University of Groningen Streptococcus pneumoniae ... · brain microvascular endothelial cells, and that PECAM-1 forms a double receptor with plgR on the plasma membr ane of endothelial

University of Groningen

Streptococcus pneumoniae interactions with endothelial cells leading to invasivepneumococcal diseaseIovino, Federico

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2013

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Iovino, F. (2013). Streptococcus pneumoniae interactions with endothelial cells leading to invasivepneumococcal disease. [Groningen]: s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 02-04-2020

https://www.rug.nl/research/portal/en/publications/streptococcus-pneumoniae-interactions-with-endothelial-cells-leading-to-invasive-pneumococcal-disease(1ef6add4-b90a-45cd-afc7-53f58a52566b).html



Streptococcus pneumoniae interactions with endothelial

cells leading to invasive pneumococcal disease

Federico Iovino

ISBN

978-90-367-6426-1

Copyright

All right reserved. No part of this publication may be reported or transmitted in any from or by any means without the permission of the author and the publisher holding the copyright of the published articles.

Cover

Streptococcus pneumoniae adhered to blood vessels in the subarachnoid space of mouse brain detected by confocal microscopy.

Layout and printing

Lovebird desgin & printing solutions www.lovebird-design.com

Ceuiralc l I Stellingen

Medischc M Behorende bij het proefschrift

B1bliothet•k C

Gruningr.•strepMc. ccus pneumoniae interactions with endothelial cells leading

to invasive pneumococcal disease"

• "A PhD degree does not turn a person into a researcher, but the universe inside the mind of a person makes this person a researcher" My wonderful grandma Clelia

"Un dottorato di ricerca 11011 fa di una persona un ricercatore, ma e l'universo nella mente di ciascuno di noi a renderci tutti ricercatori" La mia meravigliosa nonna Clelia

• "Biology gives you a brain. Life turns it into a mind" "La biologia ti dona un cervello. La vita lo trasforma in wrn mente" Jeffrey Eugenides

• "The local immune system of the brain is activated by the presence of S. pneumoniae even before meningitis develops" This thesis, Chapter 2

• "Neither the in vitro nor the in vivo studies in the published scientific literature provide unequivocal evidence that 5. pneumoniae anchors to the PAFR in vivo" This thesis, Chapter 3

• "PAFR, which facilitates pneumococcal adhesion to brain endothelial cells, does not colocalize w ith S. pneumoniae in vivo and likely has no physical interaction with this bacterium in vivo" This thesis, Chapter 4

• "It seems likely that 5. pneumoniae in the blood stream can bind to PECAM-1 expressed by brain microvascular endothelial cells, and that PECAM-1 forms a double receptor with plgR on the plasma membr ane of endothelial cells for 5. pneumoniae adhesion." This thesis, Chapter 5

• "Only the stupid needs order, the genius controls the chaos" "Solo lo stupido ha bisogno di ordine, ii genio controlla ii chaos" A. Einstein

• "By knowing the pathways by which 5. pneumoniae adheres to endothelial cells , it becomes possible to interfere with these mechanisms and to develop strategies for preventing and curing pneumococcal meningitis" This thesis, Chapter 6

• "The fun of scientific research is also to find always new borders to overcome, to build the most powerful tools of investigation, the most complex theories, to try always to move forward and to know that we get nearer and nearer to understanding the reality, without never arriving to understand it completely." "II divertimento della ricerca scientifica e anche trovare sempre altre Jrontiere da superare, costruire mezzi pit't potenti d'indagine, teorie pii't complesse, cercare sempre di progredire pur sapendo che probabilmente ci si avvicinerii sempre di piz't a comprende re la realtii, senza arrivare mai a capirla co111pleta111ente". Margherita Hack, interview by Flavia Farina / Margherita Hack, intervista di Flavia Farina

RIJKSUNIVERSITEIT GRONINGEN

Streptococcus pneumoniae interactions with endothelial

cells leading to invasive pneumococcal disease

Proefschrift

ter verkrijging van het doctoraat in de

Medische Wetenschappen

aan de Rijksuniversiteit Groningen

op gezag van de

Rector Magnificus, dr. E. Sterken,

in het openbaar te verdedigen op

rnaandag 11 novernber 2013

om 16.15 uur

door

Federico Iovino

geboren op 28 decernber 1984

te Pavia, Italie

Ccntr,11� '.\.tedi�che B1L,Ji11thcek Gruniagcn

Promo tores:

Copromotor:

Beoordelingscommissie:

Prof. Dr. J.M. van Dijl

Prof. Dr. G. Molema

Dr. J.J.E. Bijlsma

Prof. Dr. H.E. de Vries

Prof. Dr. J. Tommassen

Prof. Dr. H.C. van der Mei

Paranimfen Paranymphs Henrik Gradstedt

Vahid Farshchi Andisi

Peter Nagle

The work described in this thesis was performed in the laboratory of Molecular Bacteriology, Department of medical Microbiology of the University Medical Center Groningen and University of Groningen.

Printing of this thesis was financially supported by the Graduated school for Drug

Exploration (GUIDE) and University of Groningen.

�.,;: / rijks�.miversiteit

� gronmgen umcG

Index of contents

Chapter 1 Introduction and scope of the thesis

Chapter 2 Interactions between blood-borne Streptococcus pneumoniae and the blood-brain barrier leading to meningitis

Iovino et al., Plos One (2013) 8 (7): e68408

Chapter 3 Signalling or binding: the role of the platelet activating factor receptor in invasive pneumococcal disease

Iovino et al., Cellular Microbiology (2013) 15 (6): 870-81

Chapter 4 Novel insights in the roles of the platelet-activating factor and the poly immunoglobulin receptors in Streptococcus pneumoniae adhesion to the brain vascular endothelium

Submitted

Chapter 5 Platelet endothelial cell adhesion molecule-1 (PECAM-1), a putative receptor for the adhesion of Streptococcus pneumoniae to the vascular endothelium of the blood-brain barrier

Manuscript in preparation

Chapter 6 General summary and future perspectives

Appendices I. Nederlandse samenvating

II. List of publications

III. Acknowledgements/Ringraziamenti

9

25

51

73

97

121

135

141

145

CHAPTERl

Introduction and scope of the thesis

- 9 -

Chapter 1

Streptococcus pneumoniae: carriage, disease and virulence factors

The Gram-positive bacterium Streptococcus pneumoniae (the pneumococcus) is

an important commensal resident of the human nasopharynx. Although carriage is

usually non-symptomatic, S. pneumoniae can become invasive and spread from the

upper respiratory tract to other organs leading to serious diseases such as pneumonia,

otitis media, meningitis, or bacteremia (Figure 1).

.................... t•acterem1a

Figure 1. Schematic overview of pneumococcal diseases and sites of infection

World Health Organization (WHO) data reported that pneumonia is the single

largest cause of death in children worldwide and, every year, approximately 1.4

million children younger than 5 years old die because of this disease (http://www.

who.int/madiacentre/factsheets/fs331/en/). Pneumonia is a disease of the lungs and

can be divided into two forms, bronchial pneumonia and lobar pneumonia. Bronchial

pneumonia is most prevalent in infants, young children and aged adults and S.

pnewnoniae is one of the main causative agents of the disease. Bronchial pneumonia is

characterized by acute inflammation localized in the bronchioles. Lobar pneumonia

occurs more frequently in younger adults. More than 80% of the cases of lobar

pneumonia are caused by S. pneumoniae. This type of pneumonia affects one specific

area of the lung, called "lob" and it often happens that more lobs are involved. S.

pneumoniae is also the main cause of community-acquired pneumonia (CAP) in both

children and adults (1). Otitis media (OM) is an infection of the middle ear that can

be caused by bacteria and viruses. Main forms of this disease are acute OM (AOM),

OM with effusion (OME), and recurrent otitis media (ROM). AOM is one of the

most frequent childhood infections and the major causative agent is S. pneumoniae (2). Meningitis is an inflammation of the membranes, which cover and protect the

11

Chapter 1

brain and spinal cord. The inflammation can be caused by both viruses and bacteria. Approximately 1.2 million cases are estimated to occur annually worldwide (3). S.

pneumoniae and the Gram-negative bacterium Neisseria meningitidis are the main etiologic agents causing this disease in Europe and in the USA (3). Worldwide, many S. pneumoniae strains have developed resistance to penicillin, cephalosporins and macrolides, and some of them are resistant to multiple classes of drugs, complicating choices of treatment and therapy strategies (4-7).

The pathogenicity of S. pneumoniae has been attributed to various elements, so called virulence factors, most of which are situated on the surface of the bacterium. The polysaccharide capsule provides resistance to phagocytosis and facilitates the escape of S. pneumoniae from host immune defenses (8, 9). The capsule is recognized as the major virulence factor of S. pneumoniae. In animal experiments, unencapsulated strains showed 50% decrease in lethal dose compared to the encapsulated ones and encapsulated strains were found to be at least 105 times more virulent than strains without capsule (10, 11 ). The capsule surrounds the bacterial cell wall, which is mainly composed of peptidoglycan, a compound containing amino acids ( or peptides) linked to polysaccharides. Cell wall polysaccharides (CWPS) have been found to induce an inflammatory response similar to the response to whole pneumococci. It was shown in fact that injection in animals of purified cell wall can mimic the clinical features typical of pneumococcal pneumonia, otitis media and meningitis (12, 13, 14), and antiCWPS have been demonstrated to protect animals against S. pneumoniae challenge (15-19). Other factors, including cell wall components and the toxin pneumolysin, are involved mainly in the inflammation caused by infection (8, 9). Pneumolysin is an intracellular protein that belongs to the class of thiol-activated toxins (20, 21). It is conserved among S. pneumoniae isolates and it has a variety of toxic effects on different cell types. Although the pneumococcus does not secrete pneumolysin, this toxin can be released upon lysis of the bacteria under the influence of autolysin (20). Pneumolysin stimulates the production of inflammatory cytokines, such as tumor necrosis factor alpha and interleukin-1 by human monocytes (22), inhibits the beating of cilia on human respiratory epithelial cells, disrupts monolayers of epithelial cells from the upper respiratory tract (23) and from the alveoli (24), and decreases the bactericidal activity and migration of neutrophils (25). Pneumolysin is also capable of activating the complement pathway (26).

S. pneumoniae contains phosphorylcholine on the teichoic acids present both in the cell wall and the cytoplasmic membrane. In pneumococci, there are proteins bound to phosphorylcholine, the so called choline-binding proteins (CBPs). CbpA is the largest and most abundant representative of the CBPs (27, 28). CbpA is also secreted by pneumococci, and it has a number of biological functions that may modulate the immune system in response to pneumococcal infection (27, 28). These

12

Chapter 1

functions include binding to the secretory component of secretory IgA (29), the

complement protein C3 (30) and human factor H (31, 32). CbpA is hypothesized to

act as adhesion factor in the interactions between pneumococci and the host cell, and

it was shown to play a crucial role in S. pneumoniae adhesion to the nasopharyngeal

tract epithelium (33) and the brain endothelium (34).

Pneumococcal vaccines

The development of an effective vaccine to protect from pneumococcal

infections has been a slow process, most probably due to the poor immunogenicity

of the polysaccharides present on the bacterial surface, which represents the

primary target of neutralizing antibodies. In the early 1980s, a vaccine which

consisted of purified capsular polysaccharides derived from 23 of the known S.

pneumoniae serotypes (PPV23) was marketed in the USA and later on in Europe. The

pneumococcal serotypes included in this vaccine were responsible for 85-90% of

the invasive pneumococcal disease (IPD) cases among adults (35). A study on trials

with this vaccine in the period between 1994 and 2009 raised doubts concerning the

efficacy of PPV23, especially in developing countries where HIV infections are a

major cause of death (36, 37, 38). Actually, there is little evidence that PPV23 protects

against pneumococcal infections in the high-risk group of adults with HIV (36, 37,

38). The limitations and doubts regarding PPV23 have underscored the importance

of developing alternative vaccine strategies. A valid variant is represented by the so

called "conjugate" vaccines. These vaccines are called "conjugated" because capsular

polysaccharides are chemically conjugated to a highly immunogenic protein carrier,

such as tetanus or diphtheria toxoid. The immunological advantage of this new

class of vaccines is that they induce both B-cell-dependent and T-cell-dependent

responses, and also induce a memory response to a booster dose of the vaccine (39).

A conjugated vaccine containing capsular polysaccharide from seven S. pneumoniae

serotypes (PCV7) was licensed in the USA in 2000. The entry of PCV7 in the

pharmaceutical market tremendously decreased IPD in children (40). New conjugate

vaccines are being tested for vaccination in young children and adults, such as a

10-valent vaccine (PCVlO), which has been licensed in over 30 countries, and also a

13-valent vaccine (PCV13) (41). PCV13 in particular, due to its increased coverage of

pneumococcal serotypes, may expand the benefits of conjugate vaccines also to the

high-risk group of adults with HIV. Lastly, new trials are currently in progress for

the development of vaccines based on pneumococcal proteins, such as pneumolysin,

instead of polysaccharides. Proteins are highly conserved across different serotypes

and may thus confer more protection against most S. pneumoniae serotypes (42).

13

Chapter 1

The Central Nervous System, the brain vasculature and the blood-brain barrier

The experimental chapters in this thesis are focused on the interactions of S. pneumoniae with the endothelium of the blood-brain barrier leading to the invasion of pneumococci into the brain preceding meningitis. The Central Nervous System (CNS) consists of the brain and the spinal cord. In vertebrates, the brain is covered and protected by the skull, and enclosed in the meninges. The brain is the center of the nervous system in all vertebrate and most invertebrate animals. The physiological function of the brain is to exert a centralized control over all the compartments of the body by generating patterns of muscle activity and guiding secretion of chemicals called hormones. The brain is primarily composed of two classes of cells, the neurons and glial cells. Neurons are the basic building blocks of the nervous system (43). They consist of a central cell body and an axon, which is a thin protoplasmic fiber that extends from the cell body and projects with various branches to other areas nearby. Neurons send signals to specific target cells over long distances (44). The axons transmit signals in the form of electrochemical pulses, called action potentials (45), to other neurons through specialized neuron-neuron connections called synapses. One single axon can make thousands of synaptic connections with other neurons (46). When an action potential travels along the axon, it reaches the synaptic junction and causes the release of a chemical, called neurotransmitter, which binds to the membrane of the target cell ( 47). The crucial function of the brain is the cell-to-cell communication and synapses are the key of this communication processes in the brain (48).

The blood circulatory system of the brain comprises the microvasculature, which includes arterioles, capillaries, and venules, and the macrovasculature, containing arteries and veins (Figure 2A,B) (49).

The blood-brain barrier (BBB) is composed by a system of blood vessels which supply the whole brain, from the periphery to the inner parts, with nutrients and oxygen. The BBB separates the brain from the circulatory system of the rest of the body and protects the CNS from potentially harmful substances and regulates the transport of essential molecules for the maintenance of a stable environment (Figure 3) (50). The BBB has highly specialized properties to selectively control the permeability between blood and the CNS, a highly selective transport systems for example for mediating the flux of solutes and is also a metabolic barrier due to the presence of enzymes able to metabolize specific molecules in transit. Lastly, the BBB is also a physical barrier due to its tight endothelial cell to cell junctions (50). Endothelial cells (ECs) line all the blood vessels and they represent a crucial component of the BBB. Brain ECs express on their plasma membrane proteins and receptors which provide selective routes of entry for nutrients, glucose (GLUT-1), ions (Na, K-ATPase and Na, K) and macromolecules (e.g. insulin), and routes of exit

14

Figure 2A. Mouse brain vasculature stained with Dy Light 594 labeled Lycopersicon Esculentum Lectin, marker for blood vessels in brain of rodents

Chapter 1

SA2 Subarachnoid space

C= Cerebral cortex

SE= Septum

CP= Choroid plexus

Figure 2B. Schematic representation of mouse brain section

for potentially toxic metabolic substances (51). Astrocytes, also known as astroglia,

are characteristic star-shaped cells in the brain. They are the most abundant cells in

the human brain and they have several important functions, including biochemical

support for endothelial cells of the BBB, provision of nutrients to the nervous tissue,

and repair of the brain and spinal cord following traumatic injuries (52, 53).

The brain can be considered an "immune privileged" organ because it is

separated from the rest of the body by the BBB, which functions as a "road block"

to prevent pathogens from reaching the nervous tissue. In the case that infectious

agents or harmful substances invade the nervous tissue, the brain requires a quick

immune response to control and decrease the inflammation and eradicate the

infectious agent. Microglia are the macrophages of the brain and spinal cord, and

represent the main form of active immune defense in the CNS. Microglia are able

to phagocytose foreign materials and display these molecules for T-cell activation.

Phagocytic microglia travel to the site of the injury and secrete pro-inflammatory

factors to promote more microglial cells to proliferate and do the same. Microglia

also interact with astrocytes and neurons to fight and eradicate infections as quickly

as possible (53, 54, 55).

The endothelium

The endothelium is the layer of cells that line the interior surface of blood

vessels (56). The cells forming the endothelium are called endothelial cells. Those

endothelial cells that are in contact with the blood are called vascular endothelial

cells. Endothelial cells have a crucial role in many aspects of vascular biology. The

15

Chapter 1

meni11ge5

duro malor �7 ----..;::::::::=:::::::::::::::========,::::::::.-:;:::::. orochnold

pia matll"r

endothelial cells

- -blood

Figure 3. Schematic representation of the blood brain barrier

endothelium acts as a barrier between the lumen of the vessel and the surrounding tissue, and controls the trafficking of materials and white blood cells in and out of the bloodstream. TI1e endothelium provides a non-thrombogenic surface that contains heparan sulfate which acts as co-factor for the activation of the protease anti-thrombin which prevents thrombus formation (56). Endothelial activation, induced by pro-inflammatory cytokines, such as interleukin 1 and tumor necrosis factor a, allows endothelial cells to participate in the inflammatory response and is mostly characterized by an increase of the interactions of endothelial cells with leukocytes (57). Endothelial cells are also necessary for the formation of new blood vessels from pre-existing ones upon e.g. damage, a process known as angiogenesis, and furthermore these cells execute fundamental functions to assist in the control of the blood pressure, such as vasoconstriction and vasodilatation (56).

Itisabsolutelyimportantthattheendotheliumremainsinanunperturbedcondition to optimize the expression of anticoagulant and anti thrombotic activities. However, the endothelium is a very dynamic organ that responds to various environmental changes and is subjected to events that can cause disorders. Plasma factors such as antibodies or lipoproteins which perturb the endothelial cell functions have been identified in vitro (56). A pathological state of the endothelium can contribute to several disease processes, such as hypertension, hypercholesterolaemia, diabetes and septic shock (58). Endothelial cells express on their plasma membrane PECAM-1, one of the major endothelial adhesion molecules (59). PECAM-1 is involved in the interactions between leukocytes and endothelial cells and in leukocyte transendothelial migration during inflammation and neuroinflammation (60). PECAM-1 is also involved in development

16

Chapter 1

of infectious diseases. Recently, Lovelace et al. biochemically demonstrated that human and murine PECAM-1 bind to S. typhimurium (61). Furthermore, PECAM-1-1- mice are more resistant than wild-type mice to oral infection upon oral administration of S. hjphimurium (61 ). These novel findings indicate that PECAM-1 is not only able to bind S. typhimurium, but that PECAM-1 also has a functional role in the oral infection caused by Salmonella. Other studies showed that PECAM-1 is required for neutrophil recruitment during many inflammatory responses, but not during pneumococcal pneumonia (62, 63). However, whether PECAM-1 is capable to bind S. pneumoniae is not known. The integrity of the endothelium is maintained by the intercellular junctions between the endothelial cells. VE-Cadherin for example, is an adhesion molecule mainly expressed on the cell-cell junctions, which is of vital importance for the maintenance and control of endothelial cell contacts (64, 65).

Pathogens can invade the brain only after crossing the endothelial cell layer of the blood-brain barrier. It has been hypothesized that bacteria can adhere to epithelial and endothelial cells by a receptor-mediated adhesion process. Meningeal bacterial pathogens, such as S. pneumoniae and N. meningitidis, are able to interact with specific receptors expressed by the plasma membrane of endothelial cells and once they are bound to these receptors they can cross the endothelial layer through a process called transcytosis and transmigrate from the blood vessels of the blood-brain barrier to the nervous tissue of the brain (34, 66).

Meningitis

As mentioned above, meningitis is an inflammation of the meninges, the protective membranes covering the brain and the spinal cord (67). In adults, the most common clinical symptom of meningitis is severe headache, occurring in almost 90% of bacterial meningitis cases. The classic clinical picture of meningitis consists of nuchal rigidity, sudden high fever and an altered mental status. These three symptoms are observed in 44-46% of the bacterial meningitis cases (68, 69). Other signs associated with suspected meningitis are photophobia (intolerance to bright light), phonophobia (intolerance to loud noise) and, especially in young children, leg pain, cold extremities and abnormal skin color (70, 71). Bacterial meningitis is a disease with high morbidity and mortality worldwide despite the implementation of several vaccination programs and antimicrobial agents (72). As mentioned above, the most common etiological agents are Streptococcus pneumoniae and Neisseria meningitis, known also as the meningococcus, which are responsible for 80% of all cases (72). In particular, S. pneumoniae is responsible for two-thirds of cases in Europe and in the USA (73). Meningitis must be treated with antibiotics, which can penetrate the blood-brain barrier (BBB), such as chloramphenicol, tetracyclines, sulphonamides, erythromycin, sulphadiazine (74).

17

Chapter 1

When the immune system of the CNS tries to eradicate the pathogens, it starts defensive actions that eventually can lead to worsening of the infection and inflammation. During meningitis the BBB becomes more permeable allowing large numbers of leukocytes to enter the brain. This influx of leukocytes during the inflammatory response leads to intracerebral edema and meningeal swelling (75, 76). In particular during pneumococcal meningitis, microglia are stimulated to produce nitric oxide after interacting with streptococci, which is neurotoxic and can worsen the inner edema (77). Pro-inflammatory cytokines are molecules which are produced to enhance inflammation and they can be released by different types of cell, such as endothelial cells, macrophages, neutrophils. Concentrations of Interleukin (IL) 1 and Tumor Necrosis Factor (TNF) alpha were found to be significantly higher in the cerebrospinal fluid of patients with bacterial meningitis in comparison with controls (78). The inflammatory features and clinical complications characteristic of bacterial meningitis in humans can be reproduced using animal models and administration of bacteria directly into the cistema magna of the brain (79, 80, 81). It has been reported that the pro-inflammatory cytokines IL-1 beta, IL-6 and TNF alpha are significantly higher in animals with meningitis as compared with control groups (79, 80, 82, 83, 84).

Bacterial translocation over the blood-brain barrier

Bacteria reach the meninges by two main routes, namely through the bloodstream or through direct contact between meninges and the external environment (e. g. skull fractures), but in most cases meningitis follows invasion by the pathogenic bacteria through the bloodstream (72). Bacteria present in the bloodstream, once having reached the blood vessels in the brain, have to cross the BBB to enter the brain and cause infection. How bacterial pathogens cross the BBB is currently unclear. Several possible mechanisms have been implicated in this process, such as: 1) the destruction of the endothelial cell layers in case of, for example, N. meningitidis (85); 2) traversal of the BBB in between the endothelial cells by disruption of the tight junctions (86); and 3) traversal of the BBB by transcytosis, an intracellular transport route for the transport of molecules and vesicles through cells from the apical to basolateral side (87).

In meningococcal meningitis, it was shown that N. meningitidis requires adhesion of type IV pili to brain endothelial cells to cross the blood-brain barrier (85). The type IV pili-mediated adhesion to brain endothelial cells provokes the disruption of junctions between endothelial cells (85). N. meningitidis recruits the Par3/Par6/PKCzeta polarity complex to the cell-cell interface of the endothelial layer of the blood-brain barrier and this recruitment leads to the depletion of intercellular junctions between brain endothelial cells. The depletion of these proteins causes the

18

Chapter 1

disruption of the cell-cell junctions in the endothelium, thus opening the way for N .meningitidis to enter the brain (85). How S. pneumoniae crosses the blood-brain barrier is presently unclear and subject of debate. Pneumococci could pass between endothelial cells either via pericellular transport systems (transport between cells) or via mechanistic damage, by releasing pneumolysin. It was observed in vitro that S. pneumoniae and group A streptococci can bind to plasminogen in the blood and use this binding to cleave cell-cell junctions (86, 87) . S. pneumoniae could traverse the cellular barriers within leukocytes during leukocyte extravasion, as it was observed in vivo as a possible invasion route for Listeria monocytogenes and Mycobacterium tuberculosis (88). Another proposed mechanism is transcystosis, a specific receptor mediated process of bacterial passage through epithelial and endothelial cells (33, 34). Endothelial and epithelial cells express certain receptors on the plasma membrane, and binding of bacteria to these receptors facilitates the transmigration of the pathogens from the apical to the basolateral side of the host cell. The plateletactivating factor receptor (PAFR) and laminin receptor (LR) have been shown to mediate adhesion of S. pneumoniae to brain endothelial cells (89, 90). It was shown that S. pneumoniae binds to the poly immunoglobulin receptor (plgR) expressed on the epithelium of the upper respiratory tract (33), but whether plgR also has a role in pneumococcal translocation across the blood-brain barrier is unknown.

Receptors involved in S. pneumoniae adhesion to human cells

As mentioned above, bacterial pathogens have the capability to bind to certain receptors on the plasma membrane of epithelial and endothelial cells and this receptor-mediated binding facilitates the bacterial invasion into and translocation over cell layers. One receptor implicated in adhesion to, invasion of, and also transcytosis through endothelial cells is PAFR. PAFR is a G-protein coupled receptor with seven transmembrane domains and its natural ligand is the platelet-activating factor (PAF). PAF is a mediator in diverse pathologic processes, such as allergy, asthma, septic shock, arterial thrombosis, and inflammation (91, 92). Normally PAF binds to PAFR and activates a GTPase function, causing phospholipid turnover via the phospholipases C, D, and A2 pathways, and also activates protein kinase C and tyrosine kinases (91). PAFR has been investigated for many years and there are several inhibitors and antibodies which inhibit its biological activity. PAFR has been proposed to bind S. pneumoniae and as such facilitate adhesion to, uptake by and transcytosis through endothelial cells leading to invasive disease (34, 89, 93).

The laminin receptor (LR) is an important molecule involved in adhesion of the plasma membrane of epithelial and endothelial cells to the basement membrane. LR on endothelial cells interacts with neurotropic viruses, including Sindbis virus (94), Dengue virus (95), adeno-associated virus (96), tick-borne encephalitis virus,

19

Chapter 1

and Venezuelan equine encephalitis virus (97). LR was identified to be a common receptor for both S. pneumoniae and N. meningitidis on the surface of rodent and human brain microvascular endothelial cells (90). Fluorescent beads coated with CbpA, that were injected intravascularly into mice, adhered to the cerebral endothelium. Pre-treatment with anti LR antibody led to an inhibition of the adherence of CbpAbeads to the endothelium in mice, thus suggesting that the laminin receptor initiates bacterial contact with the blood-brain barrier endothelium (90).

The poly immunoglobulin receptor (pigR) transports immunoglobulins across mucosal epithelial barriers (98, 99, 100). S. pneumoniae can bind to pigR expressed by human nasopharyngeal epithelial cells and this binding facilitates pneumococcal translocation through the nasopharyngeal epithelium (33). Human nasopharyngeal epithelial (Detroit) cells were treated with an antibody against pigR prior to bacterial infection. The antibody treatment significantly reduced adhesion of pneumococci (33), indicating that pigR is functionally involved in pneumococcal adhesion to nasopharyngeal epithelium. Furthermore, immunoblotting analyses showed that pneumococcal CbpA binds to human pigR indicating that CbpA may be required to mediate the binding of S. pneumoniae to this receptor (101 ).

Outline and scope of the thesis

As summarized in the introductory Chapter 1 of this thesis, bacterial meningitis is thought to occur as the result of bacteria crossing the blood-brain barrier to invade the CNS. Yet, little is known about the steps preceding the disease development. In particular, where and how adhesion of S. pneumoniae to the bloodbrain barrier endothelium takes place in vivo is still unclear. The aim of the study described in Chapter 2 was to investigate the spatio-temporal interactions of bloodborne S. pneumoniae to the endothelium of the blood-brain barrier during the events leading up to meningitis, and the protective response of the CNS towards systemic pneumococcal infection. The results show that S. pneumoniae adheres to blood-brain barrier endothelium in vivo, and likely, that the adhesion to the brain microvascular endothelium is the first step by which pneumococci cross the blood-brain barrier. Although leukocyte detection in the brain during the entire course of infection was minimal, microglia and astrocytes were activated early on, already after 1 hour post infection, indicative that the local immune system in the brain was activated directly upon the presence of pneumococci in the blood.

The role PAFR in the adhesion of S. pneumoniae to human cells, especially endothelial cells, is hampered by a scarcity of biochemical characterization of the bacteria-receptor binding, and compounded by conflicting results in the literature. In Chapter 3, the current literature on PAFR, S. pneumoniae and other pathogens is discussed, including data concerning human PAFRgenetic variations related to clinical

20

Chapter 1

aspects of invasive pneumococcal disease (IPD), to shed light on the importance of PAFR in IPD. The in vitro and most of the in vivo studies analyzed and discussed in this chapter indicate that PAFR is involved in development of IPD. However, there is no unequivocal evidence that a direct interaction between S. pneumoniae and PAFR occurs in vivo.

Chapter4 documents investigations aimed at elucidating a possible role of PAFR in S. pneumoniae adhesion to the endothelium of the blood-brain barrier. Previous studies have shown that pigR is expressed in the epithelium of the respiratory tract (33, 100, 101) and neurons (102, 103, 104). Although there are no data about the expression of plgR on the brain vasculature in vivo, pigR expression was not detectable in cultured Human Brain Microvascular Endothelial Cells (HBMEC) (33). Importantly, at the start if this PhD research it had not been investigated yet whether pigR plays a role in S. pneumoniae adhesion to the brain vascular endothelium. Therefore, a study aimed at clarifying the potential role of pigR in S. pneumoniae

adhesion to the endothelial cells of the blood-brain barrier was undertaken. The results presented in this chapter show that a direct interaction between S. pneumoniae

and PAFR on the brain microvascular endothelium is not likely to occur in vivo.

Furthermore, it is demonstrated that pigR is expressed on the brain microvascular endothelial cells and that endothelial pigR may act as receptor for S. pneumoniae

adhesion to the blood-brain barrier endothelium Chapter 5 reports on a study conducted to investigate the role of PECAM-1/

CD31 in S. pneumoniae pathogenesis during pneumococcal meningitis. In particular, it was aimed at determining whether PECAM-l/CD31 could be a receptor for S

pneumoniae adhesion to the endothelium of the blood-brain barrier. The results presented in this chapter show that S. pneumoniae co-localizes with PECAM-1 in the brain of intravenously infected mice and furthermore that S. pneumoniae can bind to PECAM-1 in vitro. Taken these results together it is hypothesized that PECAM-1 may play a role as adhesion receptor for S. pneumoniae on the blood-brain barrier endothelium.

In Chapter 6 a new model for pneumococcal adhesion to endothelial cells is presented in which a cooperative role of PAFR, pigR and PECAM-1 is proposed. Furthermore, it is suggested that, by defining the pathways by which S. pneumoniae

adheres to endothelial cells, it may be possible to interfere with the respective mechanisms and develop strategies to prevent and cure pneumococcal meningitis.

Summarizing, the PhD research described in this thesis has provided novel insights in the adhesion of S. pneumoniae to the brain vascular endothelium, a crucial step in the passage of pneumococci across the blood-brain barrier leading to meningitis.

21

Chapter 1

References 1. Mandell LA, Bartlett JG, Dowen SF, File TM Jr, Muscher DM, et al. (2003) Update of practice guidelines

for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis 37: 1405-1433.

2. Pichichero ME (2000) Recurrent and persistent otitis media. Pediatric Infect Dis J 19: 911-916.

3. van de Beek D, de Gans J, Tunkel AR, Wijdicks EF (2006) Community-acquired bacterial meningitis in adults. N Engl J Med 354: 44-53.

4. Welte T, Kohnlein T (2009) Global and local epidemiology of community-acquired pneumonia: the experience of the CAPNETZ Network. Semin Respir Crit Care Med 30: 127-135.

5. Song JH, Chang HH, Suh JY et al. , ANSORP Study Group (2004) Macrolide resistance and genotypic characterization of Streptococcus pneumoniae in Asian countries: a study of the Asian Network for Surveillance of Resistant Pathogens (ANSORP). J Antimicrob Chemother 53: 457-463.

6. Doern GV, Richter SS, Miller A, Miller N, Rice C, et al. (2005) Antimicrobial resistance among Streptococcus pneumoniae in the United States: have we begun to tum the comer on resistance to certain antimicrobial classes? Clin Infect Dis 41: 139- 148.

7. Cant CM, Rosingh AW, L6pez-HontangasJL, van der Heijden M, Gonzalez-Moran F, et al. (2012) Serotype distribution and antimicrobial resistance of invasive pneumococcal disease strains in the Comunidad Valenciana, Spain, during the winter of 2009-2010: low PCV7 coverage and high levofloxacin resistance. Antimicrob Agents Chemother 56: 4988-9.

8. Alonso De Velasco E, Verheul AF, Verhoef J, Snippe H (1995) Streptococcus pneumoniae: virulence factors, pathogenesis, and vaccines. Microbial Rev 59: 591-603

9. Mitchen AM, Mitchell 1J (2010) Streptococcus pneumoniae: virulence factors and variation. Clin Microbial Infect 16: 411-8.

10. Avery OT, and Dubas R (1931) The protective action of a specific enzyme against type III pneumococcus infection in mice. J Exp Med 54: 73-89.

11. Watson DA, and Musher OM (1990) Interruption of capsule production in Streptococcus pneu111011iac serotype 3 by insertion of transposon Tn916. Infect. Immun 58: 3135-3138.

12. Carlsen BO, Kawana M, Kawana C, Tomasz A, Giebink GS (1992) Role of the bacterial cen wan in middle ear inflammation caused by Streptococcus pneu111oniae. Infect Immun 60: 2850--2854.

13. Tuomanen E, Liu H, Hengstler B, Zak 0, Tomasz A (1985) The induction of meningeal inflammation by components of the pneumococcal cen wall. J Infect Dis 151: 859-868.

14. Tuomanen E, Rich R, Zak O (1987) Induction of pulmonary inflammation by components of the pneumococcal cell surface. Am Rev Respir Dis 135 :869-874.

15. Briles DE, Claffin JL, Schroer K, Forman C (1981) Mouse IgC3 antibodies are highly protective against infection with Streptococcus pneumoniae. Nature (London) 294: 88-90.

16. Briles DE, Forman C, Crain M (1992) Mouse antibody to phosphocholine can protect mice from infection with mouse-virulent human isolates of Streptococcus pneumoniae. Infect Immun 60: 1957-1962.

17. Briles DE, Forman C, Horowitz JC, Volanakis JE, Benjamin WH, et al. (1989) Antipneumococcal effects of C-reactive protein and monoclonal antibodies to pneumococcal cen wan and capsular antigens. Infect Immun 57: 1457 1464.

18. Kenny D, Cuelde G, Fischer RT, Longo DL (1994) Induction of phosphocholine-specific antibodies in X-linked immune deficient mice: in vivo protection against a Streptococcus pneumoniae challenge. Int Immunol 6: 561-568.

19. Nicoletti C, Yang X, Cerny J (1993) Repertoire diversity of antibody response to bacterial antigens in aged mice. III. Phosphorylcholine antibody from young and aged mice differ in structure and protective activity against infection with Streptococcus pneumoniae. J Immunol 150: 54�549.

20. Boulnois GJ (1992) Pneumococcal proteins and the pathogenesis of disease caused by Streptococcus pneumoniae. J Gen Microbiol 138 :249-259.

21. Marriott HM, Mitchell 1J, Dockrell DH (2008) Pneumolysin: a double-edged sword during the host-

22

Chapter 1

pathogen interaction. Curr Mo! Med 8: 497-509.

22. Houldsworth S, Andrew PW, Mitchell lJ (1994) Pneumolysin stimulates production of tumor necrosis factor alpha and interleukin-lb by human mononuclear phagocytes. Infect Immun 62: 1501-1503.

23. Feldman C, Mitchell lJ, Andrew PW, Boulnois CJ, Read RC, et al. (1 990) The effect of Streptococcus pnc11111011iac pneumolysin on human respiratory epithelium in vitro. Microb Pathog 9: 275-284.

24. Rubins JB, Duane PG, Clawson D, Charboneau D, Young}, Niewoehner DE (1993) Toxicity of pneumolysin to pulmonary alveolar epithelial cells. Infect Immun 61: 1352-1358.

25. Paton JC, Ferrante A (1983) Inhibition of human polymorphonuclear leukocyte respiratory burst, bactericidal activity, and migration by pneumolysin. Infect Immun 41 : 1212-1216.

26. Paton JC, Rowan-Kelly B, Ferrante A (1 984) Activation of human complement by the pneumococcal toxin pneumolysin. Infect Immun 43: 1085-7.

27. Gosink KK, Mann ER, Guglielmo C, Tuomanen EI, Masure HR (2000)

28. Role of novel choline binding proteins in virulence of Streptococcus p11eu111oniae. Infect Immun 68: 5690-5.

29. Rosenow C, Ryan P, Weiser JN, Johnson S, Fontan P, et al. (1997) Contribution of novel choline-binding proteins to adherence, colonization, and immunogenicity of Streptococcus p11eu111011iac. Mo! Microbial 25: 819-825.

30. Hammerschmidt S, Talay SR, Brandtzaeg P, Chhatwal GS (1997) SpsA, a novel pneumococcal surface protein with specific binding to secretory immunoglobulin A and secretory component. Mo! Microbiol 25: 1113-24.

31. Cheng Q Finkel D, Hostetter MK (2000) Novel purification scheme and functions for a C3-binding protein from Streptococcus p11eu111oniae. Biochemistry 39: 5450-7.

32. Janulczyk R, Iannelli F, Sjoholm AG, Pozzi G, Bjorck L (2000) Hie, a novel surface protein of Streptococcus pneumoniae that interferes with complement function. J Biol Chem 275: 37257-63.

33. Dave S, Brooks-Walter A, Pangburn MK, McDaniel LS (2001) PspC, a pneumococcal surface protein, binds human factor H. Infect Immun 69: 3435-7.

34. Zhang JR, Mostov KE, Lamm ME, Nanna M, Shimida S, et al. (2000) The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102: 827-37.

35. Ring A, Weiser JN, Tuomanen EI (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 102: 347-60.

36. CDC. Prevention of pneumococcal disease: recommendations of the Advisory Committee on Immunization Practice (ACIP) (1997) MMWR Morb Wkly Rep 46: 1-24.

37. Feikin DR, Elie CM, Goetz MB, Lennox JL, Carlone GM, et al. (2001) Randomized trial of the quantitative and functional antibody responses to a 7-valent pneumococcal conjugate vaccine and/or 23-valent polysaccharide vaccine among HIV-infected adults. Vaccine.20: 545-553.

38. Dransfield MT, Nahm MH, Han MK, Hamden S, Criner GJ, et al. (2009) Superior immune response to protein-conjugate versus free pneumococcal polysaccharide vaccine in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 180: 499-505.

39. Kumar D, Chen MH, Welsh B, Siegal D, Cobos I, et al. (2007) A randomized, double-blind trial of pneumococcal vaccination in adult allogeneic stem cell transplant donors and recipients. Clin Infect Dis 45: 1576-1582.

40. Musher OM, Manof SB, Liss C, Mcfetridge RD, Marchese RD et al. (2010) Safety and antibody response, including antibody persistence for 5 years, after primary vaccination or revaccination with pneumococcal polysaccharide vaccine in middle-aged and older adults. J Infect Dis 201 : 516-524.

41. Harboe ZB, Benfield TL, Valentiner-Branth P, Hjuler T, Lambertsen L, et al. (2010) Temporal trends in invasive pneumococcal disease and pneumococcal serotypes over 7 decades. Clin Infect Dis 50: 329-337.

42. Blasi F, Mantero M, Santus P, Tarsia P (2012) Understanding the burden of pneumococcal disease in adults. Clin Microbial Infect 5: 7-14.

43. Tai SS (2006) Streptococcus p11eu111011iae protein vaccine candidates: properties, activities and animal studies. Crit Rev Microbial 32: 139-153

23

Chapter 1

44. Principles of Neural Science, first published in 1981 by Elsevier, edited by Eric R. Kandel, James H. Schwartz, and Thomas M. Jessell, p20.

45. Principles of Neural Science, first published in 1981 by Elsevier, edited by Eric R. Kandel, James H. Schwartz, and Thomas M. Jessen, p21.

