimboom et al, 2000). 

Figure 4.  EHEC receptors. The receptor for EHEC turns out to be nucleolin (Sinclair & O'Brien, 2002), this makes sense as cells in the crypts divide and express nucleolin on their surfaces while they do so.  Cells migrate from the crypts towards the lumen as cells are lost.  Targeting dividing cells therefore allows the EHEC to target cells that will remain in place the longest.  Thus EHEC use not only TIR to adhere to cells but also nucleolin.
Commensal Salmonella -Oil on troubled waters.
When a British politician (Edwina Currie) caused uproar by suggesting that many eggs were contaminated by Salmonella (see BBC page), she was correct.  However, not all Salmonella are "bad" and most we get along with most without trouble.  It has been shown that commensal strain of Salmonella blocks the ubiquitination a host protein I
kB and this ultimately leads to the inhibition of inflammation of the gut by suppression of the NF-kB transcription factor (Neish et al, 2000) (Figure 3).  It has been suggested that these commensal bacteria may be used to inoculate the guts of patient suffering from various inflammatory bowel disorders such as Crone's disease (Xavier  & Podolsky, 2000).

Pathogenic Salmonella
In common with the other major nasty types of bacteria this one is named after a person, Daniel E. Salmon, an American pathologist.  Salmonella infections produce symptoms ranging from mild food poisoning (gastro-enteritis) to fatal typhoid fever.  Salmonellosis is never good news, but it is particularly nasty for those with sickle cell anaemia, presumably because the bacteria produces a haemolysin (called typholysin).  Like other bacteria, Salmonella has evolved the habit of invading host cells in order to hide from the immune system and to gain nutrients.  However, Salmonella also invades the epithelial cells in order to escape the gut into the surrounding tissues.  Pathogenic Salmonella such as S. typhimurium, have a pathogenicity island that contains some 28 genes that encode both the type III secretory mechanism and the various toxins that mediate invasion and intra-cellular survival.  In addition to actively attempting to invade cells by inducing phagocytosis, Salmonella avoid being phagocytosed by professional phagocytes such as neutrophils and macrophages by the production of toxins (see below).  Whereas E.coli binds to the gut epithelial cells, Salmonella binds to M cells, present in the Peyer’s patches (Lymph nodes) in the gut wall.  Once inside the cell, Salmonella resides in a compartment that does not fuse with lysosomes so that the bacteria are not exposed to proteases. 

Figure 5.  Salmonella produces Actin Binding Proteins to Induce Phagocytosis.  Salmonella introduces many toxins into the host cell by type III secretion.  SipB & SipC are the most abundant. SipB alters phospholipid metabolism.  SipA is an actin binding protein that increases the bundling protein plastins ability to bundle actin filaments (See Plastin /Fimbrin section below).  Sip C is an actin bundling protein.   SopE directly antagonises the action of SptP (Fu & Galan 1999).  Both SopE and SptP are GTP exchange proteins that act on Rac and Cdc42.  SopE activates the pathway by increasing the amount of G-protein in the GTP state while SptP does the opposite!  When cells become infected with Salmonella is it seen that the hyper-ruffling takes place only at the beginning, thereafter the cell becomes quiescent.  It is because SptP turns the system off again!

Bacteria use the same or similar mechanisms in order to induce actin polymerization.  In the case of EPEC, the G-protein Chp is thought to activate WASP to produce actin polymerization. Whereas Salmonella uses Rac and Cdc42 activation to do exactly the same.  In the first case this leads to the formation of a static pedestal on which the EPEC is stably held, and in the second case the bacterium is taken into the cell.  Obviously, we are only just beginning to understand how the pathways are switched on, and in Salmonella, off again.  How the pathways result in the very different host cell behaviours will be very much harder to sort out!  The study of bacterial invasion has also highlighted the fact that there appear to be many different types of phagocytosis.  So, as you will appreciate from the next lecture, there is a two way benefit from the study of how micro-organisms commandeer the cytoskeleton of our cells: we learn how the bacteria do this and at the same time we can use the bacteria as tools to understand normal cytoskeletal function!


This is a family of actin bundling proteins first discovered in chicken intestinal microvilli (Matsudaira & Burgess, 1979).  There are three human isoforms L, T and I that differ in tissue distribution.  Structurally, plastins are simple monomeric proteins that are able to cross-link microfilaments because they contain two homologous actin binding domains of the Calponin-Homology (CH) type, also found in actin binding proteins (ABPs) such as a-actinin, dystrophin, spectrin, cortexillins and calponin (Figure 9).  Plastin is involved in invasion by Salmonella (see above), and is also implicated in invasion by Shigella flexneri (Adam et al, 1995).  There are a number of suggestions that plastin is directly involved in cell signalling in a number of surprising ways. The alkylating agent bromophenacyl bromide (BPB) binding to plastin inhibits IP3 dependent Ca2+ in human neutrophils (Rosales et al, 1994) (the specificity of this reagent was shown by blotting extracts and probing with an anti-BPB monoclonal which lit up only plastin!).  Plastin is itself regulated by calcium, its bundling activity drops in its presence. Phosphorylation of plastin at serine 5 by PKA results in integrin activation in neutrophils stimulated by Fc receptor crosslinking (Wang & Brown, 1999).  At the moment these are vaguely connected facts that smell like they are telling us something about plastin’s role in phagocytosis!


