Listeriolysin O activates Listeria monocytogenes internalization into human hepatocytes
through a novel pore-dependent mechanism
Dissertation
Presented in partial fulfillment of the requirements for the degree doctor of philosophy in
the Graduate School of The Ohio State University
By
Stephen Vadia, M.S.
Graduate Program in Microbiology
The Ohio State University
2014
Dissertation Committee:
Dr. Stephanie Seveau, Advisor
Dr. Robert Munson
Dr. Larry Schlesinger
Dr. Susheela Tridandapani
Copyright by
Stephen Vadia
2014
Abstract
The causative agent of listeriosis, Listeria monocytogenes, is a frequent contaminant of fruits, vegetables, cheeses, and processed foods. In certain high risk groups, L. monocytogenes can cause serious and potentially fatal infections. The bacterium can traverse the intestinal epithelial barrier and spread throughout the body via the bloodstream and lymphatic system. From there, it can infect the liver, cross the blood- brain barrier, and in pregnant women, the placental barrier. L. monocytogenes is able to cross these biological barriers due to its ability to proliferate within certain phagocytic and nonphagocytic cells. Bacterial invasins, most notably internalin (InlA) and InlB, bind to receptors on the surface of nonphagocytic cells to induce endocytosis of the bacterium through a zipper-like mechanism. L. monocytogenes can then escape from the endocytic vacuole and enter the cytosol through the activity of the pore-forming toxin listeriolysin
O (LLO). In the cytosol, the bacterium can replicate and disseminate to adjacent cells by cell-to-cell spreading.
LLO belongs to the cholesterol-dependent cytolysin (CDC) family of bacterial pore-forming toxins. These toxins are secreted as monomers, oligomerize on cholesterol- rich membranes, and form large transmembrane pores. In addition to mediating vacuolar escape, LLO stimulates a number of signaling pathways in host cells. We hypothesized
ii
that the signaling activity of extracellular LLO affects the intracellular lifecycle of L. monocytogenes. Using quantitative immunofluorescence microscopy to assess the importance of LLO for bacterial association to and internalization into epithelial cells, we found that LLO is required for efficient L. monocytogenes internalization into hepatocytes, and is sufficient to induce internalization of noninvasive bacteria and polystyrene beads. Using novel LLO variants that bind to host cells but are unable to form the pore complex, we demonstrated that pore formation was required for LLO- mediated internalization.
The CDC streptolysin O is known to stimulate a Ca2+-dependent membrane repair response that involves endocytosis of the toxin from the perforated plasma membrane.
We found that LLO stimulates a similar Ca2+-dependent membrane repair response, but that this response is not sufficient to account for L. monocytogenes internalization. LLO- mediated internalization and membrane repair share the requirement for the influx of extracellular Ca2+, but differ in their requirement for F-actin and K+ efflux. Surprisingly, using ionophores to stimulate Ca2+ and K+ fluxes, as happens when the plasma membrane is perforated by LLO, was sufficient to induce internalization of large particles. LLO mediates bacterial internalization in all human hepatocytes we tested and in additional epithelial cell lines. However, LLO does not induce bacterial internalization in all epithelial cells that it can perforate.
In light of this data, LLO emerges as a significant invasion factor expressed by L. monocytogenes, which is coexpressed with InlA and InlB during infection. With the same approaches used to assess the role of LLO, we also determined the contribution of InlA and InlB to bacterial association, internalization, and intracellular viability in iii
nonphagocytic cells that express the InlA and InlB receptors. We found significant variability in the involvement of each of these invasins in bacterial association, internalization, and intracellular viability, demonstrating that the activity of invasins is not governed solely by the presence or absence of their receptors.
iv
Dedication
This dissertation is dedicated to my mom for her neverending support and her encouragement to pursue a career in science. I would never have started or finished this
journey without her there to help me along the way. And to my Grandma and Grandpa
Iacobucci, for believing in me at times when I did not believe in myself.
v
Acknowledgements
First and foremost, I would like to thank my advisor, Dr. Stephanie Seveau, for her guidance, mentorship, and patience with me during my graduate career. I would also like to acknowledge my committee members, Dr. Robert Munson, Dr. Susheela
Tridandapani, and Dr. Larry Schlesinger, for their comments and insights during committee meetings, and for recommendations which have supported me greatly in securing funding while at Ohio State and during my job search.
Former lab members have helped immensely with this project. Anne-Cecile
Haghighat began this work before I joined the lab, trained me as an incoming graduate student, and with the help of Andrew Bruggeman, constructed LLO variants used in this project. Dr. Eusondia Arnett performed invaluable experimental work for these studies during particularly busy times.
This work also benefitted greatly from the help of many collaborators here at
Ohio State University. I would like to thank Dr. Prosper Boyaka for his immunological expertise and for performing mouse immunizations. From the Boyaka lab, Astrid
Bonnegarde-Bernard, and Mike Fial, provided critical hands on assistance with mouse infections. We were also greatly assisted in these experiments by Dr. Abhay Satoskar, as well as Dr. Steve Oghumu and Sanjay Varikuti from the Satoskar lab. Dr. John Robinson
vi
and Dr. William Ackerman provided protocols, advice, cell lines, and placental villi.
Brian Kemmenoe at the OSU Campus Microscopy and Imaging Facility provided precious help with scanning electron microscopy. Dr. Elizabeth Wilson-Kubalek (The
Scripps Research Institute, La Jolla, CA) performed transmission electron microscopy and provided us with wonderful images. I also wish to thank Dr. Daniel Portnoy
(University of California, Berkeley, CA), Dr. Pascale Cossart (Pasteur Institute, Paris,
France), Dr. Rodney Tweten (University of Oklahoma Health Sciences Center,
Oklahoma City, OK), and Dr. Nancy Freitag (University of Illinois at Chicago, USA) for generously providing us with bacterial strains and plasmids.
I also must thank Dr. Chad Rappleye and members of his lab, especially Dr. Eric
Holbrook and Dr. Jessica Edwards, for all of their advice over the years, as well as Dr.
Mary Anne Rubio and Dr. Juan Alfonzo for support, recommendations, and the occasional pep talk.
vii
Vita
2003...... B.A. Anthropology
Case Western Reserve University
2007 to 2008...... University Fellowship
The Ohio State University
2008 to 2012...... Graduate Teaching Associate
Department of Microbiology
The Ohio State University
2010...... M.S. Microbiology
The Ohio State University
2012 to 2013...... Presidential Fellowship
The Ohio State University
2013 to present...... Graduate Teaching Associate
Department of Microbiology
The Ohio State University
viii
Publications
Vadia S, Arnett E, Haghighat AC, Wilson-Kubalek EM, Tweten RK, and Seveau S.
2011. The pore-forming toxin listeriolysin O mediates a novel entry pathway of L. monocytogenes into human hepatocytes. PLoS Pathogens, 7:e1002356. Chapter 2.
Vadia S, Boyaka PN, Arnett E, Satoskar AR, Tweten RK, and Seveau S. 2013.
Immunization with a listeriolysin O toxoid protects mice against Listeria monocytogenes.
(Manuscript submitted for publication). Chapter 3.
Vadia, S and Seveau, S. 2013. Fluxes of Ca2+ and K+ are required for the LLO-dependent internalization pathway of Listeria monocytogenes. (Manuscript submitted for publication). Chapter 4.
Arnett E, Vadia S, Nackerman CC, Oghumu S, Satoskar A, McLeish KR, Uriarte SM, and Seveau S. 2013. The pore-forming toxin listeriolysin O is degraded by neutrophil proteases and fails to protect L. monocytogenes against intracellular killing. (The Journal of Immunology. In press).
ix
Vadia S, Arnett E, and Seveau S. Assessing the roles of LLO, InlA, and InlB during the
L. monocytogenes intracellular lifecycle in hepatocytes and placental cells (Manuscript in preparation). Chapter 5.
Fields of Study
Major Field: Microbiology
x
Table of Contents
Abstract...... ii
Dedication...... v
Acknowledgements...... vi
Vita...... viii
Table of Contents...... xi
List of Tables...... xviii
List of Figures...... xix
Chapter 1. Introduction...... 1
1.1 Listeria monocytogenes...... 1
1.1.1 Listeriosis...... 3
1.1.2 The L. monocytogenes intracellular lifecycle...... 4
1.1.3 The virulence factors of L. monocytogenes...... 5
1.1.3.1 The adhesins...... 6
1.1.3.2 The invasins...... 6
1.1.3.3 Intracellular growth and spread...... 11
1.1.3.4 The cholesterol-dependent cytolysin listeriolysin O...... 12
1.1.3.5 Regulation of listeriolysin O...... 13
xi
1.1.3.6 Listeriolysin O is critical for vacuolar escape...... 14
1.2 L. monocytogenes crosses critical barriers in the host...... 15
1.2.1 The intestinal barrier...... 16
1.2.2 The placental barrier...... 17
1.2.3 The blood-brain barrier...... 19
1.3 Host cell responses to pore-forming toxins...... 20
1.3.1 Signaling responses to pore-forming toxins...... 20
1.3.2 Plasma membrane repair...... 22
1.3.3 Listeriolysin O as a signaling molecule...... 24
1.4 Research aims and significance...... 26
Chapter 2. The pore-forming toxin listeriolysin O mediates a novel entry pathway of
Listeria monocytogenes into human hepatocytes...... 28
2.1 Introduction...... 28
2.2 Materials and Methods...... 30
2.2.1 Bacterial strains and plasmids...... 30
2.2.2 Mammalian cell culture...... 31
2.2.3 Coating of bacteria and polystyrene beads with LLO...... 32
2.2.4 Gentamicin survival assay...... 32
2.2.5 Measurement of bacterial association and entry into host cells...... 33
2.2.6 Construction and purification of recombinant toxins...... 34
2.2.7 Hemolytic assays...... 34
2.2.8 Western blotting analysis of LLO expression by the bacterial strains...... 35
2.2.9 HepG2 perforation assay...... 36 xii
2.2.10 LDH and cell viability measurements...... 36
2.2.11 Toxin oligomerization (NativePAGE)...... 37
2.2.12 Toxin oligomerization (transmission electron microscopy)...... 37
2.2.13 Invasion assays using polystyrene beads. F-actin and EEA1 labeling...... 37
2.2.14 Silencing of clathrin heavy chain in HepG2 cells...... 38
2.2.15 Cell treatment with chemical inhibitors...... 38
2.2.16 Transferrin uptake...... 39
2.2.17 Cholesterol depletion and repletion, and LLO treatment with neutralizing antibodies...... 39
2.2.18 Image acquisition...... 40
2.2.19 Live cell imaging...... 40
2.2.20 Scanning electron microscopy...... 41
2.2.21 Statistics...... 41
2.3 Results...... 43
2.3.1 LLO is critical for efficient enty of L. monocytogenes into host cells...... 43
2.3.2 Direct and dose-dependent activity of LLO in L. monocytogenes entry into host cells...... 45
2.3.3 LLO is sufficient to induce bacterial and bead entry into host cells...... 47
2.3.4 LLO perforates host cells at physiological temperature and pH...... 50
2.3.5 Construction and characterization of novel LLO variants unable to form pores...... 54
xiii
2.3.6 LLO-induced bacterial and bead entry requires host cell membrane
perforation...... 56
2.3.7 LLO activates a clathrin-independent, but dynamin-dependent internalization
pathway...... 59
2.3.8 F-actin remodeling is induced by LLO in a pore-dependent fashion and is
required for bead internalization...... 64
2.4 Discussion...... 65
Chapter 3. Immunization with a listeriolysin O toxoid protects mice against Listeria monocytogenes...... 72
3.1 Introduction...... 72
3.2 Materials and Methods...... 73
3.2.1 Generation of LLOT...... 73
3.2.2 Cholesterol binding assay...... 74
3.2.3 Hemolysis assays...... 74
3.2.4 Bacterial and mammalian cell culture...... 75
3.2.5 Mouse immunization and infection...... 75
3.2.6 L. monocytogenes internalization assay...... 75
3.3 Results...... 76
3.3.1 Generation of a listeriolysin O toxoid (LLOT)...... 76
3.3.2 Immunization with LLOT plus adjuvant protects mice against L.
monocytogenes...... 77
3.3.3 LLO neutralizing Abs inhibit cell invasion...... 78
3.4 Discussion...... 80 xiv
Chapter 4. Fluxes of Ca2+ and K+ are required for the LLO-dependent internalization pathway of Listeria monocytogenes...... 83
4.1 Introduction...... 83
4.2 Materials and Methods...... 86
4.2.1 Bacterial and mammalian cell culture...... 86
4.2.2 Toxin purification and polystyrene bead coating...... 87
4.2.3 Bacterial association with and internalization into host cells...... 87
4.2.4 Invasion assay using polystyrene beads...... 88
4.2.5 Propidium iodide incorporation assays...... 89
4.2.6 Microscope equipment...... 89
4.2.7 Statistics...... 89
4.3 Results...... 90
4.3.1 Extracellular L. monocytogenes perforates host cells in a LLO-dependent
manner...... 90
4.3.2 Host cells damaged by LLO reseal their plasma membrane via a Ca2+-
dependent but K+-independent mechanism...... 92
4.3.3 Host cells damaged by L. monocytogenes undergo Ca2+-dependent
membrane resealing...... 92
4.3.4 L. monocytogenes internalization into host cells requires Ca2+ and K+
fluxes...... 94
4.3.5 Ca2+ influx and K+ efflux induced by ionophores are sufficient to activate the
internalization of large cargoes...... 97 xv
4.4 Discussion...... 99
Chapter 5. Assessing the roles of LLO, InlA, and InlB during the L. monocytogenes intracellular lifecycle in hepatocytes and placental cells...... 104
5.1 Introduction...... 104
5.2 Materials and methods...... 105
5.2.1 Bacterial strains and plasmids...... 105
5.2.2 Mammalian cell culture...... 106
5.2.3 BeWo cell fusion...... 106
5.2.4 Measurement of bacterial association and entry into host cells...... 107
5.2.5 Isolation and infection of placental villi...... 108
5.2.6 Gentamicin survival assays...... 108
5.2.7 Propidium iodide incorporation assays...... 109
5.3 Results...... 110
5.3.1 LLO-mediated internalization is not restricted to HepG2 cells...... 110
5.3.2 LLO is not required for L. monocytogenes internalization into BeWo
cells...... 111
5.3.3 LLO is critical for infection of human placental villous explants...... 113
5.3.4 Roles of InlA and InlB in L. monocytogenes association and entry into human
hepatocytes and placental cells...... 116
5.3.5 Measuring the contribution of LLO and InlB to intracellular survival in
HepG2 and BeWo cells...... 118
5.4 Discussion...... 120 xvi
Chapter 6. Synthesis and future goals...... 123
6.1 LLO-mediated internalization is mechanistically distinct from plasma
membrane repair...... 123
6.2 Assessing the role of LLO during bacterial internalization in vivo...... 124
6.3 LLO-mediated internalization as a paradigm of host cell invasion...... 125
6.4 Investigating the host cell range of LLO-mediated internalization...... 126
6.5 Variability in the activity of L. monocytogenes invasion factors...... 127
xvii
List of Tables
Table 1. Primer and siRNA sequences used in this study...... 42
Table 2. Serum IgG antibody responses in mice immunized with LLOT...... 78
xviii
List of Figures
Figure 1.1 Human listeriosis...... 3
Figure 1.2 The intracellular lifecycle of L. monocytogenes...... 5
Figure 1.3 InlA binds E-cadherin to stimulate L. monocytogenes internalization...... 8
Figure 1.4 InlB binds c-Met to stimulate L. monocytogenes internalization...... 10
Figure 1.5 The structure of cholesterol-dependent cytolysins...... 13
Figure 2.1 LLO is required for efficient entry of L. monocytogenes into HepG2 cells..... 44
Figure 2.2 L. monocytogenes coating with recombinant LLO...... 46
Figure 2.3 Direct and dose-dependent role of LLO in L. monocytogenes entry...... 47
Figure 2.4 LLO is sufficient to induce the entry of noninvasive L. innocua into HepG2 cells...... 49
Figure 2.5 LLO is sufficient to induce entry of polystyrene beads into HepG2 cells...... 50
Figure 2.6 LLO-coated beads do not affect HepG2 cell viability...... 51
Figure 2.7 Extracellular LLO perforates erythrocytes and HepG2 cells at 37°C, pH 7.4...... 53
Figure 2.8 Characterization of the LLO variants...... 55
Figure 2.9 Formation of pore complexes is required for efficient bacterial and bead entry into HepG2 cells...... 58
xix
Figure 2.10 LLO-coated beads form small pores in HepG2 cells...... 59
Figure 2.11 PLY induces L. monocytogenes entry into HepG2 cells...... 60
Figure 2.12 LLO-coated beads are internalized into EEA1 positive endosomes...... 61
Figure 2.13 LLO-coated beads are internalized by a clathrin-independent, dynamin-, F- actin-, and tyrosine kinase-dependent pathway...... 63
Figure 2.14 F-actin, dynamin, and clathrin are dispensable for the membrane repair pathway...... 65
Figure 3.1 Characterization of LLOT...... 77
Figure 3.2 LLOT protects mice against L. monocytogenes...... 79
Figure 3.3 LLO neutralizing Abs inhibit L. monocytogenes internalization into hepatocytes...... 80
Figure 4.1 Extracellular L. monocytogenes perforates host cells in a LLO-dependent manner...... 91
Figure 4.2 Host cells damaged by LLO undergo Ca2+-dependent, but K+- independent
membrane resealing...... 93
Figure 4.3 Host cells damaged by L. monocytogenes undergo Ca2+-dependent membrane resealing...... 94
Figure 4.4 L. monocytogenes internalization into host cells requires Ca2+ and K+ fluxes...... 96
Figure 4.5 Ca2+ influx and K+ efflux induced by ionophores are sufficient to activate the internalization of large cargoes...... 98
Figure 5.1 LLO is required for efficient entry of L. monocytogenes into hepatocytes.... 111
Figure 5.2 LLO is required for efficient entry of L. monocytogenes into HeLa cells..... 112 xx
Figure 5.3 BeWo cells are perforated by LLO and undergo Ca2+-dependent membrane resealing...... 113
Figure 5.4 LLO is not required for efficient entry of L. monocytogenes into BeWo cells...... 114
Figure 5.5 LLO is critical for infection of human placental villous explants...... 115
Figure 5.6 Roles of InlA and InlB in L. monocytogenes entry into hepatocytes...... 117
Figure 5.7 The role of InlA and InlB for entry of L. monocytogenes into placental cells...... 118
Figure 5.8 The role of LLO and InlB for intracellular survival of L. monocytogenes..... 119
xxi
Chapter 1. Introduction
1.1 Listeria monocytogenes
The genus Listeria is comprised of eight species of Gram-positive bacilli, two of which are pathogenic and possess similar intracellular lifecycles, but have very different host ranges [1]. Listeria ivanovii is an intracellular pathogen that mainly affects ruminants [2], while Listeria monocytogenes has a far more extensive host range and is the cause of a potentially fatal foodborne disease in humans [2]. L. monocytogenes is widespread in the environment and is frequently found in soil or decaying plant matter, where it is believed to grow as a saprophyte [3, 4]. Contaminated silage is a common source of infection for cows and sheep, from which the bacterium may be spread to humans through the food chain. Once it encounters a susceptible host, L. monocytogenes transitions from a saprophytic to a pathogenic lifestyle [3].
L. monocytogenes was originally identified as a pathogen in 1924, when E.G.D.
Murray isolated the infectious agent — which he originally named Bacterium monocytosis — from diseased rabbits and guinea pigs [5]. Infection with L. monocytogenes was rarely described in the medical literature until 1949, when its serious pathogenic potential in humans was finally appreciated following an outbreak affecting
1
85 infants in Germany [6, 7]. Termed "Granulomatosis infantiseptica," the disease was caused by a motile bacterium that had infiltrated the livers, spleens, and brains of those infected [8]. Incorrectly identified at the time as a species of Corynebacterium, it was soon after recognized as an outbreak of L. monocytogenes [9].
L. monocytogenes outbreaks in humans remained uncommon until the late
1970s, when the first in a series of epidemics was recognized in North America and
Europe, including an outbreak in Novia Scotia caused by contaminated coleslaw that resulted in over 40 cases of listeriosis and nearly 20 deaths [10-12]. This outbreak was significant because it marked the first time that investigators unequivocally identified contaminated food products as the primary method of transmission to humans [10].
Foods most often implicated in the spread of L. monocytogenes include raw fruits and vegetables, unpasteurized dairy products, lunch meats, and many processed ready-to- eat foods. The bacterium is a particular problem for food manufacturers because it can persist in biofilms on many surfaces in processing plants, where it can contaminate food products. L. monocytogenes may then multiply during the time between processing and consumption due to its ability to grow under conditions often used for food preservation.
With an aging population, and the advent of corticosteriods, chemotherapeutics, organ transplantation, and the emergence of HIV, the pool of those susceptible to L. monocytogenes is increasing in size [13-15]. Despite growing awareness and stringent guidelines for food preparation and storage designed to limit exposure, sporadic L. monocytogenes outbreaks still occur. Most recently, outbreaks of listeriosis in the UnitedStates in 2011, 2012, and 2013, linked to cantaloupes and cheeses, lead to illness in 175 individuals and 38 deaths [16-18]
2
1.1.1 Listeriosis
Foodborne pathogens are a major cause of illness in the United States [19].
While the incidence of listeriosis is quite low compared to some other foodborne infections, the mortality rate is among the highest at 20-50% [20, 21]. Surveillance studies for Listeria in food and in the human gastrointestinal tract indicate that exposure is not uncommon, however, healthy adults typically remain uninfected or only develop mild flu-like symptoms [21, 22]. The illness is far more serious in patients who develop invasive listeriosis, which can lead to serious medical conditions including meningoencephalitis, endocarditis, hepatitis, and liver abscesses.
Figure 1.1 Human listeriosis. L. monocytogenes is acquired through contaminated food. It may then cross the intestinal barrier and traffic through the portal vein, lymphatic system, and bloodstream to the liver and spleen. If infection is not controlled at this point, bacteria can disseminate further, and potentially cross the fetoplacental and blood-brain barriers.
The severity of infection depends in large part on the immune status of the host.
Several high-risk groups including immunocompromised individuals and the elderly
3
are at a much higher risk than the general population for developing invasive listeriosis. The disease is particularly dangerous for pregnant women, who are twenty times more likely to become infected compared to the general population [23]. While mothers typically experience only mild gastroenteritis or flu-like symptoms, infection may cause abortion or stillbirth, and leads to death in 20-30% of the infants infected in utero [22].
1.1.2 The L. monocytogenes intracellular lifecycle
The ability of L. monocytogenes to cause disease is the result of a sophisticated intracellular lifestyle which facilitates spread of the bacterium throughout the body, while avoiding extracellular host defenses. L. monocytogenes may survive within professional phagocytes such as macrophages, and also invade nonphagocytic cells, including epithelial, endothelial and fibroblastic cells. Bacterial surface proteins, most notably internalin (InlA) and InlB, bind host cell receptors to induce bacterial invasion of nonphagocytic cells [24, 25]. The bacterium may then escape from the endocytic vacuole through the activity of a pore-forming toxin listeriolysin O (LLO). Two phospholipases — a broad-range phospholipase C (PC-
PLC) and a phosphatidylinositol specific phospholipase C (PI-PLC) — facilitate escape. In the cytosol, the bacterial protein ActA polymerizes host cell actin, which generates the force to propel the bacterium through the cytosol and mediates cell-to- cell spreading [26, 27]. Following cell-to-cell spread, L. monocytogenes escapes from the secondary vacuole, again through the activity of LLO and the phospholipases, and continues its spread through the host [28].
4
InlA InlB 1 2 LLO LLO PC-PLC 6 PC-PLC PI-PLC ActA PI-PLC 3 4 5
Figure 1.2 The intracellular lifecycle of L. monocytogenes. L. monocytogenes expresses surface associated invasion factors that bind to receptors on the host cell (1), and induce bacterial internalization (2). Intracellular bacteria escape from the endocytic vacuole and enter the cytosol (3), where they divide (4), polymerize F- actin, and undergo cell-to-cell spreading (5). In the newly infected cell, L. monocytogenes escapes from the double-membrane vacuole and continues the infectious cycle (6).
1.1.3 The virulence factors of L. monocytogenes
Many of the critical virulence factors of L. monocytogenes were discovered by several labs that were screening transposon mutant libraries for virulence defects.
Most of these factors cluster together in two regions of the chromosome: A pathogenicity island (Listeria pathogenicity island - 1; LIPI-1) consisting of six genes that are critical for the Listeria intracellular lifecycle, and the internalin operon, composed of two genes — inlA and inlB — that encode factors responsible for bacterial entry into nonphagocytic cells. More recently, additional virulence determinants have been identified using comparative genomics of L. monocytogenes with nonpathogenic Listeria species, and by advances in transcriptional profiling [29-
34].
5
1.1.3.1 The adhesins
The ability to attach to host cells is the first step of bacterial internalization.
Strong attachment to host surfaces is critical for enteroinvasive pathogens like
Listeria, which must resist mechanical clearing from the gastrointestinal tract.
Furthermore, attachment to surface receptors can trigger signaling events that result in bacterial uptake. L. monocytogenes expresses an array of molecules that mediate attachment to the host. Internalin (InlA) is a major adhesin that mediates attachment to cells expressing E-cadherin, a Ca2+-dependent cell-cell adhesion molecule present in the adherens junctions of polarized epithelial cells. An array of additional proteins has been implicated in adhesion of L. monocytogenes to host cells. The listeria adhesion protein (LAP) enhances attachment to intestinal epithelial cells by binding to
Hsp60 and is critical for translocation [35, 36]. Listeria adhesion protein B (LapB) binds an unidentified receptor to mediate attachment to multiple nonphagocytic cell types [37]. Other internalins (InlB, InlC, and InlJ) have been implicated in binding to
MUC2, the primary component of the intestinal mucus layer [38, 39], while fibronectin binding protein A (FbpA) and ActA mediate binding to components of the extracellular matrix (fibronectin and heparan sulfate, respectively) [40, 41]. Other factors, including Ami, DltA, and p60, modify the bacterial cell wall, and may affect binding directly or by modifying the structure, orientation, or availability of other adhesins [42-45]. The cysteine transport associated protein (CtaP) has also been implicated in bacterial adherence.
1.1.3.2 The invasins
In order to establish an infection, L. monocytogenes must cross protective barriers in the host, a feat that it achieves in large part by actively invading 6
nonphagocytic cells. InlA and InlB were the first two factors determined to be sufficient for invasion of epithelial cells by L. monocytogenes [46, 47]. Both proteins belong to the internalin family, which are abundant in the L. monocytogenes genome
[48]. The internalins contain a series of N-terminal leucine rich repeats (LRRs), and are categorized according to their cell wall anchoring mechanism. The majority are anchored to the cell wall, through a LPXTG motif. The other groups are noncovalently associated via GW repeats, or lack apparent cell wall association motifs [49].
InlA is covalently anchored to the bacterial cell wall through a C-terminal
LPXTG motif, and interacts with E-cadherin via its N-terminus. This N-terminal region, consisting of LRRs and an inter-repeat (IR) region, is both necessary and sufficient to induce entry [25]. The engagement of InlA and E-cadherin, and the clustering of E-cadherin require and occur within lipid rafts [50]. The downstream signaling cascade leads to ubiquitination of E-cadherin and activation of the endocytic machinery of the cell. InlA activates Src kinase, which phosphorylates the cytoplasmic domain of E-cadherin, displacing the receptor stabilizing molecule p120, and leading to the recruitment of the ubiquitin-ligase Hakai, ubiquitination of E- cadherin, and internalization of the receptor [51, 52]. Additionally, E-cadherin is linked to the actin cytoskeleton through interactions with β-catenin and α-catenin
[53]. InlA binding stimulates a localized increase in the concentration of myosin VIIA and vezatin through interactions formed with α-catenin, which functions as an adaptor protein for the recruitment of these downstream molecules that mediate cytoskeletal rearrangements [54]. Actin rearrangements that occur during bacterial internalization are driven by the Arp2/3 complex, which is recruited by the actin binding protein cortactin [55]. InlA-mediated internalization is partially dependent on clathrin, which 7
is recruited along with caveolin [56], a molecule that facilitates the clustering of E- cadherin at the site of entry [50, 52].
Figure 1.3 InlA binds E-cadherin to stimulate L. monocytogenes internalization. InlA binding to E-cadherin activates a signaling cascade within lipid raft microdomains. E-cadherin is phosphorylated by Src kinase, leading to recruitment of the ubiquitin ligase Hakai and clathrin. The cytoplasmic portion of E-cadherin directly interacts with β-catenin, which recruits α-catenin that is directly interacting with actin. Vezatin links the plasma membrane to myosinVIIA, which provides force to promote internalization. Actin polymerization is promoted by the Arp2/3 complex, which is activated by cortactin, downstream of Src.
8
InlB binds to multiple receptors on the host cell surface that cooperate to promote efficient bacterial entry, including the receptor for the globular domains of complement component C1q (gC1qR), glycosaminoglycans (GAGs), and its primary receptor, the hepatocyte growth factor receptor c-Met [57-59]. The receptor c-Met, is expressed by a broader range of cells than E-cadherin. Binding of c-Met by InlB co- opts the growth factor receptor signaling response leading to activation of the host ubiquitination and F-actin polymerization machineries [60, 61]. InlB contains a series of C-terminal GW repeats that mediate the non-covalent interaction between the protein and lipoteichoic acids in the cell wall, and that bind to host cell GAGs [59, 62,
63]. Because it is only non-covalently associated with the cell wall, some InlB detaches from the bacterial surface. This soluble InlB has been shown to bind host
GAGs to promote InlB clustering and activation of c-Met [59, 64, 65]. Similar to the
InlA/E-cadherin signaling cascade, InlB/c-Met signaling requires the integrity of, and occurs within lipid rafts [50, 66]. Lipid rafts are not required for the clustering of c-
Met, but are critical for downstream signaling events leading to Rac1 activation [50].
Interaction of InlB with c-Met stimulates dimerization and autophosphorylation of the receptor tyrosine kinase [66]. InlB also stimulates PI 4-kinase, PI 3-kinase and the phosphorylation and recruitment of the adaptor proteins Gab1, Cbl, Shc, and CrkII to form signaling complexes that promote L. monocytogenes entry [60, 67-69].
Downstream of PI 3-kinase, signaling molecules controlling F-actin polymerization including Rac1 [50], the Arp 2/3 complex [70] Cdc42, N-WASP and WAVE
[71]regulate bacterial endocytosis.
9
Figure 1.4 InlB binds c-Met to stimulate L. monocytogenes internalization. Cortactin activates the Arp 2/3 complex to polymerize F-actin. Actin polymerization is also controlled by PI 3-kinase through a signaling cascade involving Rac1, Cdc42, WAVE, and N-WASP. Phagocytic cup closure occurs after actin depolymerization, a process controlled by LIM kinase and cofilin. Endocytosis involves clathrin and dynamin. Adaptor molecules, Cbl, Gab1, Shc, and CrkII form signaling complexes, and the downstream signaling activity is dependent upon lipid rafts.
In addition to InlA and InlB, there are some less well characterized factors that mediate invasion of nonphagocytic cells. Some of these factors, including the autolysin Auto, were discovered by searching for putative surface proteins that are present in L. monocytogenes but absent from L. innocua [72]. Vip, LapB, the 10
membrane lipoprotein LpeA, GtcA, which glycosylates teichoic acid, and Auto, all enhance entry of L. monocytogenes into various nonphagocytic cell lines in vitro, and are important for virulence in vivo. Finally, the motility protein ActA can also enhance attachment and entry into some phagocytic and nonphagocytic cell types and is sufficient to mediate entry of L. innocua. [41, 73].
1.1.3.3 Intracellular growth and spread
Successful intracellular pathogens must be able to replicate in the intracellular environment and disseminate through the host. Within a cell, L. monocytogenes is shielded from extracellular defense mechanisms. However, it still faces challenges in the intracellular environment, including the need to acquire nutrients from the host, spread to adjacent cells, and evade intracellular defense mechanisms.
Intracellular growth of L. monocytogenes is quite rapid, with a doubling time of about 45 minutes [74]. The inability of some bacteria to replicate after microinjection directly into the cytosol of mammalian cells indicates that pathogens like L. monocytogenes are adapted for intracellular survival [75]. However, very little is known about how L. monocytogenes grows in the cytosol. The hexose phosphate transporter (Hpt) controls cytosolic replication and is critical for virulence [76]. Hpt is only present in pathogenic Listeria species, is upregulated in the host cell cytosol, and is under the control of the master regulator of virulence genes, PrfA [33, 77]. Deletion of hpt leads to a marked decrease in the replication rate of intracellular L. monocytogenes and attenuation in mice [76].
Once L. monocytogenes has invaded the host cell cytosol, the polar protein
ActA recruits the actin nucleating Arp2/3 complex, which stimulates the formation of
F-actin rich 'comet' tails. This mechanism propels L. monocytogenes through the 11
cytosol and mediates spread via the formation of membrane protrusions that are taken up by neighboring cells. This process is facilitated by InlC, a secreted internalin that is highly expressed in the intracellular environment [78]. InlC is believed to facilitate cell-to-cell spreading through interactions with actin binding proteins. InlC deficient strains are still able to form F-actin tails, but are less efficient at cell-to-cell spread
[79].
ActA also functions to limit autophagy of L. monocytogenes by recruiting host proteins that shield the bacterium from ubiquitination and subsequent degradation [80,
81]. This function is shared by InlK, an internalin that is found in L. monocytogenes but is absent from L. innocua [82]. InlK is expressed at very low levels in vitro, but is induced in mammalian hosts, where it recruits the Major Vault Protein (MVP), which masks Listeria from autophagy [82, 83].
