Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2010; 5(12): e15775.
Published online 2010 December 20. doi:  10.1371/journal.pone.0015775
PMCID: PMC3004962

Exit Mechanisms of the Intracellular Bacterium Ehrlichia

Deepak Kaushal, Editor



The obligately intracellular bacterium Ehrlichia chaffeensis that resides in mononuclear phagocytes is the causative agent of human monocytotropic ehrlichiosis. Ehrlichia muris and Ixodes ovatus Ehrlichia (IOE) are agents of mouse models of ehrlichiosis. The mechanism by which Ehrlichia are transported from an infected host cell to a non-infected cell has not been demonstrated.

Methodology/Principal Findings

Using fluorescence microscopy and transmission and scanning electron microscopy, we demonstrated that Ehrlichia was transported through the filopodia of macrophages during early stages of infection. If host cells were not present in the vicinity of an Ehrlichia-infected cell, the leading edge of the filopodium formed a fan-shaped structure filled with the pathogen. Formation of filopodia in the host macrophages was inhibited by cytochalasin D and ehrlichial transport were prevented due to the absence of filopodia formation. At late stages of infection the host cell membrane was ruptured, and the bacteria were released.


Ehrlichia are transported through the host cell filopodium during initial stages of infection, but are released by host cell membrane rupture during later stages of infection.


The obligately intracellular bacterium Ehrlichia chaffeensis that resides in mononuclear phagocytes is the etiologic agent of human monocytotropic ehrlichiosis (HME). HME is an emerging and often life-threatening tick-transmitted infectious disease in the United States [1]. Ehrlichia are round or ovoid gram negative bacteria, and form a characteristic vacuole-contained microcolony (morula) in macrophages [2]. HME was first reported in 1987 [3]. Since then, development of murine models of persistent and lethal ehrlichiosis has greatly facilitated understanding of the pathogenesis and mechanisms of host defenses against ehrlichial infections. In general microorganisms can disseminate after host cell lysis via necrotic or apoptotic cell death, or by spreading from cell-to-cell [4]. The mechanism by which Ehrlichia are released from host cells has not been demonstrated [5][8].

Recently, in a mouse model of monocytotropic ehrlichiosis, we demonstrated by eastern blotting that the heat shock protein 60 (Hsp60/GroEL) is highly post-translationally modified in E. muris, which is not virulent in immunocompetent mice compared to the highly virulent strain IOE (Ixodes ovatus Ehrlichia) [9]. Based on this observation we generated an anti-Ehrlichia specific Hsp60 antibody and used it to observe E. chaffeensis, E. muris or IOE in cell culture. In this study we demonstrated by microscopic techniques the modes by which Ehrlichia exited the host cells.


Ehrlichia are associated with the filopodia of infected DH82 cells

The intracellular pathogens E. chaffeensis and E. muris are maintained in vitro in the DH82 monocyte cell line. Previously Ehrlichia have been studied after infection of DH82 cells at high concentrations and culturing for 2–3 days when the host cells formed a confluent monolayer (no void between host cells). We observed Ehrlichia after seeding 1000–2000 infected DH82 cells per slide so that they were separated from one another (16 hours). E. muris- and E. chaffeensis-infected DH82 cells were probed with the anti-Hsp60 antibody. By 16 hours filopodia were observed in 30 percent of DH82 cells infected with Ehrlichia (3 percent in uninfected DH82 cells; p<0.0001). Filopodia extended from the polar ends of spindle-shaped Ehrlichia-infected host cells (E. chaffeensis: Fig. 1A, E. muris: Fig. 2A–D; uninfected DH82 cell: Fig. 1D, ,2E;2E; E. muris-infected DH82 cell without primary antibody: Fig. 2F) or from the non-polar sides of the cells when they contained many bacteria (E. chaffeensis: Fig. 1B). Filopodia of infected cells extended to the neighboring host cell (E. chaffeensis: Fig. 1B, E. muris: Fig. 2D). If host cells were not present in the vicinity of an infected cell, the leading edge of the filopodia of Ehrlichia-infected cells formed a flattened fan-shaped structure where the Ehrlichia morulae were contained (Fig. 1C). We have also observed that the morula-filled fan-shaped structure further developed its own filopodium (not shown). Ehrlichia infection in DH82 cells was confirmed using the Diff-Quik stain (Fig. 1F) and transmission electron microscopy (Fig. 1G).

