University of Groningen Cystic fibrosis liver disease …...Cystic fibrosis liver disease and the...
Transcript of University of Groningen Cystic fibrosis liver disease …...Cystic fibrosis liver disease and the...
University of Groningen
Cystic fibrosis liver disease and the enterohepatic circulation of bile acidsBodewes, Frank
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2014
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Bodewes, F. (2014). Cystic fibrosis liver disease and the enterohepatic circulation of bile acids. [S.l.]: [S.n.].
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 05-09-2020
CYSTIC FIBROSIS LIVER DISEASE
AND THE ENTEROHEPATIC
CIRCULATION
OF BILE ACIDS
FRANCISCUS ALBERTUS JOHANNES ANDREAS BODEWES
The work described in this thesis was performed at the Department of Pediatrics, Center for Liver,
digestive and Metabolic diseases, Beatrix Children’s Hospital, University Medical Center Groningen, the
Netherlands
The author gratefully acknowledges the unrestricted financial support for printing of this thesis by:
ABBOTT B.V.
ACTELION PHARMACEUTICALS NEDERLAND B.V.
DR FALK PHARMA BENELUX B.V.
NUTRICA BABY- EN KINDERVOEDING
TRAMEDICO
RIJKSUNIVERSITEIT GRONINGEN
Cover: Tiger and Turtle Landmarke Angermund - Duisburg
Boek (Book) ISBN: 978-90-367-6828-3
Eboek : PDF zonder DRM (PDF without DRM) ISBN: 978-90-367-6827-6
Copyright © F.A.J.A. Bodewes, 2014
All rights reserved, No part of this thesis may be reproduced, distributed, stored in a retrieval system, or
transmitted in any form or by any means, without prior permission of the author ,or when appropriate,
of the publisher of the publications
Cystic fibrosis liver disease and the enterohepatic circulation of bile acids
Proefschrift
ter verkrijging van de graad van doctor aan de
Rijksuniversiteit Groningen
op gezag van de
rector magnificus, prof. dr. E. Sterken
en volgens besluit van het College voor Promoties.
De openbare verdediging zal plaatsvinden op
woensdag 19 maart 2014 om 16.15 uur
door
Franciscus Albertus Johannes Andreas Bodewes
geboren op 14 februari 1966
te Eelde-Paterswolde
Promotor:
Prof. dr. H.J. Verkade
Copromotor:
Dr. H.R. de Jonge
Beoordelingscommissie:
Prof. dr. G.H. Koppelman
Prof. dr. E.E.S. Nieuwenhuizen
Prof. dr. R.J. Porte
Paranimfen:
Bertjan Bodewes
Bas Morselt
Ik draag dit proefschrift op aan mijn lieve ouders
Jan Bodewes en Gerry Bodewes-Valens
CONTENTS OF THIS THESIS
CHAPTER 1
General introduction
CHAPTER 2
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
Published
CHAPTER 3
Increase of serum gamma glutamyltransferase (GGT) associated with the development of cirrhotic cystic fibrosis liver disease
Submitted
CHAPTER 4
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice.
Submitted
CHAPTER 5
Bile salt metabolism in a CF mice model with spontaneous liver disease
Submitted
CHAPTER 6
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
Published
CHAPTER 7
General discussion and summary
CHAPTER 8CHAPTER 8
Nederlandse discussie en samenvatting
Curriculum Vitae
Dankwoord- acknowledgements
CHAPTER 1
GENERAL INTRODUCTION
Chapter 1
10
1
1) THE CLINICAL PERSPECTIVE
CYSTIC FIBROSIS DISEASE
Cystic Fibrosis (CF) is a severe, lifespan limiting, disease. Cystic fibrosis is one of the most
frequent autosomal inherited diseases in the world. The incidences differ globally according
to regional genetic variations (1). In the Netherlands around 1:1500-6000 inhabitants suffer
from CF (2, 3). The disease is usually already manifest at birth and progresses with age. To
date the median survival of CF patients is around 40 years (4). Most patients die from end
stage lung disease. Although severe lung disease dominates the clinical picture, CF is a multi-
organ disease. In particular diseases of intestine, pancreas and liver can be serious and
potentially life threatening (figure 1.).