1.1.3.4 The cholesterol-dependent cytolysin listeriolysin O
LLO belongs to the cholesterol-dependent cytolysin (CDC) family of toxins, which are produced by over 30 different Gram-positive and 2 Gram-negative bacteria, and are often critical for pathogenesis [84, 85]. There is no crystal structure of LLO but detailed structural information is available for other CDCs, which has aided in determining the complex structural rearrangements that take place during pore formation [86-91]. Many details of this mechanism have been worked out for two model CDCs - perfringolysin O (PFO), and pneumolysin (PLY) [92].
CDCs are secreted as monomers that bind to cholesterol containing membranes, oligomerize, then form a large (25-30 nm diameter) transmembrane β- barrel pore [84]. CDCs have a characteristic 4 domain structure (Fig. 1.5). A highly conserved Thr-Leu pair in domain 4 was recently identified as the cholesterol binding 12
motif, which is the membrane receptor for most CDCs. [93]. The few CDCs that bind to the protein receptor CD59 still require cholesterol for pore formation [94-96].
Binding of the monomer to cholesterol stimulates a conformational change in domain
3, allowing for interactions between adjacent monomers that promote toxin oligomerization and the formation of a large pre-pore complex composed of 30-50 monomers [97-100]. Pore formation occurs following the conversion of α-helices in domain 3 to amphipathic β-hairpins, and a vertical collapse of the membrane bound pre-pore complex into the host membrane [101].
PFO LLO
D1
TMH1 G80C D3 D2 I359C S213C
K344C TMH2
D4 Thr-Leu Loops 1-3
Figure 1.5 The structure of cholesterol dependent cytolysins. Ribbon diagrams of PFO (PDB 1PFO; [86]) and a model of LLO based on the PFO structure. The four domains are labeled D1-D4. In red are TMH1 and TMH2, the two α-helices that unfurl to form transmembrane β-hairpins. Loops 1-3 are in blue. The model of LLO shows the locations of Cys mutations in the LLOml (I359C and K344C) and LLOpl (G80C and S213C) variants in yellow. The Thr-Leu pair mutated in LLOT is indicated in green. Structures were generated with PyMol.
1.1.3.5 Regulation of listeriolysin O
While LLO is necessary for efficient vacuolar escape and intracellular survival, excessive LLO activity will destroy the replicative niche of L.
13
monocytogenes. Therefore, the expression and activity of the toxin must be tightly controlled. The production of LLO is regulated at multiple levels. LLO transcription is regulated by PrfA, a master regulator of L. monocytogenes virulence genes, which functions as a thermosensor [102-105]. At temperatures ≤ 30°C, the 5' UTR of the prfA transcript forms a hairpin structure that conceals the ribosome binding site. At
37°C, this hairpin is destabilized, allowing for translation [106]. At the translational level, the N-terminal sequence functions to repress LLO translation during growth in the host cytosol [107].
Additional post-translational mechanisms serve to limit the activity of LLO intracellularly. L. monocytogenes expressing the CDCs anthrolysin O (ALO), PFO, or ivanolysin O (ILO) were able to escape from endocytic vacuoles, but were cytotoxic and unable to complement virulence in vivo, suggesting that there were critical regulatory differences between LLO and other CDCs [108-110]. One of the unique features of LLO is its pH optimum [111]. At temperatures ˃30°C, LLO monomers display a tendency to denature and aggregate at pH 7.4. This pH-dependent aggregation is regulated by a trio of acidic residues in domain 3 — D208, E247, and
D320 [112]. The acidic pH optimum can also be altered by a single amino acid substitution (L461T) resulting in increased activity at neutral pH and a decrease in virulence in vivo [74].In the host cell cytosol, LLO is ubiquitinated and degraded by the host proteosome via the N-end rule pathway [113, 114].
1.1.3.6 Listeriolysin O is critical for vacuolar escape
One of the most significant host defenses encountered by an intracellular pathogen is the degradative environment of an endocytic vacuole, from which L. monocytogenes is able to escape. [27, 115, 116]. LLO is necessary for escape from 14
the primary and secondary vacuoles in many cell types, and two phospholipases —
PC-PLC and PI-PLC — cooperate with LLO to induce vacuolar escape [27, 28, 117].
Interestingly, LLO is not required for escape from primary or secondary vacuoles in human epithelial cells, where PC-PLC alone can mediate low levels of escape [118,
119]. While bacterial factors necessary for escape have been identified, the precise mechanism of vacuolar disruption is still unknown.
L. monocytogenes delays vacuolar maturation in order to limit exposure to the bactericidal contents of lysosomes. Initial perforation of the vacuole by LLO leads to the equilibration of pH and Ca2+ across the membrane, which in turn delays the recruitment of LAMP-1 [120, 121]. While limiting lysosomal fusion with the vacuole benefits the bacteria, some low level of lysosomal fusion also appears to facilitate escape. LLO is activated by the lysosomal enzyme γ-interferon-inducible lysosomal thiol reductase (GILT) [122, 123]. GILT-deficient mice are resistant to infection, and
L. monocytogenes does not replicate efficiently within GILT -/- macrophages [124].
Other host factors, including phospholipase D (PLD), protein kinase C-β (PKC-β), and the cystic fibrosis transmembrane conductance regulator (CFTR) Cl- channel also facilitate L. monocytogenes escape [125]. The effect of CFTR on escape was pH- independent, suggesting that CFTR regulated vacuolar dissolution by a novel mechanism leading to enhanced oligomerization of LLO.
1.2 L. monocytogenes crosses critical barriers in the host.
The variety of clinical presentations associated with listeriosis results from the ability of L. monocytogenes to bypass multiple defensive barriers in the host. While it has long been known that InlA and InlB are important for mediating the invasion of L.
15
monocytogenes into nonphagocytic cell lines in vitro, the precise roles of these virulence factors in vivo have been more difficult to assess due to their species specificity. InlA interacts with human and guinea pig E-cadherin, but does not recognize mouse or rat E-cadherin. InlB activates human and mouse c-Met, but not c-
Met expressed by guinea pigs. Proper evaluation of the roles for both invasion factors in vivo necessitated the development of new animal models and of mutant L. monocytogenes strains engineered to expand their host range [126-128].
1.2.1 The intestinal barrier
In order to cause systemic disease, L. monocytogenes must first cross the intestinal epithelial barrier. InlA is important for entry into enterocytes in vitro [129,
130], but as a result of the species specificity of the InlA/E-cadherin interaction, the importance of InlA for crossing the intestinal barrier was difficult to assess in vivo.
This was achieved by developing a transgenic mouse with enterocytes expressing human E-cadherin [126]. These mice were permissive to oral infection with L. monocytogenes, which could efficiently traverse the intestine and disseminate to the liver and spleen, firmly establishing the importance of InlA for crossing the intestinal barrier.
E-cadherin is present in adherens junctions, which are located near the basolateral side of the epithelium and below tight junctions, raising the question of how InlA accesses its receptor. Infection of polarized MDCK cells demonstrated that
L. monocytogenes preferentially binds muticellular junctions, where E-cadherin is transiently exposed during the release of senescent cells from the epithelium [131].
Infection of rabbit ileal loops suggested that a similar mechanism occurs in vivo, where L. monocytogenes was targeted to extruding cells at the tip of intestinal villi 16
[131]. Infection of a knock-in mouse expressing humanized E-cadherin (E16P) revealed a more general role for the InlA/E-cadherin interaction during infection
[127]. In addition to enterocytes at the tip of intestinal villi, L. monocytogenes bound most often to goblet cells, which express E-cadherin that is accessible from the intestinal lumen, and to folds in the villi epithelium [132]. L. monocytogenes entering through goblet cells did not escape from the vacuole, but instead crossed the barrier by transcytosis in an ActA and LLO-independent manner [132].
Studies concerning the role of InlB provided contradictory results. While InlB was shown to play no role in intestinal translocation in humanized transgenic E- cadherin mice, experiments with L. monocytogenes expressing a murinized InlA
(InlAm) indicated that there was cooperation between InlA and InlB during intestinal translocation, with InlA providing specificity for binding while InlB functions to enhance the rate of internalization [133]. More recently, close study of cell tropism of the InlAm strain has shown that bacteria expressing InlAm also bind N-cadherin, cross the intestinal barrier though M cells, and cause excessive inflammation and intestinal damage not typically caused by L. monocytogenes. These peculiarities of the InlAm strain may account for some of the contradictory results between studies examining the mechanism of intestinal translocation using strains expressing InlAm versus those using wild-type L. monocytogenes [134].
1.2.2 The placental barrier
After crossing the intestinal epithelium, L. monocytogenes may spread to other tissues via the lymphatic system and bloodstream. This is a major issue for pregnant women, who are particularly susceptible to L. monocytogenes, possibly due to reduced cell-mediated immunity during pregnancy [135]. Infection usually leads to a 17
self-limiting flu-like illness in mothers, but puts the fetus or neonate at extremely high risk [136].
The surface of the placenta is covered in villi, which either anchor the placenta into the maternal decidua, or are floating in the intervillous space and bathed in maternal blood. Histological analysis of infected placentas revealed bacteria within the intervillous space, as well as within the syncytiotrophoblast, which is a large, terminally differentiated, multinucleated cell that forms the outermost layer of the placental villi. Bacteria have also been identified within cytotrophoblasts, which underlie, and fuse to form the syncytiotrophoblast, or invade extravillous trophoblasts that anchor the villi to the decidua [137].
Analysis of clinical isolates suggested that InlA is critical for crossing the placental barrier, as 100% of the strains isolated from placental infections expressed full-length InlA [129]. Furthermore, E-cadherin is expressed on both the syncytiotrophoblast and cytotrophoblasts, and InlA expression was critical for invasion of human placental cell lines and placental explants [137]. A report analyzing infections of first trimester villi confirmed the importance of InlA for placental invasion, but stressed the relative impermeability of the syncytiotrophoblast, instead suggesting that L. monocytogenes preferentially crosses the placental barrier by infecting invasive extravillous trophoblasts, which efficiently killed intracellular L. monocytogenes [138, 139]. Infection of a knock-in mouse expressing humanized E- cadherin throughout the body confirmed the importance of the InlA/E-cadherin interaction in vivo, and also demonstrated the importance of InlB for crossing the placental barrier [127]. This study emphasized the importance of developing appropriate animal models where both the InlA and InlB pathways are functional, because the function of either internalin during infection may not be detected unless 18
both pathways are functional [140, 141]. The virulence factors ActA and LLO are also critical for fetal infection. LLO-deficient bacteria can survive within placental tissue, but are unable to replicate, while ActA-deficient bacteria form infectious foci, but are unable to disseminate [141].
1.2.3 The blood-brain barrier
The mechanisms used by L. monocytogenes to cross the blood-brain barrier and infect the central nervous system (CNS) are not well understood. Two non- mutually exclusive mechanisms of CNS invasion have been proposed: That circulating bacteria bind to host receptors to induce entry, similar to what occurs when L. monocytogenes crosses the intestinal and placental barriers, or that bacteria are carried across the barrier by infected leukocytes [142, 143].
Cells expressing both E-cadherin, c-Met, and Gp96 (the receptor for the invasion factor Vip) are present at the blood-brain barrier, including on brain microvascular endothelial cells (HBMECs), raising the possibility that L. moncytogenes exploits these receptors to cause CNS infections [144, 145]. In support of this hypothesis, L. monocytogenes is capable of invading endothelial cells in vivo, and in cell culture, active invasion of human umbilical vein endothelial cells
(HUVECs) and HBMECs occurs through the InlB pathway [146-149].
L. monocytogenes can be transmitted by phagocytic cells to secondary infectious foci, including the brain [150-153]. The bone marrow provides a protective niche for L. monocytogenes to survive in mice, where they infect Ly-6Chigh monocytes which carry the majority of the intracellular bacteria in the bloodstream [154, 155] .
In vivo evidence for the importance of this second mechanism of CNS infection was provided when it was discovered that bacteria disseminate to the brain in mice treated 19
with gentamicin to kill bacteria in the bloodstream [156, 157]. CNS infection was also more efficient when mice were injected i.v. with L. monocytogenes infected leukocytes than after injection of free bacteria [155, 158]. L. monocytogenes may also enhance the trafficking of phagocytes to the brain. In endothelial cells, LLO induces
NFκB translocation, and the production of ICAM-1 and E-selectin, providing a mechanism to increase adhesion of infected phagocytes to endothelial cells at the blood-brain barrier, and monocyte chemotactic protein-1 (MCP-1), which can enhance migration of infected leukocytes [151, 159]. In vitro, bacteria can spread from infected phagocytes to neurons, HBMECs, and HUVECs [147, 149, 160, 161].
Inflammation caused by the recruitment of neutrophils and the production of TNF-α and IL-8 can affect the structural integrity of the blood-brain barrier, which may also enhance CNS infection [162].
1.3 Host cell responses to pore-forming toxins
Pore-forming toxins (PFTs) constitute a major class of virulence factors produced by eukaryotic, viral, and bacterial pathogens. In general, PFTs are secreted as soluble monomers, oligomerize on host membrane, and form pores of variable size.
At high concentrations PFTs cause lysis, but at sublytic concentrations they activate cellular responses that function to defend against infection and restore homeostasis.
1.3.1 Signaling responses to pore-forming toxins
Many PFTs stimulate mitogen-activated protein kinase (MAPK) signaling.
The extracellular signal-regulated kinases (ERKs), c-Jun amino-terminal kinases
(JNKs), and p38 MAPKs regulate cellular processes in response to environmental signals, including cell division, cell migration, and cell death [163]. All three families 20
of MAPKs are activated in response to PFTs [164-168]. p38 MAPK is important for defense against many PFTs and regulates a pathway that is critical for c. elegans survival and for resealing of mammalian cells after exposure to the Cry5B toxin
[166]. However, the importance of p38 signaling for survival is not universal. While p38 is important for cell survival after exposure to S. aureus α-toxin, it is not required by cells exposed to CDCs [165, 169]. Immune responses to PFTs, both pro- and anti- inflammatory, are dependent on MAPK signaling. CDCs are known to stimulate
MAPK-dependent secretion of IL-8 [170], TNF-α [171], IL-1β [172], and IL-10 [173] in epithelial cells, monocytes and macrophages. Numerous PFTs including CDCs, aerolysin, and S. aureus α-toxin can also trigger increased cytokine expression by activating caspase-1, or NFκB [174-178].
A number of pathways are involved in promoting cell survival after challenge with PFT. The unfolded protein response (UPR) is induced by PFTs to ensure that protein synthesis and protein folding are coordinated in order to maintain homeostasis
[179]. The importance of this pathway for defense against PFTs was demonstrated in vivo with c. elegans, in which the loss of transducers that activate the UPR leads to hypersensitivity to Cry5B. This protective response also required p38 MAPK signaling [164, 179]. Multiple metabolic changes occur in perforated cells, and it is thought that a metabolic transition to a less active state promotes cell survival during times of stress [164]. Exposure to aerolysin or CDCs results in a significant decrease in protein synthesis, the induction of autophagy, the formation of lipid droplets that act as lipid storage organelles, and the activation of lipid metabolic pathways [164].
In addition to its role in the maturation of inflammatory cytokines, caspase-1, which is activated by aerolysin, also activates the sterol regulatory element binding proteins
(SREBPs) that regulate lipid metabolism [174]. This response is thought to enhance 21
cell survival, possibly by promoting membrane repair. Many of the above responses occur as a result of the loss of intracellular K+, which appears to function as a master regulator of multiple pathways that promote cell survival [164].
In addition to direct effects of the toxin, the CDC SLO can also function to translocate an effector molecule into host cells. CDCs form pores large enough for the passage of small proteins, and have been used by cell biologists for just that purpose
[180, 181]. During infection by Steptococcus pyogenes, the bacterium uses SLO to translocate the cytotoxic NAD+ glycohydrolase SPN [182, 183]. Interestingly, pore formation is not required for SPN translocation, but is required for cytotoxicity [184].
1.3.2 Plasma membrane repair
Under normal physiological conditions, many cells, including those lining the gastrointestinal tract and skin, or cells in muscle tissue, are often subjected to mechanical disruption [185, 186]. Cells possess sophisticated mechanisms to repair this mechanical damage, as well as damage caused by PFTs [181]. The repair process is dependent upon extracellular Ca2+ and exocytosis of cytoplasmic membrane, and occurs in secretory as well as nonsecretory cells [187-189].
Damaged portions of the plasma membrane may be shed via blebbing, or ectocytosis, which occurs when the plasma membrane is detached from the underlying actin cytoskeleton. While it is sometimes an indicator of apoptosis, blebbing can also facilitate mobility and cell division, and has recently been shown to promote cell survival. The CDC SLO induces Ca2+ and myosin driven blebbing, which was shown to protect against cell death [190]. Toxin treated cells visualized by deep-etch electron microscopy showed that blebs shed from the membrane are highly enriched in SLO pores [191]. Blebbing is accompanied by translocation of the Ca2+- 22
responsive annexins A1 and A6 to the site of damage in order to plug the bleb and prevent excessive leakage of cytoplasmic contents [190, 192-194]. Shedding of toxin damaged membrane has been observed both in vitro and in vivo. In a C. elegans model, membrane shedding was identified as an important determinant for survival after exposure to the pore-forming Cry5B toxin.
Ca2+ triggered exocytosis of organelles also drives a response that is necessary for plasma membrane repair. Increases in Ca2+ above 1 µM stimulate the exocytosis of Ca2+ regulated, membrane proximal lysosomes [195-197]. These Ca2+ responsive lysosomes fuse with the plasma membrane in a process involving SNARE proteins and synaptotagmin VII, which is thought to facilitate membrane resealing by multiple mechanisms [198, 199]. A patching mechanism, characterized in oocytes, operates to reseal large disruptions in the plasma membrane. Large increases in intracellular Ca2+ trigger the transport of intracellular vesicles to the damaged area, where they undergo
Ca2+-dependent homotypic fusion to form a diffusion barrier that ultimately fuses with the plasma membrane to "patch" the disruption [187]. Vesicle fusion to the membrane is also thought to reduce membrane tension, which facilitates plasma membrane resealing [200].
Recent work examining cells exposed to sublytic concentrations of CDCs revealed that plasma membrane repair also involves endocytosis, thereby providing a better understanding of how a cell may remove pores that are stably inserted across the membrane [201]. In the proposed mechanism, lysosomal exocytosis was stimulated in cells perforated by SLO, leading to the release of the lysosomal contents, including acid sphingomyelinase (ASM). ASM reacts with sphingomyelinase in the plasma membrane to form cermide rich regions, which promote membrane invagination [202]. SLO was detected in these endosomes, and 23
was trafficked to multivesicular bodies and eventually lysosomes, where the contents were degraded. [203]. A similar process was observed in cells that were mechanically damaged [202]. PFTs were also shown to enhance endocytosis in C. elegans intestinal cells, which required the RAB-5 and RAB-11 trafficking proteins for plasma membrane repair [204].
1.3.3 Listeriolysin O as a signaling molecule
The importance of LLO as a signaling molecule was first recognized following the discovery that LLO activated signal transduction pathways in epithelial cells, including ERK-1, ERK-2, and p38 MAPK signaling, and that these signaling events had consequences for infection [205-207]. Many additional signaling effects of
LLO have since been discovered, and a re-evaluation of the pore-forming activity of
LLO at physiological pH has revised ideas about the role played by LLO secreted by extracellular bacteria [208, 209].
Perforation of the plasma membrane by extracellular LLO leads to rapid changes in the intracellular concentration of ions, which elicits a variety of intracellular signaling responses. LLO pores lead to Ca2+ influx from the extracellular medium and the release of intracellular Ca2+ stores [210-212] through the activity of phospholipase C (PLC), which generates inositol triphosphate (IP3), thereby activating the IP3 receptor (IP3R) Ca2+ channels. LLO may also release Ca2+ by directly injuring the endoplasmic reticulum (ER) [210]. The role of Ca2+ in the response to LLO extends beyond the repair of LLO pores. LLO causes Ca2+- dependent inhibition of the fusion and fission dynamics of the mitochondria, leading to transient fragmentation of this organelle, which enhances infection [213, 214]. This response is unique in several ways. While mitochondrial fragmentation is often an 24
indicator of apoptosis, classical apoptotic markers are not detected after exposure to
LLO. Also, LLO-induced fragmentation does not involve proteins that typically mediate mitochondrial fusion and fission, but instead involves the actin cytoskeleton, and another target of LLO — the ER [215]. ER stress caused by extracellular LLO triggers the UPR and inhibits L. monocytogenes infection [216].
A number of cellular responses are also triggered by changes in the intracellular concentration of K+ following exposure to LLO. LLO, as well as other
CDCs, can affect gene expression by stimulating deacetylation of histone H4 and dephosphorylation of histone H3, leading to downregulation of a subset of genes involved in the immune response to L. monocytogenes [217]. Histone H3 dephosphorylation is pore-dependent and is induced by K+ efflux [209]. K+ efflux also modulates the immune response by independently activating the inflammasome and caspase-1, and by activating autophagy [164, 209].
A number of other host cell responses to LLO have not yet been linked to ion fluxes, and may be activated by other mechanisms. The immune response to LLO can be further modulated by the previously mentioned MAPK signaling, and by translocation of NFκB, which induces the expression of adhesion molecules, IL-8, and MCP-1 in endothelial cells exposed to LLO [159, 178]. Autophagy is also stimulated by LLO-induced membrane damage. While autophagy can function as a mechanism to control infection by intracellular bacteria [218], it does not appear to play a major role in controlling growth of L. monocytogenes — which possess multiple mechanisms to escape autophagic degradation — but it may be important for clearing damaged membrane and organelles from the cell [80, 82, 219].
25
LLO also influences the host response to infection by interfering with the post-translational modification of host proteins including phosphorylation and ubiquitination. Recently, it was discovered that LLO decreases SUMOylation, in a pore-dependent manner. This process likely modulates the host response to promote infection [220]. Additionally, LLO is believed to have some pore-independent signaling ability. LLO oligomerization on the membrane can aggregate lipid rafts, clustering together raft associated proteins which leads to tyrosine kinase activation.
[221, 222]
1.4 Research aims and significance
The pore-forming toxin LLO is a major virulence factor of L. monocytogenes, acting intracellularly to mediate escape from the phagosome, and extracellularly to perforate the plasma membrane and elicit a variety of host signaling pathways. We hypothesized that these extracellular activites of LLO could affect the intracellular lifecycle of L. monocytogenes.
A major aim of this work was to elucidate the role of extracellular LLO during epithelial cell invasion by L. monocytogenes. To this end, we have quantified the contribution of LLO to bacterial association and invasion of epithelial cells, and developed novel tools to determine how these events are affected by toxin binding, oligomerization, and pore formation. We went on to identify host cell factors that were stimulated by LLO, and assessed their involvement in bacterial internalization.
A second major aim of this work was to examine potential links between L. monocytogenes invasion and the host cell plasma membrane repair response.
Perforation of the plasma membrane by pore-forming toxins stimulates a membrane repair response. Previous work has shown that the repair response involves 26
endocytosis, which L. monocytogenes could potentially exploit to facilitate invasion
[201]. To determine if these two processes are linked, or if they operate independently of one another, we assessed the contribution of various signals and host cell factors to both L. monocytogenes invasion and membrane repair.
Another aim was to determine the range of cells susceptible to LLO-mediated internalization. While the importance of InlA and InlB for infection depends on the expression of its receptor on the host cell surface, the receptor for LLO — cholesterol
— is present in the plasma membrane of all cells. To determine if the toxin plays a similar role during invasion of all epithelial cells, we quantified the contribution of
LLO to invasion in a range of epithelial cell types. We also compared the roles of
LLO, with those of the established L. monocytogenes invasins InlA and InlB during bacterial association, invasion, and intracellular surivival.
The complex intracellular lifecycle of L. monocytogenes makes it an ideal pathogen for studying the cell biology of infection. Characterizing the novel LLO- mediated internalization pathway utilized by L. monocytogenes will provide a more complete understanding of the mechanisms that intracellular pathogens use to hijack host cell biology to invade cells. Additionally, these findings will help to better understand the responses stimulated by pore-forming toxins during infection. Some of these events, such as the membrane repair response, appear to be generally activated by perforation of the plasma membrane — a common event during infection, but the consequences of which are not well understood. Importantly, the LLO variants constructed in this work will be invaluable for future studies to determine which of the signaling events stimulated by LLO are due to poreformation, and which result from some other activity of the toxin.
27
Chapter 2. The pore-forming toxin listeriolysin O mediates a novel entry
pathway of Listeria monocytogenes into human hepatocytes.
2.1 Introduction
Despite the diversity of virulence factors promoting host cell invasion, only two major mechanisms of entry have been observed [223-225]. First, invasins on the bacterial cell surface bind to host cell receptors to activate complex signaling cascades that orchestrate the internalization of the bacterium. Second, some bacteria bypass the requirement for a host receptor by utilizing a secretion system that injects effectors into the host cell. These effectors subvert the host signaling machinery to trigger bacterial uptake into macropinosomes [226, 227]. Using Listeria monocytogenes as a model intracellular pathogen, we have analyzed a novel entry pathway that is activated in response to host cell perforation by a pore-forming toxin.
L. monocytogenes is a foodborne pathogen that causes a large spectrum of clinical manifestations ranging from gastroenteritis to life-threatening meningo- encephalitis and sepsis. Susceptible hosts include the elderly and immunocompromised individuals [2]. In pregnant women, the bacterium can cross the maternofetal barrier causing abortion, stillbirth, and neonatal meningitis or sepsis [2,
228, 229]. To cross host barriers and infect various organs including the liver [126,
127, 131, 230], L. monocytogenes expresses multiple virulence factors that induce its entry into, and survival within various nonphagocytic cells [231-234].
28
Two major genetic loci encode the virulence factors responsible for host cell invasion: the internalin operon and Listeria pathogenicity island 1 (LIPI-1) [46, 235].
The internalin operon encodes internalin (InlA) and InlB that bind to E-cadherin and the hepatocyte growth factor receptor (HGF-Rc/c-Met), respectively [58, 236].
Depending on the receptors expressed by the host cells, L. monocytogenes entry involves one or both internalins [47, 133]. The internalin/host receptor interactions activate signaling cascades within cholesterol-rich microdomains leading to the internalization of the bacterium [50, 56, 237, 238]. After internalization, the secreted pore-forming toxin listeriolysin O (LLO) and two phospholipases (encoded by LIPI-
1) mediate L. monocytogenes escape from the endocytic vesicle into the cytoplasm, where bacteria divide and undergo F-actin based motility to spread from cell to cell
[27, 239, 240].
InlA and InlB were defined as bacterial invasins based upon their critical role in L. monocytogenes invasion of nonphagocytic cells, and the fact that they are sufficient to induce entry of noninvasive bacteria when overexpressed from a plasmid
[24, 46]. It is well established that the internalins are critical for host cell invasion; however, they may not be sufficient for inducing efficient bacterial uptake due to their low levels of expression in L. monocytogenes. Additional virulence factors including
LLO have been proposed to regulate L. monocytogenes entry into host cells [37, 41,
73, 212, 241].
LLO is required for L. monocytogenes pathogenesis [242] and belongs to the family of the cholesterol-dependent cytolysins (CDCs) produced by numerous Gram- positive pathogens [84, 243, 244]. The CDCs are 50–70 kDa proteins synthesized as water soluble monomers that bind to cholesterol in host cell membranes [93]. Three of the CDCs — intermedilysin, lectinolysin and vaginolysin — have been shown to 29
bind to a host receptor (the complement regulatory molecule CD59) in addition to cholesterol [94, 96, 245]. Upon binding to host membranes, the CDCs diffuse laterally to form a ring-shaped oligomeric prepore complex. This complex then inserts a large β-barrel pore across the membrane in a cholesterol-dependent fashion [98].
Eukaryotic cells possess sophisticated mechanisms to repair damaged plasma membranes and survive moderate exposure to pore-forming toxins including the
CDCs [201]. A growing body of evidence demonstrates that pores formed by pore- forming proteins including perforin, S. aureus alpha hemolysin, and the CDC streptolysin O (SLO), are removed from the plasma membrane through a mechanism that involves membrane internalization [201, 204, 246]. The ability of CDCs to induce membrane internalization in eukaryotic cells to repair their membrane raised the hypothesis that LLO may affect the internalization of L. monocytogenes. In the present work we explored the link between the formation of pore complexes by LLO and bacterial internalization. Using several experimental approaches, we determined that LLO is a critical invasion factor that perforates the host cell plasma membrane to activate L. monocytogenes internalization into human hepatocytes.
2.2 Materials and Methods
2.2.1 Bacterial strains and plasmids
WT L. monocytogenes (DP10403S), isogenic Δhly (hly is the gene coding for
LLO) (DPL2161), and ΔinlAB (DPL4404) deletion mutants were gifts from Dr. Dan
Portnoy (U.C. Berkeley, California, USA) [108, 247, 248]. To construct the triple deletion mutant (Δhly ΔinlAB) we deleted hly in the DPL4404 strain by allelic exchange using the pKSV7 integrational shuttle vector (a gift from Dr. Nancy Freitag,
30
University of Illinois at Chicago, USA) [249]. A ~1000-bp DNA fragment consisting of the upstream (from bp 962 to 1463) and downstream (from bp 3029 to 3529) sequences flanking the hly open reading frame was amplified from L. monocytogenes chromosomal DNA [108]. The primers (Table 1) were designed to generate restriction sites for EcoRI and PstI. The digested fragment was ligated into pKSV7. The pKSV7 with the 1000 bp fragment was used to perform allelic exchange according to [250].
Nonhemolytic colonies were identified on 5% blood agar plates (Becton Dickinson).
The deletion of hly was further ensured by amplification of the chromosomal DNA with the primers used to generate this strain and a second set of primers that amplify the entire hly coding region. L. innocua 33090 was purchased from ATCC. Bacteria were grown overnight at 37° C in brain heart infusion (BHI) (BD Biosciences). For invasion assays, overnight cultures were diluted 1/20 in BHI and grown at 37°C until
OD600 = 0.7–0.8. Bacteria were washed three times in phosphate-buffered saline
(PBS) and diluted to the indicated multiplicity of infection (MOI) in the cell culture medium without serum. The plasmids pAM401 and pET29b coding for hly were gifts from Dr. D. Portnoy [108]. The plasmid phly/prfA* coding for hly was a gift from Dr.
Svetlana A. Ermolaeva (Gamaleya Research Institute of Epidemiology and
Microbiology, Moscow, Russia) [251]. The plasmid pQE-30 coding for ply was kindly provided by Dr. R. K. Tweten [93].
2.2.2 Mammalian cell culture
The HepG2 human hepatocyte (ATCC HB-8065) cell line was grown in minimum essential medium (MEM) (+) Earle's salts and L-glutamine (Invitrogen), supplemented with 10% heat inactivated fetal bovine serum (HI-FBS; Lonza), 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 31
µg/ml streptomycin (Invitrogen). Mammalian cells were maintained at 37°C in 5%
CO2 atmosphere. Cells were seeded in 24-well tissue culture plates and grown for 48 h (HepG2; 1×105 cells/well) before infection.
2.2.3 Coating of bacteria and polystyrene beads with LLO
Bacteria were coated with six-His-tagged toxin [28]. Briefly, 4×108 bacteria were washed twice and incubated for 10 min on ice in buffer A (20 mM Hepes pH
7.5, 50 mM NaCl, 1 nM nickel chloride). Bacteria were washed and incubated in 200
µl buffer B (20 mM Hepes pH 7.5, 50 mM NaCl) for 10 min with six-His-tagged toxin. Bacteria were then washed once with buffer B and suspended in 200 µl of buffer C (20 mM Hepes pH 7.5, 150 mM NaCl) before infecting HepG2 cells.
Carboxylate microspheres (Alexa 350, 1 µm diameter; Molecular Probes) were covalently coated with 5 mg/ml bovine serum albumin (BSA) following the manufacturer's instructions. LLO was noncovalently associated to the surface of the
BSA coated beads using the same experimental procedure used to coat bacteria. In each experiment, we verified that host membranes were not damaged by LLO, as this could cause the entry of antibodies used to label extracellular bacteria. Following fixation, cells were labeled with an anti-tubulin antibody (Sigma) and secondary fluorescent antibodies. We observed that in our experimental conditions antibodies could not enter the cells as microtubules were not labeled.
2.2.4 Gentamicin survival assay
HepG2 cells were infected with the L. monocytogenes at MOI 20, LLO-coated bacteria at MOI 20, or L. innocua and L. innocua phly/prfA* at MOI 100. The plates were centrifuged at room temperature for 5 min and incubated for 30 or 60 min at 32
37°C. Cells were washed and incubated with 15 µg/ml (for bacteria grown to OD600 =
0.8) or 100 µg/ml (For L. innocua grown to OD600 = 0.2) gentamicin for 1 h or 30 min, respectively. When measuring the intracellular survival of L. innocua in comparison to L. innocua phly/prfA*, the bacteria were grown to OD = 0.2 because this bacterial density led to the highest secretion level of LLO (data not shown). Cells were washed three times with PBS and lysed with 0.2% Triton X-100 in H2O. Serial dilutions of cell lysates were immediately performed in PBS and plated on BHI agar.
The colony forming units (CFUs) were enumerated after 48 h of incubation at 37°C.
2.2.5 Measurement of bacterial association and entry into host cells
HepG2 cells were infected with bacteria at MOI 20. The plates were centrifuged for 5 min (230 x g) at room temperature and incubated for 30 min at
37°C. Cells were washed with PBS, fixed with PBS/4% paraformaldehyde (PFA) for
15 min at room temperature, and then washed with 0.1 M glycine in PBS and incubated for 1 h in blocking solution (0.1 M glycine, 10% HI-FBS in PBS, pH 7.4).