46. Barnett MW, Larkman PM (2007) The action potential. Pract Neurol. 7: 192-7.

47. Principles of Neural Science, first published in 1981 by Elsevier, edited by Eric R. Kandel, James H. Schwartz, and Thomas M. Jessen, Ch. 10, p. 175

48. Principles of Neural Science, first published in 1981 by Elsevier, edited by Eric R. Kandel, James H. Schwartz, and Thomas M. Jessen, Ch. 10.

49. de Vries HE, Kuiper J, de Boer AG, Van Berkel 1JC, Breimer DD (1997). "The Blood-Brain Barrier in Neuroinflammatory Diseases". Pharmacological Reviews 49: 143-156.

50. Shepherd, GM (2004). The Synaptic Organization of the Brain. Oxford University Press US. "Ch. 1 : Introduction to synaptic circuits".

51. Tam SJ, Watts RJ (2010) Connecting vascular and nervous system development: angiogenesis and the blood-brain barrier. Annu Rev Neurosd 33: 379-408

52. Abbott NJ, Patabendige AA, Dolman DE, Yusof SR, Begley DJ (2010) Structure and function of the bloodbrain barrier. Neurobiol Dis 37: 13-25.

53. Volterra A, Meldolesi J (2005) Astrocytes, from brain glue to communication elements: the revolution continues. Nat Rev Neurosci 6: 626----640.

54. Liu W, Tang Y, Feng J (2011) Cross talk between activation of microglia and astrocytes in pathological conditions in the central nervous system. Life Sci 89: 141 -6.

55. Kreutzberg GW (1995) Microglia, the first line of defence in brain pathologies. Arzneimittelforschung 45: 357-60.

56. Aloisi F (2001) Immune function of microglia. Glia 36: 165-79.

57. Cines DB, Pollak ES, Buck CA, Loscalzo J, Zimmerman GA, et al. (1998) Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood 91: 3527-61 .

58. Pober JS (2002) Endothelial activation: intracellular signaling pathways. Arthritis Res 4 Suppl 3 : S109-16.

59. Mi.inzel T, Sinning C, Post F, Wamholtz A, Schulz E (2008) Pathophysiology, diagnosis and prognostic implications of endothelial dysfunction. Ann Med. 40: 180-96.

60. Newman PJ, Newman DK (2003) Signal transduction pathways mediated by PECAM-1 : new roles for an old molecule in platelet and vascular cell biology. Arterioscler Thromb Vase Biol 23: 953-64

61. Kalinowska A, Losy J (2006) PECAM-1, a key player in neuroinflammation. Eur J Neurol 13: 1284-90.

62. Lovelace MD, Yap ML, Yip J, Muller W, Wijburg 0, Jackson DE (2013) Absence of Platelet Endothelial Ce11 Adhesion Molecule 1, PECAM-l/CD31, In Vivo Increases Resistance to Sa/111011el/a enterica Serovar Typhimurium in Mice. Infect Immun 81 : 1952-63.

63. Jones MR, Simms BT, Lupa MM, Kogan MS, Mizgerd JP (2005) Lung NF-kappaB activation and neutrophil recruitment require IL-1 and TNF receptor signaling during pneumococcal pneumonia. J Immunol 175: 7530-5.

64. Tasaka S, Qin L, Saijo A, Albelda SM, DeLisser HM, Doerschuk CM (2003) Platelet endothelial cell adhesion molecule-1 in neutrophil emigration during acute bacterial pneumonia in mice and rats. Am J Respir Crit Care Med 167: 164-70.

65. Vestweber D (2008) Ve-cadherin: the major endothelial adhesion molecule controlling cellular junctions and blood vessel formation. Arterioscler Thromb Vase Biol 28: 223-32.

66. Corada M, Mariotti M, Thurston G, Smith K, Kunkel R (1999) Vascular endothelial-cadherin is an important determinant of microvascular integrity in vivo. Proc Natl Acad Sci U S A 96: 9815-20.

67. Coureuil M, Join-Lambert 0, Lecuyer H, Bourdoulous S, Marullo S, Nassif X (2012) Mechanism of meningeal invasion by Neisseria meningitidis. Virulence 3: 164-72.

68. Saez-Llorens X, McCracken GH (2003). Bacterial meningitis in children. Lancet 361: 2139-48.

24

Chapter 1

69. van de Beek D, de Gans J, Spanjaard L, Weisfelt M, Reitsma JB, Vermeulen M (2004) Clinical features and prognostic factors in adults with bacterial meningitis. N Eng J Med 351: 1849-59.

70. Attia J, Hatala R, Cook DJ, Wong JG (1999) The rational clinical examination. Does this adult patient have acute meningitis? JAMA 282: 175--81 .

71. Theilen U, Wilson L, Wilson G, Beattie JO, Qureshi S, Simpson D (2008) Management of invasive meningococcal disease in children and young people: Summary of SIGN guidelines. BMJ 336: 1367-70.

72. Edinburgh: Scottish Intercollegiate Guidelines Network (SIGN) (2008) Management of invasive meningococcal disease in children and young people.

73. van de Beek D, de Gans J, Tunkel AR, Wijdicks EF (2006) Community-acquired bacterial meningitis in adults. N Engl J Med 354: 44-53.

74. Mook-Kanamori BB, GeldhoffM, van der Poll T, van de Beek D (201 1) Pathogenesis and pathophysiologyof pneumococcal meningitis. Clin Microbiol 24: 557-91 .

75. Nau R, Sorge! F, Eiffert H (2010) Penetration o f drugs through the blood-cerebrospinal fluid/blood-brain barrier for treabnent of central nervous system infections. Clin Microbiol Rev 23: 858-83.

76. Brouwer MC, Tunkel AR, van de Beek D (2010) Epidemiology, diagnosis, and antimicrobial treabnent of acute bacterial meningitis. Clin Microbiol Rev 23: 467-92

77. Woehr! B, Klein M, Grandgirard D, Koedel U, Leib S (201 1) Bacterial meningitis: current therapy and possible future treabnent options. Expert Rev Anti Infect Tuer 9: 1053-65.

78. Aloisi F (2001 ) Immune function of microglia. Glia 36: 165-79.

79. Ohga S, Aoki T, Okada K,Akeda H, Fujioka K (1994) Cerebrospinal fluid concentrations of interleukin-I beta, tumour necrosis factor-alpha, and interferon gamma inbacterial meningitis. Arch Dis Child 70: 1 23-5.

80. Mook-Kanamori B, Geldhoff M, Troost D, van der Poll T, van de Beek D (2012) Characterization of a pneumococcal meningitis mouse model. BMC Infect Dis 12: 71.

81. Paul R, Angele B, Sporer B, Pfister HW, Koedel U (2004) Inflammatory response during bacterial meningitis is unchanged in Fas- and Fas ligand-<leficient mice. J Neuroimmuno 152: 78-82.

82. Paul R, Winkler F, Bayerlein I, Popp B, Pfister HW, Koedel U (2005) Urokinase-type plasminogen activator receptor regulates leukocyte recruibnent during experimental pneumococcal meningitis. J Infect Dis 191: 776--782.

83. Barrichello T, Fagundes GD, Generoso JS, Paula Moreira A, Costa CS, et al. (2012) Brain-blood barrier and proinflammaory mediators in neonate rats submitted meningitis by Streptococcus pneumoniae. Brain Res 1471: 162-8.

84. Barichello T, dos Santos I, Savi GD, Florentino AF, Silvestre C, et al. (2009) Tumor necrosis factor alpha (TNF-alpha) levels in the brain and cerebrospinal fluid after meningitis induced by Streptococcus pneumonine. Neurosci Lett 467: 217-9.

85. 0stergaard C, Brandt C, Konradsen HB, Samuelsson S (2004) Differences in survival, brain damage, and cerebrospinal fluid cytokine kinetics due to meningitis caused by 3 different Streptococcus p11e11111011ine serotypes: evaluation in humans and in 2 experimental models. J Infect Dis 190: 1212-20.

86. Coureuil M, Mjkaty G, Miller F, Lecuyer H, Bernard C, et al. (2009) Meningococcal type N pili recruit the polarity complex to cross the brain endothelium. Science 325: 83-7.

87. Pancholi V, Fontan P, Jin H (2003) Plasminogen-mediated group A streptococcal adherence to and pericellular invasion of human pharyngeal cells. Microb Pathog 35: 293-303.

88. Attali C, Durmort C, Vernet T, Di Guilmi AM (2008) The interaction of Streptococcus pne11111011ine with plasmin mediates transmigration across endothelial and epithelial monolayers by intercellular junction cleavage. Infect lmmun 76: 5350-6.

89. Drevets DA, Leenen PJ, Greenfield RA (2004) Invasion of the central nervous system by intracellular bacteria. Clin Microbiol Rev 17: 323-47.

90. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1995) Streptococcus p11e11111011ine anchor to activated human cells by the receptor for platelet-activating factor. Nature 377: 435-8.

25

Chapter 1

91. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, et al. (2009) Laminin receptor initiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest 1 19 : 1638-46.

92. Shukla SD (1992) Platelet-activating factor receptor and signal transduction mechanisms. FASEB J 6: 2296-301 .

93. Honda Z, Ishii S, Shimizu T (2002) Platelet-activating factor receptor. J Biochem 131: 773-9.

94. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2005) beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pncumoniac. Infect Tmmun 73:7 827-35.

95. Wang K, Kuhn R, Strauss E, Ou S, Strauss J (1992) High affinity laminin receptor is a receptor for Sindbis virus in mammalian cells. J Virol 66: 4992-5001.

96. Thepparit C, Smith D (2004) Serotype-specific entry of dengue virus into liver cells: identification of the 37 kDa/67 kDa high affinity laminin receptor as a dengue virus serotype 1 receptor. J Virol 78: 12647-12656.

97. Akache B, Grimm D, Pandey K, Yant S, Zu, H, Kay M (2006) The 37/67 kilodalton laminin receptor is a receptor for adeno-associated virus serotypes 8,2,3 and 9. J Virol 80: 9831-9836.

98. Ludwig C, Kondig J, Smith J (1996) A putative receptor for Venezuelan equine encephalitis virus from mosquito cells. J Virol 70: 5592-5599.

99. Johansen FE, Natvig Norderhaug I, R0e M, Sandlie I, Brandtzaeg P (1999) Recombinant expression of polymeric IgA: incorporation of J chain and secretory component of human origin. Eur J Immunol 29: 1 701-8. Shimad S, Kawaguchi-Miyashita M, Kushiro A, Sato T, Nanno M, et al. (1999) Generation of polymeric immunoglobulin receptor-deficient mouse with marked reduction of secretory IgA. J Immunol 163: 5367-73.

100. Asano M, Komiyama K (2011 ) Polymeric immunoglobulin receptor. J Oral Sci 53: 147-56.

101. Lu L, Lamm ME, Li H, Corthesy B, Zhang JR (2003) The human polymeric immunoglobulin receptor binds to Streptococcus pneumoniae via domains 3 and 4. J Biol Chem 278: 48178-87.

102. Ikonen E, Parton RG, Hunziker W, Simons K, Dotti CG (1993) Transcytosis of the polymeric immunoglobulin receptor in cultured hippocampal neurons. Curr Biol 3: 635-44.

103. de Hoop M, von Poser C, Lange C, Ikonen E, Hunziker W, Dotti CG (1995) Intracellular routing of wild-type and mutated polymeric immunoglobulin receptor in hippocampal neurons in culture. J Cell Biol 130: 1447-59.

104. Hemar A, Olivo JC, Williamson E, Saffrich R, Dotti CG (1997) Dendroaxonal transcytosis of transferrin in cultured hippocampal and sympathetic neurons. J Neurosci 17: 9026-34.

26

CHAPTER 2

Interactions between blood-borne Streptococcus pneumoniae and the blood-brain barrier preceding meningitis

Federico Iovino, Carlos J. Orihuela, Henk E. Moorlag, Grietje Molema, Jetta J.E. Bijlsma

PLoS One (2013) 8: e68408

27

Abstract

Streptococcus pneumoniae (the pneumococcus) is a Gram-positive bacterium and the predominant cause of bacterial meningitis. Meningitis is thought to occur as the result of pneumococci crossing the blood-brain barrier to invade the Central Nervous System (CNS); yet little is known about the steps preceding immediate disease development. To study the interactions between pneumococci and the vascular endothelium of the blood-brain barrier prior to meningitis, we used an established bacteremia-derived meningitis model in combination with immunofluorescent imaging. Brain tissue of mice infected with S. pneumoniae strain TIGR4, a clinical meningitis isolate, was investigated for the location of the bacteria in relation to the brain vasculature in various compartments. We observed that S. pneumoniae adhered preferentially to the subarachnoid vessels, and subsequently, over time, reached the more internal cerebral areas including the cerebral cortex, septum, and choroid plexus. Interestingly, pneumococci were not detected in the choroid plexus till 8 hours-post infection. In contrast to the lungs, little to no leukocyte recruitment to the brain was observed over time, though Iba-1 and GFAP staining showed that microglia and astrocytes were activated as soon as 1 hour post-infection. Our results imply that i) the local immune system of the brain is activated immediately upon entry of bacteria into the bloodstream and that ii) adhesion to the blood brain barrier is spatiotemporally controlled at different sites throughout the brain. These results provide new information on these two important steps towards the development of pneumococcal meningitis.

Chapter 2

Introduction

Streptococcus pneumoniae (the pneumococcus) is a Gram-positive human

pathogen that causes life-threatening invasive diseases such as pneumonia and

bacteremia with high morbidity and mortality throughout the world. Moreover,

5. pneumoniae is now the most common etiological agent of bacterial meningitis in

the United States and Europe [1 ]. Meningitis is an infection of the Central Nervous

System (CNS), which is characterized by inflammation of the protective membranes

covering the brain and spinal cord, collectively known as the meninges [2, 3]. One

entry route for S. pneumoniae into the CNS is thought to be via the bloodstream

by crossing the blood vessels of the blood-brain barrier. The blood-brain barrier is

composed of a specialized vasculature covered by endothelial cells that protects the

brain from harmful substances that are present in the blood stream and supplies

the brain with the required nutrients for its proper functions, making the brain

an immune privileged organ [2, 3]. The entry of pathogens into the CNS leads to

inflammation, local production of pro-inflammatory cytokines and activation of the

endothelium. The blood-brain barrier becomes more permeable allowing for the

influx of large numbers of leukocytes into the brain. This influx of leukocytes during

the inflammatory response leads to intracerebral edema and swelling of the meninges

and these pathological changes lead to neurological damage and in some cases death

of the patient [l, 3]. High levels of the pro-inflammatory cytokines, such as interleukin

1 beta (IL-1 beta), interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF alpha),

were detected in the cerebro-spinal fluid (CSF) of patients with meningitis, which is

typical of the severe inflammation of the brain [4]. Astrocytes, also known as astroglia,

are characteristic star-shaped cells in the brain. They are the most abundant cells in

the human brain and they have several important functions, including biochemical

support for endothelial ceIIs of the blood-brain barrier, provision of nutrients to the

nervous tissue, and repair of the brain and spinal cord foIIowing traumatic injuries

[5, 6]. Microglia are the resident macrophages of the brain and spinal cord and they

act as first immune defense of the CNS in response to infections [6, 7] . Microglia

are able to phagocytose foreign molecules and present them to T cells. Phagocytic

microglia travel to the site of injury and release pro-inflammatory factors to promote

the proliferation of more microglial cells. Microglia also interact with astrocytes and

neurons to fight and eradicate the infections as quickly as possible [6, 7, 8]. During

inflammation microglia and astrocytes get activated and their cellular morphology

undergoes dramatic changes in a process called astrogliosis for astrocytes: the soma

becomes more round and the dentrites shorter and thicker [9, 10, 11, 12].

How bacterial pathogens cross the blood-brain barrier is currently unclear.

Several studies have implicated 1) the destruction of the endothelial cell layers by N.

meningitidis and pneumococcal pneumolysin [13, 14], 2) traversal of the blood-brain

29

Chapter 2

barrier in between the cells by disruption of the tight junctions [15], and 3) bacterial traversal of the blood-brain barrier by transcytosis, an intracellular transport route designed to transport molecules and vesicles through cells from the apical to the basolateral side [16].

The blood circulatory system of the brain comprises the microvasculature, which includes arterioles, capillaries, and venules, and the macrovasculature, containing arteries and veins [17]. At present, it is unknown whether S. pneumoniae interacts with both the micro- and the macrovasculature prior to invasion into the brain, when it interacts with the vasculature in different areas in the brain, and whether all endothelial cells along the microvascular tree engage in bacterial adhesion and transmigration. It is thought that an important site of entry might be the choroid plexus [18], as shown for Streptococcus suis, which induces blood-brain barrier disruption in the porcine choroid plexus [19], and may also translocate intracellularly across the choroid plexus epithelium [20]. In human brain microvascular endothelial cells (HBMEC), pneumococci were able to adhere to the platelet-activating factor (PAF) receptor expressed by these cells, allowing transmigration through the endothelial cell to the basolateral site (16, 21]. Concordantly, PAF receptor-deficient mice showed a decreased incidence of pneumococcal meningitis after intravenous challenge due to less pneumococcal translocation across the blood-brain barrier (21]. In vitro and animal experiments showed in strain D39 that the pneumococcal surface protein C (PspC, also known as CbpA) is required for development of bacterial meningitis, and that TIGR4-derived recombinant PspC binds to the laminin receptor, which is also targeted by other neurotropic pathogens including prions and bacteria, such as Haemophilus

influenzae and Neisseria meningitidis (22]. Summarizing, while pneumococcal translocation occurs during natural disease

pathogenesis, it is still unclear how and where this process takes place in vivo. The aim of this study was to investigate the interactions between S. pneumoniae and the brain endothelium before the development of meningitis. Since an important route of CNS infection by bacterial pathogens is via the blood stream, we challenged mice with S. pneumoniae through intravenous injection. Herein we report that S. pneumoniae indeed adheres tightly to the vascular endothelium, preferentially to the subarachnoid vessels and over time, increasingly to the endothelium of the cerebral cortex, septum and choroid plexus. This adhesion results in almost immediate microglial and astroglial activation but does not lead to leukocyte recruitment. Thus, our studies suggest that S. pneumoniae CNS invasion is spatiotemporally separated

30

Chapter 2

Material and methods

Bacterial strains and growth conditions

An encapsulated S. pneumoniae, serotype 4 strain, TIGR4 [22], was used in this study. Pneumococci were grown standing in Todd-Hewitt broth (Oxoid Thermo Scientific, Basingstoke, England) at 37°C, bacterial growth was monitored by measuring the optical density (OD) at 600nm with a spectrophotometer. Bacteria were harvested at 0D600= 0.25-0.30 by centrifugation of 1 ml of culture at 10000 rpm for 3 minutes, and the bacteria were suspended in 1ml of sterile phosphate buffered saline (PBS) (Lonza, Verviers, Belgium). Serial dilutions in sterile PBS were made and plated on blood-agar plates to calculate the dilutions for the challenge dose of 107

colony forming units (CPU) for each mouse.

Bacteremia-derived meningitis model

All experiments involving animals were performed with the prior approval of and in accordance with guidelines of the Institutional Animal Care and Use Committee of the University of Groningen (DEC nr. 6152A). The bacteremia-derived meningitis model described by Orihuela et al. [22] was performed in the following way: four groups of 5 female Balb/c mice, 6 to 8 weeks old (Harlan, Horst, NL) were anesthetized by inhalation of 2.5% isofluorane before the challenge. Intravenous tail vein injection with 200µ1 of 107 CPU of the TIGR4 wild type was performed, as control mice were injected with PBS (mock-infected). The mice were sacrificed at 1 , 3, 8, and 14 hours after bacterial challenge, the mock-infected mice after 3 hours. After sacrifice, to remove unattached bacteria in the blood stream, perfusion was performed by injecting sterile PBS in the right ventricle via the vena cava until the blood was completely removed. Brains, lungs, and spleens were collected and stored with Shandon Cryomatrix (Thermo Scientific, Runcorn, ENG) at -80°C.

Cell lines and culture conditions

Human Brain Microvascular Endothelial Cells (HBMEC) [23] (a kind gift from Dr. KS. Kim) were cultivated in RPMI-1640 (Biochrom, Berlin, GER) supplemented with 10% Fetal Calf Serum (PCS) (Biochrom), 10% Nu-serum (BO Biosciences, Breda, NL), 2mM L-glutamine (Gibco, Grand Island, United States), lmM sodium pyruvate (Gibco ), 1 % MEM-vitamines (Gibco) and 1 % non-essential amino acids (Gibco ) . HBMEC were split in T25 flasks (TPP, Trasadingen, Switzerland) after reaching confluency and cultured till passage 36.

Pneumococcal adherence to endothelial cells

HBMEC were grown on glass disks (Thermo Scientific, Braunschweig,

31

Chapter 2

Germany) placed inside wells of 12 well-plates (TPP, Trasadingen, Switzerland).

Cells were grown to confluency at 37°C with 5% CO2

• After washing the cells with

sterile PBS, 900 µl cell culture medium was added to each well and 100 µl containing

approximately 106 CFU of S. pneumoniae TIGR4 was added. After 1 hour incubation at

37°C /5% CO2 cells were washed 3 times with PBS and fixed with paraformaldehyde

(Sigma Aldrich) 4% solution in PBS.

Antibodies, lectin and isotype controls

The following antibody combinations and dilutions were used for

immunofluorescent detection, all dilutions were made in sterile PBS with 5% Fetal

Calf Serum (FCS) (Biochrom, Berlin, GER). To detect S. pneumoniae, an anti-capsule

serotype 4 antibody (Statens Serum Institute, Copenhagen, DEN) 1 :200 diluted was

used, followed by Alexa Fluor 488 goat anti-rabbit (Invitrogen Life Technologies,

Carlsbad, CA) 1 :500 diluted. As control for the bacterial detection, brain sections from

mock-treated mice were incubated with anti-capsule serotype 4 antibody, followed

by Alexa Fluor 488 goat anti-rabbit, and no pneumococci were detected (Figure Sl ).

For the detection of endothelial cells, DyLight 594 labeled LEL (Vector Laboratories,

Burlingame, CA) was used in a 1 :200 dilution. LEL binds well to glycophorin

and Tamm-Horsfall glycoprotein and has been used effectively to label vascular

endothelium in rodents (http://www.vectorlabs.com/catalog.aspx?prodID=1715).

For the detection of VE-Cadherin in mouse tissue, a rat anti-mouse VE-Cadherin

antibody 1 :50 diluted (kind gift from Dr. E. Dejana, FIRC Institute of Molecular

Oncology, Milan, Italy) was used, while for the detection of VE-Cadherin in HBMEC

a mouse anti-human VE-Cadherin antibody (Cell Signaling, Danvers, United States)

1 :400 diluted was used, respectively followed by incubation with Alexa Fluor 594

goat anti-rat and Alexa Fluor 594 goat anti-mouse both 1 :500 diluted. As leukocyte

marker, CD45 antibody 1 :50 diluted (Dako, Ely, UK) was used followed by Alexa

Fluor 488 goat anti-rat antibody (Invitrogen Life Technologies). For the detection of

astrocytes, an anti-Glial Fibrillary Acidic Protein (GFAP) antibody (Dako) at 1 :400

dilution was used, followed by Alexa Fluor 488 goat anti-rabbit antibody (Invitrogen

Life Technologies). Microglia were detected with goat anti-mouse ionized calcium

binding adapter molecule 1 (Tba-1) (Abeam, Cambridge, United Kingdom) followed

by Alexa Fluor 488 donkey anti-goat antibody (Invitrogen Life Technologies). As

isotype controls for the primary antibodies the following antibodies were used at

the same dilution as those for specific primary antibodies: an anti avi-Tag antibody

(GenScript, Piscataway, NJ; rabbit IgG isotype control), anti-human CD4 antibody

(AbD Serotec, Martinsried, Germany; rat IgG isotype control), mouse IgG (Innovative

Research, Plymouth, United States), goat IgG (Santa Cruz Biotechnology, Heidelberg,

Germany). The avi-Tag antibody was used in combination with Alexa Fluor 488 goat

32

Chapter 2

anti-rabbit antibody, the anti-CD4 antibody was combined with Alexa Fluor 488 goat

anti-rat and the mouse IgG was combined with Alexa Fluor 594 goat anti-mouse.

No fluorescence was detected in these controls (data not shown). Incubation of the

tissue sections with only the secondary antibodies did not result in a fluorescence

signal either. For the immunohistochemical detection of CD31, rat anti-mouse CD31

antibody (Pharmigen BO Biosciences, San Jose, CA) at a 1 :50 dilution was used,

conjugated with secondary antibody rabbit anti-rat IgG 1 :40 diluted. As isotype

control a rat anti-human CD4 antibody at the same dilution of the rat anti-mouse

CD31 antibody was used.

lmmunofiuorescent detection

Brain sections of 5µm thin were cut with a cryostat, placed on microscope glass

slides (3 sections/slide) (Starfrost, Dallas, TX), and dried under a fan for at least 30

minutes for a better attachment of the section on the glass slide. Sections were fixed

with acetone for 10 minutes, dried and next incubated with primary antibody for 60

minutes. Slides were washed in PBS 3 times for 5 minutes. Subsequently, each section

was incubated with a mixture of the appropriate secondary antibodies (see Antibodies,

lectin and isohJpe controls) and DyLight 594 labeled LEL for 45 minutes in the dark.

After washing with PBS 3 times for 5 minutes, the slides were incubated in the dark

with DAPI (Roche, Mannheim, Germany) 1 :5000 diluted. After washing again twice

for 5 minutes in PBS, Citifluor solution (Science Services, Munich, Germany) was

added to each section after which the coverslip was applied.

For GFAP and Iba-1 staining, brain sections of 30µm thickness were fixed with

4% paraformaldehyde in PBS for 20 minutes. The slides were washed in PBS 3 times

for 5 minutes and preincubated using PBS with 0.3% Triton X-100 (Sigma Aldrich)

and 2% BSA (Sigma Aldrich) for 30 minutes. Each tissue section was incubated

overnight at 4°C with primary antibody diluted in PBS with 0.3% Triton X-100 and

1 % BSA. The slides were washed three times with PBS for 5 minutes. Each tissue

section was incubated with the secondary antibody diluted in PBS with 0.3% Triton

X-100 and 1% BSA.

For immunohistochemical staining of CD31, brain sections were cryostat-cut at

5µm, mounted onto glass slides (Starfrost) and fixed with acetone for 10 minutes. After

drying, sections were incubated for 45 minutes at room temperature with primary rat

anti-mouse CD31 antibody in the presence of 5% FCS. After washing, endogenous

peroxidase was blocked by incubation with 0.1% H207 in PBS for 20 minutes. This

was followed by incubation for 30 minutes at room temperature with horseradish

peroxidase (HRP) conjugated secondary antibody rabbit-anti rat IgG (Dako) 1 :40

diluted. Conjugate was diluted 1 :50 in PBS supplemented with 2% normal mouse

serum. Between incubation with antibodies, sections were washed extensively with

33

Chapter 2

PBS. Peroxidase activity was detected with 3-amino-9-ethylcarbazole (AEC) (Sigma Aldrich, St. Louis, USA) and sections were counterstained with Mayer's hematoxylin (Klinipath, Duiven, NED).

For the VE-Cadherin staining, HBMEC and HUVEC were incubated with a mixture of anti-VE-Cadherin antibody and anti-capsule serotype 4 antibody after fixation. Afterwards, cells were washed 2 times with PBS and incubated with a mixture of proper secondary antibodies. Cells were washed again 2 times with PBS. Citifluor solution (Science Services, Munich, Germany) was added to each tissue section/glass disk after which the coverslip was applied. The slides were analyzed with a Leica DM5500B microscope, using "Fluorescence" or "Contrast Phase" mode depending on immunofluorescent or immunohistochemical staining. Images were taken with a Leica DFC 360 FX camera. Additionally a Leica SP2 AOBS confocal microscope was used to confirm the presence of adhered pneumococci to the brain vascular endothelium.

Image processing

All the images obtained with fluorescence microscope Leica DM5500B were processed by Image J (http://rsb.info.nih.gov/ij/) [24]. The TIFF images obtained with the 350 nm (blue), 488 nm (green) and 594 nm (red) wavelength filters were merged using the Color-Merge Channels function. As minor manipulation, background correction was applied in those images where the background was relatively or even across the image by using the Brightness & Contrast command of Image J. In all the images subjected to this minor adjustment, the background subtraction was applied to all parts of the image. The images of the immunohistochemical staining of CD31 were taken in grayscale by the Leica QWin Standard software. The LEI z-stacks obtained with the confocal microscope Leica SP2 AOBS were merged through Imaris (Bitplane Scientific Software). No digital adjustment was performed on these images. Imaris was used also to generate the XZ and YZ plane orthogonal views.

RN A isolation

Frozen murine brains were mixed with 0.5ml TRlzol (Invitrogen Life Technologies) under liquid nitrogen and then mechanically disrupted in a Mikro-Dismembrator (Sartorius, Gottingen, Germany) for 2 minutes at 2600 rpm. After the addition of 0.5 ml TRlzol and incubation for 10 minutes at room temperature, total RNA was isolated by acid-phenol extraction and precipitated with isopropanol, washed with 70% ethanol and dissolved in 100 µl of RNase-free water. Subsequently, RNA was DNase-treated using the RNase-Free DNase Set (Qiagen, Venlo, The Netherlands) and purified using the RNA Clean-Up and Concentration Micro Kit (Norgen, Thorold, Canada). The RNA concentration was measured using a NanoDrop spectrophotometer and the RNA

34

Chapter 2

quality was assessed with an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, United States), and consistently found to be intact.

Quantitative Reverse Transcriptase PCR

One µg of total RNA was transcribed into first strand cDNA using M-MLV reverse transcriptase (Invitrogen) and (dT) 12-18 (Invitrogen) oligo's. Transcript levels were determined on the !cycler (Bio-Rad, Hercules, CA) using SYBR green supermix (Bio-Rad). Calculations were done using the comparative Ct method according to User Bulletin 2 (Applied Biosystems, Foster City, CA). Each reaction was performed in triplicate. Primers used in this study were synthesized by Biolegio (Nijmegen, Netherlands) and the sequences were as follows: GAPDH-for CATCAAGAAGGTGGTGAAGC, GAPDH-rev ACCACCCTGTTGCTGTAG, HPRTl-for GACTTGCTCGAGATGTCA, HPRTl-rev TGTAATCCAGCAGGTCAG, TNF alpha-for TCTTCTGTCTACTGAACTTCGG, TNFalpha-rev AAGATGATCTGAGTGTGAGGG, IL-1 beta-for GGCAGGCAGTATCACTCATT, IL-1 beta-rev AAGGTGCTCATGTCCTCAT, IL-6-for CCTCTCTGCAAGAGACTTCCATCCA, IL-6-rev GGCCGTGGTTGTCACCAGCA.

Bacterial quantification

The surface covered by bacteria was measured using the Threshold function of Image J, by determining the area occupied by the 488 nm (bacteria) signal. With the same function the total surface of the tissue for each image was also calculated using the 350 nm (nuclei) signal. The bacteria to tissue ratio was calculated by dividing the surface of the 488 nm signal by the total area of the tissue detected in each image. For each time point of infection, tissue sections from 3 mice were analyzed; for each mouse, 4 images of each anatomical site (subarachnoid space, cerebral cortex, septum, choroid plexus) were taken using the 350, 488 and 594 nm excitation lines. The averages of each mouse were calculated for the final quantification and statistical analysis. Quantification of the amount of bacterial signal in the brain was performed at the lowest magnification (SOX) images in the same manner. For each time point of infection, 6 tissue sections from 3 mice were analyzed. The averages of each mouse were calculated and this value for 3 mice per group used for the statistical analysis.

Scoring for co-localization

The overlap between the S. pneumoniae signal (488 nm green signal in SOX total magnification images) and the vascular endothelium signal (594 nm red signal in SOX total magnification images) was generated using Image J Co-localization threshold. In this analysis, bacteria that co-localize with the vascular endothelium appear white, while bacteria not associated with the vasculature remain green.

35

Chapter 2

Statistical analysis

The independent student t-test of SPSS Statistics 20 (IBM) was used for all

statistical analyses. The increase in mRNA levels of inflammatory cytokines between

14 hours post infection and mock was analyzed using student t-test. For the cerebral

cortex and septum, the differences between 1 hour and each of the other time points

(3, 8 and 14 hours) were statistically analyzed, and for the choroid plexus only

between 8 and 14 hours time points. The differences in the amounts of bacteria in

the SOX images between the 1 hour time point and each of the other time points (3,

8 and 14 hours) were tested in the same manner as described above. The 3 averages

and standard deviations corresponding to the 3 mice of each time point, including

the standard deviations, were used for the analysis.

Results

Progression of pneumococcal infection: from bacteremia to a disease stage leading to meningitis

Mice were infected intravenously with 107 CFU of S. pneumoniae and sacrificed

at fixed time points. No serious symptoms of disease were present at 1 and 3 hours,

while the CFU in the blood was 106 CFU/ml after 1 hour. As disease progressed,

recoverable CFU in the blood increased to 107 CFU/ml at 3 hours and to 108 CFU/ml

after 8 hours, and did not further increase at 14 hours after infection. At eight hours

and even more at fourteen hours post-infection, all mice showed clear evidence of

severe pneumococcal disease, which could progress towards meningitis based on

previous experience [25]. Thus, our model allowed for the uniform disease progression

necessary for our spatiotemporal studies. Of note, the development of the high-grade

bacteremia is required for pneumococcal translocation from the bloodstream into the

CNS of mice [26] .

Overview of the endothelium of the brain vasculature in the absence of infection

The brain receives blood from two main sources, the internal carotid arteries

and the vertebral arteries, which branch and narrow into the arterioles, and then

branch further still into the capillaries in the inner part of the brain [27]. To visualize

the brain vasculature, we used an anti CD31 antibody [28], and DyLight 594 labeled

Lycopersicon Esculentum Lectin (LEL) [29, 30]. Double staining with an anti CD31

antibody and the tomato lectin showed that the latter provides a more comprehensive

detection of the brain vasculature (Figure 1) . Brain tissue from mock-infected mice

was next used to study the vascular structures of the blood-brain barrier under

uninfected conditions, as represented in Figure 1. The fluorescently labeled lectin

thus enabled us to comprehensively detect the vasculature in the different mouse

brain compartments and was used throughout this study.

36

Chapter 2

Cerebral cortex Choroid plOX111

Figure 1. Vasculature in the different compartments of the mouse brain Immunofluorescent detection of CD31 (488 nm green signal}, tomato lectin (594 nm red signal) and nuclei (350 nm blue signal) in the brain of mock-infected mice; total magnification 630X. Double staining with anti-CD31 antibodies and the lectin showed that the tomato lectin provides a more comprehensive detection of the brain vasculature.

Spatiotemporal distribution of S. pneumoniae in the brain

To measure the amount of bacteria remaining in the brain after perfusion at all

time points, we applied a semi-quantitative method using the immunofluorescent

signal of the specific secondary antibodies used, as described in Material and

Methods. The amount of pneumococci in the brain steadily increased from 1

hour up to 14 hours after infection and differences between each time point were

statistically significant (Figure 2AB). Next, we determined the amount of bacteria in

the various areas of the brain using the same semi-quantification approach. As soon

as 1 hour post-infection, bacteria were found in the subarachnoid vessels and, over

the time course of infection, the amount of bacteria detected in this site of the brain

remained stable despite the 100-fold increase in bacterial titers (Figures 2C). In fact,

no statistically significant differences were observed between time points in this area

(Figure 2C). Bacteria were also found in the cerebral cortex and septum as soon as 1

hour after infection, and increased constantly over time in these two compartments

(Figure 2C); already between 1 and 3 hours the difference was statistically significant

(p value = 0.01) (Figure 2C). Remarkably, only 8 hours after infection bacteria could

be detected in the choroid plexus, after which also an increase in bacterial load was

observed (Figures 2A and 2C) with a statistically significant difference in the bacterial

ratio between 8 and 14 hours (p value = 0.002).

37

Chapter 2

c.

s- ,--o,c11r•"" C•<1,9r-1 1r,<r.

· -- '

' - · 1 ·

Stdf•ac:h••ld•••• O<D

y-----�-----

-.- . ·-· . . . . ----=-�-�-=----'-1-•-·-·-·-·-·-:---·-·-·�- . .

; ll.OI t-----------

a • '---=;: --=:�"=

- 1--=;:

'"""'

:

,,..."=

. 1--=;:

'"""'

:�:�I;:

=--=-

: =:-i 1 h- JM.n l h•un U Mll'11

i � :

- I . u� ·•

f..,J•ll>•llmn

:: c • ..,I�"'"'" .. .. -.-�-:-.., . --;a;;;--.. . .... -- ---i---a-..!....--... -.-.-i-i+·�---:: J._.!-1-·--'·"-------... >-•---------

..

- " " I - N "

,

- N " I - N " . . . . . . . . . . . . t Mllf l h•1t'11 lh-• H h•..,.

o..a1•p1-� -i=::�r==:::====ii=::= j � J1_·_1__ ;.L� �: � C:C oo., =-=-�---- ...,

I I I

I I FS=a

•--•---·=--· . . OOl .�- -· -------

- N "I

- N "I

- N "I

- N " . . . . . . . . . . . . 1 h,w > .. , .. ,. ,..._, H Mu1,

--· ·-· - " "J- " "

1- " "

I- "' " . . . . . . . . . . . .

1tl•• Jaa-1, l h••• U ha1n

Figure 2. S p a t i o t e m p o r a l distribution of S. pneumoniae in the brain (A) Quantification of the amount of pneumococci in the brain over the time course of infection using the fluorescent signal of the secondary antibody with which the bacteria were detected. For each time point, the average signal in 9 brain sections per mouse was calculated and with the 3 averages from 3 mice, the overall average was calculated, represented as a bold bar. In the graph, each black dot is the average value of 1 slide/mouse and each number (#1, 2, 3) represents one mouse . The ratio of bacteria in mock tissue was 0. * indicates p < 0.05, ** indicates p < O.Dl. (B)

Brain slides of mice challenged with S. pneumoniae were stained for vasculature using tomato lectin (red) and bacteria (green) as described in Materials and Methods. Total magnification SOX. These images are representative of the situation in i) each brain compartment during all the time points of infection, and ii) each mouse that was analyzed. (C) Quantification of pneumococci using the fluorescent signal of the bacteria measured in the brain over the time course of infection. For each time point, the average signal in 4 brain sections per mouse was calculated and with the 3 averages from 3 mice, the overall average was calculated, represented as a black line. In the graphs, each gray dot is the value of 1 section/mouse and each number (#1, 2, 3) represents one mouse. The ratio of bacteria in mock tissue was 0. * indicates p < 0.05, ** indicates p < 0.01.

S. pneumoniae adheres to the brain endothelium

Using the Co-localization Threshold function of hnageJ [24 ], we next determined the fraction of pneumococci, which co-localized with the vascular endothelium (Figure 3A). At the earliest time points most pneumococci were associated with the blood-brain barrier endothelium. However, at the later stages of infection (8 and 14 hours time points) the amount of pneumococci whose signal did not co-localize

38

Chapter 2

with the brain vasculature increased (Figure 3A). Most bacteria, which were not colocalized with the blood vessels, seemed to be present in the brain tissue instead of in the lumen of the vasculature (Figure S2) .

To gain further insight in pneumococcal interaction with the endothelium, detailed immune fluorescence analysis was performed on the brain slides. Pneumococci were associated with subarachnoid vessels at all time points (Figures 2B, 3A and 3B). Interestingly, in this area at the latest stage of infection (14 hours) the bacteria seemed to form clusters associated with the endothelium (Figure 3B). There were no quantitative differences between the various time points in the subarachnoid space (Figure 2C). In the cerebral cortex, only a few pneumococci were associated with the microvasculature at 1 and 3 hours post infection (Figure 3B). At the later stages of the infection, groups of S. pneumoniae were detected in close proximity to endothelium of the blood vessels in this area of the brain (Figure 3B). In the septum, a few bacteria were present at 1 hour time point post infection and up to 8 hours there were little changes in the amounts of bacteria which co-localized with the vasculature (Figures 2C and 3B). The main difference was observed at the latest stage of infection at 14 hours, when large numbers of bacteria were observed on the endothelium of the blood vessels of this area (Figures 2C and 3B). In the choroid plexus a completely different picture emerged, in fact no S. pneumoniae was detected in this area of the brain during the early stages of pneumococcal infection. Only at 8 and 14 hours after infection bacteria were associated with the blood vessels (Figures 2B, 3A and 3B). To confirm that S. pneumoniae adhered to, or at least was in close proximity of the endothelium of the brain vessels, confocal microscopy was used. The same brain tissue slides used for the immunofluorescence analysis were studied. At all time points post-infection, brain slides were analyzed and three-dimensional reconstruction of pneumococci interacting with the brain vasculature endothelium was obtained by assembling a series of thin slices (z-stacks) taken along the vertical axis. These imaging results showed that the staining of the bacteria and the LEL signal were completely co-localized demonstrating that S. pneumoniae is indeed adhered tightly to the brain endothelium (Figure 4 and S2; movies in the Supplementary Material). In general our tissue section analysis agreed with our results from the semi-quantitative analysis and indicated that the subarachnoid space is at first the preferential site of attachment.