In 1989, the CF disease causing gene was identified and named “cystic fibrosis transport
regulator” (CFTR). The CFTR gen encodes for the CFTR protein (5, 6). The CFTR gene is
localized on chromosome 7(7). Cystic fibrosis can be caused by a variety of mutations in the
CFTR gene. To date, over 1900 different mutations in the CFTR gene, are identified (8). The
CFTR gene mutation can be divided into separate groups. A mutation in one copy of the CFTR
gene that always causes CF, as long as it is paired with another CF-causing mutation in the
other copy of the CFTR gene, is a CF-causing mutation. A mutation in one copy of the CFTR
gene that does NOT cause CF, even when it is paired with a CF-causing mutation in the other
copy of the CFTR gene, is a non CF-causing mutation. A mutation that may cause CF, when
paired with CF-causing mutation in the other copy of the CFTR gene, is a mutation of varying
clinical consequence. A mutation for which we do not have enough information to determine
whether or not it falls into the other three categories is a mutation of unknown significance
(9).
The disease causing mutations can be divided into 5 different mutations classes according to
the type of malfunction of the CFTR protein (11). Based on this classification, difference in
clinical disease presentation and severity can be recognized (figure 2). The various mutation
classes offer different potential therapeutic targets for the treatment of Cystic fibrosis (12).
CF can develop if a person carries a disease causing mutations in each CFTR allele. The most
common CFTR mutation in humans is the 508del gene variation (13). In the 508del mutation
one nucleotide T, on the 508 position of the gene, is replaced by a G nucleotide. 508del is a,
so called, class 2 mutation. These class 2 mutations cause an almost complete failure (less
than 5% of normal) of the CFTR function, leading to the typical severe phenotype with
General introduction
11
1 1
progressive pulmonary disease, complete exocrine pancreatic insufficiency, diabetes and
cirrhosis (14).
CFTR is a cell membrane protein localized in various cell and in particularly in all epithelial
tissues. The protein can, for instance, be found in the alveolar cells of the airways, the
ductular cells of the pancreas, the enterocytes of the intestine, the cholangiocytes of the bile
ducts but also in the sweat glands of the skin(15). Mutations in the CFTR gene are directly
responsible for the symptomatology and disease development in these organs.
Figure 1. Cystic fibrosis is a multi-organ disease. Because so many bodily functions rely on normal water flow, a
disruption in water flow can cause a number of devastating effects, as shown in the "Manifestations of Cystic
Fibrosis" image above (illustration adapted from: Wikimedia Commons, 2011)(10). In particular diseases of intestine,
pancreas and liver can be serious and potentially life threatening.
In epithelial cells, CFTR functions as a chloride ion (Cl-), trans-membrane transporter protein
(17). CFTR actively pumps Cl- across the cellular membrane by concomitant with ATP
hydrolysis. The Cl- transport serves various roles in the different epithelia. In the lungs, the Cl
-
transport induces a Cl-ion gradient across the cell membrane (18). Based on this gradient,
Chapter 1
12
1
bicarbonate (HCO3-) is passively exchanged against Cl
- and transported out of the alveolar cell
in to alveolar lumen. Here HCO3- here serves a crucial role in maintaining viscosity and fluidity
the alveolar fluid layer. Disturbance of this HCO3- transport function causes a thick, highly
viscous fluid layer in the alveoli. The sticky secretion impairs lung mucus clearance and
thereby an increased susceptibility for bacterial pulmonary infections in CF disease. CFTR
functions as a Cl- channel in different epithelia (19). However, the contribution of CFTR
channel dysfunction, in disease development in different organs, for example, intestine and
liver, is not clear.