Following fixation and blocking, extracellular bacteria, total bacteria and host cells were labeled as previously described [252]. To quantitate the numbers of bacteria and mammalian cells, 40 sets of images (DAPI, Alexa 488, Alexa 568, phase contrast) were automatically acquired for each experimental condition using a 20 X objective.
MetaMorph imaging and analysis software was used to enumerate the total number of bacteria (Nt), extracellular bacteria (Ne), and mammalian cells (Nc) [252]. The percentage of internalization was calculated as (Nt - Ne)/Nt x 100. Bacterial association with host cells was calculated as Nt/Nc. The results were expressed relative to control (% control). For invasion assays in the presence of soluble LLO,
LLO was added to the cell culture medium along with L. monocytogenes. 33
2.2.6 Construction and purification of recombinant toxins
LLO variants were constructed by PCR-based site-directed mutagenesis using pET29b encoding native six-His-tagged LLO as a template. Mutagenic primers (Table
1) were used to construct LLOmL containing the substitutions K344C and I359C;
LLOpL containing the substitutions G80C and S213C; and LLO Alexa 488 with the substitutions C484A and D69C. Mutations were introduced by amplifying hly from pET29b using pfu Ultra II fusion polymerase, followed by digestion of methylated template DNA by DpnI (Stratagene). The constructs were transformed in E. coli XL1-
Blue and BL21(DE3). Mutations were confirmed by DNA sequence analysis at The
Ohio State University Plant-Microbe Genomics Facility. Recombinant six-His-tagged
LLO, LLOT and PLY were purified from E. coli BL21(DE3) as described previously
[74]. LLOmL and LLOpL were dialyzed overnight in the absence of reducing agents to allow for disulfide bond formation. Toxins were stored at −80°C in 1 M NaCL, 50 mM phosphate buffer, pH 8. To obtain LLO Alexa 488, LLOC484A/D69C was labeled by chemical coupling to Alexa 488-Maleimide (Molecular Probes) under conditions that lead to a dye to protein molar ratio >0.9, following the manufacturer's instructions. The fluorescent toxin was separated from the unconjugated dye by gel filtration chromatography. All the recombinant toxins purified in this study contain a six-His tag.
2.2.7 Hemolytic assays
Sheep erythrocytes (10% suspension; Lampire Biological) were diluted to
0.25% in PBS pH 7.4. Serial dilutions of toxins and erythrocytes were co-incubated in
PBS at 37°C for 30 min in 96 well plates. Plates were centrifuged and the absorbance 34
of the supernatant (A540) was measured in a PowerWavex340 spectrophotometer.
Erythrocytes incubated with 0.1% Triton X-100 in PBS or PBS alone served to determine the maximum (100%) and minimum (0%) hemolytic activity, respectively.
DTT alone had no effect on hemolysis (data no shown). The hemolytic activities of the bacteria were measured using a similar approach except that the indicated amounts of bacteria were added to the wells. The kinetic hemolytic assay in Fig. 2.7 was performed according to [253]. In this assay a decrease in absorbance reflects the lysis of erythrocytes.
2.2.8 Western blotting analysis of LLO expression by the bacterial strains
Overnight cultures of L. monocytogenes, L. innocua, and L. innocua phly/prfA* were diluted 1/20 in BHI and grown to OD600 = 0.8 (L. monocytogenes) or
OD600 = 0.2 (L. innocua, and L. innocua phly/prfA*) in BHI. Bacterial suspensions containing 2.0 × 108 bacteria were collected and centrifuged. The supernatants were collected for protein precipitation and bacterial pellets were washed and lysed as follows. The supernatants (0.25 ml of L. monocytogenes and 1 ml of L. innocua cultures) were subjected to trichloroacetic acid (TCA) precipitation. One volume cold
TCA was added to 4 volumes of supernatant and incubated 1 h on ice. Samples were centrifuged (11,000 g, 10 min, 4°C) and precipitates were washed twice with cold acetone. Bacterial and dried protein pellets were suspended in Laemmli's sample buffer. Bacterial lysates (107 bacteria/well, ~600 ng/well) and precipitates were subjected to SDS-PAGE and western blotting using rabbit anti-LLO (Abcam) and horseradish peroxidase-conjugated secondary antibodies (Cell Signaling).
35
2.2.9 HepG2 perforation assay
HepG2 cells (0.5×105) were cultured in glass bottom culture dishes (MatTek;
35 mm petri dish, 10 mm microwell) for 48 h. Cells were washed twice and incubated in the presence or absence of 1 mM CaCl2 in a buffer containing 150 mM NaCl, 1 mM MgCl2, 5 mM KCL, 20 mM Hepes, 10 mM Glucose, 4 µM ethidium homodimer, pH 7.4. Cells were placed on a temperature controlled microscope at 37°C and phase- contrast and fluorescence images were recorded every 10 s using a 100 X objective for 10 min. LLO (0.5 nM) was added and movies were recorded for an additional 28 min. Results were expressed as the average fluorescence intensity measured from at least 5 movies at each time point.
2.2.10 LDH and cell viability measurements
HepG2 cells were incubated with LLO (1, 5, or 20 nM) or BSA/LLO-coated beads for 30 min at 37°C. Following incubation, supernatants were recovered from each sample and centrifuged at 500 x g at 4°C for 5 minutes to pellet any cells released into the supernatant. 10 µl of each supernatant was diluted into 40 µl of cell culture medium without serum in 96 well plates, and assayed for the presence of lactate dehydrogenase with the TOX7 in vitro toxicology assay kit according to the manufacturer's instructions (Sigma). To assess cell viability immediately or 24 h after
LLO treatment, HepG2 cells were detached from wells with 0.25% Trypsin-EDTA.
Cells were then mixed 1:1 with 0.4% trypan blue for 3 minutes, after which viable
(unstained) and nonviable (stained in blue) cells were enumerated with a hemocytometer.
36
2.2.11 Toxin oligomerization (NativePAGE)
Erythrocyte ghost membranes (EGM) were prepared as described previously
[98] and stored at 4°C in resealing buffer (10 mM phosphate buffer, 5 mM MgCl).
EGM (6.75 × 108) were incubated in PBS with 157 nM LLO on ice for 1 min and were transferred to 37°C for 5 min. When indicated, 4 mM DTT was added to reduce the disulfide bond in LLOmL and LLOpL. Samples were centrifuged at 15,000 x g for 15 min and the supernatant was removed and replaced with an equal volume of
PBS. LLO oligomerization was analyzed using the NativePAGE Novex Bis-Tris Gel electrophoresis system. The samples were mixed with NativePAGE sample buffer containing 1% detergent, and run on a 4–16% Bis-Tris gel as described by the manufacturer (Invitrogen). LLO was detected by western blotting.
2.2.12 Toxin oligomerization (transmission electron microscopy)
LLO solutions (750 nM) in buffer (20 mM Hepes, pH 7.0, ±2 mM DTT) were pipetted in Teflon wells as 13 µl droplets and coated with 1 µl of a 0.5 mg/ml lipid mixture containing 50 mol% cholesterol (Avanti) and 50 mol% 1,2 dioleoyl-sn- glycero-3-phosphocholine (Avanti) in chloroform. After incubation in a humid chamber at room temperature for 1 h, the LLO complexes were transferred to carbon support films on electron microscopy (EM) grids and negatively stained with 1%
(w/v) uranyl acetate and observed with a FEI Tecnai F20 transmission electron microscope equipped with a Gatan Ultrascan 4K X4K CCD camera.
2.2.13 Invasion assays using polystyrene beads. F-actin and EEA1 labeling
Cells were washed and incubated for 30 min with BSA-, or BSA/LLO-coated beads at a MOI = 20 in MEM. Cells were washed, fixed with PFA and blocked. 37
Extracellular beads were labeled with an anti-BSA rabbit polyclonal antibody (Sigma) followed by a goat anti-rabbit secondary antibody conjugated to Alexa-568. The percentage of internalized beads was determined by fluorescence microscopy based on their unique (Alexa 350, intracellular) or dual fluorescence (Alexa 350 + Alexa
568, extracellular) and expressed as% intracellular beads ( = intracellular beads/total beads * 100). EEA1 was labeled with a primary antibody (Santa Cruz) and a fluorescent (Alexa 488) donkey anti-goat secondary antibody in permeabilized cells.
When co-labeling of the BSA beads and EEA1 was performed, a fluorescent donkey anti-rabbit secondary antibody was used to label the primary rabbit anti-BSA antibody. For F-actin labeling, cells were fixed and labeled as described in [253].
2.2.14 Silencing of clathrin heavy chain in HepG2 cells
HepG2 cells were transfected in 24-well cell tissue culture plates with specific human clathrin heavy chain siRNA (Ambion 43908824, Table 1) or with scrambled siRNA (Ambion, 4390843) (50 nM siRNA in 0.5 ml/0.5×105 cell/well) using SiPort neoFx transfection reagent according to the manufacturer instructions (Ambion).
After 24 h, cell culture medium was replaced and cells were further incubated for 24 h. Clathrin knock-down efficiency was verified in each experiment by western blotting using primary rabbit anti-clathrin heavy chain (Abcam) and mouse anti-α- tubulin (Sigma) antibodies.
2.2.15 Cell treatment with chemical inhibitors
Cells were pre-incubated with 0.5 µg/ml cytochalasin D (Sigma) for 10 min,
33 µM nocodazole (Sigma) for 60 min, 250 µM genistein (Sigma) for 60 min, 37 µM
LY294002 (EMD chemicals) for 60 min, 1 µM wortmannin (Sigma) for 60 min, 80 or 38
120 µM dynasore (Sigma) for 30 min, and 10 µM chlorpromazine (Sigma) for 30 min before the addition of coated beads, and the drugs were maintained at the same concentrations in the cell culture medium until the cells were fixed.
2.2.16 Transferrin uptake
HepG2 cells were washed three times in MEM and were serum starved for 2 h. Cells were incubated in MEM at 37°C with 5 µg/ml iron loaded Alexa 568 conjugated tranferrin (Molecular Probes). After 2 and 10 min of incubation, cells were transferred to ice, washed three times with cold medium and acid washed (200 mM NaCl, 50 mM MES, pH5.0) for 5 min. Four washes were performed with a cold buffer containing 150 mM NaCl, 1 mMCaCl2, 1 mM PBS, 5 mM KCl, 20 mM Hepes, pH 7.4. Alexa 488-conjugated transferrin was added (5 µg/ml) for 2 h at 4°C. Cells were then washed and fixed. Phase contrast and fluorescence images of the internalized (Alexa 568-Tf) and cell surface-associated (Alexa 488-Tf) fluorescent transferrin were acquired using a 40X objective. Extracellular and internalized transferrin molecules were measured by quantitative fluorescence microscopy.
Briefly, images were first background corrected and the average fluorescence intensities were measured in the cells (about 1000 cells were analyzed for each experimental condition).
2.2.17 Cholesterol depletion and repletion, and LLO treatment with neutralizing antibodies
For cholesterol depletion, HepG2 cells were washed twice with MEM
(without serum) and incubated at 37°C for 30 min with 5 mM MβCD in MEM. Cells were then washed twice and assayed as described. For cholesterol repletion, 39
cholesterol-depleted cells were washed twice and incubated for 15 min with a solution of 5 mM cholesterol-MβCD in MEM, then washed and assayed as described. A 1:1:1 mixture of three monoclonal anti-LLO neutralizing (H14-3, B8B20-3-2, and A4-8) or control anti-LLO (D21-1-4) antibodies (gift from Dr. P. Cossart, Institute Pasteur,
France) was added to purified LLO, BSA/LLO-coated beads, at concentrations of 2,
10, or 20 µg/ml.
2.2.18 Image acquisition
Images were acquired on a motorized inverted epi-fluorescence microscope
(Axio Observer D1, Zeiss) equipped with 20 X Plan Neofluar (N.A. = 0.5), 100 X
Plan Apo (N.A. = 1.4), and 40X Plan Neofluar (N.A. = 1.3) objectives, a high speed filter changer Lambda DG-4 (300 Watts Xenon Arc bulb, Sutter Instrument
Company), an optical emission filter wheel Lambda 10–3 for the fluorescence imaging, and a Smart shutter that controls the illumination for phase contrast imaging
(Sutter Instrument Company). The camera (back-illuminated, frame-transfer EMCCD
Cascade II 512) was from Photometrics. The filter sets for fluorescence were purchased from Chroma Technology Corporation: DAPI (49000),
GFP/FITC/Alexa488 (49002), Cy3/DsRed/Alexa568 (49005). The microscope was controlled by MetaMorph imaging software (Universal Imaging).
2.2.19 Live cell imaging
All movies were acquired on the microscope stage of an inverted fluorescence microscope at 37°C using a 100X objective. Phase contrast and fluorescence images of HepG2 cells were acquired every 10 s for 40 min, LLO was added 10 min after
40
starting recording. In some experiments, cells were incubated in Ca2+-free buffer as indicated. Movies were accelerated 198 times.
2.2.20 Scanning electron microscopy
Following incubation with BSA/LLO-coated beads, cells were washed in PBS, fixed and processed as described in [50] with the following modifications: Coverslips were sputter coated with gold palladium at 17 mA for 75 sec with a Cressington 108 sputter coater. Samples were examined and photographed with a FEI Nova Nano scanning electron microscope operating at 5 Kv.
2.2.21 Statistics
A minimum of three independent experiments were performed, each in duplicate, unless otherwise indicated. Data were expressed as mean ± Standard Error of the Mean (SEM). P-values were calculated using a standard two-tailed Student's t- test and determined significant if lower than 0.05. In figures, asterisks indicate a significant difference between the indicated experimental conditions (* p<0.05; ** p<0.005).
41
Table 1. Primer and siRNA sequences used in chapter 2
Constructs Amino acid 5’ to 3’ Sequences substitution LLOmL K344C CATCAAAAATTCTTCCTTCTGCGCCG TAATACGGAGGTTCCGC I359C GATGAAGTTCAAATCTGCGACGGCA ACCTCGGAGAC LLOpL G80C AATGTATTAGTATACCACTGCGATGC AGTGACAAATGTCCGCCA S213C GATGACGAAATGGCTTACTGCGAAT CACAATTAATTGCG LLO C484A ATTAATGTTTACGCTAAAGAAGCGAC Alexa 488 TGGTTTAGCTTGGGAATGG D69C GAAATCGATAAGTATATACAAGGATT GTGTTATAAAAACAATGTATTAGTAT ACCACG Primers for construction of - GGGAATTCAATTGTTGATACAATGAC DPL2161Δhly ATC-forward- GGCTGCAGGGTCTTTTTGGCTTGTGT AT-reverse- Primers to amplify the hly - CCGTCGGATCCATGAAAAAAATAAT ORF GCTAGTTTTTATTACAC-forward- ATCCGCGCTGCAGTTCGATTGGATTA TCTACTTTATTAC-reverse- Human clathrin heavy chain - GGUUGCUCUUGUUACGGAUtt –sense- siRNA AUCCGUAACAAGAGCAACCgt- antisense-
Primers used to construct DPL2161Δhly from [108].
42
2.3 Results
2.3.1 LLO is critical for efficient entry of L. monocytogenes into host cells
The gentamicin survival assay is commonly used to assess the role of virulence factors in the intracellular survival of bacterial pathogens. This assay enumerates viable intracellular bacteria after killing extracellular bacteria with the cell impermeant antibiotic gentamicin. We measured the relative intracellular survival of wild type L. monocytogenes (WT), along with an isogenic LLO-deficient (Δhly) mutant, a double deletion mutant deficient in the expression of InlA and InlB
(ΔinlAB), a triple deletion mutant (ΔhlyΔinlAB), and the LLO-complemented mutant
(Δhly + pAM401hly and ΔhlyΔinlAB + pAM401hly) strains in HepG2 cells. As presented in Fig. 2.1A, efficient intracellular survival of L. monocytogenes required the expression of the internalins as well as LLO. In the absence of the three virulence factors, intracellular survival was almost completely abrogated. The defect in intracellular survival of LLO-deficient bacteria (Δhly) was due to the lack of LLO expression, as the LLO-complemented (Δhly + pAM401hly) and WT strains displayed similar intracellular survival.
The results obtained with the gentamicin assay reflect the efficiencies of several stages of host cell invasion including L. monocytogenes association with host cells, entry into host cells, escape from the internalization vesicle, and intracellular division. LLO is known to promote L. monocytogenes intracellular survival by mediating bacterial escape from the internalization vesicle; therefore, it was difficult
43
A B Overlay Nuclei (DAPI) Relative Intracellular Survival ** ** 140 ** 120 ** 100 80
60 (% WT) (% 40 20 0 WT Δhly ΔinlAB ΔinlAB Δhly ΔinlAB Extracellular (Alexa 488) Total (Alexa 568) Δhly Δhly + pAM401hly C D Relative Association Relative Entry
* * 200 120 180 ** 160 100 140 80 120
100 60 (% WT) (% (% WT) (% 80 60 40 40 20 20 0 0 WT Δhly ΔinlAB Δhly ΔinlAB WT Δhly ΔinlAB Δhly ΔinlAB Δhly Δhly Δhly Δhly + pAM401hly + pAM401hly
Figure 2.1 LLO is required for efficient entry of L. monocytogenes into HepG2 cells. (A) HepG2 cells were infected with isogenic WT (DP10403S), LLO-deficient (Δhly), InlAB-deficient (ΔinlAB), LLO- and InlAB-deficient (ΔhlyΔinlAB), or LLO- complemented (Δhly + pAM401hly; ΔhlyΔinlAB +pAM401hly) bacteria (MOI = 20) for 30 min at 37°C. Gentamicin was added for 1 h and the CFUs were enumerated as described in methods. Results were the mean ± SEM (n≥3) and expressed relative to WT. Statistics indicated here and elsewhere are as follows: * p<0.05; ** p<0.005. (B) to (D) Cells were infected with WT, LLO-deficient (Δhly), LLO- and InlAB-deficient (ΔhlyΔinlAB), or LLO-complemented (Δhly + pAM401hly; ΔhlyΔinlAB + pAM401hly) bacteria (MOI = 20) for 30 min at 37°C. Cells were washed, fixed, and labeled with fluorescent antibodies and DAPI. (B) Representative images of WT extracellular bacteria (green), total bacteria (red), and HepG2 nuclei (blue). The arrow indicates an internalized bacterium. Scale bar = 10 µm. In (C) and (D), results were the mean ± SEM (n≥3) and were expressed relative to WT. to dissociate the role of LLO in escape from its potential role in bacterial association and/or entry using this assay. To specifically assess the role of LLO during the initial stages of the invasion process, we used an automated fluorescence-based assay that 44
measures the efficiencies of bacterial association with and internalization into host cells [252]. We found that LLO and the internalins did not significantly affect bacterial association with HepG2 cells. However, LLO significantly increased bacterial association when overexpressed from a plasmid (Fig. 2.1C). More importantly, LLO is critical for L. monocytogenes internalization as we observed a marked decrease in entry of the LLO (Δhly) and LLO/internalins (ΔhlyΔinlAB) deficient strains relative to the WT strain (Fig. 2.1D). The LLO-complemented (Δhly
+ pAM401hly) and WT strains displayed similar internalization efficiencies (Fig.
2.1D).
2.3.2 Direct and dose-dependent activity of LLO in L. monocytogenes entry into host cells
The lack of LLO in the LLO-deficient strains may indirectly decrease the efficiency of bacterial entry by affecting the expression or regulation of other virulence factors. To verify that the defect in entry was solely due to the absence of
LLO, we used LLO-deficient bacteria coated with increasing concentrations of six-
His tagged recombinant LLO. LLO was noncovalently adsorbed on the surface of
LLO-deficient L. monocytogenes using a previously described protocol [28]. We performed a noncovalent coating because the toxin monomers likely need to dissociate from the bacterial surface to freely diffuse within the host cell membrane to form oligomers and pores. To validate the coating procedure, we measured the fluorescence intensity of bacteria coated with increasing concentrations of an Alexa
488 LLO derivative (Fig. 2.2). We then measured the entry of LLO-deficient
45
A B Hemolytic activity of L.m. Coating with LLO Alexa 488 LLO Alexa 488 100 * 100 80 LLO Alexa 488 .
L.m 80
LLO ) / )
60 3 60 40
40
% Hemolysis % A.F.I (10 A.F.I 20 20
0 0 0.01 1 100 10000 0 0.45 0.9 Toxin (nM) LLO Alexa 488 (µM)
Figure 2.2 L. monocytogenes coating with recombinant LLO. (A) The hemolytic activity of LLO Alexa 488 and LLO was measured in triplicate for 30 min at 37°C and pH = 7.4. A representative experiment (of 3) is presented. (B) LLO-deficient L. monocytogenes (Δhly, DPL2161, L.m.) were coated with LLO Alexa 488 and the fluorescence intensity associated with bacteria was measured by quantitative fluorescence microscopy. Results were the average fluorescence intensity (A.F.I.) ± SEM expressed in arbitrary units (n≥3). At least 100 bacteria were analyzed in each experiment.
L. monocytogenes coated with increasing concentrations of recombinant LLO and observed that LLO increased bacterial entry into HepG2 cells in a dose-dependent fashion (Fig. 2.3A). We next determined whether LLO should be localized in the vicinity of the bacteria or whether LLO could act distally to regulate entry. When adding LLO to the culture medium along with the LLO-deficient strain, we observed an increase in the efficiency of bacterial internalization (Fig. 2.3B). Together, these data show that LLO potentiates internalization of L. monocytogenes into host cells in a dose-dependent fashion by acting locally or from a distance.
46
A B L.m. Coated With LLO LLO in the Cell Culture Medium
600 ** * ** 400 500 350 ** 300 400 250 300 200 200 150
100 Relative Relative (%Entry C) Relative Relative (%Entry C) 100 50 0 0 C 0.5 1 2 1 C 0.05 1 2 LLO (M) LLO (nM)
Control + LLO + LLO Alexa 488
Figure 2.3 Direct and dose-dependent role of LLO in L. monocytogenes entry. (A) LLO-deficient L. monocytogenes (Δhly, strain DPL2161, L.m.) treated with 1 mM nickel (II) chloride coating buffer in the absence (C) or presence of LLO (black bars) or LLO Alexa 488 (grey bar) were used to infect HepG2 cells at 37°C for 30 min (MOI = 20). (B) HepG2 cells were infected with LLO-deficient L. monocytogenes (Δhly, DPL2161) at 37°C for 30 min (MOI = 20), in the absence (C) or in the presence of LLO added exogenously to the cell culture medium (black bars). In (A) and (B) bacterial entry was measured by fluorescence microscopy. The results were expressed relative to bacteria incubated with host cells in the absence of LLO (C) and were the mean ± SEM (n≥3).
2.3.3 LLO is sufficient to induce bacterial and bead entry into host cells
We determined whether LLO is sufficient to induce bacterial entry into host cells or whether it only potentiates activity of the internalins. We used nonpathogenic and noninvasive Listeria innocua that does not express any of the known L. monocytogenes virulence factors [34]. As shown in Fig. 2.4A, L. innocua noncovalently coated with six-His tagged LLO were able to enter and survive in
HepG2 cells. As a second approach, L. innocua were transformed with a plasmid coding for hly (L. innocua phly/prfA*). As presented in Fig. 2.4B, LLO secreted from
L. innocua phly/prfA* was sufficient to induce bacterial entry and survival into
47
HepG2 cells. The efficiency of host cell invasion by L. innocua phly/prfA* was low compared to what was observed with WT L. monocytogenes and LLO-coated L. innocua (Fig. 2.4A). Because the activity of LLO in bacterial entry is dose-dependent
(Fig. 2.3), we determined whether L. innocua phly/prfA* expresses low levels of LLO compared to WT L. monocytogenes. The hemolytic activity of L. innocua phly/prfA* was approximately 40-fold lower than L. monocytogenes (Fig. 2.4C). L. innocua phly/prfA* also produced low levels of LLO compared with WT L. monocytogenes
(Fig. 2.4D). These results demonstrate that low concentrations of LLO (below the concentration secreted by L. monocytogenes) are sufficient to induce bacterial entry into host cells. To assess the role of LLO in the absence of any other bacterial factor, we measured the uptake of 1 µm fluorescent polystyrene beads coated with LLO. The beads were first covalently coated with bovine serum albumin (BSA). The BSA- coated beads were then noncovalently coated with LLO [28]. LLO was sufficient to induce internalization of the beads into HepG2 cells in a dose-dependent fashion; whereas, the beads coated with only BSA were not taken up by the cells (Fig.
2.5).This result clearly showed for the first time that a pore-forming toxin is a bacterial invasion factor. Cell viability was assessed by measuring the release of lactate dehydrogenase (LDH) and by trypan blue exclusion immediately and 24 h following toxin treatment (Fig. 2.6). These results show that the concentrations of
LLO that lead to the optimal bead uptake by host cells do not compromise cellular integrity.
48
A ** B 120 ** 120 ** 100 100 80 80
60 60
40 40
20 20 Relative Intracellular Survival Intracellular Relative Relative Intracellular Survival Intracellular Relative 0 0 L.m. L.i. L.i. L.i. buffer L.i. L.i. buffer + LLO phly/prfA* C D 120 L.m. L.i. L.i. phly/prfA* 100 80 pellet 60 L.m. supernatant 40
% Hemolysis % L.i. 20 phly/prfA* 0 105 106 107 108 109 Number of Bacteria
Figure 2.4 LLO is sufficient to induce the entry of noninvasive L. innocua into HepG2 cells. (A) HepG2 cells were infected with WT L. monocytogenes (DP10403S, L.m.), L. innocua (L.i.), or L. innocua treated with 1 mM nickel (II) chloride coating buffer in the absence (L.i. buffer) or presence of 5 µM LLO (L.i. buffer + LLO) at MOI = 20 for 30 min at 37°C. Gentamicin was added for 1 h and the intracellular CFUs were enumerated. Results were the mean ± SEM (n≥3) and expressed relative to L. monocytogenes. (B) HepG2 cells were infected with L. innocua (L.i.) or L. innocua phly/prfA* (L.i. phly/prfA*) for 60 min at 37°C (MOI = 100). Gentamicin was added for 30 min and the intracellular CFUs were enumerated. Results were the mean ± SEM (n≥3) and expressed relative to L. i. phly/prfA*. (C) Hemolytic activities of L. monocytogenes (DP10403S) and L. innocua phly/prfA* measured at 37°C for 30 min, pH 7.4. Results were the mean ± SEM (n≥3). (D) Equivalent amounts of L. monocytogenes and L. innocua lysates and proteins precipitated from their culture supernatants were analyzed by western blotting using an anti-LLO primary antibody. A representative experiment (of 3) is presented.
49
2.3.4 LLO perforates host cells at physiological temperature and pH
Previous studies have established that poreformation by LLO is pH sensitive at the host temperature (37°C), with LLO being more active at acidic pH. At 37°C, soluble LLO is inactivated by a pH-triggered unfurling of the domain 3 twin α-helical
A PC Total (Alexa 365)
Extracellular Overlay (Alexa 568)
B ** 30
BSA 20 ** BSA/LLO
10 % Intracellular Intracellular % Beads
0 0.5 1 2 5 LLO (M)
Figure 2.5 LLO is sufficient to induce entry of polystyrene beads into HepG2 cells. HepG2 cells were incubated with BSA- or BSA/LLO-coated beads for 30 min at 37°C (MOI = 20). Cells were washed, fixed, and extracellular beads were fluorescently labeled with a primary anti-BSA antibody and a secondary fluorescent antibody. (A) Representative fluorescence and phase-contrast (PC) images acquired with a 100 X objective. Scale bar = 10 µm. (B) The mean ± SEM (n≥3) percentage of intracellular beads was determined by counting a minimum of 100 beads in each sample. 50
bundles that prevents further formation of a pore [112]. We hypothesized that the pH and temperature dependent inactivation would not occur if LLO was incubated in the presence of host membranes. Presumably, LLO would bind to the cell membrane where it would assemble into functional pores rather than unfolding in a nonproductive fashion. This hypothesis was based upon the observation that LLO is active when added to a solution of erythrocytes at 37°C and neutral pH (all of our hemolytic assays were carried out at 37°C, pH 7.4). To test this hypothesis, we pre- incubated LLO for 20 min in the absence of host cell membranes (in PBS at 37°C, pH
7.4 or 5.5) before performing the hemolytic analysis at 37°C (pH 7.4 or 5.5).
Consistent with the previous observation of Schuerch et al. [112], pre-incubating the toxin at neutral pH and 37°C inactivated LLO; whereas, at acidic pH some activity
A LDH Release B Cell Viability
1.4 140 1.2 120 1 100 0.8
OD 80 0.6 60 0.4 40 0.2 20
0 ViabilitytoControl Relative 0
C Cell Viability 24 h after treatment 140 Control 120 100 TX-100 80 BSA beads 60 40 BSA/LLO beads 20
ViabilitytoControl Relative 0
Figure 2.6 LLO-coated beads do not affect HepG2 cell viability. (A) HepG2 cells were incubated for 30 min at 37°C with BSA- or BSA/LLO-coated beads (MOI = 20). LDH released into the supernatant was measured immediately after treatment. As controls, we used untreated cells (control) and cells incubated with 0.2% TX-100 for 30 min. Cell viability was assessed immediately (B) or 24 h (C) after treatment by counting viable cells that excluded trypan blue. Results were the mean ± SEM (n≥3). 51
of the toxin was retained (Fig. 2.7A). However, if LLO was pre-incubated at 4°C for
20 min before performing the hemolytic assay at 37°C, LLO retained its activity at pH 7.4 and 5.5 (Fig. 2.7B). These data show that soluble monomers of LLO are inactivated at 37°C and pH 7.4 in the absence of membrane, but in the presence of membranes LLO can rapidly bind to the membrane and form pores.
In response to membrane perforation by pore-forming toxins, host cells reseal their membranes using a repair process that is activated upon the influx of extracellular Ca2+ [201]. To further demonstrate that low concentrations of LLO efficiently perforate HepG2 cells at physiological temperature and pH, we measured host cell perforation in the presence and absence of extracellular calcium. Membrane perforation was quantified by fluorescence imaging using the membrane impermeant dye ethidium homodimer. In solution this dye is weakly fluorescent, but once host cells are perforated, it enters the cells and associates with nucleic acids, which increases its fluorescence quantum yield. As shown in Fig. 2.7C and D, a baseline level of fluorescence was detected with host cells exposed to ethidium homodimer in calcium-free buffer at 37°C, pH 7.4 (Videos S1 and S2). LLO induces a massive entry of ethidium homodimer into cells incubated in Ca2+-free buffer (Videos S3 and S4), but not in the presence of 1 mM extracellular Ca2+. The extensive perforation of host cells was not due to the toxicity of ethidium homodimer, as in its absence LLO induces cell swelling and lysis with similar kinetics (Video S5). In total, these results showed that at 37°C, extracellular LLO efficiently perforates host cells at neutral pH.
52
A B LLO pre-incubated 20 min at 37 C LLO pre-incubated 20 min at 4 C
0.7 1.25 nM, pH 7.4 0.7 0.6 2.5 nM, pH 7.4 0.6 5 nM, pH 7.4
0.5 0.5 700
700 0.4 1.25 nM 0.4 OD
OD pH 5.5 0.3 0.3 1.25 nM, pH 7.4 0.2 2.5 nM, 1.25 nM, pH 5.5 pH 5.5 0.2 0.1 0.1 5 nM,
0 pH 5.5 0
0:01 0:41 1:01 0:01 0:21 1:40 0:21 1:21 1:40 1:61 0:41 1:01 1:21 1:61 Time (h:min) Time (h:min)
C D
50 2+ 45 Ca Free Medium + LLO 2+ 40 1 mM Ca + LLO T = 10 min T = 28.5 min T = 38 min
Free MediumFree 35 Ca2+ Free Medium 2+
Ca 30 25 20 A.F.I.(x100) + LLO 15
10 Free Free
2+ 2+ 5 T = 10 min T = 28.5 min T = 38 min
Ca 0
Medium LLO + 0 5 10 15 20 25 30 35 40 Time (min)
Figure 2.7 Extracellular LLO perforates erythrocytes and HepG2 cells at 37°C, pH 7.4. LLO was diluted in PBS pH 7.4 or 5.5 at 4°C in a 96 well plate (20 µl toxin in each well). (A) The plate was incubated at 37°C for 20 min, then, 180 µl of a 37°C suspension of sheep erythrocytes (0.25%) in PBS pH 7.4 or 5.5 was added to each well. (B) The plate was kept on ice for 20 min, then, 180 µl of a 37°C suspension of erythrocytes (0.25%) in PBS pH 7.4 or 5.5 was added to each well. In (A) and (B) following addition of erythrocytes, the plates were incubated at 37°C in a Power Wave 340 spectrophotometer (Bio-Tek) and absorbance (700 nm) was acquired every min [253]. Results show representative experiments (of 3), each performed in duplicate. In (C) and (D), perforation of HepG2 cells was measured by quantitative live cell fluorescence microscopy. HepG2 cells were incubated on the microscope stage at 37°C for 40 min with ethidium homodimer and in the presence or absence of 1 mM Ca2+. Phase contrast and fluorescence images were recorded at regular time intervals using a 100X objective and LLO was added after 10 min of incubation. Results were expressed as the averagefluorescence intensity (in arbitrary units) in the cells ± SEM of 5 movies for each condition. Scale bar = 10 µm. 53
2.3.5 Construction and characterization of novel LLO variants unable to form pores
To determine the importance of toxin oligomerization and pore formation in
LLO-induced bacterial and bead entry into host cells, we constructed LLO derivatives locked at different stages of the pore-forming mechanism. Studies performed with the
CDC perfringolysin O showed that the introduction of two cysteines at specific locations to form an intramolecular disulfide bond inhibits the hemolytic activity of the toxin [97, 254]. Depending on its location, the disulfide bond impedes the conformational remodeling required for formation of oligomers and/or pores, while toxin binding to host membranes is unchanged [254]. Based on these studies, we constructed and characterized LLOmL (monomer-locked) and LLOpL (prepore- locked). LLOmL was expected to bind to host membranes as a monomer unable to rearrange into a prepore complex. LLOpL was expected to bind to host membranes and oligomerize to form a stable prepore complex that cannot undergo the final transition into a pore. Neither mutant exhibited detectable hemolytic activity up to concentrations of 50 µM (higher concentrations were not tested). The loss of activity was due to the formation of the disulfide bonds, as reduction by dithiothreitol (DTT) fully restored the native hemolytic activity of the toxins (Fig. 2.8A). We next determined the oligomerization state of the toxins associated with erythrocyte membranes. As expected, LLO and LLOpL formed detergent-resistant high molecular weight complexes, whereas LLOmL failed to form such oligomers (Fig. 2.8B).
