‘0 LIBRARY
Michigan State
Univereitv

. _ 'fl-N‘AF-

 

.— “—1--

 

 

This is to certify that the
thesis entitled

Exposure to P. aeruginosa and purified LPS alter CFTR—
dependent ion conductance in cultured 2WT2 epithelial cells in
a time and dose dependent fashion.

presented by

Michael D. Haenisch

has been accepted towards fulfillment
of the requirements for the

Master of degree in Physiology
Science

 

 

 

 

 

 

MSU is an Affirmative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5l08 K:IProj/Acc8.Pres/ClRC/DateDue.indd

 

EXPOSURE TO P. AERUGINOSA AND PURIFIED LPS ALTER CFTR-
DEPENDENT ION CONDUCTANCE IN CULTURED ZWUMHELIAL CELLS
IN A TIME AND DOSE DEPENDENT FASHION

By

Michael D. Haenisch

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of .

MASTER OF SCIENCE

Physiology

2009

Abstract

EXPOSURE TO P. AERUGINOSA AND PURIFIED LPS ALTER CFTR-
DEPENDENT ION CONDUCTANCE IN CULTURED ZWTZ EITHELIAL CELLS
IN A TIME AND DOSE DEPENDENT FASHION

By
Michael D. Haenisch

The most common heritable genetic disease in America, cystic fibrosis
(CF), is caused by mutations in the CF transmembrane conductance regulator
(CFTR), a chloride channel that interacts with and regulates a number of other
proteins. Pseudomonas aeruginosa infects 80% of patients. It is not known how
mutations in the chloride channel allow for colonization of the majority of
patients by a single pathogen. The hypothesis that CFTR interacts with toll-like
receptor 4 (TLR4) to phagocytise bacteria was tested. A competitive antagonist of
TLR4 did not alter the rate of phagocytosis of P. aeruginosa by cultured epithelia;
however, bacteria formed antibiotic resistant microcolonies in these experiments
meaning the actual rate of phagocytosis was not measured. Cultured epithelia
treated with P. aeruginosa had increased CF TR channel activity.
Lipopolysaccharide (LPS) purified from P. aeruginosa also increased CFTR
channel activity in a dose dependent manner; however, shorter exposure to LPS
resulted in reduced CF TR channel activity. This data is consistent with a model in
which CFTR is removed from the plasma membrane during phagocytosis of P.
aeruginosa followed by recruitment of the channel to the membrane to replace
channels removed during phagocytosis. More studies are needed to confirm this

model, but this is the first report of a bacterial product causing a biphasic time-

dependent and a dose-dependent alteration of CFTR channel activity.

 

 

Dedicated to everyone who gave me a shot: first second or otherwise.

TABLE OF CONTENTS

List of Figures ............................................................................... vi
Key to symbols and abbreviations ...................................................... viii
Introduction ................................................................................... 1
History ................................................................................ 2
Clinical perspective ............................................................... 5
Innate immunity ................................................................... 9
Theories on immunodeficiency ................................................ 12
Bacteria ............................................................................. 16
This study ........................................................................... 24
Methods ..................................................................................... 30
Materials and reagents ......................................................... 30
Cell culture ........................................................................ 31
Bacterial growth .................................................................. 32
Gentamicin protection assay ................................................... 35
Iodide efflux ..................................................................... 36
Data analysis ....................................................................... 39
Results ....................................................................................... 40
TLR4 antagonist LPS does not alter phagocytosis of P.
aeruginosa .......................................................................... 40
P. aeruginosa PA01 forms antibiotic resistant colonies
on epithelial cells .................................................................. 42
Alterations in CFTR mediated efflux in response to
treatment with bacteria ............................................................ 43
Discussion ................................................................................... 49
P. aeruginosa strain PA01 adheres to respiratory epithelial
cells but is not frequently phagocytised ....................................... 50
MKLPS does not effect the rate of adhesion or
phagocytosis ........................................................................ 52
P. aentginosa PA01 forms antibiotic resistant microcolonies
on pulmonary epithelial cells ................................................... 54

The ORION iodide selective electrode can detect changes in

CAMP stimulated iodide efflux in cells that overexpress

CF TR, but may not be sensitive enough to accurately detect

baseline readings in these cells or others expressing

physiologically relevant levels of CFTR ....................................... 55

iv

 

Sustained exposure to LPS during the entire period of sample
replacement causes a dose-dependent increase in CFTR

channel activity .................................................................... 57
Acute treatment with LPS reduces CFTR channel activity .................. 59
Figures ........................................................................................ 61
References ................................................................................... 88

List of Figures

Figure 1 Proposed structure of the CF TR showing its major
functional domains ......................................................................... 61

Figure 2 Age correlated incidence of airway infections in cystic
fibrosis airways ............................................................................. 63

Figure 3 Depiction of two theories explaining how alterations in the

ASL could lead to chronic bacterial infections ......................................... 64
Figure 4 Conserved structure of LPS .................................................... 66
Figure 5 Model of a possible interaction between CFTR and TLR4

during phagocytosis of P. aeruginosa ................................................... 67
Figure 6 Timecourse of efflux experiments ............................................. 68

Figure 7 Growth curve of P. aeruginosa PA01 grown in tryptic soy
broth showing entry into log phase growth ............................................. 69

Figure 8 Growth curve of P. aeruginosa PA14 grown in LB showing
entry into log phase growth ............................................................... 70

Figure 9 Optical density correlates to cell density in P. aeruginosa
PA01 ......................................................................................... 71

Figure 10 Optical density correlates to cell density in P. aeruginosa
PA14 ......................................................................................... 72

Figure 11 Treatment with the TLR4 antagonist MKLPS had no
effect on recovery of total bacteria, adhered bacteria, or
phagocytised bacteria ...................................................................... 73

Figure 12 Merged image of fluorescent and differential interference

contrast image of S9 epithelial cells inoculated for 3 hours with

GFP-labeled P. aeruginosa PA01 after treatment with 300pg/mL

gentamicin ................................................................................... 74

Figure 13 Analysis of ISE function when reading standards made
in distilled water ............................................................................ 75

Figure 14 Efflux rate constant in S9 cells stimulated with IOuM
forskolin measured using the ISE ........................................................ 76

vi

 

Figure 15 Efflux rate constant in 2WT2 cells stimulated with IOuM
forskolin measured using the ISE ........................................................ 77

Figure 16 Efflux rate constant in S9 cells stimulated with IOuM
forskolin measured using the ISE ........................................................ 78

Figure 17 Efilux rate constants from radionuclide efflux controls .................. 79

Figure 18 Normalized efflux rate constants from radionuclide efflux
controls ....................................................................................... 80

Figure 19 Dose response curve of CFTR mediated peak efflux in
response to increasing concentrations of LPS .......................................... 81

Figure 20 Dose response curve of CFTR mediated area under the
efflux curve in response to increasing concentrations of LPS ........................ 81

Figure 21 Increase in CFTR mediated peak efflux in response to
treatment with P. aeruginosa PA14 ...................................................... 83

Figure 22 Increase in CFTR mediated area under the efflux curve in
response to treatment with P. aeruginosa PA14 ........................................ 84

Figure 23 No significant difference in CFTR mediated peak efflux
was detected in response to brief treatment with lOOpg/mL LPS ................... 85

Figure 24 Decreased CFTR mediated area under the efflux curve
was detected in response to brief treatment with 100ug/mL LPS ................... 86

Figure 25 Model of a possible interaction between CFTR and TLR5
during phagocytosis of P. aeruginosa ................................................... 87

vii

CF

CFTR

AF508

G542X

N1303K

G551D

W1282X

mRNA

ATP

PKA

CAMP

ENaC

AP-2

EBPSO

ERM

ASF

PAMP

TLR

LPS

Key to symbols and abbreviations
Cystic fibrosis
Cystic fibrosis transmembrane conductance regulator
Deletion of phenylalanine at position 508
changed to at position 542
changed to at position 1303
changed to at position 551
changed to at position 1282
Messenger ribonucleic acid
Adeninenucleotide triphosphate
Protein kinase A
Cyclic adeninenucleotide monophosphate
Epithelial sodium channel
Adaptor protein 2
Ezrin binding protein 50
Ezrin radixin moesin
Airway surface fluid
Imunoglobin
Interleukin
Tumor necrosis factor
Pathogen associated molecular pattern
Toll-like receptor

Lipopolysaccharide

viii

NFKB

°C

MKLPS

MIC

MOI

CFU

S9

2WT2

C127

GFP

125-I

IBMX

AUC

FBS

BSA

G418

EDTA

HEPES

TSB

LB

DMEM

PBS

Antigen presenting cell

Nuclear factor kappa B.

Degrees centigrade

Mutant pentaacylated lipopolysaccharide
Minimum inhibitory concentration
Multiplicity of infection

Colony forming units

Human bronchial epithelial cells expressing normal CFTR
C127 cells expressing normal CFTR
Immortalized mouse epithelial cells
Green fluorescent‘protein

Iodine 125 (radioisotope)
Isobutylmethylxanthine

Area under the curve

Fetal bovine serum

Bovine serum albumin

Geneticin

Ethylenediaminetetraacetic acid
4-(2-hydroxyethyl)-1—piperazineethanesulfonic acid
Tryptic soy broth

Lysogeny broth

Dulbeco’s modified eagle medium

Phosphate buffered saline

ix

ATCC

SEM

OD600

DAP

R1

t1

AN OVA
min"- 1
CPM
min

ISE

Na+
Cl-
H20
TJ

N,K-pump

American type culture collection
Standard error of the mean
Revolutions per minute

Optical density at 600nm light
Diaminopimelic acid

Rate constant

Percentage of total iodide remaining at time 1
Percentage of total iodide remaining at time 2
Time 1

Time 2

Analysis of variance

Per minute (unit of rate)

Counts per minute

Minute

Ion selective electrode
Threonine

Arginine

Leucine

Sodium ion

Chloride ion

Water

Tight junction

Sodium / potassium ATPase

man

rha

abe

Ci

MRSA

Mannose

Rharnnose
Arabinose

Galactose

Phosphate group
Glucosamine
Oxygen

Amino group

Curie

Number of replicates

Methycilin resistant staphalococcus aureus

xi

INTRODUCTION

Cystic fibrosis (CF) is the most common fatal genetic disease in the United States
with an incidence of approximately 1 in 2,500 and a carrier frequency of about 1 in 20
individuals (Kerem et al, 1989; Welsh and Smith, 1995; Mateu er al, 2002). Without
treatment, patients die in early childhood; however, with aggressive supportive therapy,
patients now have a life expectancy over 30 years (Davis, 2006). The disease is caused
by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR), and to
date over 1000 mutations have been identified that cause the disease (Mateu et al, 2002).
A single mutation, a deletion of a phenylalanine residue at amino acid position 508
(AF 508), accounts for 2/3 of global CF chromosomes, and only four other mutations,
G542X, N1303K, G551D and W1282X, have frequencies over 1% (Mateu et al, 2002).
The high prevalence a single gene causing this autosomal recessive disease has led many
to speculate that heterozygotes bearing a single copy of the mutant allele may have had a
protective advantage that caused the gene to be selected for, in much the same way as
sickle cell anemia (Romeo et al, 1989; Pier, 1999). Although many questions still remain
unanswered, research into this field has unearthed a wealth of information about cell and

organismal physiology, and host-pathogen interactions.

HISTORY
First descriptions

CF was first described as a pathological condition based on autopsies done by
Dorothy Andersen in 193 8. She identified obstructed pancreatic ducts as a difference
between what became known as cystic fibrosis of the pancreas and celiac syndrome
(Andersen, 193 8). In 1946, the disease was further described as an autosomal recessive
disorder and in 1948 it was discovered that CF patients presented with abnormally high
concentrations of sodium chloride in their sweat (Andersen and Hodges, 1946;
diSantAgnese et al, 1953). This discovery laid the basis for the use of the sweat chloride
test for the diagnosis of CF, allowing for the disease to be detected earlier in life and in
patients with less severe phenotypes (Davis, 2006; Welsh and Smith, 1995). This
allowed monitoring and treatment to begin earlier, thus improving the life expectancy of
patients. Before the mutant gene was discovered, several groups described the epithelium
from CF patients as having decreased chloride permeability and increased sodium
reabsorption, two of which described chloride impermeability in respiratory epithelium,

and another in sweat ducts (Quinton, 1983; Knowles et al, 1983; Boucher er al, 1986).

 

Discovery of thegm: and molecular biology

When the gene was identified in 1989, it was named the cystic fibrosis
transmembrane conductance regulator for its ability to regulate ion transport (Kerem et
al, 1989; Riordan et al, 1989; Rommens er al 1989). A multi-group collaboration led to
the discovery of a 180,000 base pair gene on the long arm of chromosome 7 with an open
reading frame encoding a 1480 amino acid gene product. The group also identified the
AF 508 mutation (a deletion of three base pairs resulting in a deletion of a phenylalanine
residue at position 508 in a predicted nucleotide binding domain) which accounted for
70% of mutations in mRNA transcripts isolated from CF patients (Rowe et al, 2005).
Analysis of the amino acid sequence revealed two predicted membrane spanning
domains, two nucleotide binding domains similar to those found in the ATP-binding
cassette superfamily of proteins, and a central regulatory domain with phosphorylation
sites for protein kinase A and protein kinase C (Figure 1) (Riordan et al, 1989).

The protein is highly expressed in the apical membrane of epithelial cells of the
airway, sweat and pancreatic ducts, vas deferens, intestine and biliary tree (Davis, 2006).
It is not surprising therefore that these are the areas and organs most effected by the
disease, and many of the aspects of the disease can be explained by a secretory defect

caused by lack of functional CFTR. Normally the protein is formed in the rough

 

endoplasmic reticulum where it is glycosylated and transported to the Golgi apparatus
(Gelman and Kopito, 2002). The sugar residues are shortened at the Golgi apparatus and
the protein trafficks to the plasma membrane (Gelman and Kopito, 2002). Here it is
concentrated in lipid rafts and clathrin coated pits until it is removed from the membrane
(Kowalski and Pier, 2004; Dudez et al, 2008; Bradbury et al, 1994; Weixel and
Bradbury, 2001). CFTR is removed from the plasma membrane in a clathrin dependent
manner, and the AF 508 CFTR is sorted to early endosomes in a Rab5 dependent manner
(Gentzsch et al, 2004). After removal to an endosome, AF 508 CFTR is either sent to the
golgi, degraded, or recycled back to the membrane (Gentzsch er al, 2004).

At the membrane, CFTR performs several functions. Its best described function
is as a cyclic adenosine monophosphate (CAMP) activated chloride channel, but there is
strong evidence that it also has the ability to interact with and regulate other proteins.
Not only may CFTR regulate the activity of other ion channels such as the epithelial
sodium channel (ENaC), it also appears to interact with the cytoskeleton, several other
structural and trafficking proteins, and it has been suggested to regulate vesicle
trafficking within the cell (Kunzelman, 2001; Gentzsch et al, 2004). While the body of
evidence suggests that the first nucleotide binding domain of CFTR is important for its

regulation of ENaC, the other interactions are believed to be mediated through a PDZ

domain interaction sequence found at the C-terminus (Schreiber et al, 1999; Rowe et al,
2005). This sequence was shown to interact with adaptor protein 2 (AP-2) and ezrin
binding phosphoprotein 50 (EBPSO) (Kunzelman, 2001; Gentzsch er al, 2004). AP-2 is
involved in assembly of cargo into clathrin coated vesicles (Gentzsch et a1, 2004; Conner
and Schmid, 2003; Traub, 2003; Motley et al, 2003). EBPSO, a member of the ezrin
radixin moesin (ERM) family of proteins, has been shown to interact with polarized actin
filaments to maintain formation of polarized cellular structures (Kunzelman, 2001;

Morales er al, 2004).

