studies on airway surface liquid in connection with cystic...

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2008

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 393

Studies on Airway Surface Liquid inConnection with Cystic Fibrosis

INNA KOZLOVA

ISSN 1651-6206ISBN 978-91-554-7328-0urn:nbn:se:uu:diva-9358

�� !��" �� # $� �%� �� $ �� $&%��%� '(�� $ ��)* +%� �� ,�� -��%*

��

./�� 0* !��"* 1�� 2��,�� 1��$�� 3�4�� ,��% �� (��* 2�� * �� 5 * �! ��* ��* 01�6 5�"75�7##87� !"7�*

�� $�� '�() �� $ �%� �� $�� %�� * �( �� %� �� $� �%� �� $�� '�(+9)� ,%��% %�� ,�� :�� ,�� $�� 4�� '213)*+%� ��,�� ,��% � �%� �� $ $�� %� ��,�� $�� 4�� ,%��% �%�

�� %�* ��$�� (+9 �� $�� ,�� %��%�� ,%��% �� $�� $�� * +%� �� $ 213 �� %��%�� $ �� $�� %��$�� *�,�� %�� $ ��$$�� %� �� $ �%� 213�� * 3�� $�� %�� ;�� %��*+%�� %� �� $ � �� %� $� �� %�� $ �%� 213 � ��$$�� %��* +, ��%�4��,�� %� �� $ 213� $�� ,%��% �� $�� %�� ,�� %�� * 0 �%� $�� %�� $�� %� �� %�� $�/� �� /�� %� $�/�7%�� * 0 �%�� %�� %� 213 �� ,��% �� '1��%��) ��< �%� �� %� ��/�� =7�� * +%� 1��%��7�� %� �� %� $�/�7%�� * ��% ��%�� ,�� %��>� �� $�� $�� %� $�� %�� $�� $ �� %�� '�?��-�87) �� $�� %�� %�� '�(�-8�7)� �� $�� $�� %��%� �� :��* +%� �� $ �%� 213 ,�� % � �� %��%� %�� :��* �( ��%�� % %��%�� $ 6� �� %� %��%� ��:��* +%� 213 �� ,�� %�� ,%�� %� �� $ 6� �� %� �� ,��% �%� @(#�" �� (+9 ,�� $�� %��%�� %� ��%� �� * 0� ,�� %�� %� �� $ �%� 213�� $�� %�� * +%� �� $�� $�� %��(+9 �� * �� %�� %�� %� ��$ � �� $� �( �� %�� %� �� * �� %� �$$�� $ �� %�� %� 213 �� %��7�� %��% �� %�� /�� ,��% �� ,%�� %��*

� �� $�� (+9� ��,�� $�� 4�� %�� %�� &�� $��

�� !�� !" #$% �� &'$#%()��

A 0� ./�� !��"

0116 �?#�7?!�?01�6 5�"75�7##87� !"7��75 #" '%��>>��*;�*��>��B��C��75 #")

List of papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals: I Kozlova I, Nilsson H, Phillipson M, Riederer B, Seidler U, Col-

lege WH, Roomans GM. X-ray microanalysis of airway sur-face liquid in the mouse. Am J Physiol Lung Cell Mol Physiol 2005; 288: L874-L878.

II Kozlova I, Hühn MH, Flodström-Tullberg M, Wilke M, Scholte

BJ, Roomans GM. Anaesthesia and the elemental content of mouse nasal fluid. Submitted, 2008.

III Kozlova I, Nilsson H, Henriksnäs J, Roomans GM. X-ray mi-

croanalysis of apical fluid in cystic fibrosis airway epithelial cell lines. Cell Physiol Biochem 2006; 17: 13-20.

IV Kozlova I, Vanthanouvong V, Almgren B, Högman M, Roomans

GM. Elemental composition of airway surface liquid in the pig determined by X-ray microanalysis. Am J Respir Cell Mol Biol 2005; 32: 59-64.

V Vanthanouvong V, Kozlova I, Johannesson M, Nääs E, Nordvall

SL, Dragomir A, Roomans GM. Composition of nasal airway surface liquid in cystic fibrosis and other airway diseases de-termined by X-ray microanalysis. Microsc Res Techn 2006; 69: 271-276.

VI Kozlova I, Hjelte L, Roomans GM. Effect of mist tent on the

ion content of nasal fluid in patients with cystic fibrosis. Submitted, 2008.

Reprints were made with permission from the publishers.

Contents

Introduction.....................................................................................................9 1.1 Cystic Fibrosis......................................................................................9 1.2 The CFTR gene and CFTR protein ....................................................11 1.3 Pathogenesis of cystic fibrosis lung disease.......................................13

1.3.1 The airways and airway epithelium............................................13 1.3.2 Ion and water transport in the respiratory epithelium.................14 1.3.3 Airway surface liquid .................................................................17

1.4 Survival and pharmacological treatment of CF..................................23 1.5 Animal models ...................................................................................24

1.5.1 Mouse models of CF...................................................................24 1.5.2 Pig models of CF ........................................................................25

Aims..............................................................................................................27

Materials and methods ..................................................................................29 3.1 Animals and anesthesia ......................................................................29

3.1.1 Mice (Papers I, II).......................................................................29 3.1.2 Pigs (Paper IV) ...........................................................................30

3.2 Cell lines (Paper III) ...........................................................................30 3.2.1 Airway epithelial cell lines .........................................................30 3.2.2 Liquid-liquid and air-liquid interface cell cultures .....................31 3.2.3 Transepithelial electrical resistance (TEER) measurements.......31 3.2.4 pH measurements........................................................................31

3.3 Frozen-hydrated samples (Papers I, III, IV).......................................32 3.3.1 Pig tracheae and bronchi.............................................................32 3.3.2 X-ray microanalysis....................................................................32 3.3.3 Standards ....................................................................................33 3.3.4 Mouse tracheae ...........................................................................33 3.3.5 Liquid-liquid interface cell cultures............................................33

3.4 Ion-exchange beads (Papers I, II, III, IV)...........................................34 3.4.1 Pig tracheae.................................................................................34 3.4.2 X-ray microanalysis....................................................................34 3.4.3 Mouse tracheae ...........................................................................34 3.4.4 Nasal fluid collection in mice .....................................................35 3.4.5 Liquid-liquid interface cultures ..................................................35

3.5 Plasma and serum samples (Papers I, IV) ..........................................36

3.5.1 Pig plasma and serum .................................................................36 3.5.2 Mouse plasma and serum............................................................36

3.6 Morphological studies (Papers I, IV) .................................................36 3.7 Humans (Papers V, VI) ......................................................................37

3.7.1 Study design of paper V .............................................................37 3.7.2 Study design of paper VI ............................................................38 3.7.3 Defenition of chronic Pseudomonas aeruginosa colonization/infection ..........................................................................39 3.7.4 Nasal fluid collection from humans............................................39 3.7.5 X-ray microanalysis....................................................................39 3.7.6 Standard calibration curve ..........................................................40

3.8 Statistical analysis ..............................................................................40

Results...........................................................................................................41 4.1 Paper I ................................................................................................41 4.2 Paper II ...............................................................................................42 4.3 Paper III..............................................................................................42 4.4 Paper IV..............................................................................................44 4.5 Paper V...............................................................................................45 4.6 Paper VI..............................................................................................45

Discussion .....................................................................................................47 5.1 Ionic composition of ASL in mice, pigs, air-liquid & liquid-liquid interface cell cultures (Papers I, II, III, IV) ..............................................47 5.2 Ionic composition of nasal ASL in healthy and diseased humans (Papers V, VI) ..........................................................................................52

Conclusions and future perspectives.............................................................57

Sammanfattning på svenska..........................................................................59

Acknowledgements.......................................................................................61

References.....................................................................................................63

Abbreviations

�F508 Deletion of phenylalanine in position 508 in the CFTRstructure

16HBE14o- Human bronchial epithelial cell line (wild type CFTR)that maintains cell polarity and is transformed with an origin of replication defective simian virus (SV40) con-taining plasmid pSVori-. The number of the clone is 16 and the number of passages is 14 originally

ANOVA Analysis of variance ASL Airway surface liquid ATP Adenosine triphosphate cAMP Cyclic adenosine monophosphate CF Cystic fibrosis CFBE41o- Cystic fibrosis (�F508/�F508 homozygous) bronchial

epithelial cell line that maintains polarity and is trans-formed with an origin of replication defective simian virus (SV40) containing plasmid pSVori-. The number of passages is 41 originally

CFTR Cystic fibrosis transmembrane conductance regulator EMEM Eagle’s minimal essential medium ENaC Epithelial sodium channel PBS Phosphate buffered saline PCD Primary ciliary dyskinesia PCL Periciliary layer SEM Scanning electron microscopy Sephadex Trade name for a cross-linked dextran gel with ion-

exchange capacity TEER Transepithelial electrical resistance TEM Transmission electron microscope XRMA X-ray microanalysis

9

Introduction

1.1 Cystic Fibrosis

Cystic fibrosis (CF), the most common fatal autosomal recessive disease among the white population, is characterized by dysfunctional chloride ion transport across epithelial cells causing the formation of a thick viscous se-cretion which plugs ducts in the lungs, intestine, pancreas, liver, and repro-ductive organs. A large number of symptoms, including sinus infections, poor growth, diarrhea and infertility, are the result of the effect of CF on the other parts of the body. Most of the CF patients have complications in the respiratory system, because of bacterial colonization, damage to the epithe-lium and exposure of matrix proteins that increase bacterial adherence. The cruel series of infection, inflammation and diminished mucociliary clearance eventually results in chronic obstructive lung disease and irreversible respi-ratory deficiency. CF occurs in approximately one in 2,500 live births. Ap-proximately one in 20 whites carries a mutant CF gene allele [1]. Only about one in 17,000 black infants are afflicted, and CF is very rare in Asian popu-lations [2].

Historical reference from the Middle Ages described children whose brows tasted salty and who died prematurely, but it was not until 1936 when Fan-coni et al. were probably first to refer to the disease as “cystic fibrosis with bronchiectasis” and recognized it as a separate illness from celiac disease [3]. Andersen in 1938 described the clinical signs of pancreatic fibrosis and pulmonary disease in detail and used the term “cystic fibrosis of the pan-creas” [4]. This disease was characterized by malabsorption of fat and pro-tein, diarrhea, poor growth, and pulmonary infection. Pancreatic damage and lack of pancreatic enzyme secretion accounted for nutritional failure, which was assumed to lead to susceptibility to lung infection, often the final event. The thick, sticky mucus blocking the ducts of the airways as the main reason of the progressive lung disease was described by Farber who also gave rise to the alternative name “mucoviscidosis” [5]. The life expectancy was ap-proximately 6 months and death often occurred from the lung infection. A significant discovery was made in 1948 by Paul di Sant’Agnese, who de-tected that infants with CF had abnormal sweat with excess of sodium and chloride [6]. Elevated sweat chloride concentration offered a convenient

10

diagnostic test. The pilocarpine iontophoresis technique of Gibson and Cooke [7] allowed such testing in practice. Practically every patient with a clinical diagnosis of CF has a significantly high sweat chloride concentra-tion, and hardly any other diseases produce elevated sweat electrolytes, but these diseases are clinically quite different from CF. Even milder patients without pancreatic insufficiency could be identified with the pilocarpine iontophoresis technique. The sweat test promptly became and remains the most common and reliable single parameter for differential diagnosis of CF, probably excluding genetic analysis. Paul Quinton in 1983 used sweat ducts to identify chloride transport as the basic physiologic defect in CF [8]. In parallel, Boucher and coworkers [9] and Knowles and colleagues [10] dis-covered increased sodium absorption across airway epithelia in CF patients. These results implied that abnormal absorption of Na+ was dependent on the ENaC and was a regular characteristic of CF not only in sweat glands, but also in the airways. Only in 1989, the CF gene was discovered and its iden-tity was examined using cells that originated from the sweat duct [11, 12]. This discovery demonstrated that the basic defect was in a cAMP-regulated chloride channel, the cystic fibrosis transmembrane conductance regulator (CFTR), encoded by the CF gene, and this gave opportunities for new diag-nostic tests, for research, and a hope for using gene therapy.

At present, considerable progress in basic and clinical research has resulted in therapeutic improvements. Most patients are diagnosed before the age of two years. The life expectancy for patients with CF has improved signifi-cantly from 6 months to more then 30 years in the USA [3] and 40-45 years in Sweden [13], with a treatment that does not depend on specific knowledge of the basic defect. Researchers around the world are still working on cor-recting the basic defect in CF and trying to find out new approaches to ther-apy which could help to extend and improve life of CF patients. For the fu-ture, several approaches have been proposed: (1) to treat the CF-related ion transport defects pharmacologically, decreasing sodium absorption and in-creasing chloride secretion; (2) to use gene therapy, inserting a normal copy of the CF gene into the appropriate cells. A possible future development is the use of stem cells for therapy.

11

1.2 The CFTR gene and CFTR protein

The cystic fibrosis transmembrane conductance regulator gene is found in region q31.2 on the long (q) arm of human chromosome 7, is 250000 base pairs (bp) long and contains 27 axons (Figure 1). The CFTR gene encodes the CFTR protein which is 1480 amino acids long and has a molecular weight of 168,173 Da. The CFTR gene was identified in 1989 and RNA transcripts of CF gene were found in epithelial cells from various sources but not in brain or lymphoblast cell lines [11].

Figure 1. Gene location. Figure 2. Schematic drawing of the CFTR protein [11]

The CFTR protein is anchored to the cell membrane of cells in the sweat glands, lung, pancreas, liver, digestive and reproductive tracts. The CFTR protein spans this membrane and acts as a channel connecting the cytoplasm to the surrounding fluid. This channel is primarily responsible for controlling the movement of chloride ions across the cell membrane. Sodium is the most common ion in the extracellular space and the combination of sodium and chloride creates a salt (NaCl), which is lost in large amounts in the sweat of individuals with CF. Structurally, the CFTR protein belongs to the ABC (ATP-binding cassette) class of ion channels or transporters (Figure 2). The protein possesses five domains: two membrane spanning domains each of which comprises 6 alpha helices and forms the chloride channel, two nucleo-tide-binding domains that bind and hydrolyze ATP (adenosine triphosphate), and a regulatory (R) domain. A regulatory binding site on the protein allows

12

activation by phosphorylation, mainly by a cAMP-dependent protein kinase [14]. Cystic fibrosis occurs when there is a mutation in both alleles of the CFTR gene. Based on their effect on protein synthesis and function, the nu-merous CFTR mutations are classified as follows [15] (Figure 3):

Class I – mutations that produce no protein due to a stop mutation or fatal errors in the CFTR mRNA synthesis;

Class II – mutations in which the native CFTR fails to reach the apical membrane because of defective processing (e.g., �F508 CFTR);

Class III – mutations that produce a protein that reaches the plasma mem-brane but fails to respond to cAMP;

Class IV – mutations that produce a cAMP-responsive channel with re-duced conductance;

Class V – mutations that cause a decrease in the amount of CFTR proteins produced and that cause a small defect in maturation but normal or increased chloride transport activity (e.g., A455E, P574H);

Class VI – mutations that harbor nucleotide alterations that affect the regulatory properties of CFTR protein.

Figure 3. CFTR mutations causing CF [15]

13

There are over 1300 different mutations that can produce CF and there are several mechanisms by which these mutations cause errors in the CFTR protein [16]. The most common mutation, �F508 is a deletion (�) of three nucleotides that results in the loss of the amino acid phenylalanine (F) at the 508th position of the protein [11] (Figure 4). This mutation is present in 70% of the chromosomes of CF patients worldwide and 90% of CF patients in the United States [17]. The �F508 mutation, for instance, creates a protein that does not fold normally. 95% of the �F508-CFTR is retained in the endo-plasmic reticulum, from where it is degraded by the ubiquitin-proteasome system [18]. Once the CFTR protein reaches the cell membrane, its turnover is relatively fast and the wild-type (wt) CFTR has a half-life of ~72 hours, while the half-life of �F508-CFTR is ~4 hours [19]. The fully mature CFTR is a 180-190kDa protein and is recycled by endocytosis or degraded by the lysosomal proteases.

Figure 4. The �F508 deletion [11]

1.3 Pathogenesis of cystic fibrosis lung disease

1.3.1 The airways and airway epithelium

The airway is lined by a number of different cell types: columnar ciliated cells, non-ciliated columnar cells, goblet (mucus-secreting) cells, basal and Clara cells. Basal airway cells rest on a basement membrane and are instru-mental in attaching the columnar cells to the basement membrane and under-lying connective tissue [20]. The most abundant type of cell is the ciliated (fluid-secreting) columnar cells. Each cell has about 300 cilia on its apical surface [21] with an average length of 6 �m [22]. In CF patients the number of microvilli per ciliated cell is decreased by 40% [23]. Epithelial cells have a unique structure and function [24]. The fundamental feature that character-izes an epithelial cell is its polarity. The apical membrane of the airway

14

epithelial cells has cilia. The cilia transport the mucus, which traps inhaled particles, towards the pharynx where it can be transferred to the digestive tract. Mucus is produced by goblet cells and by the submucosal glands. The columnar cells contribute to the mucus by secretion of fluid containing ions and water which forms a fluid layer around the cilia to facilitate their move-ment. The lateral membrane may have gap junctions and is adjacent to a lateral intercellular space. These two membranes are separated by a tight junction that encircles the cells and serves as a boundary between apical and basolateral membranes. Tight junctions have a selective permeability to ions and other solutes. Thus there are two pathways that ions and solutes may follow in crossing the epithelium: the cellular pathway (through the cells) and the paracellular pathway (between the cells). The ciliated cell is the only cell type expressing CFTR in nasal, bronchial, and proximal bronchiolar surface epithelia. But CFTR was not or sparsely present in the submucosal gland cells in tissue specimens from normal subjects [25]. The apical regions of ciliated cells of the airway epithelia of CF patients (�F508 mutation) were completely devoid of CFTR immunostaining signal, a finding confirmed by the absence of mature CFTR protein in biochemical studies of the same samples [26].

1.3.2 Ion and water transport in the respiratory epithelium

It is generally assumed that the ion transport processes in the airway epithe-lium are similar in the upper and lower airways [27]. During the last decades evidence has accumulated that a defective Cl- transport across epithelial cells plays a fundamental role in the etiology of CF. An understanding of ion transport abnormalities may ultimately lead us to a more rational develop-ment of a specific therapy. In the airway epithelia ciliated cells perform two main ion fluxes: Cl- secretion and Na+ absorption (Figure 5) [28].