Importantly, the addition of DTT unlocked LLOmL, as shown by the formation of detergent-resistant high molecular weight oligomers. Finally, the arrangement of the toxins associated with cholesterol-rich lipid layers was analyzed by transmission electron microscopy (Fig. 2.8C). LLO and LLOpL formed characteristic arc- and ring- 54
shaped oligomers with a diameter of ~50 nm in the presence and absence of DTT.
LLOmL formed short linear assemblies that may be representative of the early stages of toxin oligomerization before formation of the prepore complex. Importantly,
LLOmL formed the typical arc- and ring-shaped oligomers when it was reduced by
DTT.
A B LLO LLOmL LLOpL
100 90 LLO 80 70 LLOmL + DTT 60 LLOpL + DTT 50 Oligomers LLOmL 40 LLOpL
% Hemolysis % 30 Dimer 20 10 Monomer 0 10-2 10-1 1 101 102 103 104 105 + DTT Toxin (nM) LLOpL
C LLO LLOmL + DTT +
Figure 2.8 Characterization of the LLO variants. (A) Hemolytic activity of LLO, LLOmL, and LLOpL was measured at pH 7.4, 37°C. DTT = dithiothreitol. (B) LLO, LLOmL, and LLOpL were incubated with erythrocyte ghost membranes, in the presence or absence of DTT, washed and suspended in 1% detergent. The samples were subjected to gradient electrophoresis, in the presence of 1% detergent, and LLO was detected by western blotting using anti-LLO antibodies. (C) Toxins were incubated for 1 h on cholesterol/DOPC lipid layers in the presence or absence of DTT. The samples were processed and analyzed by transmission electron transmission microscopy. Scale bar = 50 nm. TEM images provided by Dr. Elizabeth Wilson-Kubalek.
55
2.3.6 LLO-induced bacterial and bead entry requires host cell membrane perforation
We determined the role of LLO oligomerization into prepore and pore complexes in bacterial and bead entry into HepG2 cells. Native LLO induced bacterial and bead internalization, whereas neither LLOmL- nor LLOpL-coated beads were taken up by the cells (Fig.2.9A and B) even though all three bind to host cells with similar efficiency (Fig. 2.9C). We also used anti-LLO neutralizing antibodies, which were previously shown to prevent formation of LLO pores [255], to block
LLO-mediated uptake of beads. The LLO neutralizing antibodies markedly inhibited
LLO hemolytic activity and bead internalization, whereas control anti-LLO antibodies did not (Fig. 2.9D and E). The prepore to pore conversion of the CDCs is known to require high concentrations of membrane cholesterol [95]. We therefore prevented the formation of membrane pores by depleting host cholesterol using the cholesterol chelating agent methyl β-cyclodextrin (MβCD). Cholesterol depletion completely abrogated LLO-induced host cell perforation and the entry of BSA/LLO-coated beads into HepG2 cells (Fig. 2.9F and G). To demonstrate that inhibition of bead entry and host cell perforation was specifically due to cholesterol depletion and not to a secondary effect of the MβCD, we show that cholesterol repletion restored bead uptake and membrane perforation by LLO (Fig. 2.9F and G).
We also observed that BSA/LLO-coated beads formed pores in host cell membranes as detected by propidium iodide incorporation (Fig. 2.10A and B).
Membrane perforation is a key event in LLO-induced entry. Therefore, we hypothesized that a heterologous CDC should also increase L. monocytogenes internalization. LLO-deficient L. monocytogenes were coated with Streptococcus
56
Figure 2.9 Formation of pore complexes is required for efficient bacterial and bead entry into HepG2 cells. (A) LLO-deficient L. monocytogenes (Δhly, L.m.) were treated with coating buffer in the absence (white bars, C), or presence of LLO, LLOmL, or LLOpL. Cells were infected at 37°C for 30 min (MOI = 20). Samples were washed, fixed, and fluorescently labeled to enumerate bacterial entry by fluorescence microscopy. Results were the mean ± SEM (n≥3). (B) and (C) HepG2 cells were incubated with BSA- (Control; C) or BSA/toxin (LLO, LLOmL, or LLOpL)-coated beads for 30 min at 37°C (MOI = 20). Cells were washed, fixed, and labeled to enumerate bead entry (B) and association (C) by fluorescence microscopy. Results were the mean ± SEM (n≥3). (D) HepG2 cells were incubated with BSA/LLO-coated beads for 30 min at 37°C, in the presence or absence (C) of LLO-neutralizing or control antibodies. Cells were washed, fixed, and fluorescently labeled. Bead entry was measured by fluorescence microscopy and the results, mean ± SEM (n≥3) were expressed relative to the control (C). (E) Representative LLO hemolytic curves in the presence or absence (No Ab) of the neutralizing or control antibodies (10 µg/ml) were performed at pH 7.4, 37°C. (F) Control (C), cholesterol-depleted and -repleted HepG2 cells were incubated with BSA/LLO-coated beads for 30 min at 37°C. Bead entry was measured by fluorescence microscopy and the results, mean ± SEM (n≥3) were expressed relative to the control. (G) Control (C), cholesterol-depleted and -repleted HepG2 cells were incubated with various concentrations of LLO, 0.2% TX-100, or MEM (Control) at 37°C for 30 min. LDH release was measured using the TOX7 assay kit. Results are the mean ± SEM (n≥3).
57
Figure 2.9
A B C * * ** 1.4 ** 30 ** **
1.2 . 20 1 **
L.m 20 15 0.8 10 0.6 10 0.4
5 TotalBeads/Cell
0.2 % Intracellular% Beads % Intracellular% 0 0 0 C 1 2 1 C C Toxin (M) Toxin (5 M) Toxin (5 M)
Control LLO LLOmL LLOpL
** D ** E No Ab ** 100 + Control Abs 100 90 + Neutralizing Abs 90 80 80 70 70 60 60 50 50 40 40 30 (%) Hemolysis 30 20 20
RelativeEntry (% C) 10 10 0 0
C 2 10 20 2 10 20 (g/ml)
0.31
5.00
0.63 1.25 2.50
0.04
0.08 0.16
10.00 40.00 80.00 Neutralizing Control 20.00 Abs Abs LLO (nM)
F G ** TX-100 Control 100 1.8 1 nM LLO 5 nM LLO 20 nM LLO 90 1.6 80 1.4 70 60 1.2 50 1 40 0.8 30 0.6
20 0.4 RelativeEntry (% C) 10 LDHRelease (OD) 0.2 0 0 C Depletion Repletion Depletion Repletion (MCD) (MCD-chol) C (MCD) (MCD-chol)
58
pneumoniae [256] that exhibited a similar hemolytic activity to LLO (Fig. 2.11A).
Like LLO, PLY was able to mediate L. monocytogenes entry into HepG2 cells (Fig.
2.11B). These data suggested that the LLO structure did not contain unique features that were required to induce bacterial invasion
A B
Nuclei (DAPI) PI 600
500
400
AFI 300
200
100
0 BSA BSA/LLO
Figure 2.10 LLO-coated beads form small pores in HepG2 cells. HepG2 cells were incubated with BSA/LLO- and BSA-coated beads for 30 min at 37°C. Cells were pulse labeled with propidium iodide (100 µM) for 1 min, washed and fixed to quantify the fluorescence associated with the cells. (A) Representative DAPI and propidium iodide (PI) fluorescence images were acquired with a 20X objective. (B) Quantification of propidium iodide incorporation into the cells. Results represent the average fluorescence intensity (A.F.I.) per pixel in the cells and were the mean ± SEM (n≥3).
2.3.7 LLO activates a clathrin-independent, but dynamin-dependent internalization pathway
To demonstrate that host cell perforation by LLO leads to the formation of a micron sized internalization vesicle, we analyzed plasma membrane remodeling at the bead entry site. Scanning electron microscopy images showed the formation of plasma membrane extensions entrapping the BSA/LLO-coated beads after 15 min at
37°C. Within 30 min, we observed an increase in the number of beads completely enveloped by the plasma membrane (Fig. 2.12A), whereas, the formation of
59
membrane extensions was not induced by the BSA-coated beads. To demonstrate that
LLO induces entry of BSA/LLO-coated beads within a membrane-bound internalization vesicle, we quantified intracellular beads that colocalize with the early endosomal marker EEA1. After 30 min at 37°C, we observed that a large population
** A B 25 ** Control 100 . 20 LLO
80 L.m PLY 15 60 LLO 10 40 PLY
Hemolysis(%) 5
20 Intracellular %
0 0 0.1 1 10 100 C 1 1 Toxin (nM) Toxin (M)
Figure 2.11 PLY induces L. monocytogenes entry into HepG2 cells. (A) Hemolytic activity of LLO and PLY measured after 30 min at 37°C, pH = 7.4. A representative experiment (of 3) is presented. (B) HepG2 cells were incubated for 30 min at 37°C with LLO-deficient L. monocytogenes incubated in coating buffer in the presence or absence of 1 µM LLO or PLY. Bacterial internalization was measured by fluorescence microscopy. Results were expressed as the mean ± SEM (n≥3).
of intracellular beads was localized in endosomes to which the EEA1 marker had been recruited (Figs. 2.12B and C). We further characterized the endocytic molecules involved in this pathway. We measured the entry of BSA/LLO-coated beads in cells transfected with clathrin heay chain silencing RNA and in cells treated with the clathrin inhibitor chlorpromazine, or with dynasore, a dynamin inhibitor (Fig. 2.13A and B). The uptake of fluorescent transferrin was measured as a control for the inhibition of clathrin- and dynamin-dependent endocytosis (Fig. 2.13C). The results show that internalization of BSA/LLO-beads is dynamin-dependent and clathrin- independent.
60
A 15 min 30 min
C B PC + Total Extracellular (Alexa 365) (Alexa 568) 30 ** 25
20
15
10
5 % EEA1Positive% BeadsIntracellular 0 0 30 EEA1 (Alexa 488) Overlay Time (min)
Figure 2.12 LLO-coated beads are internalized into EEA1 positive endosomes. (A) HepG2 cells were incubated at 37°C with BSA/LLO-coated beads for 15 or 30 min, washed, fixed, and processed for scanning electron microscopy analysis. Scale bar = 1 µm. (B) HepG2 cells were incubated for 30 min at 37°C with BSA/LLO- coated beads, washed, fixed, and the extracellular beads were labeled with anti-BSA antibodies and a fluorescent secondary antibody (red). After permeabilization, EEA1 was labeled using anti-EEA1 antibodies and a fluorescent secondary antibody (green). Scale bar = 10 µm. PC = phase contrast. Arrows point out internalized beads and the arrowhead an internalized bead that massively recruited EEA1. (C) Results were expressed as the mean ± SEM (n≥3) percentage of intracellular beads that recruited EEA1. 61
Figure 2.13 LLO-coated beads are internalized by a clathrin-independent, dynamin-, F-actin- and tyrosine kinase-dependent pathway. (A) to (C) Clathrin heavy chain was knocked down in HepG2 cells by siRNA (siRNACHC) treatment or was inhibited by pre-incubating the cells for 30 min with 10 µM chlorpromazine (CPZ). Dynamin was inhibited by pre-incubating the cells for 30 min with 120 µM dynasore (DY). CPZ and DY were maintained in the cell culture medium throughout the experiments. (A) HepG2 cells were incubated with BSA/LLO-coated beads for 30 min at 37°C. Cells were washed, fixed, and bead entry was enumerated. Results were the mean ± SEM (n≥3) and were expressed relative to the control (C). (B) Representative western blot analysis of clathrin heavy chain (CHC) in cells treated with scrambled siRNA (siRNAc), specific siRNA (siRNACHC), or control untreated cells (C). The α-tubulin was used as a loading control. (C) Measurement of internalized transferrin (Tf)/Surface associated (Tf) and internalized (Tf) in control untreated cells (C), cells treated with siRNAc or siRNACHC, and cells treated with CPZ or DY. Only internalized (Tf) was measured in DY-treated cells due to DY fluorescence. Representative experiments (of 3) are shown. (D) HepG2 cells were incubated for 30 min with BSA/LLO-coated beads, fixed, and extracellular beads were labeled using anti-BSA antibodies and secondary fluorescent antibodies (red) and F-actin was labeled with fluorescent phalloidin (green). Scale bar = 5 µm, PC = phase contrast. The arrowhead and arrows point out extracellular and internalized beads, respectively. (E) HepG2 cells were pre-incubated in the presence or absence of cytochalasin D (CD), nocodazole (NO), genistein (GEN), LY294002 (LY), or wortmannin (W) at 37°C. Beads were added for 30 min in the presence of the inhibitors. The percentage of internalized beads was enumerated by fluorescence microscopy and the results were the mean ± SEM (n≥3). Experiments presented in A,B and C were performed by Dr. Eusondia Arnett.
62
Figure 2.13
63
2.3.8 F-actin remodeling is induced by LLO in a pore-dependent fashion and is required for bead internalization
Internalization of large particles such as bacteria generally requires the rearrangement of subcortical F-actin to form membrane extensions that engulf the particles [257]. Consistent with this idea, the membrane rearrangements observed by scanning electron microscopy (Fig. 2.12A) were accompanied by the recruitment of
F-actin at the bead entry site (Fig. 2.13D). Furthermore, bead entry was inhibited in cells treated with the F-actin depolymerizing drug cytochalasin D; whereas, microtubule integrity was not required for entry (Fig. 2.13E). F-actin remodeling involves the activation of the host signaling machinery. As a first approach to determine the transducers involved in entry, we have treated cells with inhibitors of tyrosine kinases (genistein) and phosphoinositide 3-kinases (PI3Ks) (LY294002 and wortmannin). Tyrosine kinases and PI3Ks are key transducers activated upstream from actin polymerization at the L. monocytogenes entry site [231]. We found that only tyrosine kinase(s) activation was critical for LLO-dependent entry (Fig 2.13E).
We also observed that purified LLO induces membrane ruffling in the 0.5 to 2 nM concentration range (representative Videos S6 and S7). Membrane ruffling started
117 ± 14.3 sec after the addition of 1.2 nM LLO and was optimal for 555 ± 145 sec
(calculated from 10 movies). This provided a convenient experimental model to determine whether pore formation and tyrosine kinases were involved in F-actin polymerization induced by LLO. LLO-induced membrane ruffling was F-actin- and tyrosine kinase-dependent as no ruffling was observed in cells stimulated by LLO in the presence of 0.5 µg/ml cytochalasin D or 250 µM genistein (Videos S8 and S9).
Importantly, membrane ruffling was only induced by LLO, but not by 0.5 to 50 nM
64
LLOpL (Video S10). These data provide a link between membrane perforation by
LLO and the remodeling of the F-actin cytoskeleton.
20
18 LLO, Ca2+ free medium 16 LLO LLO + CPZ 14 LLO + DY LLO + CD 12
10 A.F.I (x1000). A.F.I 8
6 + LLO 4
2
0 0 5 10 15 20 25 30 Time (min)
Figure 2.14. F-actin, dynamin, and clathrin are dispensable for the membrane repair pathway. HepG2 cell perforation was measured by quantitative live cell fluorescence microscopy. Cells were incubated on the microscope stage at 37°C for 30 min with 20 µg/ml propidium iodide (a cell impermeant nuclear dye) in the presence or absence (Ca2+ free medium) of 1 mM extracellular Ca2+. When inhibitors were used, cells were pre-incubated at 37°C with 10 µM chlorpromazine (CPZ), 160 µM dynasore (DY) for 30 min or 0.5 µg/ml cytochalsin D (CD) for 10 min and the inhibitors were maintained throughout the duration of the experiments. Phase contrast and fluorescence images were recorded at regular time intervals using a 100X objective. LLO (1.2 nM) was added after 5 min of incubation. Results were expressed as the average fluorescence intensity (in arbitrary units) in the cells ± SEM of 5 to 10 movies for each experimental condition.
2.4 Discussion
These studies revealed the existence of a novel pathway exploited by L. monocytogenes to gain entry into host cells. This pathway is activated in response to membrane perforation by the pore-forming toxin listeriolysin O (LLO). This is the first demonstration that a pore-forming toxin is able to induce the internalization of a 65
bacterial pathogen into host cells. We have used several approaches to show that LLO is crucial for efficient entry of L. monocyogenes into HepG2 cells and have used beads coated with LLO to decipher the molecular machinery underlying this novel pathway.
LLO is known to mediate L. monocytogenes escape from the endocytic vesicle following bacterial internalization into host cells [239]. The present findings demonstrate that LLO is also critical for L. monocytogenes internalization. Moreover, they show that LLO is sufficient to induce bacterial entry. A role for LLO in bacterial entry into nonphagocytic cells was previously proposed [212], whereas other studies did not identify such a role for LLO [119]. All of these studies relied upon the gentamicin survival assay that measures intracellular survival. The gentamicin assay is a powerful method, but it exhibits some limitations. First, it reports several stages of the host cell invasion process and cannot distinguish the role of LLO in bacterial entry from its role in intracellular survival. Second, LLO perforates host cells and likely allows gentamicin entry into the cells. As a result, the intracellular survival of
WT L. monocytogenes is underestimated in comparison to a LLO-deficient mutant. In the gentamicin assay performed in the present study, a low MOI (20) was used as higher amounts of bacteria led to substantial entry of gentamicin into the cells (MW
480), as reported by the incorporation of the cell impermeant dye ethidium homodimer (MW 857) (our unpublished data). Therefore, we developed an automated fluorescence-based assay to specifically and accurately measure bacterial association and entry [252]. With this approach, we demonstrate that LLO plays a critical role in bacterial entry, but not in bacterial association with host cells. Similar to LLO, we found that InlA and InlB did not significantly affect bacterial association, but affected bacterial intracellular survival and entry. This result is not surprising due to the 66
abundance of adhesins expressed by L. monocytogenes, as over ten surface adhesins have been identified [36-41, 43, 144]. When overexpressed, LLO significantly increased L. monocytogenes association with host cells. This result is in accordance with the observation that LLO promotes Bacillus subtilis attachment to epithelial cells
[258]. LLO and other CDCs were also shown to remain partially bound to the bacterial cell wall [259, 260]. Therefore, when overexpressed, the cell wall-associated toxin likely anchors the bacteria to host cells via binding to membrane cholesterol.
Indeed, LLO is well known to bind to cholesterol in biological and artificial membranes [93, 208].
Our study focused on elucidating the role of LLO in bacterial entry into host cells. We first investigated whether host cell perforation by LLO was required for activating this entry pathway. Although poreformation by LLO is pH-sensitive at
37°C, we demonstrated that extracellular LLO perforates host cells at neutral pH. This finding is supported by a previous study that also concluded that LLO is active at neutral and slightly basic pH values [208]. We constructed LLO variants to determine if LLO binding to host membranes, its oligomerization into a prepore complex, and pore formation are required for LLO-induced bacterial entry. The LLO variant unable to undergo the prepore to pore transition demonstrated that the formation of LLO pore complexes is a key event for bacterial and bead internalization into host cells. How does poreformation by LLO stimulate bacterial or bead uptake? The LLO entry pathway appears to be distinct from the canonical bacterial entry pathways. Bacteria are known to induce their entry by activating host receptors or by injecting effectors into the host cell cytosol [223, 224, 261]. Gram-positive bacteria do not have a type
III secretion system, although, they can produce CDC toxins to mediate the translocation of virulence factors [182, 184]. The observation that LLO alone was 67
sufficient to induce bacterial or bead entry ruled out this mechanism. The requirement for membrane perforation does not favor the simple model in which LLO acts by activating a signaling host receptor. Indeed, the LLOpL variant that binds to host cells and is able to rearrange into a prepore complex failed to induce bacterial entry. Also, another pore-forming toxin, PLY, could replace LLO in that function. We do not rule out the existence of a yet unknown receptor for LLO that would be shared by PLY, nevertheless, membrane perforation remains a key trigger for entry.
LLO induces the formation of internalization vesicles that accommodate large particles (bacteria or 1µm beads) via a cholesterol-, dynamin-, and F-actin-dependent, but clathrin-independent pathway. Several clathrin-independent and dynamin- dependent internalization pathways have been described including lipid raft- dependent pathways [262]. It is important to note that the role of cholesterol in this pathway might be complex, as cholesterol is a structural component of lipid rafts and is critical for LLO binding to host membranes and the formation of LLO pores.
However, the observation that LLO associates with and induces the coalescence of lipid raft microdomains favors the hypothesis of a lipid raft-mediated pathway [222].
We found that LLO induces the polymerization of actin at the bead entry site, and that
F-actin dynamics and tyrosine kinase activation are required for LLO-mediated bead entry. Using soluble LLO and LLOpL, we observed that membrane perforation by
LLO induces F-actin-dependent membrane ruffling. Interestingly, F-actin polymerization within membrane ruffles was tyrosine kinase-dependent. Together, our findings support a model in which host cell perforation by LLO leads to an internalization pathway that involves host tyrosine kinase-dependent stimulation of the actin cytoskeleton and the activity of dynamin. Similar to LLO, PLY promoted
68
bacterial entry into host cells and was shown to induce actin polymerization in neuroblastoma cells [263].
The fact that LLO induces internalization of bacteria in a pore-dependent fashion is reminiscent of the membrane repair pathway observed in eukaryotic cells exposed to pore-forming toxins. In response to the attack by pore-forming proteins, eukaryotic cells undergo membrane endocytosis to remove the pores from their plasma membranes [201, 246]. We identified host cell effectors, F-actin and dynamin, that are required for LLO-induced particle internalization but are dispensable for membrane repair (Fig. 2.14) [201]. In conclusion, membrane repair is likely a prerequisite for LLO-induced bacteria/bead entry, but the LLO-mediated entry pathway extends beyond or is distinct from the membrane repair pathway.
The mechanisms evolved by L. monocytogenes to gain entry into nonphagocytic cells are complex and involve several invasins such as InlA, InlB, and
LLO. The inlA, inlB, and hly genes are controlled by the central regulator of virulence genes, PrfA, and are highly up-regulated in vitro and in vivo during L. monocytogenes infection [33, 83]. Therefore, LLO is expressed together with InlA and InlB to mediate host cell invasion. Depending on the receptors expressed by host cells, InlA and InlB stimulate bacterial entry individually or in concert [133, 231]. LLO likely affects bacterial internalization in a large panel of cells because its major host receptor is cholesterol. Our study on LLO and previous studies on InlA and InlB showed that these molecules are individually sufficient to induce bacterial entry with high efficiency when overexpressed from a plasmid or coated on beads [46, 58, 264].
However, the expression level of these molecules is low in L. monocytogenes and their concerted activity likely ensures efficient bacterial uptake by host cells. The literature shows that most stages in the infectious lifecycle of a pathogenic bacterium 69
involve numerous factors, all working in concert [133, 265, 266]. A recent study demonstrated the importance of the cooperation between InlA and InlB during L. monocytogenes entry into host cells [133]. In this study it was shown that InlA ensured the specificity of bacterial recognition of intestinal cells, whereas InlB increased the InlA internalization rate. Likewise, we speculate that cooperation between LLO, InlA, and InlB occurs during host cell invasion. In this model, the three pathways would contribute simultaneously to the uptake of a given bacteria. Given the identified roles of LLO, InlA, and InlB in stimulating actin polymerization and endocytosis, cooperation between the three invasins likely involves both of these processes. The LLO-dependent entry pathway displays differences and similarities with the InlA and InlB pathways. They differ with respect to the mechanism used to activate the host cells, as membrane perforation is required in the LLO pathway, but not in the InlA and InlB pathways. Also, clathrin is involved in the InlA and InlB pathways [267], but not in the LLO pathway. However, the internalization induced by the three invasins shares common effectors such as host tyrosine kinases, dynamin, cholesterol, and F-actin [231]. Determining how host cells integrate the signals simultaneously generated by each invasin and whether the resulting pathway is the sum of the individual pathways or is a new pathway leading to efficient bacterial uptake constitutes an important goal for future research.
Pore-forming toxins are produced by numerous pathogens and may influence their uptake by host cells [268, 269]. In favor of this hypothesis, while this work was under revision, it has been published that the parasite Trypanosoma cruzi invades host cells by exploiting a host cell membrane repair mechanism in response to membrane damage [270]. Among CDC-producing bacteria, the genera Listeria,
Arcanobacterium, Bacillus, and Streptococcus include several pathogenic bacteria 70
that have been shown to invade nonphagocytic cells [271-276]. Interestingly, the
CDC intermedilysin (ILY) is required for internalization of Streptococcus intermedius
[272]. However, not all of the CDCs share this property, as a recent study showed that streptolysin O (SLO) inhibits internalization of Group A Streptococcus into keratinocytes [277]. Therefore, a fascinating avenue of research is to elucidate the molecular mechanisms underlying the role of pore-forming toxins in the regulation of host cell invasion by intracellular pathogens.
71
Chapter 3. Immunization with a listeriolysin O toxoid protects mice against
Listeria monocytogenes
3.1 Introduction
Listeria monocytogenes is a facultative intracellular pathogen responsible for listeriosis, a life-threatening foodborne disease in humans and numerous vertebrate species [235]. The intracellular lifecycle of L. monocytogenes is essential for pathogenesis and involves many virulence factors, including the pore-forming toxin listeriolysin O (LLO) [232]. LLO belongs to the largest family of bacterial pore- forming toxins, the cholesterol-dependent cytolysins, and plays a major role in host cell invasion in vitro and in vivo [242, 247, 278]. Therefore, inhibiting LLO pore- forming activity is expected to protect the host against L. monocytogenes infection. A previous study established that passive transfer of a LLO neutralizing monoclonal Ab significantly protected naive mice against sub-lethal and lethal L. monocytogenes infections [279]. This protection resulted from the LLO-neutralizing activity of the
Ab, since injection of a non-neutralizing anti-LLO Ab failed to protect mice against the pathogen [279]. This suggested that immunization leading to the production of
LLO neutralizing Abs could facilitate L. monocytogenes clearance. However, the natural immune response to L. monocytogenes involves limited production of Abs directed against LLO or any other L. monocytogenes antigens [279-281]. In the present study we genetically engineered a LLO toxoid (LLOT) and report that mice immunized with LLOT alone or with an adjuvant (cholera toxin) produced anti- 72
LLOT IgG1 and IgG2a Abs with potent LLO neutralizing activity. Mice immunized with LLOT plus adjuvant, produced higher levels of IgGs and were significantly protected against L. monocytogenes.
In cell culture models, a major role of LLO is to mediate the disruption of the phagosome, releasing L. monocytogenes into the cytosol, where bacteria replicate [239]. A large body of evidence supports the idea that LLO activity is not limited to mediating vacuolar escape. Indeed, LLO released by extracellular L. monocytogenes also perforates the host cell plasma membrane, which is proposed to substantially affect host cell invasion in vitro and in vivo [214, 217, 220, 282]. We hypothesized that in immunized mice, LLO neutralizing Abs could inhibit extracellular LLO, limiting the invasion of hepatocytes. In support of this hypothesis, we report that LLO neutralizing Abs efficiently block L. monocytogenes internalization into cultured hepatocytes.
3.2 Materials and Methods
3.2.1 Generation of LLOT
The construct coding for LLOT with the substitutions T515G and L516G was generated by PCR-based site-directed mutagenesis using the pET29b plasmid harboring wild type hly (the gene coding LLO) as a template and mutagenic primers
(Forward - 5-gaa ata tct cca tct ggg gca ccg ggg gtt atc cga aat ata gta ata aag-3 and
Reverse- 5-ctt tat tac tat att tcg gat aac ccc cgg tgc ccc aga tgg aga tat ttc-3) [282].
Mutations were confirmed by DNA sequence analysis at The Ohio State University
Plant-Microbe Genomics Facility and recombinant LLO and LLOT were purified as previously described [282].
73
3.2.2 Cholesterol binding assay
Spots (2 µl) of a serially diluted cholesterol solution were deposited onto a
PVDF membrane and air-dried. The membrane was blocked in 20 mM Tris buffer with 2% nonfat milk and 0.1% Tween 20 at pH 7.4. LLO or LLOT (20 µg/ml) were added at 4°C for 3 h. After washes, rabbit anti-LLO Abs (Abcam) were added for 1 h, followed by washes and incubation with horse radish peroxidase (HRP)- conjugated secondary antibodies. LLO was detected with ECL Western Blotting
Detection Kit (Amersham). We verified that the rabbit anti-LLO Abs equally recognize LLO and LLOT (data not shown).
3.2.3 Hemolysis assays
Erythrocytes (4 x 107 cells/ml) and serial dilutions of LLO or LLOT, in a final volume of 160 µl/well in a 96 well plate, were incubated in duplicate for 30 min at
37˚C and centrifuged. The absorbance (540 nm) of the supernatants was measured in a PowerWavex340 spectrophotometer. Triton X-100 and PBS served as positive and negative controls for hemolysis, respectively. The concentration of toxin leading to
50% hemolysis (EC50) was determined by nonlinear regression using Prism software (GraphPad Software, Inc.). For assays with purified serum IgGs, LLO and
IgGs were pre-incubated in duplicate on ice for 15 min before the addition of erythrocytes. Samples were transferred to the PowerWavex340 spectrophotometer at 37˚C and the absorbance (700 nm) was measured every minute for 30 min.
74
3.2.4 Bacterial and mammalian cell culture
Wild type (wt, DP10403S) and LLO-deficient (Δhly, DPL2161) L. monocytogenes were a gift from Dr. Dan Portnoy (U.C. Berkeley, California, USA).
Bacteria and human hepatocytes (HepG2 cells, ATCC HB-8065) were grown and seeded in 24-well tissue culture plates 48 h before bacterial infection as previously described [282].
3.2.5 Mouse immunization and infection
Four groups of mice, each including five 12-week old female C57BL/6 mice
(National Cancer Institute; Frederick, MD), were immunized by intraperitoneal injection of PBS (group 1), 20 µg of LLOT (group 2), 1 µg/ml cholera toxin (group 3), or 20 µg LLOT plus 1 µg/ml cholera toxin (group 4) on days 0, 7, and 14. Serum was collected on day 28 after blood centrifugation at 1,500 x g for 15 min at 4°C and levels of LLO-specific IgG Ab responses were evaluated by an antigen-specific
ELISA [283]. Mice were also inoculated with L. monocytogenes (DP10403S; 2 x 104) on day 28 by tail vein injection and were sacrificed 72 hours post-infection. Liver and spleen were collected and homogenized to enumerate the colony forming units
(CFUs) [284]. IgGs were purified from pooled serum using protein G agarose (Pierce) according to the manufacturer’s instructions. All protocols were approved by The
Ohio State University's Institutional Laboratory Animal Care and Use Committee.
3.2.6 L. monocytogenes internalization assay
HepG2 cells were incubated with wt or Δhly L. monocytogenes at a multiplicity of infection 5 (MOI 5) in the presence or absence of control or LLO- neutralizing monoclonal Ab (a gift from Dr. Pascale Cossart, Pasteur Institute, Paris, 75
France). After 30 min at 37°C, cells were washed, fixed, and labeled with fluorescent antibodies and DAPI to enumerate the numbers of extracellular and intracellular bacteria, and mammalian cells [282]. Bacterial association with and internalization into host cells was calculated by quantitative fluorescence microscopy, as described previously [252].
3.3 Results
3.3.1 Generation of a listeriolysin O toxoid (LLOT)
LLO belongs to the family of the pore-forming cholesterol-dependent cytolysins (CDCs) produced by over 30 Gram-positive bacterial species. CDCs are major virulence factors that primarily act by forming large (30-50 nm diameter) transmembrane pores across cholesterol-rich membranes [84]. These highly homologous toxins are organized in four domains; the C-terminal domain 4 contains several hydrophobic loops that mediate binding to target membranes [285]. One of these loops - loop L1 - contains a conserved threonine-leucine motif responsible for recognition of and binding to cholesterol [93]. Because cholesterol is strictly required for the formation of the pore complex, mutation in the cholesterol-binding motif is expected to inactivate all CDCs. We substituted the threonine-leucine pair with glycine residues in LLO yielding LLOT515G/L516G (LLOT). Recombinant LLOT was unable to bind to cholesterol and lyse erythrocytes, as previously reported for other CDCs with similar modification of their cholesterol-binding domain (Fig. 3.1)
[93].
76
A B 104 **
20 10 5 2.5 1.2 0.6 0.3 0.15 103
LLO ) 2
M 10 (n
LLOT 0 5
C 1
E 10
1
0.1 LLOLLOT
Figure 3.1 Characterization of LLOT. (A) Recombinant LLO and LLOT were incubated on a PVDF membrane pre-coated with a serial dilution of a cholesterol solution (20 to 0.15 mg/ml cholesterol in EtOH). LLO and LLOT binding to cholesterol was measured by adding primary anti- LLO Abs and HRP-conjugated secondary Abs, followed by chemiluminescent detection. A representative T experiment, of 2, is shown. (B) The EC50 of LLO and LLO was measured from four independent hemolytic assays, each performed in duplicate. P values were calculated using a two-tailed Student’s t-test (** = P <0.01).