CLINICAL PERSPECTIVE
Prawn and gmptoms

The disease is characterized by elevated sweat chloride levels, pancreatic
insufficiency, intestinal obstruction, congenital bilateral absence of the vas deferens, and
lung disease with frequent microbial infections and bronchiectasis (Davis, 2006). A
secretory defect explains many manifestations of the disease such as the failure of
electrolyte transport responsible for the lack of resorption of sodium chloride from the
sweat duct and the failure to secrete fluid into the duct of the pancreas leading to

improper secretion from that organ, pancreatitis and digestive deficiencies (Davis, 2006).

Several hypotheses have attempted to explain how CF lung disease could also be caused
by a secretory defect, but the situation in the lung is more controversial and will be
discussed in a separate section.

The most pronounced feature of the disease, pancreatic insufficiency, typically
becomes evident by 6 months of age due to failure to thrive or menconium ileus, an
intestinal blockage (Lyczak et al, 2002). In this early stage, intestinal blockage is the
most common cause of death while death due to respiratory infections and associated
respiratory collapse becoming more common in older patients (Andersen, 1938; Davis,
2006)

The leading cause of death in CF patients is lung disease (Davis, 2006).
Approximately 80-95% of CF patients become infected with Pseudomonas aeruginosa
and die from sepsis or respiratory failure caused by chronic respiratory infections
(Lyczack et al, 2002). Early in life, the lungs of CF patients become colonized with a
narrow range of opportunistic microbial pathogens (Chmiel and Davis, 2003). The
progression of these infections is curious because of its persistence and the narrow range
and temporal progression of colonization by a few specific species (Figure 2). Immunity
is clearly compromised although the reasons for this are a topic of debate. Despite

vigorous use of antimicrobial treatments, patients continue to maintain or relapse into

infection (Chmiel and Davis, 2003). The mucus of the airways thickens and cannot be
cleared properly (Davis, 2006). The damage caused by inflammation from chronic
infection and the buildup of debris eventually causes clogging of small airways (Davis,
2006). Clinically, this progression leads to fatal incidence of sepsis or respiratory
collapse (Davis, 2006). What is not well known is the mechanism by which defective

CFTR allows for these frequent and persistent microbial infections.

Treatments

Although the pancreatic insufficiency can be treated with enzyme replacement
therapy and nutritional supplementation, approximately 20% of children and 40% of
adults with CF can be diagnosed as having nutritional failure (Davis, 2006). This leads to
a generalized failure to thrive as well as several specific nutritional deficiencies. Despite
supportive nutritional therapies, the pancreas is still damaged in CF patients which can
lead to insulin dependent diabetes as the endocrine functions of this organ are impaired
(Davis, 2006).

The goals of treatments for pulmonary complications are to eradicate bacteria,
reduce inflammation, and clear the airways of mucus to enhance pulmonary function

(Davis, 2006). Antibiotic chemotherapy is used to treat recurrent infections; however,

despite vigorous administration, many patients are unable to clear infections of P.
aeruginosa and B. cepacia which are broadly resistant to antibiotics (Hart and
Winstanley, 2002). It is accepted that the intense inflammation that occurs in the lungs of
CF patients contributes to the decline in pulmonary function, but clinical trials have yet to
demonstrate an anti-inflammatory treatment regimen that results in long-term
improvements in pulmonary function (Balfour-Lynn and Dinwiddie, 1996). Clearance of
mucus is achieved by a variety of techniques aimed at detaching mucus and lowering its
viscosity. Chest percussion and forced coughing and breathing are the traditional
techniques used for breaking up tightly adhered mucus, but work has been done to
develop rapid chest compression vests and breathing apparatuses for more effective and
widely available use (Welsh and Smith, 1995). Inhalation of nebulized DNase has been
used to lower the viscosity of CF mucus and several attempts have been made to develop
treatments with inhaled compounds that increase osmosis and thus lower mucus viscosity
(Jaques et al, 2008).

The research community is also still trying to find ways to rescue CFTR before it
is destroyed within the epithelial cell. CF TR with the AF508 mutation can still function
as a chloride channel, but rarely gets a chance to do so since it is detected by molecular

chaperones and sent to the proteosome for degradation (Sun et al, 2008). Many still hope

to find a compound that can be used to inhibit these chaperones and thus allow the AF508
CFTR to reach the plasma membrane where it could function normally. However, the
loftiest treatment goal is the development of a gene therapy method that could be used to
replace or complement the mutant gene, allowing cells to express wild-type protein
(Wison, 1995). Work continues and treatments continue to improve, but it may still be

some time until anything approaching a true cure is found.

INNATE IMMUNITY

The innate immune system comprises the elements of immunity that are
constitutively present before the activation of adaptive immunity which involves
activation of T and B lymphocytes and antibody production (Bals et al, 1999). This
system is linked to adaptive immunity through its determination of antigen presentation
(Fearon and Locksley, 1996).

In humans, approximately lOmL of fluid cover the surface area of the lung which
is about the area of half a soccer field (Bals et al, 1999). This airway surface fluid (ASF)
is comprised of a layer of mucus that is positioned on top of an aqueous layer. Secretions
from mucus glands, goblet cells, clara cells, and type II pneumocytes comprise the upper

layer and contain an array of antimicrobial compounds and peptides including defensins,

cathelicidins, lactoferrin, lysozyme, complement, secretory IgA, protease, and
phospholipase A2 (Donaldson and Boucher, 2006; Bals et al, 1999; Hiemstra, 2007;
Gerritsen, 2000). Microbes are captured in the overlying mucus layer and removed from
the lung by movement of the mucus by cilia positioned on the underlying epithelium.
This system is effective enough to maintain sterility in the lung despite colonization of
the upper airway (Hart and Winstanley, 2002).

Epithelial cells, alveolar macrophages and neutrophils are considered part of the
innate immune system (Gerritsen, 2000; Bals et al, 1999). They act not only to block or
destroy pathogens, but also to activate and coordinate other aspects of the immune
response (Unanue and Allen, 1987; Zhang et al, 2000). This later activity is mediated
through the secretion of inflammatory cytokines such as interleukin-1 (IL-1), IL-8, IL-10,
and tumor necrosis factor or (TNFa) (Zhang et al, 2000; Bals et al, 1999; Hart and
Winstanley, 2002). Secretion of these compounds is associated with an influx of
macrophages and neutrophils into the infected area (Zhang et al, 2000). Eventually
adaptive immunity becomes activated and patients produce antibodies to P. aeruginosa
(Unanue and Allen, 1987). Despite this response, patients fail to clear bacteria, but the

reason why remains unclear.

10

Pathogen recognition in innate immunity is achieved through recognition of
pathogen associated molecular patterns (PAMPs) by toll-like receptors (TLRs) of which
10 have been described in mammals (Akira, 2003). PAMPs include molecules such as
bacterial flagella, peptidoglycan from gram-positive bacteria, lipopolysaccharide (LPS)
from gram negative bacteria, and capsid proteins and double stranded RNA (Delgado et
al, 2006). The prevalence of these molecules in many pathogens makes them logical
molecules for early detection of microbes since it does not require the activation of
antigen presenting cells (APC), antigen processing, or antibody production; although,
TLRs are involved in APC activation (VanKooyk and Geijtenbeek, 2006).

In humans, TLRs are found in a variety of immune and nonimmune cell types
including macrophages, dendritic cells, neutrophils, and epithelial cells (Hippenstiel et al,
2006). Once activated, TLRs activate a signal transduction cascade that leads to
activation of NFKB, gene transcription and the secretion of inflammatory mediators and
antimicrobial compounds (Hippenstiel et al, 2006). Although evidence is limited, there is
evidence that TLRs help mediate phagocytosis and phagosome maturation in
macrophages and it is required for clearance of P. aeruginosa from mouse lungs (Blander
and Medzhitov, 2004; Chignard et al, 2008). There is also evidence to suggest that TLRs

mediate apoptosis of epithelial cells that have phagocytised microbes (Ruckdeschel et al,

11

2004). Apoptosis and desquamation after phagocytosis of microbes by epithelial cells
has been proposed as a method to kill, or capture and remove potential pathogens

(Cannon et al, 2003).

THEORIES ON IMMUNODEFICIENCY
ngLSflt / low fluid

Quinton was the first to suggest an ion transport defect in the lung (Quinton,
1983). It was thought that the absence of CF TR lead to impaired secretion of ions and
thus impaired secretion of fluid of the ASF. This would explain the high viscosity of
mucus in the CF lung since a failure to secrete fluid would lead to mucus dehydration
(Boucher, 2007). In 1996, Welsh and colleagues proposed that the primary cause of CF
lung disease was deactivation of endogenous antimicrobial compounds in the airway
mucus caused by elevated sodium chloride concentrations, thus the primary defect would
be failure to reabsorb chloride due to a lack of CFTR-mediated chloride transport similar
to the situation in the sweat gland (Smith et al, 1996). However, in viva measurements of
electrolyte concentrations in the airway mucus and ASF are difficult due to their small
volume and the patient’s reflexes, and this remains one of the weak points in this theory

(Hanrahan, 2000). One might also predict that a generalized immunodeficiency would

12

result in colonization with a broad range of pathogens, rather than the narrow range that
is found in CF. An alternate theory was proposed shortly thereafter by Boucher and
colleagues who contended that CF TR was responsible for regulating conductance through
other ion channels (Matsui et al, 1998). They believed that the ability of CFTR to
regulate the activity of the epithelial sodium chloride channel (ENaC) was critical for the
maintenance of ASF volume and that the loss of this ability was responsible for increased
sodium reabsorption and dehydration of the mucus causing subsequent loss of
mucocilliary clearance (Figure 3). In either model, the underlying defect allows for
colonization by microbes that damage tissue and trigger inflammation to further impede
mucocilliary clearance and thus exacerbate the conditions that initially allowed for

colonization (Wine, 1999).

Intracellular acidification

The hypothesis that CFTR regulates organellar pH and could thus effect a wide
range of metabolic functions was an attempt to explain the pleiotropic presentation of CF
patients and how a mutation in a chloride channel could result in seemingly unrelated
pathologies. Some evidence does support this hypothesis, but most evidence in confined

to cultured cells and the results of some very similar studies are contradictory (Dunn et

13

a1, 1994; VanDyke et al, 1992). Several groups suggested that CFTR regulates the pH of
Golgi stacks, and that defective regulation in this compartment could explain at least one
of the defects seen in CF: increased adherance of P. aeruginosa (Barasch and al-Awqati,
1993; Chandy et a1 2001). The hypothesis is that if Golgi pH is altered due to a defective
CFTR, then glycoprotein biosynthesis in the Golgi may also be effected since the
enzymes responsible are pH sensitive. This theory does agree with the finding that the
glycolipid asialo-GM] is enriched on CF epithelial cells (Soong et al, 2004).
Furthermore, this molecule is a receptor for binding of P. aeruginosa, and is thought to
help the bacteria colonize the host airway (Imundo et al, 1995). One group has noted
defective killing of phagocytised P. aeruginosa by macrophages, and has linked this
phenomenon to defective acidification of phagosomes (Di et al, 2006). Thus CF TR may
effect pH regulation in more than one intracellular compartment, contributing to the

pleitropic effects of mutations in CFTR.

Hyperinflammation

Some evidence suggests that the inflammatory mechanisms of the innate immune
response are misregulated in CF. Both adult and infant airways of CF patients show

elevated levels of macrophages and neutrophils and adults have elevated levels of IL-8,

14

IL-6 and TNF-ot (Muhlebach et al, 2004). It has been proposed that the inflammatory
response may be constitutively active, overly vigorous, or fail to attenuate properly. It
has been shown in cell culture that secretion of IL-8 from non-CF epithelial cells in
response to treatment with P. aeruginosa culture filtrates is attenuated with repeated
stimulations, but the response fails to attenuate in CF cells (Wu et al, 2005).

Strategies to increase production of functional CFTR (transfection or incubation
at 25°C) have been shown to reduce NF-KB activation and production of inflammatory
cytokines such as IL—8 and TNF-a in immortalized cell lines and primary cell cultures
(Kube et al, 2005; Vij et a1, 2009; Venkatakrishnan et al, 2000). Data from primary cell
culture is controversial since the results could be artifacts caused by chronic bacterial
exposure. This has led several groups to investigate the inflammatory profile of naive CF
airways. One group reported increased inflammation and infiltration by neutrophils in
xenografis from CF fetuses in severely immunocompromised mice compared to wild-
type xenografts (Tirouvanziam et a1, 2000; Tirouvanziam et al, 2002). Several studies
have sampled the airways of young CF patients to monitor the levels of inflammatory
mediators (Hubeau et al, 2001). The results of these studies are contradictory; however,
it should be noted that measuring the concentration of cytokines from the ASF in vivo is

diffith for the same reasons as measurement of ion concentrations. Sampling

15

techniques are invasive and can easily disrupt the system that is being sampled

(Hanrahan, 2000).

CF TR arcts as; receptor fogphagocytosis

Several lines of evidence suggest that CFTR may play a more direct role in
defending against pulmonary infections. Pier et a1 (1997) showed that a portion of the
first extracellular loop of CFTR binds to lipopolysaccharide (LPS) from Pseudomonas
aeruginosa (Pier et al, 1996; Pier et al, 1997). The group proceeded to demonstrate that
this interaction was necessary for phagocytosis of the bacteria, activation of NF-KB, and
apoptosis of pulmonary epithelia (Kowalski and Pier, 2004; Schroeder et al, 2002). This
model suggests that CFTR may act as a receptor for phagocytosis of invading bacteria by

epithelial cells which constitute the first line of defense against pulmonary pathogens.

BACTERIA

Progression of colonization
Although Stapholococcus aureus and Hemophilus influenza occur early in the life

of CF patients, little research has been done on their role in disease progression (Lyczak

et al, 2002). It has been speculated that these pathogens may breach the defenses of the

16

lungs, weakening them enough to allow for colonization of more rarified pathogens later
in the course of disease (Bals et al, 1999). As treatments improved in the late 60s and
patients lived longer, P. aeruginosa emerged as the most common pathogen in CF
patients (Lyczak et al, 2002). The emergence of this pathogen correlates with the
founding of CF centers; however, it also correlates with an increase in life expectancy
due to the use of effective antimicrobial chemotherapy (Figure 2) (Elborn et al, 1991).
Patient to patient transmission of P. aeruginosa and other bacteria such as B. cepacia has
been documented in CF centers; however, it is not clear if the establishment of these
centers is the primary cause for the emergence of these pathogens as common infectious
agents in the CF population, or their transmissibility is just a cautionary note for

clinicians (Lyczak et al, 2002).

Pseudomonas aeruginosa

Approximately 80-90% of patients with CF become colonized by P. aeruginosa.
P. aeruginosa is a gram negative flagellated rod shaped bacteria responsible for many
nosocomial infections in immuno-compromised individuals (Sherertz and Sarubbi, 1983).
A widely distributed microbe, P. aeruginosa can inhabit many substrates including the

soil, rhizosphere, and the surfaces of medical devices and implants (Hart and Winstanley,

17

2002). It is metabolicly diverse and even has applications in bioremediation for toxic
waste cleanup (Ganuli and Tripathy, 2002). Unfortunately this metabolic diversity has
allowed the evolution of a natural resistance to several antibiotics (Carmeli et al, 1999).
Like all gram negative bacteria, the outer membrane of P. aeruginosa contains the
molecule lipopolysaccharide (LPS), a powerful activator of the immune and
inflammatory response (Kronborg, 1995). The role of this molecule in CF has previously
been studied for its ability to induce tissue inflammation, stimulate secretion of
inflammatory cytokines, and its proposed role as an antigen for the innate immune system
(Pier, 2007).

During the course of infection, P. aeruginosa can undergo several changes that
increase its ability to persist in the host. The progression goes from strains with a smooth
to rough then to a mucoid colony phenotype. These changes in colony morphology are
caused by changes in the structure of their LPS and alginate production (see the section
on LPS structure for more details on structural changes in the molecule). The conversion
from smooth to rough colony morphology correlates with the loss of the 0 side chain
which increases the pro-inflammatory activity of the molecule (Pier, 2007). This
conversion has been shown to increase adherence to corneal epithelial cells (Fletcher er

al, 1993).