39

Chapter 2

I I, Ill , •. , , 1,.,U•1t l ..

B.

s••achnel•••••

. .,_

Figure 3. Spatiotemporal distribution of S. pneumoniae in the different compartments of the brain (A} Visualization of S. pneumoniae co-localized with the vasculature in the brain during the entire course of infection, same images as shown in Figure 2B; total magnification SOX. Pneumococci that co-localized with the vascular endothelium appear as white, pneumococci not co-localized appear as green. (B) Brain slides of mice challenged with S. pneumoniae were stained for vasculature using tomato lectin (red}, bacteria (green}, and nuclei (blue) as described in Materials and Methods. Total magnification of subarachnoid space, septum, choroid plexus 630X; total magnification of cerebral cortex lOOOX. At 14 hours post infection the white arrow indicates the pneumococci forming clusters in the subarachnoid vessels. For each time point of infection, brains from 3 mice were analyzed, and of each mouse 4 brain sections were used for the immunofluorescent detection. These images are representative of the situation in i) each brain compartment during all the time course of infection and ii) each mouse that was analyzed.

40

Chapter 2

Subo,ochnold space Cerebral cortex Septum Choroid plexus

Figure 4. S. pneumoniae adheres to the brain vascular endothelium Visualization of S. pneumoniae (green) adhering to the brain vascular endothelium (red) using confocal microscopy. The scale of each image is shown by the white scale bar. White arrows points to completely overlapping staining of the bacteria and the lectin resulting in a yellow color, strongly suggesting a co-localization of S. pneumoniae with endothelial cells. For each time point of infection, brains from 3 mice were analyzed, and of each mouse 3 brain sections were used for the confocal detection. The images at the later stages of infection were used, because more bacteria were detected at those time points and thus these provide a clear picture of the bacteria anchored to the vasculature. Images of all time points after infection are shown in Figure S2.

Interaction of S. pneumoniae with the blood-brain barrier does not cause major disruption of the intercellular junctions of brain endothelial cells

The integrity of endothelium is maintained by the intercellular junctions between the endothelial cells. VE-Cadherin is an adhesion molecule expressed by the vascular endothelium and it is mainly present on the cell-cell junctions [31, 32] . For this reason this molecule was used as indicator for the loss of integrity of the endothelium [13]. VE-Cadherin staining in HBMEC after a 1-hour incubation with 106 CFU of S. pneumoniae TIGR4 did not differ from that observed in HBMEC under normal conditions. VE-Cadherin expression was homogeneous in the intercellular junctions also in proximity of adhered bacteria, indicating no major disruptions of endothelial integrity (Figures SA and SB). Especially the :XZ and YZ planes show that the VE-Cadherin signal is continuous without interruptions after incubation with pneumococci (Figures SC and SD). Furthermore, the VE-Cadherin staining pattern in the subarachnoid space and choroid plexus in brain tissue of infected mice was similar to that in the brains of mock treated mice, without major signs of endothelial disruption (Figure S3).

41

Chapter 2

HBMEC norm-11 condUl011s

0

----- -- ---- - - - - - -�

HBMEC olte1 1 hou, Incubation with S. pnemnonb• A

. . ": � " : . . . - •..., .. ) '\.. ... "� . . - . .•. ' ... • ,. ' •. "': --�·

_: I r ' �� : - �• •• J . ., . B

. . - ..

�

0 -�------ - ----------· ' - ---_- ---- -_ --...

Figure 5. Interaction of S. pneumoniae with endothelial cells does not cause serious

disruption of the intercellular junctions in vitro Immunofluorescent detection of VE-Cadherin (red) and S. pneumoniae (green) in H BMEC in normal conditions and after 1 hour incubation with 106 CFU pneumococci. Cells were stained as described in Material and Methods. Panel A shows a fluorescence microscopic image 630X total magnification; panel B shows a confocal microscopy image; panels C and D show the YZ and XZ orthogonal view planes. Leukocyte recruitment during pneumococcal infection

Inflammation is a crucial process in the defense mechanism against various infectious diseases, and leukocytes are the principal cellular mediators of inflammation. To evaluate the degree of inflammation in the brain and compare it to other tissue compartments, such as the lungs, during the course of S. pneumoniae infection we studied tissue influx of leukocytes using the leukocyte common antigen marker CD45 [33, 34, 35] . As expected, in brain and lungs of mock treated mice no leukocytes were detected (Figure S4). The CD45 staining showed almost no influx of leukocytes in any compartment of the brain at the early stages of infection (Figure S4), and even at the later stages, the presence of leukocytes was minimal (Figure 6A). In contrast, in the lungs a different picture was observed. At all time points of pneumococcal infection, large numbers of leukocytes were present in the lungs, even as soon as after 1 hour (Figure 6B). These results reveal that intravenous injection of S. pneumoniae causes distinct tissue-specific responses. Furthermore, in contrast to the presence of large numbers of leukocytes in the lungs, no such infiltration was observed in the brain during the time course of our pneumococcal infection. Presumably, this would facilitate the replication of bacteria that are able to successfully translocate to the CNS.

42

A.

U homs post infection

B,

Lungs

Chapter 2

Submachnold spoce ood cetebral conex Choroid plexus

Figure 6. Leukocyte presence in the brain and lungs of mock-treated and S. pneumoniae-infected mice at different time points after infection (A) Immunofluorescent staining of the leukocyte common antigen CD45 (green) and nuclei (blue) in the subarachnoid space/cerebral cortex, septum and choroid plexus at 14 hours-post infection; total magnification 400X. The CD45 staining showed no influx of leukocytes in all the compartments of the brain in the early stages of infection (data not shown), and even 14 hours after infection the presence of leukocytes was minimal. For each time point of infection, brain sections from 3 mice were analyzed, and of each mouse 3 brain sections were used for the imrnunofluorescent detection. The images are representative of the situation observed in each mouse that was analyzed.(B) Imrnunofluorescent staining of leukocyte common antigen CD45 (green) and nuclei (blue) in lungs of mock-treated and S. pneurnoniae-infected mice; total magnification 630X. For each time point of infection, lungs from 3 mice were analyzed, and of each mouse 3 lung sections were used for the imrnunofluorescent detection. The images are representative of the situation observed in each mouse that was analyzed.

The local immune system in the brain is activated early on

To evaluate the local status of inflammation in the CNS during pneumococcal infection, Iba-1 [36] and GFAP [37, 38] staining was performed on mouse brain sections representative of various areas in the brain. In the mock-treated tissue, microglia and astrocytes had a normal cellular morphology (Figures 7, SSA and SSB). In the infected animals activated microglia and astrocytes were detected as soon as 1 hour after pneumococcal infection (Figures 7, SSA and SSB). From then on there was a gradual increase in the amount of astrogliosis and activated microglia especially at the latest time points at 8 and 14 hours, where some microglia and astrocytes displayed an abnormal increase in the dimensions of the soma and the dendrites were almost completely retracted (Figures 7, SSA and SSB). Thus, this analysis revealed that the local immune system in the brain is activated directly upon the presence of pneumococci in the blood.

43

Chapter 2

Mock

Mlcroglla

Asuocytes

1 hou,

' . ,,.� I • ' ' 1 • -

' I r , I , �

-. I I , , ,,

3 hom,

< •

'4 •

' -

-. . . -.

-�J

. - . ' - . /�� . · 1

. - /· , �,-,. ---� ' -�

8 hours

. r_ I I ,

• '

( ' '

-·, · . • \ . -. 1 ' -'

14 hours

' . l,, ·, ' � - - -!.. j - l ' l _J ,.

, '

�_-' _· I ,

. '

r

) �

Figure 7. Activation of the local immune system in the brain upon pneumococcal infection Immunofluorescent staining of Iba-1 as marker for microglia (A) and GFAP as marker for astrocytes (B) in brain of a mock-treated mouse and during all the time points of pneumococcal infection. Total magnification 630X. The white rectangles delineate the activated microglia and astrocytes. Brains from 3 mice for each time point were analyzed, and for each mouse 3 brain sections were used for the confocal imaging analysis. Each time point is representative for the situation observed in each mouse that was analyzed.

Inflammatory cytokine profile reveals that the infection is progressing to meningitis

To confirm that inflammation was indeed occurring in the brain, RNA from brains of the mock-treated and infected mice of the 14 hours time point was used to measure changes in inflammatory cytokine production. We selected IL-1 beta, IL-6 and TNF alpha for the quantitative reverse transcriptase PCR, since they are known to be the major inflammatory cytokines detected in the CSF during bacterial meningitis [39]. As controls, the house-keeping genes GAPDH and HPRT1 were used. We observed at 14 hours post infection a statistically significant 40-fold increase in levels of IL-6 and TNF alpha, and a 4-fold increase in IL-1 beta compared to the mock-treated mice (Figure 8). The observed Ct values are reported in Table S1 of the Supporting Information.

44

Chapter 2

l 000

* l 60(1

* l M•l•

1 .l(,(,

1 200

l.O(l,)

0.80(1

I 0. ih''I,)

o .. 1t1,:,

(l ioo

o.oo,:, I I - - i A B A B A e A B A B

GAPDH HPRT1 ll-6 TNFalpho ll 1heto

A= mock B= 14 hours 1nfecflon

Figure 8. Inflammatory cytokine levels in the brain during pneumococcal infection mRNA levels (AU= arbitrary units) of the pro-inflammatory cytokines TNF alpha, lL-1 beta, IL-6 and the housekeeping genes G APDH and HTR1 in the brains of mock-treated mice and mice sacrificed at 1 4 hours after infection, as quantified by Real Time PCR. * indicates p < 0.05 .

Discussion

Translocation of bacteria from the blood stream across the blood-brain barrier is hypothesized to be the principal mechanism by which the pneumococcus and other meningeal pathogens invade the CNS. A crucial moment in this translocation mechanism is the adhesion of the bacteria to the vasculature endothelium composing the blood-brain barrier. It is still unclear how and where this process takes place in vivo. The purpose of this study was to dissect the manner by which S. pneumoniae in the blood stream adheres to

the endothelium of the blood-brain barrier during the events leading up to meningitis and not meningitis itself. For this reason the mice were intravenously challenged with S. pneumoniae and, after the sacrifice at different time points after infection, they were perfused in order to remove all bacteria not attached to the blood vessels. A second goal was to determine whether S. pneumoniac in the bloodstream elicits an inflammatory response from the brain before the development of meningitis.

There were several indications that pneumococcal infection in the mice was progressing towards meningitis. There was a significant, steady increase in the amount of bacteria in the brain from the first stage of infection up to 14 hours post challenge, which did not have a linear relation with the CFU's in the blood. Our scoring method to determine which bacteria co-localized with the brain vasculature showed that at the start of infection most pneumococci were associated with the

45

Chapter 2

vasculature. At the later stages the amount of pneumococci not co-localized with

blood vessels increased and these bacteria seemed to be present in the brain tissue.

We think that these signals represent bacteria that have already crossed the blood

brain barrier. This idea was strengthened by the confocal microscopy analysis, which

showed that at 14 hours post infection, in the subarachnoid space, septum and

choroid plexus, some pneumococci did not co-localize with the brain vasculature

and seemed already in the brain (Figure S3). Additionally, at 14 hours the expression

of pro-inflammatory cytokines was increased significantly. On the other hand, the

absence of leukocyte influx, even at 14 hours, indicates that we studied the processes

leading to meningitis, but not the pathogenesis of the actual disease.

The fact that we did not observe bacteria in the choroid plexus until after 8

hours of infection in combination with the finding that the number of bacteria

attached to subarachnoid vessels remained constant despite an increase in bacterial

load in the blood confirmed that we did indeed remove all non-attached bacteria.

The immune fluorescent analysis combined with the confocal images clearly showed

that the pneumococci are indeed adhering to the brain vascular endothelium. It

was hypothesized that within the subarachnoid space, bacteria grow and release

proinflammatory compounds [40]. In this study we observed that at 1 hour post

infection a higher number of bacteria was observed in the subarachnoid vessels than

in the other anatomical sites, implying that S. pneumoniae adheres preferentially to

blood vessels at this site. Over time conditions seemed to change, which allowed

bacterial attachment at blood vessels in distinct anatomical sites in the inner brain,

such as the cerebral cortex, septum and choroid plexus. What factors are responsible

for this preferential adhesion are currently unknown, but may include increased

basal levels of PAF or laminin receptors. Levels of these NFkB-regulated ligands may

also change during infection, explaining the increased adhesion to other sites at later

time points. Future studies are warranted to determine if this is indeed the case. No

increase of bacteria in the subarachnoid vessels was observed over time, which might

indicate that there is no change in receptor expression at this location.

In the literature it is postulated that the choroid plexus, where the cerebrospinal

fluid is produced, is the location where bacterial CNS invasion is most likely to occur

[18]. Along such lines, it was shown that S. suis is able to pass through the blood

brain barrier at this site [19, 20]. In contrast, in our study the adherent pneumococci

appeared in this area only at the later stages of infection, when levels of bacteremia

were high, indicating that it is most likely not the site of initial CNS invasion. To our

knowledge, this is the first study to specifically examine the spatiotemporal events

of CNS invasion in the brain. Unfortunately, the level of CPUs in blood in humans

that are developing pneumococcal meningitis is unknown, which makes it hard

to extrapolate this data from mice with acute bacteremia to the human situation.

46

Chapter 2

Although events in humans with a more prolonged infection might have a distinct pathogenesis, it seems l ikely that the interaction of the pneumococci with the endothelium is similar.

In meningococcal meningitis N. 111cningitidis is able to form microcolonies during the passage through the blood-brain barrier [41 ], and we also observed S. pneumoniae clusters in our study. In a neonatal rat meningitis mode, breakdown of the blood-brain barrier was observed [42] and pneumolysin was thought to be the main inducer of damage to brain microvascular endothelial cells [43]. In patients with meningitis caused by N. meningitidis, disruption of the blood-brain barrier has been shown to be a crucial event for the development of the disease [44]. In fact, during meningococcal meningitis metalloproteinases of the extracellular matrix contribute to blood-brain barrier disruption [45] . Using VE-Cadherin staining, Coureuil ct al. showed that N. meningitidis recruits the polarity complex, which also plays a role in formation of tight junctions. This leads to a depletion of tight junction proteins at other places in the cell, which leads to a disruption and opening of the intercellular junctions of brain endothelial cells. After incubation of brain endothelial cells with N. meningitidis, openings and breaks in the intercellular junctions between endothelial cells were observed using amongst others immune fluorescent staining of VEcadherin [13]. In our study, VE-Cadherin staining in HBMEC showed no disruption after 1 hour incubation with S. pnewnoniae TIGR4. Furthermore, we observed a continuous and homogeneous staining of VE-Cadherin in the brains of infected mice. These findings indicate that blood-borne S. pneumoniae might not cause endothelium disruption during the translocation across the blood-brain barrier and support the idea of pneumococcal translocation through a pericellular or transcytosis routes [46). Additionally, these data might indicate that toxicity of pneumolysin towards endothelial cells is limited under these conditions.

Both in vitro and in vivo, S. pneumoniae has been shown to lose its polysaccharide capsule during invasion into the host cell [47]. In our study, we primarily used antibodies that recognized the capsular polysaccharide of the bacteria. To investigate this issue in more detail, we also detected the pneumococci in the brain using a total anti-pneumococcal antiserum, recognizing capsulated and unencapsulated pneumococci, generated in a similar way as the antiserum described in Elm C. et

al., 2004 [48]. Brains from 3 mice were analyzed for each time point and from each brain one brain section was used for the immunofluorescent detection with this serum. Interestingly, a preliminary quantification analysis indicated no differences in the amount of bacteria associated with the brain vascular endothelium compared to the data obtained with the anti-capsule antibody (data not shown). Therefore, S. pneumoniae might maintain i ts capsule during translocation over the blood-brain barrier or translocation events via an intracellular route are rare.

47

Chapter 2

The inflammatory features and clinical complications following bacterial meningitis normally observed in humans can be reproduced using mouse models. Most in vivo models that examine the pathophysiology of bacterial meningitis consist of the direct injection of pneumococci into the brain of mice or rats [25, 49, 50] . Administration of bacteria directly into the brain bypasses the need for bloodbrain barrier translocation. Our aim was to study the interaction of blood-borne S. pneumoniae with the brain vascular endothelium in vivo preceding meningitis development, hence we used a bacteremia model. We observed clear evidence of activated microglia and astrocytes already 1 hour post challenge, which implies that the local immune system in the brain is activated early on. Pneumococcal bacteremia caused a severe systemic inflammatory response, characterized by the infiltration of leukocytes in the lungs, however very little leukocyte influx in the brain was observed. Low influx of leukocytes could mean that the vascular endothelial cells are not yet activated enough over this time frame to recruit leukocytes to transmigrate over the endothelium and reach the site of infection. Mook-Kanamori et al. showed neutrophil infiltration at 30 hours after infection, when meningitis is already established [25]. The inflammatory cytokine profile revealed that IL-1 beta, IL-6 and TNF alpha expression was increased at 14 hours post-injection compared to the control group. This increase was comparable to the increase observed in other studies in which the meningitis condition of the mice and rats was established by direct injection of bacteria into the cistema magna of the brain [25, 49, 51, 52, 53] . Moreover, this increase in mRNA levels of pro-inflammatory cytokines reflects the cytokine levels measured in the CSF of hospitalized patients with bacterial meningitis that all recovered [4]. This data indicates that during bacteremia the brains displays signs of inflammation and indicates that the mice were likely progressing towards meningitis development.

The brain is able to develop an inflammatory reaction in response to lipopolysaccharides (LPS}, a major component of the outer membrane of Gramnegative bacteria [54]. Previous studies by Orihuela et al. have shown that purified pneumococcal bacterial cell wall is alone sufficient to induce apoptosis of neurons in the dentate gyrus of the hypocampus of infected mice [55] . Thus, we can at present not determine whether the adhesion of the bacteria to the endothelial cells provokes a response from the CNS or whether solely the presence of bacterial molecules in blood is the cause of the early immune activation that we observe.

In conclusion, in this study we have shown that S. pneumoniae preferentially adheres to the subarachnoid vessels, and over time reaches other anatomical sites in the inner brain, such as the cerebral cortex and septum. Only at the later stages of infection S. pneumoniae interacts with the endothelium of the choroid plexus. Furthermore, the local immune system of the brain seems to sense bacterial adhesion to the endothelial cells and is activated immediately, even before meningitis develops. Our findings

48

Chapter 2

suggest that S. pneumoniae preferentially crosses the blood brain barrier as a result of adhesion, invasion and translocation through or between endothel ial cells without causing overt disruption of the vascular endothelium. This implies that S. pneumoniae

might indeed use an intracellular or paracel lular route for translocation of the bloodbrain barrier. One implication of this result is that use of bacterial adhesins as vaccine candidates and/or that adhesion inhibited via a pharmacological route could prevent the development of meningitis.

Acknowledgements Part of this work has been performed at the UMCG Microscopy and Imaging

center (UMIC). In particular we thank Klaas Sjollema for all the technical support during the confocal imaging. We thank Prof. Dr. E. Boddeke, E. M. Wesseling and M. Meijer (Neurophysiology Department, University Medical Center Groningen, Groningen, The Netherlands) for providing us with the lba-1 marker and the antiGFAP antibody and for the help in the interpretation of the astrocyte and microglia staining results. We also thank Dr. M. van der Kooi-Pol (UMCG) for help in the RNA isolation and Dr. H. Gradstedt (UMCG) for providing the total antipneumococcal antiserum.

Supplementary figures Available on http://www.ncbi.nlm.nih.gov/pu bmed/2387 4613

49

Chapter 2

References 1. Brouwer MC, Tunkel AR, van de Beek D (2010) Epidemiology, diagnosis,

and antimicrobial treatment of acute bacterial meningitis. Clin Microbial Rev 23: 467-92.

2. Koedel U, Scheid WM, Pfister HW (2002) Pathogenesis and pathophysiology of pneumococcal meningitis. Lancet Infect Dis 2: 721-36.

3. Woehr! B, Klein M, Grandgirard D, Koedel U, Leib S (201 1 ) Bacterial meningitis: current therapy and possible future treatment options. Expert Rev Anti Infect Tuer 9: 1053-65.

4. Ohga S, Aoki T, Okada K, Akeda H, Fujioka K (1994) Cerebrospinal fluid concentrations of interleukin-1 beta, tumour necrosis factor-alpha, and interferon gamma inbacterial meningitis. Arch Dis Child 70: 123-5.

5. Volterra A and Meldolesi J (2005) Astrocytes, from brain glue to communication elements: the revolu tion continues. Nat Rev Neurosci 6: 626 - 640.]

6. Liu W, Tang Y, Feng J (2011 ) Cross talk between activation of microglia and astrocytes in pathological conditions in the central nervous system. Life Sci 89: 141-6.

7. Kreutzberg GW (1995) Microglia, the first line of defence in brain pathologies. Arzneimittelforschung 45: 357-60.

8. Aloisi F (2001) Immune function of microglia. Glia 36: 165-79.

9. Sofroniew MV, Vinters HV (2010) Astrocytes: biology and pathology. Acta Neuropathol 1 19: 7-35.

10. Sofroniew MV (2009) Molecular dissection of reactive astrogliosis and glial scar formation. Trends Neurosci 32: 638-47.

1 1 . Zhang D, Hu X, Qian L, O,Callaghan JP, Hong JS (2010) Astrogliosis in CNS pathologies: is there a role for microglia? Mo) Neurobiol 41 : 232-41 .

12. Pekny M, Nilsson M (2005) Astrocyte activation and reactive gliosis. Glia 50: 427-34.

13. Coureuil M, Mikaty G, Miller F, Lecuyer H, Bernard C, et al. (2009) Meningococcal type IV pili recruit the polarity complex to cross the brain endothelium. Science 325: 83-7.

14. Marriott HM, Mitchell TJ, Dockrell DH (2008) Pneumolysin: a double-edged sword during the hostpathogen interaction. Curr Mol Med 8: 497-509.

15. Attali, C, Durmort C, Vernet T, Di Guilmi AM (2008) The interaction of Streptococcus pneumoniae with plasmin mediates transmigration across endothelial and epithelial monolayers by intercellular junction cleavage. Infect Imrnun 76: 5350-6.

16. Ring A, Weiser JN, Tuomanen EI ( 1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 102: 347-60.

17. Tarn SJ, Watts RJ (2010) Connecting vascular and nervous system development: angiogenesis and the blood-brain barrier. Annu Rev Neurosci 33: 379-408.

18. Hoffman 0, Weber RJ (2009) Pathophysiology and treatment of bacterial meningitis. Tuer Adv Neurol Disord 2: 1-7.

19. Tenenbaum T, Essmann F, Adam R, Seibt A, Janicke RU, et al. (2006) Cell death, caspase activation, and HMGBl release of porcine choroid plexus epithelial cells during Streptococcus suis infection in vitro. Brain Res 1 100 : 1-12.

20. Tenenbaum T, Papandreou T, Gellrich D, Friedrichs U, Seibt A, et al . (2009). Polar bacterial invasion and translocation of Streptococcus suis across the blood-cerebrospinal fluid barrier in vitro. Cell Microbiol 1 1 : 323-36.

21 . Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2006) beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pneumoniae. Infect Irnmun 73: 7827-35.

22. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, et al. (2009) Lam in in receptor initiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest 1 19: 1 638-46.

50

Chapter 2

23. Stins MF, Badger J, Sik Kim K (2001 ) Bacterial invasion and transcytosis in transfected human brain microvascular endothelial cells. M icrob Pathog 30: 19-28.

24. Schneider CA, Rasband WS, Eliceiri KW (201 2) NTH Image to lmageJ: 25 years of image analysis. Nat Methods 9: 671 -5.

25. Mook-Kanamori B, Celdhoff M, Troost D, van der Pol l T, van de Beek D (201 2) Characterization of a pneumococcal meningitis mouse model. BMC Infect Dis 12: 71 .

26. Orihuela CJ, Cao C, Francis KP, Yu J, Tuomanen El (2004) Tissue-specific contributions of pneumococcal virulence factors to pathogenesis. J Infect Dis 190: 1661-9.

27. Purves D, Augustine CJ, Fi tzpatrick D, Hall WC, LaMantia AS, et al. editors. (2001 ) Neuroscience. 2"" edition. Sunderland (MA): Sinauer Associates. The Blood Supply of the Brain and Spinal Cord.

28. Newman PJ, Newman DK (2004) Signal transduction pathways mediated by PECAM-1 : new roles for an old molecule in platelet and vascu lar cell biology. Arterioscler Thromb Vase Biol 23: 953-64.

29. Hashizume H, Baluk P, Morikawa S, McLean JW, Thurston C, et al. (2000) Openings between defective endothelial cells explain tumor vessel leakiness. Amer J Pathol 1 56: 1363-1380.

30. Debbage PL, Solder E, Seidl S, Hutzler P, Hug] B, et al. (2001 ) lntravital lectin perfusion analysis of vascu lar permeability in human micro- and macro- blood vessels. Histochemistry Cell Biol 1 1 6: 349-359.

31 . Vestweber D (2008) Ve-cadherin: the major endothelial adhesion molecule controlling cellular junctions and blood vessel formation. Arterioscler Thromb Vase Biol 28: 223-32.

32. Corada M, Mariotti M, Thurston C, Smith K, Kunkel R (1 999) Vascular endothelial-cadherin is an important determinant of microvascular integrity in vivo. Proc Natl Acad Sci U S A 96: 981 5-20.

33. Altin JC, Sloan EK (1 997) TI,e role of CD45 and CD45-associated molecules in T cell activation. lmmunol Cell Biol 75: 430-45.

34. Clement LT (1 992) lsoforms of the CD45 common leukocyte antigen family: markers for human T-cell differentiation. J Clin lmmunol 12: 1 -10.

35. Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (201 1 ) Physiology of microglia . Physiol Rev 91 :461 -553

36. Wen YR, Tan PH, ChengJK, Liu YC, Ji RR (201 1 ) Microglia : a promising target for treating neuropathic and postoperative pain, and morphine tolerance. J Formos Med Assoc 1 1 0: 487-94.

37. Bramanti V, Tomassoni D, Avitabile M, Amenta F, Avola R (2010) Biomarkers of glial cell proliferation and differentiation in cul ture. Front Biosci (Schol Ed) 2: 558-70.

38. Schiff L, Hadker N, Weiser S, Rausch C (2012) A literature review of the feasibi lity of glial fibrillary acidic protein as a biomarker for stroke and traumatic brain injury. Mol Diagn TI,er 16: 79-92.

39. Fassbender K, Ries S, Schminke U, Schneider S, Hennerici M (1 996) Inflammatory cytokines in CSF in bacterial meningitis: association with altered blood flowvelocities in basalcerebral arteries. J Neural Neurosurg Psychiatry 61 : 57-61 .

40. Gerber J, Nau R (2010) Mechanisms of injury in bacterial meningitis. Curr Opin Neural 23:31 2-8. 41 . Schwerk C, Papandreou T, Schuhmann D, Nickol L, Borkowski J, et al . (2012 Polar

invasion and translocation of Ncisscria 111c11i11gitidis and Streptococcus suis in a novel human model of the blood-cerebrospinal fluid barrier. Plos One 7: e30069.

42. Barichello T, Fagundes CD, Generoso JS, Paula Moreira A, Costa CS, et al. (2012) Brain-blood barrier breakdown and pro-inflammatory mediators in neonate rats submitted meningitis by Streptococcus p11cu111011iac. Brain Res 1471 : 162-8.

43. Zysk C, Schneider-Wald BK, Hwang JH, Bejo L, Kim KS, et al . (2001 ) Pneumolysin is the main inducer of cytotoxici ty to brain microvascular endothelial cells caused by Strcptococrns p11c11111011 iac. Infect Immun 69: 845-52.

44. Sharief MK, Ciardi M, TI1ompson EJ (1992) Blood-brain barrier damage in patients with bacterial meningitis: association with tumor necrosis factor-alpha but not interleukin-1 beta. J Infect Dis 1 66: 350-8.

45. Paul R, Lorenzi S, Koedel U, Sporer B, Vogel U, et al . (1998). Matrix metelloproteinases contribute to

51

Chapter 2

the blood-brain barrier disruption during bacterial meningitis. Ann Neurol 44: 592-600. 46. Pancholi V, Fontan P, Jin H (2003) Plasminogen-mediated group A streptococcal adherence to

and pericellular invasion of human pharyngeal cells. Microb Pathog 35: 293-303. 47. Hammerschmidt S, Wolff S, Hocke A, Rosseau S, Muller E, Rohde M. (2005) Illustration of

pneumococcal polysaccharide capsule during adherence and invasion of epithelial cells. Infect Immun 73: 4653-67.

48. Elm C, Braathen R, Bergmann S, Frank R, Vaerman JP, et al. (2004) Ectodomains 3 and 4 of human polymeric Immunoglobulin receptor (hpigR) mediate invasion of Streptococcus pneumoniae into the epithelium. J Biol Chem 279: 6296-304.

49. Paul R, Angele B, Sporer B, Pfister HW, Koedel U (2004) Inflammatory response during bacterial meningitis is unchanged in Fas- and Fas ligand-deficient mice. J Neuroimmuno 152: 78-82.

50. Paul R, Winkler F, Bayerlein I, Popp B, Pfister HW, Koedel U (2005) Urokinase-type plasminogen activator receptor regulates leukocyte recruitment during experimental pneumococcal meningitis. J Infect Dis 191: 776-782.

51 . Barrichello T, Fagundes GD, Generoso JS, Paula Moreira A, Costa CS, et al (2012) Brain-blood barrier and proinflammaory mediators in neonate rats submitted meningitis by Streptococcus pneumoniae. Brain Res 1471 : 162-8.

52. Barichello T, dos Santos I, Savi GD, Florentino AF, Silvestre C, et al (2009) Tumor necrosis factor alpha (TNF-alpha) levels in the brain and cerebrospinal fluid after meningitis induced by Streptococcus p11eu111011iae. Neurosci Lett 467: 217-9.

53. 0stergaard C, Brandt C, Konradsen HB, Samuelsson S (2004) Differences in survival, brain damage, and cerebrospinal fluid cytokine kinetics due to meningitis caused by 3 different Streptococcus pncumoniae serotypes: evaluation in humans and in 2 experimental models. J Infect Dis 190: 1212-20.

54. Minogue AM, Barrett JP, Lynch MA (2012) LPS-induced release of IL-6 from glia modulates production of IL-1 f3 in a JAK2-dependent manner. J Neuroinflammation 9: 1 26.

55. Orihuela CJ, Fillon S, Smith-Sielicki SH, El Kasmi KC, Gao G, et al. (2006) Cell wall-mediated neuronal damage in early sepsis. Infect Immun 74: 3783-9.

52

CHAPTER 3

Signalling or binding: the role of the platelet activating factor receptor in invasive pneumococcal disease

Federico Iovino, Matthijs C. Brouwer, Diederik van de Beek, Grietje Molema, Jetta J.E. Bijlsma

Cellular Microbiology (2013) 15: 870-81

53

Abstract

Streptococcus pneumoniae (the pneumococcus) is an opportunistic human pathogen, which causes serious invasive disease, such as pneumonia, bacteremia and meningitis. The interaction of the bacteria with host receptors precedes the development of invasive disease. One host receptor implicated in pneumococcal adhesion to, invasion of and ultimately translocation of cell layers is the platelet activating factor receptor (PAFR). PAFR is a G-protein coupled receptor, which binds PAF, a potent phospholipid activator involved in many leukocyte functions, platelet aggregation and inflammation. PAFR has been proposed to bind S. pneumoniae and as such facilitate adhesion to, uptake by and transcytosis of endothelial cells leading to invasive disease. However, there is a shortage of biochemical data supporting direct interaction between PAFR and the bacteria, in addition to conflicting data on its role in development of invasive pneumococcal disease (IPD). In this review, we will discuss current literature on PAFR and S. pneumoniae and other pathogens, including data concerning human PAFR genetic variation related to IPD clinical aspects, to shed light on the importance of PAFR in invasive pneumococcal disease (IPD). Clarification of the role of this receptor in IPD development has the potential to enable the development of novel therapeutic strategies for treating pneumococcal disease by interfering with the PAFR.

Chapter 3

Introduction

Streptococcus pneumoniae is an opportunistic human pathogen that causes lifethreatening invasive diseases, such as pneumonia, bacteremia and meningitis, resulting in high morbidity and mortality throughout the world. Every year, over a million people succumb to infections by S. pneumoniae worldwide (Bogaert et al., 2004). S. pneumoniae

colonizes the human nasopharynx, which is asymptomatic in most cases. However, especially in infants, the elderly and immune-compromised individuals, the bacterium can switch to an invasive lifestyle, translocate over cell layers and cause disease. As for other pathogens, specific interactions between bacterial proteins and host receptors are thought to play an important role in this switch.

One receptor implicated in adhesion to, invasion of and also transcytosis in endothelial cells is the Platelet Activating Factor Receptor (PAFR) (see Figure lA). Transcytosis is a mechanism by which various macromolecules are transported across the interior of a cell (Fishman et al., 1987) and has been proposed as mechanism for pneumococcal translocation over epithelial and endothelial cell layers (Zhang et al.,

2000; Ring et al., 1998). The ligand of PAFR is the platelet-activating factor (PAP), a phospholipid (1-0-alkyl-2-acetyl-sn-glycero-3-phosphorylcholine) that functions as a mediator in diverse pathologic processes, such as allergy, asthma, septic shock, arterial thrombosis, and inflammation (Shukla, 1992; Honda et al., 2002). PAFR is a G-protein coupled receptor with seven transmembrane domains and binding to its natural ligand, PAF, activates several signaling mechanisms. PAP bound to PAFR activates GTPase, causes phospholipid turnover via the phospholipases C, D, and A2 pathways and also activates protein kinase C and tyrosine kinases (Shukla, 1992). The role of PAFR in inflammation is based on PAP-induced pathological responses and prevention of the pathological conditions by PAFR antagonists (Honda et al., 2002). PAFR has been identified in lungs, brain, platelets and leukocytes, and was found to participate in trafficking of leukocytes and the generation of bronchospasms and inflammation in asthma (Snyder, 1990; Chao and Olson, 1993; Shirasaki et al., 1994; Stoll et al., 1994). Endothelial activation, induced by pro inflammatory cytokines such as interleukin (IL) 1 and tumor necrosis factor (TNF) cc, which allows endothelial cells to participate in the inflammatory response (Pober, 2002), causes an increase in the expression of specific molecules important for leukocyte trafficking, including PAP and the PAFR (Shukla, 1992; Chao and Olson, 1993). The PAFR has been investigated for many years and several inhibitors and antibodies that block its biological activity exist (see Table 1). To what extent these antibodies and chemical components interfere with the signaling cascade of PAFR is currently not known.

Interpretation of the importance of the PAFR in the development of invasive pneumococcal disease is hampered by a scarcity of biochemical characterization of the bacteria-receptor binding, and compounded by conflicting results in the literature.

55

Chapter 3

In this review, we summarize the current literature on S. pneumoniae and PAFR interactions and provide an overview of PAFR involvement in disease caused by other pathogens. A better understanding of the role of PAFR in IPD development will enable the development of new therapeutic strategies.

PAFR-mediated interaction of S. pneumoniae with human cells

Thefirststudytoshow aroleforPAFRintheinteractionofS. pneumoniaewithhuman cells demonstrated thatpre-trea tmentofcellswi th a PAFRantagonistreduced the adhesion and internalization of S. pneumoniae into activated endothelial cells and A549 human lung epithelialcells(Cundelletal., 1995)approximatelybyhalf (seeTable1).Subsequentanalysis of S. pneumoniae infecting the non-activated rat brain capillary endothelial cell line rBCEC6 using confocal microscopy indicated that the bacteria and PAFR are in close physical proximity, as co-localization was observed in 82% of the eukaryotic cell population studied (Radin et al., 2005). PAFR expression was markedly increased in cells infected with bacteria compared to non-infected ones, suggesting that endothelial activation and subsequent induction of receptor expression can be invoked by S. pneumoniae (Radin et al., 2005). S. pneumoniae has been shown in vitro to traverse eukaryotic cell layers via transcytosis (Ring et al., 1998; Zhang et al., 2000), which has been proposed to be a crucial event in the development of invasive disease. Transcytosis is a process by which macromolecules are transported through cells in vesicles and is mediated by a specific signaling cascade (Fishman et al., 1987). The exact mechanisms that control this process are still poorly understood, but host membrane receptors shown to be involved are plgR for epithelial and PAFR for endothelial cells (Ring et al., 1998; Zhang et al., 2000). Pretreatment of both rat and human brain microvascular endothelial cells (BMEC) with a PAFR antagonist decreased the amount of intracellular bacteria by 35% and the relative amount of transcytosed pneumococci by 25% (when using human BMEC) and 30% (rat BMEC), suggesting that pneumococcal transcytosis is reduced in the absence of PAFR (Ring et al., 1998).

Acid exposure of human tracheal epithelial cells increased both PAFR mRNA and protein levels as well as adhesion of S. pneumoniae to tracheal epithelial cells (Ishizuka et al., 2001). Treatment with a specific competitor of PAF for the binding to the PAFR, decreased the number of S. pneumoniae adhered to human tracheal epithelial cells exposed to acid and not under normal conditions (Ishizuka et al., 2001). A study that looked at pneumococcal adherence to human airway epithelial cells exposed to urban particulate matter also indicated a role for PAFR (Mushtaq et al., 2011). In both A549 airway epithelial cells and human primary bronchial epithelial cells, exposure to urban particulate matter increased the mRNA transcript level, PAFR protein expression and pneumococcal adhesion. In addition, the PM-stimulated bacterial adhesion to A549 cells was attenuated by addition of a PAFR antagonist (Mushtaq et al., 201 1).

56

Chapter 3

PAF contains a phosphorylcholine (ChoP) moiety, which is also a component of teichoic and lipotechoic acids in the cell wall of S. pneumoniae (Briles and Tomasz, 1973). 111erefore, it was hypothesized that the pneumococcal ChoP moiety is the initial pneumococcal component responsible for binding to the PAFR through molecular mimicry. In line with this, growing S. pneumoniae in medium with ethanolamine and no choline, which removes the phosphorylcholine from the cell wall, with the consequence that the choline binding proteins would also no longer be anchored to the bacterial surface (Thomas et al., 1978), reduced bacterial adhesion to human endothelial cells to levels observed with quiescent, non-stimulated cells (Cundell et al., 1995). S. pneumoniae

occurs in various phase variants such as opaque and transparent forms. Opaque variants contain more capsular polysaccharide relative to the ChoP containing cell wall teichoic acids, an increase in the amount of pneurnococcal surface protein A (PspA), the pneurnococcal autolysin LytA and a decrease in choline-binding protein (Cbp) A (also known as PspC), compared to transparent variants (Briles and Tornasz, 1973; Weiser et

al., 1994; Kim and Weiser, 1998; Gosink etal., 2000). The opaque form has a reduced ability to colonize the nasopharynx, but an increased virulence in the blood stream, as was shown in mouse models of systemic disease (Briles and Tornasz, 1973; Weiser et al., 1994; Kirn and Weiser, 1998; Gosink et al., 2000) . Transparent variants invaded both human and rat BMEC more readily than opaque variants, and activation of BMEC with TNFa

resulted in increased invasion of both transparent and opaque pneurnococci (Ring et al.,

1998). The increase in pneumococcal invasion was more marked in transparent variants and growth of these variants in medium with ethanolarnine reduced levels of adherence and invasion to the basal level observed with opaque variants and non-activated cells. Pretreatment of TNFa-activated and resting HBMEC with a PAF receptor antagonist resulted in a 35% and 15% decrease in invasion of transparent phase variants, but had a smaller effect on the invasion of opaque variants, which was expected because they contain less phosphorylcholine than transparent variants (Ring et al., 1998). Transparent pneurnococci harbor significantly larger amounts of CbpA, which is non-covalently bound to ChoP in the cell wall and shown to be important for adhesion to host cells and virulence (Rosenow et al., 1997; Jedrzejas, 2001). Deletion of CbpA had no impact on colonization of the nasopharynx, but impaired transition to the lungs, resulting in less fulminate pneumonia compared to wild-type (WT) bacteria and prevented entry into the cerebral spinal fluid (CSF) (Orihuela et al., 2006). In a bacterernia model, deletion of CpbA led to a significant increase in survival rates, as the 50% lethal dose was 250-fold higher than for the WT bacteria (Iannelli et al., 2004). It was proposed that CbpA may act as a bridge between pneurnococcal ChoP and PAFR, which could explain the observed virulence defects of CbpA mutants and the increased invasiveness of transparent variants. Thus, these studies indicate that the amount of ChoP in the cell wall of S. pneumoniae can modulate the interaction with the PAFR receptor (Thornton et al., 2010).