Figure 2. Molecular consequences of CFTR mutations. a, CFTR correctly positioned at the apical membrane of an
epithelial cell, functioning as a chloride channel. b, Class I. No CFTR messenger ribonucleic acid or no CFTR protein
formed (e.g., nonsense, frameshift, or splice site mutation). c, Class II. Trafficking defect. CFTR messenger ribonucleic
acid formed, but protein fails to traffic to cell membrane. d, Class III. Regulation defect. CFTR reaches the cell
membrane but fails to respond to cAMP stimulation. e, Class IV. Channel defect. CFTR functions as altered chloride
channel. f, Class V. Synthesis defect. Reduced synthesis or defective processing of normal CFTR. Chloride channel
properties are normal (Illustration from The Journal of Pediatrics, 127, 5, 1995)(16)
General introduction
13
1 1
GASTRO-INTESTINAL AND HEPATIC DISEASE IN CYSTIC FIBROSIS
CF presents with clinical symptomatology in various abdominal organs like pancreas, intestine
and liver (20). The clinical phenotype and contribution of different organ systems varies in CF
patients (21). In cystic fibrosis the clinical, organ specific, presentation and phenotypical
penetration depend on gene modifiers. Gene modifiers are genetic or environmental factors
that determine the actual organ specific phenotype. Different disease presentations in CF are
in greater or lesser extent dependent on the influence of gene modifiers (figure 3.). The most
important and relevant gastro-intestinal disease presentations in CF are exocrine pancreatic
insufficiency and cirrhosis (22, 23). However, also specific intestinal diseases, like neonatal
meconium ileus, the distal intestinal obstruction syndrome (DIOS) and intestinal fat- and bile
salt malabsorption, are frequently found CF patients (24-26).
Figuur 3. The relative contribution of modifier genes, CFTR, and environment on phenotype. [Adapted from Borowitz
et al.]1Borowitz D et al. Gastrointestinal outcomes and confounders in cystic fibrosis. J Pediatr Gastroenterol Nutr.
41:273-85 (2005)
EXOCRINE PANCREATIC INSUFFICIENCY AND FAT MALABSORPTION IN
CYSTIC FIBROSIS
Intestinal fat malabsorption is a serious clinical manifestation of CF. The decreased absorption
of dietary fats impairs the development and maintenance of a healthy nutritional status in
particular in growing children (27). In CF nutritional status relates to the prognosis and
Chapter 1
14
1
survival (28). Therefore, high caloric feeding and treatment of the intestinal fat malabsorption
are cornerstones in the treatment of cystic fibrosis (29).
The leading cause of the intestinal fat malabsorption in CF is exocrine pancreatic insufficiency
(PI)(22). CF causes fibrotic degeneration of the acinar tissue of the pancreas secondary to
destruction of the ductular structures due to loss of CFTR function (30). The fibrotic pancreas
is no longer able to excrete pancreatic enzymes including lipases and proteases essential for
intestinal fat and protein absorption. The pancreatic destruction, partly based on auto-
digestion, starts already in utero and, in most patients with a severe genotype, this develops
into complete PI already during infancy (31).
PI is treated with pancreas enzyme replacement therapy (PERT) (32). These products contain
pancreatic enzymes mostly of animal origin (33). However, bioengineered products are
currently developed and coming to the market (34). PERT is individually dosed based on the
dietary fat intake and its effects on intestinal fat absorption. Despite optimizing and
maximizing PERT many patients a degree of fat malabsorption persists (35, 36).
LIVER DISEASE IN CYSTIC FIBROSIS
The earliest form of liver involvement in CF is neonatal cholestasis (37). Infants can present
with prolonged jaundice and vitamin K dependent coagulopathy. Liver histology displays signs
of biliary obstruction, portal fibrosis and inflammation with bile duct proliferation. Mucous
plugs in bile ducts are described (38). The pathogenesis of CF related neonatal cholestasis is
not known, but it might be related to temporary biliary obstruction in the neonatal period.