3.3.2 Immunization with LLOT plus adjuvant protects mice against L. monocytogenes
We then determined if LLOT could be used to safely immunize mice against L. monocytogenes. Mice immunized with either LLOT alone or LLOT plus cholera toxin, an adjuvant that elicits B and T cell responses [286, 287], developed high levels of anti-LLOT IgGs (Table 2); although, the adjuvant greatly enhanced the production of anti-LLOT Abs. Furthermore, mice immunized with LLOT alone or LLOT plus cholera toxin produced more IgG1 than IgG2a, a profile of IgG subclasses previously seen when cholera toxin was given with inert antigens such as ovalbumin [283].
Interestingly, mice immunized with LLOT plus cholera toxin were significantly
77
protected against intravenous challenge with L. monocytogenes (Fig. 3.2). There was a 2-log decrease in bacterial burden in the spleen and liver in comparison to the non- immunized group and the groups that were treated with LLOT or cholera toxin alone.
Antibody responses IgG IgG1 IgG2a
Immunization Mouse Serum Serum Serum Serum Serum Serum # Dilution Log2 Dilution Log2 Dilution Log2 Titers Titers Titers Titers Titers Titers 1 131,072 17 524,288 19 16,384 14 2 262,144 18 1,048,576 20 16,384 14 LLOT 3 65,536 16 131,072 17 4,096 12 4 262,144 18 262,144 18 512 9 5 2,097,152 21 4,194,304 22 32,768 15
1 4,194,304 22 33,554,432 25 131,072 17 LLOT + 2 4,194,304 22 16,777,216 24 262,144 18 Cholera toxin 3 2,097,152 21 33,554,432 25 65,536 16 4 1,048,576 20 8,388,608 23 65,536 16
Table 2. Serum IgG antibody responses in mice immunized with LLOT. The levels of LLOT-specific IgG, IgG1 and IgG2a were determined by ELISA in serial dilutions (1:2) of sera from mice immunized with LLOT alone or LLOT plus cholera toxin. Serum dilution Ab titers were determined as the last dilutions of sera that gave an absorbance of > 0.1 above that of control sera from naïve mice. Results are expressed as log2 values of serum dilution titers.
3.3.3 LLO neutralizing Abs inhibit cell invasion
Our results are in accordance with previous studies that injected naive mice with a LLO neutralizing monoclonal Ab at the time of infection with L. monocytogenes. The LLO neutralizing Ab was reported to rapidly protect mice against L. monocytogenes, as bacterial burden in the spleen and liver was significantly decreased within the first 6 hours post-infection. The authors concluded that the Ab could act by neutralizing LLO in the extracellular and/or intracellular compartments.
A follow up study showed that LLO neutralizing Abs can block LLO activity within the phagosome of macrophages in vitro and in vivo; however, whether such Abs could
78
also inhibit extracellular LLO was not tested. Recently, several studies emphasized that extracellular LLO also plays a critical role in host cell invasion. In particular, extracellular LLO mediates bacterial internalization into hepatocytes in vitro and
A ** B ** 109 108
108 107
107 106
6 5
10 CFUs/Liver 10 CFUs/Spleen
105 104
104 103 PBS ADJ LLOT LLOT +ADJ PBS ADJ LLOT LLOT +ADJ
Figure 3.2 LLOT protects mice against L. monocytogenes. Mice (5 mice/condition) were immunized at weekly intervals for 3 consecutive weeks by intraperitoneal injection of PBS (negative control), cholera toxin (1 µg/ml) used as an adjuvant (ADJ), LLOT (20 µg/ml), or LLOT (20 µg/ml) plus cholera toxin (1 µg/ml). At week 4, mice were intravenously inoculated with 2 x 104 wt L. monocytogenes and sacrificed after 72 h to collect blood and enumerate bacterial colony forming units (CFUs) in the liver and spleen. (A and B) CFUs were enumerated in organ homogenates. Results are expressed as CFUs/organ and horizontal bars indicate medians. Statistical significance was calculated using a two-sided Mann-Whitney test, ** P < 0.01. Experiment performed with Eusondia Arnett. induces post-translational modifications in host cells that are critical for successful infection of the liver in mice. Serum IgGs purified from mice treated with LLOT along with the adjuvant cholera toxin, but not from control mice (treated with PBS or with cholera toxin alone), could neutralize LLO activity (Fig. 3.3A). Mice treated with LLOT alone could also neutralize LLO activity, but to a lower extent (data not shown), consistent with the lower level of anti-LLO Abs produced in this group of mice (Table 2).
79
A 1 B 0.9 ** 0.8 ** 0.7 100 PBS wt
0.6 ion t 80
a hly 0.5 LLO z
60 nali
OD700 0.4 LLO + IgG (LLOT + ADJ)
r e 0.3 40
LLO + IgG (ADJ) Int
0.2 % 20 Triton X-100 0.1 0 0 0 2 10 μg/mlAb 0 5 10 15 20 25 30 Time (min)
Figure 3.3 LLO neutralizing Abs inhibit L. monocytogenes internalization into hepatocytes. (A) IgGs were purified from pooled serum isolated from mice immunized with cholera toxin alone (ADJ) or LLOT plus cholera toxin (LLOT + ADJ) and tested for their ability to inhibit erythrocyte lysis by native LLO. As positive and negative controls, erythrocytes were incubated with PBS or Triton X- 100, respectively. (B) HepG2 cells were incubated for 30 min with wild type (wt) or LLO-deficient L. monocytogenes (Δhly) at MOI = 5, in the presence or absence of increasing concentrations of LLO neutralizing Abs. L monocytogenes internalization was measured by fluorescence microscopy. Results are the normalized mean ± SEM of three independent experiments. P values were calculated using a two-tailed Student's t-test (**=P < 0.01).
As an approach to determine if LLO neutralizing Abs could interfere with the activity of extracellular LLO, we measured the efficiency of LLO-mediated L. monocytogenes internalization into hepatocytes. In accordance with this hypothesis, we found that a monoclonal LLO-neutralizing Ab markedly inhibited internalization of wt L. monocytogenes, but not of the isogenic LLO-deficient strain, into hepatocytes (Fig.
3.3B).
3.4 Discussion
We generated the first LLO toxoid used to immunize mice against L. monocytogenes. Immunization with LLOT plus an adjuvant elicited substantial production of anti-LLO IgG1 and IgG2a Ab subclasses and protected mice against the pathogen. We propose that anti-LLO Abs could limit host cell invasion by neutralizing both the extracellular and intracellular LLO activities [288]. This 80
hypothesis does not exclude the potential contribution of a cell-mediated immune response. Importantly, the efficient production of anti-LLO antibodies of IgG1 subclass and class switching to IgG2a reflects the activation of Th1 and Th2 responses accompanied with the production of interferon gamma [289]. IgG2a were also produced in mice treated with LLOT alone, highlighting the immunogenic properties of the toxoid. The addition of cholera toxin increased the production of anti-LLO antibodies and was required for protection against L. monocytogenes.
Whether the adjuvant only amplified the immune response to LLOT or also qualitatively affected the immune response is yet to be determined. Future studies will characterize the immune responses to various doses of LLOT in the presence and absence of adjuvants.
LLO is indispensable for the pathogenesis of L. monocytogenes; therefore, elucidating the mechanisms of action of LLO is expected to facilitate the development of more effective strategies to fight listeriosis. In vitro, LLO is a multifunctional toxin that activates host cell signaling and bacterial internalization when secreted in the extracellular compartment [278, 282], and that mediates bacterial vacuolar escape when secreted in the phagosome [239]. It is unsettled whether both extracellular and intracellular activities of LLO are important for pathogenesis. The fact that LLO neutralizing Abs inhibit hepatocyte invasion in vivo and in vitro is consistent with a role for extracellular LLO in liver infection. However, further in vivo studies will be necessary to dissect the role of LLO in bacterial internalization into hepatocytes and its role in vacuolar escape.
L. monocytogenes has long been used as a model pathogen to study the adaptive immune response against intracellular bacteria. Early studies suggested that natural immunization against L. monocytogenes does not involve the production of 81
antibodies, but triggers T cell-mediated immunity and the activation of macrophages
[280, 290, 291]. The present study and the reports by Edelson et al. [279, 288], argue that when their production is induced, LLO neutralizing antibodies can efficiently decrease L. monocytogenes burden in vivo. Our work is also the first to show that immunization with a genetically inactivated LLO variant provides protection against the pathogen. Therefore, LLOT is a promising tool for the development of a vaccine against L. monocytogenes. Similarly, immunization with toxoids of cholesterol- dependent cytolysins expressed by other Gram-positive pathogens has been shown to protect against challenge with the toxins or infection [292-294].
LLO alone or in conjunction with L. monocytogenes, is used in several anti- tumor vaccine candidates, some of which are currently under clinical trial [286].
Indeed, LLO is a source for CD4+ and CD8+ T cell epitopes and exerts non-specific adjuvant properties [295]. The molecular basis of the LLO adjuvant properties is not well understood and may be unrelated to toxicity [296, 297]. Therefore, it will be of tremendous interest to further characterize the immunomodulatory functions of LLOT and whether in addition to being a potential vaccine antigen, it may also display adjuvant activity for unrelated antigens.
82
Chapter 4. Fluxes of Ca2+ and K+ are required for the LLO-dependent
internalization pathway of Listeria monocytogenes
4.1 Introduction
Numerous pathogenic microorganisms adopt an intracellular lifestyle to infect their host. Bacterial pathogens can induce their internalization into host cells by one of two major mechanisms [223]. The zipper mechanism involves bacterial surface molecules that specifically activate a receptor-mediated signaling cascade [223]. In the trigger mechanism, Gram-negative bacteria use a secretion apparatus that injects effectors into host cells to directly activate the internalization machinery [298]. More recently, a third invasion mechanism used by the bacterium Listeria monocytogenes and the parasite Trypanosoma cruzi was reported [270, 282]. This novel mechanism of pathogen internalization is stimulated upon perforation of the host cell plasma membrane. In the case of L. monocytogenes, membrane perforation is caused by the pore-forming toxin listeriolysin O (LLO) which belongs to the cholesterol-dependent cytolysin (CDC) family [282]. This invasion mechanism is not exclusive to LLO, since the CDC pneumolysin also stimulates L. monocytogenes internalization [282]. T. cruzi is believed to perforate host cells via an unidentified pore-forming protein or mechanical disruption [270]. Thus, damaging the host cell plasma membrane emerges as an invasion strategy shared by bacteria and parasites. This invasion process is not
83
the result of passive entry of the pathogen through a breach in the plasma membrane, but is due to the activation of the host cell endocytic machinery [270, 282]. Importantly, this invasion mechanism is relevant to numerous pathogens because pore-forming proteins are common virulence factors of viruses, bacteria, and eukaryotic intracellular pathogens [272, 285, 299-303].
The foodborne pathogen Listeria monocytogenes infects a large variety of host cells including cells that are normally nonphagocytic such as epithelial, endothelial, and fibroblastic cells [235]. L. monocytogenes can use several pathways to infect nonphagocytic cells [232, 304]. The surface invasins InlA (internalin) and InlB are known to stimulate the zipper mechanism of entry by activating the host cell receptors E-cadherin and c-Met (the hepatocyte growth factor receptor), respectively
[46, 58, 236]. In addition to InlA and InlB, LLO is sufficient to induce L. monocytogenes internalization via a pathway that requires the formation of the LLO pore complex, host cell tyrosine kinase signaling, F-actin polymerization, and dynamin [282].
The ability to mediate bacterial internalization is one of several effects of LLO on host cells. Initially identified as a major virulence factor that is critical for intracellular survival [115, 247], LLO mediates the escape of L. monocytogenes from the endocytic vacuole to the cytosol, where the bacterium replicates [115, 239]. LLO is also released by L. monocytogenes in the extracellular environment [305, 306], eliciting various host cell responses. These responses include the activation of mitogen-activated protein
(MAP) kinases [206, 207], NFκB [159], caspase-1, and the NLRP3 inflammasome
[209, 307]. Extracellular LLO also decreases SUMOylation [220] and histone phosphorylation [209, 217], and causes mitochondrial fragmentation [214] and the arrest 84
of protein synthesis [164]. A fundamental question is how can LLO exert so many activities? No protein receptor has been identified for LLO so far; therefore, it is likely that LLO acts non-specifically by punching holes across host membranes.
Perforation of the plasma membrane has multiple consequences, including, but not limited to ion fluxes, membrane depolarization, and changes in redox potential and osmotic pressure. The LLO pore is also very large (30 nm in diameter); therefore, even small proteins can be exchanged with the extracellular medium [181]. These events likely elicit multiple signaling pathways, potentially explaining the diverse host cell responses to LLO. In particular, localized changes in the intracellular concentration of Ca2+ significantly affect host-pathogen interactions [308]. LLO is known to
stimulate influx of extracellular Ca2+ [211] and the release of Ca2+ from intracellular stores [210, 309]. The influx of Ca2+ is crucial for repair pathways that restore plasma membrane integrity following perforation and many other cellular processes including gene transcription, intracellular trafficking, and cytoskeletal dynamics, all of which may affect the interaction of L. monocytogenes with host cells [201, 308, 310]. The known Ca2+-dependent activities of LLO include mitochondrial fragmentation [214] and membrane repair [282]. Variations in the intracellular concentration of K+ also affect host cell biology. Following host cell exposure to LLO, a decrease in the intracellular concentration of K+ leads to caspase-1 activation, histone H3 dephosphorylation, and the arrest of protein synthesis [209, 311]. In the following work, we evaulated the roles of Ca2+ and K+ fluxes triggered upon host cell perforation by LLO in L. monocytogenes internalization. Importantly, we tested for membrane damage caused by the bacterium
85
in the absence of ion fluxes and adapted the experimental approaches to mitigate such damage.
4.2 Materials and methods
4.2.1 Bacterial and mammalian cell culture
L. monocytogenes, wild-type (WT, 10403S) and the isogenic Δhly deletion strain
(DP-L2161) were grown overnight at 37°C in brain heart infusion (BHI) (BD
Biosciences). Overnight cultures were diluted 1/20 in BHI and grown at 37°C until
OD600 = 0.7-0.8. Bacteria were washed three times in phosphate-buffered saline (PBS) or the indicated medium. The human hepatocyte cell line HepG2 (ATCC HB-8065) was grown in minimum essential medium (MEM) (+) Earle's salts and L-glutamine
(Invitrogen), supplemented with 10% heat inactivated fetal bovine serum (HI-FBS;
Lonza), 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 µg/ml streptomycin (Invitrogen). HepG2 cells (105 cells/well) were cultured in
24 well tissue culture plates on glass coverslips at 37°C in 5% CO2 atmosphere for 48 h before infection. To test the role of Ca2+ influx and K+ efflux, standard, Ca2+-free, and high K+ media were prepared as follows. Standard medium (M1) contains 140 mM
NaCl, 5 mM KCL, 10mM Hepes, 1.5 mM CaCl2, 0.5 mM MgCl2, 0.36 mM K2HPO4,
2+ 0.44 mM KH2PO4, 5.5 mM D-glucose, and 4.2 mM NaHCO3 (pH 7.4) . Ca -free
+ medium (M2) is similar to M1, but lacks CaCl2. High K medium (M3) is similar to M1, but contains 5 mM NaCl and 140 mM KCL.
86
4.2.2 Toxin purification and polystyrene bead coating
Recombinant six-His-tagged LLO and LLOmL were purified from E. coli
BL21(DE3) as described previously [74, 282]. Carboxylate microspheres (Alexa 350, 1
µm diameter; Molecular Probes) were covalently coated with bovine serum albumin
(BSA) according to the manufacturer’s instructions. LLOmL was noncovalently bound to the surface of BSA-coated beads as previously described [28, 282].
4.2.3 Bacterial association with and internalization into host cells
HepG2 cells were incubated with 106 bacteria per well (MOI 2.5) in the indicated medium. The cell culture plates were centrifuged for 5 min (230 x g) at room temperature and incubated for 30 min at 37°C. Cells were washed with PBS, fixed with
3% paraformaldehyde (PFA) in PBS for 15 min, washed with PBS, and blocked for 1 h in 0.1 M glycine, 10% HI-FBS in PBS, pH 7.4. Extracellular bacteria were labeled with anti-L. monocytogenes rabbit polyclonal antibodies (GeneTex), then with anti- rabbit secondary antibodies conjugated to Alexa-488 (or Alexa-568) (Molecular
Probes). Samples were permeabilized with 0.2% Triton in PBS for 5 min and total bacteria were labeled with anti-L.monocytogenes antibodies and secondary antibodies conjugated to Alexa-568 (or Alexa-647) (Molecular Probes). Slides were mounted in
Prolong Gold with DAPI (Molecular Probes) to stain host cell nuclei [252]. To quantify the number of cells, 40 sets of images (phase contrast, DAPI, Alexa 488 and Alexa 568, or Alexa 568 and Alexa 647) were automatically acquired for each condition with a 20X objective. MetaMorph imaging and analysis software was used to enumerate the total number of bacteria (Nt), extracellular bacteria (Ne), and mammalian cells (Nc). The 87
percentage of internalized bacteria was calculated as (Nt - Ne) / Nt x 100. Bacterial association with host cells was calculated as Nt / Nc. In some instances, cells were labeled with an anti-TOMM20 mouse monoclonal antibody (Abcam) followed by anti- mouse Alexa 488-conjugated secondary antibodies. To quantify TOMM20 labeling,
phase contrast and TOMM20 fluorescence images were acquired with a 20X objective.
Fluorescence images were background corrected, cell perimeters were traced based upon the phase contrast images, and the average pixel fluorescence intensity (A.F.I.) of
TOMM20 labeling was quantified using the Metamorph software.
4.2.4 Invasion assay using polystyrene beads
HepG2 cells were washed and incubated with BSA-, or BSA/LLOmL-coated beads
(MOI = 20) in standard medium (M1). The cell culture plates were centrifuged for 5 min
(230 x g) at 4°C and then incubated at 37°C for 15 min before treatment with recombinant LLO, ionomycin (Sigma), and/or nigericin (Sigma) for 15 min. Cells were then washed, fixed with 3% PFA and blocked. Extracellular beads were labeled with anti-BSA rabbit antiserum (Sigma B1520) followed by goat anti-rabbit secondary antibodies conjugated to Alexa-568. The percentage of internalized beads [(number of intracellular beads/total number of beads) x 100] was quantified by fluorescence microscopy based on their unique (Alexa 350 for intracellular beads) or dual (Alexa 350
+ Alexa 568 for extracellular beads) fluorescence.
88
4.2.5 Propidium iodide incorporation assays
In fixed cells: HepG2 cells were incubated with WT or Δhly L. monocytogenes at the indicated MOI for 30 min at 37°C in standard (M1) or Ca2+-free (M2) medium, and pulse labeled with 100 µM propidium iodide (PI) for the last 5 min of incubation. Cells were then washed and fixed with 3% PFA and labeled with DAPI. Thirty sets of phase contrast and fluorescence images (DAPI and PI) were acquired with a 20X objective.
Fluorescence images were background corrected and the average pixel fluorescence intensity (A.F.I.) of the PI images was automatically quantified in the nuclear regions
(defined by the DAPI staining) using the MetaMorph analysis software. In living cells:
HepG2 cells were cultured in glass bottom culture dishes (MatTek; 35mm petri dish, 10 mm microwell) for 48 h. Dishes were transferred to the stage of an inverted fluorescence microscope and 20 µM PI was added. All movies were acquired on the microscope stage at 37°C using the 40X objective. Phase contrast and fluorescence images were acquired every 20 s for 980 s. LLO (1nM) was added 160 s after the initial acquisition. Cells were incubated in standard (M1), Ca2+-free (M2), or high K+ (M3) medium, as indicated.
Images were background corrected, cell perimeters were traced based upon the phase contrast images, and the pixel A.F.I. within the cell areas was quantified at each time point using the Metamorph software. At least 200 cells were analyzed in each condition.
4.2.6 Microscope equipment
Images were acquired on a motorized inverted epi-fluorescence microscope
(Axio Observer D1, Zeiss) equipped with 20 X Plan Neofluar (N.A. = 0.5) and 40X Plan
Neofluar (N.A. = 1.3) objectives, a high speed Xenon fluorescence emission device 89
(Lambda DG-4, 300 Watts, Sutter Instrument Company), an optical emission filter wheel Lambda 10-3 for the fluorescence imaging, a Smart shutter to control the illumination for phase contrast imaging (Sutter Instrument Company), and a back- illuminated, frame-transfer EMCCD camera (Cascade II 512, Photometrics). The filter sets for fluorescence were purchased from Chroma Technology Corporation: DAPI
(49000), GFP/FITC/Alexa488 (49002), Cy3/DsRed/Alexa568 (49005), Cy5 (49006).
Images were acquired and analyzed using the MetaMorph imaging software (Universal
Imaging).
4.2.7 Statistics
A minimum of three independent experiments were performed, each in duplicate, unless otherwise indicated. Data were expressed as mean ± Standard Error of the Mean
(SEM). P-values were calculated using a standard two-tailed Student’s t-test and determined significant if lower than 0.05. In figures, asterisks indicate a significant difference between the indicated experimental conditions (* p < 0.05; ** p < 0.01).
4.3 Results
4.3.1 Extracellular L. monocytogenes perforates host cells in a LLO-dependent manner
To determine if the amount of LLO secreted by L. monocytogenes is sufficient to perforate the host cell plasma membrane we incubated HepG2 cells for 30 min at 37°C with increasing concentrations of wild type (wt) or LLO-deficient (∆hly) L. monocytogenes and measured plasma membrane perforation. The extent of plasma 90
membrane perforation, which is proportional to the incorporation of the small, cell- impermeable dye propidium iodide (PI), significantly increased with the bacterial concentration, in a LLO-dependent manner (Fig. 4.1A and B). HepG2 cells were significantly perforated even when incubated with L. monocytogenes at a multiplicity of infection as low as 5 (MOI = 5) (Fig. 4.1B ).
Figure 4.1 Extracellular L. monocytogenes perforates host cells in a LLO- dependent manner. Wild type (wt) and LLO-deficient (Δhly) L. monocytogenes were incubated at various MOI with HepG2 cells for 30 min at 37°C. The cell impermeant dye propidium iodide (PI, 100 µM) was added to the cell culture medium during the last 5 minutes of incubation. After fixation and DAPI labeling, phase contrast (PC) and fluorescence images were acquired with a 20X objective. (A) Representative images. Scale bar is 40 µm. (B) The average fluorescence intensity of PI was automatically measured in the nuclear region by quantitative fluorescence microscopy using the Metamorph software. Results are expressed relative to HepG2 cells incubated without bacteria (-). Results are the mean ± SEM of at least three independent experiments, performed in duplicate (** P < 0.01).
91
4.3.2 Host cells damaged by LLO reseal their plasma membrane via a Ca2+- dependent, but K+-independent mechanism
The influx of extracellular Ca2+ is critical for resealing the plasma membrane of cells injured by LLO and other pore-forming proteins, or mechanical disruption [201,
282]. We investigated if K+ efflux is also required for resealing cells exposed to LLO.
Membrane perforation was measured at 37°C by live fluorescence imaging of HepG2 cells incubated with 1 nM of recombinant LLO and 20 µM PI. In standard culture medium (M1), which contains physiological concentrations of Ca2+ and K+, we observed minimal incorporation of PI. In Ca2+-free medium (M2), we observed a rapid and massive influx of PI due to the absence of membrane resealing, as expected. In high
K+ culture medium (M3), we observed minimal incorporation of PI, demonstrating that blocking K+ efflux does not prevent resealing of HepG2 cells damaged by LLO (Fig.
4.2). For representative movies, see supplemental material.
. 4.3.3 Host cells damaged by L. monocytogenes undergo Ca2+-dependent membrane resealing
We then determined if Ca2+-dependent membrane resealing observed in cells exposed to recombinant LLO also occurs in cells exposed to L. monocytogenes. HepG2 cells were incubated with increasing concentrations of wt and ∆hly L. monocytogenes for 30 min at 37°C in the presence (M1) or absence of extracellular Ca2+ (M2). Under these conditions, PI incorporation was significantly increased in cells incubated with wt
L. monocytogenes in Ca2+-free buffer compared to standard medium (M1) and to ∆hly in
92
Ca2+-free buffer (Fig. 4.3), confirming that LLO secreted by a low number of L. monocytogenes (MOI = 2.5) significantly perforates host cells and that host cells reseal their plasma membrane via a Ca2+-dependent process. At higher MOIs (MOI ≥ 50), the differences in PI incorporation between samples with and without extracellular Ca2+ becomes less pronounced, likely due to excessive perforation that is less efficiently resealed.
90 Standard (M1) 80 Ca2+-free (M2) 70 High K+ (M3) 60 50 40
30 + LLO PI (A.F.I.) (X 100) (A.F.I.) PI 20 10 0 0 200 400 600 800 1000 Time (s)
Figure 4.2 Host cells damaged by LLO undergo Ca2+-dependent, but K+- independent, membrane resealing. Perforation of HepG2 cells was measured by quantitative live cell fluorescence microscopy. Cells were incubated on the microscope stage at 37°C for 980 s with 20 µM PI in standard medium (M1), Ca2+-free medium (M2), or high K+ medium (M3). Phase contrast and fluorescence images were recorded at regular time intervals using a 40X objective and 1 nM LLO was added after 160 s of incubation. Results are expressed as the average fluorescence intensity (A.F.I.) of the cell area ± SEM. At least 200 cells were analyzed in each experimental condition. By 600 s, there is a statistically significant difference in PI incorporation in cells incubated in M1 compared to M2. There is no significant difference between samples incubated in M1 and M3 at any time points.
93
Figure 4.3 Host cells damaged by L. monocytogenes undergo Ca2+-dependent membrane resealing. Wild type (wt) and LLO-deficient (Δhly) L. monocytogenes were incubated at various MOIs with HepG2 cells for 30 min at 37°C in standard medium (M1) or Ca2+-free medium (M2). PI (100 µM) was added to the cell culture medium during the last 5 minutes of incubation. After fixation and DAPI labeling, phase contrast (PC) and fluorescence images were acquired with a 20X objective. (A) Representative images. Scale bar is 40 µm. (B) The average fluorescence intensity of PI was automatically measured in the nuclear region by quantitative fluorescence microscopy using Metamorph software. Results are expressed relative to HepG2 cells incubated without bacteria. Results are the mean ± SEM of at least three independent experiments performed in duplicate (* P<0.05; ** P< 0.01).
4.3.4 L. monocytogenes internalization into host cells requires Ca2+ and K+ fluxes
We next investigated if Ca2+ and K+ fluxes are involved in LLO-induced L. monocytogenes internalization. A previous study, which used the gentamicin survival assay, proposed that Ca2+ influx is essential for L. monocytogenes entry into nonphagocytic cells [212]. This conclusion was based upon the decreased recovery of viable intracellular bacteria incubated with host cells at MOI 50 in Ca2+-free buffer in comparison to standard medium. A difficulty with this assay is that in Ca2+-free buffer,
94
gentamicin (MW 480) is expected to massively diffuse into the cytosol of damaged cells, similar to what we observed with PI (MW 668) (Fig. 4.3), likely invalidating the gentamicin assay. To minimize the damage caused by LLO in the absence of extracellular Ca2+, we incubated HepG2 cells for 30 min at 37°C with L. monocytogenes at a low MOI (2.5) and used an immunofluorescence-based assay to quantify L. monocytogenes internalization [252]. This assay relies on antibodies (MW 150 kDa) to distinguish extracellular from intracellular bacteria, the large size of which should limit their diffusion across perforated host cells. After cell fixation and blocking, we assessed whether or not plasma membrane perforation by LLO in Ca2+-free medium allows for the diffusion of antibodies. We labeled HepG2 cells with an antibody recognizing the cytosolic mitochondrial epitope TOMM20. We observed an increase, although statistically non-significant, in permeabilization to antibodies in cells incubated with wt
L. monocytogenes in Ca2+-free medium in comparison to standard medium (Fig. 4.4A).
To ensure that this diffusion of antibodies does not affect the readout of bacterial internalization, we co-labeled TOMM20 (Alexa 488), extracellular (Alexa 568) and total
(Alexa 647) bacteria, and host cell nuclei. In standard medium, LLO plays an important role in L. monocytogenes internalization, as previously reported (Fig. 4.4B) [282]. In
Ca2+-free medium, wt L. monocytogenes internalization efficiency was significantly decreased in comparison to cells incubated in standard medium. Indeed, in these conditions the internalization level was as low as ∆hly L. monocytogenes in standard medium (Fig. 4.4B) [212]. Results were unchanged when damaged cells, defined as
TOMM20 positive cells, were excluded from the measurements (data not shown).
Together, these results show unequivocally that L. monocytogenes internalization into 95
HepG2 cells is Ca2+-dependent and that the level of perforation of the cells under these experimental conditions did not skew our ability to quantify L. monocytogenes internalization by fluorescence microscopy. To address the importance of K+ efflux, we compared the internalization efficiency of L. monocytogenes in standard and high K+ media (M3) (Fig. 4.4B). The internalization efficiency of wt L. monocytogenes was significantly decreased in high K+ medium, whereas the internalization efficiency of
Δhly L. monocytogenes remained unchanged. We obtained similar results when we added 135 mM KCL to the medium without removing NaCl (data not shown).
A B ** ** ** 100 100
80 80
60 60 **
TOMM20 (A.F.I.) TOMM20 40 40 Internalization %
20 20
0 0 M1M2 M1 M2 M3 wt ∆hly
Figure 4.4. L. monocytogenes internalization into host cells requires Ca2+ and K+ fluxes. HepG2 cells were incubated with wild type (wt) and LLO-deficient (Δhly) L. monocytogenes (MOI = 2.5) for 30 min at 37°C in standard (M1), Ca2+-free (M2), or high K+ (M3) media. (A) After fixation and labeling, the diffusion of anti-TOMM20 antibody into HepG2 cells was measured by quantitative fluorescence microscopy in cells incubated with wt L. monocytogenes in M1 and M2. Results are the A.F.I. of cells ± SEM from three independent experiments performed in duplicate. (B) Bacterial internalization was measured by quantitative fluorescence microscopy and expressed relative to wt bacteria in M1. Results are the mean ± SEM from three independent experiments performed in duplicate. (** P < 0.01).
96
Modifying ion concentrations in the cell culture medium did not affect L. monocytogenes association with host cells (data not shown). Together, these results indicate that both Ca2+ and K+ fluxes are required for LLO-dependent internalization of
L. monocytogenes.
4.3.5 Ca2+ influx and K+ efflux induced by ionophores are sufficient to activate the internalization of large cargoes
We investigated if fluxes of Ca2+ and/or K+, in the absence of host cell perforation and any virulence factors, were sufficient to induce the internalization of large cargoes.
We measured the internalization efficiency of 1 µm fluorescent polystyrene beads coated with bovine serum albumin (B-BSA) or with BSA and LLOmL (B-
BSA/LLOmL). LLOmL is a monomer-locked variant of LLO that is unable to form pores.
B-BSA and B-BSA/LLOmL were previously shown to remain extracellular when incubated with HepG2 cells [282]. Exogenously added LLO was sufficient to stimulate entry of B-BSA and B-BSA/LLOmL in a dose-dependent manner (Fig. 4.5A). This further demonstrates that the formation of LLO pores is required to induce uptake of the beads. To establish if ionic fluxes could replace LLO, we treated cells with Ca2+ and K+ ionophores. The Ca2+ ionophore ionomycin was sufficient to induce a low level of
+ internalization of both BSA and BSA/LLOmL beads, whereas the K ionophore nigericin did not induce bead internalization (Fig. 4.5B). We next treated cells with a mixture of ionomycin and nigericin to cause simultaneous fluxes of Ca2+ and K+. We found that bead internalization occurred to a much greater extent than with ionomycin alone. Importantly, in the presence of nigericin, the ionomycin concentration could be 97
decreased to achieve an entry level similar to that induced by LLO (Fig. 4.5C).
Therefore, simultaneous fluxes of Ca2+ and K+ are sufficient for the internalization of large cargoes, such as a bacterium.
A B 10 30 ** ** 9 ** 25 ** 8 ** 7 20 * ** 6 15 5 * ** 4 10 3
% % Internalization 2 % Internalization % 5 1 0 0
(μM)
0 0.5
0.5 2
1
1
1.5
2
1.5
0.5 2
5
1
10 1 (LLO nM) Io Ni
C 20 *
18 B-BSA B-BSA/LLOmL 16 14 12 * 10 8 6
% Internalization % 4 2 0 + + Io + Ni +
Figure 4.5 Ca2+ influx and K+ efflux induced by ionophores are sufficient to activate the internalization of large cargoes. B-BSA or B-BSA/LLOmL(MOI = 20) were incubated with HepG2 cells for 15 min in standard medium at 37˚C, followed by treatment with LLO or ionophores for an additional 15 min. (A) Cells were treated with LLO. (B) Cells were treated with ionomycin (Io) or nigericin (Ni). (C) Cells were treated with 1 μM ionomycin (Io) and/or 10 μM nigericin (Ni). Cells were washed, fixed, and extracellular beads were fluorescently labeled. Results are the mean ± SEM of at least three independent experiments (* P < 0.05; ** P < 0.01).