Phagocytosis is the internalisation of solid particles into the cells.  The first line of defence against invading bacteria is phagocytosis by the professional phagocytes, neutrophils and macrophages, however, other cells (non-professional phagocytes) are also capable of phagocytosis.  It is increasingly clear that there are at least two types of phagocytosis involved in bacterial internalisation, each associated with its own signal transduction machinery and cytoskeletal components.  Receptor mediated phagocytosis (RMP) involve WASP and Arp2/3 (

Phagocytosis is the internalisation of solid particles into the cells.  The first line of defence against invading bacteria is phagocytosis by the professional phagocytes, neutrophils and macrophages, however, other cells (non-professional phagocytes) are also capable of phagocytosis.  It is increasingly clear that there are at least two types of phagocytosis involved in bacterial internalisation, each associated with its own signal transduction machinery and cytoskeletal components.  Receptor mediated phagocytosis (RMP) involve WASP and Arp2/3 (May et al, 2000), and non-RMP that is probably just macropinocytosis (Brumell et al, 1999).  Non-RMP involves hyper-ruffling of the cell membrane and inert latex beads as well as normally non-invasive bacteria can enter the cell while this is going on.  Non-RMP is thought to be like macropinocytosis
as clathrin coated structures are seen.  The process of both type of phagocytosis is altered by pathogenic bacteria, Yersinia, Shigella and Salmonella switch on non RMP, while EPEC secretes a toxin that triggers tyrosine dephosphorylation paralysing the phagocytic function (Goosney et al, 1999).  Normally, the phagosome produced by RMP fuses with lysosome soon after internalisation.  This releases hydrolytic enzyme into the phagolysosome to digest and destroy the contents.

Shigella & Listeria

The story for these bacteria are similar to that of Salmonella except that once inside the cell, Shigella and Listeria escape the phagolysosome by the secretion of a variety of phospholipases and pore forming cytolysins such as listeriolysin (Goebel & Kuhn, 2000)  There are important differences, Salmonella elicits a rise in intracellular calcium whereas Shigella does not (Clerc et al, 1989).  In both cases when Shigella and Listeria gain access to the host cell cytoplasm they move rapidly around the cell by a process known as “rocketing”. 

Staphylococcus aureus

This bacterium was not named after someone!  It was discovered and named by Sir Alexander Ogston a Scottish surgeon in 1880.  Its name is derived from its perceived similarity in appearance to a bunch of grapes.  Staphylococcus is a gram positive bacterium that is able to invade host cells.  This is the bacteria that the media have dubbed the “superbug”, MRSA (Methicillin resistant Staphylococcus aureus that is causing concern as patients are succumbing to infection in hospitals. Unlike the other bacteria mentioned, Staphylococcus does not have to actively produce any compound in order to invade, bacteria fixed and killed in formaldehyde are just as invasive as live bacteria.  Invasion requires fibronectin-binding protein on the surface of Staphylococcus bacteria which acts as an intimin, and the host cytoskeleton (Sinha et al, 1999).  Once inside the host cell Staphylococcus escapes the phagolysosome and then induces apoptosis in non-professional phagocytes (Bayles et al, 1998).  The mechanism by which bacteria binding to the host cell through fibronectin leads to bacterial uptake is very sketchy at the moment, but because Staphylococcus is such a threat to health and other pathogenic bacteria such as Neisseria gonorrheae and Streptococcus pyogenes are also able to invade cells by a similar mechanism, this field is soon likely to become more intensely researched.
An opportunity for medical intervention?

The many adaptations that bacteria have evolved to become commensally or pathologically associated with the eukaryotic host cell is quite intimidating!  It is thought that Salmonella adopted us as hosts as little as 1 million years ago (Mills et al, 1995).  However, all our present antibiotics are  merely agents that kill bacteria or prevent their growth.  An understanding of the complex mechanisms that bacteria employ to invade our cells presents us with new targets against which small molecule drugs can be sought by the power of combinatorial chemistry.




Adam, T., Arpin, M., Prévost, M.-C., Gounon, P. and Sansonetti , P.J. (1995). Cytoskeletal rearrangements and the functional role of T-plastin during entry of Shigella flexneri into HeLa cells. J.Cell Biol. 129; 367-381.