Figure 5. Ion and water transport in airways

15

Ion transport depends on the maintenance of an electrochemical gradient across the membrane. A Na-K-ATPase in the basolateral membrane pumps Na+ out of the cell and K+ into the cell, and as a result the intracellular con-centration of Na+ is low (20mM), and the concentration of K+ is high (150 mM) [29]. With the help of this gradient a Na+/K+/2Cl- cotransporter may accumulate Cl- within the cell against its electrical gradient. Amassed Cl- leaves the cell through chloride channels in the apical membrane. Secretion of chloride is electrically coupled to efflux of K+ through a basal K+ conduc-tance channel [30, 31]. In normal airway epithelium the paracellular water transport may be coupled to the secretion of chloride ions via a calcium-dependent chloride channel (CaCC) and a cAMP-dependent chloride chan-nel (CFTR) [32], which are located in the apical membrane of epithelial cells. It is now widely accepted that the CFTR Cl- channel is the predomi-nant Cl- channel in the apical membrane of epithelial cells and that a genetic defect in the activity of this channel is the underlying cause of cystic fibro-sis. It has been shown that CFTR also functions as a regulator of other ion channels, principally the epithelial Na+ channel (Figure 6) [33, 34].

Figure 6. Multifunctional CFTR [32]

Sodium is absorbed across the epithelium in a two-step process – passive entry into the cell down an electrochemical gradient via a conductive path-way in the apical membrane and extrusion from the cell across the baso-lateral membrane by the activity of Na-K-ATPase pump [35]. Sodium chan-nels are ubiquitous in nature and are classified either as neuronal voltage-gated sodium channels, which are selectively blocked with neurotoxins or as epithelial sodium channels (ENaC), which are selectively blocked by amilo-ride [36]. Both the electrical and the chemical gradients favor sodium entry.

16

In cystic fibrosis airway epithelium there is excessive Na+ (and volume) absorption and an inability to secrete Cl- (and volume) across the epithelium. These abnormalities would predispose airway secretion to dehydration, which results in defective mucociliary clearance and lung defense. Conse-quently, drugs that tend to reverse these abnormalities may have therapeutic potential. Both apical and basolateral cell membranes are normally relatively impermeable for Cl- [31].

However, several substances such as adenosine and epinephrine are able to induce Cl- secretion in CF [28]. Signals that regulate ion transport in many epithelia come from the basolateral surface of the cells. In the airways it seems convenient to have some regulation also on the apical surface to be able to respond to the demands on mucociliary clearance that inhaled parti-cles may cause. Pharmacologic blocking of the epithelial Na+ channel, which is rate-limiting for volume absorption from the airway surface, constitutes a novel therapeutic target. Studies of mucus clearance both in animal models and human subjects demonstrate that blocking of the epithelial Na+ channel is associated with an acceleration of mucus clearance, suggesting that epithe-lial Na+ channel blocking may indeed constitute a rational form of therapy for chronic bronchitis and for cystic fibrosis [37].

Intracellular pH (pHi) is an important intrinsic factor controlling a multitude of cellular functions [38], and it has the potential to act as rapid signal for coordination of intracellular processes. Cellular mechanisms that contribute to regulation of the free intracellular activity of protons include metabolic production of protons and bicarbonate, an intracellular buffer system, and specialized transport mechanisms for these ions, e.g. Na+/H+ and Cl-/HCO3

- exchangers, proton ATPases, and conductive pathways for H+ and HCO3

-. The CF gene defect affects several ion transport mechanisms in the airway epithelial cell membrane, including apical Na+ channels and Cl- channels. CFTR is able to transport both chloride and bicarbonate [39, 40], as evi-denced, for example, by CFTR-dependent alkaline pancreatic fluid secretion [41, 42]. It has thus been postulated that defective CFTR-facilitated bicar-bonate transport in the airways might produce an abnormally acidic ASL, which could promote airway infection in CF by reducing mucociliary clear-ance and antimicrobial action, and increasing the viscosity of pH-sensitive mucous glycoproteins [43, 44]. There is evidence that ASL and/or gland fluid pH might be regulated by cell membrane ion transporters such as CFTR, anion and cation exchangers, and H+ pumps. In primary cell cultures of the airway surface epithelium, Coakley et al. [45] reported a relatively acidic ASL in CF cell cultures, with cAMP-induced alkalization in normal but not CF cells. They found evidence for ouabain-sensitive H+-K+-ATPase activity that could acidify ASL pH; however, its activity was similar in nor-mal and CF cells. Krouse et al. [46] also reported ouabain-sensitive H+-K+-

17

ATPase activity in Calu-3 cells, a model of submucosal gland cells that se-crete bicarbonate in response to forskolin. Fischer et al. [47] reported greater H+ secretion in human tracheal epithelial cells than in Calu-3 cells, and inhi-bition of H+ secretion by mucosal ZnCl2. In contrast, Inglis et al. [48] re-ported that luminal acidification in pig distal airways could be inhibited by bafilomycin A1, an inhibitor of the vacuolar-type H+ ATPase. Thus, al-though the evidence is somewhat conflicting, it appears that multiple H+/HCO3

- transporters, including CFTR, are involved in establishing pH in primary gland fluid secretions and in maintaining/regulating airway surface fluid pH.

Airway epithelial ion transport processes play a major role in the regulation of the volume and composition of the airway surface liquid (ASL) by gener-ating osmotic gradients that provide a driving force for transepithelial fluid movement. CF-associated epithelial ion and fluid transport abnormalities include upregulated amiloride-sensitive Na+ absorption and impaired Cl-

secretion [49, 50] that lead to ASL depletion, which promotes mucus adhe-sion and chronic bacterial colonization with Pseudomonas aeruginosa [44, 51]. The chronic bacterial infection gives rise to an inflammatory reaction in which neutrophils dominate. Gradually, the lung tissue becomes fibrotic and lung function deteriorates. Treatment of the pulmonary disease of CF is symptomatic, and includes physiotherapy to remove the obstructive mucus, aggressive treatment with antibiotics, and, more recently, treatment with antibodies against Pseudomonas aeruginosa [52]. In the end stage of the disease, lung or heart-lung transplantation may be an option.

Of note, fluid transport across the apical membrane of the respiratory epithe-lial cells is not only determined by chloride transport via CFTR but also by chloride transport by other chloride channels, and also by Na+-uptake via the ENaC channel. This has led to two alternative strategies in pharmacological treatment of CF, namely (1) attempts to activate other chloride channels (to provide compensatory chloride efflux), and (2) attempts to inhibit Na+-uptake via ENaC.

1.3.3 Airway surface liquid

The airway surface liquid (ASL) is a thin layer of fluid covering the airways. The ASL possesses a sol layer (periciliary liquid layer (PCL)) that keeps mucus at an optimal distance from the underlying epithelium and through which cilia beat freely, and an overlying mucus layer that entraps particles to be transported by coordinated ciliary movement (Figure 7) [53, 54].

18

Figure 7. Schematic model of the ASL [53]

The ASL secreted by airway epithelium and submucosal glands is the first line of defense of the airway epithelium and ensures the mucociliary trans-port of inhaled particles. ASL contains glycoproteins such as mucins, and other proteins including defensins, lactoferrin, lysozyme, lipids, peptides, ions, and water. The ionic composition of the ASL plays a crucial role in airway host defense by controlling the ciliary activity, mucin release [55], and antimicrobial activity [55] in the airways. Studies of patients with cystic fibrosis have strongly suggested that the volume and/or composition of the fluid that lines the airway surface are important components of lung defense [51, 56, 57, 58, 59].

Abnormal composition and physical properties of the ASL and glandular secretions are proposed to promote chronic bacterial colonization of the air-ways by impairing mucociliary clearance (Figure 8).

Figure 8. Normal and CF ASL [53]

19

Several hypotheses have been proposed to explain how mutations in the CFTR contribute to airway disease in CF. Based on measurements of the elemental composition of ASL obtained from CF patients and healthy sub-jects, together with in vitro and in vivo models using bronchial epithelial cells, one hypothesis proposes that mutations in CFTR cause abnormalities in electrolyte transport across airway epithelia leading to an elevation in the NaCl concentration in the ASL [56, 58, 60, 61]. The "high salt hypothesis" is based on meticulous experiments in which airway epithelial cells from nor-mal and CF individuals were grown on filters. Bacteria were then placed into the fluid covering the apical surfaces of the cultured cells. The remarkable finding was that bacteria flourished in the cultures of CF airway cells, but were killed in the cultures of normal cells. Bacteria placed in the media be-neath normal cells flourished, suggesting that normal cultured airway cells produce apical factors that kill bacteria. A further remarkable discovery was made when the salt content of the apical fluid was manipulated. Merely add-ing pure water to the fluid from CF cells rendered it bactericidal while add-ing NaCl to fluid from normal cells allowed bacteria to grow. Thus it could be concluded that both normal and CF airway epithelial cells release antibac-terial substances into the surface liquid, but in CF cultures the antibiotics are rendered ineffective by a higher salt concentration. Other studies, however, have failed to detect differences in NaCl concentrations in ASL between CF patients and healthy persons [62, 63, 64, 65]. These studies have led to a second hypothesis in which mutations in CFTR promote increased absorp-tion of isotonic ASL leading to dehydration and volume reduction of the ASL and impaired mucociliary clearance, leading to obstruction and infec-tion [66]. The "low volume hypothesis" is not based on CFTR’s function as a channel. Instead, it is based on CFTR’s ability to inhibit the activity of the epithelial sodium channel (ENaC). According to this hypothesis, both nor-mal and CF ASL have plasma-like levels of salt, but in CF, CFTR’s inhibi-tion of ENaC is eliminated, resulting in increased Na+ uptake. Increased Na+ uptake drives increased absorption of Cl- and water. Both hypotheses support the view that the ASL is abnormal in CF and that this abnormality arises as a consequence of the loss of CFTR activity. Both hypotheses have also fo-cused attention on the mechanism by which abnormalities in ASL composi-tion affect the function of airway epithelial cells. Figure 9 [67] summarizes the principal hypothesis-related abnormalities in ASL and glandular physiology in CF to chronic bacterial infection and pro-gressive deterioration in lung function. The "low pH hypothesis" postulates that the ASL is abnormally acidic in CF, inhibiting mucociliary clearance mechanisms [43]. Defective CFTR-dependent bicarbonate transport in CF is proposed to acidify the ASL.

20

Figure 9. Hypotheses linking defective CFTR function to airway disease in CF [67].

The "low oxygenation hypothesis" postulates that the oxygen content of the ASL is low in CF airway epithelial cells and possibly slows oxygen diffu-sion in the ASL, resulting in enhanced Pseudomonas aeruginosa growth and biofilm formation and impaired clearance [68]. The "defective gland func-tion hypothesis" postulates that the primary defect in CF is reduced fluid secretion by airway submucosal glands and possibly altered secretion of mucous glycoproteins [69, 70, 71]. One motivation for this hypothesis is that the much greater expression of CFTR in CF airway glands would impair salt and water secretion, resulting in reduced secretion fluid volume, increased protein concentration, and increased viscosity.

Many studies have concentrated on the composition of the ASL, but it has been difficult to determine the exact composition of the ASL. It might seem odd that there could be a disagreement about something as simple as the composition of airway surface liquid in human airways. However, there is no consensus on this point because the volume of fluid to be sampled is tiny and rapidly altered when disturbed. Airway surface liquid has been estimated to be anywhere from 20-150 μm in depth depending upon species and method. The sol layer is usually about the same depth as the cilia, or approximately 10 μm. If a typical thickness is taken as 30 μm, then a square cm area would contain only 3 microliters, and the entire surface of the conducting airways, an area about 2 m2, would be bathed in 60 milliliters or 4 tablespoons of fluid [72]. Published data on the composition of the ASL are divergent and show a range from very hypotonic to hypertonic. Tracheal and/or bronchial ASL has been collected both from humans and from other species, mainly from mouse [67]. Published data for the ionic composition of human ASL

21

are fairly closely together and range from 82-91 mM for Na+ and 82-108 mM for Cl- [60, 62, 65, 73]. For human xenographs in immunodeficient mice, however, lower values (64 mM Na+, 65 mM Cl-) have been reported [74]. For the mouse the variation is very large: 6 mM Na+ and 1 mM Cl- was reported by Baconnais and coworkers [75], 87 mM Na+ and 57 mM Cl- by Cowley and colleagues [76], and values in the isotonic range (~115-120 mM Cl-) by several groups [77, 78, 79]. For monkey and rabbit, values for Cl- of 112 and 114 mM have been found [80]. High values were found for dogs, where Na+ concentrations ranged from 153-173 mM and the Cl- concentra-tion was 134 mM [81], and in ferret where Na+ was 167 mM and Cl- 121 mM [82]. Clearly hypotonic values were found for the rat, where Na+ was 41 mM and Cl- 45 mM [83]. A number of different techniques for sampling and analysis have been used, and it would appear that, apart from possible spe-cies differences, one source of variation between the results is the sampling technique. In addition to the studies on mice mentioned above, Zahm and colleagues [84] give data in mmol/kg dry weight, which is not comparable with the absolute data in mM given in the other studies. Furthermore, studies have been performed on cell cultures of airway epithelial cells, but also these studies did not result in agreement being obtained. According to Matsui and coworkers [64], values for Na and Cl were nearly isotonic, whereas Zabner and colleagues [58] found values for Na and Cl around 50 mM, and McCray and colleagues [85] found a value for Cl of 18 mM. Widdicombe and co-workers [86] used X-ray microanalysis to measure the composition of ASL on top of cultured airway epithelial cells, but this study was purely qualita-tive.

There is still little information about ASL composition in lower airways, where airway disease in CF has been postulated to begin. Also, there are few data about the possibility that ASL ionic composition might be altered in response to various agonists and factors released by inflammatory cells/bacteria or other components of CF airway fluid. CF patients are born with apparently normal lungs, but this is followed by the acquisition of chronic, unrelenting bacterial infections of the airways (bronchi) in the first few years of life. Thus, in the simplest view, CF lung disease reflects the failure of the innate defense mechanisms of the lung against inhaled bacterial organisms [54]. Viral infections occur in all infants with cystic fibrosis and later in life, most CF patients become infected with Pseudomonas aerugi-nosa, which once established persists usually for life. More than 80% of adult CF patients are chronically infected [87] and develop lung infection and airway inflammation that lead to respiratory failure, which is the most common cause of morbidity and mortality [88]. Clearance of mucus from the diseased airway is of primary importance because it minimizes the bacterial contamination, reduces infections and improves the quality of life of the patients. Effective clearance of mucus is a critical innate airway defense

22

mechanism and under appropriate conditions can be stimulated to enhance clearance of inhaled pathogens. It has become clear that extracellular nucleo-tides (ATP and UTP) and nucleosides (adenosine) are important regulators of mucus clearance in the airways as a result of their ability to stimulate fluid secretion, mucus hydration, and ciliary beat frequency (CBF) [89].

The mist tent therapy was a recommended treatment for CF patients in the 1960s and 1970s and aimed at increasing the hydration and to change the properties of mucus in order to make it easier for the patients to remove it from airways by coughing. It was assumed that water particles present in mist were deposited in the lower respiratory tract of the patients. The effi-cacy of the mist tent therapy was supported by clinical trials and by ventila-tory function studies in patients with CF [90, 91]. Other studies re-examined the clinical effects of mist tent therapy in CF patients using clinical observa-tions, and pulmonary function tests, and concluded that the nocturnal mist tent therapy had no beneficial effect for CF patients. The severity of the pulmonary disease and the type of nebulizer used had no apparent effect on the result [92, 93, 94]. The efficiency of the method was doubted and its use was largely discontinued. However, at a few CF centers mist tent therapy was regularly recommended particularly for small children with CF [95].

Potential ways to increase antimicrobial activity in the CF lung include in-creasing the secretion of antibacterial proteins and promoting the secretion of antibacterial factors that may not be present under basal conditions. Be-cause bacterial killing activity is dependent on the protein concentration of antibacterial factors [96], this approach could be of benefit to patients with CF, irrespective of the issues about differences in the salt content of ASL. Although there are a large number of identified antibacterial proteins and peptides in ASL, and in other organs as well, a complete elucidation of these factors has not been achieved.

Defensins, one of the most intensively studied classes of antimicrobial pep-tides, have been identified in a wide variety of animals, including birds, ro-dents, and humans [97]. Defensins are small cationic peptides containing 29-47 arginine-rich amino acids with three disulfide bonds, which can be di-vided into �- and �- defensin subfamilies in human subjects [97]. The main function of defensins is believed to be to kill bacteria and fungi either on the surfaces of the epithelial cells or within phagolysosomes of phagocytes.

23

1.4 Survival and pharmacological treatment of CF

In the 1950s when CF was first described, the children were dying of pan-creatic insufficiency and failure to thrive. As the pancreatic enzyme re-placement therapy solved the problem with the pancreas and children started living longer, the full picture of the respiratory problems became evident [98].

Nowadays the mortality in CF is associated with chronic obstructive pulmo-nary disease and respiratory failure being the primary cause of death among 90% of CF patients [99]. Double lung or heart-lung transplantation evolved to a standard treatment modality for patients with cystic fibrosis suffering from end-stage lung disease. Overall 3-year survival is about 60% in lung-transplant patients with cystic fibrosis [100] although patients infected with Burkholderia cepacia complex were reported to have a poor prognosis [101]. A time point at which a decision about transplantation needs to be made that most centers would agree upon is a forced expiratory volume in 1s (FEV1) < 30% of the expected value in a patient receiving maximum medical treat-ment [102]. In another transplantation modality, living lobar lung transplan-tation, the time point of the surgery is also very critical but the decision is based on defining a point when a potential recipient is too ill to justify plac-ing two healthy donors at risk of donor lobectomy [103].

Progress in the supportive care of CF patients had more than doubled the median age of survival in the past decades from 14 years in 1969 to over 40 in 2008 [104]. Country-specific differences in survival exist, which might be associated with availability and access to specialized care, treatment strate-gies and socioeconomic status [105, 106].

The most consistent aspect of therapy for cystic fibrosis is limiting and treat-ing the lung damage caused by the thick mucus and the infections with the goal of maintaining the quality of life. Antibiotics have clinical benefits in patients with cystic fibrosis [107]. Intravenous, inhaled, and oral antibiotics are used to treat chronic and acute infections. Mechanical devices and in-haled medications are used to dilute and clear the abnormally viscous mucus. Many mechanisms have been proposed, including the effects on neutrophil function [108], IL-8 production [109], sputum rheology [110], goblet cell hypersecretion [111], the alginate biofilm produced by Pseudomonas aerugi-nosa [112], and direct antipseudomonal activity [113]. However, the precise mechanisms are still uncertain [114]. Other aspects of CF therapy involve treatment of diabetes with insulin, pancreatic disease with enzyme replace-ment, and infertility with advanced reproductive techniques. Since cystic fibrosis is a monogenic disease, gene therapy may become a possible treat-

24

ment. It is conceivable that drugs used by some of the patients, especially CF and allergy patients, could affect the salt content of the nasal fluid. However, the effect of drugs on the ion content of the nasal fluid has not been investi-gated, and it is therefore impossible to attribute particular changes to drugs used. This will require a separate, systematic investigation.