18

Mucoid colonies produce alginate, a secreted polysaccharide that protects the
microbes from killing by the host immune system by reducing the effectiveness of
secreted antimicrobial compounds (Hart and Winstanley, 2002; Pier et al, 2001).
Alginate production is commonly correlated with the formation of stable biofilms in
which individuals cooperatively evade host defenses (Hart and Winstanley, 2002).
Biofilm formation not only increases the resistance of P. aeruginosa to the host immune
system, but also increases resistance to antibiotic chemotherapy (Hart and Winstanley,
2002). Formation of biofilms is mediated through quorum sensing, a type of bacterial
intercellular communication, which allows bacteria to coordinate their defense against

and attacks on the host (Smith and Iglewski, 2003).

Burkholderia cepacia

 

Burkholderia was first described in 1950 as the cause of soft rot in onions and was
named Pseudomonas cepacia (Burkholder, 1950). These bacteria do bear many
similarities to Pseudomonas sp in that they are gram negative rod shaped bacteria that are
metabolically diverse, found in the soil and rhizoshpere, and are motile by means of
flagella (Hart and Winstanley, 2002). However, genetic analysis has led to them being

classified as a new genus, Burkholderia (Yabuuchi er al, 1992). B. cepacia was first

19

recognized as a common pathogen of CF patients in 1984 (Isles et al, 1984). Patients can
present with symptoms ranging from subclincial infections to a rapid and fatal decline
due to sepsis referred to as cepacia syndrome (Govan et al, 1996). Unlike P. aeruginosa,
B. cepacia has a genome with 2-4 circular replicons with multiple insertion elements
rather than the classical single circular chromosome (Rodley et al, 1995; Tyler er al,
1996). This confers a high degree of genetic plasticity to the bacteria, another feature
that is common in P. aeruginosa, and which seems to be enhanced in isolates from
chronically infected CF patients (Hart and Winstanley, 2002). This plasticity is thought
to confer a survival advantage by allowing infectious strains to rapidly evolve to the
harsh conditions found in the CF airway.

Infection by B. cepacia begins with pilus mediated attachment of bacteria either
to airway epithelia or mucins (Sajjan et al, 2000; Sajjan et al, 1992). B. cepacia is
naturally resistant to the host’s antimicrobial peptides as well as a number of antibiotics,
making clearance of the colonizing bacteria difficult (Baird er al, 1999; McClean and
Callaghan, 2009). LPS from both B. cepacia and P. aeruginosa has been shown to
induce inflammation in cultured cells and animal models, and serves other roles in the
pathogenesis of these bacteria (Mahenthiralingam et a1, 2005). Several genetic

knockouts have shown that the core oligosaccharide in B. cepacia LPS is required for its

20

resistance to antimicrobial peptides (Mahenthiralingam et al, 2005; Cox and Wilkinson,
1991; Loutet et al, 2006). Further evasion of host defenses is achieved through quorum
sensing and biofilm formation, much as is the case for P. aeruginosa (Tomlin et al,
2005). Vigorous antimicrobial chemotherapy is able to reduce colonization to the point
that sputtun cultures are negative for bacteria; however, cases of the same strain of B.
cepacia returning to a patient after a negative sputum culture are not uncommon (Denis et
a1, 2007). It could be that the second infection occurs through reinnoculation by bacteria
in the environment or on medical devices where B. cepacia has been shown to form
antiseptic resistant biofilms; however, it has been postulated that the bacteria evades
eradication and detection by invading host cells (Hart and Winstanley, 2002; Sajjan et al,
2006). B. cepacia has been shown to penetrate and survive within cultured macrophages
and epithelial cells (Sajjan et al 2006; Saini et al, 1999; Burns et al, 1996; Pirone et al,
2008). This localization would protect bacteria during antibiotic administration and
allow the strain to reemerge after treatment has stopped; however, this phenomenon is

difficult to document in vivo.

21

Lip_omlysaccharide

The outer leaflet of the outer membrane of gram-negative bacteria is composed
completely of lipopolysaccharides (LPS) (Doerrler, 2006). These molecules have four
structural components; the inner and outer core oligosaccharides, the O-antigen, and the
lipid A (Figure 4) (Pier, 2007). The lipid A consists of 4-7 fatty acid residues linked to a
diglucosamine. This portion of LPS varies in the number and species of fatty acids, and
modifications to the diglucosamine such as addition of other sugar residues. Such
modifications have been shown to alter not only bacterial behavior, but also the host
response to the LPS (Pier, 2007; Backhed et al, 2003). The length and diversity of sugar
residues of the inner and outer core oligosaccharides varies between strains (Pier, 2007;
Backhed et al, 2003). The O-antigen is a highly variable sequence of sugars that can
confer resistance to antimicrobial compounds, but is also important for immune
recognition and is the basis of serotype classification of P. aeruginosa and B. cepacia
(Mahenthiralingam et al, 2005; Backhed er al, 2003; Pier and Ames, 1984). The
conversion from rough to smooth colony morphology correlates with an increased
production of LPS that lacks the O-antigen (Pier, 2007). Patients chronically colonized

by P. aeruginosa have high serum levels of IgG that is reactive to the O-antigen of LPS,

22

thus conversion to a smooth colony phenotype may protect P. aeruginosa from acquired
immunity (Pier, 2000).

Immune activation by LPS can also be achieved through binding of the molecule
to toll-like receptor 4 (TLR4) (Greene et a1, 2005). The lipid A structure of LPS
interacts with TLR4 and the potency of immune activation varies depending on its
structure. It has been found that P. aeruginosa converts from the production of primarily
penta-acylated lipid A to hexa-acylated lipid A during colonization of the lung (Pier,
2007). The hexa-acylated form of the lipid A is a more potent activator of inflammation
and the immune response than the penta—acylated form and confers a higher degree of
resistance to antimicrobial peptides (Pier, 2007). Penta-acylated MKLPS from the
mutant Escherichia coli strain has even been shown to competitively antagonize
activation of TLR4 signaling thus reducing activation of the inflammatory signal
transduction cascade (Somerville et al, 1996). Certainly increased resistance to
antimicrobial peptides confers an advantage, but the role of conversion to hexaacylated

LPS and subsequent inflammation in P. aeruginosa colonization is still unclear.

23

THIS STUDY
Msiml objectives

My work has been focused on host-pathogen interactions in the CF lung
infections. Based on evidence from Gerald Pier’s group at Harvard and the ability of
CFTR to interact with many diverse proteins, its ability to regulate protein and vesicular
trafficking, and the well known role for TLR4 in the response to LPS, I set out to test the
hypothesis that both CFTR and TLR4 are both involved in phagocytosis of gram negative
bacteria by epithelia. A possible model is shown in Figure 5. To test this model, I used
an iodide efflux assay to measure the presence and activity of CFTR at the cell surface in
response to treatment with Pseudomonas aeruginosa and LPS purified from P.
aeruginosa. I predicted that if CFTR is involved in phagocytosis of P. aeruginosa
through an interaction with LPS, then treatment of cells with either LPS or live bacteria
would reduce the abundance of CF TR at the plasma membrane which would cause a
decrease in CFTR activity observable by iodide efflux. Furthermore, if LPS is the
antigen responsible for phagocytosis of P. aeruginosa, then addition of LPS should
reduce the number of phagocytised bacteria. quantified using the gentamicin protection

assay.

24

Gentamicin protection assay
The gentamicin protection assay, also known as an invasion / adhesion assay, is
capable of quantitatively differentiating between bacteria that are free living, adhered to
host cells, or inside of host cells. This is possible because of the inability of the antibiotic
gentamicin to cross eukaryotic cell membranes (Vaudaux and Waldvogel, 1979). This
allows for killing of all extracellular bacteria without killing intracellular bacteria which
represent phagocytised or invaded bacteria. This difference between phagocytised and
invaded is usually subtle since many invasive bacteria are able to invade host cells by
allowing themselves to be phagocytised and subsequently escaping into the cytosol or
inhibiting maturation of the phagosome (High et a1, 1992). Before this experiment can be
run, several preliminary experiments must be run.
1) The minimum inhibitory concentration (MIC) of gentamicin for the
bacteria must be determined.
2) The growth of the bacteria must be characterized to determine when log
phase growth occurs and to correlate optical density to bacterial density to

allow a consistent inoculum to be administered between experiments.

25

3) The sensitivity of the bacteria to reagents used in manipulating cell culture
must be tested.

4) The density of eukaryotic cells grown to confluence must be determined to
allow for the calculation of a consistent multiplicity of infection (MOI =

CF U bacteria/eukaryotic cell) to allow for comparison to other work.

Immortalized human bronchial epithelial cells taken from a CF patient and
transfected to express wild-type CFTR (S9 cells) were grown in 24-well plates and
incubated with bacteria for 3 hours. After this, bacteria from one set of wells was used to
quantify total bacteria, and the others were rinsed to remove any un-adhered bacteria.
Gentamicin was then added to kill any extracellular bacteria. Removal of antibiotic
followed by lysing of the host cells, dilution of samples and plating for CFUs allows for
quantification of internalized bacteria. In one experiment, host cells were lysed prior to
removal of antibiotic to confirm that bacteria remained susceptible to antibiotics.
Cultures were treated with MKLPS, an underacylated form of‘LPS that has been shown
to competitively inhibit TLR4 (Somerville et al, 1994). If TLR4 is involved in binding
and phagocytosis of the bacteria, then treatement with MKLPS should have reduced the

amount that were adhered and internalized; however, this was not observed.

26

To confirm that phagocytosis of bacteria actually occurred, the assay was repeated
with P. aeruginosa PA01 labeled with green fluorescent protein (GFP) and human
bronchial epithelial cells grown on glass coverslips. Observation of cultures after
treatment with gentamicin revealed the presence of adherent, and apparently antibiotic
resistant, microcolonies at the junctions between eukaryotic cells. This indicates that the
quantification of phagocytised bacteria includes extracellular microbes in addition to
intracellular microbes, and thus is not representative of phagocytosis and therefore cannot
be considered conclusive evidence that the rate of phagocytosis is unaffected by
treatment with MKLPS. It is likely that lysing of the host cells disrupts these resistant
colonies and allows them to be exposed to antibiotics again, thus explaining why lysing

of the cells before antibiotics are removed results in killing of bacteria

Iodide efflux

Iodide can be used to study the activity of CFTR by measuring the rate at which it
flows out of cells. Iodide will pass through CFTR, but little will pass through other
chloride channels and exchangers (Long and Walsh. 1997). In this assay, cultured cells
are loaded with iodide or radioactive 125-I by incubating cultures with a high

concentration of sodium iodide or radionuclide for 1 hour. The buffer is then removed

27

and extracellular iodide is rinsed away with iodide-free buffer. Cultures are then
incubated in iodide-free buffer for 1 minute after which the buffer is collected and
replaced. During this time, iodide will flow out of the cells through activated CFTR.
After establishing a stable baseline efflux rate, buffer is added containing forskolin and
IBMX. Forskolin activates adenylate cyclase which produces cyclic adenosine
monophosphate (CAMP) and IBMX inhibits phosphodiesterase, the enzyme that degrades
CAMP (Seamon er al, 1981; Wang et a1 1999). The resulting increase in CAMP
concentrations will activate protein kinase A which has been shown to activate CF TR by
phophorylation of its R-domain (Chappe et al, 2005). If CFTR is present at the plasma
membrane, this will cause the rate at which iodide leaves the cells to increase. Thus the
increase in efflux rate is due to an increase in CFTR Channel activity, and the amount of
Channel activity is dependent upon not only the probability of the Channel opening, but
also it’s abundance at the plasma membrane. Iodide in the samples can be measured
using ion selective electrodes or with a scintillation or gamma-counter, and used to
calculate efflux rates. Data can be analyzed by comparing peak efflux rates after
stimulation with forskolin and IBMX, or by comparing the area under the curve (AUC)

after treatment with the CAMP elevating agents forskolin and IBMX.

28

In this study, radioactive 125-I was found to be most effective for measuring
CFTR activity. It was found that the iodide selective electrode was not sensitive enough
to reliably detect the small amounts of iodide secreted from cell cultures during baseline
efflux. Cultures were treated with variable concentrations of LPS or bacteria at different
time points to observe the effect of the treatments on CF TR Channel activity. Acute
administration of LPS (at the same time as stimulation with forskolin and IBMX) was
found to reduce CF TR Channel activity compared to forskolin and IBMX alone, while
sustained treatment for longer time periods increased activity in a dose-dependent
manner. Similarly, sustained treatment with P. aeruginosa PA14 resulted in a modest
increase in CFTR Channel activity. This is the first report of a change in CFTR Channel

activity in response to treatment with LPS.

29

METHODS

Materials and reagents

Fetal bovine serum (FBS), bovine serum albumin (BSA), type IV human
placental collagen, geneticin (G418), saponin, ethylenediaminetetraacetic acid (EDTA),
sodium iodide, 4-(2-hydroxyethyl)—l-piperazineethanesulfonic acid (HEPES), glucose,
potassium Chloride, sodium bicarbonate, forskolin, isobutylmethylxanthine (IBMX),
tryptic soy, glycerol and lipopolysaccharides from P. aeruginosa serotype 10 were
purchased from Sigma. Dulbecco’s modified eagle medium (DMEM), gentamicin, PBS
without calcium or magnesium Chloride, PBS containing calcium and magnesium, IOOX
penicillin/streptomycin, streptomycin and tryspin/EDTA were purchased from GIBCO.
Potassium nitrate, calcium nitrate, sodium nitrate, sodium hydroxide, sodium chloride,
magnesium sulfate, sodium phosphate (monobasic) and calcium chloride were Purchased

from JT Baker.

30

Cell culture

Murine mammary epithelial cells (C127) stably expressing wtCFTR under control
of a bovine papiloma based vector, 2WT2 cells (Genzyme), and S9 cells, human
bronchial epithelial cells from a patient with a AF508 and W1282X mutations
complemented with wt human CFTR CDNA (ATCC), were routinely cultured in
Dulbeco’s Modified Eagle Medium supplemented with 10.0% fetal bovine serum,
100ug/mL penicillin, IOOug/mL streptomycin and in T25 culture flasks (Corning) coated
with 0.10mg/mL collagen and 0.01mg/mL bovine serum albumin in a water saturated 5%
carbon dioxide (Airgas) / 95% air atmosphere at 370°C. Culture media for 2WT2 cells
also contained lOOug/mL geneticin (G418). Cultures were passed by rinsing with 1.0mL
of 0.05% trypsin EDTA followed by 15 minutes incubation in 0.50mL of 0.25% trypsin
EDTA. Growth media was added at the end of incubation and cells were seeded into
35mm dishes (Corning), glass cover slips (Fisher) or 24-well plates (Corning) coated
with 0.10mg/mL collagen and 0.01mg/mL bovine serum albumin. For gentamicin
protection assays, cells were grown in the absence of antibiotics. Cells were harvested

for hemacytometer counts using the passaging protocol.

31

Bacterial Growth

Pseudomonas aeruginosa strains PA14 and PA01 were generously provided by
Dr. Todd Ciche. For all experiments, 5.0mL overnight cultures of P. aeruginosa PA01
were inoculated in tryptic soy broth (TSB) or P. aeruginosa PA14 were inoculated in
lysogeny broth (LB) (Acumedia) from a frozen stock and grown aerobically at 370°C
with shaking at l35rpm in an incubator shaker (New Brunswick Scientific).

For growth curves, 10.0mL of respective media were inoculated with IOOpL of
overnight culture and grown as before. 800pL was removed at each time-point and used
for serial dilutions and plating and to measure the OD600 using a Beckman DU52O
spectrophotometer. Dilutions were spread on plates of LB with 1.5% agar (Difco) and
grown overnight at 37°C. Plates were counted and used to calculate the bacterial density
(CFU/mL) which was plotted against time and OD600 to determine when cultures
entered log-phase growth and the concentration of bacteria at a given optical density.