57

Chapter 3

A study by Garcia Rodriguez (Garcia Rodriguez et al., 1995) indicated that

N-glycosylation facilitates expression of the PAFR on the cell membrane, as the amount

of PAFR lacking the N-linked glycosylation site on the second extracellular loop on the

surface was only 30 % of the WT. In line with this, adhesion of pneumococci to COS

cells transfected with the mutant fonn of human PAFR was also reduced to 30%. This

suggests that pneumococci may recognize a protein component of the PAFR and also

support a role for the receptor in pneumococcal binding.

Although all these in vitro studies point to a direct role of PAFR in the pneumococcal

adhesion to invasion of and transcytosis of human cells, treatment with antagonists

or antibodies against PAFR never inhibits pneumococcal interaction completely, most

studies reach an approximately 50% reduction at most. This indicates the existence of

PAFR-independent pathways for adhesion and invasion. In fact, inflammation also

leads to an increase in protein level of keratin 10, laminin receptor and pigR, all of

which are known to be mediators of pneumococcal adhesion to human cells (Hinojosa

et al., 2009; Orihuela et al., 2009; Sanchez et al., 2010; Shivshankar et al., 201 1). Therefore,

inflammation and subsequent endothelial activation can induce the expression of several

receptors, including the P AFR that may be responsible for the increased adhesion and

uptake of pneumococci.

Role of PAFR in pneumococcal disease in animal models

The involvement of PAFR in IPD was also investigated using in vivo animal

models representing various aspects of pneumococcal disease in combination with

absence of the PAFR gene (Ptafr I mice) or treatment with PAFR antagonists (see Table

2). Ptafr+ mice had more bacterial growth locally in the lungs at 24 h after inoculation

as determined using bioluminescent S. pneumoniae variants, but after 48 h, the incidence

and degree of bacteremia was much higher in WT mice than in Ptafr+mice (Radin et al.,

2005). Imaging of the infection showed that in Ptafr+ mice the infection was restricted

to the pulmonary area 48 h after challenge. In contrast, in WT mice the bacteria were

dispersed throughout the body including the central nervous system (CNS), probably

due to the higher level of bacteremia in the WT mice. Thus, in this model trafficking of

S. pneumoniae from the lungs to the blood and from the blood to the brain depends on

the presence of PAFR (Radin et al., 2005).

Another study looking at the role of PAFR in pneumococcal pneumonia, found

that all WT mice died within 85 h after infection, while mortality was delayed and

reduced in Ptafr 1- mice (Rijneveld et al., 2004). The bacterial loads in the lungs of 187

Ptajr+ mice were significantly lower than in WT mice and, in line with the previous

study, the amount of bacteria in blood was also lower in Ptafr+ mice. Moreover, at 42 h

after challenge, heavy inflammatory infiltrates were detected in the lungs of WT mice,

while lung inflammation was less pronounced in Ptafr+ mice, showing that PAFR

58

Chapter 3

influences disease severity in pneumococcal pneumonia (Rijneveld et al., 2004). A role for PAFR in IPD was also shown in an experimental sickle cell disease

mouse model. Sickle cell disease causes a 600-fold increased risk of IPD. After intranasal administration of S. pncu111011iac, 57% of mice transplanted with sickle cell bone marrow died, compared to 16'¼> of mice transplanted with normal bone marrow (Miller et al.,

2007). Histopathological analysis showedthat "sickle cell mice" expressed significantly more PAFR on the endothelium and epithelium than their healthy controls. Complete absence of PAFR led to a continuous protection from pneumococcal-induced mortality, whereas pharmacological intervention with PAFR antagonist was only transiently protective, in fact at 48 h the antagonist effect was lost (Miller et al., 2007).

Additional evidence of PAFR involvement in IPD is provided by studies on the uptake of pneumococcal cell wall into host cells by innate immunity receptors. Upon injection, choline-containing bacterial cell walls bound to endothelial cells and caused rapid lethality in WT, Tlr2 ' , and Nod2 ' mice but not in Ptafr mice (Filion ct al., 2006). Additionally, pneumococcal cell wall-mediated toxicity in WT mice was abrogated in the presence of a PAFR antagonist and partially in the presence of cytidine diphosphate (CDP)-choline, an intermediate in the generation of phosphatidykholine from choline (Filion et al., 2006). This also supports the idea that PAFR recognizes and facilitates uptake of pneumococcal choline-containing cell wall in vivo and that this might mediate binding and uptake of S. pneumoniae to non-phagocytic cells.

In the absence of infection, aged mice have increased basal levels of lung inflammation, shown by cytokine analysis and histopathology of lung sections. Keratin 10 (KlO), laminin receptor (LR) and PAFR, host proteins known to be mediators of pneumococcal adhesion to human cells, were increased in the lungs (Shivshankar et

al., 2011) . Exposure of normal A549 cells to conditioned media from senescent cells doubled the amount of PAFR protein and S. pneumoniae adhesion (Shivshankar et al.,

201 1). Similarly, aged mice and young ones pre-treated with TNFa to mimic increased basal inflammation, had increased levels of plgR and PAFR in their lungs and were more susceptible to pneumococcal infection. Whether aged mice also had lower levels of TNF-alpha during infection was not mentioned. However, during pneumonia, aged mice had reduced levels of both plgR and PAFR in comparison to the young ones, which was unexpected because the aged mice had 10,000-fold more bacteria in their lungs at the time of the tissue collection (Hinojosa et al., 2009).

Other studies established an indirect role for PAFR in pneumococcal disease. The involvement of the PAFR in pneumolysin-induced acute lung injury was studied in mice, in which the pulmonary arterial pressure (Ppa) was continuously monitored, as pulmonary hypertension is often associated with lung damage (Witzenrath et al.,

2007; Jyothula and Safdar, 2009). Pneumolysin (PLY), a cholesterol-dependent toxin, is one of the major virulence factors of S. pneumoniae and facilitates bacterial adherence

59

Chapter 3

to host cells and subsequently damages them during bacterial invasion (Marriott et al.,

2008). Ex vivo analysis performed on lungs revealed that the PLY-induced Ppa increase was significantly lower in Ptafr-l- mice compared to WT mice. Moreover, treatment with PAFR antagonist reduced the PLY-induced hypertension in the lungs of WT mice (Jyothula and Safdar, 2009).

The CCAAT/enhancer-binding protein b (C/EBPb) was recently discovered as an essential mediator of the inflammatory response to bacterial infections (Duitman et al.,

2012). C/EBPb is a member of the C/EBP family of transcription factors (Lekstrom-Himes and Xanthopoulos, 1998). Expression of C/EBPb is typically low in most cell types, but is rapidly induced upon various extracellular stimuli, such as IL-1, IL-6, LPS, and TNF-a (Alam, 1992; Juan, 1993) and several in vitro studies have suggested that C/EBPb has an important role in inflammation (Alam et al., 1992; Hu et al., 2000). Expression of P AFR was significantly reduced during pneumococcal infection in Cebpd·1· mice compared with WT controls. Of particular interest, in mice intranasally infected with S. pneumoniae it was found that PAFR expression was low than uninfected mice and increased significantly over time in epithelial cells of WT mice and, only marginally, in Cebpd+ mice (Duitman et al., 2012). Thus, these results suggest that C/EBPb-mediated PAFR expression is an important factor for S. pneumoniae dissemination into the bloodstream.

Taken together, these mouse model studies clearly show that P AFR plays a distinctive role in pneumococcal pathogenesis. However, there are also studies that argue against a role for P AFR in pneumococcal pathogenesis. Sepsis-associated neuronal damage (SAND) was induced in mice using an intravascular challenge with purified pneumococcal cell wall. This study showed that neuronal damage can occur without direct invasion of the brain by bacteria (Orihuela et al., 2006). Interestingly, SAND was still present in mice lacking the PAFR, but was less severe in mice without cell wall recognition proteins TLR2 and NOD2, and in mice overexpressing interleukin-10 (IL-10) in macrophages (Orihuela et al., 2006). These findings showed that PAFR and/or binding of pneumococcal cell wall to PAFR does not have a crucial role in inflammation caused by pneumococcal components and that inflammation outside the PAFR axis can be damaging. Recently, it has been described that the laminin receptor (LR), a 67kDa protein present on eukaryotic cell basolateral membranes, promotes contact of S. pneumoniae and other meningeal pathogens with the blood-brain barrier in a bacteremiaderived meningitis model (Orihuela et al., 2009). Furthermore, CbpA was identified as interaction partner for LR using CbpA coated beads. However, binding of CbpAcoated beads was independent of PAFR-mediated bacterial uptake, as adherence of pneumococci to CbpA-coated beads was equally high in Ptajr+ and WT mice (Orihuela et al., 2009). These studies argue against a role of PAFR in binding CbpA and/or cholinecontaining cell wall, and thus IPD development.

60

Chapter 3

Involvement of the PAFR in the development of secondary pneumonia due to S. pneumoniae and influenza virus synergism

HistoricaIIy, bacterial infections following influenza are an important cause of morbidity and mortality worldwide. S. pneumoniae is commonly associated with secondary pneumonia in humans infected with influenza virus (McCullers, 2006). However, the mechanisms that cause this lethal synergism are still not clear. In an attempt to elucidate what causes the susceptibility to pneumococcal pneumonia after an influenza infection several studies have scrutinized the role of PAFR in trus process. One study did not find any indications that the PAFR plays a role in secondary infection development (McCullers, 2006). In post-influenza pneumonia induced by S. pneumoniae, groups of mice treated with PAFR antagonist had survival rates similar to those of control mice. Moreover, bacterial titers in lung and blood were even rugher when PAFR was blocked by PAFR antagonist (McCullers and Rehg, 2002). Furthermore, PtaJr1- mice infected with influenza and challenged 7 days later with S. pneumoniae had survival rates 20-40% lower than those of WT mice (McCullers et al., 2008). Interestingly, the same study showed that neither antibody-mediated nor chemical inhibition of the PAFR affected adherence of pneumococci to A549 cells, nor did it interfere with the increased adhesion induced by pre-incubation with the virus (McCullers et al., 2008). Thus, these studies indicate that PAFR plays no role in the development of secondary pneumonia and the increased adhesion of pneumococci to respiratory epithelial cells pre-treated with influenza virus.

In contrast, Van der Sluijs et al demonstrated a role for the PAFRin the S.pneumoniae

influenza virus synergism. An increase in PAFR expression was observed in lungs of mice infected with influenza virus. Furthermore, PtaJr1-mice had a significantly reduced bacterial infection in the lungs and an increased survival rate of the secondary infection (van der Sluijs et al., 2006). The models to study the role of PAFR in S. pneumoniae

influenza virus synergism were not identical and different pneumococcal strains were used, which prevents a direct comparison of the results. Thus, whether or not the PAFR is involved in secondary pneumococcal infections still remains unclear.

Clinical significance of genetic variation in PAFR in human IPD

Adoption and twin studies have shown that genetics are major determinants of susceptibility to infectious diseases (Sorensen et al., 1988; Haralambous et al., 2003). Defects in innate immunity have been described to be associated with susceptibility to IPD within families (Johnson et al., 1992; Fijen et al., 1994). These studies support the idea that genetics are important in susceptibility to IPD. Single base-pair alterations (single nucleotide polymorphisms [SNPs]) occur regularly in genes controlling the host response to microbes, and may explain inter-individual differences in susceptibility (Kwiatkowski, 2000; Brouwer et al., 2009). A meta-analysis on genetic association studies

61

Chapter 3

in IPD showed that genetic variation in the complement (mannose-binding-lectin), and the Toll-like receptor activation pathway (NFKBIA, NFKBIE, TIR.AP) influence the risk of infection (Brouwer et al., 2009). Recently, genetic variation in the G-coupled protein receptor p2-adrenoceptor (ADRB2) was described as a risk factor for susceptibility in IPD as well (Adriani et al., 2012). Functional diversity of the PTAFR gene may as well play a role in host-pathogen interaction. A functional miss-sense mutation from Ala to Asp on position 224 of the PTAFR gene has been described in the Japanese population and was shown to influence coupling of the receptor to G-proteins, and impair the PAF-PTAFR

signaling cascade (Fukunaga et al., 2001). The impaired coupling resulted in a 66% less effective chemotaxis in cells homozygous for the variant allele (Fukunaga et al., 2009). A population-based study assessed the influence of this SNP on the risk of pneumococcal infections in 182 European Americans and 53 African Americans with IPD (Lingappa et al., 2011 ) . African American patients with IPD were more likely to have the variant allele compared to African American controls (odds ratio 2.38, 95% confidence interval 1 .29-4.39; P= 0.005). In European Americans the genotype frequencies were similar between patients and controls, with a low minor allele frequency in both groups (3%). Because no correction for multiple testing was performed in this study and 28 different SNPs were tested, the results could however be false positive (Lingappa et al., 2011). Therefore, validation of this result should be performed before any firm conclusions can be drawn on the role of genetic variation in PTAFR as a risk factor for IPD.

PAFR and disease development by other pathogenic microorganisms

The PAFR has also been implicated in disease development by other pathogenic bacteria, and some viruses also exploit the PAFR for interaction with human cells. Nontypeable H. infiuenzae (Nllii) is a common cause of human invasive disease and initiates infection by colonizing the upper respiratory tract. Pretreatment of bronchial epithelial cell monolayers with a PAFR antagonist significantly inhibited the invasion of the NTHi 2019 strain (Swords et al., 2001 ). Confocal microscopy of bronchial cell layers, showed only partial co-localization of NTHi and PAFR, suggesting that H. influenzae binding to PAFR may not be essential for the observed effects (Swords et al., 2000). Additionally, histopathologic analysis of lungs of WT and Ptajr1- mice 24 h after challenge with NTHi, showed no difference in the degree of inflammation (Branger et al., 2004), which argues against a major role of PAFR in the development of NTHi pneumonia. In addition, it has been suggested that ChoP contributes to adherence of the pneumococcus to host cells by binding to PAFR, whose natural ligand, the PAF, also contains ChoP, suggesting that ChoP might function as an adhesin also in H. infiuenzae (Weiser et al., 1998) . The role of ChoP in host-pathogen interactions has been discussed in detail by two recent reviews (Thornton et al., 2010; Clark and Weiser, 2012).

Staphylococcus aureus is a Gram-positive coccus that can cause diseases ranging

62

Chapter 3

from minor skin infections to life-threatening conditions, such as pneumonia, meningitis and bacteremia. It was observed that stapyholoccal LTA stimulates an immediate intracellular Ca2+ flux in human epidermoid (KB) cells, which depended on the presence of PAFR and mimicked the effect seen with PAFR agonist treatment. In line with this, intradermal injections of LTA and the PAFR agonist induced cutaneous inflammation in WT but not in Ptajr+ mice (Zhang et al., 2005). The role of PAFR was also investigated in a rabbit sinusitis model induced by heat-killed S. aureus. Histopathological analysis of the sinus mucosa clearly showed decreased inflammation in the group treated with a PAFR antagonist compared to the controls (Karasen et al., 2004). Similar results were obtained using a guinea pig model of otitis media, in which inflammation was induced by middle ear inoculation of killed S. aureus. Histopathology of the temporal bones showed decreased inflammation in the PAFR antagonist-treated group compared to the controls (Karasen et al., 2000). These results suggest that the PAFR might be involved in S. aureus pathogenesis; however, S. aureus does not contain PspC or ChoP, suggesting that this is probably due to its role in inflammation.

Pseudomonas aeruginosa is a major cause of nosocomial pneumonia, a serious disease associated with high morbidity and mortality worldwide, especially in cystic fibrosis patients. Upon intranasal challenge with P. aeruginosa to induce pneumonia, the host defense was compromised in Ptajr+ mice, which was reflected by increased bacterial outgrowth at the primary site of infection in compared to the WT background and increased inflammation and damage in the lungs. In addition, the ability of Ptafr-1

neutrophils to phagocytose P. aeruginosa in vitro was reduced compared to WT-derived cells. These finc)ings suggest that PAFR is an essential component for an effective host response to P. aeruginosa pneumonia (van de Zoelen et al., 2008).

Acinetobacter baumannii is a Gram-negative bacterium, which can cause severe nosocomial disease and also contains ChoP on the outer membrane protein PorinD. Removal of ChoP from the A. baumannii cells reduced adherence to and invasion of A549 by 95%. Immunofluorescent analysis of A549 cells infected with A. baumannii Chop+ showed that ChoP co-localized with A549 cells. In addition, in a mouse pneumonia model for lung infection, A. baumannii Chop+ was attenuated by the PAFR antagonist. In fact, treatment with the PAFR antagonist significantly reduced bacterial loads in infected lungs by approximately 1 log cfu/ml. Therefore, these data strongly suggest that A. baumannii interacts with PAFR to invade the host cell (Smani et al., 2012). The role of PAFR in influenza virus infection by subtypes A/WSN/33 HlNl or A H3Nl, was studied using both the administration of a PAFR antagonist after the occurrence of the influenza symptoms in WT mice and infection of Ptafr1 mice. Absence or blockade of PAFR caused significant protection against influenza-associated lethality and lung injury, which was correlated with decreased levels of neutrophil recruitment, lung edema, vascular permeability and injury (Garcia et al., 2010).

63

Chapter 3

Bacterial microbiota examinations have revealed the presence of S. pneumoniae in

the upper respiratory tract of patients infected with different types of virus, including

rhinovirus (RV). Interestingly, a specific inhibitor of PAFR, PAF and the pyrrolidine

derivative of dithiocarbamate, an inhibitor of the transcription factor nuclear factor

kappaB, decreased the number of pneumococci adhering to human tracheal epithelial

cells after RV-14 infection (Ishizuka et al., 2003). Thus, it seems that RV-14 infection

facilitates adhesion of S. pneumoniae in human tracheal epithelial cells by inducing PAFR

expression. Neural apoptosis is a clear sign of Human immunodeficiency virus type 1

(HIV-1) infection of the CNS. Activated H IV-1-infected rnonocytes produce high levels

of lNFa and PAF that cause neuronal apoptosis, which can be blocked by incubation

with PAFR antagonist (Perry et al., 1998). This suggests that the blockade of PAFR may

be a treatment option for HN-1- associated neurodegeneration, for example Alzheimer

Disease. In summary, all studies above indicate that PAFR is an important factor in

the development of invasive infections, but provide little to no evidence for the direct

binding of the pathogens to the receptor.

Concluding remarks

Currently, knowledge of the mechanisms by which various host receptors

contribute to development of IPD is far from complete. For some receptors, such as plgR

and LR, pneumococcal proteins have been clearly established as ligands, but for PAFR

the mechanisms involved are still unclear. The in vitro and most of the in vivo experiments

discussed in this review clearly indicate that PAFR contributes to the development of

IPD. However, none of them provides unequivocal evidence that S. pneumoniae anchors

to the PAFR in vivo and that this is crucial for IPD development due to invasion and

translocation of host cells. In fact when PAFR is genetically deleted or chemically

inhibited significant adhesion and invasion of human cells by S. pneumoniae still occurs,

indicating the presence of alternative receptors that are engaged by the bacteria. Only

a few indications exist for direct binding of S. pneumoniae to the PAFR, including the

confocal microscopy imaging study by Radin (Radin, 2006), which however provides

no biochemical evidence for a direct interaction, and the study by Garcia Rodriguez et

al. (Garcia Rodriguez et al., 1995), which indicates that N-linked glycosylation in the

second extracellular loop may not be a putative binding site for S. pneumoniae on the

PAFR. This is in contrast to studies performed on the interaction of PspC with plgR

and the LR, or between S. pneumoniae and extracellular matrix proteins (Zhang et al.,

2000; Hammerschmidt et al., 2007; Quin et al., 2007; Bergmann et al., 2009; Orihuela

et al., 2009; Agarwal et al., 2010). The direct interaction between human plgR and S.

pneumoniae was clearly demonstrated through a biochemical approach and CbpA was

identified as ligand for the receptor by immunoblot analysis in combination with affinity

purification (Zhang et al., 2000). Additionally, the complement regulator Factor H also

64

Chapter 3

binds to PspC, which was clearly shown using deletion constructs and surface plasmon resonance experiments (Agarwal et al., 2010). Upon contact of S. pneumoniae with hostcell-bound vitronectin, pneumococci are confined to peculiar forms such as microspikelike structures. Specific interactions of S. pneumoniae with the heparin-binding sites of purified multimeric and not monomeric vitronectin have been demonstrated using flow cytometry, immunoblotting, and peptides inhibiting the interaction (Bergmann et al., 2009). Both the interaction with vitronectin and with Factor H also facilitates uptake of the bacteria by the host cells (Hammerschmidt et al., 2007; Quin et al., 2007; Agarwal et al., 2010). In vitro and animal experiments showed that pneumococcal PspC is required for development of bacterial meningitis, and that PspC binds to laminin receptor (Orihuela et al., 2009). Thus, in comparison with other pneumococcal interactions with host proteins, the interaction of S. pneumoniae with PAFR is not well characterized. The effects observed in the papers discussed in this review, might well be exerted through the binding of PAF or pneumococcal cell wall components to the receptor and the subsequent inflammatory consequences of its activation (see Figure lB).

Additional studies using approaches similar to those employed to demonstrate the pneumococcal interaction with plgR, factor H, LR and vitronectin are needed to determine whether or not S. pneumoniae binds to the PAFR. Thus studies together with investigation of the role of PTAFR genetic variation in human populations would determine the nature of the PAFR involvement in IPD development. Most studies clearly show a role for PAFR in IPD and that treatment of mice with PAFR-antagonists exerts a beneficial pharmacological effect as it prevents or ameliorates disease. This suggests that modification of PAFR activity could be a promising therapeutic option and therefore clarification of its precise role in IPD is essential.

65

Chapter 3

A �' f.lfrf L1J1/{.}IJ/Jt

CbpA phosphorylcholine

Invasion, translocaffon

Blood

Brnin

B &.:._• , ... ,. ... .. . , )

....,/�smmsl/on

lnflsmmstlon

pneumococcal "-._ components "'-,.

/ PAF

PAFR

Inflammation leading to neutrophil influx,

invasion and trans/ocation

Figure 1. Schematic representation of the S. pneumoniae - PAFR interactions discussed in this review. (A) PAFR is proposed to directly bind S. pneumoniae via the ChoP moieties in the TA and LTA of the cell wall and/or CbpA. Binding of pneumoocci to the PAFR would lead to uptake of the bacteria followed by transcytosis and bacterial translocation of the cell layer. (B) Proposed model based on the literature reviewed of the involvement of PAFR in invasive pneumococcal disease via indirect effects. Upon infection with S. pneumoniae, inflammation develops due to the presence of the bacteria and or pneumococcal components. Subsequently, PAF and or pneumococcal components bind and activate PAFR leading to inflammation and neutrophil infiltration. This would then facilitate pneumococcal transmigration over human cell layers.

66

Chapter 3

Table 1. Overview of the in vitro studies and results discussed in the review.

Reference Cell type Experimental PAFR Results approach antagonist

L659989 Approximately 50% reduction of

Cundell, HUVEC, Treatment with adherence and internalization of S. 1995 A549 PAFR antagonist

WEB2086 pneu111011ine in cells activated by TNFa or IL-1 .

Confocal microscopy

82% co-localization of PAFR and S. Radin, 2005 Rat BMEC analysis of PAFR none p11eu111011inc. expression and

pneumococci

Invasion decreased to 35 %

Human, Treatment with a (transparent) and 20 % (opaque) in Ring, 1998

rat BMEC PAFR antagonist L659989 activated and 25 % (transparent) in resting cells (no inhibition of opaque

variants in resting cells).

Human Approximately 30-40% reduction in lshizuka, Tracheal

2001 Epithelial PAFR antagonist Y24180 adherence of S. p11cu111011inc only after

Cells acid exposure.

Musthaq, PM increased PAFR expression and led

2011 A549 PAFR antagonist WEB2086 to a 3-fold increase of pneumococcal adhesion.

Anti PAFR 3,4,5-trime- Neither antibody-mediated nor McCullers,

A549 antibody, PAFR thoxyphe- chemical inhibition of the PAFR 2008

antagonist nyl-1,3-diox- affected adherence of pneumococci to olane A549 cells.

HUVEC = Human Umbilical Vein Endothelial Cells, A549 = Human Nasopharyngeal Epithelial Cells, PAFR = Platelet-Activating Factor Receptor, TNF = Tumor Necrosis Factor, IL = Interleukin, BMEC = Brain Microvascular Endothelial Cells, PM = Particulate Matter

67

Chapter 3

Table 2. Overview of the data from in vivo experiments analyzed in the review.

Reference Experimental

Model PAFR an-

Results approach tagonist

At 48 h afler challenge, incidence and degree Comparison of Balb/e mice of bactcremia was much higher in WT than in

Radin, 2005 Pta/r· vs WT lntratracheal chal- none Ptafr 1·• mice lenge In PtaJi-/- infection was restricted only to the

pulmonary area.

Rijncveld, Comparison of

C57BL/6 mice Mortal i ty delayed and reduced in PAFR-1• .

PAFR'· vs WT none At 42 h heavy inflammatory infiltrates 2004

mice Intranasal challenge

were detected in the lungs of WT.

Comparison of Sickle cell mice expressed significantly more PtaJi-1• vs WT

C57BL/6 mice PAFR on the endothelium and epithelium. Miller, 2007 mice and PAFR

Intranasal challenge BN5202 1 Ptafr 1- resulted in protection of sickle cell mice

antagonist from mortality by S. pneumoniae.

Comparison of C57BL/6 mice

Choline-containing cell walls caused lethality in

Filion, 2006 PtaJi-1• vs WT

lntravascular admin- CV6209 WT but not in Ptaji· 1

mice and PAFR istration

Cell wall mediated toxicity abrogated with PAFR antagonist antagonist.

Shivshan-Aged mice

Balb/c mice Aged mice had increased lung inflammation and kar, 20 1 1 Intranasal chal lenge

none more PAFR expression.

Balb/c mice TN Fa increased levels of PAFR in the lungs and

Hinojosa, Aged mice Intranasal chal lenge none

increased susceptibility to pneumococcal infection. 2009 During pneumonia, aged mice had reduced levels

of PAFR.

Comparison of PLY-induced Ppa increase significantly lower in

Witzenrath, Ptaji-'· vs WT Balb/c mice

BN50730 the PtaJi-1•• mice and PAFR Direct infusion into

2007 antagonist the pulmonary artery

BM l 3505 PAF R antagonist reduced PLY-induced hypertension in the lungs of WT mice

Orihuela, Comparison of Balb/c mice

2006 Ptafr'· vs WT Intranasal and intra- none SAND still present in Ptaji· mice

mice venous challenge

Orihuela, Comparison of C57BL/6 mice

Pneumocoecal adherence was equal in WT and 2009

Ptafr1• vs WT I ntravascular cha I- none Ptafr mice lengc

McCullcrs, Balb/c mice PAFR antagonist did not change survival rates.

2002 PAFR antagonist

Intranasal chal lenge CV6209 Bacterial titers in lung and blood higher when

PAFR blocked.

McCullcrs, Comparison of

C57BL/6 mice Pta[,-'· lower rates of survival. 2008

Pta/i-'· vs WT Intranasal challenge

none No bacteria in the blood observed in PAFR·1· .

mice

Van der Comparison of

C57BL/6 mice Ptaji·-1- reduced bacterial infection in the lungs and Sluijs, 2006

Ptaji-'· vs WT Intranasal challenge

none increased survival rate of the secondary infection

mice

PLY Pneumolysin, SAND; Sepsis-associated neuronal damage, Ppa- Pulmonary Arterial Pressure

68

Chapter 3

References 1 . Adriani, K.S., Adriani, K.S., Brouwer, M.C., Baas, F., Zwinderman, A.H., van der Ende, A., et al

(2012) Genetic Variation in the (32-Adrenocepter Gene Is Associated with Susceptibility to Bacterial Meningitis in Adults. PLoS One. 7:e3761 8.

2. Agarwal, V., Asmat, T.M., Luo, S., Jensch, I., Zipfel, P.F., Hammerschmidt S. (201 0) Complement regulator Factor H mediates a two-step uptake of Streptococcus p11e11mo11iae by human cells. J Biol Chem. 285:23486-95.

3. Alam, T., An, M.R., Papaconstantinou, J. (1 992). Differential expression of three C/EBP isoforms in multiple tissues during the acute phase response. J Biol Chem. 267:5021-4.

4. Bergmann, S., Lang, A., Rohde, M., Agarwal, V., Rennemeier, C., Grashoff, C., et al. (2009) Integrinlinked kinase is required for vitronectin-mediated internalization of Streptococcus p11eumo11iae by host cells. J Cell Sci. 122:256-67.

5. Bogaert, D., De Groot, R., Hermans, PW. (2004) Streptococcus p11e11111011iae colonisation: the key to pneumococcal disease. Lancet Infect Dis. 4:144-54.

6. Branger, J., Wieland, CW., Florquin, S., Maris, NA., Pater, JM., Speelman, P., et al. (2004) Plateletactivating factor receptor-deficient mice show an unaltered clearance of nontypeable Hae111oplzil11s i11fl11e11zae from their respiratory tract. Shock. 22:543-7

7. Briles, EB., Tomasz, A. (1973) Pneumococcal Forssman antigen. A choline-containing lipoteichoic acid. J Biol Chem. 248:6394-7

8. Brouwer, M.C., de Gans, J., Heckenberg, S.G., Zwinderman, A .H., van der Poll, T., van de Beek, D. (2009) Host genetic susceptibility to pneumococcal and meningococcal disease: a systematic review and meta-analysis. Lancet Infect Dis. 9: 31-44.

9. Chao, W., Olson, M.S. (1993) Platelet-activating factor: receptors and signal transduction. Biochem J. 292:61 7-29.

10. Clark SE, Weiser JN. (2012) Microbial modulation of host immunity with the small molecule phosphorylcholine. Infect Immun. 81:392-401 .

1 1 . Cundell, D.R., Gerard, N .P., Gerard, C., Idanpaan-Heikkila, I., Tuomanen, E.I. (1995) Streptococcus p11e11111011iae anchor to activated human cells by the receptor for platelet-activating factor. Nature. 377:435-8.

12. Dessing, M.C., Hirst, R.A., de Vos, A.F., van der Poll, T. (2009) Role of Toll like receptors 2 and 4 in pulmonary inflammation and injury induced by pneumolysin in mice. PLoS One. 4:e7993

13. Duitman, J., Schouten, M., Groot, A.P., Daalhuisen, J.B., Florquin, S., van der Poll, T., et al. (2012) CCAAT/enhancer-binding protein b facilitates bacterial dissemination during pneumococcal pneumonia in a platelet-activating factor receptor-dependent manner. Proc Natl Acad Sci U S A. 109:9113-8.

14. Fijen, C.A., Kuijper, E.J., Tjia, H.G., Daha, M.R., Dankert, J. (1994) Complement deficiency predisposes for meningitis due to nongroupable meningococci and Neisseria-related bacteria. Clin Infect Dis. 18: 780-4.

15. Filion, S., Soulis, K., Rajasekaran ,S., Benedict-Hamilton ,H., Radin, J.N., Orihuela, C.J., et al. (2006) Platelet-activating factor receptor and innate immunity: uptake of gram positive bacterial cell wall into host cells and cell-specific pathophysiology. J Immunol. 177:6182-91

16. Fishman, J.B., Rubin, J.B., Handrahan, J.V., Connor, J.R., Fine, R.E. (1987) Receptor-mediated transcytosis of transferrin across the blood-brain barrier. J Neurosci Res. 18:299-304.

17. Fukunaga, K., Ishii, S., Asano, K., Yokomizo, T., Shiomi, T., Shimizu, T., et al. (2001) Single nucleotide polymorphism of human platelet-activating factor receptor impairs G-protein activation. J Biol Chem. 276: 43025-30.

18. Garcia Rodriguez C, Cundell DR, Tuomanen EI, Kolakowski LF Jr, Gerard C, Gerard NP. (1995) The role of N-glycosylation for functional expression of the human platelet-activating factor receptor.

69

Chapter 3

Glycosylation is required for efficient membrane trafficking. J Biol Chem. 270:251 78-84.

19. Garcia, C.C., Russo, R.C., Guabiraba, R., Fagundes, C.T., Polidoro, R.B., Tavares, L.P., et al. (2010) Platelet-activating factor receptor plays a role in lung injury and death caused by Influenza A in mice. PLoS Pathog. 6:e1001 1 71 .

20. Gosink, K.K., Mann, E.R., Guglielmo, C., Tuomanen, E.I., Masure, H.R. (2000) Role o f novel choline binding proteins in virulence of Streptococcus pneumoniae. Infect Immun. 68:5690-5.

21 . Hammerschmidt, S., Agarwal, V., Kunert, A., Haelbich, S., Skerka, C., Zipfel, P.F. (2007) The host immune regulator factor H interacts via two contact sites with the PspC protein of Streptococcus p11e11111011iae and mediates adhesion to host epithelial cells. J Immunol. 1 78:5848-58.

22. Haralambous, E., Weiss, H.A., Radalowicz, A., Hibberd, M.L., Booy, R., Levin, M. (2003) Sibling familial risk ratio of meningococcal disease in UK Caucasians. Epidemiol Infect. 1 30: 413-8.

23. Hinojosa, E., Boyd, AR., Orihuela, C.J. (2009) Age-associated inflammation and toll-like receptor dysfunction prime the lungs for pneumococcal pneumonia. J Infect Dis. 200:546-54

24. Honda, Z., Ishii, S., Shimizu, T. (2002) Platelet-activating factor receptor. J Biochem. 131 :773-9.

25. Hu, H.M., Tian, Q., Baer, M., Spooner, C.J., Williams, S.C., Johnson, P.F., et al. (2000) The C/ EBP bZ(P domain can mediate l ipopolysaccharide induction of the proinflammatory cytokines interleukin- 6and monocyte chemoattractant protein-I . J Biol Chem. 275:16373-81 .

26. Iannelli, F., Chiavolini, D. , Ricci, S., Oggioni, M.R., Pozzi, G. (2004) Pneumococcal surface protein C contributes to sepsis caused by Streptococcus pneumoniae in mice. Infect Immun. 72:3077-80.

27. Ishizuka, S., Yamaya, M., Suzuki, T., Nakayama, K., Kamanaka, M., Ida, S., et al. (2001) Acid exposure stimulates the adherence of Streptococcus pneumoniae to cultured human airway epithelial cells: effects on platelet-activating factor receptor expression. Am J Respir Cell Mo] Biol. 24:459-68.

28. (shizuka, S., Yamaya, M., Suzuki, T., Takahashi, H., Ida, S., Sasaki, T., et al. (2003) Effects of rhinovirus infection on the adherence of Streptococcus p11e11111011iae to cultured human airway epithelial cells. J Infect Dis. 188:1928-39.

29. Jedrzejas, M.J. (2001) Pneumococcal virulence factors: structure and function. Microbiol Mo! Biol Rev. 65:1 87-207

30. Johnson, C.A., Densen, P., Hurford, R.K. Jr., Col ten, H.R., Wetsel, R.A. (1992) Type I human complement C2 deficiency. A 28-base pair gene deletion causes skipping of exon 6 during RNA splicing. J Biol Chem 267: 9347-53.

31 . Juan, T.S., Wilson, D.R., Wilde, M.D., Darlington, G.J. (1993) Participation of the transcription factor C/ EBP delta in the acute-phase regulation of the human gene for complement C3. Proc Natl Acad Sci U S A. 90:2584-8.

32. Jyothula, S., Safdar, Z. (2009) Update on_pulmonary_hypertension complicating chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis. 4:351-63.

33. Karasen, R.M., Uslu, C., Gundogdu, C., Taysi S., Akcay, F. (2004) Effect of WEB 2170 BS, platelet activating factor receptor inhibitor, in the rabbit model of sinusitis. Ann Oto! Rhino! Laryngol. 1 13:477-82.

34. Karasen, R.M., Sutbeyaz, Y., Aktan, B., Ozdemir, H., Gundogdu, C. (2000) Effect of WEB 2170 BS, platelet activating factor receptor inhibitor, in the guinea pig model of middle ear inflammation. Ann Oto) Rhino) Laryngol. 1 09:549-53.

35. Kim, J.O., Weiser, J.N. (1998) Association of intrastrain phase variation in quantity of capsular polysaccharide and teichoic acid with the virulence of Streptococcus pneumoniae. J Infect Dis. 177:368-77.

36. Kwiatkowski, D. (2000) Science, medicine, and the future: susceptibility to infection. BMJ 321: 1061-5.

37. Lekstrom-Himes, J., Xanthopoulos, KG. (1998) Biological role of the CCAAT/enhancer-binding protein family of transcription factors. J Biol Chem. 273:28545-8.

38. Lingappa, J.R., Dumitrescu, L., Zimmer, S.M., Lynfield, R., McNicholl, J.M., Messonnier, N.E., et al.

(201 1 ) Identifying host genetic risk factors in the context of public health surveillance for invasive pneumococcal disease. PLoS One 6: e23413.

70

Chapter 3

39. Marriott, H.M., Mitchell, T.J., Dockrell, D.H. (2008) Pneumolysin: a double-edged sword during the host-pathogen interaction. Curr Mo! Med. 8:497-509

40. McCullers, J.A., Rehg, J.E. (2002) Lethal synergism between influenza virus and Streptococcus pne11111011iae: characterization of a mouse model and the role of platelet-activating factor receptor. J Infect Dis. 186:341-50.

41 . McCullers, J.A. (2006) Insights into the interaction between influenza virus and pneumococcus. Clin Microbiol Rev. 19:571-82.

42. McCullers, J.A., Iverson, A.R., McKean, R., Murray, P.J. (2008) The platelet activating factor receptor is not required for exacerbation of bacterial pneumonia following influenza. Scand J Infect Dis. 40:1 1-7.

43. Miller, M.L., Gao, G., Pestina, T., Persons, D., Tuomanen, E.I. (2007) Hypersusceptibil i ty to invasive pneumococcal infection in experimental sickle cell disease involves platelet-activating factor receptor. J Infect Dis. 195:581-4.

44. Mushtaq, N., Ezzati, M., Hall, L., Dickson, I., Kirwan, M., Png, K.M., et al. (201 1 ) Adhesion of Streptococcus p11e11111011iae to human airway epithelial cells exposed to urban particulate matter. J Allergy Clin Immunol. 127:1236-42.

45. Orihuela, C.J., Gao, G., Francis, K.P., Yu, J., Tuomanen, E.I. (2004) Tissue-specific contributions of pneumococcal virulence factors to pathogenesis. J Infect Dis. 190:1661-9.

46. Orihuela, C.J., Filion, S., Smith-Sielicki, SH., El Kasmi, K.C., Gao, G., Soulis, K., et al. (2006) Cell wallmediated neuronal damage in early sepsis. Infect Immun. 74:3783-9.

47. Orihuela, C.J., Mahdavi, J., Thornton, J., Mann, B., Wooldridge, K.G., Abouseada, N., et al. (2009) Laminin receptor initiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest. 1 19: 1638-46.

48. Perry, S.W., Hamilton, J.A., Tjoelker, L.W., Dbaibo, G., Dzenko, K.A., Epstein, L.G., et al. ( 1998) Platelet-activating factor receptor activation. An initiator step in HIV-1 neuropathogenesis. J Biol Chem. 273:17660-4.

49. Pober, J.S. (2002) Endothelial activation: intracellular signaling pathways. Arthritis Res. 4 Suppl 3:S109-16.

50. Quin, L.R., Onwubiko, C., Moore, Q.C., Mills, M.F., McDaniel, L.S., Carmicle, S. (2007) Factor H binding to PspC of Streptococc11s p11e11111oniae increases adherence to human cell lines in vitro and enhances invasion of mouse lungs in vivo. Infect Immun. 75:4082-7.

51. Radin, J.N., Orihuela, C.J., Murti, G., Guglielmo, C., Murray, P.J., Tuomanen, E.I. (2005) Beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus p11e11111011iae. Infect Immun. 73:7827-35.

52. Rijneveld, A.W., Weijer, S., Florquin, S., Speelman, P., Shimizu, T., Ishii, S., et al. (2004) Improved host defense against pneumococcal pneumonia in platelet-activating factor receptor-deficient mice. J Infect Dis. 189:71 1-6.

53. Ring, A., Weiser, J.N., Tuomanen, E.T. (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest. 102:347-60.

54. Rosenow, C., Ryan, P., Weiser, J.N., Johnson, S., Fontan, P., Ortqvist, A., et al. (1997) Contribution of novel choline-binding proteins to adherence, colonization and immunogenicity of Streptococcus p11e11111011iae. Mol.Microbiol. 25:819--829.