Figure 1
Ehrlichia are contained in the filopodia of DH82 cells.
Figure 2
E. muris is associated with the filopodia of DH82 cells.

Macrophage filopodia contain a meshwork of actin filaments and surround foreign organisms during phagocytosis [10]; cytochalasin D inhibits actin polymerization [11]. As phalloidin has a high affinity for actin, we used phalloidin conjugated to Alexa 594 for detection of actin in the filopodia. Filopodia stained with phalloidin-Alexa 594 were intensely red, whereas DAPI stained the host nucleus as well as the DNA of E. chaffeensis (Fig. 3A–D, uninfected DH82 cell: Fig. 3E). Filopodia formation was observed within an hour after culturing the infected DH82 cells. By 24 hours the average length of a filopodium in infected cells was 120 micrometers (Fig. 3F) (NS, p>0.05). We have observed filopodia in cell culture measuring more than 10 times longer than the diameter of the host cell.

Figure 3
Actin was a major protein of filopodia induced during Ehrlichia chaffeensis infection.

Scanning electron microscopy of the mechanically opened cells demonstrated the presence of Ehrlichia in the filopodia of the DH82 host cells (Fig. 4Aa–f). On contact with a new cell, the pathogens from the fan-shaped flattened structure were in a location where they can pass to the neighboring cell (Fig. 4Ae–f). Further, observation of cell membranes deformed from within by intracellular ehrlichiae revealed the opportunity for bacterial intrusion into the adjacent cells (Fig. 4Ag–i). These observations suggest that Ehrlichia passed from one host cell to another without entering the extracellular space. To detect actin in the filopodium induced by E. muris, the infected DH82 cells were treated with anti-actin antibody. The concentration of actin was high in the filopodia of the infected DH82 cells (Fig. 4B), whereas a filopodium was rarely observed in the uninfected control cells (Fig. 4C).

Figure 4
Ehrlichia is transported through the filopodia of the host cells.

Inhibition of actin polymerization in DH82 cells infected with Ehrlichia prevented filopodia formation followed by the localization of the pathogen in the periphery of macrophages

Macrophages characteristically migrate and extend pseudopodia to assume an amoeboid conformation, whereas non-activated monocytes lack such processes and appear round [12]. Since actin and microtubules are involved in the formation of filopodia, we determined the effect of the actin inhibitor, cytochalasin D on the transport of the pathogen in Ehrlichia-infected monocytes. Cytochalasin D inhibited filopodium formation in both E. chaffeensis- (Fig. 5A) and E. muris- infected cells (Fig. 5C, D, F, G). Ehrlichiae were confined to the periphery of the macrophages. The experiments suggest the hypothesis that Ehrlichia are transported through the filopodia, a potential mechanism to avoid the host immune system while the pathogen passed from cell to cell.

Figure 5
Cytochalasin D inhibited filopodium formation in Ehrlichia-infected cells.

Hybiske and Stephens [13] used latrunculin B (actin polymerization inhibitor), wiskostatin (N-WASP inhibitor) and blebbistatin (myosin II inhibitor) to demonstrate that actin, N-WASP and myosin are required for the non-lytic exit mechanism of the bacterium Chlamydia. Results from our studies also showed that latrunculin B, wiskostatin and blebbistatin inhibited filopodium formation; whereas treatment of E. muris-infected DH82 cells with nocodazole (microtubule inhibitor) did not inhibit filopodium formation (Fig. 6A–E).

Figure 6
Actin inhibition prevented filopodium formation in E. muris-infected DH82 cells.

Based on the findings that Ehrlichia are contained in the filopodium of DH82 cells, we hypothesized that filopodium formation is essential for intercellular transport of Ehrlichia. To test this hypothesis, we labeled E. muris-infected DH82 cells with carboxyfluorescein succinimidyl ester (CFSE) and seeded the infected cell culture with non-labeled uninfected DH82 cells. Alternately, we also labeled uninfected DH82 cells with CFSE and seeded with DH82 cells infected with E. muris. We observed filopodia/pseudopodia from the infected DH82 cells in close proximity to the neighboring uninfected DH82 cells and also observed Ehrlichia in the uninfected DH82 cells (Fig. 7A–C). In the presence of cytochalasin D we did not observe filopodia in the Ehrlichia-infected DH82 cells, and there was no infection in the neighboring uninfected cells (Fig. 7D–E). If Ehrlichia indeed are transported through the filopodia, we reasoned that the bacterial load will decrease in the presence of cytochalasin D when both infected and uninfected DH82 cells are seeded together as the pathogen cannot be transported to the uninfected cell. To test the hypothesis, we seeded uninfected DH82 cells with E. muris-infected DH82 cells (in the presence and absence of cytochalasin D) and analyzed the bacterial load by quantitative real time–polymerase chain reaction (RT-PCR) after 24 hours. Initial studies demonstrated that cytochalasin D was not toxic to Ehrlichia (Fig. 7F). When the uninfected DH82 cells were seeded with E. muris-infected DH82 cells in the presence of cytochalasin D, the bacterial load decreased, whereas in the absence of cytochalasin D the bacterial load increased (Fig. 7G).