Bayles, K.W., Wesson, C.A., Liou, L.E., Fox, L.K., Bohach, G.A. and Trumble, W.R. (1998). Intracellular Staphylococcus aureus escapes the endosome and induces apoptosis in epithelial cells. Infect.Immun. 66, 336-342.

Boudeau, J., Glasser, A.-L., Masseret, E., Joly, B. & Darfeuille-Michaud, A. (1999). Invasive ability of an Escherichia coli strain isolated from the ileal mucosa of a patient with Crohn’s disease. Infect.Immun. 67; 4499-4509.

Bourdet-Sicard, R., Rudiger, M., Jockusch, B. M., Gounon, P., Sansonetti, P. J. & Tran Van Nhieu, G. (1999) Binding of the Shigella protein IpaA to vinculin induces F-acin depolymerization. EMBO J. 18, 5853-5862.

Brumell, J.H., Steele-Mortimer, O. & Finlay, B.B. (1999). Bacterial invasion: Force feeding by Salmonella. Current Biol. 9; R277-R280.

Clerc, P., Berthon, B., Claret, M. and Sansonetti, P.J. (1989). Internalization of Shigella flexneri into HeLa cells occurs without an increase in intracellular calcium. Infect.Immun. 57; 2919-2922.

Cooper, J.A. & Schafer, D.A. (2000). Control of actin assembly and disassembly at filament ends. Curr.Op.Cell Biol. 12; 97-103. (Much information on Arp2/3 complex and how it works with other proteins).

Cornelis, G. R. (2000) Molecular and cell biology aspects of plaque. PNAS. 97, 8778-8783.

de la Cruz, F. & Davies, J. (2000) Horizontal gene transfer and the origin of species: lessons from bacteria., Trends Microbiol. 8, 128-133

Duménil, G., Sansonetti, P. & Tran Van Nhieu, G. (2000) Src tyrosine kinase activity down-regulates Rho-dependent responses during Shigella entry into epithelial cells and stress fibre formation., J.Cell Sci. 113, 71-80.

Freeman, N. L., Zurawski, D. V., Chowrashi, P., Ayoob, J. C., Huang, L., Mittal, B., Sanger, J. M. & Sanger, J. W. (2000) Interaction of the enteropathognic Escherichia coli protein, translocated intimin receptor (Tir), with focal adhesion proteins., Cell Motility Cytoskeleton. 47, 307-318.

Galán, J.E. & Collmer (1999). Type III secretion machines: Bacterial devices for protein delivery into host cells.  Science 284; 1322-1328.

Goosney, D.L., de Grado, M. and Finlay, B.B. (1999). Putting E.coli on a pedestal: a unique system to study signal transduction and the actin cytoskeleton. Trends in Cell Biology 9;11-14.

Goosney, D.L., Celli, J., Kenny, B. & Finlay, B.B, (1999). Enteropathogenic Escherichia coli inhibits phagocytosis. Infect.Immun. 67; 490-495.

Goebel, W. and Kuhn, M. (2000). Bacterial replication in the host cell cytosol. Curr.Op.Microbiol. 3; 49-53.

Hayward, R. D. & Koronakis, V. (1999) Direct nucleation and bundling of actin by the SipC protein of invasive Salmonella., EMBO J. 18, 4926-4934.

Kalman, D., Weiner, O. D., Goosney, D. L., Sedat, J. W., Finlay, B. B., Abo, A. & Bishop, J. M. (1999) Enteropathogenic E.coli acts through WASP and Arp2/3 complex to form actin pedestals., Nature Cell Biol. 1, 389-391.

Keen, N., Staskawicz, J., Mekalanos, J., Ausubel, F. & Cook, R. J. (2000) Pathogens and hosts: The dance is the same, the couples are different.  PNAS. 97, 8752-8753.

Kerr, J.R. (1999).  Type III (contact-dependent) secretion in gram-negative bacteria. Reviews in Medical Microbiol.  10; 155-164.

Knutton, S., Rosenshine, I., Pallen, M.J., Nisan, I. Neves, B.C., Bain, C., Wolff, C., Dougan, G. and Frankel, G. (1998). A novel EspA-associated surface organelle of enteropathogenic Eschericia coli involved in protein translocation into epithelial cells.  EMBO J. 17; 2166-2176.

Lee, S. H., Hava, D. L., WAldor, M. K. & Camilli, A. (1999) Regulation and temporal expression patterns of Vibrio cholerae virulence genes during infection.  Cell. 99, 625-634.

Machesky, L.M. Atkinson, S.J., Ampe, C., Vandekerckhove, J. and Pollard, T.D. (1994). Purification of a complex containing two unconventional actins from Acanthamoeba by affinity chromatography on profilin-agarose. J.Cell Biol. 127; 107-115.