55. Sanchez, C.J., Shivshankar, P., Stol, K., Trakhtenbroit, S., Sullam, P.M., Sauer, K., et al. (2010) The pneumococcal serine-rich repeat protein is an intra-species bacterial adhesin that promotes bacterial aggregation in vivo and in biofilms. PLoS Pathog. 6:e1001044.

56. Shirasaki, H., Nishikawa, M., Adcock, J.M., Mak, J.C., Sakamoto, T., Shimizu, T., et al. (1994) Expression of platelet-activating factor receptor mRNA in human and guinea pig lung. Am J Respir Cell Mo! Biol. 10:533-7.

57. Shivshankar, P., Boyd, A.R., Le Saux, C.J., Yeh, LT., Orihuela, C.J. (2011 ) Cellular senescence increase expression of bacterial ligands in the lungs and is positively correlated with increased susceptibility

71

Chapter 3

to pneumococcal pneumonia. Aging Cell. 10:798-806.

58. Shukla, S.D. (1992) Platelet-activating factor receptor and signal transduction mechanisms. FASEB J. 6:2296-301

59. Smani, Y., Docobo-Perez, F., Lopez-Rojas, R., Dominguez-Herrera, J., Ibanez-Martinez, J., Pach6n, J. (2012) Platelet-activating factor receptor initiates contact of Acinetobacter baumannii expressing phosphorylcholine with host cells. J Biol Chem. 287:26901-10.

60. Snyder, F. (1990) Platelet-activating factor and related acetylated lipids as potent biologically active cellular mediators. Am J Physiol. 259:C697-708.

61 . Sorensen, T.I., Nielsen, G.G., Andersen, P.K., Teasdale, T.W. (1988) Genetic and environmental influences on premature death in adult adoptees. N Engl J Med. 318: 727-32.

62. Stoll, L.L., Denning, G.M., Kasner, N.A., Hunninghake, G.W. (1994) Platelet-activating factor may stimulate both receptor-dependent and receptor-independent increases in [Ca2+] in human airway epithelial cells. J Biol Chem. 269:4254-9.

63. Swords, W.E., Buscher, B.A., Ver Steeg Ii, K., Preston, A., Nichols, W.A., Weiser, J.N., et al. (2000) Non-typeable Haemophilus influenzae adhere to and invade human bronchial epithelial cells via an interaction of lipooligosaccharide with the PAF receptor. Mol Microbial. 37:13-27.

64. Swords, W.E., Ketterer, M.R., Shao, J., Campbell, C.A., Weiser, J.N., Apicella, M.A. (2001) Binding of the non-typeable Hae111ophi/11s influenzae lipooligosaccharide to the PAF receptor initiates host cell signaling. Cell Microbiol. 3:525-36.

65. Thomas, A.M., Lambert, P.A., Poxton, I.R. (1978) The uptake of choline by Streptococcus pneumoniae. J Gen Microbial. 109:313-7.

66. Thornton, J.A., Durick-Eder, K., Tuomanen, E.I. (2010) Pneumococcal pathogenesis: «innate invasion» yet organ-specific damage. J Mo] Med (Berl). 88:103-7.

67. van der Sluijs, K.F., van Elden, L.J., Nijhuis, M., Schuurman, R., Florquin, S., Shimizu, T., et al. (2006) Involvement of the platelet-activating factor receptor in host defense against Streptococcus p11eu111011iae during postinfluenza pneumonia. Am J Physiol Lung Cell Mo] Physiol. 290:L194--9.

68. van Zoelen, M.A., Florquin, S., Meijers, J.C., de Beer, R., de Vos, A.F., de Boer, O.J .. , et al. (2008)

72

CHAPTER 4

Novel _insights in the roles of the platelet-activating factor and poly immunoglobulin receptors in Streptococcus pneumoniae adhesion to

the brain microvascular endothelium

Federico Iovino, Grietje Molema, Jetta J.E. Bijlsma

Submitted

73

Abstract

Streptococcus pneumoniae (the pneumococcus), the main causative agent of

bacterial meningitis in Europe and in the USA, is thought to invade the

Central Nervous System via the bloodstream by crossing the endothelium

of the blood-brain barrier. Receptor-mediated adhesion of the bacteria to the

brain endothelium is considered a key event leading to the development of

meningitis. The platelet-activating factor receptor (PAFR) is implicated in

pneumococcal adhesion to brain endothelial cells. However, absence or blocking

of PAFR reduces pneumococcal adhesion at most 50% and it is unclear whether

pneumococci directly bind to PAFR in vivo. S. pneumoniae was shown to bind to

the poly immunoglobulin receptor (plgR) on human nasopharyngeal epithelial

cells. Whether pigR mediates pneumococcal adhesion to the blood-brain barrier

endothelium in vivo has not yet been fully investigated. Therefore, we studied

the role of PAFR and pigR in S. pneumoniae adhesion to brain endothelial

cells in a blood-home model of meningitis. Mice were intravenously infected

with pneumococci and sacrificed at various time points prior to the onset of

meningitis. Immunofluorescent analysis of brain tissue showed almost no co

localization of endothelial PAFR with S. pneumoniae. Brain endothelium cells

produced plgR and pneumococci frequently co-localized with the receptor

both in the brains and in vitro. Blocking of pigR with antibodies in vitro reduced

pneumococcal adhesion to brain endothelial cells and incubation of the bacteria

with cell lysates showed that they bind pigR. In conclusion, we showed in

vivo and in vitro that S. pneumoniae does not co-localize with endothelial PAFR

implying that pneumococci may not bind to PAFR. On the other hand, we

demonstrated that plgR is expressed by brain endothelial cells. Additionally,

S. pneumoniae co-localizes with this receptor in vivo and the bacteria physically

interact with plgR in vitro. These findings indicate that plgR on the blood-brain

barrier endothelium represents a novel adhesion receptor for S. pneumoniae.

Chapter 4

Introduction

Bacterial meningitis is a life-threatening infectious disease leading to cerebral dysfunctions and, in the most serious cases, death (1, 2). The Gram-positive bacterium Streptococcus pncumoniac (the pneumococcus) is a main causative agent of this disease (3, 4). S. pncumoniae is thought to invade the Central Nervous System (CNS) via the bloodstream by crossing the blood vessels of the blood-brain barrier, a specialized system of endothelial cells that protects the brain from harmful substances present in the bloodstream and supplies the brain with the required nutrients for its proper functions (1, 2, 3). How S. pneumoniae adheres to the blood brain barrier is unclear and subject of debate. The platelet-activating factor receptor (PAFR) is implicated in pneumococcal adhesion to endothelial cells (5, 7, 8). PAFR is a G-protein coupled receptor and the binding of its natural ligand, p la telet-acti va ting factor (PAF), activates several signaling mechanisms which mediate diverse pathological processes, such as allergy, asthma, septic shock, arterial thrombosis, and inflammation (9-14). In vitro blocking and transfection studies and most in vivo experiments using mice that genetically lack the receptor clearly indicate that PAFR contributes to the development of invasive pneumococcal disease (IPD) (5, 7, 8, 15). The question that still remains is whether S. pneumoniae binds directly to the PAFR.

When PAFR is genetically deleted or chemically inhibited, pneumococci still adhere to and invade human cells and cause disease in mice (5, 7, 8) indicating that S. pneumoniae can engage alternative receptors besides PAFR (15). One candidate might be the poly immunoglobulin receptor (plgR), which is known to bind to pneumococci in human nasopharyngeal epithelial cells (6). Within the host, plgR mediates the transport of immunoglobulins across mucosa] epithelial barriers (16-19). Treatment of human nasopharyngeal epithelial cells (Detroit) with anti-plgR antibody significantly blocked pneurnococcal adherence and translocation over epithelial cells (6) . Furthermore, irnrnunoblotting analysis showed that pneumococcal cholinebinding protein (Cbp) A binds to human plgR (20). PigR was previously shown to be expressed in neurons (21, 22, 23), but was not detected in Human Brain Microvascular Endothelial Cells (HBMEC) (6). However, there are no data on expression of plgR on the brain vasculature in vivo.

In this study, we investigated whether pneumococci co-localize with PAFR expressed by the brain vascular endothelium during events preceding meningitis. Immunofluorescent analysis performed on brain tissue from infected mice failed to show co-localization between PAFR and S. pneumoniae suggesting that a direct interaction between PAFR and pneumococci is unlikely to occur in vivo. However, we showed that plgR is present on HBMEC and on the vascular endothelium of mouse brain tissue. The majority of pneumococci co-localized with endothelial plgR and inhibition of the receptor reduced pneumococcal adherence. Furthermore we

75

Chapter 4

biochemically demonstrated that S. pneumoniae binds to pigR from endothelial cell lysates. Taken together, we have indications that direct interaction of S. pneumoniae

with PAFR is unlikely to occur in vivo and that pigR, which is expressed on brain vascular endothelium, may act as adhesion receptor for S. pneumoniae on the bloodbrain barrier.

Results

PAFR is heterogeneously expressed by the endothelium of the blood-brain barrier and S.

pneumoniae does not co-localize with this receptor

In agreement with previous studies (5, 8), treatment of HBMEC with antiPAFR antibodies significantly reduced pneumococcal adherence by approximately 50 % compared to treatment with IgG from non-immunized rabbit or the absence of added antibodies (Figure lA). When PAFR was blocked in Human Umbilical Vein Endothelial Cells (HUVEC) S. pneumoniae adhesion was approximately 30% less compared to treatment with rabbit IgG or in absence of blocking antibody, although this was statistically not significant (data not shown). This result confirms that PAFR plays a role in pneumococcal adherence to brain endothelial cells. Subsequently, we investigated whether S. pneumoniae co-localized with PAFR in vitro.

Immunofluorescent detection of PAFR in HBMEC showed heterogeneous expression by certain clusters of cells, whereas other areas of the monolayer did not express PAFR (Figure 1B). Interestingly, co-localization analysis using imageJ showed that most pneumococci adherent to HBMEC did not co-localize with PAFR (Figure lC). In brain tissue of mock-infected mice PAFR was detected mainly on the endothelium although expression was not homogenous in the various compartments of the brain (Figure 2). This distribution of PAFR expression was not influenced by S. pneumoniae

infection (Figure 3). Throughout the time course of infection, most bacteria did not co-localize with the PAF receptor in the various brain compartments (Figure 3). Specifically, analysis of the brain slides using ImageJ indicated that only about 0.8% of the pneumococci were co-localized with PAFR at all time points of infection in all brain compartments. These findings suggest that, although the PAFR is involved in pneumococcal adhesion to brain endothelial cells, direct association of S. pneumoniae

with PAFR is not likely to occur in vivo.

76

A

0

� C: 0

�

B

1 0E-1 9 0E-2 B 0E-2 7 0E-2 6 0E-2 * 5 0E-2 4 0E-2 3 0E-2 2 0E-2 1 OE-2

0 .f---'-----''- _ __ .. __

C

Chapter 4

Figure 1. PAFR is indirectly involved in S. pneumoniae adhesion to human endothelial cells (A) Blocking of PAFR (white column) in HBMEC cells leads to a reduction of pneumococcal adhesion in comparison with HBMEC treated with isotype control (hatched column) and with HBMEC treated without blocking antibody (black column). * P value < 0.05. (B) Immunofluorescent staining of PAFR (red), adherent S. pneumoniae

(green) and cellular nuclei (blue) in HBMEC. After 1 hour incubation with pneumococci, HBMEC cells were washed with PBS in order to remove the non-adherent bacteria, Total magnification 400X. (C) Co-localization of pneumococci and PAFR detected in panel B. White pixels represent the areas of bacterial signal co-localized with PAFR while green pixels represent the area of bacterial signal not co-localized with PAFR. White arrows indicate the only pneumococci co-localized with PAFR signal observed over all tissue sections of all mice analyzed.

!ubara.:tanolcf space

Cl'loroidplexu1

Figure 2. Heterogeneous expression of PAFR by brain endothelium in absence of

infection Irnmunofluorescent detection of PAFR (blue) and endothelial cells (red) in the different brain compartments of mice in the absence of bacterial infection. Total magnification 630X

77

Chapter 4

:-.rA I!'"' : :

!-t:· ... -

�.iCA'A:•.-:; s:t:t

1 hour infectlon

8 hours Infection

C:·- ::a .:.s� :r 1,..11,ss 5 ;:r�._r,;,: H·=-==t

3 hours Infection

14 hours Infection

C:-0 ::1 J' !'"i'"A 15 S .s ; .. � .. l"",:'o"·,!!�-=-='=t

Figure 3. S. pneumoniae does not co-localize with PAFR on brain endothelium during the course of infection Brain slides of mice intravenously challenged with S. pneumoniae were stained for vasculature endothelium (red), pneumococci (green), and PAFR (blue). Total magnification 630X. For each time point, the panel "Co-localization analysis S. pneumoniae-PAFR" shows bacteria co-localizing with PAFR in white, and bacteria not co-localizing with PAFR in green; white arrows indicate the only pneumococci co-localized with PAFR. For each time point of infection, brains from three mice were analyzed, and of each mouse three brain sections were used for the immunofluorescent detection. The images are representative of the situation in each brain compartment during the entire time course of infection and in each mouse that was analyzed.

S. pneumoniae co-localizes with plgR in HBMEC and blockade of this receptor reduces pneumococcal adherence

To assess whether the anti-mouse plgR antibodies could be used for immunohistochemistry/ immunofluorescence detection of plgR in mouse tissue, we applied it to lung sections and showed that epithelial cells do express plgR, as was previously reported (6, 18) (Figure S1). Immunofluorescent analysis with anti-human plgR antibodies showed that plgR was also present on HBMEC (Figure 4A). As it was previously reported that endothelial KC cells do not express plgR (6), a Western blot with the anti-human plgR antibody was performed. This analysis showed that plgR was produced by Detroit cells, and that a band of the same molecular weight was detectable in HBMEC cells. As expected no plgR expression was observed in A549 cells (6) (Figure S2),

78

Chapter 4

confirming that the immune fluorescent analysis indeed detected plgR on HBMEC.

Furthermore, co-localization analysis showed that most pneumococci adherent to

HBMEC co-localized with pigR (Figure 4B). HUVEC also expressed pigR (Figure 4C)

and, S. pneumoniae adherent to these cells also mostly co-localized with this receptor

(Figure 4D). To exclude the possibility that the antibody non-specifically bound S.

pneumoniae, we stained the pneumococci with the anti-human pigR antibody only

and consistently found that the bacteria were not detectable with this antibody ( data

not shown). Importantly, blocking of pigR in HBMEC and HUVEC cells with the

anti-human plgR antibody led to a significant reduction of pneumococcal adhesion

compared to cells incubated without this antibody and cells treated with isotype

control antibodies. Together, these findings show that plgR is functionally involved

in S. pneumoniae adhesion to endothelial cells (Figures 4E and 4F).

A B

C D

E � G

• OE-2 1 ,e.1 �.oe.2

1 21!.-t

5! lOE-2 I 1 OE-2

> =2

j 2 oe.2 c: 1 0!!-1 2 0�2 " . i G OE-t

:I; t 0!!- 2 � 4 0e-1 1 01!.-2 -2 01!1

0

Figure 4. PlgR is involved in S. pneumoniae adhesion to human endothelial cells

and pneumococci co-localize with plgR (A) Immunofluorescent detection of plgR (green), adherent S. pneumoniae (red) and cellular nuclei (blue) in HBMEC cells after 1 hour incubation with pneumococci. After 1 hour incubation with pneumococci, HBMEC cells were washed with PBS to remove the non-adherent bacteria. Total magnification 400X. (B) Co-localization between pneumococci and plgR detected in panel A. White pixels represent the areas of bacterial signal co-localized with plgR, while red pixels indicate the area of bacterial signal not colocalized with plgR. (C) Immunofluorescent detection of plgR (green), adherent S. pneumoniae (red) and cellular nuclei (blue) in HUVEC cells after 1 hour incubation with pneumococci. After 1 hour incubation with pneumococci, HUVEC cells were washed with PBS to remove the non-adherent bacteria. Total magnification 400X. (D) Co-localization between pneumococci and plgR detected in panel C. White pixels represent the areas of bacterial signal co-localized with plgR, while red pixels indicate the area of bacterial signal not co-localized with plgR. E-G. Blocking of plgR (white column) in Detroit cells (E), HBMEC (E) and HUVEC (F) leads to a significant reduction of pneumococcal adhesion in comparison with HBMEC/HUVEC treated with isotype control IgG (hatched column) or with HBMEC/HUVEC cells incubated without a blocking antibody (black column). * P value < 0.05.

79

Chapter 4

S. pneumoniae co-localizes with plgR expressed on the brain vascular endothelium

Based on the finding that HBMEC and HUVEC express plgR, we analyzed plgR expression in brain tissue of healthy mice that had been 'mock-treated' with PBS as a control for the infection experiments described below. PigR was indeed detected in the healthy brain and, amongst others, associated with endothelial cells (Figure SA). To confirm that plgR was indeed expressed by the vascular endothelium, we analyzed brain sections of healthy mice, 1 and 14 hours after mock-treatment using a three-dimensional reconstruction after confocal microscopy. This analysis showed that plgR expression is indeed associated with endothelial cells (Figure SB).

To investigate whether pneumococci co-localized with plgR in the brain of infected mice, it was necessary to label the anti-capsule antibodies with Alexa Fluor 350. When S. pneumoniae was detected with Alexa Fluor 350 (Figure 6A) the coccoid shape and chain structure typical of the pneumococcus were not always as well distinguishable as when pneumococci were detected with Alexa Fluor 488 (Figure 3). To confirm that the Alexa Fluor 350 signal indeed represented pneumococci, we stained S. pneumoniae in brain tissue using first the anti-capsule serotype 4 antibody labeled with Alexa Fluor 350, followed by the same antibody labeled with Alexa Fluor 488. This showed that the bacteria signal detected with Alexa Fluor 350 always colocalized with the signal obtained with Alexa Fluor 488 (Figure S3). In brain sections of intravenously-infected mice, the expression pattern of plgR was similar to that in mock-infected mice. Notably, most pneumococci co-localized with plgR expressed on the vascular endothelium in all brain compartments at all time points after infection (Figure 6A) . An estimation of the signals in the brain slides with ImageJ indicated that > 95% of pneumococci present in the brain were co-localized with plgR at all time points of infection in all brain compartments. Confocal microscopy confirmed that S. pneumoniae indeed co-localized with plgR on the endothelial cells (Figure 6B and movies in the Supporting Information). This implies that systemically administered S. pneumoniae anchors to plgR expressed on brain vascular endothelium prior to development of meningitis.

80

"

.. _

a

Swfla .. cnn.1.pwit

-

,... -�· - ' � �.. . ._

... }_ .. �t� ... �p�·�

'"

;"'� -.:.r. - - --�-----

' � � - � r •

_.. .. �_;, .,._ - . - � - - .

.. :..�� ·:: r ��- • , . . ..... - -

Chapter 4

Figures 5. Heterogeneous expression of plgR by brain endothelium in the absence of infection (A) lmmunofluorescent detection of plgR (green) and endothelial cells (red) in the brain in absence of bacterial infection shows heterogeneous expression of plgR on brain endothelium in vivo. Total magnification 630X. (B) Confocal microscopy images visualizing plgR (green) on the brain vascular endothelium (red). Overlaps in the plgR signal and endothelial cell staining resulted in a yellow color, suggestive of plgR expression by endothelial cells. The white scale bar in each image represents 20 µm. The images are representative of the situation in each brain compartment during the entire time course of infection and in each mouse that was analyzed.

81

Chapter 4

A

�

� <,

-·iJr.,. · � : . ._ -

B

1 hour Infection

8 houra Infection

Co ocacatiO"lan� :;; ,---11-.lgfl

.,;

1 4 --\ ,._

.. �

3 houra fnf9ctk>n

tl;,,.,._ • . plgft

1 .. houra lnfoctlon

�� r' _/ _., /

• �

" '

' �� -

Figure 6. S. pneumoniae co-localizes with plgR on brain endothelium during the

entire course of infection (A) Brain slides of mice challenged with S. pneumoniae were stained for vasculature endothelium using tomato lectin (red), pneumococci (blue), and plgR (green). Total magnification 630X. For each time point of infection, brains from 3 mice were analyzed, and of each mouse 3 brain sections were used for the immunofluorescent detection. The images are representative of the situation in each brain compartment during the entire time course of infection and in each mouse that was analyzed. For each time point, the panel "Co-localization analysis S. pneumoniae-plgR" shows bacteria co-localizing with plgR in white, and bacteria not co-localizing with plgR in blue. (B) Confocal visualization of S. pneumoniae (red) and plgR (green) in the different brain compartments during the whole time course of pneumococcal infection detected by confocal microscopy. The white scale bar in each image represents 5 µm. The yellow color of the bacteria is the result of co-localization of the red-colored S. pneumoniae and green-colored plgR. For each time point of infection, brains from 3 mice were analyzed, and of each mouse 3 brain sections were used for the confocal detection. The images are representative of the situation in each brain compartment during the entire time course of infection and in each mouse that was analyzed.

82

Chapter 4

S. pneumoniae binds to plgR expressed by human endothelial cells

Since S. pneumoniae co-localized with plgR expressed by the brain vascular endothelium in vivo, we wondered whether pneumococci can physically interact with plgR expressed by endothelial cells. Thus, we set up a ligand-receptor interaction assay as detailed in the Materials and Methods section (Figure S4). Since binding of pneumococci to plgR expressed by Detroit cells was previously described (5), we first tested our method using Detroit cell lysates as a positive control. After incubation of the pneumococci with Detroit cell lysate, plgR was indeed detected on the bacteria (Figure 7 A). As a negative control, we used an anti-human tubulin antibody. Tubulin was not detected on the pneumococci, indicating that the interaction was specific for plgR and that endothelial proteins did not become associated with the bacteria through non-specific interactions for instance through precipitation or centrifugation (Figure 7B). Subsequently, S. pneumoniae cells incubated with either HBMEC or HUVEC lysates were stained with anti-pneumococcal antiserum and the anti-human plgR antibody, which showed that in both cases plgR was present on the pneumococci (Figure 7C).

83

Chapter 4

A

Dctro 1

C

HBMEC

HUVEC

B

Oetro,1

Anh pneumococcal ant serum

Antl pneumococcal antlse,um

Anti pncumococ:cal ant serum

Anti a tubulln antibody

Overlay

Anti plgR anllbody

Anti plgR antibody

Enlarged overlay

overlay 2 channelS

.. l.. • � .. - .

C ,- .._. • . . . . . . . . -.

Overlay 2 channels

HB!-.1EC

HUVEC

- .

I · • . • •

Anti pneumocOCCl!III antiserum

Anti o tubulln ent1body

Ant.I pneumococcal antiserum

Anti a tubulln antibody

Figure 7. S. pneumoniae binds to plgR in vitro

Enlarged ow-nay

. ·-� -- �.:.; • ·- � 'I;. Enlarged overley

.,. . .. - ...

(A) lmrnunofluorescent detection of S. pneumoniae (red) and plgR (green) after pneumococci were incubated with Detroit, HBMEC or HUVEC cell lysates. Total magnification 400X. The overlay indicates plgR signal on the pneumococci. (B) Immunofluorescent detection of S. pneumoniae (red) and a tubulin (green) after pneumococci were incubated with Detroit, HBMEC or HUVEC cell lysates. Total magnification 400X. Discussion

The aim of this research was to clarify whether pneumococci adhering to the brain endothelium co-localize with either P AFR or plgR during the events preceding meningitis. In line with previous data (5, 8), we demonstrated that the blocking of PAFR significantly reduces S. pneumoniae adhesion to HBMEC. Yet, we did not observe a co-localization between endothelial PAFR and pneumococci, neither in vitro nor in vivo. In contrast, we observed plgR on brain endothelial cells, which co-localized with S. pneumoniae in vitro and in vivo. Blocking of plgR by antibodies reduced pneumococcal adherence to endothelial cells and pneumococci bound plgR present in endothelial cell lysates. These data suggest that a direct interaction

84

Chapter 4

between PAFR and S. pneumoniae is not likely to occur at the blood-brain barrier endothelium. Instead, plgR seems to represent a crucial and so far overlooked receptor for pneumococcal adhesion to the blood-brain barrier endothelium.

Data from previous studies clearly indicate that PAFR is involved in invasive pneumococcal disease. In vitro, blocking of the receptor reduced adherence to and invasion of endothelial cells. In addition, disease severity was reduced in PAFR·1

mice, in particular the translocation from blood to the brain was impaired (7). These findings were interpreted as a role for PAFR in pneumococcal adherence to the blood brain barrier. However, no biochemical evidence for a direct interaction between pneumococci and PAFR has thus far been published (15). In our present study, the possible co-localization of pneumococci with PAFR was not detectable, neither in vitro not in vivo, which lead us to conclude that S. pneumoniae is unlikely to bind to PAFR directly at least at the blood-brain barrier. This is in contrast with a study that reported considerable co-localization of PAFR and S. pneumoniae in rat brain endothelial cells in vitro (7). Unfortunately, no overview images from the latter study are available that show the level of PAFR expression in rBCEC6 cells. However, it is conceivable that these cells might express PAFR at higher levels than the cells we used in our present studies, which would increase the chance of co-localization. In any case, our finding that the pattern and levels of PAFR expression by the HBMEC cells was similar to that observed in the brains of infected and uninfected mice (heterogeneous and patchy), indicates to us that the PAFR expression that we observed in vitro likely represents the physiological situation in mice and man. Another difference is that whereas we used anti-capsule antibodies to detect the bacteria, Radin et al. used an anti-ChoP antibody (7). Although ChoP is a part of the cell wall of the bacteria, this antibody could in principle also detect ChoP not associated with bacteria. The amino acid sequence of rat PAFR has 79% identity with human PAFR (25) and 91 % identity with mouse PAFR rat. Non-conserved amino acid residues are present also in the extracellular portion of the protein, in particular in the second extracellular loop (26). These differences in the PAFR amino acid sequences may also account for the apparent S. pneumoniae colocalizing to PAFR on the rat endothelium and the absence of this co-localization with human or murine PAFR.

Although PAFR has been implicated in the adhesion of several pathogens to human cells there is a scarcity of data that demonstrates direct interaction of the bacteria with this receptor (15). Recently, it was shown that Neisseria meningitidis is capable of binding to PAFR on human airway cells through the ChoP and glycan modification of its pili. As in our study, PAFR was also shown to be heterogeneously expressed in bronchial epithelial cells, but in case of N. meningitidis there was a 100% co-localization between the bacteria and the receptor. Furthermore, immune precipitation and ELISA using purified components confirmed that PAFR is an

85

Chapter 4

adhesion receptor for N. meningitidis (27). As we do not observe any co-localization between PAFR and pneumococci in HBMEC and in the brain of mice, and colocalization is a prerequisite for a direct interaction, we consider it highly unlikely that S. pneumoniae binds to PAFR. Thus, it seems that PAFR is an important receptor for respiratory pathogens, either as a direct receptor for meningococci or indirectly for S. pneumoniae. The inflammation induced by the presence of pneumococci leads to the release of proteins and cytokines by the endothelium, including inflammation mediators such as PAF, which binds to and activates its receptor, PAFR. Alternatively, ChoP-containing fragments of the pneumococcal cell wall, released during growth and or destruction of the bacteria, could bind to PAFR and activate the receptor (28, 29). The signaling cascade of PAFR then leads to a pro-inflammatory state and events, such as neutrophil infiltration. This activation of the brain endothelial cells might subsequently facilitate transmigration of S. pneumoniae over cell layers and lead to invasive disease.

PlgR is expressed by the respiratory epithelium (6, 18, 19, 20) and on neurons (21, 22, 23), and it is a well-known receptor for S. pneumoniae (6). It has been implicated in the translocation of pneumococci over the epithelium through an intracellular pathway known as transcytosis (6). Furthermore, through interaction with the bacterial ligand CbpA, also known as PspC, plgR is necessary and sufficient for pneumococcal adherence to epithelial cells (30). Binding of the bacteria to plgR leads to the activation of signaling cascades that drive pneumococcal invasion of epithelial cells (6, 30, 31). An absence of plgR was reported in human alveolar epithelial cells (A549) and the human brain microvascular endothelial cell line KC (6), which led to the suggestion that plgR would be an exclusively epithelial receptor for S. pneumoniae. In addition, the interactions between plgR and S. pneumoniae have been studied using Madin-Darby Canine Kidney (MOCK) or Calu-3 cells which are epithelial cells, but also did not investigate the presence of plgR on endothelial cells (31, 32). On the other hand, we detected plgR both on Detroit cells and in HBMEC cells, but not A549 cells as reported before (Figure S4). This discrepancy might be due to the use of different endothelial cell lines, but is more likely to be due to the use of different antibodies by Zhang et al. and us. Zhang et al. prepared a rabbit antiserum against human plgR and a sheep antiserum against mouse plgR (6), while we used commercial antibodies specific for human or mouse plgR (see Antibodies and lectin in Material and Methods) .

PlgR acts as receptor for transport of immunoglobulins across the mucosa} epithelium (16-19). It is presently unclear what the function of plgR might be on the brain endothelium. As well as epithelial cells, also endothelial cells employ specific mechanisms for the transport of immunoglobulins. It is known that endothelial cells express the MHC class I-related receptor, FcRn, which plays an essential role

86

Chapter 4

in the transcytosis of IgG (33, 34, 35). It is thus conceivable that pigR may have a

physiological function in endothelial cells that is analogous to the one exercised by

FcRn.

To the best of our knowledge, our immunofluorescence and confocal analyses

of the co-localization of S. pneumoniae with pigR in brains of infected mice represent

the first demonstration of pneumococcal adhesion to a receptor in vivo. Furthermore,

our data reflects the findings obtained with pigR on epithelial cells, suggesting that

S. pneumoniae wi11 also have a 100 % co-localization with pigR on lung epithelium

after a nasopharyngeal challenge. It also implies that adhesion to pigR may

induce invasion of endothelial cells possibly also through interaction with PspC.

Pneumococci lacking PspC (also known as CbpA) were shown to be less adhesive to

rat BMEC than wild-type, and PspC was shown to be involved in the transition from

the lungs to the blood and from the blood into the cerebra-spinal fluid (CSP) (5, 36,

37). This indicates that, at least in the rat model, the binding of PspC to pigR might be

important for the development of both sepsis and meningitis. Thus inclusion of PspC

in experimental protein vaccines could perhaps induce a broad protection against

IPD. After intranasal chaIIenge, mice lacking pigR showed less nasal colonization and

decreased levels of bacteremia compared to wild-type mice (6) but, unfortunately, no

data was provided on the presence of the bacteria in the brain and or CSF. To finally

assess whether the absence of pigR significantly reduces bacterial translocation

into the brain in vivo, intravenous administration of pneumococci in pigR1· and WT

mice should be performed. Based on our present findings and previously published

observations from others, we hypothesize that pigR is involved in pneumococcal

adhesion, not only to the mucosal epithelial barrier of the respiratory tract (6), but

also to the blood-brain barrier endothelium.

In conclusion, in this study we have shown that PAFR, which facilitates

pneumococcal adhesion to brain endothelial cells, does not co-localize with S.

pneumoniae in vivo and likely has no physical interaction with the bacteria in vivo. On the other hand, S. pneumoniae does co-localize to pigR, which is expressed by the

brain vascular endothelium, and can physically interact with this receptor in vitro. We thus conclude that pigR may act as a novel receptor for S. pneumoniae adhesion to

the blood-brain barrier endothelium.

Material and Methods

Cell lines, primary cells and culture conditions

The HBMEC line (kind gift from Dr. K.S. Kim) (38) was cultivated in RPMI-

1640 (Biochrom, Berlin, Germany) supplemented with 10% Fetal Calf Serum

(PCS) (Biochrom), 10% Nu-serum (BD Biosciences, Breda, The Netherlands), 2mM

L-glutamine (Gibco, Grand Island, United States), lmM sodium pyruvate (Gibco),

87

Chapter 4

1 % MEM-vitamins (Gibco) and 1 % non-essential amino acids (Gibco ). HUVEC (obtained from Endothelial Cell Facility, University Medical Center Groningen) were cultivated in DMEM (Cellgro, Huissen, The Netherlands) supplemented with 1 % L-glu ta mine (Gibco ), 1 % non-essential amino acids (Cellgro ), 1 % heparin (Sigma) and 20% FCS (Sigma). Detroit cells (obtained from Department of Molecular Virology, University Medical Center Groningen) (39) were cultivated in RPMI-1640 (Biochrom) supplemented with 10% FCS (Biochrom), 2mM L-glutamine (Gibco), lmM sodium pyruvate (Gibco) and 1% non-essential amino acids (Gibco). A549 (obtained from Department of Molecular Virology, University Medical Center Groningen) (40) were cultivated in RPMI-1640 (Biochrom) supplemented with 10% FCS (Biochrom), 2mM L-glutamine (Gibco) and 1% non-essential amino acids (Gibco). All cells were cultured at 5% CO

2 at 37°C.


For the in vivo experiments, an encapsulated S. pneumoniae serotype 4 strain, TIGR4 (36) was used, while for the adhesion assays and the in vitro physical interaction between bacteria and receptors an unencapsulated S. pneumoniae, serotype 4 strain, TIGR4 was used. Encapsulated TIGR4 was grown in Todd-Hewitt broth (Oxoid Thermo Scientific, Basingstoke, United Kingdom), un-encapsulated TIGR4 in M17 medium supplemented with 0,5% glucose. Both strains were grown at 37°C, bacterial growth was monitored by measuring the optical density (OD) at 600 nm with a spectrophotometer. Bacteria were harvested at OD600

= 0.25-0.30 by centrifugation of 1 ml of culture at 10,000 rpm for 3 minutes. Subsequently, encapsulated TIGR4 was resuspended in 1ml of sterile phosphate-buffered saline (PBS) (Lonza, Verviers, Belgium) till reaching a challenge dose of 107 colony-forming units (CFU) for each mouse, while un-encapsulated TIGR4 was re-suspended in HBMEC cell culture medium till reaching a bacterial concentration of approximately 106 CFU.


All experiments involving animals were performed with the prior approval of and in accordance with guidelines of the Institutional Animal Care and Use Committee of the University of Groningen (DEC nr. 6152A). The bacteremia-derived meningitis model described by Orihuela et al. (36) was adapted in the following way: four groups of 5 female Balb/c mice, 6 to 8 weeks old (Harlan, Horst, The Netherlands) were anesthetized by inhalation of 2.5% isoflurane before the challenge. Intravenous tail vein injection with 200µ1 of 107 CFU of the TIGR4 wild-type was performed, and as a control 2 mice were injected with PBS (mock-infected). The mice were sacrificed at 1, 3, 8, or 14 hours after the bacterial challenge. After sacrifice, to remove unattached bacteria in the blood stream, perfusion was performed by injecting sterile PBS in the

88

Chapter 4

right ventricle via the vena cava until the blood was completely removed. Brains, lungs, and spleens were collected and stored in Shandon Cryomatrix (Thermo Scientific, Runcorn, The Netherlands) at -80°C.

Antibodies and lectin

The following antibody combinations and dilutions in sterile PBS with 5% Fetal Calf Serum (FCS) (Biochrom, Berlin, Germany) were used for immunofluorescent detection. To detect S. pneumoniae, an anti-capsule serotype 4 antibody (Statens Serum Institute, Copenhagen, Denmark) 1 :200 diluted was used in combination with an Alexa Fluor 488 goat anti-rabbit antibody (Invitrogen Life Technologies, Carlsbad, United States) 1 :500 diluted. For the detection of endothelial cells, Dy Light 594-labeled Lycopersicon Esculentum Lectin (tomato lectin) (Vector Laboratories, Burlingame, United States) was used in a 1 :200 dilution. LEL binds well to glycophorin and TammHorsfall glycoprotein and has been used effectively to label vascular endothelium in rodents (http://www.vectorlabs.com/catalog.aspx?prodID=17l5). For the detection of nuclei DAPI (Roche, Mannheim, Germany) 1 :5,000 diluted was used. For the detection of PAFR in human cells and mouse tissue, a polyclonal rabbit anti-PAFR antibody (Cayman Chemicals, Ann Arbor, United States) was used in a 1 :50 dilution. For the detection of pigR in human cells and mouse tissue, respectively, a goat anti-human pigR antibody and a goat anti-mouse pigR antibody (R & D Systems, Abingdon, United Kingdom) were used in a 1 :50 dilution. As isotype controls for the primary antibodies, rabbit IgG (Innovative Research, Plymouth, United States) and a goat IgG (Sa_nta Cruz Biotechnology, Dallas, United States) were used at the same dilution as those for specific primary antibodies. For the detection of PAFR and S. pneumoniae (both on HBMEC and brain tissue), both primary polyclonal antibodies were labeled using the Zenon Rabbit IgG Labeling Kit (Invitrogen Life Technologies, Carlsbad, United States), and the anti-PAFR antibody was labeled with Alexa Fluor 350 (blue fluorescence), while the anti-capsule serotype 4 antibody was labeled with Alexa Fluor 488 (green fluorescence). Subsequently, the labeled antibodies were diluted 1 :50. Also isotype controls were labeled with the Zenon Rabbit IgG Labeling Kit and used at the same dilution as were used for specific primary antibodies; no fluorescence was detected in all controls (data not shown). For the detection of pigR and S. pneumoniae (both on HBMEC and brain tissue) through fluorescence microscopy, a goat anti-mouse plgR antibody was used in combination with an Alexa Fluor 488 donkey anti-goat antibody (Invitrogen Life Technologies), while the anti-capsule serotype 4 antibody was labeled with Alexa Fluor 350 with the Zenon Rabbit IgG Labeling Kit. The goat IgG isotype control was used at the same dilution as used for the anti-plgR antibody in combination with an Alexa-fluor 488 donkey anti-goat antibody (Invitrogen Life Technologies); no fluorescence was detected in

89

Chapter 4

all controls (data not shown). To detect plgR and S. pneumoniae on brain tissue by

confocal microscopy, the goat anti-mouse plgR antibody was used in combination

with an Alexa Fluor 488 donkey anti-goat antibody (Invitrogen Life Technologies),

and the anti-capsule serotype 4 antibody was labeled with Alexa Fluor 488 using the

Zenon Rabbit IgG Labeling Kit.

For studies on the in vitro physical interaction between S. pneumoniae and

particular receptors, bacteria were detected with a rabbit anti-pneumococcal

antiserum (Eurogentec, Maastricht, the Netherlands), and for the detection of the

receptors the same antibodies as mentioned above were used. As negative control

for the physical interaction between S. pneumoniae and particular receptors, a mouse

anti-human a tubulin antibody (Sigma Aldrich) in combination with Alexa Fluor 488

Goat anti-Mouse were used.

For Western blotting analyses, goat anti-human plgR (R & D Systems) and an a

tubulin antibody (Sigma Aldrich) were used for the detection of respectively human

plgR and a tubulin.

Immunofluorescent detection

For immunofluorescent staining of brain and lung cells, sections of 5µm thin

were cut with a cryostat, placed on microscope glass slides and dried under a fan

for at least 30 minutes. Sections were fixed with acetone for 10 minutes and dried

for 5 minutes. For immunofluorescent staining of HBMEC, after 1 hour incubation

with S. pneumoniae, cells were fixed with 4% paraformaldehyde (Sigma Aldrich)

and washed 2 times with PBS prior starting the staining procedure.

For the detection of PAFR and S. pneumoniae in brain tissue (Table 1 ), the slides

were first incubated with an anti-PAFR antibody labeled with Alexa Fluor 350 for 1

hour in the dark at room temperature (RT). Slides were then washed in PBS 3 times

for 5 minutes. Subsequently, the slides were incubated with anti-capsule serotype 4

antibody labeled with Alexa Fluor 488 for 1 hour in the dark at room temperature

(RT). After washing with PBS 3 times for 5 minutes, the slides were incubated at RT

in the dark with DyLight 594 labeled LEL. Slides were washed 3 times with PBS for

5 minutes and incubated with DAPI (Roche, Mannheim, Germany) for 10 minutes at

RT in the dark. Lastly, the slides were washed twice for 5 minutes in PBS.

For the detection of plgR and S. pneumoniae in brain tissue (Table 1 ), the slides

were first incubated with anti-capsule serotype 4 antibody labeled with Alexa Fluor

350 for 1 hour in the dark at RT. Slides were then washed in PBS 3 times for 5 minutes.

Subsequently, the slides were incubated with anti-plgR antibody for 1 hour in the

dark at RT. After washing with PBS 3 times for 5 minutes, the slides were incubated

at RT in the dark with a mixture of DyLight 594 labeled LEL and Alexa Fluor 488

donkey anti-goat. Slides were washed 3 times with PBS for 5 minutes, incubated with

90

Chapter 4

DAPI for 10 minutes at RT in the dark, and washed again twice for 5 minutes in PBS

at RT in the dark.