Figure 7
Inhibition of filopodium formation prevented Ehrlichia intercellular transport.

Ehrlichia are contained within the filopodia of infected mouse macrophages

Lacking a technique to demonstrate the mechanism of Ehrlichia transport between cells in vivo, we cultured in vitro for five days splenocytes of mice that had been infected for 7 days prior to harvesting the cells. The cytoplasm of the macrophages from E. muris-infected mice harbored few pathogens on days 1–3 in cell culture, whereas by day 5 the macrophages were highly populated with E. muris (Fig. 8A, C–F, and M). E. muris-infected mouse macrophages had filopodia that contained the pathogen similar to those observed in the infected DH82 cells. Similar results were observed when macrophages from the highly lethal Ixodes ovatus Ehrlichia (IOE)-infected mice (infected for 7 days prior to harvesting) were cultured for five days in vitro. IOE were also observed in the filopodia of infected mouse macrophages (Fig. 8B, G and H).

Figure 8
Ehrlichia are observed in the filopodia of mouse macrophages.

Ehrlichia infection led to a ruptured overlying host cell membrane at a late stage of infection

We cultured E. muris in DH82 cells for 60 hours on coverslips and observed specimens under a scanning electron microscope. At 60 hours the size of the morula enlarged probably due to fusion of adjacent morulae (Fig. 9B; Morulae of E. muris at 24 hours: Fig. 9A). The cell membrane of E. muris infected DH82 cell ruptured at 60 hours (Fig. 9C) and the bacteria were released through the pores on the host cell membrane (Fig. 9D). The pathogens released after membrane rupture were observed attached to the filopodium of neighboring cells (Fig. 9E). Attached ehrlichiae were observed in association with ruffled cell membrane characteristic of entry by endocytosis (Fig. 9F). TEM of IOE infected spleen confirmed morula fusion (Fig. 9G).

Figure 9
Ehrlichia morulae inside a DH82 cell with an overlying ruptured host cell membrane.


Successful establishment of infection by obligately intracellular bacterial pathogens requires adhesion to host cells, and cellular invasion followed by intracellular multiplication, dissemination to other tissues, or persistence [14]. An understanding of the specific pathways of pathogen exit is of fundamental importance to microbial pathogenesis because of its intimate association with dissemination, transmission and inflammation. Microbial exit is an organized and directed process mediated by both bacterial and cellular factors. Shigella and Listeria promote their escape from phagosomes through the action of pore-forming cytolysins: IpaB for Shigella, listeriolysin O and C-type phospholipases for Listeria. Cellular release then occurs as these bacteria use actin polymerization to protrude out of the cell [13]. Hybiske and Stephens [13] demonstrated that the intracellular bacteria, Chlamydia, are released by two mutually exclusive pathways-lysis and extrusion. The pathogenic Mycobacteria are also released by multiple exit mechanisms. Lasunskaia et al. [15] showed that Mycobacteria induce filopodia formation in macrophages. Recent studies by Hagedorn et al. [4] showed that Mycobacteria are released from the host cell by an actin-based ejectosome. The ejectosome was released into the phagocytic cup of the neighboring host cell.

The mechanism by which Ehrlichia exits the cell or is transported between cells is not known [7], [8]. In general, microorganisms are released by a lytic or non-lytic mechanism to infect neighboring host cells. This investigation demonstrated mechanisms by which Ehrlichia exited host cells. Our results based on microscopic analyses showed that Ehrlichia were contained in the filopodium of macrophages which were transported to neighboring cells. An advantage of Ehrlichia transport through the filopodia (in the initial stages) is that the pathogen evades the host immune system while it passed from cell to cell.