Matsudaira, P.T. and Burgess, D.R. (1979) Identification and organization of the components in the isolated microvillus cytoskeleton. J.Cell Biol. 83; 667-673.

May, R.C., Caron, E., Hall, A. and Machesky, L.M. (2000). Involvement of the Arp2/3 complex in phagocytosis mediated by FcgR or CR3. Nature Cell Biol. 2; 246-248.

May, R.C. & Machesky, L.M. (2001). Phagocytosis and the actin cytoskeleton. J.Cell Sci. 114; 1061-1077 (A recent review).

Mills, D.B., Bajaj, V. & Lee, C.A. (1995). A 40 kilobase chromosomal fragment encoding Salmonella typhimurium invasion genes is absent from the corresponding region of the Escherichia coli K-12 chromosome. Mol.Microbiol. 15, 749-759.

Mitra, K., Zhou, D. & Galan, J. E. (2000). Biophysical characterization of SipA, an actin-binding protein from Salmonella entericaFEBS letters. 482, 81-84.

Mullins, R.D. & Pollard, T.D. (1999) Structure and function of the Arp2/3 complex. Curr.Op.Cell Biol. 9; 244-249.

Mullins, R.D. (2000). How WASP-family proteins and the Arp2/3 complex convert intracellular signals into cytoskeletal structures. Curr.Op.Cell Biol. 12; 91-96. 

Mulvey, M. A. & Hultgren, S. J. (2000) Bacterial spelunkers.  Science. 289, 732-733.

Nataro, J.P. & Kaper, J.B. (1998). Diarrheagenic Escherichia coli.  Clin.Microbiol.Rev. 11; 142-201.  (A very complete account of everything you might ever want to know about when E.coli goes bad! Getting slightly dated now)

Neish, A. S., Gewirtz, A. T., Zeng, H., Young, A. N., Hobert, M. E., Karmali, V., Rao, A. S. & Madara, J. L. (2000) Prokaryotic regulation of epithelial responses y inhibition of IkB-a ubiquitination. Science. 289, 1560-1563.

Page, A.-L., Ohayon, H., Sansonetti, P. J. & Parsot, C. (1999) The secreted IpaB and IpaC invasins and their cytoplasmic chaperone IpgC are required for intercellular dissemination of Shigella flexneri. Cellular Microbiol. 1, 183-193.

Pennington, T.H. (1997). Death in Wishaw 1996- E.coli O157 and food safety. 2nd Opinion 8; 10-11. (An interesting account of the out break and its political outcome).

Pruimboom-Brees, I.M., Morgan, T.W., Ackermann, M.R., Nystrom, E.D., Samuel, J.E., Cornick, N.A. & Moon, H.W. (2000) Cattle lack vascular receptors for Escherichia coli 0157:H7 Shiga toxins. PNAS 97, 10325-10329.

Rosales, C., Jones, S.L., McCourt, D. and Brown, E.J. (1994). Bromophenacyl bromide binding to the actin-bundling protein I-plastin inhibits inositol trisphosphate-independent increase in Ca2+ in human neutrophils. PNAS. 91; 3534-3538.

Rozelle, A.L. et al, (2000) Phosphatidyinositol 4,5 –bis phosophate induces actin-based movement of raft-enriched vesicles through WASP-Arp2/3. Curr.Biol. 10; 311-320.

Shin, J.-S., Gao, Z. & Abraham, S. N. (2000) Involvement of cellular caveolae in bacterial entry into mast cells., Science. 289, 785-788.

Sinha, B., Francois , P.P., Nübe, O., Foti, M., Harford, O.M., Vaudaux, P., Foster, T.J., Lew, D.P., Herrman, M. and Krause, K.-H. (1999). Fibronectin-binding protein acts as Staphylococcus aureus invasin via fibronectin bridging to integrin a5b1. Cellular Microbiol. 1, 101-117

Sinclair, J. F. & O'Brien, A. D. (2002) Cell surface-localized nucleolin is a eukaryotic receptor for the adhesin intimin-g of enterohemorrhagic Escherichia coli 0157:H7., J.Biol.Chem. 277, 2876-2885.

Vallance, B. A. & Finlay, B. B. (2000) Exploitation of host cells by enteropathogenic Escherichia coli. PNAS. 97, 8799-8806.

Xavier, R. J. & Podolsky, D. K. (2000) How to get along - Friendly microbes in a hostile world. Science. 289, 1483-1484.

  EDInfo Biomedical Sciences Cytoskeletal Links Encyclopaedia of A.B.P.s The Amoebae Protozoology links Glossary of Amoeba terms   Maciver Lab Home