98
4.4 Discussion
The present study demonstrates that host cells are significantly perforated by extracellular L. monocytogenes in a LLO-dependent manner. Importantly, perforation is substantial even within 30 min and at a multiplicity of infection as low as 2.5. We show for the first time that host cells undergo Ca2+-dependent membrane resealing during L. monocytogenes infection. Importantly, Ca2+ influx and K+ efflux are essential regulators of LLO-dependent L. monocytogenes internalization into HepG2 cells. Finally, Ca2+ influx alone, induced by ionomycin in the absence of plasma membrane perforation, is sufficient to induce plasma membrane rearrangements and internalization of polystyrene beads; however with a low efficiency. This suggested that in addition to
Ca2+ influx, K+ efflux is required for efficient internalization. In support of this, significant uptake of beads was observed in cells treated with a combination of ionomycin and nigericin to induce simultaneous Ca2+ and K+ fluxes. Together, these results support the hypothesis that host cell plasma membrane perforation by pore- forming proteins induces non-specific uptake of large cargoes via inducing ion fluxes.
One important message of the present study is that LLO produced by L. monocytogenes perforates host cells during the early stage of invasion. This process was previously neglected, likely because perforation was unnoticed due to rapid Ca2+- dependent membrane resealing that maintains cell integrity and viability during L. monocytogenes infection. This finding further confirms that the activity of extracellular
LLO during the intracellular lifecycle of L. monocytogenes deserves further study. One such activity is the control of bacterial internalization. Previous studies demonstrated that host cell perforation by LLO activates the internalization of large cargoes such as L. 99
monocytogenes [282]. This activation requires tyrosine kinase activity, F-actin polymerization, and dynamin, ultimately leading to the formation of an early endosome that contains the bacterium or 1 µm bead. The present work identified Ca2+ and K+ fluxes as critical factors that regulate membrane dynamics and internalization following host cell perforation.
The influx of extracellular Ca2+ subsequent to plasma membrane injury caused by pore-forming toxins or mechanical disruption is essential for the activation of membrane repair [181, 188]. If Ca2+ influx is prevented, the plasma membrane is not resealed and cells can rapidly lyse. The resealing machinery of cells exposed to diverse pore-forming toxins has recently been addressed and it appears that plasma membrane trafficking is intimately linked to the resealing process via endocytosis or shedding of microvesicles that contain the toxin pores [165, 190, 204, 246, 312]. It was proposed that cells injured by the CDC streptolysin O (SLO), produced by Streptococcus pyogenes, are repaired via
Ca2+-dependent endocytosis of SLO pores [201]. In this model, an increase in cytosolic
Ca2+ activates rapid exocytosis of lysosomes, which release acid sphingomyelinase
(ASM). ASM catalyzes the synthesis of ceramides on the outer leaflet of the plasma membrane leading to endocytosis of the SLO pores via a clathrin-, dynamin-, and F- actin-independent route [202]. We expect that cells injured by LLO and SLO, which form a similar pore, use a similar repair mechanism [285]. In support of this, we report that resealing of cells injured by LLO, similar to SLO, is Ca2+-dependent, and clathrin-, dynamin-, and F-actin- independent [282]. It was proposed that T. cruzi injures host cells to exploit the Ca2+- and ASM-dependent, but clathrin-, dynamin-, and F-actin- independent resealing process to gain entry into host cells [270]. Based upon the 100
findings reported in the present manuscript and past studies [282], this model is not sufficient to account for LLO-dependent L. monocytogenes internalization into host cells. While LLO-mediated internalization and plasma membrane resealing have a shared requirement for Ca2+ influx and are both independent of clathrin, there are important differences between the two processes. Unlike the membrane resealing process, LLO-induced L. monocytogenes internalization is F-actin-, dynamin-, and K+- dependent [282]. Our results confirm that ion fluxes are critical regulators of endocytosis following host cell perforation, but reveal important differences between the sensu stricto plasma membrane resealing process and LLO-mediated L. monocytogenes internalization. It will be necessary to further dissect the events occurring during the membrane repair process in order to establish if additional events, beyond sensu stricto resealing, involve F-actin, dynamin, and K+ efflux. Alternatively, LLO-induced bacterial uptake may use a pathway that occurs downstream of, or that is distinct from the repair of the plasma membrane.
K+ efflux subsequent to cell perforation by LLO and other pore-forming toxins is known to be responsible for the activation of caspase-1 [174, 217], histone modification
[209] autophagy, and protein synthesis [164]. The finding reported here that K+ efflux also regulates the uptake of the pathogen is novel. How host cells detect and respond to variations in intracellular K+ is also unsettled. Whereas Ca2+ is a central regulator involved in a large variety of processes including endocytosis; few studies have addressed the role of K+ in endocytosis. It was previously shown that lowering intracellular K+ concentrations decreases clathrin-coated pit formation and receptor endocytosis [313]; however, other cargoes could still be internalized [314]. The LLO- 101
dependent uptake of L. monocytogenes is clathrin-independent. Therefore it is likely that low K+ levels facilitate an alternative, clathrin-independent pathway.
It appears that signaling downstream of Ca2+ and K+ fluxes is complex and involves several pathways that may, or not, cross-talk with one another. For example, histone dephosphorylation induced by LLO is Ca2+-independent and K+-dependent; whereas bacterial internalization is both Ca2+- and K+-dependent. When cells were treated with ionophores in an attempt to reproduce the effect of LLO, without creating membrane pores, the Ca2+ ionophore ionomycin was sufficient to induce the internalization of polystyrene beads. However, the ionomycin concentration required to induce this event approached amounts that were detrimental to the cells, as cells started to round up during the course of the experiment. The Ca2+ influx stimulated by this concentration of ionomycin was capable of influencing membrane dynamics and internalization, but with low efficiency. In contrast, treating the cells with the K+ ionophore nigericin had no such effect, even at high concentrations. In an attempt to initiate multiple fluxes at once, as occurs following perforation of the plasma membrane by LLO, we treated cells with a combination of nigericin and ionomycin. The addition of both ionophores at concentrations that were insufficient to induce internalization when added alone lead to enhanced levels of bead internalization, approaching levels induced by LLO. This result provides evidence that host cells likely integrate responses generated from multiple signals that are mobilized by ion fluxes. The way in which host cells sense changes in K+ levels, and the nature and extent of cross-talk between the pathways induced by ion fluxes following membrane perforation are exciting avenues
102
for future work. Finally, it will be important to determine if additional pathogens also cause local plasma membrane lesions to facilitate their internalization into host cells.
103
Chapter 5. Assessing the roles of LLO, InlA, and InlB during the L. monocytogenes
intracellular lifecycle in hepatocytes and placental cells
5.1 Introduction
The ability to invade nonphagocytic cells is a critical feature of L. monocytogenes pathogenesis. In contrast to entry into phagocytes such as macrophages and neutrophils,
L. monocytogenes must express virulence factors to activate the host cell endocytic machinery of nonphagocytic cells [304]. The proteins of the internalin operon — InlA and InlB — are well characterized virulence factors that are responsible for mediating entry into nonphagocytic cells in vitro, and for crossing epithelial and endothelial barriers in vivo [126, 127]. InlA binds to its receptor, E-cadherin, to induce L. monocytogenes entry into epithelial cells, while InlB binds to its primary receptor, c-Met, as well as
GAGs (glucosaminoglycans) and the receptor for the globular domains of complement component C1q (gC1qR) to induce entry into a wide variety of cell types including fibroblasts, endothelial cells, and epithelial cells [46, 47]. In vivo, InlA and InlB are vital for crossing the intestinal, placental, and possibly blood-brain barriers [126, 127, 145].
More recently, listeriolysin O (LLO) was identified as a bacterial invasin that is necessary for efficient invasion of the human hepatocyte cell line, HepG2, and is sufficient to induce internalization by a novel mechanism that requires pore formation
104
which stimulates actin- dynamin- and tyrosine kinase-dependent endocytosis [282]. The ability of InlA and InlB to induce entry depends on the expression of their receptors on the target cell. However, the receptor for LLO, cholesterol, is expressed in all cells that are infected by L. monocytogenes, suggesting that LLO may mediate invasion into all eukaryotic cells, but it has yet to be determined if this pathway is widespread.
In the following studies, we have determined and compared the roles of LLO,
InlA, and InlB in bacterial association with and internalization into human hepatocytes and placental cells, two cell types that are infected by L. monocytogenes during pathogenesis. Using a quantitative immunofluorescence approach, we found that the
LLO-mediated internalization pathway is not restricted to HepG2 cells, but also, that
LLO does not function as an invasin in all nonphagocytic cells [252]. Surprisingly, the internalins InlA and InlB were not critical for bacterial internalization into all cells expressing their receptors, suggesting that the role for these factors might be distinct from internalization. Our results suggest that the roles for these three invasins during infection may be more variable and more complex than is currently appreciated.
5.2 Materials and methods
5.2.1 Bacterial strains and plasmids
WT L. monocytogenes (DP10403S), isogenic Δhly (DPL2161), ΔinlA (DPL4405), and ΔinlB (DPL4406) deletion mutants were kindly provided by Dr. Dan Portnoy (U.C.
Berkeley, California, USA). WT LO28 L. monocytogenes and the isogenic transposon insertion LO28 hly::Tn917 mutant were gifts from Dr. Pascale Cossart (Pasteur
Institute, Paris, France) [315]. Bacteria were grown overnight at 37° C in brain heart 105
infusion (BHI) (BD Biosciences). For invasion assays, overnight cultures were diluted
1/20 in BHI and grown at 37°C until OD600 = 0.7–0.8. Bacteria were washed three times in phosphate-buffered saline (PBS) and diluted to the indicated multiplicity of infection
(MOI) in cell culture medium without serum.
5.2.2 Mammalian cell culture
The human hepatocyte cell lines HepG2 (ATCC HB-8065), Hep3B (ATCC HB-
8064), and PLC5 (ATCC CRL-8024), and the cervical epithelial HeLa cell line (ATCC
CCL-2), were grown in minimum essential medium (MEM) (+) Earle's salts and L- glutamine (Invitrogen), supplemented with 10% heat inactivated fetal bovine serum (HI-
FBS; Lonza), 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 µg/ml streptomycin (Invitrogen). The hepatocyte cell line Huh7 was grown in Dulbecco's modified Eagle's medium (DMEM) with 4 mM L-glutamine adjusted to contain 1.5 g/L sodium bicarbonate and 4.5 g/L D-glucose (Invitrogen), and supplemented with 10% HI-FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin.
Human derived placental epithelial BeWo cells (ATCC CCL-98) were grown in
DMEM/F12 (1:1) (Invitrogen) supplemented with 10% HI-FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin. Mammalian cells were maintained at 37°C in 5% CO2 atmosphere.
5.2.3 BeWo cell fusion
To chemically fuse BeWo cells, 1.5 x 104 cells/well were seeded in 24-well tissue culture plates on coverslips coated with 0.2% gelatin. After 24 h, cell culture medium 106
was replaced with medium containing 2.5 µM 5AZC. 24 and 48 h later, cell culture medium was exchanged with fresh medium containing 2.5 µM 5AZC and 0.5 µM forskolin. After an additional 24 h, the cell culture medium was exchanged with fresh medium containing 0.5 µM forskolin. Fused BeWo cells were infected 96 h after seeding. BeWo cell fusions were performed by Dr. Eusondia Arnett.
5.2.4 Measurement of bacterial association and entry into host cells
HepG2, Hep3B, Huh7, and PLC5 hepatocyte cell lines were seeded in 24-well tissue culture plates and grown for 48 h (HepG2; 1x105 cells/well. Hep3B; 0.75x105 cells/well. Huh7; 0.75x105 cells/well. PLC5; 0.75x105 cells/well) 24 h (HeLa; 0.5×105 cells/well), or 96 h (fused BeWo; 1.5x104 cells/well. nonfused BeWo; 0.85x104 cells/well). Hepatocytes were infected at MOI 20. Confluent monolayers of BeWo cells were infected with 1 x 106 bacteria/well.The plates were centrifuged for 5 min (230 x g) at room temperature and incubated for 30 min at 37°C. Cells were washed with PBS, fixed with PBS/4% paraformaldehyde (PFA) for 15 min at room temperature, then washed with 0.1 M glycine in PBS and incubated for 1 h in blocking solution (0.1 M glycine, 10% HI-FBS in PBS, pH 7.4). Following fixation and blocking, extracellular bacteria were labeled with a rabbit polyclonal anti-L. monocytogenes antibody
(GeneTex) followed by an anti-rabbit secondary antibody conjugated to Alexa488
(Molecular Probes). Cells were then permeabilized with 0.5% Triton for 5 min and total bacteria were labeled with a rabbit polyclonal anti-L. monocytogenes antibody followed by an anti-rabbit secondary antibody conjugated to Alexa568 (Molecular Probes). After immunofluorescence labeling, cells were mounted on coverlips with ProLong Gold
107
antifade containing DAPI (Molecular Probes). To quantify bacteria and mammalian cells, 40 sets of images (DAPI, Alexa 488, Alexa 568, phase contrast) were acquired for each experimental condition using a 20 X objective. MetaMorph imaging and analysis software was used to enumerate the total number of bacteria (Nt), extracellular bacteria
(Ne), and mammalian cells (Nc) [252]. The percentage of internalization was calculated as (Nt - Ne)/Nt x 100. Bacterial association with host cells was calculated as Nt/Nc.
5.2.5 Isolation and infection of placental villi
Placental villi were isolated from term human placenta with IRB approval, using standard protocols. Villi were dissected from placenta in ice-cold RPMI (Invitrogen).
Immediately following dissection, single villi were placed in wells of 24-well tissue culture plates in RPMI and infected with 5 x 106 L. monocytogenes (wt or Δhly) per explant, for 1 h at 37°C. Villi were then washed 3 times for 5 min each, and treated with
15 µg/ml gentamicin, then incubated further at 37°C. At 1, 3, 10, and 20 h post- infection, villi were washed with PBS and homogenized. Homogenates were serially diluted in PBS and plated on BHI agar. CFUs were enumerated after 48 h at 37°C.
5.2.6 Gentamicin survival assays
HepG2 cells were seeded at 1 x 105 cells/well in 24-well tissue culture plates 48 h before infection. HepG2 were infected with the L. monocytogenes strains at MOI 20.
BeWo, fused and nonfused, were seeded in 24-well tissue culture plates 95 h before infection at 1.5 x 104 cells/well and 0.85 x 104 cells/well, respectively. BeWo seeded for fusion were chemically treated as described above. BeWo cells were infected with 1 x 108
106 bacteria/well. The plates were centrifuged at room temperature for 5 min (230 x g) and incubated for 30 min at 37°C. Cells were then washed and incubated with 15 µg/ml gentamicin for 1 h. Cells were washed three times with PBS and lysed with 0.2% Triton
X-100 in H2O. Serial dilutions of cell lysates were immediately performed in PBS and plated on BHI agar. CFUs were enumerated after 48 h at 37°C.
5.2.7 Propidium iodide incorporation assays
Nonfused or fused BeWo were washed in buffer containing150 mM NaCl, 1 mM
MgCl2, 5 mM KCL, 20 mM Hepes, 10 mM Glucose, and with or without 1 mM CaCl2, then incubated in buffer with or without 1mM CaCl2 + 20 µM PI ± 1 nM LLO for 30 min in a 37°C water bath. Cells were then washed and fixed with 4% PFA and labeled with DAPI. Thirty sets of phase contrast and fluorescence images (DAPI and PI) were acquired with a 20X objective. Fluorescence images were background corrected and the average pixel fluorescence intensity (A.F.I.) of the PI images was automatically quantified in the nuclear regions (defined by the DAPI staining) using the MetaMorph analysis software.
109
5.3 Results
5.3.1 LLO-mediated L. monocytogenes internalization is not restricted to HepG2 cells
As a first step to determine if LLO functions as an invasin into multiple cells, we compared bacterial association and entry of wt and LLO-deficient strains of L. monocytogenes into various hepatocyte cell lines. Four human hepatocyte cell lines
(HepG2, Huh7, PLC5, and Hep3B) were infected for 30 min at MOI 20, and bacterial association and entry were quantified by immunofluorescence microscopy. While LLO had no effect on bacterial association in all tested cell lines, entry of the LLO-deficient strains was significantly inhibited into all of the hepatocytes tested. Thus, the the LLO- mediated internalization pathway is shared by multiple human hepatocyte cell lines (Fig.
5.1). Infection of human epithelial HeLa cells, which are commonly used as a model to study the intracellular lifestyle of L. monocytogenes, with two bacterial strains (10403S;
LO28) and their isogenic LLO-deficient mutants (10403S Δhly; LO28 hly::Tn917), demonstrated that the role for LLO in entry was not specific to hepatocytes or to one strain of L. monocytogenes (Fig. 5.2) [282].
110
HepG2 Huh7
Relative Association Relative Entry Relative Association Relative Entry 160 140 250 140 140 120 ** 120 200 120 ** 100 100 100 80 150 80 80
60 % WT % % WT % 60 % WT % 60
100 WT % 40 40 40 50 20 20 20 0 0 0 0 WT Δhly Δhly + WT Δhly Δhly + WT Δhly Δhly + WT Δhly Δhly + pAM401hly pAM401hly pAM401hly pAM401hly
PLC5 Hep3B
Relative Association Relative Association Relative Entry Relative Entry 180 180 160 160 160 ** 140 140 140 140 120 120 * 120 120 100 100 100 100 80 80
% WT % 80 80 % WT %
% WT % 60 60 60 60 WT % 40 40 40 40 20 20 20 20 0 0 0 0 WT Δhly Δhly + WT Δhly Δhly + WT Δhly Δhly + WT Δhly Δhly + pAM401hly pAM401hly pAM401hly pAM401hly
Figure 5.1 LLO is required for efficient entry of L. monocytogenes into hepatocyes. HepG2, Huh7, PLC5, and Hep3B hepatocytes were infected with WT, LLO-deficient (Δhly), or LLO-complemented (Δhly + pAM401hly) bacteria (MOI = 20) for 30 min at 37°C. Cells were washed, fixed, and labeled with fluorescent antibodies and DAPI. Results show the mean ± SEM (n ≥ 3) expressed relative to wt (* P < 0.05; ** P < 0.01).
5.3.2 LLO is not required for L. monocytogenes internalization into BeWo cells
While the liver stage of infection is important for the pathogenesis of L. monocytogenes, listeriosis is characterized by the ability of the bacterium to invade multiple host tissues. Given the severity of fetal infections, it is important to better understand the mechanisms by which L. monocytogenes crosses the placental barrier.
Using BeWo cells, a human epithelial choriocarcinoma cell line, we examined the role
111
HeLa HeLa
Relative Association Relative Entry Relative Association Relative Entry * * 140 140 100 120 100 120 80 100 80 100 80 60 80 60
% WT % 60 60
% WT % WT % % WT % 40 40 40 40 20 20 20 20 0 0 0 0 WT Δhly WT Δhly WT hly::Tn917 WT hly::Tn917 Parental strain: 10403S Parental strain: 10403S Parental strain: LO28 Parental strain LO28
Figure 5.2 LLO is required for efficient entry of L. monocytogenes into HeLa cells. HeLa cells were infected with WT (DP10403S or LO28) or corresponding isogenic LLO-deficient (Δhly or hly::Tn917) bacteria (MOI = 20) for 30 min at 37°C. Cells were washed, fixed, and labeled with fluorescent antibodies and DAPI. Bacterial association and entry are expressed relative to WT strains. Results show mean ± SEM (n≥3) expressed relative to WT (* P < 0.05). of LLO for invasion of placental cells by measuring bacterial association and entry of
WT and LLO-deficient L. monocytogenes. BeWo cells were infected for 30 min at 37°C and association and entry were quantified by immunofluorescence microscopy. In order to better simulate infection of the syncytiotrophoblast, which is the multinucleated epithelium that covers the placental villi, we also infected BeWo cells that were chemically fused to model this specialized epithelium. While BeWo cells were efficiently perforated by LLO (Fig 5.3), the toxin was not required for efficient entry of
L. monocytogenes. In the absence of LLO, there was no decrease in bacterial association or entry into unfused BeWo cells (Fig 5.4). There was an overall decrease in the number of bacteria associated with fused BeWo cells, but again, LLO was not required for efficient bacterial association or entry (Fig 5.4).
112
140 * 0 nM LLO 120 1 nM LLO 100
80
** 60
40 ** * PI (Relative A.F.I.) (Relative PI
20
0 +Ca 2+ -Ca2+ +Ca 2+ -Ca2+
Nonfused Fused
Figure 5.3 BeWo cells are perforated by LLO and undergo Ca2+-dependent membrane resealing. BeWo cells were incubated with 20 µM PI and 0 or 1 nM LLO for 30 min at 37°C. Cells were washed, fixed, and nuclei were labeled with DAPI. Phase contrast and fluorescence images were acquired with a 20X objective. The average fluorescence intensity of PI was automatically measured in the nuclear region by quantitative fluorescence microscopy using Metamorph software. Results are expressed relative to nonfused BeWo cells + Ca2+ without LLO. Results show the mean ± SEM (n≥4). (* P<0.05; ** P< 0.01). Experiments performed by Dr. Eusondia Arnett.
5.3.3 LLO is critical for infection of human placental villous explants
To further determine the contribution of LLO to placental infection, we infected human placental villous tissue explants with WT and LLO-deficient L. monocytogenes.
Villi were dissected from term human placenta, and incubated with 5 x 106 bacteria for 1 h, washed, then incubated with gentamicin to kill extracellular bacteria. At the indicated time points, villous tissue was homogenized and plated on BHI agar to enumerate intracellular CFUs. Villi were efficiently infected by both strains, however, LLO- deficient L. monocytogenes were cleared from the villi, while WT L. monocytogenes
113
BeWo BeWo Relative Association Relative Entry 200 200 * 150 150
100 100
% WT % % WT % 50 50
0 0 WT Δhly WT Δhly WT Δhly WT Δhly
Nonfused Fused Nonfused Fused
Figure 5.4 LLO is not required for efficient entry of L. monocytogenes into placental cells. BeWo cells were infected with WT or LLO-deficient (Δhly) bacteria (1 x 106 bacteria/well) for 30 min at 37°C. Cells were washed, fixed, and labeled with fluorescent antibodies and DAPI. Results show the mean ± SEM (n = 4) expressed relative to wt (* P < 0.05; ** P < 0.01). Assays performed by Dr. Eusondia Arnett
114
was able to replicate intracellularly (Fig. 5.5). Assays with all three infection models
(nonfused BeWo cells, fused BeWo cells, and placental villi) indicate that LLO is not required for efficient internalization of L. monocytogenes into placental cells. However,
LLO is necessary to establish an infection in human placental villi.
108 10403S DP-L2161 107 median 106
105
4 cfu / villi / cfu 10
103
102
10
1 1 3 10 20 1 3 10 20 Hours post infection
Figure 5.5 LLO is critical for infection of human placental villous explants. Term human placental villi were infected with 5 x 106 WT (DP10403S) and LLO-deficient (Δhly; DP-L2161) bacteria for 1 h at 37°C, then washed and treated with 15 µg/ml gentamicin. At 1, 3, 10, and 20 h post-infection, villi were washed, homogenized, and plated on BHI agar to enumerate CFUs. Results show avg CFUs recovered from 3 independent experiments performed in duplicate. Placental villi obtained by Dr. William Ackerman.
115
5.3.4 Roles of InlA and InlB in L. monocytogenes association and entry into human hepatocytes and placental cells
We next assessed the contribution of InlA and InlB to bacterial association and internalization into four hepatocyte cell lines. HepG2, Huh7, PLC5, and Hep3B human hepatocytes were infected with WT, InlA-deficient (ΔinlA), and InlB-deficient (ΔinlB) strains. Bacterial association and entry was then quantified by immunofluorescence.
InlA had the most significant effect on bacterial association to Hep3B cells. In the other three cell lines, average association was decreased, but did not reach statistical significance. InlA could also function as an invasin, as entry of the InlA-deficient strain was significantly reduced in two (PLC5 and HepG2) of the four hepatocyte cell lines.
Intriguingly, InlA was not acting as an adhesin in PLC5 and HepG2 cells.
Unexpectedly, we detected no significant reduction in bacterial association or entry of the InlB-deficient strain in any hepatocyte cell line tested (Fig. 5.6). E-cadherin and c-
Met were detected by western blotting, and by immunofluorescence on the surface of all four hepatocyte cell lines (data not shown), confirming that the lack of a phenotype does not result from the absence of receptors.
116
HepG2 Huh7
Relative Association Relative Entry Relative Association Relative Entry 250 140 160 140 * 200 120 140 120 100 120 100 150 100 80 80 80
100 60 60
% WT %
% WT % % WT %
60 WT % 40 50 40 40 20 20 20 0 0 0 0 WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB
PLC5 Hep3B
Relative Association Relative Entry Relative Association Relative Entry 180 180 160 140 160 160 140 120 140 140 * 120 ** 120 120 100 100 80 100 100 80
80 80 60
% WT %
% WT % % WT % % WT % 60 60 60 40 40 40 40 20 20 20 20 0 0 0 0 WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB
Figure 5.6 Roles of InlA and InlB in L. monocytogenes entry into hepatocytes. HepG2, Hep3B, Huh7, and PLC5 hepatocytes were infected with WT, InlA-deficient (DPL4405), or InlB-deficient (DPL4406) bacteria (MOI = 20) for 30 min at 37°C. Cells were washed, fixed, and labeled with fluorescent antibodies and DAPI. Results show the mean ± SEM (n ≥ 3) expressed relative to WT (* P < 0.05; ** P < 0.01).
We next measured bacterial association and entry of WT, InlA-deficient (ΔinlA), and InlB-deficient (ΔinlB) L. monocytogenes into nonfused and fused BeWo cells. In the absence of InlA, bacterial association to fused and nonfused BeWo cells was significantly decreased, but internalization was not significantly affected. As with invasion of hepatocytes, InlB was not involved in either bacterial association, or internalization into BeWo cells (Fig. 5.7).
117
BeWo BeWo
Relative Association Relative Entry 160 250 140 ** 200 120 100 150 * 80
* 100 % WT % % WT % 60 40 50 20 0 0 WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB WT ∆inlA ∆inlB
Nonfused Fused Nonfused Fused
Figure 5.7 The role of InlA and InlB for entry of L. monocytogenes into placental cells. Cells were infected, washed, fixed, and labeled with fluorescent antibodies and DAPI. BeWo cells were infected with WT or LLO-deficient (Δhly) bacteria (1 x 106 bact/well) for 30 min at 37°C. Results show the mean ± SEM (n = 4) expressed relative to wt (* P < 0.05; ** P < 0.01). Experiments performed by Dr. Eusondia Arnett.
5.3.5 Measuring the contribution of LLO and InlB to intracellular survival in
HepG2 and BeWo cells
InlB is important for colonization of the liver in mice [316], and the placenta in gerbils and humanized E-cadherin knock-in mice [127]. However, in our in vitro assays
InlB is not required for L. monocytogenes entry into human hepatocyte and placental cell lines. We therefore determined if InlB was important for the following stages of infection, such as intracellular survival. Using a gentamicin assay, we measured survival of WT, LLO-deficient, and InlB-deficient strains in HepG2, nonfused BeWo cells, and fused BeWo cells. Significantly fewer LLO-deficient L. monocytogenes were recovered compared to WT in all cell types tested. InlB was also important for bacterial recovery
(Fig 5.8A and B). So while InlB is not required for efficient internalization of L.
118
monocytogenes into HepG2 cells, it is important for the viability of intracellular bacteria. In contrast, InlB had no effect on CFU recovery from BeWo cells after
HepG2 BeWo A Relative CFU/ml B Relative CFU/ml ** ** 120 400 ** 350 100 300 80 250
60 WT % 200 % WT % 150 40 100 20 50 0 0 WT Δhly ∆inlB WT ∆inlB ∆hly WT ∆inlB ∆hly
Nonfused Fused
C BeWo BeWo
1000 Nonfused Fused WT 1000 WT 100 100 ∆inlB ** ∆inlB 10 ** ** ∆hly 10 ∆hly
1 1
Relative Growth Relative Relative Growth Relative 0.1 0.1 0 2 4 6 8 10 12 0 2 4 6 8 10 12 Time (hrs) Time (hrs)
Fig 5.8 The role LLO and InlB for intracellular survival of L. monocytogenes. (A) HepG2 cells were infected for 30 min with WT, InlB-deficient, and LLO-deficient bacteria at MOI 20. After 30 min, cells were washed and then incubated for 60 min with 15 µg/ml gentamicin. Cell lysates were then plated on BHI agar to enumerate cfu (B) Survival assay performed in BeWo cells as in (A), except with 1 x 106 bacteria/well (C) Intracellular survival over time in BeWo. Gentamicin assay performed as in (B), but CFUs were enumerated after 1.5, 5, and 10 h infections. Results show the mean ± SEM (n ≥ 3). (* P < 0.05; ** P < 0.01). Assays with BeWo cells were performed by Dr. Eusondia Arnett.
a 90 min infection. Compared to the WT strains, higher numbers of LLO-deficient bacteria were recovered from BeWo cells. The increased recovery of the Δhly strain likely results from a combination of increases in bacterial assocation, and in the 119
percentage of internalized bacteria seen after infection of BeWo cells with LLO- deficient L. monocytogenes (Fig. 5.4). Indeed, when samples were incubated for 5 and
10 hours after infection, Δhly L. monocytogenes showed only a slight increase in the number of intracellular bacteria (Fig 5.8C).
5.4 Discussion
One active area of research involves determining the range of cells that are invaded through the LLO-mediated internalization pathway. In this study, we demonstrated that LLO is important for mediating internalization of L. monocytogenes into epithelial cells derived from different tissues. LLO was required for efficient internalization into four different hepatocyte cell lines. While always significant, the extent of the decrease in internalization in the absence of LLO varied somewhat between cell lines, indicating that LLO may not be equally important for entry into all cells. We showed that LLO was also critial for efficient entry into HeLa cells, a human cervical cell line that is commonly used as a model cell type for studying L. monocytogenes invasion. This finding confirms that LLO-mediated entry is not limited to hepatocytes.
However, we also determined that LLO is not required for efficient entry into all nonphagocytic cells. In fact, during BeWo infection, L. monocytogenes association and entry was enhanced in the absence of LLO. The mechanism for this is not understood, but the observation that a cholesterol-dependent cytoslyin can inhibit internalization is not unprecedented. Hakansson et al. reported that in the absence of streptolysin O
(SLO), group A Streptococcus (GAS) was internalized in greater numbers by 120
keratinocytes, and that internalization was significantly decreased by the addition of purified SLO [317]. Despite our finding that LLO does not enchance L. monocytogenes entry into BeWo cells, LLO is still critical for some stage of the intracellular lifecycle in placental tissue, as evidenced by the LLO-dependent survival of L. monocytogenes in human placental villi.
Earlier work demonstrated that LLO-mediated internalization occurs through a pore-dependent mechanism. LLO can perforate BeWo cells, showing the formation of a transmembrane pore is required, but not sufficient to induce entry of L. monocytogenes.
The CDCs intermedilysin and vaginolysin, bind to the membrane protein CD59 [94, 96].
These toxins first engage the cell membrane through interactions with their protein receptor, which is important for the initial tethering of the toxin to the target cell, then contact cholesterol, which is required for pore formation. Evidence from our previous studies suggests that LLO does not induce internalization by binding to a host membrane receptor other than cholesterol. First, pore-deficient LLO variants that could bind as monomers or oligomerize into a pre-pore complex on the host membrane could not induce internalization [282]. Second, ion fluxes induced by Ca2+ and K+ ionophores were sufficient to induce the internalization of large cargoes with similar efficiency as
LLO. The observation that LLO does not induce internalization of L. monocytogenes into all nonphagocytic cells is compatible with the hypothesis that membrane perforation by LLO activates an ion flux induced effector molecule that is either not present, or not functional in all nonphagocytic cells.
InlA and InlB were identified as invasins long before the importance of LLO for internalization was appreciated. If and how these three invasins cooperate during L. 121
monocytogenes entry is not understood. With this in mind, we determined the contribution of each factor to bacterial association and entry into multiple cell types. We found that InlA can act as an adhesin and an invasin. The importance of InlA for association and entry varied depending on the cell type infected. In some cases, both
LLO and InlA were necessary for efficient entry, while in other cases, one factor was sufficient. The features dictating which of these virulence factors are important for bacterial association or entry are still unknown. Surprisingly, for InlA and InlB, it was not dependent solely on the presence of their receptors on the cell surface. Both E- cadherin and c-Met were easily detected by immunofluorescence on all of the cell lines tested, yet InlB was not involved in bacterial internalization in hepatocytes, and there was no detectable phenotype in BeWo cells. This raises questions about the function of
InlB during infection. While it is still necessary to determine if BeWo cells express a functional InlB receptor, there is some indication in the literature that InlB may not function as an invasin during pathogenesis. Both in vitro and in vivo studies suggest that in mice, InlB may play its greatest role in intracellular proliferation of bacteria in the liver, not during entry into hepatic cells [318, 319]. Whether a similar mechanism is operating in human nonphagocytic cells is an area for future research.
122
Chapter 6. Synthesis and future goals
In this work, we have identified and begun to characterize a novel bacterial internalization pathway that is mediated by the pore-forming toxin listeriolysin O.
Invasion of nonphagocytic cells classically occurs by one of two previously characterized mechanisms. Bacteria that use the "zipper" mechanism express surface proteins that interact with host cell receptors. Binding of the bacterial ligand to the host receptor elicits signaling cascades that culminate in bacterial internalization. Bacteria that use the "trigger" mechanism possess type III secretion systems, which they employ to inject bacterial effector proteins into target cells. These effector proteins stimulate the host cell cytoskeleton, leading to plasma membrane rearrangements and ultimately, bacterial internalization [223]. LLO stimulates internalization into nonphagocytic cells by a unique mechanism that is pore-dependent. We have made progress in characterizing this internalization pathway but many areas for future research remain, some of which are discussed in this chapter.