Filopodia are thin cell surface protrusions containing bundles of parallel actin filaments. Filopodia are designed for exploring the extracellular matrix and surfaces of other cells, identifying targets of adhesion, or navigating to its appropriate target. In epithelial cells they act as a zipper to fuse to one another [16]. Filopodia of the host cells have been shown to be utilized for cellular entry and exit by pathogenic microorganisms [17]. Bacteria also induce filopodia-like structures to enhance the cohesion of the bacterial microcolonies and therefore blood vessel infection under the harsh conditions of the bloodstream [18]. Listeria grow in the cytoplasm of macrophages and recruit and polymerize host cell actin, which provides the driving force for movement through the cytoplasm and into nearby cells by means of filopodia-like projections [19]. We have not observed ejectosomes or phagocytic cups as observed in Mycobacteria-infected host cells or a pinching off mechanism as observed in Chlamydia-infected host cells. We observed by electron microscopy Ehrlichia-containing cells with an extension to the neighboring cell mediated by the filopodia. Adhesion of one host cell to another mediated through the filopodium may be a mechanism of transport of Ehrlichia. Future investigations will focus on how ehrlichial proteins influence the host cell actin to induce filopodium elongation.

Once the host cell contains a large number of the bacterial pathogen, the host cell may not have a functional system which can induce actin polymerization for filopodium formation. Hence the only mechanism for the bacteria to exit the host cell is by lysing the host cell membrane. Blouin and Kocan [20] demonstrated the exit mechanism in Anaplasma. The authors showed that Anaplasma are released by membrane rupture of the host cell without apparent loss of host cell cytoplasm. Perforin-like proteins induce pore formation of host cells and are expressed by many bacterial and protozoan pathogens. Kafsack et al. [21] had demonstrated that Toxoplasma gondii secretes a perforin-like protein which disrupts the host cell membrane facilitating its exit. Perforin-like proteins are not documented in Ehrlichia; future studies may reveal if a similar protein is associated with the pathogen. Finlay and Falkow [22] had demonstrated that the actin cytoskeleton but not the microtubule network plays an active role in bacterial entry into host cells. SEM results (Fig. 9E) demonstrated that Ehrlichia have high affinity to the actin cytoskeleton. Based on our observation that phalloidin binds to the actin-rich filopodium, we hypothesize that the actin cytoskeleton may play a major role in Ehrlichia entry into host cells. Future study will determine the ehrlichial protein which binds actin cytoskeleton.

Materials and Methods

Cell culture

Three monocytotropic ehrlichial strains were used in this study. The human pathogen, Ehrlichia chaffeensis (a gift from Dr. Jere McBride, UTMB, Texas), and mildly virulent E. muris (provided by Dr. Y. Rikihisa, Ohio State University, Columbus, OH) were cultivated in DH82 cells [9] at 37°C in DMEM supplemented with 5% heat-inactivated bovine calf serum. Ehrlichiae were harvested when approximately 90 to 100% of the cells were observed to be infected. Infected cells were scraped, and 1000–2000 infected cells were cultured for 16 hours on culture slides. The non-culturable IOE was maintained in mice. The animals were sacrificed after 7 days and the spleen used as source of the pathogen.

Fluorescence microscopy

Hsp60 (GroEL) peptides were synthesized by Biosynthesis, Inc. (Lewisville, Texas) and injected (i.p. - four times) into C57BL/6 mice. Antibody was obtained 40 days after the first injection. After fixation in 50% methanol-acetone, both E. muris- and E. chaffeensis-infected DH82 cells were incubated with the anti- Ehrlichia Hsp60 antibody (1[ratio]125) (45 min), and after further washes they were reacted with anti-mouse immunoglobulin G conjugated to Alexa 488. After several washes they were mounted in mounting medium containing DAPI (Vectashield, Vector Labs, Burlingame, CA).

Phalloidin has high affinity for actin. Alexa 594-conjugated phalloidin (Invitrogen, CA) was incubated with E. chaffeensis-infected or uninfected DH82 cells following the manufacturer's instructions. Alternatively, for detection of actin, rabbit anti-actin antibody (1[ratio]100) (Sigma, St. Louis, MO) was used for staining E. muris-infected or uninfected cells followed by treatment with goat anti-rabbit TRITC (1[ratio]400). After several washes specimens were mounted in medium containing DAPI (Vectashield, Vector Labs, Burlingame, CA).