1.5 Animal models

1.5.1 Mouse models of CF

The cloning of the CFTR gene made it possible to generate mice, in which gene function was deleted either by homologous recombination, or by intro-duction of specific mutations in the murine CFTR gene [115, 116, 117]. This strategy was the first attempt to generate an animal model that reproduces the basic ion transport defects of human CF airways, such as deficient cAMP-dependent Cl- secretion and increased Na+ absorption, to study their role in the in vivo pathogenesis of chronic airway disease. In several studies it was shown that CF mice exhibit characteristic defects in intestinal ion transport producing a severe CF-like gastrointestinal phenotype similar to meconium ileus in CF infants [115, 116]. Surprisingly, even when CF mice were rescued from neonatal mortality due to intestinal blockage by treatment with an osmotic laxative, lack of functional CFTR did not cause CF-like airway disease [117]. In contrast to CF patients, in whom chronic obstructive lung disease remains the major cause of morbidity and mortality, CF mice do not exhibit airway mucus plugging, goblet cell metaplasia or chronic in-flammation and bacterial infection of the lower airways. Subsequently, de-tailed functional analyses of epithelial ion transport in different regions of the airways demonstrated that CF mice exhibit Na+ hyperabsorption and deficient cAMP-dependant Cl- secretion, and an associated phenotype of ASL depletion and goblet cell metaplasia in the nasal epithelium [79, 117, 118]. Some studies demonstrated that a deficient CFTR-mediated Cl- secre-tion is compensated by CFTR-independent Ca2+-activated Cl- channels (CaCC) in the lower airways of CF mice [119, 120]. More recently, it be-came possible to measure mucociliary transport in small animals by a tech-nique that determines the clearance of a fluorescent dye deposited into the lower airways with microdialysis [121]. Consistent with normal ion transport properties and lack of pathology in lower airways, it was demonstrated that in vivo mucus clearance rates are normal in CFTR-deficient mice [121]. In a normal airway, an appropriate balance between Na+ absorption (mediated by ENaC in the apical membrane) and anion secretion (mediated by apical CFTR and alternative anion channels) determines the volume of fluid on the

25

airway surface. This transport balance regulates the height of the periciliary liquid (PCL) in which the cilia beat to permit effective mucus clearance (Figure 10) [122]. Mall and coworkers created a �-ENaC overexpressing mouse to increase sodium absorption, this reduced periciliary liquid depth, impeding ciliary movement and mucus clearance. Normally, CFTR regulates ENaC activity, a process that does not occur in cystic fibrosis because CFTR is absent or defective. In �-ENaC transgenic mice, and perhaps in cystic fibrosis airways, failure of mucus clearance promotes accumulation of air-way secretory products, including glycosaminoglycans (GAGs) and a pool of chemokines (e.g., interleukin-8), neutrophil proteases and growth factors that promote airway inflammation and mucus cell hyperplasia [123].

Figure 10. Modeling cystic fibrosis [122]

1.5.2 Pig models of CF

Despite nearly 20 years of research and more than 5000 PubMed–cited pa-pers [124], knowledge of the underlying CFTR defect has not been effec-tively translated into improved clinical therapies. One reason has been poor understanding of how the CFTR defect produces CF lung disease. Trans-genic mouse models of CF manifest subtle to no CF lung disease [125]. The long-awaited CF pig holds great promise as a model of CF (Figure 11) [126].

26

Recent success in applying somatic cell nuclear transfer technology to gen-erate heterozygous CFTR null and �F508-CFTR knock-in pigs [126] assures that homozygous CF pigs are on their way. Pigs have already been used ex-tensively to study a wide variety of lung functions and pathologies, such as surfactant biology, asthma, acute lung injury, and gene transfer [124]. Of relevance to CF, information already exists in pigs on airway salt transport, nasal potential differences, submucosal gland function, and mucociliary clearance mechanisms [127, 128, 129, 130]. Will CF pigs provide new in-sight into the pathogenesis of CF lung disease? Research into the link be-tween CFTR dysfunction and CF lung disease has had adverse history.

Figure 11. Photo of the first CFTR +/ - piglet taken at 1 day of age [126]

The two major expected uses of CF pigs include research on the pathogene-sis of CF lung disease and the testing of CF therapies. Hopefully, pigs will develop the major manifestations of human lung disease. Rogers and co-workers [131] showed many similarities between human and pig lungs: simi-lar anatomy, histology, electrolyte transport, submucosal gland function, and immune and inflammatory responses. If CF pigs do not develop lung disease spontaneously, then exposure to Pseudomonas aeruginosa might initiate the process.

However, a study reported by Ostedgaard and coworkers indicated partial cellular processing and significant plasma membrane targeting of porcine �F508-CFTR [132], raising concerns about the phenotypic consequences of the �F508 pig model. Another study by Liu and coauthors confirmed the species difference in cellular processing of �F508-CFTR reported by Ost-edgaard et al., but their results indicated rather mild processing defect of �F508-pCFTR and suggested that the transgenic �F508-CFTR pig model may not develop the desired CF phenotypes [133]. It was estimated that plasma membrane targeting of 10% �F508-CFTR protein could avoid the CF phenotype in human subjects [134].

27

Aims

The aims of this study were to:

• To determine the composition of the ASL in mice, the apical fluid of normal (16HBE14o-) and CF (CFBE41o-) bronchial epithelial cell lines, and pigs, using two different methods: X-ray microanalysis of frozen-hydrated samples and X-ray microanalysis of dextran beads equilibrated with the ASL or apical fluid, respectively.

• To find similarities and differences in structure and physiology of mouse, pig, and human airways.

• To carry out an ultrastructural study of the airway epithelium at various osmotic values.

• To investigate CFTR-mediated HCO3- conductance and the role of HCO3

- in regulating ASL pH.

• To study the effects of agonists and antagonists of ion transport on changes in ion concentrations in mouse nasal fluid.

• To investigate the influence of repeated anaesthesia on the ion concentra-tion in the mouse nasal fluid.

• To study the effects of antagonists on ion concentrations in the apical fluid of normal and CF bronchial epithelial cell lines.

• To determine the elemental composition of nasal fluid in healthy subjects, CF patients, CF heterozygotes, patients with rhinitis, or with primary ciliary dyskinesia (PCD).

• To investigate whether CF patients become colonized with Pseudomo-nads during mist tent treatment, and what effect a night spent in a mist tent has on the ion content of nasal ASL in CF patients and healthy con-trols.

29

Materials and methods

3.1 Animals and anesthesia All animals were housed in an environmentally controlled vivarium and allowed free access to a standard pellet diet and water. All experimental procedures were conducted in accordance with guidelines of the Swedish National Board for Laboratory Animals and had previously been approved by the Swedish Laboratory Animal Ethical Committee in Uppsala.

3.1.1 Mice (Papers I, II)

For the physiological experiments, ~50 NMRI mice (B&K, Sollentuna, Sweden) of both sexes, ~1 month old, were used. In addition, five NMRI mice of ~15-18 months were used as controls for comparison with CFTR(-/-) mice of similar age. The CFTRtm1Cam(-/-) mice (female, ages 15-18 months) were raised in Hannover (Germany), transported by air to Uppsala (Sweden), and kept for 1 week before the start of the experiment on a crude fiber-deficient diet with addition of a laxating salt solution (Oralav; B. Braun, Melsungen, Germany). These mice were anesthetized by spontaneous inha-lation of isoflurane (Forene; Abbott Scandinavia, Kista, Sweden). The inha-lation gas was administered continuously through a breathing mask (Simtec Engineering, Switzerland) and contained a mixture of 40% oxygen, 60% nitrogen, and 2.2% isoflurane. Before harvesting of the trachea, the mice were terminated by spinal translocation. The NMRI mice were anesthetized with pentobarbital. In some experiments, the anesthetized animals were in-jected intraperitoneally with pilocarpine (50 mg/kg body weight) or isopro-terinol (10 mg/kg body weight), and the trachea was removed 15 minutes after injection. In experiments where nasal fluid was collected, the animals received an intraperitoneal injection with pilocarpine, isoproterenol (dose as above), or phenylephrine (10 mg/kg body weight). C57BL/6J mice (F12 backcross) heterozygous for the �F508 mutation (CftrTm1EUR) [135] were bred as heterozygotes and the offspring was genotyped by PCR analysis. Homozygous F508del mice (CftrTm1EUR) and wild-type control mice were

30

temporally anesthetized with a solution of 2,2,2 tribromoethanol in 2-methylbutan-2-ol (Avertin). The mice were anesthetized on two subsequent days and nasal fluid was collected after anesthesia. After the first sampling, the animals were allowed to recover and it was noted that their behavior was normal, and no negative symptoms of the anesthesia were observed. After the last sampling, the animals were sacrificed, and nasal fluid was collected post-mortem.

3.1.2 Pigs (Paper IV)

Ten healthy pigs of mixed breed (Hamshire, Yokeshire, and Swedish land-race) with a body weight of 20-30 kg were used in this study. Before trans-port to the laboratory, premedication with 40 mg azaperon (Stresnil; Janssen Pharmaceutical, Beerse, Belgium) was given by intramuscular injection. Anesthesia was induced with 0.5 mg atropine (Atropine; NM Pharma AB, Stockholm, Sweden) in a mixture of 100 mg tiletamin and 100 mg zolaze-pam (Zoletile forte vet; Virbac Laboratories; Carros, France) diluted in 5 ml medetomidin 1 mg ml-1 (Domitor, Orion Corporation, Farmos, Finland); 1 ml per 20 kg body weight intramuscularly. The animals were placed in su-pine position on a heating pad and intubated with a cuffed endotracheal tube with 6.0 mm inner diameter and ventilated mechanically in a volume- or pressure-controlled mode with 3 cm H2O of positive end expiratory pressure (PEEP) (Siemens Servo 900C; Siemens-Elema AB, Solna, Sweden). The I:E ratio was 1:2. The tidal volume was 14 ml/kg plus compensation for dead space volume and the frequency was adjusted to an end tidal CO2 of 5.5 kPa. A bolus injection of 0.2 mg fentanyl (Fentanyl; Antigen Pharmaceuticals, Roscrea, Ireland) was given intravenously after the intubation. Anesthesia was maintained by infusion of 5 ml kg/h of 4 g ketamin (Ketamin Veteri-naria, Zürich, Switzerland), 1 mg fentanyl in 1,000 ml Rehydrex with glu-cose (Pharmacia and Upjohn, Stockholm, Sweden). The animals had no signs of respiratory infections or inflammation on visual inspection. The animals were also used for another study and sacrificed at the end of the experiment.

3.2 Cell lines (Paper III)

3.2.1 Airway epithelial cell lines

16 HBE14o- (16HBE) and CFBE41o- (CFBE) (homozygous for the �F508 mutation) cells were kind gifts of Dr D. C. Gruenert, San Francisco [136].

31

The cells were grown to confluence in Eagle’s Minimal Essential Medium (EMEM, SVA; Uppsala, Sweden) in tissue culture flasks. The confluent cells were trypsinized, split, and at least 1.6x105 cells were seeded on special inserts (Costar Transwell, Corning, NY, USA; 6.5 mm diameter, 0.4 �m pore size; 0.33 cm2 growth area) on a tissue culture 24-well plate (Sarstedt, Newton, NC, USA; 1.2 cm diameter). The inserts had been coated with 100 �l of a solution containing 0.01% collagen type I from calf skin (Sigma, St. Louis, MO, USA). The collagen was allowed to bind overnight at 37°C. Excess fluid was removed from the coated surface, which was then dried for 30-60 minutes. The coated surface was sterilized by ultraviolet light for 2-3 hours.

3.2.2 Liquid-liquid and air-liquid interface cell cultures

Some membranes with cells were chosen for analysis as soon as confluence was reached in liquid-liquid interface cultures (i.e., with growth medium both apically and basally). Other cultures were allowed to differentiate in air-liquid interface on the membranes for 3-4 weeks. For this, the medium was removed from the apical surface, and 700 �l of medium consisting of 48.5% Dulbecco’s Modified Eagle’s Medium (DMEM) with 1000 mg/L glucose (Gibco, Paisley, Scotland), 48.5% Ham’s F12 medium with L-glutamine (Gibco), 2% Ultroser G (Gibco), and 1% of a penicil-lin/streptomycin solution (Sigma, St. Louis, MO, USA) to give a final con-centration of 60 mg/ml penicillin and 50 mg/ml streptomycin was added basally. No medium was added to the apical side of the culture but the cul-ture was rinsed with 100 �l sterile isotonic phosphate-buffered saline (PBS) (140 mM NaCl, 5 mM Na2HPO4, 4.2 mM KH2PO4) every other day.

3.2.3 Transepithelial electrical resistance (TEER) measurements

The transepithelial resistance of both liquid-liquid and air-liquid cultures was measured with an AVOM epithelial voltmeter in an ENDOHM-6 chamber (World Precision Instruments, Sarasota, FL, USA). In all TEER experiments on cell cultures EMEM was used. The temperature was kept at 37°C in the ENDOHM chamber and the cell cultures were left at this temperature during all preparations and experimental steps.

3.2.4 pH measurements

The pH measurements on the fluid covering the apical surface of airway epithelial cells in culture were done with a solid state PHR-46 Micro combi-

32

nation pH electrode (Lazar Research Laboratories, Inc., Los Angeles, CA, USA). For this series of experiments both cell lines under liquid-liquid con-ditions were used after 3-4 days of seeding. Medium from the apical surface was exchanged for a simple unbuffered solution, which did not contain HCO3

- ions [137]. Liquid-liquid interface cultures were permitted to equili-brate with 10-50 �l of simple unbuffered solution [138], with or without added agonists or antagonists, or CFTRinh-172 [139], a specific blocker of CFTR (a kind gift of Prof. A. Verkman, Los Angeles, CA, USA) for 3 hours maximum, because approximately after 3.5 hours the TEER in confluent cell cultures decreased to below 100 �m2. The pH measurements were done every hour until the tight junctions broke down in the unbuffered solution.

3.3 Frozen-hydrated samples (Papers I, III, IV)

3.3.1 Pig tracheae and bronchi

The pieces of tracheae or bronchi were removed and dissected in a specially designed chamber, in which the humidity could be kept constant at close to 100%. The chamber consisted of a Perspex box (manufactured by the work-shop, Biomedical Center, Uppsala University, Sweden), with a retractable plastic sheath on the side, which could be opened for handling of the speci-men. A water reservoir at 37°C was placed at the bottom of the chamber, and the specimen was placed on a perforated shelf. Humidity was monitored with a hygrometer. Tissue pieces were frozen in liquid propane cooled by liquid nitrogen. The pieces were stored in liquid nitrogen until analysis.

3.3.2 X-ray microanalysis

For analysis, the tissue pieces were placed with the mucosal side up onto a specially designed holder and transferred to a Philips 525 scanning electron microscope (Philips Electron Optics, Eindhoven, The Netherlands), equipped with a Bio-Rad (Hemel Hempstead, UK) Polaron 7400E cold stage. The samples were coated with a thin carbon layer in the cold stage, at a temperature of -190°C, and kept at this temperature throughout analysis. The lowest possible temperature was selected to minimize the risk of subli-mation. After preliminary experiments, an accelerating voltage of 9 or 10 kV was chosen to minimize overpenetration of the beam. The samples were analyzed by a LINK (Oxford Instruments, Oxford, UK) AN 10000 energy-dispersive spectrometer system. Analysis was performed for 500 s with a beam size of 200 nm, a beam emission current of ~15 �A, a count rate of

33

230-235 cps, and a detector dead time of 5%. Typically, 8-10 analyses were performed per sample. For quantitative analysis, the data were compared with the results obtained on a standard.

3.3.3 Standards

The standard consisted of a solution of 150 mM NaCl containing 5% albu-min. The solution was smeared over an aluminium planchet to achieve a relatively thin fluid layer, to mimic the situation for the ASL. In this way, mass loss due to irradiation by the electron beam can be assumed to be com-parable between specimens and standard. The standard was shock-frozen, transferred to the cold stage of the electron microscope, and analyzed under the same instrumental conditions as the specimen. Quantitative analysis was performed using the ratio of characteristic to continuum intensity in the specimen and by comparing this ratio with that obtained by analysis of the standard salt solution [140]. Concentrations of elements other than Na or Cl were determined from binary standards containing these elements and Na or Cl, using the ratio method [140].

3.3.4 Mouse tracheae

The tracheae were removed and immediately frozen in liquid propane cooled by liquid nitrogen to avoid compression during dissection. The tracheae were then dissected into tracheal rings under liquid nitrogen. The pieces were stored in liquid nitrogen until analysis. The analysis was performed exactly in the same way as for pig tracheae (see above).

3.3.5 Liquid-liquid interface cell cultures

Liquid-liquid cultures of confluent 16HBE and CFBE cells were allowed to equilibrate with isotonic PBS for 24 hours prior to the experiment. 7 �l of PBS was added to the apical surface of the cells, which gave a layer with a depth of about 200 �m. After 24 hours the membrane was frozen in liquid propane cooled by liquid nitrogen to -190°C. The membrane was placed onto a specially designed holder and transferred to a Philips 525 scanning electron microscope equipped with a Bio-Rad Polaron 7400E cold stage and analyzed under the same conditions as pig or mouse tracheae (see above).

34

3.4 Ion-exchange beads (Papers I, II, III, IV)

3.4.1 Pig tracheae

Sephadex G-25 beads (diameter 20-40 �m; Pharmacia, Uppsala, Sweden) were spread evenly on the surface of the dissected pieces of the pig trachea and left during 30 minutes in the humidity chamber described above. Sephadex is the trade name for a cross-linked dextran gel with ion-exchange capacity. Saturation of the beads with a salt solution was obtained after 5 minutes [141]. After absorption of the ASL, the beads were recovered by flushing with a hydrophobic volatile oil (DC 200, 0.65 cSt; Dow Corning, Midland, MI) and collecting the beads in a watch-glass [142, 143]. Under a preparation microscope, all adhering fluid and debris was removed from the beads, and single beads were transferred onto specimen grids, which had been submerged into the oil. The specimen grids used were nylon grids (Agar Scientific, Stansted, UK) covered with a thin Formvar (Merck, Darm-stadt, Germany) film. The grid with beads was slowly lifted out of the oil bath and mounted onto an aluminum holder covered with round carbon ad-hesive tape and left at room temperature for evaporation of the oil. Grids with Sephadex beads were carbon coated before analysis.