The disease is mostly self-limiting and does not seem to be related to the development of
severe liver CF related liver disease later in life. To date CF related neonatal cholestasis is
probably earlier recognized in countries with a neonatal screenings program for CF (39).
CF patients frequently show signs of hepatic steatosis or fatty liver (40). Steatosis is often
diagnosed based during routine ultrasonography of the liver in the clinical follow up of CF
patients. An elevated fat fraction has also been observed in over 80% of subjects with cystic
fibrosis on MRI scanning (41). Histologically proven steatosis can be found in up to 35% of CF
patients (42). There is no direct correlation between steatosis and the later development of
cirrhotic liver disease (43). Steatosis in infants with CF is often suggested being related to
poor nutritional status in particular early or late recognized disease (44). It has further been
suggested that steatosis in CF has to do with to essential fatty acid deficiency due to the CF
related intestinal fat malabsorption (45).
Cholelithiasis or bile stone disease is frequent in cystic fibrosis patients (46). Asymptomatic
stones are often seen on routine ultrasonography studies (47, 48). However, CF patients do
General introduction
15
1 1
regularly present with symptomatic cholelithiasis that necessitate cholecystectomy (46).
Cystic fibrosis related cholelithiasis does not usually respond to non-lithogenic treatment like
for instance ursodeoxycholic acid (UDCA) probably because cholesterol is not the main
component of stone or sludge (49). It is hypothesized that cholelithiasis in CF is related to
CFTR dependent alterations in the biliary bile composition. However, the exact pathogenesis
of the susceptibility of CF patients for bile stone disease has remained unknown (50).
During routine laboratory checkups in CF patients often elevations of liver enzymes (AST, ALT
and GGT) are found (51). If these laboratories abnormalities persist, in repeated
measurements, they are sometime classified as signs of CF related liver disease (52).
Persistent elevation of liver enzymes above two times the upper limit of normal are suggested
as an indication to start UDCA as potential treatment option (53). However, the predictive
value and the relation of elevation of liver enzymes to the presence of relevant liver disease
has remained a subject of controversy (54).
CIRRHOSIS IN CYSTIC FIBROSIS
The most severe hepatic complication in CF is the development of hepatic fibrosis into
cirrhosis. This potentially life threatening liver disease develops in about 10% of the CF
patients (40). The clinical presentation is dominated by symptoms of portal hypertension such
as splenomegaly, hypersplenism, gastrointestinal variceal disease and bleeding (42, 53). Less
frequent are ascites and hepatopulmonary syndrome found in CF related cirrhosis (CCFLD)
(55). The liver parenchymal functions, including protein synthesis and detoxification, are
usually spared (4, 56).
The disease is not yet clinically present at birth and develops during childhood. The majority
of patients have developed clinically manifest cirrhosis before adulthood with a peak
incidence around the age of 10 years (57). Although CF related cirrhosis is rapidly progressive
during childhood, the disease tends to stabilize into adulthood. Cirrhotic decomposition is a
rare event and liver transplantation while feasible in CF patients, is relatively rarely indicated
(56).
The clinical diagnosis of CCFLD is generally made on the basis of multi nodular irregular aspect
of the liver on ultrasound in combination with the presence of splenomegaly and signs of
hypersplenism including thrombocytopenia (58). Liver biopsy can be used for histological
diagnosis of cirrhosis (59). Since CCFLD can be localized in a segmental manner, liver biopsy
poses a risk for sampling error. Therefore, it has been advised to perform multiple biopsies in
patients suspected of CCFLD (60). For the clinical situation, it is not necessary to perform liver
biopsy in the situation of established cirrhosis. However to evaluate developing fibrosis for
diagnostic of therapeutic studies biopsy is probably needed to serve as the gold standard.