For filopodium inhibition studies, 30 minutes after transfer of Ehrlichia-infected DH82 cells to culture slides they were treated with cytochalasin D (0.5 micrograms/ml) (Fisher, Fairlawn, NJ). Cells were washed after 16 hours and fixed and stained as described above. Experiments were repeated three times. The cells were viewed by epifluorescence microscopy (Olympus BX51, Japan). For inhibition studies using the inhibitors lantrunculinB, wiskostatin, nocodazole, and blebbistatin, the method of Hybiske and Stephens [13] was followed. To determine that filopodia are essential for transport of Ehrlichia we treated infected or uninfected DH82 cells with CFSE (Invitrogen, CA) for 15 minutes following the manufacturer's instructions and seeded in the presence or absence of cytochalasin D. The cells were incubated for 24 hours and stained with Ehrlichia HSP60 antibody, further stained with anti-mouse Alexa 594, and mounted in medium containing DAPI (Vectashield, Vector Labs, Burlingame, CA).

Determination of ehrlichial copy numbers in infected cells

Ehrlichial copy numbers in infected DH82 cells seeded with uninfected DH82 cells (in the presence or absence of cytochalsin D for 24 hours) were determined by quantitative real-time PCR method by analyzing the dsb gene [23].

Electron microscopy

For transmission electron microscopy (TEM), the monolayers (Ehrlichia-infected cells) in T25 flasks were fixed in 2.5% paraformaldehyde and 0.1% glutaraldehyde in 0.05 M cacodylate buffer, pH 7.3, to which 0.03% trinitrophenol and 0.03% CaCl2 were added. After fixation the cells were washed with 0.1 M cacodylate buffer, scraped off the flasks and pelleted. The pellets were post-fixed in 1% OsO4 in 0.1 M cacodylate buffer, en bloc stained with 1% uranyl acetate in 0.1 M maleate buffer, dehydrated in ethanol and embedded in Poly/Bed 812 (Polysciences, Warrington, PA). Ultrathin sections were cut on Reichert-Leica Ultracut S ultramicrotome, stained with lead citrate and examined in a Philips 201 or CM-100 electron microscope at 60 kV.

For scanning electron microscopy (SEM), Ehrlichia-infected DH82 cells were cultured on Thermanox coverslips (Nunc, Rochester, NY) and were fixed similarly as for TEM, dehydrated in ethanol and processed through hexamethyldisilazane followed by air drying and mounting on stubs. The cells were mechanically opened with a touch of scotch tape (3M, St. Paul, MN) to expose the interior of the cells. Specimens were sputter-coated with iridium and observed with a Hitachi S4700 (Japan) scanning electron microscope.

In vitro studies of mouse macrophages

E. muris (provided by Dr Y. Rikihisa, Ohio State University, Columbus, OH) or IOE (a gift from Dr M. Kawahara, Nagoya City Public Health Research Institute, Nagoya, Japan) were inoculated (i.p) into five week old C57/BL6 mice, and the animals were sacrificed after 7 days. The spleen was harvested, and the splenocytes cultured on a slide or flask in DMEM medium for 5 days. The infected or uninfected mouse cells were processed for epifluorescence microscopy, TEM and SEM. The mice were housed and cared for in the Animal Research Center at the University of Texas Medical Branch in accordance with the Institutional Animal Care and Use Committee guidelines under whose review and approval the experiments were conducted.


The experiments were performed in triplicate, and each figure is the representative of approximately 15–20 images.


We thank Dr. Jere W. McBride for providing E. chaffeensis, Jeeba A. Kuriakose for the technical assistance, and Julie Wen for electron microscopic expertise. We are also thankful for the technical suggestions by Nagaraja Thirumalapura and Fang Rong for improving the manuscript.


Competing Interests: The authors have declared that no competing interests exist.

Funding: This research was supported by a grant from the National Institute of Allergy and Infectious Diseases (AI031431). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