For treatment of eukaryotic cells, a 10.0mL culture was inoculated with IOOuL of
overnight culture and grown for 6 hours. The OD600 was measured and used to calculate
the volume of culture needed to obtain a multiplicity of infection (MOI) of 50 during
each time-point of the efflux experiment, or for each well of a 24-well plate. For efflux

experiments, the required volume was removed from the culture and cells were pelleted

32

by centrifugation at 3500rpm for 3.0 minutes in an Eppendorf 5415 centrifuge. The
supernatant was discarded and cells were rinsed with efflux buffer and re-pelleted as
before. Cells were re-suspended in efflux buffer to a density of 10"8CFU/mL. Bacteria
culture was added directly to culture media for gentamicin protection assays.

The minimum inhibitory concentration (MIC) of gentamicin was determined by
growing P. aeruginosa PA01 as described for the gentamicin protection assay. Bacteria
were diluted in DMEM to the same density that would be used for the gentamicin
protection assay and SOOuL of suspension was added to 24-well plates with serially
increasing concentrations of gentamicin added to groups of four wells. The plates were
then incubated for 1.0 hour under the same conditions used in the gentamicin protection
assay (see below). After incubation, three LB agar plates were spread with 200p.L from
each well and incubated overnight at 37°C . The MIC was defined as the concentration of
gentamicin that resulted in no visible growth in all four wells of a given concentration in
three replicates of the experiment. The concentration of gentamicin used in the
gentamicin protection assay was 300ug/mL, SOug/mL higher than the MIC.

The sensitivity of P. aeruginosa PA01 to 1% saponin and 21.2mM EDTA
dissolved in calcium free PBS was determined by incubating bacteria suspended in

DMEM in the working concentration of saponin solution for 15 minutes in a 24-well

33

plate similar to the treatment in the MIC experiment. Samples were serially diluted,
plated, and incubated overnight at 37°C after which colonies were counted and compared
to untreated controls.

GFP was stably expressed in P. aeruginosa PA01 using a triparental mating
system to induce recombination of the gene into the Chromosome. Donor cells,
Escherichia coli bearing either pURR25 or pUX-BF13, were generously provided by Dr.
Todd Ciche and were grown in LB containing 300pg/mL diaminopirnelic acid (DAP) and
100ug/mL arnpicillin (Roche), and recipient cells, P. aeruginosa PA01, were grown in
LB as described previously. Overnight cultures were used to inoculated lOmL of
respective media with lOOuL of donor and recipient bacteria. E. coli strains were grown
for 10.5 hours, and P. aeruginosa strains were grown for 3.5 hours, at which time all
cultures had an OD600 greater than 1. Cells were pelleted by 2 minutes of centrifugation
at 2400rpm, and washed twice with their respective media. Pellets were resuspended in
the remaining fluid, mixed, the suspension was spotted on a plate containing LB and
300ug/mL DAP, and the plate was incubated overnight at 37°C. Cells were washed off
of the plate with 1.5mL LB and washed twice by pelleting at 2400rpm for 2min and

resuspending in 1.5mL LB. Bacteria were plated on LB agar containing 30ug/mL

34

kanarnicin (Roche) and 60ug/mL streptomycin, and incubated overnight at 37°C.

Transformants were streaked onto LB agar and used to make freezer stocks.

Gentamicin protection assay

S9 cells were grown to 100% confluence in 24—well plates in culture media
lacking antibiotics. P. aeruginosa PA01 were suspended in DMEM to the cell density
required to achieve an MOI of 100 and split into two tubes. MKLPS (Invivogen) was
added to one tube to a concentration of 50pg/mL. A SOOpL volume of suspension was
added to each well and cultures were incubated for 3 hours in a water saturated 5%
carbon dioxide / 95% air atmosphere at 37°C. After incubation, plates were removed and
samples were collected to quantify total bacteria. The remaining wells were rinsed 5X in
room temperature PBS (with calcium Chloride). A microscope was used to confirm that
cells were still adhered after rinsing. Half of the remaining cells were lysed in a solution
of 1% saponin and 21.2mM EDTA dissolved in calcium free PBS (saponin solution) for
15min and collected, and the other half were incubated for 1 hour with 300pg/mL
gentamicin in a water saturated 5% carbon dioxide/95% air atmosphere at 37°C. Cells
were lysed as before and samples collected. Samples were serially diluted, plated on LB

agar, and incubated overnight at 37°C. Colony counts were used to calculate the density

35

 

of total bacteria, adhered bacteria, and protected bacteria in each treatment. One
experiment was performed in which cells were lysed before the removal of antibiotic.

For microscopy, S9 cells were grown on glass coverslips under normal conditions
for the assay. GFP labeled P. aeruginosa PA01 was prepared as before to achieve an
MOI of l and added to cultured cells. After rinsing or treatment with gentamicin,
cultures were not lysed, but were viewed using fluorescent and differential interference
contrast using a Leica DM5000 compound microscope (Leica Microsystems, Wetzlar,
Germany) equipped with an X-Cite 120 fluorescence illuminator (EXPO, Quebec,
Canada), a Spot Pursuit Charge-coupled device camera (Diagnostic Instruments, Sterling

Heights, MI), and a GFP filter set (Leica) as described in Ciche et a1 (2008).

Iodide efflux

The ORION iodide selective electrode was calibrated with fresh sodium iodide
standard solutions of lOmM, 1.0mM, 0.10mM, 0.01mM, and 0.001mM made in nitrate
based efflux buffer as per the manufacturer’s instructions. Nitrate based efflux buffer
contains 3mM potassium nitrate, 2mM calcium nitrate, 20mM HEPES, lmM glucose and
either 136mM sodium iodide or 136mM sodium nitrate. To verify electrode function,

serial dilutions were made from the same stock solution used to standardize the electrode

36

 

initially. These solutions were read with the meter and the ratio between the measured
concentration and the actual concentration was calculated. Since the meter would read a
trace concentration of iodide in distilled water rather than reporting a null concentration,
the limit of detection had to be defined as the concentration below which the ratio
between the measured concentration and the actual concentration was greater than 1.2,
after which the deflection away from the actual concentration was sharp. Similar
experiments were performed with the electrode being calibrated and measurements being
made in standards prepared in distilled water and Chloride based efflux buffer.
Experiments were performed in tandem with three cultures at a time. Each day an
experiment was run, a replicate set of three cultures of control cells treated with only
forskolin and IBMX was ran to standardize the results for comparison to treatments. To
begin, growth media was removed and cultures were rinsed with efflux buffer containing
lSOmM sodium Chloride, 5.4mM potassium Chloride, 0.80mM magnesium sulfate, lOmM
HEPES, 1.0mM sodium phosphate (monobasic), 1.8mM calcium Chloride, and 1.0mM
glucose. Cultures were incubated in 1.0mL of efflux buffer with 2p.Ci of 125-1
(PerkinElmer) for 1-1.5 hours at 37°C. After loading with radionuclide, buffer was
removed to a collection tube and cells were rinsed 4X with 1mL of 4°C efflux buffer. A

1mL volume of room temperature efflux buffer was added to each culture for one minute,

37

after which the buffer was removed to a collection tube and replaced with 1mL of room
temperature efflux bufi‘er. This process was repeated for 15 minutes. Beginning at
11min, buffer was replaced with efflux buffer containing IOpM forskolin and IOOpM
IBMX. After 15 minutes, 1mL of 2M NaOH was added for 10 minutes to lyse the cells.
While cells were lysing, the next set of 3 dishes were rinsed and loaded with radionuclide
as before. Effluxed 125-I in samples was counted with a Cobra II auto-gamma counter
(Packard). In experiments where cells were treated with lipopolysaccharide (Sigma) or
bacteria, the treatment was added to the efflux buffer at the concentration indicated for
the entire period of sample replacement beginning after the 4 rinses with 4°C efflux
buffer. A schematic of the time-course is shown below (figure 6).

After running control experiments, samples were only collected between 5min
and 15min of the efflux experiment and peak efflux was measured as the fold change
over baseline of each culture at 12min. All measurements of the area under the curve
were taken from the start of the stimulus until time 13, just before values returned to
basal levels. The efflux rate constant was calculated as described by Becq er a1 (2003).
Briefly, the total radioactivity in the timecourse was summed (from the first sample
replacement to that collected in the lysate) and used to calculate the percentage of

intracellular iodide left at each time point. To calculate the rate constant for each time

38

point, the equation r = [ln(R1)— ln(R2)] / (t1-t2) was used where r is the rate constant, R1

and R2 are the percentages of intracellular iodide left at time point t1 and t2 respectively.

Data analysis

Data analysis was performed using StatView (Abacus Inc). Single comparisons
were made using an unpaired 2-tailed student’s t-test, and multiple comparisons were
made using ANOVA using the Bonferoni correction for post-hoe analysis. A p-value of

s 0.1 was considered significant.

39

RESULTS

TLR4 antagonist LPS does not alter recovery of P. aeruginosa in the gentamicin
protection assay.

To determine if TLR4 is involved in phagocytosis of P. aeruginosa, a gentamicin
protection assay was used to quantify bacteria that were internalized by human bronchial
epithelial cells that express wild-type CFTR (S9 cells). Bacterial growth curves were
performed to ensure reproducible production of bacterial inocula, and the minimum
inhibitory concentration of gentamicin was determined to establish a concentration
required for killing of extracellular bacteria. Lastly, phagocytosis was confirmed with

fluorescent microscopy.

Cell counts, bacterial growth curves, gentamicin MIC, and treatment sensitivity.

Growth curve analysis was performed for P. aeruginosa strains PA01 and PA14
to allow for treatment of cultures with a consistent amount of bacteria in log phase
growth. Growth curves for P. aeruginosa PA01 and PA14 are shown in Figures 7 and 8,
and the correlation between bacterial cell density and optical density (OD600) is shown

in figures 9 and 10. Both strains reached log phase growth by 6 hours and had a linear

40

absorbance between 0 and 1 OD600. An optical density of 1 correlated to approximately
10"9 colony forming units (CFU)/mL of bacteria. To calculate how many bacteria were
required for a specific multiplicity of infection (MOI), S9 cells grown to confluency were
counted using a hemacytometer. Confluent cultures of S9 cells in 24-well plates were
found to have approximately 250,000 cells/well. Confluent cultures of 2WT2 epithelial
cells grown in 35mm dishes were found to have approximately 950,000 cells/dish.

To determine what concentration of gentamicin was required to kill extracellular
bacteria, the minimum inhibitory concentration (MIC) of gentamicin was determined by
adding increasing concentrations of gentamicin to cultures of P. aeruginosa PA01. The
MIC was found to be 250ug/mL, so 300pg/mL gentamicin was used in all experiments.
Similar experiments were performed to confirm that other agents used to treat cultures in
the course of the gentamicin protection assay (0.25% trypsin EDTA, 1% Triton X-100,
phosphate buffered saline, 1% saponin with 21.2mM EDTA, and DMEM) did not have
an effect on P. aeruginosa PA01 viability. Cultures of P. aeruginosa PA01 treated with
trypsin EDTA had fewer bacteria than untreated controls; however, they were not

effected by treatment with saponin and EDTA.

41

T LR4 antagonist LPS does not alter invasion or adhesion of P. aeruginosa PA 01.
Quantification of media used to inoculate cells showed that S9 cultures were
treated with 6*10"6CFU/mL P. aeruginosa PA01, or an MOI of about 13. During the
incubation, bacteria roughly tripled to a total of about 2*10A7CFU/mL or an MOI of
about 40. Bacteria adhered at a rate of 1% of the total inoculum (at the end of
incubation) and about 0.03% of the total inoculum were protected from gentamicin. This
means that on average, approximately 40% of S9 cells had a bacteria adhered to their
surface, but only 1% had engulfed bacteria. No significant difference was seen in
protection or adhesion between untreated controls and cultures treated with the TLR4
antagonist MKLPS (Figure 11). Lysis of cells before the removal of antibiotic resulted in

no recovery of bacteria.

P. aeruginosa PA01 forms antibiotic resistant colonies on epithelial cells.

The continued susceptibility of bacteria to 300ug/mL gentamicin after incubation
with S9 cells was confirmed by lysing epithelial cells in the presence of the antibiotic.
No bacteria were recovered from this treatment. P. aeruginosa PA01 was labeled with
GFP and used to inoculate S9 cells grown on glass coverslips at an MOI of one bacteria

per epithelial cell. Similar examination after treatment with 300p.g/mL gentamicin

42

revealed the formation of antibiotic resistant microcolonies at eukaryotic cell junctions

(Figure 12).

Alterations in CFTR mediated efflux in response to treatment with bacteria.

An iodide efflux assay was used to monitor Changes in the abundance and activity
of CFTR at the plasma membrane in response to treatment with P. aeruginosa or purified
LPS. Measurement of iodide in samples was used to calculate the rate at which the ion
was secreted from cultured cells. This data was then used to calculate the efflux rate
constant at each time point. By measuring the efflux as a rate constant (minA-l) instead
of a rate (CPM/min), the analysis compensates for variability in the amount of iodide
loaded between trials. If CFTR is involved in phagocytosis of the bacteria, then these
treatments should result in depletion of CFTR at the plasma membrane and a
corresponding decrease in the rate of iodide efflux. Two methods were employed for this
experiment, one using an ion selective electrode to detect iodide released during the

course of the experiment, and the other using radioactive 125-Iodide as a tracer.

43

 

The iodide selective electrode was not sensitive enough for detection of iodide efflux.

The ORION iodide selective electrode was not sensitive enough for use in ion
efflux experiments. The limit of detection was determined experimentally using either a
buffer containing Chloride, nitrate, or with distilled water as described in the methods.
Although the manufacturer states that the electrode can detect 50nM iodide, the meter
was only able to reliably detect iodide down to 60uM in Chloride buffer, and at 10-fold
lower concentrations in either nitrate buffer or distilled water (Figure 13). Despite the
use of nitrate buffer in efflux experiments, the baseline efflux readings for S9 cells were
at or below the limit of detection of the meter when using the sample replacement
technique. Stimulation with forskolin had no detectable effect on efflux rates in these
cells (Figure 14). When the experiment was performed with 2WT2 cells, baseline efflux
rates were similarly low, but forskolin was able to elicit a detectable increase in efflux
rate (Figure 15).

Direct continuous measurement of iodide in the extracellular buffer without
sample replacement was attempted by placing the meter directly in the buffer in the cell
culture dish. This technique was problematic for several reasons. First, the solution
needed to be agitated to facilitate diffusion of iodide throughout the buffer. Second, the

meter can take anywhere from 20 seconds to 5 minutes to reach a stable reading, far too

44

slow for direct measurement of efflux from cultured cells. Third, the solution came to
equilibrium between intracellular and extracellular iodide concentrations too quickly to
allow for iodide to continue to efflux for the amount of time required to allow the meter
to generate a stable reading or baseline efflux rate. Despite the ability to read higher
concentrations of extracellular iodide during the efflux experiment, no detectable

response was observed in response to forskolin (Figure 16).

Radionuclide efllux controls

In radionuclide efflux experiments, the initial efflux rate constant declined
steadily from approximately 0.12minA-1 until stabilizing 5 minutes into the timecourse at
about 0.05minA-1 after which the rate constant declined slightly over another 15 minutes
to 0.03minA-1 (at 20 minutes into the timecourse). Treatment of cells with lOuM
forskolin and IOOuM IBMX at 10 minutes into the time-course stimulated an increase in
the efflux rate beginning at 11 minutes, peaking 3.3 fold higher than baseline (a rate
constant of 0.2minA-1) at 12 minutes, and decreasing steadily back to baseline by 14
minutes (4 minutes after addition of 10pM forskolin and 100uM IBMX). Acute
treatment of cultures with 100ug/mL LPS alone at 10 minutes (the same time interval as

previous treatment with IOpM forskolin and IOOpM IBMX) had no effect on efflux rate

45

 

compared to untreated cultures. Figure 17 shows the calculated rate constants from
unstimulated, lOpM forskolin and 100uM IBMX stimulated, and acute LPS stimulated
cultures and Figure 18 shows these values normalized to the baseline of approximately
0.05minA-l.