X-ray microanalysis of the beads was performed with the instrumentation described above, at 20 kV for 100 s with a beam size of 100 nm. Typically 10-12 beads were analyzed from each sample. For quantitative analysis, the data were compared with the results from X-ray microanalysis obtained on beads soaked in salt solutions of different concentrations (50-250 mM) [141], and with beads soaked in serum from the same pigs that was analyzed chemically.

3.4.3 Mouse tracheae

Dextran (Sephadex G-25) beads (diameter 20-40 �m) were equilibrated for 10 minutes with the ASL in the mouse trachea in the following way. A small amount of beads was placed in the opening at the base of a Microlance 3 needle (0.8 x 40 mm; Beston Dickinson, Dublin, Ireland). The needle was connected to a syringe previously filled with air, and the beads were sprayed evenly over the tracheal surface by pressure on the syringe [144]. During the experiment the pieces of dissected mouse trachea were located in the humid-

35

ity box (described above) to avoid evaporation of the ASL. After absorption of the ASL, the beads were prepared for analysis in the same way as de-scribed above.

3.4.4 Nasal fluid collection in mice

Nasal fluid from the mouse was collected in Sephadex beads as follows. The Sephadex G-25 beads were applied to double-sided tape (3M, Minneapolis, MN) attached to a filter paper that was cut in triangular strips (base about 0.2-0.3 mm, length about 0.8 mm) with a narrow tip. The filter paper with the beads was inserted into one or both nostrils of the mouse. Beads were facing the nasal septum and kept there for 10 minutes. Then, the filter paper with saturated beads was removed from the nostril and carefully washed in silicon oil to ascertain that no fluid was left on the outside of the beads. In some cases, the beads were separated and each bead was individually moved to a nylon electron microscopy grid until the grid contained 10-15 beads. The grid was then carefully removed from the oil, dried by evaporation of the oil at room temperature, and mounted on a specimen holder. In other cases, when it was difficult to separate the beads from the tape, the filter paper with the beads was carefully washed in the silicon oil and mounted on a specimen holder with the beads facing upwards. Grids with Sephadex beads or filter paper with beads on it were carbon coated before analysis and prepared for the analysis in the same way as described above.

3.4.5 Liquid-liquid interface cultures

Liquid-liquid interface cultures were allowed to equilibrate with 10 �l of isotonic PBS (with or without added agonists/antagonists) for 24 hours. Sephadex G-25 (diameter 20-40 �m) beads were equilibrated for 10 minutes with the apical fluid in the following way. A small amount of beads was placed in to the opening at the base of a Microlance 3 needle (0.8 x 40 mm). The needle was connected to a syringe previously filled with air, and beads were spread evenly over the airway surface by pressure on the syringe [144]. After absorption of the ASL, the beads were recovered by flushing in the silicon oil. All preparations for the X-ray microanalysis were done exactly in the same way as described above. For quantitative analysis, the data were compared to the results obtained on beads soaked in salt solutions of differ-ent concentrations (50 mM – 250 mM).

36

3.5 Plasma and serum samples (Papers I, IV)

3.5.1 Pig plasma and serum

Venous and arterial blood was collected from the pigs, and plasma and se-rum was collected either after centrifugation or after clotting for 24 hours at 4°C. Analysis of Na+, Mg2+, phosphate, Cl-, K+, and Ca2+ in the plasma or serum was performed using a Konelab (Espoo, Finland) analyzer in the Laboratory of Clinical Chemistry, Swedish Agricultural University, Uppsala.

3.5.2 Mouse plasma and serum

Venous and arterial blood was collected from the mice, and both plasma and serum (after clotting for 24 hours at 4°C) were collected. Sephadex beads were equilibrated with serum or plasma for 10 minutes. After that, the beads were recovered by being flushed with silicon oil and transferred to grids, carbon coated and prepared for X-ray microanalysis as described above.

3.6 Morphological studies (Papers I, IV) For morphological studies, tissue was removed from the anesthetized ani-mals (pig, mouse) and immediately fixed in 2.5% glutaraldehyde in water or different concentrations of sodium cacodylate buffer (0.025, 0.05, 0.1, or 0.15 mM). The tissues were kept in fixative for 24 hours at 4°C and then postfixed with osmium tetroxide, dehydrated in a graded ethanol series, and embedded in epoxy resin. Ultrathin sections were cut for electron micros-copy, contrasted with uranyl acetate and lead citrate, and viewed at 75 kV in a Hitachi 7100 transmission electron microscope. Pig tissue was cut perpen-dicularly to the epithelium (sections 2 �m thick), and stained with toluidine blue for light microscopy. The height of epithelial cells (50 cells for each fixative concentration) was measured by a semi-automatic image analysis system (VIDS; Synoptics, Cambridge, UK). Pieces of pig trachea exposed for 30 minutes to Sephadex beads as described above were fixed in 2.5% glutaraldehyde in 0.1 mM sodium cacodylate overnight. Tissue was post-fixed and embedded as described above. Thin sections (2 �m) were cut and stained with toluidine blue for light microscopy.

37

3.7 Humans (Papers V, VI)

3.7.1. Study design of paper V

Nasal fluid was collected from several different groups of subjects. The con-trol group consisted of healthy volunteers (non-smokers; 5 males and 14 females, age ranging from 7 to 54 years, mean age 32 ± 3 years). In addition, CF patients (6 males and 11 females, age ranging from 12 to 43 years, mean age 25 ± 2 years), CF heterozygotes (12 mothers of CF patients, age ranging from 28 to 53 years, mean age 43 ± 2 years), patients with primary ciliary dyskinesia (PCD) (7 males and 3 females, age raging from 6 to 40 years, mean age 24 ± 5 years), and allergy/rhinitis patients (15 females and 13 males, age raging from 4 to 55 years, mean age 24 ± 3 years) were included in the study.

All CF and PCD patients were treated by the Cystic Fibrosis Center, Uppsala University Hospital, and the CF heterozygotes were mothers of the CF pa-tients. The CF patients were classified according to the severity of their dis-eases (taking into account lung function and the need for intravenous antibi-otics) as “mild”, “medium”, and “severe”. Data for forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1) for the CF patients were recorded on or close to the date of sampling of the nasal fluid and compared with appropriate reference values and expressed as percentage of the refer-ence value [145]. CF patients classified as having “mild” disease had FVC and FEV1 above 75% and a number of acute exacerbations requiring intra-venous antibiotics over the last year was between zero and two, those classi-fied as “moderate” disease had FVC and FEV1 between 50 and 75% and number of acute exacerbations requiring intravenous antibiotics over the last year was between three and five, and those classified as “severe” disease had FVC and FEV1 less than 50% and the number of acute exacerbations requir-ing intravenous antibiotics over the last year was above six.

All but one of the rhinitis patients were children or teenagers on regular con-trol at the Allergy Clinic, Children’s Hospital, Uppsala University Hospital, and one of them was an adult staff member at the Clinic. All subjects in the group were allergic to inhalants, the vast majority against several important allergens in the area as judged by positive skin tests, IgE-antibody tests and/or clinical history. All subjects had a perennial allergy to fur-coated animals or seasonal allergy with Birch-Timothy pollen sensitivity, most of them both. Corticosteroids had been prescribed in most cases, both for asthma and rhinitis. Ongoing treatment with antibiotics or corticosteroids was not interrupted for the collection of nasal fluid. Sampling was performed

38

during or immediately after the pollen season in May-June, that is, during relevant exposure for the vast majority of the patients.

The study protocol was approved by the Ethics Board of Uppsala University and all subjects and/or their parents gave informed consent.

3.7.2 Study design of paper VI

Another study with humans was aimed at investigating whether there was a relation between mist tent treatment and time point of chronic coloniza-tion/infection with pseudomonads. All patients, born between 1980 and 1998, and monthly attending Stockholm CF Center were included in the study (n=39). Their records were evaluated retrospectively including annual check ups with specific questions about mist tent therapy. Furthermore, home visits were made to check hygienic and technical issues. Bacteriologi-cal data were obtained through patient records and from the bacteriological data base of the hospital (1989 and onwards).

In the study on nasal fluid, 5 CF-patients and 4 controls were included. The group of CF patients consisted of patients from the Stockholm Cystic Fibro-sis Center, Karolinska University Hospital Huddinge, Sweden (3 females and 2 males, age ranging from 16 to 23, mean age 18.6years). The control group consisted of healthy volunteers (nonsmokers; 3 females and 1 males, age ranging from 24 to 59, mean age 41 years). All patients were classified according to the severity of their disease (lung function, number of treat-ments with intravenous antibiotics). The CF patients and healthy controls spent a night (8 hours) in the mist tent and samples of the nasal ASL were taken before the experiment, after the period in the mist tent, and then at each hour during 4 hours after the persons had left the tent. Prior to this ex-periment, samples were taken from 4 CF patients before they went into the mist tent, and then at each hour during their stay in the mist tent. Also sam-ples were taken from the healthy subjects immediately after they left the mist tent and then every 2 hours during 8 hours, and then every 24 hours during two days.

The mist tent was designed and manufactured by Pharmalink (Stockholm, Sweden) and consisted of a polyeten plastic bag, which covered the head and shoulders of the patient and was supplied with a mist of distilled water. In this tent humidity could be kept constant at close to 100%.

39

3.7.3 Defenition of chronic Pseudomonas aeruginosa colonization/infection

Chronic colonization/infection was defined as six consecutely positive spu-tum cultures during 6-7 months and/or increased antibody levels (Pseudo-monas aeuruginosa Exotoxin A) Exotoxin A ELISA (ExoA) [146].

3.7.4 Nasal fluid collection from humans

The subjects were asked to close their nose with a nose clip for 10 minutes, and subsequently, nasal fluid was collected by inserting ion exchange beads (Sephadex G-25; 20-40 �m diameter) in the vestibule of the nose, mounted on 3M double-sided tape on filter paper (2 x 2 mm2; Whatman, Springfield Mill, UK). During the insertion of the filter paper with the beads, subjects were asked to either hold their breath or to breath through their mouth to prevent the warm expired air from causing evaporation of the nasal fluid. One filter paper with beads then was inserted into each nostril of the subject. The nostrils were kept closed for another 10 minutes. Normally this is suffi-cient to saturate the filter paper and the beads with nasal fluid. Beads were saturated after 5 minutes [141], but we used 10 minutes to make sure that we obtained sufficient fluid. In some cystic fibrosis patients, the nostrils were kept closed for a longer time (15-20 minutes), because of their reduced se-cretion. In some of the allergy/rhinitis patients also this longer time was nec-essary, if use of a nasal decongestant had dried out the mucous membranes of the nose. As shown elsewhere [141], the prolonged exposure time of the beads to fluids does not affect the results of the quantitative elemental analy-sis. At the end of the period, the filter papers were removed, while the sub-jects held their breath, and the filter papers were transferred and carefully washed in hydrophobic volatile silicon oil. The beads were cleaned from adhering fluid and mucus by moving them with a thin needle in the silicon oil. With a forceps, single beads were transferred onto nylon specimen grids. The grids with the beads were left to air dry at room temperature overnight and were mounted onto aluminum specimen holders covered with carbon adhesive tape. The specimens were carbon coated prior to analysis to prevent charging in the electron microscope.


The Sephadex beads were analyzed in a Philips 525 scanning electron mi-croscope (SEM) (Philips Electron Optics, Eindhoven, The Netherlands) with LINK AN 10000 energy-dispersive detector (Oxford Instruments, Oxford,

40

UK) at 20 kV with a 100 nm beam for 100 s. Twelve beads were analyzed per sample, one spectrum from each bead.

3.7.6 Standard calibration curve

Standard curves were made as described previously [141, 144] by soaking Sephadex beads in sodium chloride (NaCl) or potassium chloride (KCl) so-lutions of different concentrations (25-250 mM) dissolved in distilled water. Beads were soaked in solutions for 10 minutes and then processed as de-scribed earlier. The accuracy of the method and standard curves were veri-fied by testing the results obtained on beads soaked in human blood serum, where they returned physiological values: [Na]: 130 ± 22 mM, [Cl]: 105 ± 14 mM, [K]: 8 ± 1 mM.

3.8 Statistical analysis

Data are presented as mean values and standard error (SE) of the mean. To compare data from more than two groups, one-way analysis of variance (ANOVA) followed by Dunnett’s Multiple Comparison Test was used. To compare data from two groups, the statistical analysis was performed using an unpaired Student’s t-test.

41

Results

4.1 Paper I

X-ray microanalysis of the ASL layer in frozen-hydrated mouse trachea (Pa-per I, Figure 1) showed a Na concentration of ~80 mM and a Cl concentra-tion of ~50 mM. The concentration of K was considerably higher than ex-pected for an extracellular fluid. The fluid also contained large amounts of P and S. There was no significant difference in elemental composition between the young (1 month) control animals and the old (15-18 months) control animals. The ASL in CF mice had significantly higher concentrations of Na and Cl compared with their age-matched controls (Paper I, Figure 2). Data for Mg, P, S, and Ca were not significantly different among any of the groups (Mg, P, and S not shown).

Transmission electron microscopy of CF mouse trachea showed concretions in what appears to be a gland duct (Paper I, Figure 3) but otherwise a normal ultrastructure of the surface epithelial cells.

Whereas X-ray microanalysis of ASL in frozen-hydrated trachea mainly samples the upper mucous layer of the ASL, the Sephadex beads sample the aqueous component. The data on the ionic composition of the aqueous com-ponent of the tracheal ASL also showed concentrations of Na and Cl, each ~60 mM. As a comparison, data on mouse serum or plasma prepared for analysis in a way similar to the ASL are given (Paper I, Figure 4). The con-centration of K in tracheal ASL was much higher than that in serum.

Cholinergic stimulation of the animal by pilocarpine resulted in a significant decrease of all elemental concentrations in the tracheal ASL, with the excep-tion of Ca, of which the concentration in tracheal ASL from normal mice is at the limit of the detection method. Adrenergic stimulation with isoprotere-nol had a similar, but less pronounced, effect (Paper I, Figure 4).

Transmission electron microscopy of the surface epithelium of mouse tra-chea fixed in a buffer with a strength of 200 mosmol/kgH2O (equivalent to

42

100 mM NaCl) did not show any damage to the cells, compared with tissue fixed in a buffer with a strength of 300 mosmol/kg H2O (Paper I, Figure 5). In surface epithelium fixed in a buffer with a strength of 100 mos-mol/kgH2O, minor damage was observed in the form of small vacuoles in the apical part of some of the cells (not shown).

The elemental composition of nasal fluid from mice was not significantly different from that of ASL from the trachea. Stimulation with pilocarpine caused a significant increase in Na, Cl, and K in the nasal fluid, whereas stimulation with isoproterenol and phenylephrine only caused a significant increase in the K concentration (Paper I, Figure 6).

4.2 Paper II

There was no significant difference in the composition of the nasal fluid collected from the living mice during the first or second anesthesia, or at the post-mortem sampling (Paper II, Figures 2 and 3). Hence, repeated anesthe-sia did not influence the elemental content of the nasal fluid. The results show that the content of Na, Cl and K in the nasal fluid of �F508 homozy-gous mice is higher than in wild-type mice, in line with previous findings that CF null mice have significantly higher concentrations of these ions in their nasal fluid [147, paper I]. Intraperitoneal (i.p.) injection of 2,2,2 tribro-moethanol (TBE) was used for the anesthesia. TBE resulted in the simple and rapid induction of short-term anesthesia and did not influence the ion concentration in the nasal fluid of the experimental animals. During experi-ments all animals survived and no negative symptoms of the anesthesia were observed.

4.3 Paper III

Analysis of the apical fluid in frozen-hydrated specimens of liquid-liquid interface cultures showed that after equilibration for 24 hours, the Na and Cl concentrations in the apical fluid had decreased, both in CF cells and in con-trol cells, compared to the original PBS, in which the Na+ concentration was 150 mM, and the Cl- concentration 140 mM. The decrease was less in the CF cultures, and the Na and Cl concentrations were significantly higher in the apical fluid of CF cells compared to controls (Paper III, Figure 1). Both in control cells and CF cells, the apical fluid was clearly hypotonic.

43

When the time of equilibration in liquid-liquid interface cultures was varied, and the ionic composition of the apical fluid was determined with the Sephadex bead method, the results showed that equilibration had taken place after 4 hours, and that no further changes occurred after that point in time. At all three time points chosen, the apical fluid of CF cell cultures contained significantly more Na and Cl than control cultures but similar concentrations of K (Paper III, Figure 2). Addition of 60 �M glibenclamide to control cell cultures resulted in an increase of Na, Cl and K concentrations in the apical fluid after 24 hours. Glibenclamide did not have any effect on CF cell cul-tures. Exposure of control and CF cells to 10 mM NaCN, which inhibits cell metabolism and leads to a cell death of about 40%, resulted in significantly higher Na and Cl concentrations than in cultures not exposed to cyanide (Paper III, Figure 3).The concentrations of Na and Cl as determined by the Sephadex bead method in liquid-liquid interface were somewhat higher than those determined by X-ray microanalysis of frozen-hydrated samples, but the fluid in control cultures was still clearly hypotonic.

When cells are grown in air-liquid interface, they develop cilia (Paper III, Figure 4). Although no fluid is applied to the apical surface of cell cultures, an apical fluid layer produced by the cells appears. Analysis of frozen-hydrated air-liquid interface cultures showed higher Na+ concentrations (134 vs 89 mM) as well as higher Cl- concentrations (66 vs 40 mM) in the apical fluid of the CF culture compared to control, but also prominent peaks for the intracellular elements P, S and K, indicating that the electron beam had completely penetrated the fluid layer and excited the underlying cells. Analysis of the fluid layer in air-liquid interface cultures sampled with the Sephadex bead method showed that it was hypotonic both in the control and the CF cultures, but that the concentrations of Na, Cl and K were signifi-cantly higher in the CF cultures (Paper III, Figure 5). The ionic concentra-tions in the apical fluid of the air-liquid interface cultures collected with the Sephadex bead method were somewhat lower than in the liquid-liquid inter-face cultures. pH measurements in the apical fluid of 16HBE cells, or CFBE cells, respectively, showed that under unstimulated conditions, the apical fluid of both the 16HBE cells and the CFBE cells become more acid from a starting value of 7.35 for both 16HBE and CFBE cells; when chloride efflux was stimulated with forskolin and IBMX, the apical fluid covering the 16HBE cells became more alkaline, whereas the fluid covering the CFBE cells still became more acid, indicating that CFTR transported bicarbonate in the wild-type cells, but not in the CF cells. The alkalinization of the apical fluid covering the 16HBE cells was blocked by the specific CFTR-blocker CFTRinh172, but no effect was noted when the inhibitor was added to CFBE cells (Paper III, Table 1). Neither glibenclamide nor amiloride affected sig-nificantly the pH changes in the apical fluid in the absence of cAMP elevat-ing agents (not shown).