For detection of pigR in lung (Table 1), slides were first incubated with anti

plgR antibody for 1 hour at RT. Slides were then washed 3 times in PBS for 5 minutes.

Slides were incubated with Alexa Fluor 488 donkey anti-goat for 1 hour at RT in

the dark, and then washed 3 times in PBS for 5 minutes. At the end, slides were

incubated with DAPI for 10 minutes at RT in the dark and washed again with PBS 2

times for 5 minutes.

For the detection of PAFR and S. pneumoniae in HBMEC and HUVEC (Table

1), cells were first incubated with anti-PAFR antibody labeled with Alexa Fluor 594

following incubation with anti-capsule serotype 4 antibody labeled with Alexa Fluor

488. For the detection of plgR and S. pneumoniae in HBMEC and HUVEC (Table 1),

cells were first incubated with a mixture of the anti-plgR and anti-capsule serotype 4

antibodies, following incubation with a mixture of Alexa Fluor 488 donkey anti-goat

and Alexa Fluor 594 goat anti-rabbit antibodies. For the immune fluorescent staining

of human cells, the same procedure applied for the tissue sections was used, the only

difference being that the PBS used to wash the cells was immediately removed.

For immunostaining on mouse tissue, the slides were washed 3 times with

PBS for 5 minutes in between each incubation step and 2 times with PBS for 5

minutes after DAPI incubation. For immunostaining on HBMEC and HUVEC, cells

were washed once with PBS for a few seconds after each incubation step. Citifluor

solution (Science Services, Munich, Germany) was added to each tissue section/glass

disk after which the coverslip was applied. The slides were analyzed with a Leica

DM5500B microscope and images were recorded with a Leica DFC 360 FX camera.

Additionally a Leica SP2 AOBS confocal microscope was used to confirm the presence

of pneumococci adhering to the brain vascular endothelium.

Image processing

All images recorded with the Leica DM5500B fluorescence microscope were

processed with ImageJ (41). The TIFF images obtained with the 350nm (blue), 488nm

(green) and 594nm (red) wavelength filters were merged using the Color-Merge Channels function. As minor manipulation, background correction was applied in

those images where the background was across the image by using the Brightness &

Contrast command of ImageJ. In all the images subjected to this minor adjustment,

the background subtraction was applied to all parts of the image. The LEI z-stacks

obtained with the confocal microscope Leica SP2 AOBS were merged through Imaris

(Bitplane Scientific Software). No digital adjustment was performed on these images.

91

Chapter 4

Co-localization analysis

Co-localization of S. pneumoniae with plgR and PAFR was analyzed with ImageJ (41). The images with the bacterial signal and plgR/PAFR were opened separately and analyzed with the ImageJ plugin Analyze/Co-localization analysis. White pixels were automatically generated on the areas of the bacterial signals co-localizing with the receptor signals. Bacteria that did not co-localize with the plgR signal remained blue and bacteria not co-localized with the PAFR remained green.

Bacterial quantification

The surface covered by bacteria was measured using the Threshold function of ImageJ, by determining the area occupied by the 488 nm bacterial signal and the white pixels of the bacteria co-localized with receptors. The percentage of the bacteria co-localized with receptors compared to the total bacterial signal was calculated by dividing the surface area of white bacteria by the surface area of the green bacteria in each image and multiplication with 100.

In vitro interaction of S. pneumoniae with receptors

Lysis buffer was prepared with 50 mM tris-HCl (Promega, Mannheim, GER) pH 7.4, 150 mM NaCl, 1 % triton 100X, 1 % sodium deoxycholate, 0.1 % SDS (all from Sigma Aldrich), 1 mM EDTA (Merck Millipore, Billerica, United States), protease inhibitors 1 X (Roche, Mannheim, Germany). 250 µl of lysis buffer was added to confluent monolayers of HBMEC or HUVEC cells grown in small T25 flasks (TPP, Trasadingen, Switzerland). After 1 minute incubation at room temperature, monolayers were scraped with a cell scraper (Nalge Nunc International, Rochester, United States) and the lysate harvested. After centrifugation at 14,000 rpm for 15 minutes at 4°C, the cell lysate in the supernatant was harvested and the pellet discarded. coomassie brilliant blue staining was used as quality control to visualize the proteins of the cell lysates separated by SDS-page. A solution of approximately 106 CFU of un-encapsulated TIGR4 was prepared in sterile phosphate buffered saline (PBS) (Lonza, Verviers, Belgium). 50 µl of HBMEC or HUVEC lysate was added to 50 µl of the bacterial solution and the mixture was incubated at 4°C with gentle agitation for 1 hour. The mixture of cell lysate and bacteria was then centrifuged at 14,000 rpm for 20 minutes at 4°C. All the supernatant was removed and the bacterial pellet was washed 2 times with 100 µl PBS. After washing, the pellet was re-suspended with 100 µl of an antipneumococcal antiserum (1:100 dilution) (Eurogentec, Maastricht, The Netherlands) directly labeled with Alexa Fluor 594 (Zenon Rabbit IgG Labeling Kit), following an incubation at 4 °C for 1 hour. After washing, the bacterial pellet was stained for plgR using an anti-human pigR antibody, followed by incubation with an Alexa Fluor Donkey anti-Goat 488 antibody. The same procedure has been adopted to check the

92

Chapter 4

physical interaction between S. pneumoniae and the PAFR, the only difference being that the anti-PAFR antibody was labeled with 594 Alexa Fluor (Zenon Rabbit IgG Labeling Kit) instead of using a secondary antibody for the detection. As a negative control for the physical interaction between S. pneumoniac and receptors, an antihuman a-tubulin antibody was used. Incubation with the anti-a-tubulin antibody was followed by incubation with an Alexa Fluor 488 Goat anti-Mouse antibody. Finally, the bacterial pellet was re-suspended in 100 µl of distilled water. A 5 µl drop was pipetted on a microscope glass slide, covered with a coverslip and analyzed by fluorescence microscopy. The slides were inspected with a Leica DM5500B microscope and the images were recorded with a Leica DFC 360 FX camera.


HBMEC and HUVEC cells were grown in 12-well plates (TPP) . When immunofluorescent staining was performed after the adherence assay, cells were grown on glass disks placed inside each well. For adherence of HUVEC on glass, disks were coated with gelatin for 45 minutes. After removal of the gelatin, disks were incubated with 0.5% glutaraldehyde (Sigma Aldrich) for 45 minutes. Disks were then washed 2 times with PBS and 2 times with HUV EC culture medium. Cells were grown to confluency at 37°C under 5% CO2• To block the PAFR or pigR receptors, cells were incubated with the respective receptor-specific antibody at a final antibody concentration of 50 µg/ml. As controls, cells were incubated with a rabbit IgG (Innovative Research, Plymouth, United States) or a goat IgG (Santa Cruz biotechnology, Heidelberg, Germany) at the same concentration as the antiPAFR or anti-pigR antibodies. As a further control, we also used cells that had not been incubated with antibody/IgG. Cells were grown to confluency in an incubator at 37°C with 5% CO2 overnight. After washing the cells with sterile PBS, 900 µl cell culture medium was added to each well and 100 µl of approximately 106 CFU of unencapsulated S. pneumoniae TIGR4 was added. After 1 hour incubation at 37°C at 5% CO2, cells were treated with a 50/50 mix of 1 % saponin and trypsin-EDTA (0.05%-0.02%) and Iysed. CPUs were determined by plating serial dilutions of lysed cells on blood agar plates. The ratio of adherent bacteria was calculated by dividing the adhered bacteria by the total amount of bacteria (adherent + non-adherent bacteria).

Western blotting procedure

Lysates of Detroit, A549 or HBMEC cells were prepared as described above for in vitro studies on the interactions between S. pneumoniae and particular receptors. Proteins of the cell lysates were separated by SDS-PAGE using NuPAGE gels (Invitrogen), and then blotted (75 min, 100 mA per gel) onto a nitrocellulose membrane (Protran®, Schleicher & Schuell, Bath, United Kingdom). Subsequently,

93

Chapter 4

human plgR was detected using a specific antibody and a- tubulin was used as loading control (see the above section on antibodies and lectin). Antibody detection was performed with fluorescent IgG secondary antibodies IRDye 800 CW donkey anti-goat (LiCor Biosciences, Bad Homburg, Germany) for human plgR detection or 700 CW goat anti-mouse antibodies for a-tubulin detection, in combination with the Odyssey Infrared Imaging System (LiCor Biosciences). Fluorescence was recorded at 700 and 800 nm.

Statistical analyses

The independent student t-test of SPSS Statistics 20 (IBM) was used for the statistical analysis of the adherence assay results. The comparison between ratio of adherent bacteria in HBMEC/HUVEC cells treated with receptor antibodies and HBMEC/HUVEC cells treated with isotype IgG or no antibody was statistically tested.

Acknowledgements

Part of the work was performed at the UMCG Microscopy and Imaging Center (UMIC). We thank Henk Moorlag from the UMCG Endothelial Cell Facility for providing HUVEC and for cryostat cutting of tissue sections. We also thank Dr. Henrik Gradstedt for providing the total anti pneumococcal antiserum and Prof. Dr. Jan Maarten van Dijl for critically reading the manuscript.

Supplementary Figures

Available upon request

94

C'1apter 4

Table 1 . Immunofluorescent detection scheme

Detection of 1st 2nd 3rd 4th

PAFR and S. p11eu11ro11iae in Anti PAFR antibody Anti capsule serotype tomato lectin DAPI brain tissue labeled with Alexa Fluor 4 antibody labeled

350 (Zenon Kit) with Alexa Fluor 488 Dilution 1 :200 1 :5000 (Zenon Kit)

Dilution 1 :50 Dilution 1 :500

Detection of 1st 2nd 3rd 4th

plgR and S.

p11e111no11iae in Anti capsule serotype 4 Anti pigR antibody Mixture tomato DAPI brain tissue labeled with Alexa Fluor lectin and

350 (Zenon Kit) Dilution 1 :50 Alexa Fluor 488 1 :5000 Donkey anti

Dilution 1 :50 Goat

Dilution 1 :500

Detection of 1st 2nd 3rd

plgR in lung tissue Anti pigR antibody Alexa Fluor 488 DAPI

Donkey anti Goat Dilu tion 1 :50 Dilution 1 :5000

Dilution 1 :500


PAFR and S. p11e11mo11iae in Anti PAFR antibody Anti capsule serotype DAPI HBMEC and labeled with Alexa Fluor 4 antibody labeled HUVEC 594 (Zenon Kit) with Alexa Fluor 488 1 :5000

(Zenon Kit) Dilution 1 :50

Dilution 1 :50


plgR and S. p11e11111011iae in Mixture of anti pigR Mixture of Alexa Fluor DAPI HBMEC and antibody (dilution 1 :50) 488 Donkey anti Goat HUVEC and anti capsule serotype and Alexa Fluor 594 Dilution 1 :5000

4 antibody (dilution 1:200) Goat anti Rabbit

Dilution 1 :200

95

Chapter 4

References 1 . Woehr! B, Klein M, Grandgirard D, Koedel U, Leib S (2011 ) Bacterial meningitis: current therapy and

possible future treatment options. Expert Rev Anti Infect Tuer 9: 1053-65.


3. Mook-Kanamori BB, Geldhoff M, van der Poll T, van de Beek D (2011) Pathogenesis and pathophysiology of pneumococcal meningitis. Clin Microbiol Rev 24: 557-91 .

4 . Bogaert D, De Groot R, Hermans PW (2004) Streptococcus pneumoniac colonization: the key to pneumococcal disease. Lancet Infect Dis 4: 144-54.

5. Ring A, Weiser JN, Tuomanen EI (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 1 02: 347-60.

6. Zhang JR, Mostov KE, Lamm ME, Nanno M, Shimida S, Ohwaki M, Tuomanen E (2000) The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102: 827-37.

7. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2005) Beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pneumoniac. Infect Immun 73: 7827-35.

8. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1995) Streptococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377: 435-8.

9. Honda Z, Ishii S, Shimizu T (2002) Platelet-activating factor receptor. J Biochem 131: 773-9.

10. Stoll LL, Denning GM, Kasner NA, Hunninghake GW (1994) Platelet-activating factor may stimulate both receptor-dependent and receptor-independent increases in [Ca2+) in human airway epithelial cells. J Biol Chem 269: 4254-9.

1 1 . Shirasaki H, Nishikawa M, Adcock IM, Mak JC, Sakamoto T, et al. (1994) Expression of plateletactivating factor receptor mRNA in human and guinea pig lung. Am J Respir Cell Mo! Biol 10: 533-7.

12. Chao W, Olson MS (1 993) Platelet-activating factor: receptors and signal transduction. Biochem J 292: 617-29.

13. Shukla SD (1992) Platelet-activating factor receptor and signal transduction mechanisms. FASEB J 6: 2296-301 .

14. Snyder F (1990) Platelet-activating factor and related acetylated lipids as potent biologically active cellular mediators. Am J Physiol 259: C697-708.

15. Iovino F, Brouwer MC, van de Beek D, Molema G, Bijlsma JJ (2013) Signalling or binding: the role of the platelet activating factor receptor in invasive pneumococcal disease. Cell Microbiol 15:870-81

16. Johansen FE, Natvig Norderhaug I, R0e M, Sandlie I, Brandtzaeg P (1999) Recombinant expression of polymeric IgA: incorporation of J chain and secretory component of human origin. Eur J Immunol 29: 1701-8.

17. Shimad S, Kawaguchi-Miyashita M, Kushiro A, Sato T, Nanno M, et al. (1999) Generation of polymeric immunoglobulin receptor-deficient mouse with marked reduction of secretory IgA. J Immunol 163: 5367-73.

1 8. Kim KJ, Malik AB (2003) Protein transport across the lung epithelial barrier. Am J Physiol Lung Cell Mo! Physiol 284: L247-59.

19. Asano M, Komiyama K (201 1) Polymeric immunoglobulin receptor. J Oral Sci 53: 147-56.

20. Lu L, Lamm ME, Li H, Corthesy B, Zhang JR (2003) The human polymeric immunoglobulin receptor binds to Streptococcus pncumoniac via domains 3 and 4. J Biol Chem 278: 48178-87.

21 . Ikonen E, Parton RG, Hunziker W, Simons K, Dotti CG (1993) Transcytosis of the polymeric immunoglobulin receptor in cultured hippocampal neurons. Curr Biol 3: 635-44.

22. de Hoop M, von Poser C, Lange C, Ikonen E, Hunziker W, Dotti CG (1995) Intracellular routing of wild-type and mutated polymeric immunoglobulin receptor in hippocampal neurons in culture. J Cell Biol 130: 1447-59.

96

Chapter 4

23. Hemar A, Olivo JC, Williamson E, Saffrich R, Dotti CG (1 997) Dendroaxonal transcytosis of transferrin in cultured hippocampal and sympathetic neurons. J Neurosci 1 7: 9026-34.

24. Rijneveld, A.W., Weijer, S., Florquin, S., Speelman, P., Shimizu, T., Ishii, S., et al. (2004) Improved host defense against pneumococcal pneumonia in platelet-activating factor receptor-deficient mice. J Infect Dis 189:71 1 -6.

25. Bito H, Honda Z, Nakamura M, Shimizu T (1994) Cloning, espression and tissue distribution of rat platelet-activating factor-receptor cDNA. Eur J Biochem 221 : 21 1 -8.

26. Ishii S, Shimizu T (2000) Platelet-activating factor (PAF) receptor and genetically engineered PAF receptor mutant mice. Prog Lipid Res 39: 41 -82.

27. Jen FE, Warren MJ, Schulz BL, Power PM, Swords WE, et al. (2013) Dual Pili Post-translational Modifications Synergize to Mediate Meningococcal Adherence to Platelet Activating Factor Receptor on Human Airway Cells. PLoS Pathog 295: e1 003377.

28. Thornton JA, Durick-Eder K, Tuomanen EI (2010) Pneumococcal pathogenesis: "innate invasion" yet organ-specific damage. J Mol Med (Berl) 88: 103-7.

29. Filion S, Soulis K, Rajasekaran S, Benedict-Hamilton H, Radin JN, et al. (2006) Plateletactivating factor receptor and innate immunity: uptake of gram positive bacterial cell wall into host cells and cell-specific pathophysiology. J Immunol 1 77:61 82-91

30. Agarwal V, Hammerschmidt S (2009) Cdc42 and the phosphatidylinositol 3-kinase-Akt pathway are essential for PspC-mediated internalization of pneumococci by respiratory epithelial cells. J Biol Chem 284: 1 9427-36.

31. Agarwal V, Asmat TM, Oierdorf NI, Hauck CR, Hammerschmidt S (2010) Polymeric immunoglobulin receptor- mediated invasion of Streptococcus p11e11111011iae into host cells requires acoordinate signaling of SRC family of protein-tyrosine kinases, ERK, and c-Jun N-terminal kinase. J Biol Chem 285: 35615-23.

32. Elm C, Braathen R, Bergmann S, Frank R, Vaerman JP, et al. (2004) Ectodomains 3 and 4 of human human polymeric Immunoglobulin receptor (hpigR) mediate invasion of Streptococcus p11e11mo11iae into the epithelium. J Biol Chem 279: 6296-304.

33. Ghetie V, Ward ES ( 1997) FcRn: the MHC class I-related receptor that is more than an IgG transporter. Immunol Today 18: 592-8.

34. Borvak J, Richardson J, Medesan C, Antohe F, Radu C, et al. (1 998) Functional expression of the MHC class I-related receptor, FcRn, in endothelial cells of mice. Int Immunol 10: 1 289-98.

35. Fi ran M, Bawdon R, Radu C, Ober RJ, Eaken D, et al . (2001 ) The MHC class I-related receptor, FcRn, plays an essential role in the maternofetal transfer of gamma-globulin in humans. Int Immunol 1 3: 993-1002.

36. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, et al. (2009) Laminin receptorjnitiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest 1 1 9: 1638-46.

37. Orihuela CJ, Gao G, Francis KP, Yu J, Tuomanen EI (2004) Tissue-specific contributions of pneumococcal virulence factors to pathogenesis. J Infect Dis 1 90: 1661-9.

38. Stins MF, Badger J, Sik Kim K (2001 ) Bacterial invasion and transcytosis in transfected human brain microvascular endothelial cells. Microb Pathog 30: 1 9-28.

39. Romero-Steiner S, Caba J, Rajam G, Langley T, Floyd A, et al. (2006) Adherence of recombinant pneumococcal surface adhesion A (rPsaA) coated particles to human nasopharyngeal epithelial cells for the evaluation of anti-PsaA functional antibodies. Vaccine 24: 3224-31 .

40. Giard DJ, Aaronson SA, Todaro GJ, Amstein P, Kersey JH, et al. (1973) In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors. J Natl Cancer Inst 51 : 1417-23.

41 . Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9: 671-5.

97

CHAPTER S

Platelet endothelial cell adhesion molecule-I (PECAM-1), a putative receptor for the adhesion of Streptococcus pneumoniae to the vascular

endothelium of the blood-brain barrier

Federico Iovino, Grietje Molema, Jetta J.E. Bijlsma


99

Abstract

Bacterial meningitis is a life threatening infectious disease leading to cerebral dysfunction and consequently death. The Gram-positive bacterium Streptococcus

pneumoniae is the main causative agent of this disease. S. pneumoniae is thought to invade the Central Nervous System via the bloodstream by crossing the vascular endothelium of the blood-brain barrier, a specialized system of endothelial cells that protects the brain from harmful substances that are present in the bloodstream and supplies the brain with the required nutrients for its proper function. The exact mechanism by which pneumococci cross endothelial cell barriers before meningitis develops is unknown. Here we investigated the role of PECAM-l/CD31, one of the major endothelial cell adhesion molecules, in S. pneumoniae adhesion to vascular endothelium of the blood-brain barrier. Mice were intravenously infected with pneumococci and sacrificed at various time points to mimic the stages preceding meningitis. Immunofluorescent analysis of brain tissue of infected mice showed that pneumococci co-localized with PECAM-1. Also in Human Brain Microvascular Endothelial Cells (HBMEC) incubated with S. pneumoniae we observed a clear colocalization between PECAM-1 and pneumococci. Blocking of PECAM-1 reduced the adhesion of S. pneumoniae to endothelial cells in vitro, implying that PECAM-1 is involved in pneumococcal adhesion to endothelial cells. Furthermore, we demonstrate using endothelial protein lysates that S. pneumoniae physically binds to PECAM-1 in vitro. In vitro we also show that PECAM-1 co-localizes with the S. pneumoniae adhesion receptor plgR, which we previously demonstrated to be present on endothelial cells. In brain tissue of infected mice S. pneumoniae was associated with both receptors. Lastly, immunoprecipitation experiments revealed that PECAM-1 can physically interact with plgR. In summary, we show for the first time that bloodborne S. pneumoniae can bind to PECAM-1 expressed on the brain microvascular endothelium and that PECAM-1 can interact with plgR. We hypothesize that this interaction may play a major role in pneumococcal binding to the blood-brain barrier vasculature prior to invasion into the brain.

Chapter 5

Introduction

Streptococcus pneumoniae (the pneumococcus) is a Gram-positive bacterial pathogen that causes life-threatening invasive diseases in humans, such as pneumonia and bacteremia. Every year, over a million people worldwide succumb to diseases caused by S. pneumoniae (1). Moreover, this bacterium is the most common causative agent of bacterial meningitis, an inflammation of the protective membranes covering the brain and spinal cord, collectively known as the meninges (2, 3). S. pneumoniae

is thought to invade the brain mainly via the bloodstream by crossing the vascular endothelium of the blood-brain barrier, a specialized system of endothelial cells that protects the brain from harmful substances that are present in the bloodstream and supplies the brain with the required nutrients for its proper function (2, 3) .

Meningeal pathogens, such as S. pneumoniae and Neisseria meningitidis, bind to receptors expressed on the plasma membrane of epithelial and endothelial cells and, through this binding, they can invade and translocate over human cell layers (4- 6). The platelet-activating factor receptor (PAFR) has been described as one of the major receptors involved in the interaction between S. pneumoniae and endothelial cells (7-9). Blocking of PAFR leads to a significant reduction of pneumococcal adhesion to endothelial cells in vitro, and the complete absence of PAFR leads to less invasive pneumococcal disease in mouse models ( 4, 7). However, in Chapter 4 of this thesis, we showed data suggesting that physical interaction between S. pneumoniae and PAFR is not likely to occur as we did not observe co-localization between the receptor and the bacteria in the brain tissue of intravenously infected mice. Orihuela et al. showed that the laminin receptor initiates the contact of S. pneumoniae to the brain vascular endothelium in vivo (10). The poly immunoglobulin receptor (plgR) was furthermore shown to mediate binding of pneumococci to the epithelium of the upper respiratory tract (5) and in Chapter 4 we showed that plgR expressed by brain endothelial cells is capable of binding S. pneumoniae. It is unclear at the moment whether more receptors are involved in pneumococcal adherence to the blood brain endothelium.

The platelet endothelial cell adhesion molecule-1 (PECAM-1; also known as CD31}, is a pan-endothelial protein that is present in the intercellular junctions of the endothelial cells (11, 12). PECAM-1 is involved in leukocyte migration, angiogenesis, and integrin activation (11, 12). In particular, the involvement of PECAM-1 in leukocyteendothelium interaction and leukocyte trans-endothelial migration in the vascular system makes PECAM-1 a key molecule in inflammation and neuroinflammation (13, 14). Recently PECAM-1 was implicated in Salmonella typhimurium infections (15). This raised the question whether PECAM-1 might also play a role in host-pneumococcal interactions. Accordingly, the specific aim of the present study was to investigate whether PECAM-1 plays a role in S. pneumoniae adhesion to the vascular endothelium of the blood-brain barrier. To study this, Balb/c mice were intravenously infected

101

Chapter 5

with S. pneumoniae strain TIGR4 and sacrificed at different time-points to mimic the stages preceding meningitis (16). In brief, the present immunofluorescent and confocal microscopic analyses of brain sections in combination with in vitro blocking experiments and immunoprecipitation show that PECAM-1 may well be a novel adhesion receptor for pneumococci that possibly exerts this action in conjunction with plgR. Thus, novel insights in the receptors that S. pneumoniae can employ to adhere to the vascular endothelium of the blood-brain barrier are presented.

Results

S. pneumoniae co-localizes with PECAM-1 expressed by the vascular endothelium of the

blood-brain barrier in vivo and by HBMEC in vitro

Previously we have shown that, as soon as 1 hour after infection, pneumococci adhere to the blood-brain endothelium. However, there are significant spatiotemporal differences in the invasion of the brain. Most noticeable, the bacteria occurred in the choroid plexus only after 8 hours post infection (16) . It was therefore of interest to investigate the possible co-localization of S. pneumoniae with PECAM-1 at all time points during infection. Unexpectedly, our present immunofluorescent analyses of brain tissue from the infected mice revealed that, at all time points of infection, most adherent bacteria co-localized with PECAM-1 (Figures lA,B and SlA,B). In fact, the assessment of the bacterial and PECAM-1 signals with ImageJ showed that > 95% of the pneumococci were co-localized with PECAM-1 in all brain compartments at all time-points of infection (Figures lA,B and SlA,B). This observation was subsequently verified by confocal microscopy, which confirmed that S. pneumoniae was indeed co-localized with PECAM-1 (Figure S2 and movies in Supporting Information). In

vitro, co-localization analyses also showed that most pneumococci co-localized with PECAM-1 in HBMEC cells (Figure 2). Our immunofluorescent analyses furthermore showed that in brain tissue of mock-treated healthy mice the PECAM-1 signal was weak and patchy (Figure 3A). Interestingly, the PECAM-1 signal increased over the time course of infection and, in particular, it seemed that this increase of the PECAM-1 signal was detectable in association with pneumococci or nearby areas where the bacteria were present (Figure 3B).

To confirm that the PECAM-1 protein levels increased upon the presence of pneumococci and not because of inflammation caused by bacterial infection, we immunofluorescently stained PECAM-1 in lung tissue. We previously showed that intravenous injection of S. pneumoniae TIGR4 causes a high influx of leukocytes in the lungs right from 1 to 14 hours post challenge, which is a clear sign of inflammation (16). No differences were, however, observed in PECAM-1 signal between lungs of mock-treated mice and lungs of infected mice up to 8 hours post challenge (Figure 4A). Bacteria were occasionally detected in the lungs only at 14 hours post challenge

102

Chapter 5

(Figure 4B), and this was paralleled by a more intense PECAM-1 fluorescent signal nearby the areas where bacteria were visible than in areas where no bacteria were detected (Figure 4B). Together, these results suggest that the presence of pneumococci induces PECAM-1 protein expression in the endothelium.

A

Subarachno1d space

Cerebral cortex

Septum

B

Subarachno1d space

Cerebral cortex

Septum

Choroid plexus

PECAM-1 + DAPI

PECAM-1 + DAPI

1 hour Infection S pneumon,se

14 hours lnfecdon

S pneumomae

Overlay

Overlay

Figure 1. In vivo co-localization of S. pneumoniae and PECAM-1 on brain

endothelium Brain slides of mice systemically infected with S. pneumoniae were immunofluorescently stained for PECAM-1 (red), pneumococci (green), and cellular nuclei (blue). The images are representative for the situation in each brain compartment at 1 hour (A) and 14 hours (B) post bacterial challenge. At 1 hour no pneumococci were detected in the choroid plexus. For each time point of infection, 6 brain sections from 3 mice in each group were analysed. Total magnification of subarachnoid space, septum and choroid plexus 630X; total magnification of cerebral cortex lOOOX.

103

Chapter 5

Figure 2. Co-localization of S. pneumoniae and PECAM-1 in HBMEC Immunofluorescent detection of S. pneumoniae (green}, PECAM-1 (red) and cellular nuclei (blue) in HBMEC after 1 hour incubation with pneumococci.

B

1 hour

3 hours

8 hOIJfS

14 hours

Tomato l•ctm

s

Tc-mato '&ct1n

Mock

c ·

PEC.Af,l-1

S. pneumon;ae Infection PECAl,1-1

-'c

s

s '

.

C Figure 3. Increased PECAM-1 signal in the presence of pneumococci in the brain hnmunofluorescent detection of tomato lectin (red}, PECAM-1 (green} in the brain of mock-treated mice (A) and S. pneumonia-infected mice at different time points post infection (8). Total magnification SOX Brains from 2 mock-treated and 3 S. pneumoniae-infected mice were analyzed, and of each mouse 3 brain sections were analyzed. The images are representative of the situation in the brains of healthy mock-treated mice, and at each time-point of infection with S. pneumoniae among all mice that have been analyzed.

104

Chapter 5 A B

Mock 1 hour 3 hours 8 hours

14 hours

PECAM-1 + DAPI S pneumonrae

Figure 4. Specific induction of PECAM-1 in the presence of pneumococci in the

lungs Immunofluorescent detection of PECAM-1 (red) and cellular nuclei (blue) in lungs of (A) mock-treated mice, (B) during systemic infection with S. pneumoniae up to 8 hours post challenge, and (C) in lungs of infected mice at 14 hours post challenge. As expected, no pneumococci were detected in lungs of mock-treated mice up to 8 hours post infection (data not shown). Bacteria were occasionally detected in the lungs at 14 hours post challenge Total magnification 400X. Lungs from 3 infected mice were analyzed, and of each mouse 3 lung sections were stained, the images showing representative outcomes of the situation in the lungs at each time point and condition.

Blocking of PECAM-1 reduces adhesion of S. pneumoniae to endothelial cells The immunofluorescent analysis showed co-localization between pneumococci

and PECAM-1 on brain endothelium in vivo and on brain endothelial cells in vitro. Therefore, the next step was to investigate whether PECAM-1 is functionally

involved in pneumococcal adhesion to human endothelial cells. Indeed, in HBMEC

and Human Umbilical Vein Endothelial Cells (HUVEC) treated with anti-PECAM-1

antibody prior to incubation with pneumococci, the bacterial adhesion was - 45-50%

reduced compared to control conditions (Figures SA,B). This shows that PECAM-1 is

involved in the adherence of S. pneumoniae to human endothelial cells in vitro.

105

Chapter 5

A 8 2.stf.-01 •OJE<i

'""'" 2((£-GI

),:.X� ' . J!(f..(1

,.,.., HllC<2

1C(£.(1 1 !4: (!7

'""'" ICllCG!

,0£.,

Figure 5. Blocking of PECAM-1 significantly reduces pneumococcal adhesion to human endothelial cells Blocking of PECAM-1 with anti-PECAM-1 antibody (white bars) led to a significant reduction of pneumococcal adhesion to HBMEC (A) and HUVEC (B) compared to isotype IgG treahnent (hatched bars) and untreated HBMEC/HUVEC (black bars). Data represent averages of triplicate repeats per condition plus standard deviations. The Y- axis shows the ratio of adherent bacteria. The ratio of adherent bacteria was calculated by dividing the adherent bacteria by the total amount of bacteria in the well (adherent + not adherent bacteria). * P value < 0.05 as determined by the students t-test. Assays have been repeated three times in each endothelial cell type.

Physical interaction between S. pneumoniae and PECAM-1 from human endothelial cell

lysates

After the observation that blocking of PECAM-1 with a specific antibody leads to a significant reduction of pneumococcal adhesion to endothelial cells, we investigated whether S. pneumoniae can physically interact with PECAM-1 present in lysates of endothelial cells. To this end, the bacteria were incubated with the HBMEC or HUVEC lysates as described in chapter 4 of this thesis. Briefly, the bacteria were incubated with a lysate of cultured cells and collected by centrifugation. The supernatant was discarded from the bacterial pellet and, after extensive washing steps, the bacteria were probed for the presence of bound eukaryotic proteins from the cell lysate using immuno-fluorecence. Indeed, a specific PECAM-1 signal was detected on the surface of pneumococci incubated with HBMEC or HUVEC cells (Figure 6A), which indicates that the PECAM-1 protein binds to pneumococci. In contrast, the control protein a-tubulin was not detected on the pneumococcal cells (Figure 6B), indicating that the interaction was specific for PECAM-1 and that endothelial proteins like tubulin did not generally become associated with the bacteria through non-specific interactions. Importantly, these results unequivocally demonstrate that S. pneumoniae can physically interact with the PECAM-1 protein as present in lysates of HBMEC and HUVEC cells.

106

A.

HBMEC

HUVEC

B.

S pnoumon,ae PECAM-1

S pneumoroN - a tubulln over1ay

HBMEC ov-.iay HBMEC Entarged overlay

Chapter 5

Overtay Enlarged over1ay

HUVEC overtay HWEC Enlarged overlay

Figure 6. S. pneumoniae bound to PECAM-1 present in HBMEC and HUVEC

lysates (A) Immunofluorescent detection of S. pneumoniae (green) and PECAM-1 (red) after pneumococci were incubated with HBMEC and HUVEC lysates. Total magnification 400X. The co-localization of the fluorescence in the overlay (yellow) indicates that PECAM-1 signal is associated with the pneumococci. Parts of the overlay images (within white rectangles) are enlarged and shown on the right. (B) Immunofluorescent detection of S. pneumoniae (green) and a-tubulin (red) after pneumococci were incubated with HBMEC and HUVEC lysates. Total magnification 400X. The overlay indicates that the a-tubulin signal is not detectable on the pneumococci. Parts of the overlay images (within white rectangles) are enlarged and shown on the right. PECAM-1 interacts with plgR, which may lead to the formation of a double receptor for the adhesion of S. pneumoniae

Our previous studies have shown that S. pneumoniae binds the plgR receptor on brain endothelial cells and that this binding is important for pneumococcal adherence to such cells (chapter 4 of this thesis). In the light of our present results, this would suggest that plgR and PECAM-1 could be co-localized. We therefore investigated the possible co-localization of these two receptors. Indeed, confocal imaging showed that PECAM-1 and plgR are frequently co-localized in HBMEC and HUVEC in the absence of infecting bacteria (Figures 7 A,B). Furthermore, after incubation with S. pneumoniae for 1 hour, pneumococci adherent to endothelial cells co-localized with both PECAM-1 and plgR at the same time (Figures 7C,D). Immunofluorescent analysis of brain tissues from healthy mice showed that PECAM-1 and plgR were frequently co-localized in all brain compartments. Analysis of brains from infected mice also showed that these two receptors were frequently co-localized (Figures SA-C and S3A,B) and, on top of this, most S. pneumoniae co-localized with both PECAM-1

107

Chapter 5

and plgR (Figures 8B,C and S3A,B). These observations strongly support the view that PECAM-1 and plgR are co-localized and may both have a role of pneumococcal adherence to endothelial cells.

To determine whether PECAM-1 and plgR interact with each other, we performed immunoprecipitation experiments in which we pulled down one protein from HBMEC or HUVEC lysates and detected the other one by Western blotting, or vice versa, As a positive control f-actin was used, which is known to interact with PECAM-1 (17). As a negative control p53 was used, since there is no evidence in the literature for a direct interaction between p53 and PECAM-1 (Figure 9A). Indeed, PECAM-1 and plgR were co-immunoprecipitated with either antibody, which demonstrated that PECAM-1 and plgR can physically interact in HBMEC and HlNEC lysates (Figures 9A,B). Lastly, preliminary data indicate that the adhesion of S. pneumoniae to HBMEC and HUVEC cells was more strongly impaired when PECAM-1 and plgR were simultaneously blocked with specific antibodies than when either PECAM-1 or plgR were blocked individually (Figures lOA,B). This suggests that PECAM-1 and plgR could cooperatively facilitate the pneumococcal adhesion to endothelial cells .

..

C

nc,.u.,

I ' "

D

PECAtA-1

108

!..-··· . �- - . - .. · ..

!,· .. ' -· . .

Figure 7. Endothelial

PECAM-1 and plgR

are co-localized in vitro

and S. pneumoniae is co-localized with

PECAM-1 close to plgR Confocal imaging of PECAM-1 and plgR in HBMEC (A) and HUVEC (B) in the absence of bacteria, and in HBMEC (C) and HUVEC (D) after a 1 hour incubation with pneumococci. The overlay shows PECAM-1 colocalized with plgR. For each cell type, 10 cells have been randomly selected for imaging. The images are representative of the situation in the monolayer, each image shows one cell.

A

Subarachnotd space

Cerabral COttH

C PECAIA-1

space

COf;H

.. �. B

ili>arachnotd spa::e

PECAM-1

14 hours lnftd:lon p1gR PECAM-h·plgR S ?')1tuma11H PECAM-1

Seplum

Choroid pllXUI

Chapter 5

plgll

PECAM-hP'QR S P,lt\lfflOl/H

Figure 8. In vivo PECAM-1 and plgR co-localization on brain microvascular

endothelium and preferential co-localization of S. pneumoniae with plgR and

PECAM-1 at the same time Immunofluorescent detection of PECAM-1 and plgR in brain tissue of mock-treated mice (A) and of PECAM-1, plgR and S. pneumoniae in brain tissue of infected mice at 1 hour (B) and 14 hours (C) post challenge. Total magnification of subarachnoid space, septum and choroid plexus was 630X; total magnification of cerebral cortex was lO00X. Brains from 2 mock-treated and 3 intravenously infected mice were analyzed, and of each mouse 3 brain sections were used for fluorescence staining. The images are representative of the situation in the brains of healthy mock-treated mice, and at 1 and 14 hours post challenge with S. pneumoniae.

A PuH down: anti-PECAM-1

B PuH down: anti- plgR

HBMEC lysate HUVEC lysate HBMEC lysate HUVEC tysate dotacllon dotatllon �---�,----�

��=,� -�, 1 r ail ��= I ••• 11 l,aJPt;:.I

.______,I I C

dttactlon

Pull down: Mou11 lgG

HBMEC lysate HUVEClysete

I I �� =1 =====,,.--�, =====

Figure 9. PECAM-1 interacts with plgR in HBMEC and HUVEC cell lysates Immunoprecipitation of either PECAM-1 or plgR followed by Western blot analysis of various proteins showed that PECAM-1 interacts with plgR present in HBMEC and HUVEC lysates (A) as well as with f-actin, while no interaction with p53 is detectable (B). Immunoprecipitation of plgR in HBMEC and HUVEC lysates followed by detection of PECAM-1. Control immunoprecipitations with mouse IgG-coated beads do not show a PECAM-1 signal (C).

109

Chapter 5

A

14(PC � 8 1:?0% <-- ---------

� lO�C

j so•; � ! 60'C p i 40•,

8

FECAl,1 l+plgR PED-'.1�1 ptqR

ti

12�

10l%

BO!o

40%

20%

0%

FEC�I.I 1+pl9R PE.C-'-1.f-1

Figure 10. Blocking of PECAM-1 and plgR reduces the pneumococcal adhesion to human endothelial cells to lower levels than blocking of the individual receptors Simultaneous blockong of PECAM-1 and plgR reduces the pneumococcal adherence to H BMEC (A) and HUVEC (B) to a greater extent than blocking of the individual receptors. White bars represent the percentage of adhesion in the presence of anti-receptor antibodies, hatched bars in the presence of antibody isotype controls, and black bars in the absence of added antibodies (medium incubation). The y-axis shows the percentage of the adhesion ratio where the antibody isotype control adhesion ratio was set at 100%. Data sets obtained for adhesion in the presence of individual or a mix of both receptor antibodies were obtained independently; each set of experiments was repeated three times for each endothelial cell type.

Discussion

The aim of this study was to investigate whether S. pneumoniae has the capacity to bind to the PECAM-1 protein expressed by the vascular endothelium of the blood-brain barrier. To study this, we mimicked the events preceding meningitis by intravenously infecting mice with pneumococci and sacrificing them at various time points after challenge (16). In addition, we performed in vitro bacteria-endothelial adhesion assays. In the course of these analyses, we have demonstrated that S. pneumoniae co-localizes with PECAM-1 in vivo and in vitro, and that PECAM-1 blocking with a specific antibody significantly reduces adhesion of S. pneumoniae to endothelial cells in vitro. The idea that PECAM-1 has a role in pneumococcal adhesion to endothelial cells is strongly supported by immunoprecipitation experiments, which show that S. pneumoniae binds the PECAM-1 present in HBMEC and HUVEC cell lysates. Importantly, PECAM-1 and a second receptor on brain endothelial cells, the poly immunoglobulin receptor plgR, both interact with S. pneumoniae in

vitro, and are co-localized at the sites of bacterial adhesion in vivo and in vitro. Coimmunoprecipitation studies show that PECAM-1 and plgR are non-covalently bound to each other in endothelial cell extracts. Altogether, these observations imply that the interacting PECAM-1 and plgR proteins mediate the adhesion of S. pneumoniae to the blood-brain barrier endothelium.