Based on these results, samples were only collected between 5 minutes and 15
minutes of the efflux experiment and peak efflux was measured as the fold Change over
baseline of each culture at 12 minutes. All measurements of the area under the curve
(AUC) were taken from the start of the stimulus until time 13, just before values returned
to basal levels. Treatments were administered at two times: either acute (at 10 minutes
into the efflux experiment until the end at 15 minutes) or continuous (beginning at the

start of the efflux time-course and continuing until the end at 15 minutes).

P. aeruginosa LPS alters CF TR channel activity in a dose-dependent manner.

Cultures treated continuously with P. aeruginosa PA14 at an MOI of 100 showed
a 25% increase in CAMP stimulated peak efflux and a 35% increase in AUC compared to
control cultures (Figures 21 and 22). Cultures were treated with LPS from P. aeruginosa,
to determine if LPS was responsible for the response observed with live bacteria.

Continuous treatment of cultures with increasing concentrations of LPS resulted in an

46

 

increase in CAMP-stimulated peak efflux and AUC in a dose dependent fashion
compared to controls treated only with CAMP elevators. This trend is shown in Figures
19 and 20. Cultures continuously treated with SOug/mL LPS and had a 30% increase in
CAMP stimulated peak efflux (3.25 fold compared to a 2.5 fold increase in control
cultures) and a 60% increase in AUC. Cultures treated with lOOug/mL LPS had a 40%
increase in CAMP stimulated peak efflux (3.5 fold compared to a 2.5 fold increase in
control cultures) and an 85% increase in AUC. Cultures treated with 200ug/mL LPS had
a 50% increase in CAMP stimulated peak efflux (3.75 fold compared to a 2.5 fold

increase in control cultures) and a 110% increase in AUC.

P. aeruginosa LPS alters CF TR channel activity in a time-dependent manner.

To determine if alteration of CF TR channel activity by LPS from P. aeruginosa
was time-dependent, experiments were performed in which cultures were not treated with
LPS until 10 minutes into the time course of sample replacement (instead of continuous
administration over the entire duration of sample replacement). As stated previously,
acute treatment of cultures with 100pg/mL LPS alone (in place of IOpM forskolin and
IOOpM IBMX at 10 minutes) had no response that could be differentiated from untreated

(no forskolin or IBMX) cultures (Figures 17 and 18). Cultures treated acutely with a

47

 

cocktail of lOOpg/mL LPS, IOpM forskolin and IOOuM IBMX at 10 minutes showed a
25% decrease in peak efflux that was not statistically significant, and a 40% reduction in
area under the curve that was significantly different from cultures treated with CAMP

elevating agents alone (Figure 23 and 24).

48

 

DISCUSSION

There is evidence to support the hypothesis that the cystic fibrosis transmembrane
conductance regulator (CF TR) acts as a receptor for lipopolysaccharide (LPS) on the
surface of gram-negative pathogens, and this interaction leads to phagocytosis of bacteria
by epithelial cells (Pier et al, 1997; Schroeder et al, 2002). This interaction is thought to
be important for Clearance of Pseudomonas aeruginosa from the airway, and explains
why cystic fibrosis (CF) patients without CF TR become infected with these bacteria and
can not Clear them from the airway. The idea that CFTR would serve as a receptor for
LPS seems odd since toll-like receptor 4 (TLR4) is a known receptor for LPS found on
pulmonary epithelial cells that has been shown to be involved in phagocytosis of gram
negative bacteria and activation of the immune response (Blander and Medzhitov. 2004).
Since both CF TR and TLR4 seem to be involved in the cell’s response to P. aeruginosa, I
hypothesized that these proteins form a multi-protein complex that assembles to facilitate
phagocytosis of bacteria.

I tested this hypothesis using two experimental models: a gentamicin protection
assay, and an ion efflux assay. The gentamicin protection assay allows for the

quantification of intracellular (phagocytised) bacteria, and the iodide efflux assay allows

49

 

It. .

for a measurement of CFTR Channel activity at the plasma membrane. Based on my
model in which CFTR and TLR4 both participate in LPS-mediated phagocytosis of P.
aeruginosa, I predicted that treatment with MKLPS, a competitive antagonist of TLR4,
would reduce the quantity of bacteria phagocytised during the gentamicin protection
assay. I would also predict that treatment of cells with either purified LPS, or live
bacteria would result in a decrease in CFTR channel activity at the plasma membrane that

could be detected with the ion efflux assay.

P. aeruginosa strain PA01 adheres to respiratory epithelial cells but is not fregprently
phagocytised.

Human bronchial epithelial cells that stably express wt-human CFTR (S9 cells)
were infected with P. aeruginosa and then treated to kill extracellular bacteria. The
remaining bacteria were then presumed to be intracellular (phagocytised). To quantify
extracellular adherence of bacteria to the epithelial cells, several samples were treated to
remove unadhered bacteria by rinsing with PBS, but were not exposed to the antibiotic to
kill extracellular bacteria.

A small amount of P. aeruginosa PA01 was found to adhere to pulmonary

epithelial cells after 3 hours of incubation. This result is consistent with the

50

 

Characterization of the bacteria as an adherent pathogen (Taminiau et al, 2002).
Although the frequency of adherence was small, the inoculum used in these experiments
was large (MOI of 50), and the small frequency of adhesion could be skewed lower due
to depletion of binding sites by the large number of bacteria added. It is also possible that
a longer incubation time could increase the frequency of adherence.

The amount of bacteria protected from gentamicin was much smaller than that of
adhesion and is consistent with previous reports of the intracellular occurrence of this
pathogen (Darling and Evans, 2001; Pier et al, 1997). Several studies have concluded
that P. aeruginosa is an invasive pathogen; however, the in vivo significance of the small
frequencies of invasion seen in these studies has not been determined. Just because a
small fraction of bacteria are found to invade host cells during a gentamicin protection
assay does not mean that invasion of cells is important for colonization and infection by
the bacterium. It is possible that invasion by a small fraction of bacteria allows enough
individuals to evade the host immune system and re-emerge periodically to sustain
infection despite a strong immune response. However, it is also possible that the results
of these experiments are artifacts. It may be that a small amount of intracellular bacteria
are phagocytised by the host cell, but that the endosomes have not yet matured enough to

kill them. The large inoculums used in these experiments (MOI of 100) more Closely

51

 

resemble the situation when infection has already been well established and the host
immune system overcome. The results of these conditions may therefore be more
representative of how pathogens damage a host when infection has been established
rather than how bacteria behave to initially colonize and maintain infection in a host.
Both situations are relevant to the treatment of CF, but the underlying question of how

early colonization is established is more relevant to the underlying defect in the CF lung.

MKLPS doegot effect the rate of adhesion or phagocmsis.

MKLPS is an underacylated form of LPS from an E. coli mutant. This molecule
has been shown to bind to, but not activate TLR4, thus it acts as a competitive inhibitor
of TLR4 signaling. Pre-treatrnent of bronchial epithelia cells (S9) with MKLPS
beginning 15 minutes before the addition of bacteria and continuing over the 3 hour
period of incubation had no effect on the rate of adhesion or phagocytosis of P.
aeruginosa strain PA01. This would seem to indicate that under the conditions tested in
this study, TLR4 is not necessary for binding or phagocytosis of P. aeruginosa to
epithelial cells; however, with the rate of invasion being as low as it was, it is possible
that any difference was lost to noise. This result is not consistent with Pier et a1 (1997)

since his group showed that LPS reduced the rate of phagocytosis of P. aeruginosa;

52

 

however it does agree with other studies that have shown P. aeruginosa to have a low
frequency of invasion (phagocytosis) (Darling and Evans, 2001). It is possible that P.
aeruginosa was able to bind to other receptors at the cell surface at a high enough
frequency to mask the effect of TLR4 blockade (Figure 25). For example, TLR5 is a
receptor for bacterial flagella and asialo-GMI glycolipids have been studied for their
ability to bind pili and promote adherance of P. aeruginosa (Saiman and Prince, 1993;
Ramphal et al, 2008). It is possible that several or all of these receptors are involved in
phagocytosis of P. aeruginosa. This would mirror the work done by Chignard et al
(2008), who initially concluded that TLR5 was not involved in the host response to P.
aeruginosa because of evidence from a TLR5 knockout mouse. However, in a later
paper when the TLR5 knockout mice were crossed with TLR4 knockout mice, it was
shown that both receptors are involved in the host response to P. aeruginosa. This
redundancy that is built into the immune system complicates elucidation of the

mechanisms responsible for host immune response.

53

 

P. aeryginosa PA01 forms antibiotic resistant microcolonies on pulmonary epithelial

 

_C_e_l_l§

S9 cells grown on glass coverslips were inoculated with GFP-labeled P.
aeruginosa strain PA01, treated to kill extracellular bacteria using the same techniques as
the gentamicin protection assay, and viewed under a differential interference contrast
fluorescent microscope. Observation of GFP-labeled P. aeruginosa after gentamicin
treatment of inoculated S9 cells revealed the presence of extracellular bacterial
microcolonies at the junctions between epithelial cells. This pattern of growth of P.
aeruginosa has been observed in lungs from CF transplant patients and in animals
infected in the laboratory (de Bentzmann et al, 1996; Haubler, 2003). However, this
pattern has not been noted in the context of a gentamicin protection assay. The presence
of extracellular bacteria after treatment with gentamicin indicates that these bacteria were
resistant to the treatment and calls into question not only the results of the gentamicin
protection assay performed in this study, but the results of several decades of experiments
in which this assay has been used to quantify invasion / phagocytosis of P. aeruginosa
and B. cepacia (Pier, 1996; Sajjan et al, 2006; Pirone et al, 2008). Since the presence of
extracellular bacteria would increase the apparent frequency of invasion in these

experiments, the resulting uncertainty about where those colonies came from makes that

54

 

data unreliable without performing control experiments in which microscopy is used to
confirm effective killing of extracellular bacteria. I have not read a single paper in which
this control was performed as part of a gentamicin protection assay which may explain
why my results are different from those of Gerald Pier (1997). A common control is to
lyse host cells before removal of antibiotic to confirm that the remaining bacteria are still
susceptible to killing by the antibiotic. My results confirmed that lysis of epithelial cells
before removal of gentamicin resulted in death of all remaining bacteria; however, if
biofilm formation was necessary for antibiotic resistance, then lysing host cells would
disrupt the biofilm rendering the bacteria susceptible to antibiotics again. Thus

microscopy is the only way to definitively confirm antibiotic susceptibility in this assay.

The ORION iodide selective electrode can detect Changes in CAMP stimulated iodide

efflux in cells that overexpress CFTR. but may not be sensitive enoagh to accurately

detect baseline readings in these cells or others expressing physiologically relevant levels

of CFTR.
One method for measuring CFTR Channel activity at the plasma membrane is
with an ion efflux assay. In this experimental system, cultured cells are “loaded” by

incubation in buffer with sodium iodide in place of sodium chloride, or in buffer

55

 

containing small amounts of radioactive 125-I. Repeated replacement of the buffer with
iodide-free buffer allows intracellular iodide to efflux from the cells, which it does
through CFTR, thus the rate of ion efflux is proportional to CFTR Channel activity (Becq
et al, 2003). Iodide can be quantified using ion selective electrodes, or with any means of
detecting gamma radiation, depending on which technique is used.

Newer studies have used ion selective electrodes to measure iodide concentrations
in the iodide efflux assay since this saves the hazard of working with radioactivity.
According to the manufacturer, the ORION iodide selective electrode (ISE) should be
able to detect 50nM iodide; however, in our lab, the ISE was not accurate below 60uM
and was found to be unreliable below concentrations of 6uM, almost lOO—fold higher
than the reported limit listed by the manufacturer. However, rather than reading 0, the
ISE continues to give readings that are greater than the actual concentration of iodide
when measuring samples that are below its limit of detection. This is problematic in
efflux studies since it can create an artificially elevated baseline efflux rate. The baseline
values measured on the ISE are at or just below the experimentally determined limit of
detection for the meter. When the concentration of iodide drops below the limit of
detection of the meter, the ISE will show a higher concentration of iodide than what is

really there, thus elevating and leveling the baseline efflux reading.

56

in..- .

 

Depending on what cells type is used (cells that overexpress CFTR, or cells that
express physiologically relevant levels of CFTR), stimulation of cells with CAMP
elevating agents may be able to raise efflux rates above the lower limit of detection of the
ISE. No detectable Change in CAMP stimulated efflux rates was ever observed in S9
cells using this technique; however, the response from 2WT2 cells which overexpress
CFTR was detectable using the ISE. Still, if future work is to be done with this model of
instrument, trials with iodide concentrations below IOuM should be excluded from

analysis.

SLstained exposure to LPS during the entire period of sample replacement causes a dose-
dependent increase in CFTR Channel activity.

Efflux rates were measured in 2WT2 epithelial cells using the radionuclide efflux
technique. These cells were used in this experiment because they over-express human
CFTR, giving them a robust response to CAMP elevating agents in the iodide efflux
assay. Treatment of cells with LPS or live bacteria was either acute (10 minutes into
sample replacement) or sustained (over the entire duration of sample replacement), but

treatment with the CAMP elevating agents forskolin and IBMX was always acute.

57

 

Sustained treatment of cultures by the addition of a range of LPS concentrations
to the sample replacement buffer resulted in a significant dose-dependent increases in
CAMP stimulated efflux rate. The same result was obtained, although to a lesser extent,
when cultures were treated with live P. aeruginosa PA14. These results are the opposite
of those predicted by my hypothesis. If LPS is the antigen responsible for CFTR
mediated phagocytosis of P. aeruginosa, then treatment of cultures with LPS should have
reduced the abundance of CF TR at the plasma membrane resulting in a concomitant
reduction in CAMP stimulated efflux rate.

The results of these experiments could be caused by increasing the activity of
CFTR at the plasma membrane, or by increasing the abundance of CFTR at the plasma
membrane. An increase in activity could be due to an increased probability of opening or
a decrease in Channel inactivation. The abundance of CF TR at the plasma membrane is
determined by the rate at which the protein is removed and the rate at which it is added.
If this is the reason for the increase in CFTR activity, then all that can be concluded from

the efflux data is that the ratio of the removal rate to the recruitment rate decreased.

58

 

Acute treatment with LPS reduces CF TR Channel activity.

It is possible that treatment with LPS activates a signaling mechanism that
increases the abundance of CF TR at the plasma membrane to replace molecules removed
through phagocytosis. In this case, Changes in CFTR Channel activity in response to LPS
treatment might show time dependence. To test this hypothesis, cells were treated
acutely (10 minutes after the start of sample replacement, rather than over the entire
duration of sample replacement) with a cocktail of LPS and CAMP elevating agents and
compared to cultures treated with CAMP elevating agents alone. Acute treatment of
cultures with a cocktail of LPS, forskolin and IBMX resulted in a significant reduction in
AUC, but not in peak efflux when compared to cultures treated with just forskolin and
IBMX. This indicates that CFTR Channel activity is decreased in response to LPS,
consistent with the hypothesis that CFTR is involved in LPS mediated phagocytosis. The
biphasic nature of this response is also consistent with the presence of a mechanism to
replace CFTR after it is removed as a result of LPS mediated phagocytosis.

The idea that CFTR can be removed as part of a phagocytic event in this short
amount of time is not inconsistent with previous reports. Phagocytosis has been shown to
begin within as little as 45 seconds of stimulation isolated macrophages, and can be

completed in as little as 5 minutes (Matsumura et al, 2006). This conclusion also agrees

59

 

with the work done by Pier et al (1997) who showed not only greater rates of
phagocytosis of P. aeruginosa in cells expressing wild-type CFTR compared to those
expressing the AF508 CFTR, but also showed that phagocytosis could be reduced by the

addition of purified LPS or a peptide identical to the first extracellular loop of CFTR.