44

4.4 Paper IV

The results of the X-ray microanalysis of frozen-hydrated pig trachea, placed with the mucosal side up, are shown in Figure 1 (Paper IV). Figure 2 shows an image of frozen-hydrated pig trachea, mounted sideways. This image allows an estimation of the thickness of the ASL layer, which varies between 50 and 100 �m. X-ray microanalysis of the specimen mounted sideways (in the frozen-hydrated state) gave results close to those obtained with the ASL pointing upwards (Paper IV, Figure 1). Results on pig bronchi (measured in the frozen-hydrated state with the mucosal side up) showed somewhat higher concentrations for Na, Cl, and K compared with the trachea (Paper IV, Fig-ure 1). In contrast, as expected, according to data from X-ray microanalysis of the epithelial cells in the frozen-dried state, P and K are the main intracel-lular elements, and intracellular Na and Cl concentrations are low (data not shown).

Transmission electron microscopy of pig trachea fixed in solutions of differ-ent osmolarity showed that at concentrations of 50 mM sodium cacodylate and less, evident damage to cellular organelles occurred, in comparison to the normal morphology at the highest concentrations (100-150 mM) of so-dium cacodylate (Paper IV, Figure 3). At 50 mM, fluid-filled vesicles were formed in the epithelial cells and at lower osmolarity, the mitochondria showed clear signs of damage. In addition, it was observed that the cell size was dependent on the osmolarity of the fixative solution (Paper IV, Figure 4).

When Sephadex beads were placed onto the trachea and left there for 30 minutes, the beads sank through the ASL layer and reached the tips of the cilia (Paper IV, Figure 5). Minor mechanical compression of the epithelium due to the weight of the beads could be observed, but the epithelium was still intact and continuous. It can be seen in this figure that the mucus layer con-tained cells or cell debris. Figure 6 shows the results of the X-ray micro-analysis of Sephadex G-25 beads that had absorbed ASL for 30 minutes, compared with beads that had absorbed plasma or serum from the same pigs. The elemental concentrations in plasma were not significantly different from those in serum and the values were pooled. The Na and Cl concentrations in the ASL were only little lower than in plasma. The K concentration was higher than in plasma, but lower than in the measurements on frozen-hydrated ASL. The concentrations of P and S were similar to the values from plasma but lower than in the measurements on frozen-hydrated ASL.

45

4.5 Paper V

The analysis of beads with nasal fluid showed that in healthy subjects the elemental values were [Na]: 141 ± 8 mM, [Cl]: 170 ± 12 mM, [K]: 61 ± 8 mM, which was similar to those reported previously using the same tech-nique [141]. In CF patients, CF heterozygotes, and in rhinitis and PCD pa-tients the levels of Na and Cl in the nasal fluid were significantly higher than in healthy controls. In CF and PCD patients also the levels of K were higher than in healthy controls (P < 0.05) (Paper V, Figure 2).

No significant difference in Na, Cl, or K was observed between healthy males and females. However, females with CF had significantly higher con-centrations of Na, Cl, and K in their nasal fluid compared with those concen-trations in CF males (Paper V, Figure 3).

Although the sex difference in ion concentrations was not significant for PCD and rhinitis patients, there was a significant difference between male and female patients when all patients with airway diseases (CF, PCD, and rhinitis) were pooled (Paper V, Figure 3). The female CF patients classified as “severe” had significantly higher K concentrations in their nasal fluid than the female patients classified as “mild” or “medium” (Paper V, Figure 4). Within the group of cystic fibrosis patients, there was no significant correla-tion between the elemental concentrations and FEV1% or FVC% (Paper V, Figures 5a and 5b). There was also no significant correlation between age and elemental content of the nasal fluid (Paper V, Figure 5c). This was also the case for the other groups (results not shown).

4.6 Paper VI

Twenty-seven out of 39 patients (mean 5.2 years, range 0.1-16 years) had been treated in the mist tent. Two of these patients initiated the treatment when they were already chronically colonized with Pseudomonads. Seven of 39 patients became chronically colonized with Pseudomonads. Three of these seven patients did not have any mist tent therapy. The remaining four patients became chronically colonized after having ended their mist tent therapy (2, 0.5, 1.5, and 0.5 years after end of treatment). No patients be-came chronically colonized during mist tent treatment.

The analysis of beads with nasal fluid showed that in the nasal fluid of CF patients the concentration of Na, Cl, and K was 144 ± 4, 159 ± 4, and 58 ± 6 mM, respectively, before the patients entered the tent, significantly higher

46

than the levels in the nasal fluid of the controls, where the values were [Na]: 128 ± 1 mM, [Cl]: 121 ± 4 mM, and [K]: 22 ± 2 mM, similar to those re-ported previously using the same technique (Paper V). During the period in the tent, the ion content significantly decreased in both groups, to levels of [Na]: 45 ± 8 mM, [Cl]: 68 ± 9 mM, and [K]: 14 ± 2 mM in CF patient group and [Na]: 35 ± 2 mM, [Cl]: 47 ± 1 mM, and [K]: 7 ± 4 mM in control group immediately after the period in the mist tent. After leaving the mist tent the ion levels in the nasal fluid significantly increased, reaching after 4 hours values of [Na]: 162 ± 12 mM, [Cl]: 159 ± 6 mM, and [K]: 63 ± 5 mM in the CF patient group and [Na]: 130 ± 2 mM, [Cl]: 126 ± 11 mM, and [K]: 25 ± 4 mM in control group, which was for both groups slightly higher than before entering the mist tent. No major changes in the ion content of the airway surface liquid occurred after 4 hours.

47

Discussion

5.1 Ionic composition of ASL in mice, pigs, air-liquid & liquid-liquid interface cell cultures (Papers I, II, III, IV)

The ionic composition of the airway surface liquid (ASL) is of importance in cystic fibrosis. However, the elemental composition of the ASL has been debated and a wide range of values has been published. It has been difficult to determine the exact composition of the ASL due to its small volume and imperfect methods. Two techniques were developed to measure the elemen-tal composition of the ASL in mice (Papers I and II), cell cultures (Paper III), pigs (Paper IV), and humans (Papers V and VI). In one method, the composition of the ASL was measured by X-ray microanalysis of frozen-hydrated samples. In the second method, small dextran beads were equili-brated with the ASL and analyzed by X-ray microanalysis. The results from paper IV indicate that in the pig, which is close to man with respect to fluid transport in the airway, the Na+ concentration in the ASL is close to iso-osmolar. X-ray microanalysis of the ASL in frozen-hydrated mouse trachea indicates a hypotonic composition of the ASL (Paper I). This finding is con-firmed by the analysis of the watery fraction of the ASL, absorbed by the Sephadex beads. Analysis of frozen-hydrated trachea indicates that the ASL in CF mice has a higher content of NaCl than that in control mice. There are, however, major differences in the results of the ASL measured in situ in frozen-hydrated specimens, and the results obtained on the Sephadex beads. The ASL in situ contains substantially higher concentrations of, for example, P, S, and K than the ASL in the Sephadex beads. The elements P, S, and K are “cellular”, rather than “extracellular” elements. One could suspect that the measurements on the ASL in the frozen-hydrated state, underlying tissue, or at least cilia, are excited by the electron beam, and thus contribute to the spectrum. There are, however, several arguments against this notion. First, the spatial resolution of analysis at 9 or 10 kV in a frozen-hydrated specimen can be calculated to be in order of 2-3 �m [148]. This is much less than the actual thickness of the ASL that can be measured from micrographs where the trachea is mounted sideways. Second, the results from the analysis where

48

the trachea is mounted sideways, rather than with the ASL pointing upwards, give the same result, and in this situation overpenetration would not excite underlying tissue. Finally, if the results were due to overpenetration of the electron beam and excitation of the underlying epithelium, one would expect a negative correlation between, e.g., Na and P, or Na and K. A measurement with little or no overpenetration would sample the overlying ASL, and show high Na, and low P and K, whereas a measurement with much overpenetra-tion would mainly excite the epithelium and show high P and K and low Na. Analysis of the data, however, shows that the correlation between Na and P is not negative (Paper IV, Figure 7), and the same result is obtained for the relation between Na and K. The difference, then, between the measurements of ASL in situ in frozen-hydrated specimens and in the beads must be due to the fact that actually different substances are measured.

The measurements of the ASL in frozen-hydrated specimens measure the upper (mucous) layer of the fluid, whereas the beads absorb the watery com-ponent of the layer. The upper part of the ASL likely contains glycoproteins, exfoliated cells, and cell debris that could add “cellular” elements such as P, S, and K to the layer. The exfoliated cells and cell debris can be clearly seen in Figure 5 (Paper IV). If mucus and debris are left on the beads, higher val-ues for P, S, and K are obtained. The fact that the chloride concentration in the upper layer is somewhat lower might be due to the presence of nega-tively charged macromolecules in this layer that would tend to attract cations and repel anions. The total ion concentration in the ASL appears to be slightly higher then in serum. It could be argued that during analysis of the frozen-hydrated samples there is a risk for sublimation of the ice in the vac-uum of the electron microscope, which would result in an increase of the ionic concentrations during analysis. To minimize this risk, the samples were kept at the lowest possible temperature, and the standard solution (a salt solution containing albumin to mimic the organic part of the ASL) was spread out in a thin layer to resemble the analytical conditions for the ASL. The fact that the concentrations of Na and K in the ASL are higher than in serum does not necessarily mean that the fluid is hypertonic, because the presence of large negatively charged organic macromolecules from mucus and cell debris will bind cations and lower their activity. The result un-equivocally confirms the notion that the Na and Cl concentrations in the watery phase of the ASL of the pig are close to those found in plasma. The CF value of around 90 mM in the ASL in the beads agrees reasonably well with data of Giljam et al. [60], Hull et al. [62], Knowles et al. [65], and Joris et al. [73] for the human and confirms that the pig is a good model. The ele-mental composition of the watery layer of the ASL does, however, differ from that of plasma in a number of respects. Concentrations of Mg, P, and K are higher than in plasma. For K, where we find a concentration of 20 mM, some literature data are available. In human ASL, Knowles et al. [65] found

49

18 mM and Joris et al. [73] found 29 mM, which agrees well with our data. Both techniques agree that the ASL in mouse is hypotonic (Paper I). The data also show that neither age nor the method of anesthesia affects the com-position of the ASL. Our results agree reasonably well with data obtained by Cowley et al.[76], who found 87 mM for Na and 57 mM for Cl. However, our concentrations are lower than those published by Verkman [77], Song et al. [78] , Tarran et al. [79], and Caldwell et al. [80], who found a chloride concentration of ~115-120 mM. On the other hand, our values are much higher than those given by Baconnais et al. [75], who found concentrations < 10 mM. It may seem unreasonable that airway surface epithelial cells could be continuously exposed to a hypotonic fluid, but ultrastructural investiga-tions appear to confirm that the cells are not noticeably damaged by a fluid with a salt concentration of ~100 mM (200 mosmol/kg H2O), which is close to the 90 mM Na plus K measured. The relatively low ionic concentrations in mouse ASL are in contrast with our findings in the human [141] and the pig (Paper IV), but agree with data on the rat [149]. It is well known that the mouse airway has much fewer submucosal glands than pig and human air-ways [150]. Light microscopy of the control mouse tracheae used in the study (Paper I) failed to find a noticeable number of submucosal glands. It could be hypothesized that if much of the ASL is produced by glands as an isotonic fluid and if the surface epithelium absorbs ions from the ASL, this would result in lower ionic concentrations in the ASL compared with air-ways with many glands, such as pig and human airways. CF mice have sig-nificantly higher concentrations of Na and Cl in their ASL than the controls. This appears to disagree with findings of Zahm et al. [84], who found no significant differences between ASL in control and CF mice, but since those data are presented as mmol/kg dry weight, whereas our data are in mmol/kg wet weight, these studies are difficult to compare. One should be careful, however, in extending data from the mouse to the human, because of the differences in airway architecture. In one of the CFTR (-/-) mice, a concre-tion in a duct of a submucosal gland was observed (Paper I, Figure 3).

In general, transgenic mice with CF have been reported to show no evident signs of airway disease that would be comparable to the human airway with dilated gland ducts and solidified mucus [117], but a recent report found that in long-living CFTR (-/-) mice, airway pathology developed that resembled that in CF patients [151]. Our findings could thus be related to the age of the mice. The ionic composition of the ASL can be changed by pharmacological treatment. The effect of stimulation on the ionic composition of tracheal and nasal fluid, respectively, is markedly different. The effects on the nasal fluid can be explained by assuming secretion by glands in the nose, where the cholinergic stimulation gives rise to secretion of a NaCl-rich fluid, whereas the �-adrenergic agonist isoproterenol mainly causes secretion of protein-rich secretory granules, to which, apparently, K is bound. The �-adrenergic

50

agonist phenylephrine had only small effects. In the trachea, however, both pilocarpine and isoproterenol caused a significant decrease in ionic concen-trations. This remarkable effect can only be explained by assuming that the ASL is diluted due to secretion of water (fluid secretion can be observed microscopically). The origin of this water (the tracheal wall or the distal airways or alveoli) remains to be elucidated. In the trachea, different from the situation in the nose, submucosal glands would not contribute signifi-cantly to the fluid under stimulated conditions. Admittedly, the doses of the agonists used in the paper I are very high and some unspecific effects may be present. However, the data provide “proof of principle” that the ionic concentrations in the ASL can be manipulated by pharmacological treat-ment. The experimental system used in paper I offers the possibility to di-rectly test the effect of drugs on the ionic composition and water content of the ASL, which may be helpful for research on a disease where one wishes to increase the hydration of the fluid lining the airway wall.

In view of the high cost of CF mice, it would be advantageous to be able to take multiple samples from a single animal. Therefore, we investigated the influence of anesthesia (2,2,2 tribromoethanol (TBE)) on the ion concentra-tion in the mouse nasal fluid during a time-course study design in such a way that the animal could survive the testing (Paper II). During experiments all animals survived and no negative symptoms or influence of the anesthesia on the elemental content of the nasal ASL were observed. Although TBE solutions are often referred to as Avertin, this is misnomer. Avertin was the trade name for Winthrop Laboratories’ proprietary TBE formulation, which is no longer available [152]. TBE is an attractive choice of anesthetic for many researchers because it is easy and inexpensive and more importantly, i. p. injection of TBE results in good anesthesia which permits many surgical procedures on laboratory animals. Adverse reports about the efficacy and safety of TBE, however, combined with the availability of effective pharma-ceutical-grade alternatives, have made the continued routine use of TBE for rodent anesthesia controversial [153]. One study concluded that high losses due to death among mice after recovery were attributable to fluid distension of the stomach and small intestine, suggesting intestinal ileus as the cause of death [154]. Many of the adverse effects of TBE possibly occur because of improper preparation and storage. Meyer and Fish [152] in their review con-cluded that if intraperitoneal TBE is chosen by an investigator for survival animal studies, its use should be at the lowest practical concentration and effective dose.

CF mice are a convenient model for testing novel therapies such as gene therapy and new pharmaceuticals and their survival in long term experiments will be beneficial for many researches. The methods presented in these stud-

51

ies will allow the use of CF mice in studies, where repeated measurements of parameters on a single animal have to be carried out.

The data obtained from the cell culture study (Paper III) show that the apical fluid covering cultured normal bronchial epithelial cells is hypotonic and there is no great difference between air-liquid and liquid-liquid cultures, and that the apical fluid covering CF cells has a significantly higher Na and Cl content than that of control cells. The values obtained for Na and Cl concen-trations on control cultures in the different experimental set-ups used in the paper III range from about 80 to 100 mM for Na+ and 60-80 mM for Cl-. This is somewhat lower than the values obtained by Jayaraman et al. [63] who found values of 97 mM for Na+ and 117 mM for Cl- in air-liquid inter-face primary cultures of bovine trachea. Our values are also lower than those obtained by Knowles et al. [65] in primary cultures from canine trachea (around 120 mM), as well as found by Matsui et al. [64] and Tarran et al. [79] in primary cultures from human airway (125-140 mM) but much higher than those obtained by McCray et al. [85] (18 mM Cl-) in primary cultures from mouse tracheal epithelial cells. Apart from the difference in techniques to measure the composition of the apical fluid, an important difference be-tween our study (Paper III) and the studies referenced above is that those earlier studies used primary cultures of human and animal airway cells, whereas our study used cell lines. The cell lines used in our study are homo-geneous and equivalent to absorptive surface epithelial cells [136]. The ho-mogeneity of the system is an advantage, but also a limitation, since the data obtained may not be representative for the human airway in situ, where both glandular (secretory) and surface epithelial (absorptive) cells determine the composition of the ASL. It should be noted that bicarbonate and organic anions are not measured by X-ray microanalysis.

The studies in which the pH of the apical medium was determined give sup-port to the notion that in the wild-type 16HBE cells, CFTR transports (some) bicarbonate ions, and that this transport can be inhibited by a specific blocker of CFTR; in CF cells, this transport is absent. The fact that amiloride had no significant effect on the pH of the apical fluid under unstimulated (basal) conditions, may be explained by a low activity of the Na+-H+ ex-change mechanism if CFTR is not stimulated.

We could not find support for the notion that air-liquid cultures would be less suitable for these studies than liquid-liquid cultures [155]. In both cases, TEER values in excess of 300 �cm2 were obtained, meaning that tight junc-tions had been formed. A hypotonic apical fluid can only be obtained if the paracellular pathway is closed, since leakage of the isotonic medium from the basolateral compartment would increase the ionic concentrations in the apical fluid. In the apical fluid of the air-liquid cultures, the Na+ and Cl- con-

52

centrations were less than in the liquid-liquid cultures; hence, there is no indication for increased leakage. The data presented in paper III show that the apical fluid of CF cultures has a higher content of Na and Cl than that of control cells. It has been shown previously by Andersson et al. [156, 157] that the control 16HBE14o- cell line has CFTR in its plasma membrane and has a cAMP-activated chloride transport mechanism, whereas the CF cell line CFBE41o- does not transport chloride after stimulation with cAMP. These findings are in line with studies by Reddy et al. [158] and Kunzel-mann et al. [159] stating that activation of the epithelial Na+-channel (ENaC) requires activation of CFTR, hence in cultured CF cells, the activity of ENaC would be expected to be reduced. Addition of glibenclamide, an in-hibitor of CFTR, to the apical medium resulted in an increase of the Na+ and Cl- concentrations over a 24 h period, strengthening the notion that in this cell model, inhibition of CFTR results in higher Na+ and Cl- concentrations in the apical fluid. Also the data from the cells exposed to sodium cyanide (NaCN) indicate that metabolically inactive or dead cells do not cause a de-crease in the ionic concentrations of the apical fluid. The system is useful not only for studies related to CF, but also for studies related to other airway diseases where changes in the ionic composition of the ASL are relevant [160].