PECAM-1 is involved in leukocyte migration during neuroinflammation by mediating the interaction of leukocytes with the vascular endothelium. HIV infection

110

Chapter 5

leads to enhanced transmigration of HIV-infected leukocytes into the central nervous

system (CNS) and changes in the blood-brain barrier permeability through alterations

in PECAM-1 interactions between endothelial cells and between leukocytes and

endothelial cells (18). Recently, we showed that pneumococcal interaction with the

blood-brain barrier endothelium did not cause a major disruption of endothelial

integrity (16). Furthermore, we also showed that influx of leukocytes in the brain was

low up to 14 hours post intravenous challenge (16). Thus, while PECAM-1 alterations

in HIV infection lead to leukocyte recruitment into the CNS and consequent alterations

in blood-brain permeability, interaction of blood-borne S. pneumoniae with PECAM-1

on the brain microvascular endothelium does neither seem to cause a serious influx

of leukocytes nor a major disruption in blood-brain barrier integrity.

Relatively little is known about PECAM-1 as a putative a receptor for pathogens.

PECAM-J-1- mice were more resistant to gastrointestinal infection with Salmonella typhimurium than wild-type mice, resulting in lower bacterial loads in systemic organs,

such as the liver and spleen (15). Whole bacterial cell ELISA experiments revealed

that human and murine PECAM-1 can bind to S. typhimurium, which indicates that

PECAM-1 can bind to bacteria and that this may have pathogenic consequences (15).

In our study, S. pneumoniae is clearly shown to co-localize with PECAM-1 both in vitro and in vivo, and incubation of S. pneumoniae with endothelial cell lysates leads

to PECAM-1 association with the bacteria. This implies that PECAM-1 may also

serve as a receptor for S. pneumoniae which, in tum, may have consequences for the

development of invasive pneumococcal disease (IPD).

As shown by our immunofluorescence detection method the PECAM-1 signal in

the healthy mouse brain is relatively weak (Figure 3A). However, upon pneumococcal

infection, more cells display a fluorescent PECAM-1 signal, in particular when S.

pneumoniae is attached to the cells (Figure 3B). Using lung tissues of the same mice,

we showed that inflammation per se may not be enough to induce a PECAM-1 signal,

but that the presence of the bacteria is needed. This implies that the induction of this

receptor may be a direct consequence of bacterial binding to the receptor. Additional

experiments will be needed to further establish whether such a causal role between

bacterial binding and (transcriptional or post-transcriptional) control of PECAM-1

expression exists.

After intravenous injection of 107 colony forming units (CFU) S. pneumoniae TIGR4, the systemic numbers of bacteria were reduced by 10 fold 1 hour after the

challenge (16), but as the disease progressed, the recoverable CFU in the blood

increased. A significant drop in CFU in the blood after intravenous administration

of pneumococci was previously observed also in a rabbit model (19). PECAM-1 is

present in the endothelial cells of the entire vascular system, so it may be possible

that this apparent 10-fold drop in CFU in the blood within 1 hour is the result of

an immediate attachment of bacteria to PECAM-1 in the blood vessels in specific

111

Chapter 5

compartments of the body, such as the heart and kidney vasculature where

PECAM-1 is highly expressed (20), which would sequester the bacteria. In principle,

the immediate attachment of bacteria to the vasculatures of organs expressing high

levels of PECAM-1 may lead to bacterial invasion into these organs. However, S.

pneumoniae is known to mainly cause pneumonia, bacteremia, otitis media and

meningitis. It would, therefore, be interesting to investigate whether blood-borne

pneumococci are also able to translocate over the endothelium of organs with a high

PECAM-1 expression, such as the heart and kidney.

Receptor-mediated adhesion to human cells is a mechanism described also

in the meningeal invasion by N. meningitidis. It was reported that Type IV-pili of

meningococci adhere first to a receptor, possibly PAFR (21), and afterwards to the

/3-adrenoceptor as expressed on the cell membrane of HBMEC. These two receptors

together allow meningococci to adhere to and traverse the endothelial cells leading to

invasion of the brain ( 6, 21 ). A similar receptor cooperation concept may also apply to

S. pneumoniae adhesion to and invasion of endothelial cells of the blood-brain barrier.

The results presented in Chapter 4 of this thesis show that S. pneumoniae co-localizes

with endothelial plgR in all brain compartments at all time-points of infection. Here

we show that plgR and PECAM-1 co-localize in the healthy murine brain and in

the presence of infecting bacteria. Furthermore, preliminary data indicate that the

blocking of both receptors with specific antibodies reduces pneumococcal adhesion

approximately 20% more than blocking each receptor separately. Moreover, the

present immunoprecipitation analyses show that PECAM-1 and plgR can interact

with each other. Thus, it may well be that these two proteins form a double receptor on

the plasma membrane of endothelial cells, which may have a functional consequence

for bacterial adhesion to the endothelium.

The hypothetical double receptor PECAM-1-plgR might provide two

independent binding sites for the bacteria. Thus, when only PECAM-1 is blocked, S.

pneumoniae can still bind to plgR and vice versa. This could explain why blocking both

receptors reduced pneumococcal adherence more than blocking only one receptor.

To unequivocally demonstrate a role for these receptors per se in S. pneumoniae pathology, intravenous challenges with S. pneumoniae in PECAM-1 + mice and plgR-1-

mice should be performed. In case of a direct role, fewer bacteria would be expected

to translocate the blood-brain barrier, and accordingly meningitis symptoms should

be reduced. Lastly, systemic challenges in mice lacking both PECAM-1 and plgR

should be performed to determine whether pneumococcal adhesion to the brain

endothelium and disease symptoms are diminished to a greater extent in the double

mutant mice than in mice lacking only one receptor. This would establish whether

and, if so, to what extent the two receptors cooperate, and whether this has functional

and pathological consequences.

112

Chapter 5

In conclusion, based on the data presented, we propose that S. pneumoniae in the blood stream can bind to PECAM-1 expressed by brain microvascular endothelial cells, and that PECAM-1 may form a double receptor with pigR on the plasma membrane of endothelial cells for S. pneumoniae adhesion. Additional experiments need to be executed to shed light on the functional role of both receptors in facilitating S. pneumoniae invasion into the brain in vivo, and to answer the question whether their blockade may have any therapeutic advantage for patients with bacteremia and meningitis.

Materials and Methods

Cell lines, primary cells and culture conditions

The HBMEC cell line (kindly provided by Dr. K.S. Kim) (22) was cultivated in RPMI-1640 (Biochrom, Berlin, Germany) supplemented with 10% Fetal Calf Serum (FCS) (Biochrom), 10% Nu-serum (BD Biosciences, Breda, The Netherlands), 2mM L-glutamine (Gibco, Grand Island, United States), lmM sodium pyruvate (Gibco), 1% MEM-vitamines (Gibco) and 1% non-essential amino acids (Gibco). HUVEC (obtained from the Endothelial Cell Facility, University Medical Center Groningen) were cultivated in DMEM (Cellgro, Huissen, The Netherlands) supplemented with 1 % L-glutamine (Gibco), 1 % non-essential amino acids (Cellgro), 1 % heparin (Sigma) and 20% FCS (Sigma). Detroit cells (obtained from the Department of Molecular Virology, University Medical Center Groningen) (23) were cultivated in RPMI-1640 (Biochrom) supplemented with 10% FCS (Biochrom), 2mM L-glutamine (Gibco), lmM sodium pyruvate (Gibco) and 1% non-essential amino acids (Gibco). All cells were cultured at 5% CO2 at 37°C


For the in vivo experiments an encapsulated S. pneumoniae, serotype 4 strain, TIGR4 (24, and Chapter 4) was used. For the adhesion assays and the in vitro studies on physical interactions between bacteria and receptors, an unencapsulated S. pneumoniae, serotype 4 strain, TIGR4 was used. Encapsulated TIGR4 was grown in Todd-Hewitt broth (Oxoid Thermo Scientific, Basingstoke, United Kingdom), and unencapsulated TIGR4 in M17 medium supplemented with 0.5% glucose. Both strains were grown at 37°C, bacterial growth was monitored by measuring the optical density at 600 nm (OD600) with a spectrophotometer. Bacteria were harvested at OD

600 = 0.25-0.30 by centrifugation of 1 ml of culture at 10,000 rpm for 3 minutes.

Subsequently, encapsulated TIGR4 was resuspended in 1ml of sterile phosphate buffered saline (PBS) (Lonza, Verviers, Belgium) till reaching a challenge dose of 107

CFU for each mouse, while unencapsulated TIGR4 was resuspended in HBMEC cell culture medium till reaching a bacterial concentration of approximately 106 CFU.

113

Chapter 5


All experiments involving animals were performed with the prior approval of,

and in accordance with guidelines of the Institutional Animal Care and Use Committee

of the University of Groningen (DEC nr. 6152A). The bacteremia-derived meningitis

model described by Orihuela et al. (24) was adapted as described before (16).

Antibodies and lectin

The following antibody combinations and dilutions in sterile PBS with 5% Fetal

Calf Serum (FCS) (Biochrom, Berlin, Germany) were used for immunofluorescent

detection. To detect S. pneumoniae, either a rabbit anti-capsule serotype 4 antibody

(Statens Serum Institute, Copenhagen, Denmark) 1 :200 diluted or a rabbit anti

pneumococcal antiserum (Eurogentec, Maastricht, The Netherlands) 1 :50 diluted

were used, either combined with an Alexa Fluor 488 goat anti-rabbit antibody

(Invitrogen Life Technologies, Carlsbad, United States) 1 :500 diluted or labeled with

Alexa Fluor 350 (Zenon Rabbit IgG Labeling Kit, Invitrogen Life Technologies). For

the detection of endothelial cells, DyLight 594 labeled Lycopersicon Esculentum

Lectin (tomato lectin) (Vector Laboratories, Burlingame, United States) was used

in a 1 :200 dilution. LEL binds to glycophorin and Tamm-Horsfall glycoprotein and

has been used effectively to label vascular endothelium in rodents (http://www.

vectorlabs.com/catalog.aspx?prodID=1715). For the detection of nuclei DAPI (Roche,

Mannheim, Germany) 1 :5,000 diluted was used. For the detection of mouse PECAM-1,

a rat anti-mouse PECAM-1 antibody (BO Biosciences, Breda, The Netherlands) 1 :50

diluted was used, followed by incubation with either an Alexa Fluor 594 goat anti-rat

antibody or an Alexa Fluor 488 goat anti-rat antibody (Invitrogen Life Technologies)

1 :500 diluted. For the detection of human PECAM-1, a mouse anti-human PECAM-1

antibody (Dako) was used, followed by incubation with an Alexa Fluor 594 goat anti

mouse antibody (Invitrogen Life Technologies). For the detection of mouse pigR a

goat anti-mouse pIGR antibody (R & D Systems, Abingdon, United Kingdom) was

used, followed by incubation with an Alexa Fluor 488 donkey anti-goat antibody

(Invitrogen Life Technologies) . For the detection of human pigR a goat anti-human

pigR antibody (R & D Systems, Abingdon, United Kingdom) was used, followed by

incubation with an Alexa Fluor 488 donkey anti-goat antibody. For the detection of

a-tubulin, a mouse anti-human a-tubulin (Sigma Aldrich) 1 :200 diluted was used,

followed by incubation with an Alexa Fluor 594 goat anti-mouse antibody (Invitrogen

Life Technologies) 1 :500 diluted. As isotype controls for the primary antibodies

mouse IgG (Innovative Research, Plymouth, United States), rat anti-human CD4

(AbD Serotec, Martinsried, Germany) and a goat IgG (Santa Cruz Biotechnology,

Dallas, United States) were used at the same dilutions as those for specific primary

antibodies.

114

Chapter 5

Western blotting

Human PECAM-1, plgR and a-tubulin were detected by Western blotting with the same antibodies that were also used for immunofluorescence staining. AntiPECAM-1 and anti-plgR antibodies were diluted 1:200, while the anti-a-tubulin antibody was 1:1000 diluted. For detection of p53, a mouse anti-p53 antibody (Abeam, Cambridge, United Kingdom) 1:1000 diluted was used. Incubation with primary antibodies was followed by incubation with secondary antibodies IRDye 800 CW goat anti-mouse (LiCor Biosciences, Bad Homburg, Germany) for PECAM-1 and p53 detection, donkey anti-goat (LiCor Biosciences) for plgR detection, or IRDye 700 CW goat anti-mouse (LiCor Biosciences) for a-tubulin detection. All secondary antibodies were 1 :5000 diluted.

Immunofluorescent detection

For immunofluorescent staining of brain, sections of 5µm thin were cut with a cryostat, placed on microscope glass slides and dried under a fan for at least 30 minutes. Sections were fixed with acetone for 10 minutes and dried for 5 minutes. For immunofluorescent staining of HBMEC (Table 1), after 1 hour incubation with S. pneumoniae, cells were fixed with paraformaldehyde (Sigma Aldrich) 4% and washed 2 times with PBS prior to starting the staining procedure.

For the detection of PECAM-1 and S. pneumoniae in brain tissue (Table 1), the slides were first incubated with a mixture of anti-PECAM-1 and anti-capsule serotype 4 antibodies for 1 hour at room temperature (RT). Slides were then washed in PBS 3 times for 5 minutes. Subsequently, the slides were incubated with a mixture of Alexa Fluor 594 goat anti-rat and Alexa Fluor 488 goat anti-rabbit for 1 hour at room temperature (RT) in the dark. After washing with PBS 3 times for 5 minutes, the slides were incubated with DAPI (Roche, Mannheim, Germany) for 10 minutes at RT in the dark. For the detection of PECAM-1, plgR and S. pneumoniae in brain tissue (Table 1), the slides were first incubated with a mixture of anti-PECAM-1 and anti-plgR antibodies for 1 hour at room temperature (RT). Slides were then washed in PBS 3 times for 5 minutes. Subsequently, the slides were incubated with a mixture of the appropriate secondary antibodies for 1 hour at RT in the dark. After washing with PBS 3 times for 5 minutes, the slides were incubated with anti-capsule serotype 4 antibody labeled with Alexa Fluor 350, and then washed again 3 times with PBS for 5 minutes.

For the detection of PECAM-1 and S. pneumoniae in HBMEC (Table 1 ), HBMEC were first incubated with mixture of anti-PECAM-1 antibody and anti-pneumococcal antiserum at RT for 1 hour. After washing 2 times with PBS, cells were incubated with mixture of Alexa Fluor 594 goat anti-mouse and Alexa Fluor 488 goat anti-rabbit antibodies for 1 hour at RT in the dark. Cells were then washed again 2 times with

1 15

Chapter 5

PBS prior to incubation with DAPI for 10 minutes at RT in the dark. For the detection of PECAM-1, pigR and S. pneumoniae in HBMEC and HUVEC

(Table 1), cells were first incubated with a mixture of anti-PECAM-1 and anti-pigR antibodies at RT for 1 hour. After washing 2 times with PBS, cells were incubated with a mixture of appropriate secondary antibodies for 1 hour at RT in the dark. Cells were then washed again 2 times with PBS prior incubation with anti-pneumococcal antiserum labeled with Alexa Fluor 350, then washed again 2 times with PBS for 5 minutes. Citifluor solution (Science Services, Munich, Germany) was added to each tissue section/glass disk after which the coverslip was applied. The slides were analyzed with a Leica DM5500B microscope and images were recorded with a Leica DFC 360 FX camera. Additionally, a Leica SP2 AOBS confocal microscope was used.

Image processing

All the images recorded with the Leica DM5500B fluorescence microscope were processed with ImageJ (25). The TIFF images obtained with the 350nm (blue), 488nm (green) and 594nm (red) wavelength filters were merged using the Color-Merge Channels function. As a minor digital manipulation, background correction using the Brightness & Contrast command of Image} was applied to reduce the background only when the background was across the image. In all images subjected to this minor adjustment, the background subtraction was applied to all parts of the image. The LEI z-stacks obtained with the Leica SP2 AOBS confocal microscope were merged with Imaris (Bitplane Scientific Software). No digital adjustment was performed on these images.

Bacterial quantification and co-localization analysis using Image]

The surface covered by bacteria was measured using the Threshold function of Image} (25), by determining the area occupied by the 488 nm bacterial signal and the white pixels of the bacteria co-localizing with receptors. The percentage of bacteria co-localized with receptors compared to the total bacterial signal was calculated by dividing the surface area of white bacteria by the surface area of the green bacteria in each image. The images representing the bacterial and PECAM-1 signals were analyzed with the Image} (25) application Co-localization analysis. White pixels were automatically generated on the areas of the bacterial signal co-localized with the PECAM-1 signal.

Physical interaction of S. pneumoniae with PECAM-1 protein

Lysis buffer consisted of 50 mM tris-HCl (Promega, Mannheim, Germany) pH 7.4, 150 mM NaCl, 1% triton lO0X, 1% sodium deoxycholate, 0.1% SDS (all from Sigma Aldrich), 1 mM EDTA (Merck Millipore, Billerica, United States), and

116

Chapter 5

protease inhibitors 1 X (Roche, Mannheim, Germany). 250 µI of lysis buffer was added to confluent monolayers of HBMEC or HUVEC grown in small T25 flasks (TPP, Trasadingen, Switzerland) . After 1 minute incubation at RT, monolayers were scraped with a cell scraper (Nalge Nunc International, Rochester, United States), harvested, and centrifuged at 20,600 rpm for 15 minutes at 4°C. The supernatant fraction was used for further analyses. As a quality control, the lysates were separated by SOS-PAGE and proteins were visualized by coomassie brilliant blue staining. A solution of approximately 106 CPU of unencapsulated S. pneumoniae TIGR4 was prepared in sterile PBS. 50 µI of HBMEC or HUVEC lysate was added to 50 µI of the bacterial solution and the mixture was incubated at 4°C with gentle agitation for 1 hour. The mixture of cell lysate and bacteria was then centrifuged at 20,600 rpm for 20 minutes at 4°C. The supernatant was removed and the bacterial pellet washed 2 times with 100 µI PBS. After washing, the pellet was resuspended in 100 µI of a mixture of anti-pneumococcal antiserum and anti-PECAM-1 antibody in PBS, followed by incubation at 4°C for 1 hour. After washing in PBS, the bacterial pellet was resuspended in a mixture of Alexa Fluor 488 goat anti-rabbit and Alexa Fluor 594 goat antip10use antibodies in PBS. As a negative control mouse anti-human a-tubulin antibody was used. Incubation with anti-a-tubulin antibody was followed by incubation with Alexa Fluor 488 goat anti-mouse antibody. The bacterial pellet was finally resuspended in 100 µI of distilled water. A 5 µI drop was pipetted onto a microscope glass slide, and Citifluor solution (Science Services, Munich, Germany) was added to the slide after which a coverslip was applied. The slide was then analyzed with a Leica DM5500B fluorescence microscope and images were recorded with a Leica DFC 360 FX camera.


HBMEC and HUVEC were grown in 12-well plates (TPP) at 37°C with 5% CO2

until confluency. Prior seeding HUVEC, each well was coated with gelatin for 45 minutes. When immunofluorescent staining was performed after the adherence assay, cells were grown on glass disks placed inside each well. For adherence of HUVEC on glass, disks were coated with gelatin for 45 minutes. After removal of the gelatin, disks were incubated with 0.5% glutaraldehyde (Sigma Aldrich) for 45 minutes. Disks were then washed 2 times with PBS and 2 times with HUVEC culture medium. After washing 2 times with PBS, anti PECAM-1 antibody diluted in RPMI was added to each well to a final concentration of 50 µg/ml and cells were incubated for 1 hour at 37°C with 5% CO2• As control we used HBMEC and HUVEC treated with mouse IgG (Innovative Research, Plymouth, United States) at 50 µg/ml or just RPMI. After washing the cells with sterile PBS, 900 µI cell culture medium was added to each well and 100 µI of approximately 106 CPU of unencapsulated TIGR4 was

117

Chapter 5

added. After 1 hour incubation at 37°C at 5% CO2

cells were treated with a 50/50 mix

of 1 % saponin (Sigma Aldrich) and trypsin-EDTA (0.05%-0.02%) (Gibco) and lysed.

CFU were determined by plating serial dilutions of lysed cells on blood agar plates.

The ratio of adherent bacteria was calculated by dividing the adherent bacteria by the

total amount of bacteria in the well (adherent + not adherent bacteria).

Detection of PECAM-1 protein in HBMEC and HUVEC lysates

Proteins of the cell lysates were separated by SDS-PAGE using NuPAGE gels

(Invitrogen), and then blotted (75 min, 100 mA per gel) onto a nitrocellulose membrane

(Protran®, Schleicher & Schuell, Bath, United Kingdom). Antibody detection was

performed with IRDye 800 CW IgG secondary antibodies (LiCor Biosciences, Bad

Homburg, Germany), in accordance with the IgG isotype of the primary antibodies,

and 700 CW goat anti-mouse for a-tubulin detection (see Antibodies and lectin),

in combination with the Odyssey Infrared Imaging System (LiCor Biosciences).

Fluorescence was recorded at 700 and 800 nm.

Protein-protein interaction studies

Approximately 5µg of anti-PECAM-1, anti-pigR antibody or mouse IgG were

incubated with 15µ1 of Protein A/G Plus-Agarose beads (Santa Cruz Biotechnology)

at 4"C for 1 hour under gentle agitation. Subsequently, to remove unbound antibody,

the antibody-coated beads were centrifuged at 6,700g for 30 seconds at 4°C and

washed with 500µ1 of the lysis buffer used to make cell lysates. This procedure

was repeated 3 times, and afterwards the pellet was resuspended in 50µ1 of lysis

buffer. Aliquots of cell lysates (-200µg protein) were incubated with 10µ1 of Protein

A/G Plus-Agarose beads (Santa Cruz Biotechnology) at 4"C for 1 hour under gentle

agitation. After centrifugation at 6,700 rpm for 30 seconds at 4 ·c, the cell lysates were

mixed with antibody-coated beads and the total mix was incubated at 4°C for 1 hour

under gentle agitation. Samples were then centrifuged at 8000 rpm for 30 seconds

at 4°C and the pellets were washed with 500µ1 of lysis buffer. This procedure was

repeated 3 times. Lastly, the pellets were resuspended in 50µ1 of lysis buffer and

boiled at 95°C for 5 minutes. Proteins in the different samples were separated on

NuPAGE gels (Invitrogen}, and then blotted (see above Western blotting procedure).

Statistical analysis

The independent student t-test of SPSS Statistics 20 (IBM) was used for the

statistical analysis of the bacterial-endothelial adhesion assay results. The comparison

between ratio of adherent bacteria in HBMEC/HUVEC cells treated with receptor

antibodies and HBMEC/HUVEC cells treated with isotype IgG or RPMI without

antibody was statistically tested. Differences were considered significant when p < 0.05.

118

Chapter 5

Acknowledgements

Part of the work was performed at the UMCG Microscopy and Imaging Center (UMIC). We thank Henk Moorlag for providing HUVEC and for the cryostat cutting of tissue sections. We also thank Henrik Gradstedt for providing the total anti pneumococcal antiserum and Prof. Dr. Jan Maarten van Dijl for critically reading the manuscript.

Table 1 . lmmunofluorescent detection scheme

Detection of J•t 2nd 3rd 4th

PECAM-l and S. pneumoniae Anti PECAM-1 Anti capsule Alexa Fluor goat DAPI

in brain tissue, antibody serotype 4 anti rabbit 488

HBMEC and antibody 1 :5000

HUVEC Dilution 1 :50 Dilution 1 :500 Dilution 1 :50

Detection of J•t 2nd 3rd

PECAM-1 and S.

pneumoniae in lung Anti PECAM-1 Anti capsule DAPI

tissue antibody serotype 4 antibody Dilution 1 :5000

Dilution 1 :50 Dilution 1 :50

Detection of J•t 2nd 3rd

PECAM-1, plgR and S. pneumoniae Mixture anti Mixture of Alexa Anti capsule

in brain tissue, PECAM-1 and anti Fluor 594 donkey serotype 4

HBMEC and plgR antibodies anti mouse and antibody labeled

HUVEC Alexa Fluor 488 with Alexa Fluor Dilution of each antibody 1:50

donkey anti goat 350 (Zenon Kit)

Dilution 1 :500

119

Chapter 5

References 1. Bogaert D, De Groot R, Hermans PW (2004) Streptococcus pneumoniae colonization: the key to

pneumococcal disease. Lancet Infect Dis 4:144-54.

2. Koedel U, Scheid WM, Pfister HW (2002) Pathogenesis and pathophysiology of pneumococcal meningi tis. Lancet Infect Dis 2:721-36.

3. Woehr) B, Klein M, Grandgirard D, Koedel U, Leib S (2011) Bacterial meningitis: current therapy and possible future treatment options. Expert Rev Anti Infect Ther 9:1 053-65.

4 . Ring A, Weiser JN, Tuomanen EI (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 102:347-60.

5. Zhang JR, Mostov KE, Lamm ME, Nanno M, Shimida S, et al. (2000) The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102: 827-37.

6. Coureuil M, Join-Lambert 0, Lecuyer H, Bourdoulous S, Marullo S, Nassif X (2012) Mechanism of meningeal invasion by Neisseria meningitidis. Virulence 3: 164-72.

7. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1995) Streptococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377:435-8.

8. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2005) Beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pneumoniae. Infect Immun 73:7827-35.

9. Iovino F, Brouwer MC, van de Beek D, Molema G, Bijlsma JJ. Signalling or binding: the role of the platelet-activating factor receptor in invasive pneumococcal disease. Cell Microbiol 15: 870-81.

10. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, Abouseada N, et al. (2009) Laminin receptor initiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest 1 19:1638-46.

11. DeLisser HM, Newman PJ, Albelda SM (1994) Molecular and functional aspects of PECAM-1/CD31 . Immunol Today 15: 490-5

12. Woodfin A, Voisin MB, Nourshargh S (2007) PECAM-1 : a multi-functional molecule in inflammation and vascular biology. Arterioscler Thromb Vase Biol 27: 2514-23.

13. Muller WA, Weigl SA, Deng X, Phillips DM (1993) PECAM-1 is required for trans-e11dothe/ial migration of leukocytes. J Exp Med 178: 449-60.


15. Lovelace MD, Yap ML, Yip J, Muller W, Wijburg 0, Jackson DE (2013) Absence of Platelet Endothelial Cell Adhesion Molecule 1, PECAM-l/CD31, In Vivo Increases Resistance to Salmonella enterica Serovar Typhimurium in Mice. Infect Immun 81 : 1952-63.

16. Iovino F, Orihuela CJ, Moorlag HE, Molema G, Bijlsma JJE (2013) Interactions between blood-borne Streptococcus pneumoniae and the blood-brain barrier leading to meningitis. Plos One. In press.

17. Newman PJ, Newman DK (2004) Signal transduction pathways mediated by PECAM-1 : new roles for an old molecule in platelet and vascular cell biology. Arterioscler Thromb Vase Biol 23: 953-64.

18. Eugenin EA, Gamss R, Buckner C, Buono D, Klein RS, et al (2006) Shedding of PECAM-1 during HIV infection: a potential role for soluble PECAM-1 in the pathogenesis of NeuroAIDS. J Leukoc Biol 79: 444-52.

19. 0stergaard C, O,Reilly T, Brandt C, Frimodt-M0ller N, Lundgren JD (2006) Influence of the blood bacterial load on the meningeal inflammatory response in Streptococcus pneumoniae meningitis. BMC Infect Dis 6:78.

20. Sheibani N, Sorenson CM, Frazier WA (1999) Tissue specific expression of alternatively spliced murine PECAM-1 isoforms. Dev Dyn 214: 44-54.

21. Jen FE, Warren MJ, Schulz BL, Power PM, Swords WE, et al. (2013) Dual pili post-translational modifications synergize to mediate meningococcal adherence to platelet activating factor receptor on human airway cells. PLoS Pathog. 9: e1003377.

120

Chapter 5

22. Stins MF, Badger J, Sik Kim K (2001) Bacterial invasion and transcytosis in transfected human brain microvascular endothelial cells. Microb Pathog 30: 19-28.

23. Romero-Steiner S, Caba J, Rajam G, Langley T, Floyd A, et al. (2006) Adherence of recombinant pneumococcal surface adhesion A (rPsaA) coated particles to human nasopharyngeal epithelial cells for the evaluation of anti-PsaA functional antibodies. Vaccine 24: 3224-31 .

24. Orihuela CJ, Mahdavi J, Thornton J, Mann B, Wooldridge KG, et al. (2009) Laminin receptor initiates bacterial contact with the blood brain barrier in experimental meningitis models. J Clin Invest 1 19: 1638-46.

25. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to Image]: 25 years of image analysis. Nat Methods 9: 671-5.

1 21

CHAPTER 6

General summary and future perspectives

- 123 -

Chapter 6

General summary

Despite the introduction of antimicrobial agents and vaccines, bacterial meningitis still remains a life-threating disease with high morbidity and mortality worldwide. The Gram-positive bacterium Streptococcus pneumoniae is the main causative agent of this disease (1). Bacterial meningitis is an inflammation of the meninges, the layers covering and protecting the brain and spinal cords. The inflammation is caused by bacteria, which enter the Central Nervous System (CNS) and spread in the brain. The blood-brain barrier is a specialized vasculature system, which separates the brain from the circulating blood and, as such, it has critical functions in both the protection and nutrient supply of the brain (2, 3). Endothelial cells form the layer that lines the interior surface of the blood vessels (4). To invade the brain from the blood, meningeal pathogens must first face the blood-brain barrier endothelium and develop strategies to pass this barrier. Transcytosis has been proposed as a mechanism employed by the bacteria for adhesion to, invasion into and translocation across human cell barriers (6, 12). The aim of the PhD research described in this thesis was to investigate where and how the interactions between S. pneumoniae and the blood-brain barrier take place during the events preceding the development of meningitis.

S. pneumoniae is able to adhere to brain endothelial cells (5-8) but, when the present studies were started, the interactions between S. pneumoniae and brain endothelial cells had not been investigated in vivo. It was therefore first investigated whether S. pneumoniae adheres to the vascular endothelium in vivo, and whether pneurnococci have preferential sites of adhesion to the blood-brain barrier endothelium in vivo. The results documented in Chapter 2 show that S. pneumoniae indeed adheres to the blood-brain barrier endothelium in a murine in vivo model. Specifically, bloodborne pneumococci adhered to the microvascular endothelium of the subarachnoid space, cerebral cortex and the septum, though the spatiotemporal distribution of S. pneumoniae over these sites differed. Notably, the choroid plexus had previously been proposed to be the area where the main bacterial invasion of the CNS occurs during bacterial meningitis (9, 10, 11). However, we observed pneumococci in the choroid plexus only during the later stages of pneumococcal infection. Although leukocyte influx in the brain during the entire course of infection was minimal, microglia and astrocytes were activated early on, already after 1 hour post infection, indicating that the local immune system in the brain was activated directly upon the presence of pneumococci in the blood. Signs of astrogliosis and microglial activation became even more prominent over the time course of infection. This implies that the local immune system of the brain immediately sensed the bacterial presence in the blood and/or bacterial adhesion to the endothelial cells, and was activated during the entire time course of pneumococcal infection.

125

Chapter 6

Endothelial and epithelial cells express certain receptors on the plasma membrane to which bacteria can bind (6, 12). After investigating the spatiotemporal interactions of blood-borne S. pneumoniae to the brain microvascular endothelium, the role of specific receptors in S. pneumoniae adhesion to endothelial cells was investigated. The platelet-activating factor receptor (PAFR) is a G-protein coupled receptor of the platelet-activating factor (PAF), a mediator in diverse pathological processes, such as allergy, asthma, septic shock, arterial thrombosis, and inflammation (13, 14). S. pneumoniae has been proposed to adhere to PAFR and this adhesion may facilitate the passage of bacteria through endothelial cells leading to invasive disease (5, 6, 7). However, the current literature is characterized by conflicting data concerning the role of PAFR in invasive pneumococcal disease (IPD). Therefore, the current literature on PAFR interactions with S. pneumoniae and other pathogens was analyzed, summarized and discussed in Chapter 3. This included studies concerning genetic variations in the human PAFR related to clinical features of IPD. The in vitro

and most of the in vivo experiments described in the papers discussed in Chapter 3 indicated that PAFR is involved in the development of IPD. However, no unequivocal evidence that S. pneumoniae binds to PAFR in vivo was so far documented in the literature.

In Chapter 4, evidence is provided that a direct interaction between PAFR and S. pneumoniae is not likely to occur. Immunofluorescent analysis of brain tissue from intravenously infected mice showed that S. pneumoniae does not co-localize with PAFR. This was corroborated by immunofluorescent detection of S. pneumoniae

and PAFR in Human Brain Microvascular Endothelial Cells (HBMEC) and Human Umbilical Vein Endothelial Cells (HUVEC) incubated with pneumococci, which showed almost no co-localization between bacteria and the PAFR receptor. Thus, it was proposed that, although PAFR is involved in the adhesion of S. pneumoniae

to endothelial cells, a direct binding of pneumococci to PAFR is unlikely. This is in line with published studies, which showed that when PAFR is genetically deleted or chemically inhibited, S. pneumoniae is still able to adhere to human cells in vitro and cause invasive infections in vivo. Importantly, this implies that pneumococci have the capability to bind to alternative receptors (5, 6, 7).

The poly immunoglobulin receptor (pigR) mediates the transport of immunoglobulins across mucosa} epithelium (15, 16) and is known to be involved in pneumococcal adhesion to the epithelium of the human nasopharynx (12). However, it had not been investigated yet whether pigR also plays a role in S. pneumoniae adhesion to the brain vascular endothelium in vivo. This was therefore assessed by immunofluorescent analysis of brain tissue of intravenously infected mice. Importantly, the results presented in Chapter 4 show that most pneumococci are co-localized with pigR. The co-localization of bacteria and pigR was observed

126

Chapter 6

also in HBMEC and HUVEC after incubation with pneumococci. Furthermore, by incubating pneumococci with HBMEC and HUVEC lysates, it was shown that S. pneumoniae can directly bind to pigR. The combined results therefore suggest that plgR could be a direct adhesion receptor for S. pneumoniae on the blood-brain barrier endothelium.

In the present PhD research, attention was also focused on a third receptor molecule expressed on the cell membrane of brain vascular endothelial cells and its possible involvement in S. pneumoniae adhesion, namely the platelet endothelial cell adhesion molecule (PECAM-1). PECAM-1 is one of the major adhesion molecules expressed by endothelial cells and it has a crucial role in neuroinflammation and subsequent blood-brain barrier disruption (17, 18). Very recently, PECAM-1 was implicated in gastrointestinal infection caused by Salmonella typhimurium. Chapter 5 describes a study conducted to investigate whether PECAM-1 could potentially be a receptor for S pneumoniae adhesion to the endothelium of the blood-brain barrier. Immunofluorescent analysis performed on brain tissue from intravenously infected mice showed that most pneumococci did indeed co-localize with PECAM-1. A similar co-localization was observed in vitro in HBMEC and HUVEC cells upon incubation with pneumococci. Blocking of PECAM-1 with a specific antibody in HBMEC and HUVEC cells led to a significant reduction of bacterial adhesion, indicating that PECAM-1 is functionally involved in S. pneumoniae adhesion to human endothelial cells. Using the same biochemical approach that was used to demonstrate the S. pneumoniae-pigR association (Chapter 4), it was then shown that S. pneumoniae can directly bind the PECAM-1 expressed by HBMEC and HUVEC cells. Importantly, in a subsequent series of experiments, it was shown that PECAM-1 co-localizes with pigR and infecting pneumococci and, moreover, that both receptors can directly interact with each other as was evidenced by co-immunoprecipitation experiments. Taken together, these observations imply that S. pneumoniae binds to PECAM-1 and pigR on brain microvascular endothelial cells, which possibly (co-)facilitates its passage across the blood-brain barrier and invasion of the brain.

In conclusion, the results obtained in this PhD research imply that S. pneumoniae can cross the brain vascular endothelium as a result of adhesion to the PECAM-1 and pigR receptors, followed by translocation through or between endothelial cells without causing major disruptions of the vascular endothelium. However, the question whether the pneumococcus either uses an intracellular route, a paracellular route or both routes for translocation across the brain microvascular endothelium remains to be answered. Lastly, while direct interactions were shown between S. pneumonia, pigR and PECAM-1, no evidence for direct interactions of S. pneumoniae with PAFR was obtained. This suggests that PAFR is only indirectly involved in the pneumococcal translocation across the blood-brain barrier.

127

Chapter 6

Future perspectives

Co-operation of PAFR, plgR and PECAM-1 during pneumococcal adhesion to endothelial

cells: a new model

From the studies presented in this thesis and previous publications by others, it remains unclear what the exact role of PAFR is in invasive pneumococcal disease (IPD). The involvement of PAFR in S. pneumoniae adhesion to endothelial cells might well be exerted through the binding of PAF or pneumococcal cell wall components to the receptor and the subsequent inflammatory consequences of its activation. Inflammation leads to an upregulation of receptors expressed on the plasma membrane of cells (19), and PECAM-1 and pigR may be among them. In the presence of more pigR and PECAM-1, S. pneumoniae has more sites to adhere to the endothelial cells. Like others we have shown that the blocking of PAFR with a specific antibody leads to a significant reduction of pneumococcal adhesion to endothelial cells in vitro. We therefore hypothesize that, due to PAFR blocking, the upregulation of pigR and PECAM-1 may be prohibited. Possibly, the presence of pneumococci may trigger the autocrine production of PAF, directly or indirectly,

which activates PAFR leading to an upregulation of pigR and PECAM-1. This would increase pneumococcal adherence to pigR and PECAM-1 and, in tum, more adherent bacteria would increase the chances of bacterial invasion into the cells. Besides PAF, secreted pneumococcal components, such as pneumolysin and CbpA, could bind to PAFR and activate this receptor (20). Clearly this remains to be demonstrated, and an in-depth investigation of secreted and surface-attached pneumococcal proteins, including proteomics and structural analyses, will be required to identify any protein(s) of S. pneumoniae that could be recognized and bound by PAFR. Once these as yet hypothetical proteins are identified, mutant S. pneumoniae strains lacking these proteins can be created to verify their possible roles in PAFR binding. The absence of such pneumococcal proteins would prevent PAFR activation and thus PECAM-1 and pigR upregulation. The direct consequence would be that these S. pneumoniae

mutants have a reduced ability to adhere to endothelial cells than the respective parental wild-type bacteria. In particular, the expression of pigR and PECAM-1 in cells incubated with pneumococci lacking the hypothetical proteins and wildtype pneumococci should be quantified to validate whether the expression of these two receptors is indeed not upregulated when the PAP-binding site is blocked. A parallel approach would be to use/develop specific antibodies directed against the PAF-binding site of PAFR. By treating endothelial cells with such antibodies in vitro,

PAFR activation should be blocked and, consequently, the upregulation of pigR and PECAM-1 as well. The expected outcome would be that pneumococci adhere less to antibody-treated cells than to cells that were not treated with the antibody. If so, PECAM-1 and pigR would not be upregulated when PAFR is not activated.

128

Chapter 6

The present studies have shown that, if pigR and PECAM-1 are blocked, the pneumococcal adhesion to endothelial cells is reduced. Bacteria cannot adhere to pigR and PECAM-1 upon their blocking by the respective antibodies, but PAFR activation may still be possible and could thus lead to upregulation of PECAM-1 and pigR expression. Pneumococci could next adhere to pigR and PECAM-1, which would be increasingly expressed after PAFR activation. Using an experimental set up that allows a precise analysis of the sequence of events, we should be able to dissect this presumed co-operation of PAFR, pigR and PECAM-1 in pneumococcal adhesion to endothelial cells. It would furthermore be interesting to determine whether more bacterial adhesion leads to more invasion into and transmigration over endothelial cells, which could be studied in vitro using cell culture assays. Invasion into cells can be investigated by adding antibiotics into cell culture medium after cells have been incubated with bacteria. In this way all extracellular bacteria, including adherent bacteria, would be killed while only the intracellular bacteria survive. Transmigration can be studied in vitro using transwells, which are often applied in cell culture assays to study the migration of molecules from the apical to the basolateral side of confluent cell layers.

In order to prevent the onset of meningitis and other invasive diseases upon pneumococcal bacteremia, the main goal would be to completely inhibit adhesion of pneumococci to endothelial cells. Clearly, if pneumococci cannot adhere to endothelial cells, bacterial invasion will most likely be prevented. In Chapter 5 data are provided, which suggest that PECAM-1 and pigR may form a double receptor on the plasma membrane of endothelial cells. To determine whether these two receptors do indeed form a double receptor, further interaction and structural analyses are needed that would also serve to pinpoint the relevant interfaces for protein-protein interactions. It would be important to complement crystallographic studies with other experimental and computational approaches (21). Current methods for comparative modeling allow the accurate construction of three-dimensional models for proteins of unknown structure, and binding site predictors can be used to propose binding regions (22) . Intriguingly, it is well conceivable that the putative double receptor composed of PECAM-1 and pigR could have a double function. Thus, it might exercise an immunoglobulin transfer function, the main physiological role of pigR and, at the same time, regulate endothelial cell-cell connections, known as one of the main functions of PECAM-1.