60

Figure 1: Proposed structure of the CFTR showing its major functional domains. The N-
terminus is on the left. Membrane spanning domains 1 and 2 are thought to form a pore
that allows the passage of Chloride ions, and membrane-spanning domain 2 is the site of
protein glycosylation at its first extracellular loop. Binding of ATP at nucleotide binding
domains 1 and 2 increases the Channel’s open probability. The location of the AF508
mutation is shown in nucleotide binding domain 1. Channel activation is regulated
through phosphorylation of the central R-domain. The PDZ domain interacting motif
(TRL or threonine, arginine, leucine) is shown at the C-terminus of the protein interacting
with a PDZ domain that links the protein to other cellular components. Taken from
Rowe et al (2001).

61

 

Extracellular

Membrane Spanning Membrane Spanning ,
Domain 2 Apical
plasma

membrane

Domain 1

 

 

 

 

 

Nucleotide
Binding
Domain 2

 

     
    

Nucleotide
Binding
Domain 1

4

Regulatory Domain
(Phosphorylation site) I

\
“—-—-
<-_____,

‘1
Other ion channels [I] Cytoskeleton
/

I

Signal transduction
Proteins

Intracellular

62

 

too,

 

 

 

 

 

 

 

 

 

 

P. aeruginosa
so. ___
fi
00. A
g ‘0 H. Influenza s. aureus
204
B.cepacta “-
O—fl I 1 I 1
M01 zips omo «tot-r 13mm 250034 emu 49

Age (years)

Figure 2: Age correlated incidence of airway infections in cystic fibrosis airways.
Staphalococcus aureus is the most common pathogen early in life, including infections by
methycilin resistant S. aureus, but is soon surpassed in frequency by Pseudomonas
aeruginosa. Haemophilus influenza is more common early in life, but infections by other
organisms like Burkholderia cepacia increase in frequency later in life. Taken from the
Cystic Fibrosis Foundation (2005).

63

 

Figure 3: Depiction of two theories explaining how alterations in the ASL could lead to
Chronic bacterial infections. The high salt hypothesis shown in panel A postulates that
airway epithelium are impermeable to water (H20) and that Chloride (Cl-) must move
through CFTR to be reabsorbed. Since the sodium potassium (K+) ATPase (N ,K-pump)
maintains a constant gradient favoring the reabsorption of sodium, the amount of sodium
that actually is reabsorbed is regulated by the amount of Chloride that can be absorbed.
Lack of functional CFT R causes prevents reabsorption of Chloride thus keeping the
concentration of sodium (Na+) Chloride in the ASL higher than normal, a situation which
has been shown to inactivate antimicrobial compounds. Panel B shows the low volume
hypothesis. In this model, Chloride can transit from the apical to basolateral suface by
CF TR-independent mechanisms like tight junctions (Tl) between cells. However, the
loss of CFTR-mediated reduction in sodium currents through the epithelial sodium
Channel (ENaC) results in hyperabsorption on water, resulting in a reduced ASL volume
and impaired mucociliary transport. Taken from Wine (1999).

64

 

Na+ Cl-Cl-
Na+Na+ CL H20

UL}

    

 

 

 

 

 

XL Na+ CI- H20 CI- 7
ENaC ,

  

 

65

O-antigen
repeat
40 units

Core polysaccharide

Disaccharide
diphosphate

 

Lipid A Fatty acids

 

Figure 4: Conserved structure of LPS. This figure shows the conserved domains
of LPS, including the O-antigen, a region of variable sugar residues; the outer
core polysaccharide, a conserved link between the lipid A and the O-antigen; the
core disaccharide to which fatty acids are linked; and the fatty acids that together
with the core disaccharide make up the lipid A. Variations have also been
described in the lipid A portion of the molecule in both the number and type of
fatty acids. Taken From Fox (2009).

66

 

\
5 CFTR TRL4

 

E Adaptor Protein(s)
llIIlD ERM P aeru inosa CWOSkeleIal
* g —‘ elements

 

 

449:“
.z

7 5 E 5 “i Pha-oc osis
r. r I x 7—3
lllll- um =. IIIIII=
'
Second '. .
Inactivation
Messengers

 

 

 

Figure 5: Model of a possible interaction between CFTR and TLR4 during phagocytosis

of P. aeruginosa. P. aeruginosa is thought to interact with TLR4 through LPS located in
its outer membrane. This activates signal transduction pathways that activate
inflammation. I propose that the bacteria / receptor complex then forms a complex with
CFTR to recruit cytoskeletal components and to lead to inactivation of the inflammatory
signal. The complex is also able to phagocytise the bacterium.

67

 

 

’ 3 -I- Rinse out media
i -0- Load with 125—l

+ Remove loading
buffer; Rinse

3 0 -)(- Begin sample
0.0 ; replacement

. -6- CAMP elevating
1 agents
A -0- Begin lysis

. -r3- Rinse media from
- . next set of cultures;

 

 

-80 -70 -60 -50 ~40 -30 -20 -10 0 10 20 30
Time (min)

Figure 6: Timecourse of efflux experiments. Efflux experiments were run with sets of
three 35mm dishes of 250,000 cultured 2WT2 cells at a time. First, growth media was
removed and cultures were rinsed with efflux buffer. Next, cultures were incubated in
efflux buffer containing 2uCi of 125-1 for one hour. After incubation, the loading buffer
was removed and cultures were rinsed four times with efflux buffer. Over the duration of
sample replacement beginning at time 0, cultures were incubated in 1mL of efflux buffer
for one minute intervals, after which the buffer was removed and replaced with fresh
buffer. Application of CAMP elevating agents was achieved through sample replacement
with buffer containing the CAMP elevating agents forskolin and IBMX beginning at 10
minutes into sample replacement. After five minutes of treatment with CAMP elevating
agents, cultures were lysed by incubation in 1M sodium hydroxide for fifteen minutes.
During lysis, the efflux experiment was set up in the next set of three 35mm dishes by
removing grth media and repeating the treatment outlined above.

68

‘c ':‘--.i‘ E'EIU! 'l.‘ ‘. II in.‘ V .1 I
E , .. . i1 hi 3‘

 

 

 

 

 

Growth of P. aeruginosa PA01 In Tryptic Soy Broth
.. 7000
° i

6000
3
E 5000
an i I
335‘ 4000
3 E
3 o 3000
3 2000 ~ 1., .
t t
g 1000 . . ......
0 b i
0 2 4 5 8 10
Time (hours)

 

 

Figure 7: Growth curve of P. aeruginosa PA01 grown in tryptic soy broth showing entry
into log phase growth. Samples were taken from lOmL cultures of P. aeruginosa PA01
that had been inoculated with lOOuL from an overnight culture. Samples were diluted
and plated on LB agar and incubated overnight. Colonies were counted and used to
calculate bacterial cell density in relation to time. The growth curve shows entry into
log-phase growth by 6 hours. Data are presented as mean :t SEM (n = 3 to 6).

69

Growth of P. aeruginosa PA14 in LB

9000 '
8000
7000
6000 law...“ .. .. ,. ,.
3000 .
2000
1000 - ;

 

 

 

 

 

Bacterlal Density (mllllone of
CFU/mL)

 

 

 

0 2 4 6 8 10
Time (hours)

 

Figure 8: Growth curve of P. aeruginosa PA14 grown in LB showing entry into log
phase growth. Samples were taken from 10mL cultures of P. aeruginosa PA14 that had
been inoculated with IOOpL from an overnight culture. Samples were diluted and plated
on LB agar and incubated overnight. Colonies were counted and used to calculate
bacterial cell density in relation to time. The grth curve shows entry into log-phase
growth by 6 hours. Data are presented as mean 1 SEM (n = 3 to 6).

70

Optical density of P. aeruginosa PA01 over the duration
of growth In tryptic soy broth

 

 

 

R2 = 0.99

 

Bacterial Density (millions
of CFU/ mL)
8
8

 

 

 

0 0.2 0.4 0.6 0.8 1 1.2
00600

 

 

Figure 9: Optical density correlates to cell density in P. aeruginosa PA01. Samples were
taken from 10mL cultures of P. aeruginosa PA01 that had been inoculated with lOOuL
from an overnight culture. The optical density of samples was measured at 600nm, and
aliquots were diluted and plated on LB agar and incubated overnight. Colonies were
counted and used to calculate bacterial cell density in relation to OD600. The equation of
the best fit line was used to calculate the volume of bacteria needed to inoculate
eukaryotic cell cultures. Data are presented as mean 1 SEM (n = 3 to 6).

71

 

Optical density of P. aeruginosa PA14 over the duration
3 of growth in LB

 

 

 

 

 

 

 

'5

I

.5

7:"

U

E t

I

C

3

E

g
o 0.5 1 1.5 2
00600

i
I
s

Figure 10: Optical density correlates to cell density in P. aeruginosa PA14. Samples
were taken from 10mL cultures of P. aeruginosa PA14 that had been inoculated with
lOOuL from an overnight culture. The optical density of samples was measured at
600nm, and aliquots were diluted and plated on LB agar and incubated overnight.
Colonies were counted and used to calculate bacterial cell density in relation to OD600.
The equation of the best fit line was used to calculate the volume of bacteria needed to
inoculate eukaryotic cell cultures. Data are presented as mean :1: SEM (n = 3 to 6).

72

I; --

TLR4 antagonist LPS (MKLPS) has no effect on adhesion
or protection from gentamicin

 

 

 

 

 

3 100000000
g rooooooo ,,
3 1000000 :
8 100000 E
'6
1’5 10000
B
n 1000 -
g 100 -
>
8 1o
a 1 g

E a a a '0 1, a 8 B A

g i5 .5 2i g g 5 t e E

g M 4: J: x B B x

z 3 2 z E o z

E 3 :2 .. E :

 

 

Figure 11: Treatment with the TLR4 antagonist MKLPS had no effect on recovery of
total bacteria, adhered bacteria, or phagocytised (protected) bacteria. Immortalized
human bronchial epithelial cells taken from a CF patient and transfected with wtCFTR
(S9 cells) were grown to confluence in 24-well plates. Cultures were inoculated with 10
bacteria per epithelial cell and incubated for 3 hours. After incubation, cultures were
treated to isolate total bacteria, adhered bacteria, and phagocytised (protected) bacteria.
No significant difference was detected between the recovery rates of either set of bacteria
in the presence of MKLPS. Data are presented as mean :1: standard deviation (n = 41 to
56).

73

 

 

Figure 12: Merged image of fluorescent and differential interference contrast photograph
of S9 epithelial cells incubated for 3 hours with an MOI of l GFP-labeled P. aeruginosa
PA01 per S9 cell followed by treatment with 300ug/mL gentamicin for 1 hour.
Extracellular bacteria were observed (arrows) adhered to the junctions between epithelial
cells in antibiotic resistant microcolonies. Images in this thesis are presented in color.

74

 

 

 

 

 

Analysis of ISE function: actual concentration of Na!
standard solutions compared to measured
concentrations
-25 ————~~~~~~ .
z 1
3 i
I; 20 - C i
2 i
"' i
g 15 . .
I: :
l” 3
8 10 ~ 0 g
a . E

Q '1

< 0 Y Q a e i

O 1 2 3 4 5
Measured:ActuaI Concentration (measured/actual)

 

 

 

Figure 13: Analysis of ISE function when reading standards made in distilled water. The
ORION iodide selective electrode (ISE) was calibrated with standards prepared in
distilled water as per the manufacturer’s instructions. A dilution series of standard
sodium iodide solutions was prepared and the concentrations measured using the ISE.
The ratios between the actual concentrations of sodium iodide and the measured
concentrations of sodium iodide were calculated and show that the meter loses sensitivity
at concentrations below 5uM sodium iodide. (n = 1)

75

 

 

ukw- - H7? -vvv-.- V-.L__ .._.__.‘_

Efflux rate in 59 cells measured using the ISE

 

 

 

 

El'llux Rate constant (mlnA-l)
O
H
U'I

 

 

 

 

0.1
10uM Forskolin
o , . r
0 2 4 6 8 10 12

Time (min)

 

Figure 14: Efflux rate constant in S9 human bronchial epithelial cells stimulated with
10uM forskolin measured using the ISE. The ORION iodide selective electrode (ISE)
was used to read samples from efflux experiments run with S9 cells using the sample
replacement technique. Forskolin was unable to stimulate a detectable increase in efflux
using this technique. Values are mean :i: standard deviation (n = 4).

76

 

 

 

Efflux rate in 2WT2 cells measured using the ISE

 

 

 

 

 

 

 

o 2 4 6 8 1o 12
Time (30sec)

 

 

Figure 15: Efilux rate constant in 2WT2 mouse mammary epithelial cells stimulated with
10uM forskolin measured using the ISE. The ORION iodide selective electrode (ISE)
was used to read samples from efflux experiments run with 2WT2 cells using the sample
replacement technique. Forskolin stimulated an approximately 4-fold increase in efflux
using this technique. Values are mean 1 standard deviation (n = 4).

77

‘m.-.

 

 

 

 

  
 

 

 

 

 

 

 

 

Continuous measurement of efflux rate using the ISE In
59 cells shows no detectable response to forskolin
0.25 ,
‘r
A 0.2 mm.--“ __,---.._,,-,,,_ ..._.,_‘, .
.7 10uM Forskolin
‘5 0.15
E
3 0.1 4
a
g 0.05 ..
(f 1 2 3 4 5 6 7 8 99
-o.os m.- .—
Tlme (305cc)

 

 

Figure 16: Efflux rate constant in S9 human bronchial epithelial cells stimulated with
10uM forskolin measured using the ISE. The ORION iodide selective electrode (ISE)
was used to directly monitor the concentration of iodide in the extracellular buffer in
cultured S9 cells. Forskolin was unable to stimulate a detectable increase in efflux using
this technique. Values are mean :3: standard deviation (n = 6).

78

 

 

 

 

 

 

 

 

 

 

Efflux rate in 2WT2 cells stimulated with CAMP
elevators, LPS, or untreated
0.25 ~~~«~—~~~-
i?
E forskolin and
u \ IOOuM IBMX
50.15 _ ____ ,,, ,, m _ m I ,7 i p - I unstimulated
= I
3 ‘s
O 1 a. ,. t -—k '100ug/ml LPS
3 ' I I
a I
g 0 05 _ ‘9. Treatment
5 i
o . , . i
0 5 10 15 20
Time (min)

 

 

 

Figure 17: Efflux rate constant in 2WT2 mouse mammary epithelial cells stimulated with
10uM forskolin measured using radionuclide tracer technique. Experiments were done
with cultured 2WT2 cells which were unstimulated, treated with 10uM forskolin and
lOOuM IBMX, or treated with lOOug/mL LPS over the duration indicated. Only cultures
treated with the CAMP elevating agents forskolin annd IBMX showed a change in efflux
rate. Values are mean :i: SEM of the efflux rate constant (n = 3).

79

 

 

Normalized efflux rate constants in 2WT2 cells treated
wlth cAMP elevating agents, LPS, or unstimulated

 

l

I

—O- 10uM

g forskolin and
; 100uM IBMX
i

l

I ~ unstimulated

*‘ -A- -100ug/mi LPS

 

 

Efflux Rate (fold baseline)

Treatment

 

 

 

 

 

20

Time (min)

 

 

 

Figure 18: Normalized efflux rate constant in 2WT2 mouse mammary epithelial cells
stimulated with 10uM forskolin measured using radionuclide tracer technique. Efflux
experiments were done with cultured 2WT2 cells which were unstimulated, treated with
10uM forskolin and lOOuM IBMX, or treated with IOOug/mL LPS over the duration
indicated. Only cultures treated with the cAMP elevating agents forskolin annd IBMX
showed a change in efflux rate. Values are mean :i: SEM of the efflux rate constant
normalized to the baseline efflux rate (n = 3).