5.2 Ionic composition of nasal ASL in healthy and diseased humans (Papers V, VI)

The ASL layer is very thin and its composition may be sensitive to distur-bance by the sampling procedure. Collecting samples of nasal fluid by using ion-exchange beads does not cause any disturbance to the airway epithelium. The nose is a good model system to study ion concentrations in the ASL, as the nose has similar ion transport characteristics and a similar epithelium compared to the lower airways. It has been shown that the ionic composition of nasal fluid resembles that of ASL taken from the lower airways, even through there are small differences [62, 65]. In paper V, data show the con-centrations of Na, Cl, and K in nasal fluid in healthy subjects, CF patients, CF heterozygotes, rhinitis patients and patients with PCD. In healthy sub-jects, Na, Cl, and K values were similar to those reported previously using the same technique [141], and in line with the theory that ASL is isotonic [59, 79]. In CF patients, CF heterozygotes, and in PCD and rhinitis patients, the concentrations of Na and Cl were significantly higher than that in healthy controls, and in CF and PCD patients, even K was significantly higher than that in healthy subjects. The results for nasal fluid from healthy controls and CF patients in paper V show values for Na and Cl concentrations compara-

53

ble to those of earlier publications [56, 62, 65]. The ionic concentrations in the nasal ASL of healthy controls are hypertonic compared to that in serum. This does not contradict the notion that an isotonic fluid is produced, but it is likely that some evaporation of water occurs because of the flow of air, lead-ing to an increase in salt concentration. The sum of [Na] + [K] is larger than [Cl], which may be due to the presence of bicarbonate ions and organic ani-ons (amino acids, glycoproteins) in the mucus. The K concentration is much higher than in serum. This is likely due to K leaking out from damaged cells shed from the epithelium.

In the model of Matsui et al. [64], it is predicted that the ASL normally is isotonic, and that it is isotonic in CF patients as well, but has a reduced vol-ume (“low volume hypothesis”). In contrast, in the model of Smith et al. [56] the ASL is normally hypotonic, but salt concentrations are increased in CF patients (“high salt hypothesis”). Our findings that in a healthy person nasal ASL is slightly hypertonic would agree best with the “low volume hypothesis”. On the other hand, we find that the salt concentration in nasal ASL from CF patients and CF heterozygotes is significantly higher than in a normal person, which would be more in line with the alternative hypothesis. The fact, that the values for Na and Cl found in CF heterozygotes are higher than in normal persons may be due to the fact that in many CF heterozygotes chloride transport in nasal epithelial cells is abnormal even through this does not give rise to clinical problems [161]. It is uncertain how much of this increase in Na and Cl is due to a defect in CFTR, and how much is due to epithelial damage caused by chronic inflammation, since patients with rhini-tis have element concentrations that are not significantly different from those found in CF patients and have significantly higher values of Na and Cl than controls. Since, apart from the defect in CFTR also chronic inflammation plays a role in determining the final ion concentrations in the ASL, the ASL in (adult) patients does not present an ideal system to distinguish between the “low volume hypothesis” [64] and the “high salt hypothesis” [56]. The finding that CF patients with severe symptoms have higher ion concentra-tions than patients with mild or moderate symptoms does not contradict the hypothesis that inflammation is an important factor determining the ionic concentrations in the ASL.

So far, the composition of the ASL has not been studied in any other dis-eases besides CF. The hypothesis that epithelial damage caused by inflam-mation could cause a leakage of ions into the ASL, resulting in abnormally high Na and Cl concentrations seems reasonable, but expects further testing. Smith et al. [56] have also proposed that increased salt concentrations in the ASL would reduce the ability of defensins to kill bacteria, and in this way, increased Na and Cl concentrations in the ASL might actually increase the risk for infection. Female CF patients have significantly higher ionic concen-

54

trations in their nasal ASL than male patients. Female CF patients are known to be infected earlier by mucoid strains of Pseudomonas than male patients [162]. In a recent study on mice, it was shown that female mice were more susceptible to Pseudomonas aeruginosa infection than male mice [163].

The ASL provides a first line of defense against bacteria. The ASL contains a variety of factors, secreted by airway epithelial cells, which directly kill pathogens or modulate the inflammatory response. Known components of the ASL are mucins, epithelial glycoproteins [164], and antimicrobial pep-tides like lysozymes, lactoferrins [165] and wide range of defensins [166]. Defensins, one of the most intensively studied classes of antimicrobial pep-tides, have been identified in a wide variety of animals, including birds, ro-dents, and humans [97]. Defensins are small cationic peptides containing 29-47 arginine-rich amino acids with three disulfide bonds, which can be di-vided into the �- and �- defensin subfamilies in human subjects [97]. The main function of defensins is believed to be to kill bacteria and fungi either on the surfaces of the epithelial cells or within phagolysosomes of phago-cytes. Some studies have demonstrated that chemokine ligand 20 (CCL20) is a bi-functional peptide in airway host defense, participating in both the in-nate and adaptive immune system. It is secreted both apically and basolater-ally by airway epithelia in response to inflammatory stimuli, and exerts an-timicrobial activity against a wide spectrum of mainly Gram-negative bacte-ria [167, 168]. All these peptides exhibit a broad spectrum of activity and may be inactive in CF ASL due to its elevated salt concentration [85, 96, 165]. Some previous studies have shown that lysozyme and lactoferrin levels are not decreased in CF airway secretions; they may even be increased, but these studies did not examine the antimicrobial activity of these compounds [169, 170]. Therefore it can be concluded that the greater the amount of an-timicrobial factor or the lower the ionic strength, the greater the bacterial killing effect. This approach could be of benefit to patients with CF, irre-spective of the issues about differences in the salt content of ASL. Although there are a large number of identified antibacterial proteins and peptides in the ASL, and in other organs as well, a complete elucidation of these factors has not been achieved.

Antibiotics have clinical benefits in patients with cystic fibrosis [107]. Many mechanisms have been proposed, including the effects on neutrophil func-tion [108], IL-8 production [109], sputum rheology [110], goblet cell hyper-secretion [111], the alginate biofilm produced by Pseudomonas aeruginosa [112], and direct antipseudomonal activity [113]. However, the precise mechanisms are still uncertain [114]. It is conceivable that drugs used by some of the patients, especially CF and allergy patients, could affect the salt content of the nasal fluid. However, the effect of drugs on the ion content of

55

the nasal fluid has not been investigated, and it is therefore impossible to attribute particular changes to drugs used.

The results of these studies may provide a direction for focusing future re-search which may result in an improved treatment strategy for the CF dis-ease and other airway diseases, such as asthma or rhinitis. Since data from the present study indicate that CF patients have a fluid with a composition different from normal, it would theoretically be possible to investigate the effect of an experimental treatment for CF by investigating the effect of the treatment on the composition of the nasal fluid. Thus, the methods to deter-mine the ionic/elemental composition of the ASL, presented in this thesis, could be helpful in studies of the efficiency of possible pharmacological treatment of cystic fibrosis.

57

Conclusions and future perspectives

� The elemental composition of the ASL in mice is hypotonic. Al-though mouse airways have few similarities with human airways, a mouse could still be a good model for investigations. The data shown in the present study provide “proof of principle” that the ionic concentrations in the ASL can be manipulated by pharmacological treatment, and also that CF mice have higher concentrations of Na and Cl in their ASL compared to normal mice, which is an observa-tion similar to that in humans. Anesthesia does not influence the ionic concentration of mouse ASL, thus mice could be used in long term experiments in which the animals could survive the testing. The experimental system used in the present study offers the possibility to directly test the effect of drugs on the ionic composition and water content of the ASL, which may be helpful for research on disease where one wishes to increase the hydration of the fluid lining the airway wall.

� The pig is a good model for investigations on airways because pig

airways share many structural and physiologic similarities with hu-man airways. The ASL in pigs is isotonic.

� In a model system consisting of absorptive surface epithelial cells,

the apical fluid becomes hypotonic. Inhibition (by glibenclamide or NaCN) in control cells or absence of functional CFTR (in a CF cell line) results in slightly but significantly higher concentrations of Na+ and Cl- in the apical fluid. The cell culture system is useful not only for studies related to CF, but also for studies related to other airway diseases where changes in the ionic composition of the ASL are relevant.

� Higher ion concentrations in nasal ASL of CF patients could be par-

tially due to the known ion transport defect in CF epithelial cells. Pa-tients with other inflammatory airway diseases also display higher ionic concentrations in the ASL. Probably epithelial damage, caused by chronic inflammation, plays a part in the disturbance of ion transport across the airway epithelium. The salt content of the ASL may be relevant in CF, since the ASL is known to contain anti-

58

bacterial proteins (defensins) that are sensitive to the salt concentra-tion. Therefore, the higher salt concentrations in the ASL of CF pa-tients may have negative consequences for the anti-bacterial defense system in the lung, and conversely, the decrease in ion concentra-tions, caused by spending time in a mist tent, may have positive ef-fects. However, with currently used procedures, the effect of sleep-ing in a mist tent on the ion content of the nasal ASL is short-lived.

59

Sammanfattning på svenska

Cystisk fibros (CF) är en medfödd, ärftlig, dödlig sjukdom som främst drab-bar den vita rasen. CF orsakas av en mutation i en gen, CFTR (cystic fibros transmembrane conductance regulator), som är ansvarig för bildandet av en kloridkanal och som indirekt är ansvarig för bildandet av vätskan som täcker luftvägarna, vätskan i tarmkanalen, och svettens sammansättning. Luftvägarna täcks av ett tunt vätskeskikt i vilket flimmerhåren badar. Överst i skiktet finns slem som fångar upp bakterier och andra partiklar i inand-ningsluften, så att de kan avlägsnas av flimmerhåren. Om kloridkanalen CFTR saknas eller är defekt leder detta till en defekt transport av joner och vatten i luftvägsslemhinnan, vilket i sin tur leder till segt slem, problem med flimmerhårens verksamhet, inflammationer, bakterietillväxt, och vävnads-skador. Volymen och sammansättningen hos den vätska som täcker luftvä-garna är därför viktiga för hur CF uppstår, och det är därför relevant att be-stämma vätskans sammansättning. Detta innebär en svårighet, eftersom skik-tet är så tunt (20-50 tusendels millimeter). När denna studie inleddes varie-rade de publicerade siffrorna om jonkoncentrationerna i vätskan i mycket hög grad, bl a beroende på att olika mättekniker och olika djurslag användes. Detta problem ledde till att en relativt enkel metod för att bestämma jonkon-centrationerna i luftvägsvätskan i olika djurslag, inklusive människan ut-vecklades. Två mättekniker utvecklades för att mäta luftvägsvätskans sam-mansättning, så att indirekt information om jontransport i luftvägsslemhin-necellerna kunde samlas in. När det gällde den första metoden avlägsnades vävnad från luftstrupen från det nedsövda försöksdjuret. Denna frystes, och analyserades med hjälp av röntgenmikroanalys i svepelektronmikroskopet. När det gällde den andra metoden samlades luftvägsvätskan in med hjälp av små dextrankulor (Sephadexkulor) som sedan torkades och analyserades. Denna metod gav värden för luftvägsvätskan (utan slem) och kunde även användas hos människan, vilket var en stor fördel. Båda metoderna användes för att bestämma jonsammansättningen i vätskan från luftstrupen och näsan hos grisar, normala och transgena CF-möss, väts-kan som täckte cellodlingar av normala (16HBE14o-) eller CF-luftvägsslemhinneceller (CFBE41o-), samt slutligen näsvätskan hos friska

60

människor och patienter med CF eller andra luftvägssjukdomar. Luftvägs-vätskan har ungefär samma jonkoncentrationer som blodserum hos grisar och friska människor. CF-patienter har däremot mycket högre halter av na-trium- och kloridjoner än friska människor. Eftersom luftvägsvätskan inne-håller antibakteriella proteiner, som fungerar sämre om salthalten är för hög, kan detta ha betydelse för det faktum att CF-patienter är så infektionskänsli-ga. Det kunde också visas att jonkoncentrationerna i luftvägsvätskan kunde ändras genom farmakologisk stimulering. Studien kunde också visa att CFTR spelar roll för transporten av bikarbonat-joner, och därmed påverkar luftvägsvätskans pH. Slutligen studerades effekten av dimtältbehandling på CF-patienter. Denna behandling hade som syfte att späda ut det sega slemmet, så att det blev lät-tare att hosta upp. Studien kunde visa att behandlingen minskade saltkon-centrationen i luftvägsvätskan, vilket i sig var positivt, men effekten var tyvärr kortvarig.

61

Acknowledgements

I would like to express my sincere gratitude to the many people who have in different ways contributed to this thesis. In particular, I would like to thank:

My superb supervisor, Professor Godfried Roomans, for introducing me to the field of electron microscopy and airway diseases, for being so supportive and encouraging both professionally and privately. Thank you also for al-ways making time to listen, for always being there to help, and creating a nice atmosphere in the group.

My two co-supervisors: Professor Marieann Högman, thank you for your tuition and advice, for great support, ideas and sharing your joy. Thank you, Dr Anca Dragomir, for your patience to teach me, especially during my first years in the lab, for support in research and fruitful discussions, for your smiles and optimism.

Anders Ahlander, Marianne Ljungkvist, Leif Ljung, Barbro Einarsson for their excellent laboratory and technical assistance. Their help with tissue sectioning, cell culturing, and electron microscopy was as valuable as their ability to create such a pleasant lab atmosphere.

Elsbeth Scholtes, thank you for all your support and help, for sharing nice moments and memorable dates during my stay in Sweden.

My co-PhD-students: Igor Oliynyk, Georgia Varelogianni, Harriet Nilsson, Andrei Malinovschi, Dieter Fuchs, and all the other present and former PhD students, students, and researchers at the department, thank you all for creat-ing a fun and pleasant atmosphere. Colleagues and corridor seniors Mats Hjortberg, Agneta Lukinius, Faranak Azarbayjani, and Jan Westman, for a friendly and inspiring atmosphere.

My co-authors and contributors to the studies – thank you for fruitful col-laborations! Special thanks go to Viengphet Vanthanouvong, Harriet Nils-son, Johanna Henriksnäs, Mia Phillipson, Marie Johannesson, Lena Hjelte, Bob Scholte, and Malin Flodström-Tullberg.

62

The administrative staff of the Department of Medical Cell Biology: Agneta Sandler Bäfwe, Kärstin Flink and Marianne Ljungkvist for help and positive attitude.

Cystic fibrosis patients at Uppsala University Hospital and Huddinge CF center in Stockholm who kindly participated in my projects.

Elena Kozlova-Aldskogius for introducing me to Professor Godfried Roomans, for insisting on my starting doing research in biology, and all the help.

Thank you, Oleg, for being my friend, for your advice and all the help with the computer, providing literature, pleasant coffee-breaks, for your hospital-ity and excursions in Crimea. Thank you, Inna and Pasha, Igor and Georgia, for nice company outside the work skiing, partying and just enjoying life together. My best friends from Kherson: Svetlana and Leonid, for keeping in contact despite the distance and always being close to me, for wonderful times spent together during summer vacation! Other friends and relatives from Ukraine, all remembered, and none forgotten!

I would also like to thank my family, my mother and father Nadezda and Ivan, my brother Oleg and his wife Tatjana, my nephew Igor, my niece Anna, for being proud of me whatever I am doing and never letting me down, for all love, support and constant encouragement.

Thank you, my lovely daughters, Alina and Jana, for full happiness and joy. Everything what I have done in my life is only for your sake!

and Sasha, I love you.

Financial support for these studies was provided by the Cystic Fibrosis Foundation (USA), the Swedish Heart Lung Foundation, the Swedish Sci-ence Research Council, the Swedish Asthma and Allergy Association and the Swedish Association for Cystic Fibrosis.

Uppsala, October 2008

� ��

63

References

[1] C. R. Marino, F. S. Gorelick, Scientific advances in cystic fibrosis, Gastroen-terology 103 (1992) 681-693.

[2] T. F. Boat, M. J. Welsh, A. L. Beaudet, Cystic fibrosis, In: C. R. Scriver, A. L. Beaudet, W. S. Sly, D. Valle, eds., The metabolic basis of inherited disease, Vol 2, New York, NY: McGraw-Hill (1989) 2649-2680.

[3] G. Fanconi, E. Uehlinger, C. Knauer, Das coeliakiesyndrom bei angeborener zysticher pankreasfibromatose und bronchiectasien, Wiener Klin Wochen 86 (1936) 753.

[4] D. H. Andersen, Cystic fibrosis of the pancreas and its relation to celiac disease: a clinical and pathological study, Am J Dis Child 56 (1938) 344-399.

[5] S. Farber, Pancreatic insufficiency and the celiac syndrome, N Engl J Med 229 (1943) 653-682.

[6] P. A. di Sant’Agnese, R. C. Darling, G. A. Perera, E. Shea, Abnormal electrolyte composition of sweat in cystic fibrosis of the pancreas. Clinical significance and relationship to disease, Pediatrics 12 (1953) 549-563.

[7] L. E. Gibson, R. E. Cooke, A test for concentration of electrolytes in sweat in cystic fibrosis of the pancreas utilizing pilocarpine by iontophoresis, Pediatrics 23 (1959) 545-549.

[8] P. M. Quinton, Chloride impermeability in cystic fibrosis, Nature 301 (1983) 421-422.

[9] R. C. Boucher, M. J. Stutts, M. R. Knowles, L. Cantley, J. T. Gatzy, Na+ trans-port in cystic fibrosis respiratory epithelia: abnormal basal rate and response to adenylate cyclase activation, J Clin Invest 78 (1986) 1245-1252.

[10] M. R. Knowles, M. J. Stutts, A. Spock, N. Fischer, J. T. Gatzy, R. C. Boucher, Abnormal ion permeation through cystic fibrosis respiratory epithelium, Science 221 (1983) 1067-1070.

[11] J. R. Riordan, J. M. Rommens, B. Kerem, N. Alon, R. Rozmahel, Z. Grzelczak, J. Zielenski, S. Lok, N. Plavsic, J. L. Chou, et al., Identification of the cystic fi-brosis gene: cloning and characterization of complementary DNA, Science 245 (1989) 1066-1073.

[12] B. Kerem, J. M. Rommens, J. A. Buchanan, D. Markiewicz, T.K. Cox, A. Chakravarti, M. Buchwald, L. C. Tsui, Identification of the cystic fibrosis gene: genetic analysis, Science 245 (1989) 1073-1080.