6.1 LLO-mediated internalization is mechanistically distinct from plasma membrane resealing
We investigated the molecular mechanisms of LLO-mediated internalization and plasma membrane resealing, and identified differences between the two responses. The laboratory of Dr. Norma Andrews first observed that cells perforated with streptolysin O
123
repaired their plasma membranes through a Ca2+-dependent mechanism involving rapid endocytosis of the perforated membrane [201]. They later demonstrated that this repair response occurs via a pathway that is triggered by lysosome exocytosis, and that culminates in endocytosis of the pore [202]. The parasite Trypanasoma cruzi is thought to induce this repair pathway to stimulate invasion of host cells [270]. We determined that LLO also stimulates a Ca2+-dependent membrane repair response in host cells (Fig.
2.7). However, when identifying and comparing host cell factors required for LLO- mediated invasion of L. monocytogenes, with factors required for membrane resealing, we identified a number of differences. Both responses occured independent of clathrin, but differed in their requirements for F-actin polymerization, dynamin, and K+ efflux.
All of these host cell factors were required for bacterial internalization (Figs. 2.13 and
4.4), but not membrane resealing (Figs. 4.2 and 2.14), demonstrating that the resealing pathway was not sufficient to induce bacterial internalization, as L. monocytogenes internalization could be blocked under conditions where resealing still occurred. LLO- mediated internalization and plasma membrane resealing do however share a requirement for Ca2+ influx (Figs. 4.2 and 4.4). While we know it is not sufficient, future work is needed to determine if L. monocytogenes internalization requires host cell factors in addition to the machinery driving membrane resealing, or if these two events occur along separate pathways.
6.2 Assessing the role of LLO during bacterial internalization in vivo
We have identified LLO as a bacterial invasin, using similar approaches as those used to first identify InlA and InlB as invasins [24, 25]. Like the internalins, expression 124
of LLO was required for efficient bacterial internalization into nonphagocytic cells, and was sufficient to induce the internalization of a noninvasive species of Listeria (Listeria innocua; Fig. 2.4) and of polystyrene beads (Fig. 2.5). But it was not until the development of new animal models that the significance of InlA and InlB in vivo was fully revealed [126, 127]. Another important future aim is to assess the importance of the LLO-mediated internalization pathway in vivo. LLO is required for L. monocytogenes pathogenesis, but it is not known how blocking the activity of extracellular LLO would affect host cell invasion in an animal model. To that end, we generated a LLO toxoid (LLOT), which we used to immunize mice (Fig. 3.1).
Immunization resulted in the generation of LLO-neutralizing antibodies (Fig.3.3), and in a significant reduction of the bacterial load in the livers and spleens of infected mice
(Fig. 3.2). It is possible that the LLO-neutralizing antibodies prevented the extracellular and/or intracellular activities of LLO. More work must be done to determine the mechanism behind this protection, and whether the reduced bacterial load results from a decrease in invasion, or from enhanced clearance by the host.
6.3 LLO-mediated internalization as a paradigm of host cell invasion
Another intriguing result from our studies is the observation that a second cholesterol-dependent cytoslysin, pneumolysin, was sufficient to induce internalization of L. monocytogenes when added exogenously (Fig. 2.11). This demonstrates that the ability to induce internalization is not unique to LLO. Another group has shown that the CDC intermedilysin is critical for invasion of nonphagoctyic cells by Streptococcus intermedius [272]. And the use of pore-forming toxins to induce 125
pathogen internalization is not unique to bacteria. Trypanasoma cruzi perforates host cells and hijacks the membrane repair response to gain entry, while Plasmodium berghei sporozoites express a pore-forming protein with a membrane attack complex/perforin
(MACPF)-related domain that is essential for liver cell invasion [270, 300]. Taken as a whole, these studies reveal that perforation of the plasma membrane by pore-forming toxins may be a widely distributed mechanism utilized by pathogens to gain access to the intracellular niche. Further investigation of the LLO-mediated internalization pathway may therefore yield information that is relevant to internalization pathways utilized by a broad range of pathogens.
6.4 Investigating the host cell range of LLO-mediated internalization
Much of the work characterizing the entry pathway has been performed in the human hepatocyte HepG2 cell line, but we have shown that this entry pathway is not restricted to HepG2 cells. It functions in all of the hepatocyte cell lines we tested, and in
HeLa cells (Fig. 5.1 Fig. 5.2). Given that this internalization pathway is mediated by pore formation, we hypothesized that all cells perforated by LLO would be susceptible to this internalization mechanism. However, we identified cells that are perforated by
LLO, but are not susceptible to this entry pathway. In the human placental BeWo cell line, L. monocytogenes actually entered more efficiently in the absence of LLO (Fig.
5.4). And while other groups have identified pore-forming proteins that can induce endocytosis, as described above, one group has demonstrated that streptolysin O inhibits internalization of group A Streptotoccus [277, 317]. Another important area for future research is to develop a better understanding of the factors that dictate whether or not 126
LLO induces bacterial internalization into a target cell. It is important to further define the host cell machinery that is required for this internalization pathway.
6.5 Variability in the activity of L. monocytogenes invasion factors
It is believed that the importance of the internalins InlA and InlB during host cell invasion is dependent up the expression of their receptors, E-cadherin and c-Met, respectively. Using immunofluorescence microsopy to quantify bacterial association and internalization, we found variability in the roles of the internalins during invasion (Figs.
5.5 and 5.6). InlA can be involved in either bacterial association or internalization. In contrast, we found no role for InlB in entry of any of the cells types tested, and a role for intracellular survival in HepG2 cells, indicating that InlB may function primarily to mediate intracellular surival in some cells (Figs. 5.6 and 5.7). In BeWo cells, we found no role for InlB in entry or intracellular survival (Figs. 5.7 and 5.8). As a whole, our data reveals that the importance of each individual invasion factor is variable depending on the type of cell infected. This is true even if the host cell receptor for that invasion factor is present. Additionally, we show that multiple invasion factors are required for efficient infection, with the influence of different factors predominating at different stages of the intracellular lifecycle (association, internalization, or intracellular survival), or during infection of different cell types.
127
Bibliography
1. Guillet, C., et al., Human listeriosis caused by Listeria ivanovii. Emerg Infect Dis, 2010. 16(1): p. 136-8. 2. Vazquez-Boland, J.A., et al., Listeria pathogenesis and molecular virulence determinants. Clin Microbiol Rev, 2001. 14(3): p. 584-640. 3. Freitag, N.E., G.C. Port, and M.D. Miner, Listeria monocytogenes - from saprophyte to intracellular pathogen. Nat Rev Microbiol, 2009. 7(9): p. 623-8. 4. Schuchat, A., B. Swaminathan, and C.V. Broome, Epidemiology of human listeriosis. Clin Microbiol Rev, 1991. 4(2): p. 169-83. 5. Murray, E.G.D., Webb, R.E., Swann, M.B.R., A disease of rabbits characterized by a large mononuclear leucocytosis, caused by a hitherto undescribed bacillus Bacterium monocytogenes (n. sp.). . J. Pathol. Bacteriol, 1926. 29: p. 407– 439. 6. Pirie, J.H., The Genus Listerella Pirie. Science, 1940. 91(2364): p. 383. 7. Seeliger, H.P., Listeriosis--history and actual developments. Infection, 1988. 16 Suppl 2: p. S80-4. 8. Reiss, H.J., J. Potel, and A. Krebs, [Granulomatosis infantiseptica; a general infection of infants and the newborn characterized by the presence of miliary granuloma]. Z Gesamte Inn Med, 1951. 6(15-16): p. 451-7. 9. Seeliger, H.P., Zur Atiologie der Granulomatosis infantiseptica und pseudotuberkuloser Erkrankungen. Dtsch Med Wochenschr, 1952. 77: p. 587. 10. Schlech, W.F., 3rd, et al., Epidemic listeriosis--evidence for transmission by food. N Engl J Med, 1983. 308(4): p. 203-6. 11. Linnan, M.J., et al., Epidemic listeriosis associated with Mexican-style cheese. N Engl J Med, 1988. 319(13): p. 823-8. 12. Fleming, D.W., et al., Pasteurized milk as a vehicle of infection in an outbreak of listeriosis. N Engl J Med, 1985. 312(7): p. 404-7. 13. Jurado, R.L., et al., Increased risk of meningitis and bacteremia due to Listeria monocytogenes in patients with human immunodeficiency virus infection. Clin Infect Dis, 1993. 17(2): p. 224-7. 14. Louria, D.B., et al., Listeriosis complicating malignant disease. A new association. Ann Intern Med, 1967. 67(2): p. 260-81. 15. Stamm, A.M., et al., Listeriosis in renal transplant recipients: report of an outbreak and review of 102 cases. Rev Infect Dis, 1982. 4(3): p. 665-82. 16. Centers for Disease, C. and Prevention, Multistate outbreak of listeriosis associated with Jensen Farms cantaloupe--United States, August-September 2011. MMWR Morb Mortal Wkly Rep, 2011. 60(39): p. 1357-8.
128
17. Multistate Outbreak of Listeriosis Linked to Imported Frescolina Marte Brand Ricotta Salata Cheese (Final Update). 2012; Available from: http://www.cdc.gov/listeria/outbreaks/cheese-09-12/index.html. 18. Multistate Outbreak of Listeriosis Linked to Crave Brothers Farmstead Cheeses. 2013; Available from: http://www.cdc.gov/listeria/outbreaks/cheese-07-13/index.html. 19. Mead, P.S., et al., Food-related illness and death in the United States. Emerg Infect Dis, 1999. 5(5): p. 607-25. 20. Temple, M.E. and M.C. Nahata, Treatment of listeriosis. Ann Pharmacother, 2000. 34(5): p. 656-61. 21. Swaminathan, B. and P. Gerner-Smidt, The epidemiology of human listeriosis. Microbes Infect, 2007. 9(10): p. 1236-43. 22. Schlech, W.F., 3rd, Foodborne listeriosis. Clin Infect Dis, 2000. 31(3): p. 770-5. 23. Southwick, F.S. and D.L. Purich, Intracellular pathogenesis of listeriosis. N Engl J Med, 1996. 334(12): p. 770-6. 24. Braun, L., H. Ohayon, and P. Cossart, The InIB protein of Listeria monocytogenes is sufficient to promote entry into mammalian cells. Mol Microbiol, 1998. 27(5): p. 1077- 87. 25. Lecuit, M., et al., Internalin of Listeria monocytogenes with an intact leucine-rich repeat region is sufficient to promote internalization. Infect Immun, 1997. 65(12): p. 5309-19. 26. Theriot, J.A., et al., The rate of actin-based motility of intracellular Listeria monocytogenes equals the rate of actin polymerization. Nature, 1992. 357(6375): p. 257-60. 27. Tilney, L.G. and D.A. Portnoy, Actin filaments and the growth, movement, and spread of the intracellular bacterial parasite, Listeria monocytogenes. J Cell Biol, 1989. 109(4 Pt 1): p. 1597-608. 28. Gedde, M.M., et al., Role of listeriolysin O in cell-to-cell spread of Listeria monocytogenes. Infect Immun, 2000. 68(2): p. 999-1003. 29. Doumith, M., et al., New aspects regarding evolution and virulence of Listeria monocytogenes revealed by comparative genomics and DNA arrays. Infect Immun, 2004. 72(2): p. 1072-83. 30. Severino, P., et al., Comparative transcriptome analysis of Listeria monocytogenes strains of the two major lineages reveals differences in virulence, cell wall, and stress response. Appl Environ Microbiol, 2007. 73(19): p. 6078-88. 31. Milohanic, E., et al., Transcriptome analysis of Listeria monocytogenes identifies three groups of genes differently regulated by PrfA. Mol Microbiol, 2003. 47(6): p. 1613-25. 32. Chatterjee, S.S., et al., Intracellular gene expression profile of Listeria monocytogenes. Infect Immun, 2006. 74(2): p. 1323-38. 33. Joseph, B., et al., Identification of Listeria monocytogenes genes contributing to intracellular replication by expression profiling and mutant screening. J Bacteriol, 2006. 188(2): p. 556-68. 34. Glaser, P., et al., Comparative genomics of Listeria species. Science, 2001. 294(5543): p. 849-52. 35. Jaradat, Z.W., J.W. Wampler, and A.W. Bhunia, A Listeria adhesion protein-deficient Listeria monocytogenes strain shows reduced adhesion primarily to intestinal cell lines. Med Microbiol Immunol, 2003. 192(2): p. 85-91.
129
36. Burkholder, K.M. and A.K. Bhunia, Listeria monocytogenes uses Listeria adhesion protein (LAP) to promote bacterial transepithelial translocation and induces expression of LAP receptor Hsp60. Infect Immun, 2010. 78(12): p. 5062-73. 37. Reis, O., et al., LapB, a novel Listeria monocytogenes LPXTG surface adhesin, required for entry into eukaryotic cells and virulence. J Infect Dis, 2010. 202(4): p. 551-62. 38. Linden, S.K., et al., Listeria monocytogenes internalins bind to the human intestinal mucin MUC2. Arch Microbiol, 2008. 190(1): p. 101-4. 39. Sabet, C., et al., The Listeria monocytogenes virulence factor InlJ is specifically expressed in vivo and behaves as an adhesin. Infect Immun, 2008. 76(4): p. 1368-78. 40. Dramsi, S., et al., FbpA, a novel multifunctional Listeria monocytogenes virulence factor. Mol Microbiol, 2004. 53(2): p. 639-49. 41. Alvarez-Dominguez, C., et al., Host cell heparan sulfate proteoglycans mediate attachment and entry of Listeria monocytogenes, and the listerial surface protein ActA is involved in heparan sulfate receptor recognition. Infect Immun, 1997. 65(1): p. 78-88. 42. Pilgrim, S., et al., Deletion of the gene encoding p60 in Listeria monocytogenes leads to abnormal cell division and loss of actin-based motility. Infect Immun, 2003. 71(6): p. 3473-84. 43. Milohanic, E., et al., The autolysin Ami contributes to the adhesion of Listeria monocytogenes to eukaryotic cells via its cell wall anchor. Mol Microbiol, 2001. 39(5): p. 1212-24. 44. Milohanic, E., et al., Sequence and binding activity of the autolysin-adhesin Ami from epidemic Listeria monocytogenes 4b. Infect Immun, 2004. 72(8): p. 4401-9. 45. Abachin, E., et al., Formation of D-alanyl-lipoteichoic acid is required for adhesion and virulence of Listeria monocytogenes. Mol Microbiol, 2002. 43(1): p. 1-14. 46. Gaillard, J.L., et al., Entry of L. monocytogenes into cells is mediated by internalin, a repeat protein reminiscent of surface antigens from gram-positive cocci. Cell, 1991. 65(7): p. 1127-41. 47. Dramsi, S., et al., Entry of Listeria monocytogenes into hepatocytes requires expression of inIB, a surface protein of the internalin multigene family. Mol Microbiol, 1995. 16(2): p. 251-61. 48. Kobe, B. and A.V. Kajava, The leucine-rich repeat as a protein recognition motif. Curr Opin Struct Biol, 2001. 11(6): p. 725-32. 49. Bierne, H., et al., Internalins: a complex family of leucine-rich repeat-containing proteins in Listeria monocytogenes. Microbes Infect, 2007. 9(10): p. 1156-66. 50. Seveau, S., et al., Role of lipid rafts in E-cadherin-- and HGF-R/Met--mediated entry of Listeria monocytogenes into host cells. J Cell Biol, 2004. 166(5): p. 743-53. 51. Fujita, Y., et al., Hakai, a c-Cbl-like protein, ubiquitinates and induces endocytosis of the E-cadherin complex. Nat Cell Biol, 2002. 4(3): p. 222-31. 52. Bonazzi, M., et al., Successive post-translational modifications of E-cadherin are required for InlA-mediated internalization of Listeria monocytogenes. Cell Microbiol, 2008. 10(11): p. 2208-22. 53. Lecuit, M., et al., A role for alpha-and beta-catenins in bacterial uptake. Proc Natl Acad Sci U S A, 2000. 97(18): p. 10008-13. 54. Sousa, S., et al., Unconventional myosin VIIa and vezatin, two proteins crucial for Listeria entry into epithelial cells. J Cell Sci, 2004. 117(Pt 10): p. 2121-30.
130
55. Sousa, S., et al., Src, cortactin and Arp2/3 complex are required for E-cadherin-mediated internalization of Listeria into cells. Cell Microbiol, 2007. 9(11): p. 2629-43. 56. Veiga, E., et al., Invasive and adherent bacterial pathogens co-Opt host clathrin for infection. Cell Host Microbe, 2007. 2(5): p. 340-51. 57. Braun, L., B. Ghebrehiwet, and P. Cossart, gC1q-R/p32, a C1q-binding protein, is a receptor for the InlB invasion protein of Listeria monocytogenes. EMBO J, 2000. 19(7): p. 1458-66. 58. Shen, Y., et al., InIB-dependent internalization of Listeria is mediated by the Met receptor tyrosine kinase. Cell, 2000. 103(3): p. 501-10. 59. Jonquieres, R., J. Pizarro-Cerda, and P. Cossart, Synergy between the N- and C-terminal domains of InlB for efficient invasion of non-phagocytic cells by Listeria monocytogenes. Mol Microbiol, 2001. 42(4): p. 955-65. 60. Ireton, K., B. Payrastre, and P. Cossart, The Listeria monocytogenes protein InlB is an agonist of mammalian phosphoinositide 3-kinase. J Biol Chem, 1999. 274(24): p. 17025- 32. 61. Li, N., et al., The Listeria protein internalin B mimics hepatocyte growth factor-induced receptor trafficking. Traffic, 2005. 6(6): p. 459-73. 62. Jonquieres, R., et al., Interaction between the protein InlB of Listeria monocytogenes and lipoteichoic acid: a novel mechanism of protein association at the surface of gram- positive bacteria. Mol Microbiol, 1999. 34(5): p. 902-14. 63. Braun, L., et al., InlB: an invasion protein of Listeria monocytogenes with a novel type of surface association. Mol Microbiol, 1997. 25(2): p. 285-94. 64. Marino, M., et al., GW domains of the Listeria monocytogenes invasion protein InlB are SH3-like and mediate binding to host ligands. EMBO J, 2002. 21(21): p. 5623-34. 65. Banerjee, M., et al., GW domains of the Listeria monocytogenes invasion protein InlB are required for potentiation of Met activation. Mol Microbiol, 2004. 52(1): p. 257-71. 66. Ferraris, D.M., et al., Ligand-mediated dimerization of the Met receptor tyrosine kinase by the bacterial invasion protein InlB. J Mol Biol, 2010. 395(3): p. 522-32. 67. Ireton, K., et al., A role for phosphoinositide 3-kinase in bacterial invasion. Science, 1996. 274(5288): p. 780-2. 68. Sun, H., et al., Host adaptor proteins Gab1 and CrkII promote InlB-dependent entry of Listeria monocytogenes. Cell Microbiol, 2005. 7(3): p. 443-57. 69. Pizarro-Cerda, J., et al., Type II phosphatidylinositol 4-kinases promote Listeria monocytogenes entry into target cells. Cell Microbiol, 2007. 9(10): p. 2381-90. 70. Bierne, H., et al., A role for cofilin and LIM kinase in Listeria-induced phagocytosis. J Cell Biol, 2001. 155(1): p. 101-12. 71. Bierne, H., et al., WASP-related proteins, Abi1 and Ena/VASP are required for Listeria invasion induced by the Met receptor. J Cell Sci, 2005. 118(Pt 7): p. 1537-47. 72. Cabanes, D., et al., Auto, a surface associated autolysin of Listeria monocytogenes required for entry into eukaryotic cells and virulence. Mol Microbiol, 2004. 51(6): p. 1601-14. 73. Suarez, M., et al., A role for ActA in epithelial cell invasion by Listeria monocytogenes. Cell Microbiol, 2001. 3(12): p. 853-64. 74. Glomski, I.J., et al., The Listeria monocytogenes hemolysin has an acidic pH optimum to compartmentalize activity and prevent damage to infected host cells. J Cell Biol, 2002. 156(6): p. 1029-38.
131
75. Goetz, M., et al., Microinjection and growth of bacteria in the cytosol of mammalian host cells. Proc Natl Acad Sci U S A, 2001. 98(21): p. 12221-6. 76. Chico-Calero, I., et al., Hpt, a bacterial homolog of the microsomal glucose- 6-phosphate translocase, mediates rapid intracellular proliferation in Listeria. Proc Natl Acad Sci U S A, 2002. 99(1): p. 431-6. 77. Joseph, B. and W. Goebel, Life of Listeria monocytogenes in the host cells' cytosol. Microbes Infect, 2007. 9(10): p. 1188-95. 78. Engelbrecht, F., et al., A new PrfA-regulated gene of Listeria monocytogenes encoding a small, secreted protein which belongs to the family of internalins. Mol Microbiol, 1996. 21(4): p. 823-37. 79. Rajabian, T., et al., The bacterial virulence factor InlC perturbs apical cell junctions and promotes cell-to-cell spread of Listeria. Nat Cell Biol, 2009. 11(10): p. 1212-8. 80. Yoshikawa, Y., et al., Listeria monocytogenes ActA-mediated escape from autophagic recognition. Nat Cell Biol, 2009. 11(10): p. 1233-40. 81. Rich, K.A., C. Burkett, and P. Webster, Cytoplasmic bacteria can be targets for autophagy. Cell Microbiol, 2003. 5(7): p. 455-68. 82. Dortet, L., et al., Recruitment of the major vault protein by InlK: a Listeria monocytogenes strategy to avoid autophagy. PLoS Pathog, 2011. 7(8): p. e1002168. 83. Camejo, A., et al., In vivo transcriptional profiling of Listeria monocytogenes and mutagenesis identify new virulence factors involved in infection. PLoS Pathog, 2009. 5(5): p. e1000449. 84. Tweten, R.K., Cholesterol-dependent cytolysins, a family of versatile pore-forming toxins. Infect Immun, 2005. 73(10): p. 6199-209. 85. Hotze, E.M., et al., Identification and characterization of the first cholesterol-dependent cytolysins from Gram-negative bacteria. Infect Immun, 2013. 81(1): p. 216-25. 86. Rossjohn, J., et al., Structure of a cholesterol-binding, thiol-activated cytolysin and a model of its membrane form. Cell, 1997. 89(5): p. 685-92. 87. Polekhina, G., et al., Crystallization and preliminary X-ray analysis of the human-specific toxin intermedilysin. Acta Crystallogr D Biol Crystallogr, 2004. 60(Pt 2): p. 347-9. 88. Rossjohn, J., et al., Structures of perfringolysin O suggest a pathway for activation of cholesterol-dependent cytolysins. J Mol Biol, 2007. 367(5): p. 1227-36. 89. Bourdeau, R.W., et al., Cellular functions and X-ray structure of anthrolysin O, a cholesterol-dependent cytolysin secreted by Bacillus anthracis. J Biol Chem, 2009. 284(21): p. 14645-56. 90. Xu, L., et al., Crystal structure of cytotoxin protein suilysin from Streptococcus suis. Protein Cell, 2010. 1(1): p. 96-105. 91. Feil, S.C., et al., Structure of the lectin regulatory domain of the cholesterol-dependent cytolysin lectinolysin reveals the basis for its lewis antigen specificity. Structure, 2012. 20(2): p. 248-58. 92. Hotze, E.M. and R.K. Tweten, Membrane assembly of the cholesterol-dependent cytolysin pore complex. Biochim Biophys Acta, 2012. 1818(4): p. 1028-38. 93. Farrand, A.J., et al., Only two amino acids are essential for cytolytic toxin recognition of cholesterol at the membrane surface. Proc Natl Acad Sci U S A, 2010. 107(9): p. 4341-6. 94. Giddings, K.S., et al., Human CD59 is a receptor for the cholesterol-dependent cytolysin intermedilysin. Nat Struct Mol Biol, 2004. 11(12): p. 1173-8.
132
95. Giddings, K.S., A.E. Johnson, and R.K. Tweten, Redefining cholesterol's role in the mechanism of the cholesterol-dependent cytolysins. Proc Natl Acad Sci U S A, 2003. 100(20): p. 11315-20. 96. Gelber, S.E., et al., Functional and phylogenetic characterization of Vaginolysin, the human-specific cytolysin from Gardnerella vaginalis. J Bacteriol, 2008. 190(11): p. 3896- 903. 97. Ramachandran, R., R.K. Tweten, and A.E. Johnson, Membrane-dependent conformational changes initiate cholesterol-dependent cytolysin oligomerization and intersubunit beta-strand alignment. Nat Struct Mol Biol, 2004. 11(8): p. 697-705. 98. Shepard, L.A., et al., The mechanism of pore assembly for a cholesterol-dependent cytolysin: formation of a large prepore complex precedes the insertion of the transmembrane beta-hairpins. Biochemistry, 2000. 39(33): p. 10284-93. 99. Tilley, S.J., et al., Structural basis of pore formation by the bacterial toxin pneumolysin. Cell, 2005. 121(2): p. 247-56. 100. Olofsson, A., H. Hebert, and M. Thelestam, The projection structure of perfringolysin O (Clostridium perfringens theta-toxin). FEBS Lett, 1993. 319(1-2): p. 125-7. 101. Czajkowsky, D.M., et al., Vertical collapse of a cytolysin prepore moves its transmembrane beta-hairpins to the membrane. EMBO J, 2004. 23(16): p. 3206-15. 102. Leimeister-Wachter, M., et al., Identification of a gene that positively regulates expression of listeriolysin, the major virulence factor of listeria monocytogenes. Proc Natl Acad Sci U S A, 1990. 87(21): p. 8336-40. 103. Mengaud, J., et al., Pleiotropic control of Listeria monocytogenes virulence factors by a gene that is autoregulated. Mol Microbiol, 1991. 5(9): p. 2273-83. 104. Chakraborty, T., et al., Coordinate regulation of virulence genes in Listeria monocytogenes requires the product of the prfA gene. J Bacteriol, 1992. 174(2): p. 568- 74. 105. Freitag, N.E., P. Youngman, and D.A. Portnoy, Transcriptional activation of the Listeria monocytogenes hemolysin gene in Bacillus subtilis. J Bacteriol, 1992. 174(4): p. 1293-8. 106. Johansson, J., et al., An RNA thermosensor controls expression of virulence genes in Listeria monocytogenes. Cell, 2002. 110(5): p. 551-61. 107. Schnupf, P., et al., Regulated translation of listeriolysin O controls virulence of Listeria monocytogenes. Mol Microbiol, 2006. 61(4): p. 999-1012. 108. Jones, S. and D.A. Portnoy, Characterization of Listeria monocytogenes pathogenesis in a strain expressing perfringolysin O in place of listeriolysin O. Infect Immun, 1994. 62(12): p. 5608-13. 109. Wei, Z., et al., Characterization of Listeria monocytogenes expressing anthrolysin O and phosphatidylinositol-specific phospholipase C from Bacillus anthracis. Infect Immun, 2005. 73(10): p. 6639-46. 110. Frehel, C., et al., Capacity of ivanolysin O to replace listeriolysin O in phagosomal escape and in vivo survival of Listeria monocytogenes. Microbiology, 2003. 149(Pt 3): p. 611-20. 111. Geoffroy, C., et al., Purification, characterization, and toxicity of the sulfhydryl-activated hemolysin listeriolysin O from Listeria monocytogenes. Infect Immun, 1987. 55(7): p. 1641-6. 112. Schuerch, D.W., E.M. Wilson-Kubalek, and R.K. Tweten, Molecular basis of listeriolysin O pH dependence. Proc Natl Acad Sci U S A, 2005. 102(35): p. 12537-42.
133
113. Schnupf, P., et al., Listeriolysin O secreted by Listeria monocytogenes into the host cell cytosol is degraded by the N-end rule pathway. Infect Immun, 2007. 75(11): p. 5135-47. 114. Schnupf, P., D.A. Portnoy, and A.L. Decatur, Phosphorylation, ubiquitination and degradation of listeriolysin O in mammalian cells: role of the PEST-like sequence. Cell Microbiol, 2006. 8(2): p. 353-64. 115. Gaillard, J.L., et al., In vitro model of penetration and intracellular growth of Listeria monocytogenes in the human enterocyte-like cell line Caco-2. Infect Immun, 1987. 55(11): p. 2822-9. 116. Bielecki, J., et al., Bacillus subtilis expressing a haemolysin gene from Listeria monocytogenes can grow in mammalian cells. Nature, 1990. 345(6271): p. 175-6. 117. Dancz, C.E., et al., Inducible control of virulence gene expression in Listeria monocytogenes: temporal requirement of listeriolysin O during intracellular infection. J Bacteriol, 2002. 184(21): p. 5935-45. 118. Marquis, H., V. Doshi, and D.A. Portnoy, The broad-range phospholipase C and a metalloprotease mediate listeriolysin O-independent escape of Listeria monocytogenes from a primary vacuole in human epithelial cells. Infect Immun, 1995. 63(11): p. 4531-4. 119. Grundling, A., M.D. Gonzalez, and D.E. Higgins, Requirement of the Listeria monocytogenes broad-range phospholipase PC-PLC during infection of human epithelial cells. J Bacteriol, 2003. 185(21): p. 6295-307. 120. Beauregard, K.E., et al., pH-dependent perforation of macrophage phagosomes by listeriolysin O from Listeria monocytogenes. J Exp Med, 1997. 186(7): p. 1159-63. 121. Shaughnessy, L.M., et al., Membrane perforations inhibit lysosome fusion by altering pH and calcium in Listeria monocytogenes vacuoles. Cell Microbiol, 2006. 8(5): p. 781-92. 122. Poussin, M.A. and H. Goldfine, Involvement of Listeria monocytogenes phosphatidylinositol-specific phospholipase C and host protein kinase C in permeabilization of the macrophage phagosome. Infect Immun, 2005. 73(7): p. 4410-3. 123. Goldfine, H., S.J. Wadsworth, and N.C. Johnston, Activation of host phospholipases C and D in macrophages after infection with Listeria monocytogenes. Infect Immun, 2000. 68(10): p. 5735-41. 124. Singh, R., A. Jamieson, and P. Cresswell, GILT is a critical host factor for Listeria monocytogenes infection. Nature, 2008. 455(7217): p. 1244-7. 125. Radtke, A.L., et al., Listeria monocytogenes exploits cystic fibrosis transmembrane conductance regulator (CFTR) to escape the phagosome. Proc Natl Acad Sci U S A, 2011. 108(4): p. 1633-8. 126. Lecuit, M., et al., A transgenic model for listeriosis: role of internalin in crossing the intestinal barrier. Science, 2001. 292(5522): p. 1722-5. 127. Disson, O., et al., Conjugated action of two species-specific invasion proteins for fetoplacental listeriosis. Nature, 2008. 455(7216): p. 1114-8. 128. Wollert, T., et al., Extending the host range of Listeria monocytogenes by rational protein design. Cell, 2007. 129(5): p. 891-902. 129. Jacquet, C., et al., A molecular marker for evaluating the pathogenic potential of foodborne Listeria monocytogenes. J Infect Dis, 2004. 189(11): p. 2094-100. 130. Olier, M., et al., Expression of truncated Internalin A is involved in impaired internalization of some Listeria monocytogenes isolates carried asymptomatically by humans. Infect Immun, 2003. 71(3): p. 1217-24.
134
131. Pentecost, M., et al., Listeria monocytogenes invades the epithelial junctions at sites of cell extrusion. PLoS Pathog, 2006. 2(1): p. e3. 132. Nikitas, G., et al., Transcytosis of Listeria monocytogenes across the intestinal barrier upon specific targeting of goblet cell accessible E-cadherin. J Exp Med, 2011. 208(11): p. 2263-77. 133. Pentecost, M., et al., Listeria monocytogenes internalin B activates junctional endocytosis to accelerate intestinal invasion. PLoS Pathog, 2010. 6(5): p. e1000900. 134. Tsai, Y.H., et al., Murinization of internalin extends its receptor repertoire, altering Listeria monocytogenes cell tropism and host responses. PLoS Pathog, 2013. 9(5): p. e1003381. 135. Issekutz TB, E.J., Bortolussi R., The immune response to human neonates to Listeria monocytogenes infection. Clin Invest Med, 1984. 7: p. 281-6. 136. Janakiraman, V., Listeriosis in pregnancy: diagnosis, treatment, and prevention. Rev Obstet Gynecol, 2008. 1(4): p. 179-85. 137. Lecuit, M., et al., Targeting and crossing of the human maternofetal barrier by Listeria monocytogenes: role of internalin interaction with trophoblast E-cadherin. Proc Natl Acad Sci U S A, 2004. 101(16): p. 6152-7. 138. Robbins, J.R., et al., Placental syncytiotrophoblast constitutes a major barrier to vertical transmission of Listeria monocytogenes. PLoS Pathog, 2010. 6(1): p. e1000732. 139. Zeldovich, V.B., et al., Invasive extravillous trophoblasts restrict intracellular growth and spread of Listeria monocytogenes. PLoS Pathog, 2011. 7(3): p. e1002005. 140. Bakardjiev, A.I., J.A. Theriot, and D.A. Portnoy, Listeria monocytogenes traffics from maternal organs to the placenta and back. PLoS Pathog, 2006. 2(6): p. e66. 141. Le Monnier, A., et al., ActA is required for crossing of the fetoplacental barrier by Listeria monocytogenes. Infect Immun, 2007. 75(2): p. 950-7. 142. Drevets, D.A. and P.J. Leenen, Leukocyte-facilitated entry of intracellular pathogens into the central nervous system. Microbes Infect, 2000. 2(13): p. 1609-18. 143. Kim, K.S., Mechanisms of microbial traversal of the blood-brain barrier. Nat Rev Microbiol, 2008. 6(8): p. 625-34. 144. Cabanes, D., et al., Gp96 is a receptor for a novel Listeria monocytogenes virulence factor, Vip, a surface protein. EMBO J, 2005. 24(15): p. 2827-38. 145. Disson, O. and M. Lecuit, Targeting of the central nervous system by Listeria monocytogenes. Virulence, 2012. 3(2): p. 213-21. 146. Kirk, J., Diagnostic ultrastructure of Listeria monocytogenes in human central nervous tissue. Ultrastruct Pathol, 1993. 17(6): p. 583-92. 147. Drevets, D.A., et al., Listeria monocytogenes infects human endothelial cells by two distinct mechanisms. Infect Immun, 1995. 63(11): p. 4268-76. 148. Parida, S.K., et al., Internalin B is essential for adhesion and mediates the invasion of Listeria monocytogenes into human endothelial cells. Mol Microbiol, 1998. 28(1): p. 81- 93. 149. Greiffenberg, L., et al., Interaction of Listeria monocytogenes with human brain microvascular endothelial cells: InlB-dependent invasion, long-term intracellular growth, and spread from macrophages to endothelial cells. Infect Immun, 1998. 66(11): p. 5260- 7.