80'

 

Lipopolysaccharide causes a dose-dependent increase
in cAMP stimulated peak efflux

 

 

 

’5‘ 4.25 g
5 0.0009 <0-0001
3 ‘1
3 3.75 l
3 i
g 3 25 i
g ' i
E 2.75
3 1
9' l

2.25 .

 

 

0 50 100 150 200
LPS (ug/mL)

 

 

 

Figure 19: Dose response curve of CFTR mediated peak efflux in response to increasing
concentrations of LPS. Cultured 2WT2 cells were treated with the concentration of LPS
indicated for the entire duration of sample replacement. Values represent the average
peak efflux normalized to baseline in response to treatment with the cAMP elevating
agents forskolin and IBMX which were administered from 10 minutes into sample
replacement until the end of the experiment. The numbers above each point indicate the
p-value compared to cultures treated with only cAMP elevating agents. Values are mean
:t SEM (n = 3 to 6).

81

 

Lipopolysaccharide causes a dose-dependent increase
in cAMP stimulated CFTR activity

 

 

 

 

 

 

 

 

 

 

12 ~ — w—
0.0001
’3
.5 i
'6 ..,_;.._-_ M.-.
I
2
\
u
a
S.
x
I
E
III
6 .
0 50 100 150 200

LPS concentration (ug/mL)

 

 

Figure 20: Dose response curve of CF TR mediated iodide efflux area under the curve in
response to increasing concentrations of LPS. Cultured 2WT2 cells were treated with the
concentration of LPS indicated for the entire duration of sample replacement. Values
represent the average AUC normalized to baseline in response to treatment with the
cAMP elevating agents forskolin and IBMX which were administered from 10 minutes
into sample replacement until the end of the experiment. The numbers above each point
indicate the p-value compared to cultures treated with only cAMP elevating agents.
Values are mean :1: SEM (n = 3 to 6). 0.0629 bt 200 and 50

82

 

Treatment with P. aeruginosa PA14 increases
cAMP stimulated peak efflux

3.2 ——————__0.0918

Peak efflux (fold baseline)

 

 

   

N

 

Control pA14

 

Figure 21: Increase in CFTR mediated peak efflux in response to treatment with P.
aeruginosa PA14. Cultured 2WT2 cells were treated at an MOI of 100 bacteria /
epithelial cell for the entire duration of sample replacement. Values represent the average
peak efflux normalized to baseline in response to treatment with the cAMP elevating
agents forskolin and IBMX which were administered from 10 minutes into sample
replacement until the end of the experiment. The numbers above each point indicate the
p-value compared to cultures treated with only cAMP elevating agents. Values are mean
1 SEM (n = 3).

83

 

Treatment with P. aeruginosa PA14 increases
cAMP stimulated area under the efflux curve

 

0.0402

5”
U1

00

Efflux (AUC/ baseline)
\l
\l U1

9‘
U1

 

 

   

Control PA14 l

i

 

Figure 22: Increase in CF TR mediated iodide efflux area under the curve in response to
treatment with P. aeruginosa PA14. Cultured 2WT2 cells were treated at an MOI of 100
bacteria / epithelial cell for the entire duration of sample replacement. Values represent
the average AUC normalized to baseline in response to treatment with the cAMP
elevating agents forskolin and IBMX which were administered from 10 minutes into
sample replacement until the end of the experiment. Values are mean 1 SEM (n = 3).

84

 

 

 

Brief treatment with LPS does not reduce CFTR
mediated peak efflux

 

 

Peak efflux (fold baseline)

 

   

FSK LPS + FSK

Figure 23: No significant difference in CFTR mediated peak efflux was detected in
response to brief treatment with IOOug/mL LPS. Cultured 2WT2 cells were treated with
a cocktail of lOOug/mL LPS, 10uM forskolin and IOOuM IBMX beginning 10 minutes
into sample replacement until the end of the experiment. Values represent the average
peak efflux normalized to baseline in response to treatment with the cocktail. Values are
mean 1 SEM (n = 3).

85

 

 

 

Brief treatment with LPS reduces CFTR mediated area
under the efflux curve

 

0.0732

Efflux Rate (AUC/ baseline)

 

 

   

FSK LPS + FSK

Figure 24: Decreased CF TR mediated iodide efflux area under the efflux curve was
detected in response to brief treatment with IOOug/mL LPS. Cultured 2WT2 cells were
treated with a cocktail of 100ug/mL LPS, 1011M forskolin and 100uM IBMX beginning
10 minutes into sample replacement until the end of the experiment. Values represent the
average AUC normalized to baseline in response to treatment with the cocktail. Values
are mean 1 SEM (n = 3).

86

 

 

CFTR S TLR4 E Adaptorprotein(s)
nub ERM P. aeruginosa _.Cytoske|eta|
elements

 

 

Phagocytosis

10”]

Figure: 25 Model of a possible interaction between CFTR and TLR3 during phagocytosis

 

 

of P. aeruginosa. P. aeruginosa is thought to interact with TLR4 through LPS located in
its outer membrane and with TLR5 through flagella. This activates signal transduction
pathways that activate inflammation. I propose that the bacteria / receptor complex then
forms a complex with CFTR to recruit cytoskeletal components and to lead to
inactivation of the inflammatory signal. One signal that is generated is responsible for
recruiting CFTR to the membrane to replish channels removed during the phagocytic
event. The complex is also able to phagocytise the bacterium.

87

 

REFERENCES
Akira S. 2003. Toll-like receptor signaling. JBC. 278-40: 38105-38108

Andersen DH. 1938. Cystic fibrosis of the pancreas and its relation to celiac
disease. Am J Dis Child. 56: 344

Andersen DH and RG Hodges. 1946. Celiac syndrome v. genetics of cystic
fibrosis of the pancreas with a consideration of etiology. Am J Dis Child.
72:62

Backhed F, S Normark, EKH Schweda, S Oscarson, A Richter-Dahlfors. 2003.
Structural requirements for TLR4-mediated LPS signalling: a biological
role for LPS modifications. Microbes and Infection. 5: 1057—1063

Baird RM, H Brown, AW Smith, ML Watson. 1999. Burkholderia cepacia is
resistant to the antimicrobial activity of airway epithelial cells.
Immunopharmacology 44: 267—272

Balfour-Lynn IM, R Dinwiddie. 1996. Role of corticosteroids in cystic fibrosis
lung disease. Journal of the Royal Society of Medicine. 89: 8—13

Bals R, DJ Weiner, JM Wilson. 1999. The innate immune system in cystic
fibrosis lung disease. JCI. 103-3: 303-307

Barasch J, Q Al-Awqati. 1993. Defective acidification of the biosynthetic
pathway in cystic fibrosis. J Cell Sci Suppl. 17: 229-233

Becq F, C Auzanneau, C Norez, R Derand, L Bulteau-Pignoux. 2003. Radiotracer
flux method to study CFTR charmel activity: regulation, pharmacology
and drug discovery. JOURNAL?

Blander JM, R Medzhitov. 2004. Regulation of phagosome maturation by signals
from toll-like receptors. Science. 304: 1014-1018

Boucher RC, MJ Stutts, MR Knowles, L Cantley, JT Gatzy. 1986. Na+ transport
in cystic fibrosis respiratory epithelia: abnormal basal rate and response to
adenylate cyclase activation. J Clin Invest. 78: 1245-1252

Boucher RC. 2007. Airway surface dehydration in cystic fibrosis: pathogenesis
and therapy. Annual review of medicine. 5 8: 157-170

Bradbury NA, JA Cohn, CJ Venglarik, RJ Bridges. 1994. Biochemical and
biophysical identification of cystic fibrosis transmembrane conductance
regulator chloride channels as components of endocytic clathrin-coated
vesicles. The Journal of Biological Chemistry. 269-11: 8296-8302

88

 

Burkholder WH. 1950. Sour skin, a bacterial rot of onion bulbs. Phytopathology.
40: 115—7

Burns JL, M Jonas, EY Chi, DK Clark, A Berger, A Griffith. 1996. Invasion of
respiratory epithelial cells by Burkholderia (Pseudomonas) cepacia. Infect.
Immun.64-10: 4054-4059

Cannon CL, MP Kowalski, KS Stopak, GB Pier. 2003. Pseudomonas aeruginosa—
Induced Apoptosis Is Defective in Respiratory Epithelial Cells Expressing
Mutant Cystic Fibrosis Transmembrane Conductance Regulator. Am J
Respir Cell Mol Biol. 29: 188—197

Carmeli Y, N Troillet, GM Eliopoulos, MH Samore. 1999. Emergence of
antibiotic- resistant Pseudomonas aeruginosa: comparison of risks
associated with different antipseudomonal agents. Antimicrobial agents
and chemotherapy. 43-6: 1379- 1382

Chandy G, M Grabe, HPH Moore, TE Machen. 2001. Proton leak in regulation of
golgi pH in respiratory epithelial cells. Am J Cell Physiol. 28]: C908-
C921

Chappe V, T Irvine, J Liao, A Evagelidis, JW Hanrahan. 2005. Phosphorylation
of CFTR by PKA promotes binding of the regulatory domain. The EMBO
Journal. 24: 2730—2740

Chmiel JF, PB Davis. 2003. State of the art: why do the lungs of patients with
cystic fibrosis become infected and why can’t they clear the infection?
Respiratory Research. 4: 8-20

Ciche TA, K Kim, B Kaufmann-Daszczuk, KCQ Nguyen, DH Hall. 2008. Cell
Invasion and Matricide during Photorhabdus luminescens Transmission by

Heterorhabditis bacteriophora Nematodes. Applied and Environmental
Microbiology. 74-8: 2275— 2287

Conner SD, SL Schmid. 2003. Differential requirements for AP-2 in clathrin-
mediated endocytosis. JCB. 162-5: 773-779

Cox AD. SG Wilkinson. 1991. Ionizing groups in lipopolysaccharides of
Pseudomonas cepacia in relation to antibiotic resistance. Molecular
Microbiology. 5-3: 641-646

Cystic Fibrosis Foundation. 2005. Patient Registry Annual Data Report, Bethesda,
Maryland

89

Darling KE, TJ Evans. 2001. Novel Non-CFTR-Dependent Pseudomonas
aeruginosa Epithelial Intemalization Results in Intracellular Bacterial
Replication and Transmigration without Host Cell Death. Abstr Intersci
Conf Antimicrob Agents Chemother Intersci Conf Antimicrob Agents
Chemother. Dec 16-19; 41: abstract no. B-973

Davis PD. 2006. Cystic fibrosis since 1938. Am J Respir Crit Care Med. 173:
475-482

de Bentzrnann S, P Roger, F Dupuit, O Bajolet-Laudinat, C Fuchey, MC
Plotkowski, E Puchelle. 1996. Asialo GM] is a receptor for Pseudomonas

aeruginosa adherence to regenerating respiratory epithelial cells. Infect.
Immun. 64-5: 1582-1588

Delgado MA, JF Poschet, V Deretic. 2006. Nonclassical pathway of
Pseudomonas aeruginosa DNA-induced interleukin-8 secretion in cystic
fibrosis airway epithelial cells. Infection and immunity. 74-5: 2975-2984

Denis M, K Rarnotar, K Vandemheen, E Tullis, W Ferris, F Chan, C Lee, R
Slinger, SD Aaron. 2007. Infection With Burkholderia cepacia Complex

Bacteria and Pulmonary Exacerbations of Cystic Fibrosis. Chest. 131:
1188-1196

Di A, ME Brown, LV Deriy, C Li, FL Szeto, Y Chen, P Huang, J Tong, AP
Naren, V Bindokas, HC Palfrey, DJ Nelson. 2006. CFTR regulates
phagosome acidification in macrophages and alters bactericidal activity.
Nature Cell Biology. 8-9: 933-944

di Sant’Agnese PA, RC Darling, GA Perera, E Shea. 1953. Abnormal electrolyte
composition of sweat in cystic fibrosis of the pancreas. Clinical
significance and relationship to disease. Pediatrics. 12: 549-563

Doerrler WT. 2006. Lipid trafficking to the outer membrane of Gram-negative
bacteria. Molecular Microbiology. 60-3: 542—552

Donaldson SH, RC Boucher. 2006. Pathophysiology of Cystic Fibrosis. Ann
Nestlé. 64: 101—109

Dudez T, F Borot, S Huang, BR Kwak, M Bacchetta, M Ollero, BA Stanton, M
Chanson. 2008. CFTR in a lipid raft-TNFRI complex modulates gap
junctional intercellular communication and IL-8 secretion. Biomchimica
et Biophysica Acta. 1783: 779- 788

90

Dunn KW, J Park, CE Sernrad, DL Gelman, T Shevell, TE McGraw. 1994.
Regulation of endocytic trafficking and acidification are independent of
the cystic fibrosis transmembrane regulator. The Journal of Biological
Chemistry. 269-7: 5336- 5345

Elborn JS, DJ Shale, JR Britton. 1991. Cystic fibrosis: current survival and
population estimates to the year 2000. Thorax. 46: 881-885

Fearon DT, RM Locksley. 1996. The instructive role of innate immunity in the
acquired immune response. Science. 272-5258: 50-54

Fletcher EL, SMJ Fleiszig, NA Brennnan. 1993. Lipopolysaccharide in adherence

of Pseudomonas aeruginosa to the cornea and contact lenses. Invest
Ophthalmol Vis Sci. 34: 1930-1936

Fox A. Bacteriology — Chapter Four: Cell envelope, spores and macromolecular
biosynthesis. Microbiology and immunology online. Accessed 5/5/09

Ganguli A, AK Tripathy. 2002. Bioremediation of toxic chromium from
electroplating effluent by chromate-reducing Pseudomonas aeruginosa
A2Chr in two bioreactors. Appl Microbiol Biotechnol. 58: 416-420

Gelman MS, RR Kopito. 2002. Rescuing protein conformation: prospects for
pharmacological therapy in cystic fibrosis. JCI. 110: 1591-1597

Gentzsch M, XB Chang, L Cui, Y Wu, VV 02013, A Choudhury, RE Pagano, JR
Riordan. 2004. Endocytic trafficking routes of wild type and dF508 cystic

fibrosis transmembrane conductance regulator. Molecular Biology of The
Cell. 15: 2684- 2696

Gerritsen J. 2000. Host defence mechanisms of the respiratory system. Peadiatric
respiratory reviews. 1: 128—134

Greene, Carroll, Smith, Taggart, Devaney, Griffin, O’Neill, McElvaney. 2005.
TLR- induced inflammation in cystic fibrosis and non-cystic fibrosis
airway epithelial cells. Journal of Immmunology. 174: 1638-1646

Govan JRW, JE Hughes, P Vandamme. 1996. Burkholderia cepacia: medical,
taxonomic and ecological issues. J. Med. Microbiol. 45: 395-407

Hanrahan JW. 2000. Airway plumbing. JCI. 105: 1343-1344

Hart AC, C Winstanley. 2002. Persistant and aggressive bacteria in the lungs of
cystic fibrosis children. British Medical Bulletin. 61: 81-96

91

 

Haubler S, I Ziegler, A L0ttel, F G0tz, M Rohde, D Wehmhtihner, S
Saravanamuthu, B Tilmmler, I Steinmetz. 2003. Highly adherent small-

colony variants of Pseudomonas aeruginosa in cystic fibrosis lung
infection. J Med Microbiol. 52: 295-301

Hiemstra PS. 2007. Antimicrobial peptides in the real world: implications for
cystic fibrosis. Eur Respir J. 29: 617—61 8

High N, J Mounier, MC Prevost, PJ Sansonetti. 1992. IpaB of Shigella flexneri
causes entry into epithelial cells and escape from the phagocytic vacuole.
The EMBO Journal. 11-5: 1991-1999

Hippenstiel S, B Opitz, B Schmeck, N Suttorp. 2006. Lung epithelium as a
sentinel and effector system in pneumonia= molecular mechanisms of
pathogen recognition and signal transduction. Respiratory Research. 7-97

Hubeau C, E Puchelle, D Gaillard. 2001. Distinct pattern of immune cell
population in the lung of human fetuses with cystic fibrosis. J Allergy 01in
immunol. 108: 524- 529

Imundo L, J Barasch, A Prince, Q Al-Awqati. 1995. Cystic fibrosis epithelial cells
have a receptor for pathogenic bacteria on their apical surface. PNAS. 92:
3019-3023

Isles A, I Maclusky, M Corey, R Gold, C Prober, P Fleming, H Levison. 1984.
Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J
Pediatr. 104-2: 206-210

Jaques A, E Daviskas, JA Turnton, K McKay, P Cooper, RG Stirling, CF
Robertson, PTP Bye, PN LeSouef, B Shadbolt, SD Anderson, B Charlton.