[13] H. Kollberg, Cystic fibros urgammal sjukdom med ny aktualitet, Historik och klinik, Läkartidningen 96 (1999) 3324-3329.

[14] S. M. Rowe, S. Miller, E. J. Sorscher, Cystic fibrosis, N Engl J Med 352 (2005) 1992-2001.

[15] A. Vankeerberghen, H. Cuppens, J. J. Cassiman, The cystic fibrosis transmem-brane conductance regulator: an intriguing protein with pleiotropic function, J Cystic Fibrosis 1 (2002) 13-29.

[16] T. Hegedus, A. Aleksandrov, L. Cui, M. Gentzsch, X. B. Chang, J. R. Riordan, F508del CFTR with two altered RXR motifs escapes from ER quality control

64

but its channel activity thermally sensitive, Biochem Biophys Acta 1758 (2006) 565-572.

[17] J. L. Bobadilla, M. H. Farrell, P. M. Farrell, Applying CFTR molecular genetics to facilitate the diagnosis of cystic fibrosis through screening, Adv Pediatr 49 (2002) 131-190.

[18] C. L. Ward, R. R. Kapito, Intracellular turnover of cystic fibrosis transmem-brane conductance regulator. Inefficient processing and rapid degradation of wild-type and mutant proteins, J Biol Chem 269 (1994) 25710-25718.

[19] G. D. Heda, M. Tanwani, C. R. Marino, The Delta F508 mutation shortens the biochemical half-life of plasma membrane CFTR in polarized epithelial cells, Am J Physiol Cell Physiol 280 (2001) C166-C174.

[20] E. Shebani, S. Shahana, C. Janson, G. M. Roomans; BHR group, Attachment of columnar airway epithelial cell in asthma, Tissue Cell 37 (2005) 145-152.

[21] J. R. Harkema, A. Marriassy, J. St.Goerge, D. M. Hyde, C. G. Plopper, Epithe-lial cells in conducting airways: a species comparision., ed., Marcel-Dekker, New York (1991).

[22] S. M. Serafini, E.D. Michaelson, Length and distribution of cilia in human and canine airways, Bull Eur Physiopathol Respir 13 (1977) 551-559.

[23] H. Gilljam, A. M. Motakefi, B. Robertson, B. Strandvik, Ultrastructure of the bronchial epithelium in adult patients with cystic fibrosis, Eur J Respir Dis 71 (1987) 187-194.

[24] M. J. Welsh, Mechanisms of airway epithelial ion transport, Clin Chest Med 7 (1986) 273-283.

[25] N. Kälin, A. Claass, M. Sommer, E. Puchelle, B. Tümmler, DeltaF508 CFTR protein expression in tissues from patients with cystic fibrosis, J Clin Invest 15 (1999) 1379-1389.

[26] S. M. Kreda, M. Mall, A. Mengos, L. Rochelle, J. Yankaskas, J. R. Riordan, R. C. Boucher, Characterization of wild-type and deltaF508 cystic fibrosis trans-membrane regulator in humanrespiratory epithelia, Mol Biol Cell 16 (2005) 2154-2167.

[27] P. G. Middleton, K. A. Pollard, E. Donohoo, J. R. Wheatley, D. M. Geddes, E. W. Alton, Airway surface liquid calcium modulates chloride permeability in the cystic fibrosis airway, Am J Respir Crit Care Med 168 (2003) 1223-1226.

[28] M. J. Welsh, Electrolyte transport by airway epithelia, Physiol Rev 67 (1987) 1143-1184.

[29] J. H. Widdicombe, C. B. Basbaum, E. Highland E, Ion contents and other prop-erties of isolated cells from dog tracheal epithelium, Am J Physiol Cell Physiol 241 (1981) C184-C192.

[30] J. D. McCann, M. J. Welsh, Regulation of Cl- and K+ channels in airway epi-thelium, Annu Rev Physiol 52 (1990) 115-135.

[31] P. M. Quinton, Cystic Fibrosis: a disease in electrolyte transport, FASEB J 4 (1990) 2709-2717.

[32] S. Blouquit, A. Regnier, L. Dannhoffer, C. Fermaian, E. Naline, R. C. Boucher, T. Chinet, Ion and fluid transport properties of small airways in cystic fibrosis, Am J Respir Crit Care Med 174 (2006) 299-305.

[33] E. M. Schwiebert, D. J. Benos, M. E. Egan, M. J. Stutts, and W. B. Guggino, CFTR is a conductance regulator as well as a chloride channel, Physiol Rev 79 (1999) S145-S166.

[34] M. J. Stutts, C. M. Canessa, J. C. Olsen, M. Hamrick, J. A. Cohn, B. C. Rossier, B. C. Boucher, CFTR as a cAMP-dependent regulator of sodium channels, Sci-ence 269 (1995) 847-850.

65

[35] M. R. Knowles, M. J. Stutts, J. R. Yankaskas, J. T. Gatzy, R. C. Boucher, Ab-normal respiratory epithelial ion transport in cystic fibrosis, Clin Chest Med 7 (1986) 285-297.

[36] A. J. Hirsh, B. F. Molino, J. Zhang, N. Astakhova, W. B. Geiss, B. J. Sargent, B. D. Swenson, A. Usyatinsky, M. J. Wyle, R. C. Boucher, R. T. Smith, A. Za-murs, M. R. Johnson, Design, synthesis, and structure-activity relationships of novel 2-substituted pyrazinoylguanidine epithelial sodium channel blockers: drugs for cystic fibrosis and chronic bronchitis, J Med Chem 49 (2006) 4098-4115.

[37] R. C. Boucher, Relationship of airway epithelial ion transport to chronic bron-chitis, Proc Am Thorac Soc 1 (2004) 66-70.

[38] A Roos, W. F. Boron, Intracellular pH, Physiol Rev 61 (1981) 296-424. [39] A. M. Paradiso, R. D. Coakley, R. C. Boucher, Polarized distribution of HCO3

- transport in human normal and cystic fibrosis nasal epithelia, J Physiol 548 (2003) 203-218.

[40] M. J. Welsh, J. J. Smith, cAMP stimulation of HCO3- secretion across airway

epithelia. J Pancreas 2 (2001) 291-293. [41] S. D. Freedman, H. F. Kern, G. A. Scheele, Pancreatic acinar cell dysfunction

in CFTR (-/-) mice is associated with impairments in luminal pH and endocyto-sis, Gastroenterology 121 (2001) 950-957.

[42] J. H. Poulsen, H. Fischer, B. Illek, T. E. Machen, Bicarbonate conductance and pH regulatory capability of cystic fibrosis transmembrane conductance regula-tor, Proc Natl Acad Sci USA 91 (1994) 5340-5344.

[43] R. D. Coakley, R. C. Boucher, Regulation and functional significance of airway surface liquid pH, J Pancreas 2 (2001) 294-300.

[44] H. Matsui, S. H. Randell, S. W. Peretti, W. Davis, R. C. Boucher, Coordinated clearance of periciliary liquid and mucus from airway surface, J Clin Invest 102 (1998) 1125-1131.

[45] R. D. Coakley, B. R. Grubb, A. M. Paradiso, J. T. Gatzy, L. G. Johnson, S. M. Kreda, W. K. O’Neal, R. C. Boucher, Abnormal surface liquid pH regulation by cultured cystic fibrosis bronchial epithelium, Proc Natl Acad Sci USA 100 (2003) 16083-16088.

[46] M. E. Krouse, J. F. Talbott, M. M. Lee, N. S. Joo, J. J. Wine. Acid and base secretion in the Calu-3 model of human serous cells, Am J Physiol Lung Cell Mol Physiol 287 (2004) L1274-L1283.

[47] H. Fischer, J. H. Widdicombe, B. Illek, Acid secretion and proton conductance in human airway epithelium, Am J Physiol Cell Physiol 282 (2002) C736-C743.

[48] S. K. Inglis, S. M. Wilson, R. E. Olver, Secretion of acid and base equivalents by intact distal airways, Am J Physiol Lung Cell Mol Physiol 284 (2003) L855-L862.

[49] M. J. Stutts, B. C. Rossier, R. C. Boucher, Cystic fibrosis transmembrane con-ductance regulator invert protein kinase A-mediated regulation of epithelial so-dium channel single channel kinetics, J. Biol Chem 272 (1997) 14037-14040.

[50] R. C. Boucher, New concepts of the pathogenesis of cystic fibrosis lung dis-ease, Eur J Respir 23 (2004) 146-158.

[51] R. C. Boucher, An overview of the pathogenesis of cystic fibrosis lung disease, Adv Drug Deliv Rev 54 (2002) 1359-1371.

[52] D. Carlander, Avian IgY antibody: In vitro and in vivo, Doctoral Thesis, Uni-versity of Uppsala (2002).

[53] www.med.unc.edu/cystfib/Staff/paradiso.htm [54] M. R. Knowles, R. C. Boucher, Mucus clearance as a primary innate defence

mechanisms for mammalian airways, J Clin Invest 109 (2002) 571-577.

66

[55] J. H. Widdicombe, J. G. Widdicombe, Regulation of human airway surface liquid, Respir Physiol 99 (1995) 3-12.

[56] J. J. Smith, S. M. Travis, E. P. Greenberg, M. J. Welsh, Cystic fibrosis airway epithelial fail to kill bacteria because of abnormal airway surface fluid, Cell 85 (1996) 229-236.

[57] P. Quinton, Viscosity versus composition in airway pathology, Am J Respir Crit Care Med 149 (1994) 6-7.

[58] J. Zabner, J. J. Smith, P. H. Karp, J. H. Widdicombe, M. J. Welsh, Loss of CFTR chloride channels alters salt absorption by cystic fibrosis airway epithelia in vitro, Mol Cell 2 (1998) 397-403.

[59] R. C. Boucher, Molecular insights into the physiology of the "thin film" of airway surface liquid, J Physiol (Lond) 516 (1999) 631-638.

[60] H. Gilljam, A. Ellin, B. Strandvik, Increased bronchial chloride concentration in cystic fibrosis, Scand J Clin Lab Invest 49 (1989) 121-124.

[61] M. J. Goldman, G. M. Anderson, E. D. Stolzenberg, U. P. Kari, M. Zasloff, J. M. Wilson, Human �-defensin-1 is a salt-sensitive antibiotic in lung that is inac-tivated in cystic fibrosis, Cell 88 (1997) 553-560.

[62] J. Hull, W. Skinner, C. Robertson, P. Phelan, Elemental content of airway sur-face liquid from infants with cystic fibrosis, Am J Respir Crit Care Med 157 (1998) 10-14.

[63] S. Jayaraman, Y. Song, L. Vetrivel, L. Shankar, A. S. Verkman, Noninvasive in vivo fluorescence measurement of airway-surface liquid depth, salt and concen-tration, and pH, J Clin Invest 107 (2001) 317-324.

[64] H. Matsui, B. R. Grubb, R. Tarran, S. H. Randell, J. T. Gatzy, C. W. Davis, R. C. Boucher, Evidence for pericilliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic fibrosis airway disease, Cell 95 (1998) 1005-1015.

[65] M. R. Knowles, J. M. Robinson, R. E. Wood, C. A. Pue, W. M. Mentz, G. C. Wager, J.T. Gatzy, R. C. Boucher, Ion composition of airway surface liquid of patients with cystic fibrosis as compared with normal and disease-control sub-jects, J Clin Invest 100 (1997) 2588-2595.

[66] W. B. Guggino, Cystic fibrosis and the salt controversy, Cell 96 (1999) 607-610.

[67] A. S. Verkman, Y. Song, J. R. Thiagarajah, Role of airway surface liquid and submucosal glands in cystic fibrosis lung disease, Am J Physiol Cell Physiol 284 (2003) C2-C15.

[68] D. Worlitzsch, R. Tarran, M. Ulrich, U. Schwab, A. Cekici, K. C. Meyer, P. Birrer, G. Bellon, J. Berger, T. Weiss, K. Botzenhart, J. R. Yankaskas, S. Ran-dell, R. C. Boucher, G. Döring, Effect of reduced mucus oxygen concentration in airway Pseudomonas infections of cystic fibrosis patients, J Clin Invest 109 (2002) 317-325.

[69] S. Jayaraman, N. S. Joo, B. Reits, J. J. Wine, A. S. Verkman, Submucosal gland secretions in airways from cystic fibrosis patients have normal Na+ and pH but elevated viscosity, Proc Natl Acad Sci USA 98 (2001) 8119-8123.

[70] J. A. Nadel, B. Davis, R. J. Phipps, Control of mucus secretion and ion trans-port in airways, Annu Rev Physiol 41 (1979) 369-381.

[71] L. Trout, M. King, W. Feng, S. K. Inglis, S. T. Ballard, Inhibition of airway liquid secretion and its effect on the physical properties of airway mucus, Am J Physiol Lung Cell Mol Physiol 274 (1998) L258-L263.

[72] http://www.stanford.edu/class/psych121/humangenome-CF.htm

67

[73] L. Joris, I. Dab, P. M. Quinton, Elemental composition of human airway sur-face fluid in healthy and diseased airways, Am Rev Respir Dis 148 (1993) 1633-1637.

[74] S. Baconnais, R. Tirouvanzium, J. M. Zahm, S. de Bentzmann, B. Peault, G. Balossier, E. Puchelle, Ion composition and rheology of airway liquid from cys-tic fibrosis fetal tracheal xenografts, Am J Respir Cell Mol Biol 20 (1999) 605-611.

[75] S. Baconnais, J. M. Zahm, L. Kilian, P. Bonhomme, D. Gobillard, A. Perchet, E. Puchelle, G. Ballossier, X-ray microanalysis of native airway surface liquid collected by cryothechnique, J Microsc 191 (1998) 311-319.

[76] E. A. Cowley, K. Govindaraju, D. K. Lloyd, D. H. Eidelman, Airway surface fluid composition in mice, Am J Physiol Lung Cell Mol Physiol 278 (2000) L1213-L1220.

[77] A. S. Verkman, Lung disease in cystic fibrosis: is airway surface liquid compo-sition abnormal? Am J Physiol Cell Mol Physiol 281 (2001) L306-L308.

[78] Y. Song, J. Thiagarajah, A. S. Verkman, Sodium and chloride concentrations, pH, and depth of airway surface liquid in distal airways, J Gen Physiol 122 (2003) 511-519.

[79] R. Tarran, B. R. Grubb, D. Parsons, M. Picher, A. J. Hirsh, C.W. Davis, R. C. Boucher, The CF salt controversy: in vivo observations and therapeutic ap-proaches, Mol Cell 8 (2001) 149-151.

[80] R. A. Caldwell, B. R. Grubb, R. Tarran, R. C. Boucher, M. R. Knowles, P. M. Barker, In vivo airway surface liquid Cl- analysis with solid-state electrodes, J Gen Physiol 119 (2002) 3-14.

[81] R. C. Boucher, M. J. Stutts, P. A. Bromberg, J.T.Gatzy, Regional differences in airway surface liquid composition, J Appl Physiol 50 (1981) 613-620.

[82] N. P. Robinson, H. Kyle, S. E. Webber, J. G. Widdicombe, Electrolyte and other chemical concentrations in tracheal airway surface liquid and mucus, J Appl Physiol 66 (1989) 2129-2135.

[83] E. A. Cowley, K. Govindaraju, D. K. Lloyd, D. H. Eidelman, Airway surface fluid composition in the rat determined by capillary electrophoresis, Am J Physiol, 273 (1997) L895-L899.

[84] J. M. Zahm, S. Baconnais, D. J. Davidson, S. Webb, J. Dorin, N. Bonnet, G. Balossier, E. Puchelle, X-ray microanalysis of airway surface liquid collected in cystic fibrosis mice, Am J Physiol Lung Cell Mol Physiol 281 (2001) L309-L313.

[85] P. B. Jr. McCray, J. Zabner, H.P. Jia, M. J. Welsh, P. S. Thorne, Efficient kill-ing of inhaled bacteria in DeltaF508 mice: role of airway surface liquid compo-sition, Am J Physiol 227 (1999) L183-L190.

[86] J. H. Widdicombe, S. J. Bastacky, D. X. Wu, C. Y. Lee, Regulation of depth and composition of airway surface liquid, Eur Respir J 10 (1997) 2892-2897.

[87] A. Bush, E. W. Alton, J. C. Davies, U. Griesenbach, A. Jaffe (eds), Cystic Fi-brosis in the 21st Century, Prog Respir Res Basel Karger 34 (2006) 122-130.

[88] D. McShane, J. C. Davies, T. Wodehouse, A. Bush, D. Geddes, E. W. Alton, Normal nasal mucociliary clearance in CF children: evidence against a CFTR-related defect, Eur Repir J 24 (2004) 95-100.

[89] B. Button, R. C. Boucher, Role of mechanical stress in regulating airway sur-face hydration and mucus clearance rates, Respir Physiol Neurobiol 8 (2008) [Epub ahead of print].

[90] L. W. Matthews, C. F. Doershuk, S. T. Miller, S. Pittman, M. Lough, Mist tent therapy of cystic fibrosis, J Sathma Res 5 (1968) 267-276.

68

[91] A. P. Norman, M. Hall, Nocturnal mist therapy in cystic fibrosis, Practitioner 206 (1971) 786-796.

[92] E. K. Motoyama, L. E. Gibson, C. J. Zigas, Evaluation of mist tent therapy in cystic fibrosis using maximum expiratory flow volume curve, Pediatrics 50 (1972) 299-306.

[93] N. Chang, H. Levison, K. Cunningham, D. N. Crozier, O. Grosett, An evalua-tion of nightly mist tent therapy for patients with cystic fibrosis, Am Rev Respir Dis 107 (1973) 672-675.

[94] M. A. Bureau, D. M. McDougall, P. H. Beaudry, M. M. Belmonte, Late effect of nocturnal mist tent therapy related to the severity of airway obstruction in children with cystic fibrosis, Pediatrics 61 (1978) 842-846.

[95] B. Jakobsson, L. Hjelte, B. Nystrom, Low level of bacterial contamination of mist tents used in home treatment of cystic fibrosis patients, J Hosp Infect 44 (2000) 37-41.

[96] S. M. Travis, B. A. Conwey, J. Zabner, J. J. Smith, N. N. Anderson, P. K. Singh, E. P. Greenberg, M. J. Welsh, Activity of abundant antimicrobials of the human airway, Am J Respir Cell Mol Biol 20 (1999) 872-879.

[97] S. Van Westering, P. J. Sterk, K. F. Rabe, P. S. Hiemstra, Defensins: key play-ers or bystanders in infection, injury, and repair ion the lung? J Allergy Clin Immunol 104 (1999) 1131-1138.