1. Paddock CD, Childs JE. Ehrlichia chaffeensis: a prototypical emerging pathogen. Clin Microbiol Rev. 2003;16:37–64. [PMC free article] [PubMed]
2. Popov VL, Han VC, Chen SM, Dumler JS, Feng HM, et al. Ultrastructural differentiation of the genogroups in the genus Ehrlichia. J Med Microbiol. 1998;47:235–251. [PubMed]
3. Maeda K, Markowitz N, Hawley RC, Ristic M, Cox D, et al. Human infection with Ehrlichia canis, a leukocytic rickettsia. N Engl J Med. 1987;316:853–856. [PubMed]
4. Hagedorn M, Rohde KH, Russell DG, Soldati T. Infection by tubercular mycobacteria is spread by nonlytic ejection from their amoeba hosts. Science. 2009;323:1729–1733. [PMC free article] [PubMed]
5. Li JS, Winslow GM. Survival, replication, and antibody susceptibility of Ehrlichia chaffeensis outside of host cells. Infect Immun. 2003;71:4229–4237. [PMC free article] [PubMed]
6. Maender JL, Tyring SK. Treatment and prevention of rickettsial and ehrlichial infections. Dermatol Ther. 2004;17:499–504. [PubMed]
7. Rikihisa Y. Molecular events involved in cellular invasion by Ehrlichia chaffeensis and Anaplasma phagocytophilum. Vet Parasitol. 2010a;167:155–166. [PMC free article] [PubMed]
8. Rikihisa Y. Anaplasma phagocytophilum and Ehrlichia chaffeensis: subversive manipulators of host cells. Nat Rev Microbiol. 2010b;8:328–339. [PubMed]
9. Thomas S, Thirumalapura N, Crossley EC, Ismail N, Walker DH. Antigenic protein modifications in Ehrlichia. Parasite lmmunol. 2009;31:296–303. [PMC free article] [PubMed]
10. Hartwig JH, Davies WA, Stossel TP. Evidence for contractile protein translocation in macrophage spreading, phagocytosis, and phagolysosome formation. J Cell Biol. 1977;75:956–967. [PMC free article] [PubMed]
11. Rosania GR, Swanson JA. Microtubules can modulate pseudopod activity from a distance inside macrophages. Cell Motil Cytoskeleton. 1996;34:230–245. [PubMed]
12. Srivastava KD, Magazine HI. Thrombin receptor activation inhibits monocyte spreading by induction of ET(B) receptor-coupled nitric oxide release. J Immunol. 1998;161:5039–5044. [PubMed]
13. Hybiske K, Stephens RS. Mechanisms of host cell exit by the intracellular bacterium Chlamydia. Proc Natl Acad Sci USA. 2007;104:11430–11435. [PubMed]
14. Pizarro-Cerda J, Cossart P. Bacterial adhesion and entry into host cells. Cell. 2006;124:715–727. [PubMed]
15. Lasunskaia EB, Campos MN, de Andrade MR, Damatta RA, Kipnis TL, et al. Mycobacteria directly induce cytoskeletal rearrangements for macrophage spreading and polarization through TLR2-dependent PI3K signaling. J Leukoc Biol. 2006;80:1480–1490. [PubMed]
16. Wood W, Martin P. Structures in focus-filopodia. Int J Biochem Cell Biol. 2002;34:726–730. [PubMed]
17. Van Nhieu GT, Caron E, Hall A, Sansonetti PJ. IpaC induces actin polymerization and filopodia formation during Shigella entry into epithelial cells. EMBO J 18: 1999:3249–3262. [PubMed]
18. Mikaty G, Soyer M, Mairey E, Henry N, Dyer D, et al. Extracellular bacterial pathogen induces host cell surface reorganization to resist shear stress. PLoS Pathog. 2009;5:e1000314. [PMC free article] [PubMed]
19. Tilney LG, Portnoy DA. Actin filaments and the growth, movement, and spread of the intracellular bacterial parasite, Listeria monocytogenes. J Cell Biol. 1989;109:1597–1608. [PMC free article] [PubMed]
20. Blouin EF, Kocan KM. Morphology and development of Anaplasma marginale (Rickettsiales: Anaplasmataceae) in cultured Ixodes scapularis (Acari: Ixodidae) cells. J Med Entomol. 1998;35:788–797. [PubMed]
21. Kafsack BF, Pena JD, Coppens I, Ravindran S, Boothroyd JC, et al. Rapid membrane disruption by a perforin-like protein facilitates parasite exit from host cells. Science. 2009;323:530–533. [PMC free article] [PubMed]
22. Finlay BB, Falkow S. Comparison of the invasion strategies used by Salmonella cholerae-suis, Shigella flexneri and Yersinia enterocolitica to enter cultured animal cells: endosome acidification is not required for bacterial invasion or intracellular replication. Biochimie. 1988;70:1089–1099. [PubMed]
23. Stevenson HL, Jordan JM, Peerwani Z, Wang HQ, Walker DH, et al. An intradermal environment promotes a protective type-1 response against lethal systemic monocytotropic ehrlichial infection. Infect Immun. 2006;74:4856–4864. [PMC free article] [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science