135
150. Pron, B., et al., Dendritic cells are early cellular targets of Listeria monocytogenes after intestinal delivery and are involved in bacterial spread in the host. Cell Microbiol, 2001. 3(5): p. 331-40. 151. Drevets, D.A., et al., The Ly-6Chigh monocyte subpopulation transports Listeria monocytogenes into the brain during systemic infection of mice. J Immunol, 2004. 172(7): p. 4418-24. 152. Drevets, D.A., et al., Innate responses to systemic infection by intracellular bacteria trigger recruitment of Ly-6Chigh monocytes to the brain. J Immunol, 2008. 181(1): p. 529-36. 153. Drevets, D.A., et al., IFN-gamma triggers CCR2-independent monocyte entry into the brain during systemic infection by virulent Listeria monocytogenes. Brain Behav Immun, 2010. 24(6): p. 919-29. 154. Drevets, D.A. and M.S. Bronze, Listeria monocytogenes: epidemiology, human disease, and mechanisms of brain invasion. FEMS Immunol Med Microbiol, 2008. 53(2): p. 151- 65. 155. Join-Lambert, O.F., et al., Listeria monocytogenes-infected bone marrow myeloid cells promote bacterial invasion of the central nervous system. Cell Microbiol, 2005. 7(2): p. 167-80. 156. Drevets, D.A., P.J. Leenen, and R.A. Greenfield, Invasion of the central nervous system by intracellular bacteria. Clin Microbiol Rev, 2004. 17(2): p. 323-47. 157. Drevets, D.A., T.A. Jelinek, and N.E. Freitag, Listeria monocytogenes-infected phagocytes can initiate central nervous system infection in mice. Infect Immun, 2001. 69(3): p. 1344- 50. 158. Drevets, D.A., Dissemination of Listeria monocytogenes by infected phagocytes. Infect Immun, 1999. 67(7): p. 3512-7. 159. Kayal, S., et al., Listeriolysin O-dependent activation of endothelial cells during infection with Listeria monocytogenes: activation of NF-kappa B and upregulation of adhesion molecules and chemokines. Mol Microbiol, 1999. 31(6): p. 1709-22. 160. Dramsi, S., et al., Entry of Listeria monocytogenes into neurons occurs by cell-to-cell spread: an in vitro study. Infect Immun, 1998. 66(9): p. 4461-8. 161. Greiffenberg, L., et al., Interaction of Listeria monocytogenes with human brain microvascular endothelial cells: an electron microscopic study. Infect Immun, 2000. 68(6): p. 3275-9. 162. Doran, K.S., et al., Concepts and mechanisms: crossing host barriers. Cold Spring Harb Perspect Med, 2013. 3(7). 163. Johnson, G.L. and R. Lapadat, Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science, 2002. 298(5600): p. 1911-2. 164. Gonzalez, M.R., et al., Pore-forming toxins induce multiple cellular responses promoting survival. Cell Microbiol, 2011. 13(7): p. 1026-43. 165. Husmann, M., et al., Differential role of p38 mitogen activated protein kinase for cellular recovery from attack by pore-forming S. aureus alpha-toxin or streptolysin O. Biochem Biophys Res Commun, 2006. 344(4): p. 1128-34. 166. Huffman, D.L., et al., Mitogen-activated protein kinase pathways defend against bacterial pore-forming toxins. Proc Natl Acad Sci U S A, 2004. 101(30): p. 10995-1000. 167. Porta, H., et al., Role of MAPK p38 in the cellular responses to pore-forming toxins. Peptides, 2011. 32(3): p. 601-6.
136
168. Ratner, A.J., et al., Epithelial cells are sensitive detectors of bacterial pore-forming toxins. J Biol Chem, 2006. 281(18): p. 12994-8. 169. Stringaris, A.K., et al., Neurotoxicity of pneumolysin, a major pneumococcal virulence factor, involves calcium influx and depends on activation of p38 mitogen-activated protein kinase. Neurobiol Dis, 2002. 11(3): p. 355-68. 170. Dogan, S., et al., Pneumolysin-induced CXCL8 production by nasopharyngeal epithelial cells is dependent on calcium flux and MAPK activation via Toll-like receptor 4. Microbes Infect, 2011. 13(1): p. 65-75. 171. Hackett, S.P. and D.L. Stevens, Streptococcal toxic shock syndrome: synthesis of tumor necrosis factor and interleukin-1 by monocytes stimulated with pyrogenic exotoxin A and streptolysin O. J Infect Dis, 1992. 165(5): p. 879-85. 172. Houldsworth, S., P.W. Andrew, and T.J. Mitchell, Pneumolysin stimulates production of tumor necrosis factor alpha and interleukin-1 beta by human mononuclear phagocytes. Infect Immun, 1994. 62(4): p. 1501-3. 173. Bebien, M., et al., The pore-forming toxin beta hemolysin/cytolysin triggers p38 MAPK- dependent IL-10 production in macrophages and inhibits innate immunity. PLoS Pathog, 2012. 8(7): p. e1002812. 174. Gurcel, L., et al., Caspase-1 activation of lipid metabolic pathways in response to bacterial pore-forming toxins promotes cell survival. Cell, 2006. 126(6): p. 1135-45. 175. McCoy, A.J., et al., Differential regulation of caspase-1 activation via NLRP3/NLRC4 inflammasomes mediated by aerolysin and type III secretion system during Aeromonas veronii infection. J Immunol, 2010. 185(11): p. 7077-84. 176. Dragneva, Y., et al., Subcytocidal attack by staphylococcal alpha-toxin activates NF- kappaB and induces interleukin-8 production. Infect Immun, 2001. 69(4): p. 2630-5. 177. Walev, I., et al., Resealing of large transmembrane pores produced by streptolysin O in nucleated cells is accompanied by NF-kappaB activation and downstream events. FASEB J, 2002. 16(2): p. 237-9. 178. Kayal, S., et al., Listeriolysin O secreted by Listeria monocytogenes induces NF-kappaB signalling by activating the IkappaB kinase complex. Mol Microbiol, 2002. 44(5): p. 1407- 19. 179. Bischof, L.J., et al., Activation of the unfolded protein response is required for defenses against bacterial pore-forming toxin in vivo. PLoS Pathog, 2008. 4(10): p. e1000176. 180. Bhakdi, S., et al., A guide to the use of pore-forming toxins for controlled permeabilization of cell membranes. Med Microbiol Immunol, 1993. 182(4): p. 167-75. 181. Walev, I., et al., Delivery of proteins into living cells by reversible membrane permeabilization with streptolysin-O. Proc Natl Acad Sci U S A, 2001. 98(6): p. 3185-90. 182. Madden, J.C., N. Ruiz, and M. Caparon, Cytolysin-mediated translocation (CMT): a functional equivalent of type III secretion in gram-positive bacteria. Cell, 2001. 104(1): p. 143-52. 183. Ghosh, J. and M.G. Caparon, Specificity of Streptococcus pyogenes NAD(+) glycohydrolase in cytolysin-mediated translocation. Mol Microbiol, 2006. 62(4): p. 1203- 14. 184. Magassa, N., S. Chandrasekaran, and M.G. Caparon, Streptococcus pyogenes cytolysin- mediated translocation does not require pore formation by streptolysin O. EMBO Rep, 2010. 11(5): p. 400-5.
137
185. McNeil, P.L., Cellular and molecular adaptations to injurious mechanical stress. Trends Cell Biol, 1993. 3(9): p. 302-7. 186. McNeil, P.L. and R. Khakee, Disruptions of muscle fiber plasma membranes. Role in exercise-induced damage. Am J Pathol, 1992. 140(5): p. 1097-109. 187. McNeil, P.L. and R.A. Steinhardt, Plasma membrane disruption: repair, prevention, adaptation. Annu Rev Cell Dev Biol, 2003. 19: p. 697-731. 188. Bi, G.Q., J.M. Alderton, and R.A. Steinhardt, Calcium-regulated exocytosis is required for cell membrane resealing. J Cell Biol, 1995. 131(6 Pt 2): p. 1747-58. 189. Coorssen, J.R., H. Schmitt, and W. Almers, Ca2+ triggers massive exocytosis in Chinese hamster ovary cells. EMBO J, 1996. 15(15): p. 3787-91. 190. Babiychuk, E.B., et al., Blebbing confers resistance against cell lysis. Cell Death Differ, 2011. 18(1): p. 80-9. 191. Keyel, P.A., et al., Streptolysin O clearance through sequestration into blebs that bud passively from the plasma membrane. J Cell Sci, 2011. 124(Pt 14): p. 2414-23. 192. Monastyrskaya, K., et al., Down-regulation of annexin A1 in the urothelium decreases cell survival after bacterial toxin exposure. J Urol, 2013. 190(1): p. 325-33. 193. Potez, S., et al., Tailored protection against plasmalemmal injury by annexins with different Ca2+ sensitivities. J Biol Chem, 2011. 286(20): p. 17982-91. 194. Draeger, A., K. Monastyrskaya, and E.B. Babiychuk, Plasma membrane repair and cellular damage control: the annexin survival kit. Biochem Pharmacol, 2011. 81(6): p. 703-12. 195. Reddy, A., E.V. Caler, and N.W. Andrews, Plasma membrane repair is mediated by Ca(2+)-regulated exocytosis of lysosomes. Cell, 2001. 106(2): p. 157-69. 196. Rodriguez, A., et al., Lysosomes behave as Ca2+-regulated exocytic vesicles in fibroblasts and epithelial cells. J Cell Biol, 1997. 137(1): p. 93-104. 197. Jaiswal, J.K., N.W. Andrews, and S.M. Simon, Membrane proximal lysosomes are the major vesicles responsible for calcium-dependent exocytosis in nonsecretory cells. J Cell Biol, 2002. 159(4): p. 625-35. 198. Rao, S.K., et al., Identification of SNAREs involved in synaptotagmin VII-regulated lysosomal exocytosis. J Biol Chem, 2004. 279(19): p. 20471-9. 199. Martinez, I., et al., Synaptotagmin VII regulates Ca(2+)-dependent exocytosis of lysosomes in fibroblasts. J Cell Biol, 2000. 148(6): p. 1141-49. 200. Togo, T., T.B. Krasieva, and R.A. Steinhardt, A decrease in membrane tension precedes successful cell-membrane repair. Mol Biol Cell, 2000. 11(12): p. 4339-46. 201. Idone, V., et al., Repair of injured plasma membrane by rapid Ca2+-dependent endocytosis. J Cell Biol, 2008. 180(5): p. 905-14. 202. Tam, C., et al., Exocytosis of acid sphingomyelinase by wounded cells promotes endocytosis and plasma membrane repair. J Cell Biol, 2010. 189(6): p. 1027-38. 203. Corrotte, M., et al., Toxin pores endocytosed during plasma membrane repair traffic into the lumen of MVBs for degradation. Traffic, 2012. 13(3): p. 483-94. 204. Los, F.C., et al., RAB-5- and RAB-11-dependent vesicle-trafficking pathways are required for plasma membrane repair after attack by bacterial pore-forming toxin. Cell Host Microbe, 2011. 9(2): p. 147-57. 205. Tang, P., I. Rosenshine, and B.B. Finlay, Listeria monocytogenes, an invasive bacterium, stimulates MAP kinase upon attachment to epithelial cells. Mol Biol Cell, 1994. 5(4): p. 455-64.
138
206. Tang, P., et al., Listeriolysin O activates mitogen-activated protein kinase in eucaryotic cells. Infect Immun, 1996. 64(6): p. 2359-61. 207. Tang, P., et al., Listeria monocytogenes invasion of epithelial cells requires the MEK- 1/ERK-2 mitogen-activated protein kinase pathway. Infect Immun, 1998. 66(3): p. 1106- 12. 208. Bavdek, A., et al., Sterol and pH interdependence in the binding, oligomerization, and pore formation of Listeriolysin O. Biochemistry, 2007. 46(14): p. 4425-37. 209. Hamon, M.A. and P. Cossart, K+ efflux is required for histone H3 dephosphorylation by Listeria monocytogenes listeriolysin O and other pore-forming toxins. Infect Immun, 2011. 79(7): p. 2839-46. 210. Gekara, N.O., et al., The multiple mechanisms of Ca2+ signalling by listeriolysin O, the cholesterol-dependent cytolysin of Listeria monocytogenes. Cell Microbiol, 2007. 9(8): p. 2008-21. 211. Repp, H., et al., Listeriolysin of Listeria monocytogenes forms Ca2+-permeable pores leading to intracellular Ca2+ oscillations. Cell Microbiol, 2002. 4(8): p. 483-91. 212. Dramsi, S. and P. Cossart, Listeriolysin O-mediated calcium influx potentiates entry of Listeria monocytogenes into the human Hep-2 epithelial cell line. Infect Immun, 2003. 71(6): p. 3614-8. 213. Stavru, F. and P. Cossart, Listeria infection modulates mitochondrial dynamics. Commun Integr Biol, 2011. 4(3): p. 364-6. 214. Stavru, F., et al., Listeria monocytogenes transiently alters mitochondrial dynamics during infection. Proc Natl Acad Sci U S A, 2011. 108(9): p. 3612-7. 215. Stavru, F., et al., Atypical mitochondrial fission upon bacterial infection. Proc Natl Acad Sci U S A, 2013. 110(40): p. 16003-16008. 216. Pillich, H., et al., Activation of the unfolded protein response by Listeria monocytogenes. Cell Microbiol, 2012. 14(6): p. 949-64. 217. Hamon, M.A., et al., Histone modifications induced by a family of bacterial toxins. Proc Natl Acad Sci U S A, 2007. 104(33): p. 13467-72. 218. Choy, A. and C.R. Roy, Autophagy and bacterial infection: an evolving arms race. Trends Microbiol, 2013. 21(9): p. 451-6. 219. Meyer-Morse, N., et al., Listeriolysin O is necessary and sufficient to induce autophagy during Listeria monocytogenes infection. PLoS One, 2010. 5(1): p. e8610. 220. Ribet, D., et al., Listeria monocytogenes impairs SUMOylation for efficient infection. Nature, 2010. 464(7292): p. 1192-5. 221. Gekara, N.O. and S. Weiss, Lipid rafts clustering and signalling by listeriolysin O. Biochem Soc Trans, 2004. 32(Pt 5): p. 712-4. 222. Gekara, N.O., et al., The cholesterol-dependent cytolysin listeriolysin O aggregates rafts via oligomerization. Cell Microbiol, 2005. 7(9): p. 1345-56. 223. Cossart, P. and P.J. Sansonetti, Bacterial invasion: the paradigms of enteroinvasive pathogens. Science, 2004. 304(5668): p. 242-8. 224. Pizarro-Cerda, J. and P. Cossart, Bacterial adhesion and entry into host cells. Cell, 2006. 124(4): p. 715-27. 225. Valdez, Y., R.B. Ferreira, and B.B. Finlay, Molecular mechanisms of Salmonella virulence and host resistance. Curr Top Microbiol Immunol, 2009. 337: p. 93-127. 226. Lara-Tejero, M., et al., A sorting platform determines the order of protein secretion in bacterial type III systems. Science, 2011. 331(6021): p. 1188-91.
139
227. Galan, J.E., Common themes in the design and function of bacterial effectors. Cell Host Microbe, 2009. 5(6): p. 571-9. 228. Gellin, B.G., et al., The epidemiology of listeriosis in the United States--1986. Listeriosis Study Group. Am J Epidemiol, 1991. 133(4): p. 392-401. 229. Jackson, K.A., M. Iwamoto, and D. Swerdlow, Pregnancy-associated listeriosis. Epidemiol Infect, 2010. 138(10): p. 1503-9. 230. Gaillard, J.L., F. Jaubert, and P. Berche, The inlAB locus mediates the entry of Listeria monocytogenes into hepatocytes in vivo. J Exp Med, 1996. 183(2): p. 359-69. 231. Stavru, F., C. Archambaud, and P. Cossart, Cell biology and immunology of Listeria monocytogenes infections: novel insights. Immunol Rev, 2011. 240(1): p. 160-84. 232. Seveau, S., J. Pizarro-Cerda, and P. Cossart, Molecular mechanisms exploited by Listeria monocytogenes during host cell invasion. Microbes Infect, 2007. 9(10): p. 1167-75. 233. Pizarro-Cerda, J. and P. Cossart, Listeria monocytogenes membrane trafficking and lifestyle: the exception or the rule? Annu Rev Cell Dev Biol, 2009. 25: p. 649-70. 234. Ireton, K., Entry of the bacterial pathogen Listeria monocytogenes into mammalian cells. Cell Microbiol, 2007. 9(6): p. 1365-75. 235. Vazquez-Boland, J.A., et al., Pathogenicity islands and virulence evolution in Listeria. Microbes Infect, 2001. 3(7): p. 571-84. 236. Mengaud, J., et al., E-cadherin is the receptor for internalin, a surface protein required for entry of L. monocytogenes into epithelial cells. Cell, 1996. 84(6): p. 923-32. 237. Seveau, S., et al., A FRET analysis to unravel the role of cholesterol in Rac1 and PI 3- kinase activation in the InlB/Met signalling pathway. Cell Microbiol, 2007. 9(3): p. 790- 803. 238. Pizarro-Cerda, J., M. Bonazzi, and P. Cossart, Clathrin-mediated endocytosis: what works for small, also works for big. Bioessays, 2010. 32(6): p. 496-504. 239. Schnupf, P. and D.A. Portnoy, Listeriolysin O: a phagosome-specific lysin. Microbes Infect, 2007. 9(10): p. 1176-87. 240. Henry, R., et al., Cytolysin-dependent delay of vacuole maturation in macrophages infected with Listeria monocytogenes. Cell Microbiol, 2006. 8(1): p. 107-19. 241. Bergmann, B., et al., InlA- but not InlB-mediated internalization of Listeria monocytogenes by non-phagocytic mammalian cells needs the support of other internalins. Mol Microbiol, 2002. 43(3): p. 557-70. 242. Gaillard, J.L., P. Berche, and P. Sansonetti, Transposon mutagenesis as a tool to study the role of hemolysin in the virulence of Listeria monocytogenes. Infect Immun, 1986. 52(1): p. 50-5. 243. Billington, S.J., B.H. Jost, and J.G. Songer, Thiol-activated cytolysins: structure, function and role in pathogenesis. FEMS Microbiol Lett, 2000. 182(2): p. 197-205. 244. Mitchell, T.J. and P.W. Andrew, Biological properties of pneumolysin. Microb Drug Resist, 1997. 3(1): p. 19-26. 245. Wickham, S.E., et al., Mapping the intermedilysin-human CD59 receptor interface reveals a deep correspondence with the binding site on CD59 for complement binding proteins C8alpha and C9. J Biol Chem, 2011. 286(23): p. 20952-62. 246. Thiery, J., et al., Perforin activates clathrin- and dynamin-dependent endocytosis, which is required for plasma membrane repair and delivery of granzyme B for granzyme- mediated apoptosis. Blood, 2010. 115(8): p. 1582-93.
140
247. Portnoy, D.A., P.S. Jacks, and D.J. Hinrichs, Role of hemolysin for the intracellular growth of Listeria monocytogenes. J Exp Med, 1988. 167(4): p. 1459-71. 248. Bakardjiev, A.I., et al., Listeriosis in the pregnant guinea pig: a model of vertical transmission. Infect Immun, 2004. 72(1): p. 489-97. 249. Smith, K. and P. Youngman, Use of a new integrational vector to investigate compartment-specific expression of the Bacillus subtilis spoIIM gene. Biochimie, 1992. 74(7-8): p. 705-11. 250. Camilli, A., L.G. Tilney, and D.A. Portnoy, Dual roles of plcA in Listeria monocytogenes pathogenesis. Mol Microbiol, 1993. 8(1): p. 143-57. 251. Pushkareva, V.I. and S.A. Ermolaeva, Listeria monocytogenes virulence factor Listeriolysin O favors bacterial growth in co-culture with the ciliate Tetrahymena pyriformis, causes protozoan encystment and promotes bacterial survival inside cysts. BMC Microbiol, 2010. 10: p. 26. 252. Haghighat, A.C. and S. Seveau, Quantification of host-microbe interactions by automated fluorescence microscopy. J Immunol Methods, 2010. 352(1-2): p. 186-91. 253. Arnett, E., et al., Defensins enable macrophages to inhibit the intracellular proliferation of Listeria monocytogenes. Cell Microbiol, 2011. 13(4): p. 635-51. 254. Hotze, E.M., et al., Arresting pore formation of a cholesterol-dependent cytolysin by disulfide trapping synchronizes the insertion of the transmembrane beta-sheet from a prepore intermediate. J Biol Chem, 2001. 276(11): p. 8261-8. 255. Nato, F., et al., Production and characterization of neutralizing and nonneutralizing monoclonal antibodies against listeriolysin O. Infect Immun, 1991. 59(12): p. 4641-6. 256. Marriott, H.M., T.J. Mitchell, and D.H. Dockrell, Pneumolysin: a double-edged sword during the host-pathogen interaction. Curr Mol Med, 2008. 8(6): p. 497-509. 257. Kumari, S., S. Mg, and S. Mayor, Endocytosis unplugged: multiple ways to enter the cell. Cell Res, 2010. 20(3): p. 256-75. 258. Krawczyk-Balska, A. and J. Bielecki, Listeria monocytogenes listeriolysin O and phosphatidylinositol-specific phospholipase C affect adherence to epithelial cells. Can J Microbiol, 2005. 51(9): p. 745-51. 259. Bielecki, J., Association of listeriolysin O with the cell surface of Listeria monocytogenes. Acta Microbiol Pol, 1994. 43(3-4): p. 279-89. 260. Price, K.E. and A. Camilli, Pneumolysin localizes to the cell wall of Streptococcus pneumoniae. J Bacteriol, 2009. 191(7): p. 2163-8. 261. Rottner, K., T.E. Stradal, and J. Wehland, Bacteria-host-cell interactions at the plasma membrane: stories on actin cytoskeleton subversion. Dev Cell, 2005. 9(1): p. 3-17. 262. Mayor, S. and R.E. Pagano, Pathways of clathrin-independent endocytosis. Nat Rev Mol Cell Biol, 2007. 8(8): p. 603-12. 263. Iliev, A.I., et al., Cholesterol-dependent actin remodeling via RhoA and Rac1 activation by the Streptococcus pneumoniae toxin pneumolysin. Proc Natl Acad Sci U S A, 2007. 104(8): p. 2897-902. 264. Cossart, P., J. Pizarro-Cerda, and M. Lecuit, Invasion of mammalian cells by Listeria monocytogenes: functional mimicry to subvert cellular functions. Trends Cell Biol, 2003. 13(1): p. 23-31. 265. Ibarra, J.A. and O. Steele-Mortimer, Salmonella--the ultimate insider. Salmonella virulence factors that modulate intracellular survival. Cell Microbiol, 2009. 11(11): p. 1579-86.
141
266. Parsot, C., Shigella type III secretion effectors: how, where, when, for what purposes? Curr Opin Microbiol, 2009. 12(1): p. 110-6. 267. Cossart, P. and E. Veiga, Non-classical use of clathrin during bacterial infections. J Microsc, 2008. 231(3): p. 524-8. 268. Iacovache, I., M. Bischofberger, and F.G. van der Goot, Structure and assembly of pore- forming proteins. Curr Opin Struct Biol, 2010. 20(2): p. 241-6. 269. Gonzalez, M.R., et al., Bacterial pore-forming toxins: the (w)hole story? Cell Mol Life Sci, 2008. 65(3): p. 493-507. 270. Fernandes, M.C., et al., Trypanosoma cruzi subverts the sphingomyelinase-mediated plasma membrane repair pathway for cell invasion. J Exp Med, 2011. 208(5): p. 909-21. 271. Talbot, U.M., A.W. Paton, and J.C. Paton, Uptake of Streptococcus pneumoniae by respiratory epithelial cells. Infect Immun, 1996. 64(9): p. 3772-7. 272. Sukeno, A., et al., Intermedilysin is essential for the invasion of hepatoma HepG2 cells by Streptococcus intermedius. Microbiol Immunol, 2005. 49(7): p. 681-94. 273. Zhang, J.R., et al., The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell, 2000. 102(6): p. 827-37. 274. Ammendolia, M.G., et al., Invasive pathway of Listeria ivanovii in human amnion-derived WISH cells. Int J Immunopathol Pharmacol, 2007. 20(3): p. 509-18. 275. Russell, B.H., et al., Bacillus anthracis internalization by human fibroblasts and epithelial cells. Cell Microbiol, 2007. 9(5): p. 1262-74. 276. Jost, B.H. and S.J. Billington, Arcanobacterium pyogenes: molecular pathogenesis of an animal opportunist. Antonie Van Leeuwenhoek, 2005. 88(2): p. 87-102. 277. Logsdon, L.K., et al., Streptolysin O inhibits clathrin-dependent internalization of group A Streptococcus. MBio, 2011. 2(1): p. e00332-10. 278. Hamon, M.A., et al., Listeriolysin O: the Swiss army knife of Listeria. Trends Microbiol, 2012. 20(8): p. 360-8. 279. Edelson, B.T., P. Cossart, and E.R. Unanue, Cutting edge: paradigm revisited: antibody provides resistance to Listeria infection. J Immunol, 1999. 163(8): p. 4087-90. 280. Mackaness, G.B., Cellular resistance to infection. J Exp Med, 1962. 116: p. 381-406. 281. Hage-Chahine, C.M., et al., Hemolysin-producing Listeria monocytogenes affects the immune response to T-cell-dependent and T-cell-independent antigens. Infect Immun, 1992. 60(4): p. 1415-21. 282. Vadia, S., et al., The pore-forming toxin listeriolysin O mediates a novel entry pathway of L. monocytogenes into human hepatocytes. PLoS Pathog, 2011. 7(11): p. e1002356. 283. Bonnegarde-Bernard, A., et al., IKKbeta in intestinal epithelial cells regulates allergen- specific IgA and allergic inflammation at distant mucosal sites. Mucosal Immunol, 2013. 284. Disson, O., et al., Modeling human listeriosis in natural and genetically engineered animals. Nat Protoc, 2009. 4(6): p. 799-810. 285. Dunstone, M.A. and R.K. Tweten, Packing a punch: the mechanism of pore formation by cholesterol dependent cytolysins and membrane attack complex/perforin-like proteins. Curr Opin Struct Biol, 2012. 22(3): p. 342-9. 286. Boyaka, P.N., et al., Chimeras of labile toxin one and cholera toxin retain mucosal adjuvanticity and direct Th cell subsets via their B subunit. J Immunol, 2003. 170(1): p. 454-62.
142
287. Meza-Sanchez, D., et al., Intradermal immunization in the ear with cholera toxin and its non-toxic beta subunit promotes efficient Th1 and Th17 differentiation dependent on migrating DCs. Eur J Immunol, 2011. 41(10): p. 2894-904. 288. Edelson, B.T. and E.R. Unanue, Intracellular antibody neutralizes Listeria growth. Immunity, 2001. 14(5): p. 503-12. 289. Xu, W. and J.J. Zhang, Stat1-dependent synergistic activation of T-bet for IgG2a production during early stage of B cell activation. J Immunol, 2005. 175(11): p. 7419-24. 290. Miki, K. and G.B. Mackaness, The Passive Transfer of Acquired Resistance to Listeria Monocytogenes. J Exp Med, 1964. 120: p. 93-103. 291. Lane, F.C. and E.R. Unanue, Requirement of thymus (T) lymphocytes for resistance to listeriosis. J Exp Med, 1972. 135(5): p. 1104-12. 292. Paton, J.C., et al., Purification and immunogenicity of genetically obtained pneumolysin toxoids and their conjugation to Streptococcus pneumoniae type 19F polysaccharide. Infect Immun, 1991. 59(7): p. 2297-304. 293. Jost, B.H., et al., Immunization with genetic toxoids of the Arcanobacterium pyogenes cholesterol-dependent cytolysin, pyolysin, protects mice against infection. Infect Immun, 2003. 71(5): p. 2966-9. 294. Cowan, G.J., et al., Immunisation with anthrolysin O or a genetic toxoid protects against challenge with the toxin but not against Bacillus anthracis. Vaccine, 2007. 25(41): p. 7197-205. 295. Sun, R. and Y. Liu, Listeriolysin O as a strong immunogenic molecule for the development of new anti-tumor vaccines. Hum Vaccin Immunother, 2013. 9(5): p. 1058-1068. 296. Carrero, J.A., H. Vivanco-Cid, and E.R. Unanue, Listeriolysin o is strongly immunogenic independently of its cytotoxic activity. PLoS One, 2012. 7(3): p. e32310. 297. Wallecha, A., et al., Listeria monocytogenes-derived listeriolysin O has pathogen- associated molecular pattern-like properties independent of its hemolytic ability. Clin Vaccine Immunol, 2013. 20(1): p. 77-84. 298. Galan, J.E. and H. Wolf-Watz, Protein delivery into eukaryotic cells by type III secretion machines. Nature, 2006. 444(7119): p. 567-73. 299. Sanchez-Martinez, S., et al., Membrane-active peptides derived from picornavirus 2B viroporin. Curr Protein Pept Sci, 2012. 13(7): p. 632-43. 300. Ishino, T., Y. Chinzei, and M. Yuda, A Plasmodium sporozoite protein with a membrane attack complex domain is required for breaching the liver sinusoidal cell layer prior to hepatocyte infection. Cell Microbiol, 2005. 7(2): p. 199-208. 301. Fernandez-Puentes, C. and L. Carrasco, Viral infection permeabilizes mammalian cells to protein toxins. Cell, 1980. 20(3): p. 769-75. 302. Hayes, C.S., S.K. Aoki, and D.A. Low, Bacterial contact-dependent delivery systems. Annu Rev Genet, 2010. 44: p. 71-90. 303. Kafsack, B.F. and V.B. Carruthers, Apicomplexan perforin-like proteins. Commun Integr Biol, 2010. 3(1): p. 18-23. 304. Pizarro-Cerda, J., A. Kuhbacher, and P. Cossart, Entry of Listeria monocytogenes in mammalian epithelial cells: an updated view. Cold Spring Harb Perspect Med, 2012. 2(11). 305. Bubert, A., et al., Differential expression of Listeria monocytogenes virulence genes in mammalian host cells. Mol Gen Genet, 1999. 261(2): p. 323-36.
143
306. Moors, M.A., et al., Expression of listeriolysin O and ActA by intracellular and extracellular Listeria monocytogenes. Infect Immun, 1999. 67(1): p. 131-9. 307. Meixenberger, K., et al., Listeria monocytogenes-infected human peripheral blood mononuclear cells produce IL-1beta, depending on listeriolysin O and NLRP3. J Immunol, 2010. 184(2): p. 922-30. 308. TranVan Nhieu, G., et al., Calcium signalling during cell interactions with bacterial pathogens. Biol Cell, 2004. 96(1): p. 93-101. 309. Gekara, N.O., et al., Listeria monocytogenes desensitizes immune cells to subsequent Ca2+ signaling via listeriolysin O-induced depletion of intracellular Ca2+ stores. Infect Immun, 2008. 76(2): p. 857-62. 310. Berridge, M.J., P. Lipp, and M.D. Bootman, The versatility and universality of calcium signalling. Nat Rev Mol Cell Biol, 2000. 1(1): p. 11-21. 311. Bischofberger, M., I. Iacovache, and F.G. van der Goot, Pathogenic pore-forming proteins: function and host response. Cell Host Microbe, 2012. 12(3): p. 266-75. 312. Husmann, M., et al., Elimination of a bacterial pore-forming toxin by sequential endocytosis and exocytosis. FEBS Lett, 2009. 583(2): p. 337-44. 313. Larkin, J.M., et al., Depletion of intracellular potassium arrests coated pit formation and receptor-mediated endocytosis in fibroblasts. Cell, 1983. 33(1): p. 273-85. 314. Moya, M., et al., Inhibition of coated pit formation in Hep2 cells blocks the cytotoxicity of diphtheria toxin but not that of ricin toxin. J Cell Biol, 1985. 101(2): p. 548-59. 315. Cossart, P., et al., Listeriolysin O is essential for virulence of Listeria monocytogenes: direct evidence obtained by gene complementation. Infect Immun, 1989. 57(11): p. 3629-36. 316. Khelef, N., et al., Species specificity of the Listeria monocytogenes InlB protein. Cell Microbiol, 2006. 8(3): p. 457-70. 317. Hakansson, A., et al., Cytolysin-dependent evasion of lysosomal killing. Proc Natl Acad Sci U S A, 2005. 102(14): p. 5192-7. 318. Gregory, S.H., A.J. Sagnimeni, and E.J. Wing, Expression of the inlAB operon by Listeria monocytogenes is not required for entry into hepatic cells in vivo. Infect Immun, 1996. 64(10): p. 3983-6. 319. Gregory, S.H., A.J. Sagnimeni, and E.J. Wing, Internalin B promotes the replication of Listeria monocytogenes in mouse hepatocytes. Infect Immun, 1997. 65(12): p. 5137-41.
144