[98] P. L. Zeitlin, Emerging drug treatments for cystic fibrosis, Expert Opin Emerg Drugs 8 (2003) 523-535.

[99] S. D. Grosse, C. A. Boyle, J. R. Botkin, A. M. Comeau, M. Kharrazi, M. Rosenfeld, B. S. Wilfond, Newborn screening for cystic fibrosis evaluation of benefits and risks and recommendations for state newborn screening programs, MMWR Recomm Rep 53 (2004) 1-36.

[100] L. E. Bennett, B. M. Keck, O.P. Daily, R. J. Novick, J. D. Hosenpud, World-wide thoracic organ transplantation: a report from the UNOS/ISHLT Interna-tional Registry for Thoracic Organ Transplantation, Clin Transpl (2000) 31-44.

[101] B. D. Alexander, E. W. Petzold, L. B. Reller, S. M. Palmer, R. D. Davis, C. W. Woods, J. J. Lipuma, Survival after lung transplantation of cystic fibrosis pa-tients infected with Burkholderia cepacia complex, Am J Transpl 8 (2008) 1025-1030.

[102] E. Kerem, J. Reisman, M. Corey, G. J. Canny, H. Levison, Prediction of mor-tality in patients with cystic fibrosis, N Engl J Med 326 (1992) 1187-1191.

[103] M. E. Bowdish, M. L. Barr, Living lobar lung transplantation, Respir Care Clin N Am 10 (2004) 563-579.

[104] I Durieu, R. Nove Josserand, [Cystic fibrosis in 2008], Rev Med Interne (2008).

[105] R. Mahadeva, K. Webb, R. C. Westerbeek, N. R. Carroll, M. E. Dodd, D. Bilton, D. A. Lomas, Clinical outcome in relation to care in centres specializing in cystic fibrosis: cross sectional study, Bmj 316 (1998) 1771-1775.

[106] M. S. Schechter, B. J. Shelton, P. A. Margolis, S. C. Fitzsimmons, The asso-ciation of socioeconomic status with outcomes in cystic fibrosis patients in the United States, Am J Respir Crit Care Med 163 (2001) 1331-1337.

[107] A. Equi, I. M. Balfour-Lynn, A. Bush, M. Rosenthal, Long term azithromycin in children with cystic fibrosis: a randomized, placebo-controlled crossover trial, Lancet 16 (2002) 597-605.

[108] J. Kadota, O. Sakito, S. Kohno, H. Sawa, H. Mukae, H. Oda, K. Kawakami, K. Fukushima, K. Hiratani, K. Hara, A mechanism of erythromycin treatment in patients with diffuse panbronchiolitis, Am Rev Respir Dis 147 (1993) 153-159.

69

[109] O. A. Khair, J. L. Devalia, M. M. Abdelaziz, R. J. Sapsford, R. J. Davis, Ef-fect of erythromycin on Hoemophilus influenzae endotoxin-induced release of IL-6, IL-8 and sICAM-1 by cultured human bronchial eoithelial cells, Eur Res-pir J 8 (1995) 1452-1457.

[110] M. J. Dupont, J. R. Lapointe, Quantitative effect of roxithromycin and rifam-picin on mucoid cultures from directly plated sputum of cystic fibrosis patients chronically colonized with Pseudomonas aeruginosa, Drugs Exp Clin Res 16 (1990) 597-605.

[111] J. Tamaoki, K. Takeyama, I. Yamawaki, M. Kondo, K. Konno, Lipopolysac-charide-induced goblet cell hyper secretion in the guinea pig trachea: inhibition by macrolides, Am J Physiol Lung Cell Mol Physiol 272 (1997) L15-L19.

[112] H. Kobayashi, Biofilm disease: its clinical manifestation and therapeutic pos-sibilities of macrolides, Am J Med 99 (1995) 26S-30S.

[113] K. Tateda, Y. Ishii, T. Matsumoto, N. Furuya, M. Nagashima, T. Matsunaga, A. Ohno, S. Miyazaki, K. Yamaguchi, Direct evidence for antipseudomal activ-ity of macrolides: exposure-dependent bacterial activity and inhibition of protein synthesis by erythromycin, clarithromycin, and azithromycin, Antimicrob Agent Chemother 40 (1996) 2271-2275.

[114] A. Jaffer, A. Bush, The anti-inflammatory effects of macrolides in lung dis-ease: state of the art, Pediatr Pulmonol 31 (2001) 464-473.

[115] J. N. Snouwaert, K. K. Brigman, A. M. Latour, N. N. Malouf, R. C. Boucher, O. Smithies, B. H. Koller, An animal model for cystic fibrosis made by gene targeting, Science 257 (1992) 1083-1088.

[116] W. H. Colledge, B. S. Abella, K. W. Southern, R. A. Ratcliff, C. Jiang, S. H. Cheng, L. J. MacVinish, J. R. Anderson, A. W. Cuthberg, M. J. Evans, Genera-tion and characterization of a deltaF508 cystic fibrosis mouse model, Nat Genet 10 (1995) 445-452.

[117] B. R. Grubb, R. C. Boucher, Pathophysiology of gene-targeted mouse models for cystic fibrosis, Physiol Rev 79 (1999) S193-S214.

[118] B. R. Grubb, R. N. Vick, R. C. Boucher, Hyperabsorption of Na+ and raised Ca2+-mediated Cl- secretion in nasal epithelia of CF mice, Am J Physiol 266 (1994) C1478-C1483.

[119] B. R. Grubb, A. M. Paradiso, R. C. Boucher, Anomalies in ion transport in CF mouse tracheal epithelium, Am J Physiol 267 (1994) C293-C300.

[120] L. L. Clarke, B. R. Grubb, J. R. Yankaskas, C. U. Cotton, A. McKenzie, R. C. Boucher, Relationship of a non-cystic fibrosis transmembrane conductance regulator-mediated chloride conductance to organ-level disease in cftr (-/-) mice, Proc Natl Acad Sci USA 91 (1994) 479-483.

[121] B. R. Grubb, J. H. Jones, R. C. Boucher, Mucociliary transport determined by in vivo microdialysis in the airways of normal and CF mice, Am J Physiol Lung Cell Mol Physiol 286 (2004) L588-L595.

[122] R. A. Frizzell, J. M. Pilewski, Finally, mice with CF lung disease, Nat Med 10 (2004) 452-454.

[123] M. A. Mall, J. R. Herkema, J. B. Trojanek, D. Treis, A. Livraghi, S. Schubert, Z. Zhou, S. M. Kreda, S. L. Tilley, E. J. Hudson, W. K. O'Neal, R. C. Boucher, Development of chronic bronchitis and emphysema in �-Epithelial Na+ channel-overexpressing mice, Am J Respir Crit Care Med 177 (2008) 730-742.

[124] A. S. Verkman, From the farm to the lab: the pig as a new model of cystic fibrosis lung disease, Am J Physiol Lung Cell Mol Physiol 295 (2008) L238-L239.

[125] C. Guilbault, Z. Saeed, G. P. Downey, D. Radzioch, Cystic fibrosis mouse models, Am J Respir Cell Mol Biol 36 (2007) 1-7.

70

[126] C. S. Rogers, Y. Hao, T. Rokhlina, M. Samuel, D. A. Stoltz, Y. Li, E. Petroff, D. W. Vermeer, A. C. Kabel, Z. Yan, L. Spate, D. Wax, C. N. Murphy, A. Ricke, K. Whiworth, M. L. Linville, S. W. Korte, J. F. Engelhard, M. J. Welsh, R. S. Prather, Production of CFTR-null and CFTR-�F508 heterozygous pigs by adeno-associated virus-mediated gene targeting and somatic cell nuclear trans-fer, J Clin Invest 118 (2008) 1571-1577.

[127] S. T. Ballard, L. Trout, Z. Bebok, E. J. Sorscher, A. Crews, CFTR involve-ment in chloride, bicarbonate, and liquid secretion by airway submucosal glands, Am J Physiol Lung Cell Mol Physiol 277 (1999) L694-L699.

[128] S. T. Ballard, L. Trout, J. Garrison, S. K. Inglis. Ionic mechanism of forskolin-induced liquid secretion by porcine bronchi, Am J Physiol Lung Cell Mol Physiol 290 (2006) L97-L104.

[129] D. B. Salinas, N. Pedemonte, C. Muanprasat, W. F. Finkbeiner, D. W. Nielson, A. S. Verkman, CFTR involvement in nasal potential differences in mice and pigs studied using a thiazolidinone CFTR inhibitor, Am J Physiol Lung Cell Mol Physiol 287 (2004) L936-L943.

[130] J. R. Thiagarajah, Y. Song, P. M. Haggie, A. S. Verkman, A small molecule CFTR inhibitor produces cystic fibrosis –like submucosal gland fluid secretions in normal airways, FASEB J 18 (2004) 875-877.

[131] C. S. Rogers, W. M. Abraham, K. A. Brogden, J. F. Engelhardt, J. T. Fisher, P. B. Jr. McCray, G. McLennan, D. K. Meyerholz, E. Namati, L. S. Ostedgaard, R. S. Prather, J. R. Sabater, D. A. Stoltz, J. Zabner, M. J. Welsh, The porcine lung as a potential model for cystic fibrosis. Am J Physiol Lung Cell Mol Physiol 295 (2008) L240-L263.

[132] L. S. Ostedgaard, C. S. Rogers, Q. Dong, C. O. Randak, D. W. Vermeer, T. Rokhlina, P. H. Karp, M. J. Welsh, Processing and function of CFTR-Delta508 are species-dependent, Proc Nat Acad Sci USA 104 (2007) 15370-15375.

[133] Y. Liu, Y. Wang, Y. Jiang, N. Zhu, H. Liang, L. Xu, X. Feng, H. Yang, T. Ma, Mild processing defect of porcine �F508-CFTR suggests that �F508 pigs may not develop cystic fibrosis disease, Biochem Biophys Res Commun 373 (2008) 113-118.

[134] L. G. Johnson, J. C. Olsen, B. Sarkadi, K. L. Moore, R Swanstrom, R. C. Boucher, Efficiency of gene transfer for restoration of normal airway epithelial function in cystic fibrosis, Nat Genet 2 (1992) 21-25.

[135] P. French, J. H. Doorninck, R. H. P. C. Peters, E. Verbeek, N. A. Ameen, C. R. Marino, H. R. Jorge, J. Bijman, B.J. Scholte, A �F508 mutation in mouse cystic fibrosis transmembrane conductance regulator results in a temperature-sensitive processing defect in vivo, J Clin Invest 98 (1996) 1304-1312.

[136] A. L. Cozens, M. J. Yezzi, K. Kunzelmann, T. Ohrui, L. Chin, K. Eng, W. E. Finkbeiner, J. H. Widdicombe, D. C. Gruenert, CFTR expression and chloride secretion in polarized immortal human bronchial epithelial cells, Am J Respir Cell Mol Biol 10 (1994) 38-47.

[137] S. K. Inglis, M. R. Corboz, S. T. Ballard, Effect of anion secretion inhibitors on mucin content of airway submucosal gland ducts, Am J Physiol 274 (1998) L762-L766.

[138] S. Jayaraman, Y. Song, A. S. Verkman, Airway surface liquid pH in well-differentiated airway epithelial cell cultures and mouse trachea, Am J Physiol Cell Physiol 281 (2001) C1504-C1511.

[139] T. Ma, J. R. Thiagarajah, H. Yang, N. D. Sonawane, C. Folli, L. J. Galietta, A. S. Verkman, Triazolidinone CFTR inhibitor identified by high-throughput screening blocks cholera toxin-induced intestinal fluid secretion, J Clin Invest 110 (2002) 1651-1658.

71

[140] G. M. Roomans, Quantitative X-ray microanalysis of biological specimens, J Electron Microsc Tech 9 (1988) 19-43.

[141] V. Vanthanouvong, G. M. Roomans, Methods for determining the composi-tion of nasal fluid by X-ray microanalysis, Microsc Res Tech 63 (2004) 123-128.

[142] B. O. Nilsson, L. Ljung, Electron probe micro-X-ray analysis of electrolyte composition of fluid microsamples by use of Sephadex bead, Ups J Med Sci 84 (1979) 1-2.

[143] B. O. Nilsson, L. Ljung, X-ray micro analysis of cations (Na, K, Ca) and ani-ons (S, P, Cl) in uterine secretions during blastocyst implantation in the rat, J Exp Zool 234 (1985) 415-421.

[144] H. Nilsson, I. Kozlova, V. Vanthanouvong, G. M. Roomans, Collection and X-ray microanalysis of airway surface liquid in the mouse using ion exchange beads, Micron 35 (2004) 701-705.

[145 ] P. H. Quanjer, Standardized lung function testing, Report from Working Party “Standardisation of Lung Function Tests”, European Community for Coal and Steel, Bull Eur Physiopathol Respir 19 (1983) 45-51.

[146] M. Granström, A. Ericsson, B. Strandvik, B. Wretlind, O. R. Pavlovskis, R. Berka, M. L. Vasil, Relation between antibody response to Pseudomonas aeruginosa exoproteins and colonization/infection in patients with cystic fibro-sis, Acta Pediatr Scand 73 (1984) 772-777.

[147] T. N. Hilliard, J. Zhu, R. Farley, S. Escudero-Garcia, B. J. Wainwright, P. K. Jeffery, U. Griesenbach, A. Bush, J. C. Davies, E. W. Alton, Nasal abnormali-ties in cystic fibrosis mice independent of infection and inflammation, Am J Respir Cell Mol Biol 39 (2008) 19-25.

[148] S. J. B. Reed, Electron Microprobe Analysis, Cambridge, UK, Cambridge Univ., (1975).

[149] V. Vanthanouvong, I. Kozlova, G. M. Roomans, Ionic composition of rat airway surface liquid determined by X-ray microanalysis, Microsc Res Tech 68 (2005) 6-12.

[150] H. K. Choi, W. E. Finkbeiner, J. H. Widdicombe, A comparative study of mammalian tracheal mucous glands, J Anat 197 (2000) 361-372.

[151] P. R. Durie, G. Kent, M. J. Phillips, C. A. Ackerley, Characteristic multiorgan pathology of cystic fibrosis in a long-living cystic fibrosis transmembrane regu-lator knockout murine model, Am J Pathol 164 (2004) 1481-1493.

[152] R. E. Meyer, R. E. Fish, A review of tribromoethanol anesthesia for produc-tion of genetically engineered mice and rats, Lab Anim 34 (2005) 47-52.

[153] J. Silverman, Protocol review, Anesthetic in GEM: does TBE make the grade? Lab Anim 32 (2003) 19-21.

[154] D. Tarin, A. Sturdee, Surgical anesthesia of mice: evaluation of tribromo-ethanol, ether, halothane and methoxylflurane and development of a reliable technique, Lab Anim 6 (1979) 79-84.

[155] C. Ehrhardt, C. Friegel, U. F. Schaefer, K. J. Kim, C-M. Lehr, Influence of apical fluid volume on the development of functional intercellular junctions in the human epithelial cell line 16HBE14o-: implications for the use of this cell line as an in vitro model for bronchial drug absorption studies, Cell Tissue Res 308 (2002) 391-400.

[156] C. Andersson, B. Gaston, G. M. Roomans, S-Nitrosoglutathione induces func-tional �F508-CFTR in airway epithelial cells, Biochem Biophys Res Commun 297 (2002) 552-557.

72

[157] C. Andersson, J. Servetnyk, G. M. Roomans, Activation of CFTR by genistein in human airway epithelial cell lines, Biochem Biophys Res Commun 308 (2003) 518-522.

[158] M. M. Reddy, M. J. Light, P. M. Quinton, Activation of the epithelial Na+ channel (ENaC) requires CFTR Cl- channel function, Nature 402 (1999) 301-304.

[159] K. Kunzelmann, R. Schreiber, R. Nitschke, M. Mall, Control of epithelial Na+ conductance by the cystic fibrosis transmembrane conductance regulator, Pflügers Arch Eur J Physiol 440 (2000) 193-201.

[160] S. D. Anderson, E. Daviskas, The mechanism of exercise-induced asthma is…, J Allergy Clin Immunol 106 (2000) 453-459.

[161] I. Sermet-Gaudelus, M. Dechaux, B. Vallee, A. Fsjac, E. Girodon, T. Nguyen-Khoa, S. Marianovski, I. Hurbain, J. L. Bresson, G. Lenoir, A. Edelman, Chlo-ride transport in nasal ciliated cells of cystic fibrosis heterozygotes, Am J Respir Crit Care Med 171 (2005) 1026-1031.

[162] C. A. Demko, P. J. Byard, P. B. Davis. Gender differences in cystic fibrosis: Pseudomonas aeruginosa infection, J Clin Epidemiol 48 (1995) 1041-1049.

[163] C. Moser, P. O. Jensen, H. Calum, T. Bjarnsholt, H. Hougen, N. Hoiby, M. Johannesson, Gender differences during chronic Pseudomonas aeruginosa lung infection in mice, Pediatr Pulmonol 38 (2004) 272-273.

[164] K. Felgentreff, C. Beisswenger, M. Griese, T. Gulder, G. Bringmann, R. Bals, The antimicrobial peptide cathelicidin interacts with airway mucus, Peptides 27 (2006) 3100-3106.

[165] P. K. Singh, B. F. Tack, P. B. Jr McCray, M. J. Welsh, Synergistic and addi-tive killing by antimicrobial factors found in human airway surface liquid, Am J Physiol Lung Cell Mol Physiol 279 (2000) L799-L805.

[166] M. Pazgier, D. M. Hoover, D. Yang, W. Lu, J. Lubkowski, Human �-defensins, Cell Mol Life Sci 63 (2006) 1294-1313.

[167] T. D. Starner, C. K. Barker, H. P. Jia, Y. Kang, P. B. Jr. McCray, CCL20 is an inducible product of human airway epithelia with innate immune properties, Am J Respir Cell Mol Biol 29 (2003) 627-633.

[168] P. H. Gilligan, Microbiology of airway disease in patients with cystic fibrosis, Clin Microbiol, 4 (1991) 35-51.

[169] J. Gawel, J. Rudnik, J. Pryjma, J. Zebrak, J. Haluszka, I. Rudnik, H. Ma-jewska-Zalewska, Protein in bronchial secretion of children with chronic pul-monary disease: I. Relation to clinical diagnosis, Scand J Respir Dis 60 (1979) 63-68.

[170] J. Zebrak, T. Herman, R. Werys, J. Pryjma, J. Gawel, Proteins in bronchial secretion of children with chronic disease: II. Relation to bronchoscopic and bronchographic examination, Scand J Respir Dis 60 (1979) 69-75.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 393

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-9358

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2008

studies on airway surface liquid in connection with cystic...

Documents