2008. Inhaled mannitol improves lung function in cystic fibrosis. Chest.
133: 1388-1396

Kerem B, JM Rommens, JA Buchanan, D Markiewicz, TK Cox, A Chakravarti,
M Buchwald, LC Tsui. 1989. Identification of the cystic fibrosis gene:
genetic analysis. Science. 245: 1073-1080

Knowles MR, MJ Stutts, A Spock, N Fischer, JT Gatzy, RC Boucher. 1983.
Abnormal ion permeation through cystic fibrosis respiratory epithelium.
Science. 221: 1067- 1070

Kowalski MP, GB Pier. 2004. Localization of cystic fibrosis transmembrane
conductance regulator to lipid rafts of epithelial cells is required for
Pseudomonas aeruginosa— induced cellular activation. The Journal of
Immmunology. 172: 418-425

92

_ —-—-_==__—_

 

Kronborg G. 1995. Lipopolysaccharide (LPS), LPS-immune complexes and
cytokines as inducers of pulmonary inflammation in patients with cystic

fibrosis and chronic Pseudomonas aeruginosa lung infection. APMIS
Suppl. 50: 1-30

Kube DM, D Fletcher, PB Davis. 2005. Relation of exaggerated cytokine

responses of CF airway epithelial cells to PA01 adherence. Respiratory
Research. 6-69

Kunzelman K. 2001. CFTR: interacting with everything? News Physiol. Sci. 16:
167-170

Long KL, KB Walsh. 1997. Iodide efilux measurements with an iodide-selective
electrode: A non-radioactive procedure for monitoring cellular chloride
transport. Methods in Cell Science. 19: 207—212

Loutet SA, RS Flannagan, C Kooi, PA Sokol, MA Valvano. 2006. A Complete
Lipopolysaccharide Inner Core Oligosaccharide Is Required for
Resistance of Burkholderia cenocepacia to Antimicrobial Peptides and
Bacterial Survival In Vivo. Journal of Bacteriology. 188-6: 2073—2080

Lyczak JB, CL Cannon, GB Pier. 2002. Lung infections associated with cystic
fibrosis. Clinical Microbiology Reviews. April 2002: 194-222

Mahenthiralingam E, TA Urban, JB Goldberg. 2005. The multifarious,
multireplicon burkholderia cepacia complex. Nature. 3: 144-156

Mateu E, F Calafell, MD Ramos, T Casals, J Bertranpetit. 2002. Can a place of
origin of the main cystic fibrosis mutations be identified? Am J Hum
Genet. 70: 257-264

Matsui H, BR Grubb, R Tarran, SH Randell, JT Gatzy, CW Davis, RC Boucher.
1998. Evidence for periciliary liquid layer depletion, not abnormal ion

composition, in the pathogenesis of cystic fibrosis airway disease. Cell.
95: 1005-1015

Matsumura K, K Orita, Y, Wakamoto, K Yasuda. 2006. Phagocytic response to
fully controlled plural stimulation of antigens on macrophage using on-
chip microcultivation system. Journal of Nanobiotechnology. 4:7

McClean S, M Callaghan. 2009. Burkholderia cepacia complex: epithelial cell—

pathogen confrontations and potential for therapeutic intervention. Journal
of Medical Microbiology. 58: 1-12

93

 

Morales FC, Y Takahashi, EL Krelmann, MM Georgescu. 2004. Ezrin-radixin-
moesin (ERM)-binding phosphoprotein 50 organizes ERM proteins at the
apical membrane of polarized epithelia. PNAS. 101-51: 17705-17710

Motley A, NA Bright, MNJ Seaman, MS Robinson. 2003. Clathrin-mediated
endocytosis in AP-2-depleted cells. JCB. 162-5: 909-918

Muhlebach MS, W Reed, TL Noah. 2004. Quantitative cytokine gene expression
in CF airway. Pediatr Pulmonol. 37-5: 393-399

Pier GB, P Ames. 1984. Mediation of the killing of rough, mucoid isolates of
Pseudomonas aeruginosa from patients with cystic fibrosis by the
alternative pathway of complement. J Infect Dis. 150-2: 223-8

Pier GB, M Grout, TS Zaidi, JC Olsen, LG Johnson, JR Yankaskas, JB Goldberg.
1996. Role of mutant CF TR in hypersusceptibility of cystic fibrosis
patients to lung infections. Science. 271: 64-67

Pier GB, M Grout, TS Zaidi. 1997. Cystic fibrosis transmembrane conductance
regulator is an epithelial cell receptor for clearance of Pseudomonas
aeruginosa from the lung. PNAS. 94: 12088-12093

Pier GB. 1999. Evolution of the AF508 CFTR mutation: Response. Trends in
microbiology. 56-58

Pier GB. 2000. Peptides, Pseudomonas aeruginosa, polysaccharides and
lipopolysaccharides — players in the predicament of cystic fibrosis
patients. Trends in Microbiology. 8-6: 247-250

Pier GB, F Coleman, M Grout, M Franklin, DE Ohman. 2001. Role of alginate O
acetylation in resistance of mucoid Pseudomonas aeruginosa to apsonic
phagocytosis. Infection and Immunity. 69-3: 1895-1901

Pier G. 2007. Pseudomonas aeruginosa lipopolysaccharide: a major virulence

factor, initiator of inflammation and target for effective immunity. Int. J
Med Microbiol. 297(5): 277-295.

Pirone L, A Bragonzi, A Farcomeni, M Paroni, C Auriche, M Conese, L Chiarini,
C Dalmastri, A Bevivino, F Ascenzioni. 2008. Burkholderia cenocepacia
strains isolated from cystic fibrosis patients are apparently more invasive
and more virulent than rhizosphere strains. Environmental Microbiology.
10-10: 2773 — 2784

Quinton PM. 1983. Chloride impermeability in cystic fibrosis. Nature. 301: 421-
422

94

 

Ramphal R, V Balloy, J Jyot, Amrisha Verma, M Si-Tahar, M Chignard, 2008.
Control of Pseudomonas aeruginosa in the Lung Requires the Recognition

of Either Lipopolysaccharide or Flagellin. The Journal of Immunology.
181: 586-592

Riordan JR, JM Rommens, B Kerem, N Alon, R Rozrnahel, Z Grzelczak, J
Zielenski, S Lok, N Plavsic, JL Chou, ML Drumm, MC Iannuzzi, FS
Collins, LC. 1989. Identification of the cystic fibrosis gene: cloning and
characterization of complementary DNA. Science. 245: 1066-1073

Rodley PD, U Romling, B Tummler. 1995. A physical genome map of the
Burkholderia cepacia type strain. Molecular Microbiology. 17-1: 57-67

Romeo G, M Devoto, LJV Galietta. 1989. Why is the cystic fibrosis gene so
frequent? Human genetics. 84: 1-5

Rommens JM, MC Iannuzzi, B Kerem, ML Drumm, G Melmer, M Dean, R
Rozmahel, JL Cole, D Kennedy, N Hidaka, M Zsiga, M Buchwald, JR
Riordan, LC Tsui, FS Collins. 1989. Identification of the Cystic Fibrosis
Gene: Chromosome Walking and Jumping. Science. 245: 1059-1065

Rowe SM, S Miller, EJ Sorcher. 2005. Cystic fibrosis: mechanisms of disease.
NEJM. 352: 1992-2001

Ruckdeschel K, G Pfaffinger, R Haase, A Sing, H Weighardt, G Hacker, B
Holzrnann, J Heesemann. 2004. Singaling of apoptosis through TLRs
critically involves toll/IL-l receptor domain-containing adapter inducing
IFN-B, but not MyD88, in bacteria-infected murine macrophages. The
Journal of Immunology. 173: 3320- 3328

Saiman L, A Prince. 1993. Pseudomonas aeruginosa pili bind to asialoGMl which
is increased on the surface of cystic fibrosis epithelial cells. J Clin Invest.
92-4: 1875-80

Saini LS, SB Galsworthy, MA John, MA Valvano. 1999. Intracellular survival of
Burkholderia cepacia complex isolates in the presence of macrophage cell
activation. Microbiology. 145: 3465-3475

Sajjan US, MC, MA Karmali, JF Forstner. 1992. Binding of Pseudomonas
cepacia to Normal Human Intestinal Mucin and Respiratory Mucin from
Patients with Cystic Fibrosis. J. Clin. Invest. 89: 648-656

Sajjan U, Wu Y, Kent G, Forstner J. 2000. Preferential adherence of cable-
piliated burkholderia cepacia to respiratory epithelia of CF knockout mice

and human cystic fibrosis lung explants. J Med Microbiol. 49-10: 875-
885.

95

 

Sajjan US, JH Yang, MB Hershenson, JJ LiPuma. 2006. Intracellular trafficking
and replication of Burkholderia cenocepacia in human cystic fibrosis
airway epithelial cells. Cell Microbiol. 8-9: 1456-1466

Seamon KB, W Padgett, JW Daly. 1981. Forskolin: Unique diterpene activator of
adenylate cyclase in membranes and in intact cells. PNAS. 78-6: 3363-
3367

Sherertz RJ, FA Sarubbi. 1983. A three year study of nosocomial infections
associated with Pseudomonas aeruginosa. Journal of clinical
microbiology. July 1983: 160- 164

Schreiber R, A Hopf, M Mall, R Greger, K Kunzehnann. 1999. The first-
nucleotide binding domain of the cystic-fibrosis transmembrane
conductance regulator is important for inhibition of the epithelial Na+
channel. PNAS. 96: 5310—5315

Schroeder TH, MM Lee, PW Yacono, CL Cannon, AA Gerceker, DE Golan, GB
Pier. 2002. CFTR is a pattern recognition molecule that extracts
Pseudomonas aeruginosa LPS from the outer membrane into epithelial
cells and activates NF- KB translocation. PNAS. 99: 6907-6912

Smith JJ, SM Travis, IP Greenberf, MJ Welsh. 1996 Cystic fibrosis airway
epithelia fail to kill bacteria because of abnormal airway surface fluid.
Cell. 85: 229—236

Smith RS, BH Iglewski. 2003. Pseudomonas aeruginosa quorum sensing as a
potential antimicrobial target. JCI. 112: 1460-1465

Somerville JE, L Cassiano, B Bainbridge, MD Cunningham, RP Darveau. 1996.
A novel Escherichia coli lipid A mutant that produces an antiinflarnmatory
lipopolysaccharide. J Clin Invest. 97-2: 359-365

Soong G, B Reddy, S Sokol, R Adamo, A Prince. 2004. TLR2 is mobilized into
an apical lipid raft receptor complex to signal infection in airway epithelial
cells. JCI. 113: 1482-1489

Sun F, Z Mi, SB Condliffe, CA Bertrand, X Gong, X Lu, R Zhang, JD Latoche,
JM Pilewski, PD Robbins, RA Frizzell. 2008. Chaperone displacement
from mutant cystic fibrosis transmembrane conductance regulator restores
its function in human airway epithelia. FASEB. 22-9: 3225-3263

96

 

Taminiau B, M Daykin, S Swift, ML Boschiroli, A Tibor, P Lestrate, X De Bolle,
D O’Callaghan, P Williams, JJ Letesson. 2002. Identification of a
Quorum-Sensing Signal Molecule in the Facultative Intracellular Pathogen
Brucella melitensis. Infection and Immunity. 70-6: 3004—3011

Tirouvanziam R, S de Bentzmann, C Hubeau, J Hinrasky, J Jacquot, B Peault, E
Puchelle. 2000. Inflammmation and infection in naive human cystic
fibrosis airway grafts. Am J Respir Cell Mol Biol. 23: 121-127

Tirouvanziam R, I Khazaal, B Peault. 2002. Primary inflammation in human
cystic fibrosis small airways. Am J Lung Cell Mol Physiol. 283: L445-
451

Tomlin KL, RJ Malott, G Ramage, DG Storey, PA Soko, H Ceri. 2005. Quorum-
Sensing Mutations Affect Attachment and Stability of Burkholderia

cenocepacia Biofilms. Applied and environmental microbiology. 71-9:
5208-5218

Traub LM. 2003. Sorting it out: AP-2 and alternate clathrin adaptors in endocytic
cargo selection. JCB. 163-2: 203-208

Tyler SD, KR Rozee, WM Johnson. 1996. Identification of 181356, a New
Insertion Sequence, and Its Association with 18402 in Epidemic Strains of
Burkholderia cepacia Infecting Cystic Fibrosis Patients. Journal of
Clinical Microbiology. 34-7: 1610—1616

Unanue ER, PM Allen. 1987. The Basis for the Immunoregulatory Role of
Macrophages and Other Accessory Cells. Science. 236-4801: 551-557

VanDyke RW, KV Root, JH Schreiber, JM Wilson. 1992. Role of CFTR in
lysosomal acidification. Biochemical and Biophysical Research
Communications. 184-1 : 300-305

VanKooka, TBH Geijtenbeek. 2006. Toll-like receptors keep antigen sorting on
the right track. Immunity. 10/25/06: 525-527

Vaudaux P, FA Waldvogel. 1979. Gentamicin Antibacterial Activity in the
Presence of Human Polymorphonuclear Leukocytes. Antimicrobial agents
and chemotherapy. Dec: 743-749

Venkatakrishnan A, AA Stecenko, G King, TR Blackwell, KL Brigham, JW
Christrnan, TS Blackwell. 2000. Exagerated activation of nuclear factor-
kB and altered IkB-B processing in cystic fibrosis bronchial epithelial
cells.

97

 

 

Vij N, S Mazur, PL Zeitlin. 2009. CFTR Is a Negative Regulator of NFkB
Mediated Innate Immune Response. PLoS ONE. 4-2: e4664

Wang Z, EJ Bilsky, D Wang, F Porreca, W Sadée. 1999. 3-lsobutyl-1-
methylxanthine inhibits basal u—opioid receptor phosphorylation and
reverses acute morphine tolerance and dependence in mice. European
Journal of Pharmacology. 371-1: 1-9

Weixel KM, NA Bradbury. 2001. Endocytic adaptor complexes bind the C-
terminal domain of CFTR. Eur J Physiol. 443. 870-874

Welsh MJ, JJ Smith. 1995. Cystic fibrosis. Sci Am. 273: 52-59
Wilson JM. 1995. Gene therapy for cystic fibrosis. JCI. 96: 2547-2554
Wine JJ. 1999. The genesis of cystic fibrosis lung disease. JCI. 103: 309-312

Wu Q, L Zhong, MW Verghese, SH Randell. 2005. Airway epithelial cell
tolerance to Pseudomonas aeruginosa. Respiratory Research. 6-26

Yabuuchi E, Y Kosako, H Oyaizu, I Yano, H Hotta, Y Hashimoto, T Ezaki, M
Arakawa. 1992. Proposal of Burkholderia gen. nov. and transfer of seven
species of the genus Pseudomonas homology group II to the new genus,
with the type species Burkholderia cepacia (Palleroni and Holmes 1981)
comb. nov. Microbiol Immunol. 36-12: 1251-1275

Zhang P, WR Summer, GJ Bagby, S Nelson. 2000. Innate immunity and
pulmonary host defense. Immunological Reviews. 173: 39—51

98