Blockade of S. pneumoniae receptors: a new horizon for the prevention of pneumococcal

meningitis

Knowledge of the mechanisms by which the pneumococcus interacts with the blood-brain barrier before invading the brain is fundamental in order to develop

129

Chapter 6

therapeutic strategies to avoid the adhesion of S. pneumoniae to the blood-brain

barrier and, thus, the invasion of the pathogen into the CNS. The results presented

in this thesis provide a better understanding of the events preceding pneumococcal

meningitis and, in particular, of S. pneumoniae receptor-mediated adhesion to the

blood-brain barrier endothelium. Specifically, it is shown that in mice intravenously

infected with S. pneumoniae, the bacteria adhere to the blood-brain barrier as early

as 1 hour post challenge, and that the amount of bacteria in the brain increases over

the time of infection. Nevertheless, the influx of leukocytes into the brain remained

low, even at 14 hours after challenge. Despite the low presence of leukocytes in the

brain, clinical symptoms of the infected mice, astrogliosis and microglial activation

were observed already at 1 hour post challenge, and a significant increase of pro

inflammatory cytokines in the brain at 14 hours post challenge compared to mock

conditions, were all signs of a possible progression to meningitis (8). The bacteremia

derived meningitis model was used in this PhD research to investigate the events

preceding meningitis. To study meningitis itself, intracisternal administration

of bacteria was previously used to induce meningitis in vivo. Using this model,

neutrophil influx in the brain was detected only at 30 hours after challenge (23),

indicating that white blood cells can be detected in the brain only at the late stages

of the disease. A high count of white blood cells in the cerebrospinal fluid (CSF) is

often used as a clinical test for the diagnosis of meningitis (24). Moreover, classic

symptoms of meningitis, such as rash, neck stiffness, photophobia, severe headache

and impaired consciousness develop late (24). While these symptoms represent the

classical picture of suspected meningitis, the disease may be already at an advanced

stage in patients displaying them. Thus, it is not really surprising that at this stage

the risks of brain damage is considerable even in case the infection is successfully

fought with antibiotics (25). World Health Organization (WHO) data indicate that,

even if the disease is diagnosed early and adequate treatment is started immediately,

5% to 10% of the patients will die, typically within 24 to 48 hours after occurrence

of the first symptoms. When left untreated, up to 50% of the patients may die. In

case of recovery, brain damage, hearing loss or a learning disability appears in 10%

to 20% of the survivors. It is therefore crucial to start interventions before bacteria

have invaded the brain in order to optimize the survival chances of patients and

to prevent permanent damage. Treatment with antibiotics is a common method to

cure meningitis, but bacteria can develop resistance to antibiotics quickly. Routine

vaccination against pneumococcal infections with the pneumococcal conjugate

vaccine (PVC), which is active against seven common serotypes of this pathogen, has

significantly reduced the incidence of pneumococcal meningitis (26, 27). Notably,

in this thesis, the roles of PAFR, plgR and PECAM-1 in S. pneumoniae adhesion to

the vascular endothelium of the blood-brain barrier are described. Hence, a possible

130

Chapter 6

approach complementary to vaccines and antibiotics for curing or preventing pneumococcal meningitis could be the development of therapies that interfere with these receptors. This could for example be achieved with monoclonal antibodies blocking PAFR, plgR and PECAM-1, which would limit the adhesion of pneumococci to the blood-brain barrier.

Bacteremia is the presence of bacteria in the blood stream, and it is shown in this thesis that pneumococci in the blood can reach and adhere to the blood-brain barrier after which they possibly invade the brain. A major cause of bacteremia is pneumonia, which can occur when an infection of the lungs grows out of control (28, 29, 30). Patients with bacteremia and pneumonia can be considered at "high-risk" for progression towards brain infection and, for this reason, it could be valuable for them to receive innovative treatments that block pneumococcal receptors. This would prevent the bacteria from adhering to the blood-brain barrier and invading the brain. Such a treatment would even be of relevance for patients who are hospitalized with meningitis as it would still reduce the numbers of bacteria adhering to the bloodbrain barrier and, in combination with antibiotic therapy, both the blood-borne bacteria and the bacteria that have already reached the brain could be eliminated.

To evaluate the validity of such a new therapeutic strategy in mice, the best approach would be to intravenously inject the bacteria then proceed with the systemic administration of the blocking antibodies and observe whether the systemic administration of blocking antibodies leads to less severe disease symptoms. The use of mice lacking the receptors could be used as an alternative approach, but the receptors cannot be genetically deleted in humans, while receptor-specific antibodies could be administered as therapeutic agents. Importantly, it has been reported that the absence of PAFR is not lethal in mice, which can grow and live as well as wildtype mice (31). In PAFR-1• mice modifications of known angiogenic factors, such as VEGF, have not been observed, but a sustained rise of the chemokines CXCL2 and CCL2 has been detected, which may compensate for the absence of an important molecule in maintaining immunological surveillance (32). These findings indicate that the absence of PAFR in mice is compatible with a healthy life, although it has to be noted that this was observed under controlled conditions. This is a crucial issue, because the blocking of this receptor should prevent pneumococcal infection without causing lethal or dangerous side effects in patients. Mice lacking plgR were of normal size and fertility as well, but displayed increased IgG levels in their sera, suggesting a triggering of systemic immunity (33, 34). These findings indicate that the absence of plgR is not lethal either, but at the same time it seems that the absence of plgR is sensed as a potential danger by systemic immunity. PECAM-1 is expressed not only by endothelial cells, but also by platelets, neutrophils and macrophages, and its signaling pathways might mediate resistance to apoptosis

131

Chapter 6

and promote cell survival (17). Absence of PECAM-1 may therefore lead to several cellular dysfunctions. Interestingly, Duncan et al. showed that PECAM-1 is neither required for embryogenesis, fetal maturation, nor fertility in mice and, furthermore, PECAM-1 expression on platelets is not essential for platelet-platelet aggregation (35) . PECAM-1 1 animals develop normally and do not show any signs of an abnormal physiology under normal baseline conditions. However, as the animals age they can develop an autoimmune lupus-like syndrome and, when stressed, a variety of abnormal responses have also been noticed. In particular, PECAM-1 ·1• animals exhibited a prolonged bleeding time (36). Thus, although in certain circumstances the absence of the PAFR, pigR or PECAM-1 receptors did not cause serious problems in mice, dysfunctions and abnormalities due to a lack of these receptors cannot be excluded. Thus, while the blocking of these pneumococcal receptors could provide protection against meningitis, it will be crucial to develop strategies to avoid possible dysfunctions caused by their blocking before this approach can be implemented in the clinic. A possible way would be to first determine the site of each receptor to which bacteria bind, and then to investigate how to block this specific site in order to preclude pneumococcal adhesion without altering the physiological functions of the respective receptors. It has been shown that computational methods can accurately predict peptide binding sites on protein surfaces (37) and, by using these informatics tools, the pneumococcal binding sites of PAFR, plgR and PECAM-1 can in principle be predicted. Through site-directed mutagenesis, these binding sites of the receptors can be deleted and, in this way, various mutant versions of the respective receptors can be generated. Then with expression vectors, cell lines can be transfected in order to generate cells expressing the mutated versions of the receptors. For example, COS-7 cells have been previously used for transfection with PAFR-expressing vectors and can be used also for PECAM-1 expression (7, 38), while transfection with pigR-expressing vectors has been reported in MDCK and Calu-3 cells (39, 40, 41). The receptor sites involved in the binding to the bacteria can thus be found by evaluating the adhesive behavior of S. pneumoniae to the cells expressing the different versions of the mutated receptors. When these binding sites are no longer present, bacterial adhesion should be significantly decreased. Genetically modified mice expressing the receptors lacking the binding sites for S. pneumoniae should then be generated to evaluate whether the deletion of the binding sites alters the normal physiological functions of the respective receptors. Based on the literature data described above, mutated PAFR is not likely to cause serious dysfunctions. However, mutated pigR may interfere with the normal transport of immunoglobulins across cells. It has been shown that treatment with phorbol myristyl acetate (PMA) activates isozymes of protein kinase C (PKC), a downstream protein in the signaling cascade of pigR (42). This activation stimulates not only pigR transcytosis, but more generally the pathway of a variety of

132

Chapter 6

molecules in the transcytotic pathways (42). Thus, treahnent with PMA can stimulate

cells to use other transcytotic systems and their receptors, in case immunoglobulin

transfer mediated by pigR is altered. In this way transport of immunoglobulins can

be maintained as well through other transcytotic ways. Regarding the vascular role

of PECAM-1, it is important to bear in mind that the PECAM-1 protein is present in

the junctions between endothelial cells in order to avoid problems, such as long-term

bleeding, which could be dangerous for patients. However, in principle, it should

be possible to develop monoclonal antibodies blocking the pneumococcal binding

site of the PECAM-1 receptor. These could be administered systemically in mice

prior to an intravenous challenge with S. pneumoniae to determine whether treahnent

with these specific antibodies reduces the progression towards meningitis. Mice

expressing the mutated receptors can also be generated but, clearly, genetic deletion of

receptor genes in humans is not an option, while systemic administration of blocking

antibodies has a realistic potential to become a new therapeutic approach for the

prevention and cure of meningitis. While the blocking of pigR and PECAM-1 showed

only a partial reduction of pneumococcal adherence to endothelial cells, the blocking

of the PAFR, pigR and PECAM-1 receptors at the same time could be a potential

way to completely preclude the adhesion of S. pneumoniae to the blood-brain barrier

endothelium and, therefore, prevent completely the pneumococcal invasion into the

brain. This, however, remains to be assessed experimentally. Clearly, an important

advantage of conceivable therapies based on monoclonal antibodies against PAFR,

plgR and PECAM-1 would be that these antibodies need to be administered only

as long as pneumococci are detectable in the blood. Once these have been fought

successfully with regular antibiotics, the antibody therapy can be stopped, which

will reduce possible unwanted side effects to a minimum.

In conclusion, by knowing the pathways by which S. pneumoniae adheres to

endothelial cells it may become possible to interfere with these mechanisms and to

develop innovative strategies for preventing and curing pneumococcal meningitis.

This is an important objective, because pneumococcal meningitis is still a very serious

infectious disease with high mortality and morbidity worldwide.

133

Chapter 6

References 1 . Brouwer MC, Tunkel AR, van de Beek D (2010) Epidemiology, diagnosis,

and antimicrobial treatment of acute bacterial meningitis. Clin Microbiol Rev 23: 467-92.


3. Woehr! B, Klein M, Grandgirard D, Koedel U, Leib S (201 1) Bacterial meningitis: current therapy and possible future treatment options. Expert Rev Anti Infect Ther 9: 1053-65.

4. Cines DB, Pollak ES, Buck CA, Loscalzo J, Zimmerman GA, et al (1998) Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood. 91: 3527-61.

5. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1 995) Streptococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377: 435-8.

6. Ring A, Weiser JN, Tuomanen EI (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 102: 347-60.

7. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2005) beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pneumoniae. Infect Immun 73: 7827-35.

8. Iovino F, Orihuela CJ, Moorlag HE, Molema G, Bijlsma JJ (2013) Interactions between blood-borne Streptococcus pneumoniae and the blood-brain barrier preceding meningitis. Plos One in press.

9. Hoffman 0, Weber RJ (2009) Pathophysiology and treatment of bacterial meningitis. Ther Adv Neurol Disord 2: 1-7.

10. Tenenbaum T, Essmann F, Adam R, Seibt A, Janicke RU, et al. (2006) Cell death, caspase activation, and HMGBl release of porcine choroid plexus epithelial cells during Streptococcus suis infection in vitro. Brain Res 1 100 :1-12.

11. Tenenbaum T, Papandreou T, Gellrich D, Friedrichs U, Seibt A, et al. (2009). Polar bacterial invasion and translocation of Streptococcus s11is across the blood-cerebrospinal fluid barrier in vitro. Cell Microbiol 1 1 : 323-36.

12. Zhang JR, Mostov KE, Lamm ME, Nanno M, Shimida S, et al (2000) The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102: 827-37.

13. Shukla SD (1992) Platelet-activating factor receptor and signal transduction mechanisms. FASEB J 6:2296-301

14. Honda Z, Ishii S, Shimizu T (2002) Platelet-activating factor receptor. J Biochem 131 :773-9.

15. Kim KJ, Malik AB. (2003) Protein transport across the lung epithelial barrier. Am J Physiol Lung Cell Mol Physiol 284: L247-59.

16. Asano M, Komiyama K. (2011 ) Polymeric immunoglobulin receptor. J Oral Sci 53: 147-56.

17. Newman PJ, Newman DK (2003) Signal transduction pathways mediated by PECAM-1 : new roles for an old molecule in platelet and vascular cell biology. Arterioscler Thromb Vase Biol 23: 953-64


19. Shivshankar P, Boyd AR, Le Saux CJ, Yeh IT, Orihuela CJ (201 1) Cellular senescence increase expression of bacterial ligands in the lungs and is positively correlated with increased susceptibility to pneumococcal pneumonia. Aging Cel1 10: 798-806.

20. Choi CW, Lee YG, Kwon SO, Kim HY, Lee JC, et al. (201 2) Analysis of Streptococcus pne11111011iae secreted antigens by immuno-proteomic approach. Diagn Microbiol Infect Dis 72: 318-27.

21 . Kobe B, Guncar G, Buchholz R, Huber T, Maco B, et al. (2008) Crystallography and protein-protein interactions: biological interfaces and crystal contacts. Biochem Soc Trans 36: 1438-41 .

22. Pierri CL, Parisi G, Porcell i V (2010) Computational approaches for protein function prediction: ombined strategy from multiple sequencealignment to molecular docking-based virtual screening. Biochim Biophys Acta 1804: 1695-712.

23. Mook-Kanamori B, Geldhoff M, Troost D, van der Poll T, van de Beek D (2012) Characterization of

134

Cltapter 6

a pneumococcal meningitis mouse model. BMC Infect Dis 12: 71 .

24. Fitch, M.T. and D. van de Beek (2007) Emergency diagnosis and treatment of adult meningitis. Lancet Infect Dis 7: 191-200.

25. Mook-Kanamori BB, Geldhoff M, van der Poll T, van de Beek D (201 1 ) Pathogenesis and pathophysiology of pneumococcal meningitis. Clin Microbial Rev 24: 557-91 .

26. Segal S, Pollard AJ (2004) Vaccines against bacterial meningitis. British Medical Bulletin 72: 65-81 .

27. Weisfelt M, de Gans J, van der Poll T, van de Beek D (2006) Pneumococcal meningitis in adults: new approaches to management and prevention. Lancet Neurol 5: 332-42.

28. Metersky ML, Ma A, Bratzler OW, Houck PM (2004) Predicting bacteremia in patients with community-acquired pneumonia. Am J Respir Crit Care Med 169: 342-7.

29. Musher OM, Alexandraki I, Graviss EA, Yanbeiy N, Eid A, et al. (2000) Bacteremic and nonbacteremic pneumococcal pneumonia. A prospective study. Medicine (Baltimore). 79: 210-21 .

30. Pesola GR, Charles A (1992) Pneumococcal bacteremia with pneumonia. Mortality in acquired immunodeficiency syndrome. Chest. 1 01 : 150-5.

31. Hikiji H, Ishii S, Shindou H, Takato T, Shimizu T (2004) Absence of platelet-factor receptor protects mice from osteoporosis following ovariectomy. J Clin Invest. 1141 : 85-93.

32. Marina Z (2004) Absence of PAF actions increases angiogenesis. Br J Pharmacol. 141 : 1085-1086.

33. Shimada S, Kawaguchi-Miyashita M, Kushiro A, Sato T, Nanno M, et al. (1999) Generation of polymeric immunoglobulin receptor-deficient mouse with marked reduction of secretory IgA. J Immunol. 163: 5367-73.

34. Finn-Eirik Johansen, Marcela Pekna, Inger Natvig Norderhaug, Bj0m Haneberg, Max Albert Hietala, et al. (1999) Absence of Epithelial Immunoglobulin a Transport, with Increased Mucosa! Leakiness,

in Polymeric Immunoglobulin Receptor/Secretory Component-Deficient Mice. J Exp Med. 190: 915-922.

35. Duncan GS, Andrew DP, Takimoto H, Kaufman SA, Yoshida H, et al. (1999) Genetic evidence for functional redundancy of Platelet/Endothelial cell adhesion molecule-1 (PECAM-1) : CD31-deficient mice reveal PECAM-1-dependent and PECAM-1-independent functions. J Immunol. 1 62: 3022-30.

36. Wu Y, Welte T, Michaud M, Madri JA (2007) PECAM-1 : a multifaceted regulator of megakaryocytopoiesis. Blood. 110: 851-9.

37. Petsalaki E, Stark A, Garcia-Urdiales E, Russel RB (2009) Accurate prediction of peptide binding sites on protein surfaces. PLoS Comput Biol 5: e1000335.

38. Albelda SM, Muller WA, Buck CA, Newman PJ (1991) Molecular and cellular properties of PECAM-1 (endoCAM/CD31): a novel vascular cell-cell adhesion molerule. J Cell Biol 1 14: 1059--68.

39. Giffroy D, Courtoy PJ, Vaerman JP (2001) Polymeric IgA binding to the human pigR elicits intracellular signalling, but fails to stimulate pigR-transcytosis. Scand J Immunol 53: 56-64.

40. Agarwal V, Asmat TM, Dierdorf NI, Hauck CR, Hammerschmidt S (2010) Polymeric immunoglobulin receptor- mediated invasion of Streptococcus p11c11111011iae into host cells requires acoordinate signaling of SRC family of protein-tyrosine kinases, ERK, and c-Jun N-terminal kinase. J Biol Chem 285: 35615-23.

41 . Elm C, Braathen R, Bergmann S, Frank R, Vaerman JP, et al. (2004) Ectodomains 3 and 4 of human human polymeric Immunoglobulin receptor (hpigR) mediate invasion of Streptococcus p11e11111011iae into the epithelium. J Biol Chem. 279: 6296-304.

42. Cardone MH, Smith BL, Mennitt PA, Mochly-Rosen D, Silver RB, Mostov KE (1996) Signal transduction by the polymeric immunoglobulin receptor suggests a role in regulation of receptor transcytosis.

135

Chapter 6

136

APPENDIX I

N ederlandse samenvatting

- 137 -

Appendix I

Ondanks de introductie van antimicrobacteriele middelen en vaccinaties blijft

bacteriele meningitis nog steeds een wereldwijd levensbedreigende ziekte met een

hoge morbiditeit en mortaliteit. De Gram-positieve bacterie Streptococcus pneumoniae is de hoofdveroorzaker van deze ziekte (1). Bacteriele meningitis is een ontsteking

van de hersenvliezen, de lagen die de hersenen en het ruggenmerg bedekken en

beschermen. De ontsteking wordt veroorzaakt door bacterien die het centrale

zenuwstelsel binnendringen en zich vervolgens verspreiden naar de hersenen. De

bloed-hersenbarriere is een specifiek vasculair systeem, dat de hersenen scheidt

van de bloedsomloop en deze barriere heeft belangrijke functies op het gebied van

de bescherming van de hersenen en de voedselvoorziening naar het brein (2,3).

De bloed-hersenenbarriere bestaat uit een laag endotheelcellen welke zich aan de

binnenkant van bloedvaten bevindt (4). Om de hersenen binnen te dringen, moeten

de ziekteverwekkers van het hersenvlies dus eerst door deze bloed-hersenbarriere.

Om hier doorheen te dringen hebben de pathogenen verschillende strategieen

ontwikkeld. Het doel van dit promotieonderzoek was te onderzoeken waar en hoe

interacties tussen S. pneumoniae en de bloed-hersenbarriere plaatsvinden voordat

deze wordt doorbroken en het hersenvlies ontstoken raakt.

Het is bekend dat S. pneumoniae (de pneumokok) zich kan hechten aan

endotheelcellen van de bloedvaten van de hersenen (5-8), maar een daadwerkelijke

interactie tussen S. pneumoniae en de endotheelcellen van de hersenen is nirnmer

in een levend organisme (in vivo) aangetoond. Derhalve hebben wij onderzocht of

S. pneumoniae door de bloed-hersenbarriere binnendringt door zich te hechten aan

het vasculaire endotheel en of de pneumokok voorkeuren heeft voor bepaalde

bindingsplaatsen gevormd door zogenaarnde adhesiernoleculen aan de bloed

hersenbarriere. Hoofdstuk een bevat een kort overzicht van wat er bekend is over

bovengenoemde processen en wat heeft gediend als startpunt voor het experirnentele

onderzoek.

Het tweede hoofdstuk laat zien dat S. pneumoniae zich inderdaad hecht aan

het endotheel van de bloed-hersenbarriere. Daamaast hebben wij in dit hoofdstuk

la ten zien dat sommige bacterien niet met het endotheel van de bloed-hersenbarriere

co-lokaliseren, wat impliceert dat zij deze barriere zijn overgestoken. Derhalve lijkt

het hechten aan het micro-vasculaire endotheel de manier waarop S. pneumoniae de

bloed-hersenbarriere passeert. Dit proces vindt plaats in zowel de subarachno'idale

ruimte, de hersenschors als het septum. De huidige wetenschappelijke consensus

gaat ervan uit, dat de choroid plexus de plek is waar de belangrijkste bacteriele

invasie plaatsvindt die de bacteriele meningitis veroorzaakt (9, 10, 11). We hebben

echter alleen in latere stadia van een pneumokokkeninfectie de pneumokokken in

de choroid plexus aan kunnen tonen. Gedurende het hele verloop van de infectie

namen de leukocyten in de hersenen slechts minimaal toe, terwijl de rnicroglia en

139

Appendix I

astrocyten in de hersenen al binnen een uur na infectie actief werden. Dit toont aan dat de hersenen de mogelijkheid bezitten om een lokale immuunreactie te ontwikkelen tegen een bacteriele infectie.

Bovenstaand onderzoek gaf aanleiding om de moleculaire mechanismen te onderzoeken die dit proces van bacteriele hechting aan de bloedvaten vormgeven. Het lijkt er op, dat de bacterien het mechanisme van transcytose gebruiken om zich eerst te hechten aan het endotheel, vervolgens binnen te dringen en zich uiteindelijk te verspreiden door menselijke cellen (6, 12). De specifieke receptoren op de membranen van de endotheelcellen die betrokken kunnen zijn bij invasieve pneumokokkenziekte (IPZ) omvatten onder andere de zogenoemde Platelet Activating Factor (Bloedplaatjes Activerende Factor) Receptor PAFR. PAFR is een G-prote'ine gekoppelde receptor van PAF, dat een intermediair is bij verschillende pathologische processen zoals allergie, astma, septische shock, arteriele trombose en ontstekingen (13, 14). Er wordt gedacht dat PAFR betrokken is bij het hechten van S. pneumoniae aan menselijke endotheelcellen, waardoor deze receptor de passage van bacterien door de endotheelcellen faciliteert, met een invasieve ziekte tot gevolg (5, 6, 7). Ondanks veel eerder onderzoek zijn er echter geen eenduidige data betreffende de exacte rol van PAFR in IPZ voorhanden.

In hoofdstuk 3 hebben wij derhalve de bestaande literatuur over de wisselwerkingen tussen S. pneumoniae en andere ziekteverwekkers geanalyseerd. Tevens zijn wij ingegaan op onderzoeken naar menselijke genetische varianten van PAFR met betrekking tot IPZ. De eerdere onderzoeken in cellen en levende organismen hebben la ten zien, dat PAFR hoogstwaarschijnlijk betrokken is bij de ontwikkeling van IPZ. Er is echter geen hard bewijs geleverd, dat S. pneumoniae in deze organismen ook echt bindt aan PAFR. Wanneer PAFR genetisch verwijderd of chemisch veranderd wordt, is S. pneumoniae namelijk nog steeds in staat om te binden aan menselijke cellen. Dit impliceert dat pneumokokken de mogelijkheid hebben zich te hechten aan alternatieve receptoren.

In hoofdstuk 4 bewijzen wij dat directie interactie tussen PAFR en S. pneumoniae

niet waarschijnlijk is. Een blokkade van de PAFR leidt weliswaar tot een significante afname van binding van pneumokokken aan bloed-hersenbarrierecellen in weefselkweek, daarentegen laten immunofluorescentie analyses van hersenweefsel van intraveneus ge'infecteerde muizen zien, dat S. pneumoniae niet co-lokaliseert met PAFR in de bloedvaten. Dit hebben we bevestigd metbehulpvan immunofluorescentieanalyses in menselijke bloedvatenendotheelcellen die waren ge'incubeerd met pneumokokken. Concluderend denken wij dat, hoewel PAFR zeker betrokken zal zijn bij binding van S. pneumoniae aan endotheelcellen, een directe associatie van pneumokokken met PAFR niet waarschijnlijk is.

De zogenoemde poly immunoglobuline receptor (pigR) regelt het transport van immunoglobuline door mucosale epitheelcellen (15, 16) en het is bekend dat deze

140

Appendix I

receptor ook betrokken is bij de binding van pneumokokken aan het epitheel van de menselijke nasopharynx (12). Omdat onbekend was of pigR ook een rol speelt bij de binding van S. pneumoniae in vivo, hebben wij eveneens de rol van pigR in de binding van S. pneumoniae aan endotheelcellen van de bloed-hersenbarriere onderzocht. Immunofluorescentie-analyses van hersenweefsel van intraveneus ge'infecteerde muizen laten zien, <lat S. pneumoniae co-lokaliseert met pigR. Co-lokalisatie tussen bacterien en pigR is tevens vastgesteld bij bloedvatendotheelcellen in kweekvaatjes. Tevens vonden wij biochemisch bewijs <lat S. pneumoniae zich kan hechten aan pigR. Een combinatie van beide resultaten laat duidelijk zien <lat pigR wellicht een rol speelt als directe receptor voor de binding van S. pneumoniae aan het endotheel van de bloed-hersenbarriere.

In dit proefschrift hebben wij als laatste aandacht besteed aan PECAM-1. PECAM-1 is een van de belangrijkste adhesiemoleculen in bloedvatendotheelcellen en het is bekend dat deze receptor een cruciale rol inneemt bij neuro-ontstekingen en daaropvolgende verstoringen van de bloed-hersenbarriere (17, 18). Hoofdstuk 5 gaat in op de rol van PECAM-1 in S. pneumoniae gemedieerde meningitis pathogenese. In het bijzonder is onderzocht of PECAM-1 een receptor voor de hechting van S. pneumoniae aan het endotheel van de bloed-hersenbarriere is. Immunofluorescentieanalyses van hersenweefsel van intraveneus ge·infecteerde muizen laten zien, <lat de meeste S. pneumoniae cellen co-lokaliseren met PECAM-1. Eenzelfde co-lokalisatie hebben wij in vitro in met pneumokokken ge'incubeerde endotheelcellen aangetoond. Een blokkade van PECAM-1 in deze endotheelcellen zorgde voor een significante afname van bacteriele binding. Dit wijst erop dat PECAM-1 functioneel betrokken is bij binding van S. pneumoniae aan menselijke endotheelcellen. Tenslotte hebben wij biochemisch bewijs geleverd dat S. pneumoniae kan binden aan PECAM-1.

Resumerend hebben we laten zien wat de rol van PAFR, pigR en PECAM-1 is in de binding van S. pneumoniae aan endotheelcellen, in weefselkweek en in een muizenmodel van meningitis. PAFR speelt waarschijnlijk een indirect rol in dit proces, terwijl de pneumokok met zowel pigR als PECAM-1 een directe interactie aangaat. Of deze kennis kan worden vertaald naar therapeutische mogelijkheden om het ontstaan respectievelijk de verdere ontwikkeling van meningitis te blokkeren zal uit vervolgonderzoek moeten blijken.

141

Appendix I

Referenties

1 . Brouwer MC, Tunkel AR, van de Beek D (2010) Epidemiology, diagnosis, and antimicrobial treatment of acute bacterial meningitis. Clin Microbiol Rev 23: 467-92.


3. Woehr! B, Klein M, Grandgirard D, Koedel U, Leib S (2011 ) Bacteria l meningitis: current therapy and possible future treatment options. Expert Rev Anti Infect Tuer 9: 1 053-65.

4. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1995) Stnytococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377: 435-8.

5. Ring A, Weiser JN, Tuomanen EI (1998) Pneumococcal trafficking across the blood-brain barrier. Molecular analysis of a novel bidirectional pathway. J Clin Invest 1 02: 347-60.

6. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen EI (2005) beta-Arrestin 1 participates in platelet-activating factor receptor-mediated endocytosis of Streptococcus pneumoniae. Infect Immun 73: 7827-35.

7. Hoffman 0, Weber RJ (2009) Pathophysiology and treatment of bacterial meningitis. Tuer Adv Neurol Disord 2: 1-7.

8. Tenenbaum T, Essmann F, Adam R, Seibt A, Janicke RU, et al. (2006) Cell death, caspase activation, and HMGBl release of porcine choroid plexus epithelial cells during Streptococcus suis infection in vitro. Brain Res 1100 :1-12.

9. Tenenbaum T, Papandreou T, Gellrich D, Friedrichs U, Seibt A, et al. (2009). Polar bacterial invasion and translocation of Streptococcus suis across the blood-cerebrospinal fluid barrier in vitro. Cell Microbiol 1 1 : 323-36.

10. Zhang JR, Mostov KE, Lamm ME, Nanno M, Shimida S, et al (2000) The polymeric immunoglobulin receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102: 827-37.

1 1 . Shukla SD (1992) Platelet-activating factor receptor and signal transduction mechanisms. FASEB J 6:2296-301

12. Honda Z, Ishii S, Shimizu T (2002) Platelet-activating factor receptor. J Biochem 131:773-9.

13. Kim KJ, Malik AB. (2003) Protein transport across the lung epithelial barrier. Am J Physiol Lung Cell Mol Physiol 284: L247-59.

14. Asano M, Komiyama K. (2011) Polymeric immunoglobulin receptor. J Oral Sci 53: 147-56.

15. Newman PJ, Newman DK (2003) Signal transduction pathways mediated by PECAM-1: new roles for an old molecule in platelet and vascular cell biology. Arterioscler Thromb Vase Biol 23: 953-64


142

APPENDIX II

List of publications

- 143 -

List of publications

• Iovino F, Brouwer MC, van de Beek D, Molema G, Bijlsma JJE

Appendix II

Signalling or binding: the role of the platelet activating factor receptor in invasive

pneumococcal disease.

Cellular Microbiology (2013) 15 (6): 870-81. Review

• Iovino F, Orihuela CJ, Moorlag HE, Molema G, Bijlsma JJE

Interactions between blood-borne Streptococcus pneumoniae and the blood-brain

barrier preceding meningitis.

Plos One (2013) 8 (7): e68408

• Gradstedt H, Iovino F, Bijlsma JJE

Streptococcus pneumoniae invades endothelial host cells via multiple pathways and

is killed in a lysosome dependent manner.

Plos One (2013) 8 (6): e65626

• Iovino F, Molema G, Bijlsma JJE

Novel insights in the roles of platelet-activating factor and poly immunoglobulin

receptors in Streptococcus pneumoniae adhesion to the brain microvascular

endothelium.

Submitted

• Iovino F, Molema G, Bijlsma JJE

Platelet endothelial cell adhesion molecule-1 (PECAM-1), a putative receptor for

the adhesion of Streptococcus pneumoniae to the vascular endothelium of the blood

brain barrier.


145

Appendix III

APPENDIX III

Acknowledgements Ringraziamenti

- 147 -

Appendix Ill

Acknowledgements /Ringraziamenti

First of all, I want to thank Jetta, Ingrid and Jan Maarten for their supervision, for supporting me and my work throughout my entire PhD. I would like to thank the members of the thesis assessment committee, Prof. dr. J. Tommassen, Prof. dr. H.E. de Vries and Prof. dr. H.C. van der Mei, for reading and approving my thesis. I would like to thank my collaborators, Dr. Carlos J. Orihuela and Henk Moorlag. Carlos, thank you for giving me the opportunity to spend two months in your lab during my first year of PhD, your input has been precious to build up this thesis. Henk, you have been introduced me by Ingrid as "the technician with the golden hands", and during my PhD I realized that it is indeed true! I made under your supervision my first steps in the world of cell culturing and tissue staining, and most importantly you were always available for helping me during my experiments, thank you very much.

Thanks to my fella Peter B. for writing the dutch general summary of the thesis. Thank you to the reviewer who rejected my first experimental work. Sometimes rejections help more than positive responses. As in life, what does not kill you makes you stronger.

Thanks to all the Mol Bae members, working with all of you in the lab has been great. Thanks to Viv (and Geoff), Lakshmi, Emma, Corinna, Annette, Eleni (and Mark), Ruben, Ewoud, Magda, Monica, Xiaomei, Dennis, Rense, Giorgio, Gosia, Girbe, Jolanda, Jessica, Sjouke, Nana Arna. A special thanks to Carmine, my first Italian friend in Groningen. Only two words can describe how I have appreciated all the time that we spent together, inside and outside the lab: bella zio! And what can I say about you guys, Vahid and Henrik, my paranymphs, colleagues and dear friends: together we made up a great team inside and outside the lab, the Strep Boys. I could fill a lot of pages describing all the crazy adventures we had together, the "dream team" dinners and parties, and all the laughing we had together at work. I'm happy to leave this office, because without you this office is like a "broodje" without its "frikandel", I'm sure you understand the metaphor. I will never forget anything, and never forget this guys: we have been, we are and we will always be the Strep Boys! During my stay in Groningen, I had the pleasure to meet people who became special friends. Thanks to Peter B., Melanie, Peter N., Martti, Vaishali, Pranav, Despina and Kostas. All the time we spent together during our "non-working" life was great, thanks for all the happy, crazy and unforgettable moments. Peter B., Peter N. and Martti, I could not write these pages without mentioning our crazy weekend in Dresano/Milan/Pavia: maxi cocktails, the Skorpion Club, Peter B. "sneezing" in my car, Peter N. and Martti in their typical "drunk-dressing up" (and we know what it

149

Appendix I I I

means) in the back seat of my father's car, the derby in San Siro. And of course, the crazy "Eurovision song contest" night. And many other moments. Thank you guys!

A special acknowledgement to my other fella, Peter N. First of all, thank you for being one of my paranymphs. And besides that, thank you for being my "schatje". I added this part after you went through the acknowledgements . . . yeah!

This thesis is a mirror of what I am, and I really believe that we have to fight for what we believe in. This is what I got from my family. Thanks to my mom and my dad, who have always supported and encouraged me. And most of all, thank you for always believing in me. Thanks to Filippo, my little brother, because we have always supported each other during the recent years, especiaIIy in the very sad moments. Thanks to Chiara, my big sister, Martin and Clelia, together we laughed and cried a lot during the last few years, thanks for all your support. Thanks to my grandfather Giancarlo, I strongly believe that I love research so much because of you.

This time in Groningen taught me that sometimes life is a bitch. I lost two pieces of my heart while I was here. Grandma and my uncle Bibo, wherever you are now, I have never felt you so close to me as in the last years. Thanks for being with me in my dreams at night; this has been a great force for doing my work as well as I could.

And of course, thanks to all my family for aII the great support: Zio Claudio & Zia Rosy, Zia Grazia & Zio Ernesto, Andrea & Daniele, Zia Rosetta, Zio Paolo & Nonna Jolanda. Thanks to Rita and Luigi, for aII your kindness, support and love.

Questa tesi e lo specchio di quello che sono, io credo veramente che bisogna combattere per

quello in cui si crede. Questa e quello che la mia Jamiglia mi ha sempre trasmesso. Grazie all

mia mamma e al mio papa che mi hanno sempre sostenuto e incoraggiato. E soprattutto, grazi

di credere sempre in me. Grazie a Filippo, il mio fratellino, perche ci siamo sempre sostenut

a vicenda negli ultimi anni, soprattutto nei momenti piu tristi. Grazie a Chiara, la mia sorel

Iona, Martin e Clelia, abbiamo riso e pianto tanto insieme in questi ultimi anni, grazie per

tutto il vostro sostegno. Grazie a mio nonno Giancarlo, credo fortemente che amo cosi tan.to

la ricerca per merito tuo.

Questa periodo a Groningen mi ha insegnato che talvolta la vita e una stronza. Ho perso due

pezzi del mio cuore mentre ero qui. Nonna e Bibo, ovunque voi siate adesso, non vi ho mai

sentito cosi vicini come in questi ultimi anni. Grazie per essere insieme a me nei miei sogni

notturni, questo mi ha data la forza per fare al meglio il mio lavoro. E ovviamente, grazie a

tutto il resto della mia famiglia: Zia Claudio & Zia Rosy, Zia Grazia e Zia Ernesto, Andrea

& Daniele, Zia Rosetta, Zia Paolo & Nonna Jolanda. Grazie a Rita e Luigi, per la vostra gen

tilezza, il vostro supporto e il vostro affetto.

Thanks to all my italian friends. It hasn't been the same as when I was still in Italy with you guys. The long distances allowed us to see each other only a few times

150

Appendix lll

every year. But still, friendship and true relationships are not built on quantity but on quality. Thanks to Giova, Barzo, Biki, Lando, Cava, Tommy, Arrigo, Cape, Italo, Andrea, Daniele, Matteo and Jessica, immortal friends who are always present. And thank you Gabry, "friendship is a difficult thing to find, true friendship is almost impossible, thanks for making possible the impossible" . Thanks Lele, you are one of the biggest presents that my uncle Bibo left for me.

Grazie a tutti i miei amici in Italia. Non e stato come quando ero li con voi. La lunga distanza ha pennesso di vederci solo poche volte l'anno. Ma nonostante ci6, la vera amicizia si basa sulla qualita e non sulla quantitti. Grazie a Giova, Barzo, Biki, Lando, Cava, Tommy, Arrigo, Cape, Andrea, ltalo, Matteo, Daniele, Jessica. E grazie Gabry, "l'amicizia e una cosa difficile da trovare, quella vera e quasi impossibile, grazie per aver reso possibile l'impossibile". Grazie Lele, tu sei uno dei regali pht belli che Bibo mi ha lasciato.

Thanks to my Dutch water polo team Trivia. Water polo is not a sport, but a way of life, and it was wonderful to have the opportunity to continue this sport outside of Italy. A special thanks to my coach Engel and to my team mates Ruben and Lude.

And now is your tum in the list, Melania, amore mio. I still remember the night when we had our first date together. It was freezing outside (strange) and I was waiting for you at the Grote Markt. I was there before you (strange). After a few minutes I saw you, cycling on that funny mountain bike that was too big for you. That evening I had promised to myself that I would have returned home early, because the day after I had a busy day in the lab. I came home at 2 o' clock in the night, and everything between us started. The word "thanks" to you is not sufficient. It seems amazing that we have always lived very close to each other in Italy, but we met in Groningen. I will be eternally grateful to Groningen for making this happen. I could fill at least 10 pages about you and us, but you would probably get bored of reading too many romantic words, I know you. So I will be very brief, with just a few words: Thank you for supporting me, for kissing and hugging me every morning and every night, for the funny voices that we always make to each other, for loving me, for making me conscious of the real meaning of the word "love". Before meeting you, I always called "home" my hometown, where my family and friends are. Now my home is you, "where" is not important.

I love you, you are my life.

E ora e il tuo turno nella Lista, Melania, amore mio. Ancora ricordo la sera del nostro primo appuntamento. Faceva freddo fuori (strano) e ti stavo aspettando a Grote Markt. Ero li prima di te (strano). Dopo pochi minute ti ho vista che stavi pedalando su quella mountain bike che era troppo grande per te. Quella sera mi ero promesso di rientrare a casa presto, perche il giorno dopo avrei avuto una giornata pesante al lavoro. Rientrai a casa alle 2 di notte, e da li e

151

Appendix l l l

iniziato tutto tra noi. La parola "grazie' per te non e sufficiente. Sembra incredibile che abbiamo sempre vissuto cosi vicini in Italia, ma ci siamo incontrati a Groningen. Saro sempre grato a Groningen per aver reso questo possibile. Potrei riempire almeno 10 pagine su di te e su di noi, ma probabilmente ti stuferesti di leggere trope parole mielose, ti conosco. Quindi sar6 breve, solo poche parole: Grazie per il tuo supporto, grazie dei baci e degli abbracci ogni mattina e ogni sera, grazie delle vocine che ci facciamo sempre, grazie di amarmi, grazie per avermi reso cosciente del vero significato della parola "amore". Prima di incontrarti, ho sempre chiamato "casa" la mia cittti, dove ci sono la mia famiglia e i miei amici. Ora casa mia sei tu, il "dove" non e importante.

Ti amo, sei la mia vita.

152

Documents

University of Groningen Streptococcus pneumoniae ... · brain microvascular endothelial cells, and that PECAM-1 forms a double receptor with plgR on the plasma membr ane of endothelial