Chapter 1
16
1
To date ursodeoxycholic acid (UDCA) is the only medical treatment used in CF related liver
disease. UDCA is an endogenous, relatively hydrophilic bile salt with choleretic and
antifibrotic properties (61). UDCA is given to CF patients with persistently increased liver
enzymes during routine laboratory follow up and/or hepatomegaly (53). It is proven that
UDCA is capable of reversing liver enzyme elevation (62). However, the benefit of UDCA in the
treatment or prevention of CCFLD is not known (63). Therefore, the use of UDCA in CF
remains subject of discussion (64).
THE ROLE OF CFTR IN CYSTIC FIBROSIS RELATED LIVER DISEASES.
Despite the fact that, in recent years, there has been a rapid increase in the knowledge on CF
and CFTR protein function, the pathogenesis of CCFLD remains unknown. To improve the
treatment and clinical outcome of CCFLD a more fundamental understanding of the
mechanisms causing the disease is crucial.
Proliferation and destruction of the bile ducts is are prominent histological features of CCFLD
(65, 66). In the past, these observations lead to the assumption that biliary obstruction lays at
the basis of the disease (67). In CF lung disease, due to CFTR malfunction, the mucus in the
alveoli is thick and sticky. In the liver, CFTR is exclusively expressed in the cholangiocytes or
bile duct cells. In parallel to the lung disease, it was assumed that the observed bile duct
obstruction was caused by thick and sticky bile obstructing the bile ducts (68). This hypothesis
formed one of the substantiations to try the choleretic bile acid UDCA as a treatment in CFLD
(69). It was believed that UDCA would make the bile more fluid and increase the bile flow,
thereby preventing the bile duct obstruction. However, it was never been shown that indeed
the increased viscosity of bile in CF is the primary cause of the disease.
It is not known why only up to 10% of CF patients develop CCFLD. It has been demonstrated
that only patients with a severe phenotype, including pancreatic insufficiency, can develop
CCFLD (70). However, within the group of CF patients with a severe phenotype the risk for
developing CCFLD is not “further” genotype related. It is assumed that additional genetic risk
factors or external factors have co-responsibility for the development of CCFLD (71).
Extensive genetic modifiers studies in patients with CCFLD have shown for instance that
carrier ship of the Pi allele for alpha-1 antitrypsin deficiency adds to risk for CCFLD (57).
Another striking and yet unexplained clinical phenomenon of CCFLD is the young and distinct
age of presentation of the disease. CCFLD is not present at birth. In most patients, cirrhosis
develops before the age of 18 years with a peak incidence of about 10 years (40). In
adulthood, hardly any new case of cirrhosis develops. It is known from historical autopsy
studies that liver fibrosis and biliary obstruction are relatively common in adult CF patients
(72). However not all patients did eventually develop clinical manifest cirrhosis. To date we do
General introduction
17
1 1
not know what the risks factors are that trigger the disease to start development in children.
Most CCFLD patients have a severe form of portal hypertension with splenomegaly. Gastro-
intestinal variceal disease is common and in particular young patients are at high risk for
hemorrhages (58, 73). This risk for variceal bleeding diminishes with age and is less common
in adult CF patients (74).
2) THE ROLE OF BILE SALT IN CYTOTOXICITY AND BILE FLOW
A potential mechanism for the biliary liver disease in CF involves the concept of bile salt
related cytotoxicity. Before discussing the contribution of bile salts in pathology, the
physiological roles of bile salts will be addressed.
THE ROLE OF BILE SALT IN DIETARY FAT DIGESTION
Bile salts are polarized steroids that play a vital role in intestinal fat absorption and biliary
disposal of endogenous and exogenous compounds (75). In the intestine bile salts function as
essential surfactants used to solubilize dietary fats in the hydrophilic milieu of gut (76).
However, based on the same properties, bile salts can also act as detergents to liver tissue,
for example, in the situation of bile salt accumulation (cholestasis) (77). Corresponding with
their potentially toxic effects to cells, the bile salt synthesis and their concentrations are
tightly regulated (78).
Bile salts are synthesized in the hepatocytes from cholesterol. Bile salts are excreted into the
bile and transported, to the intestine, via the intra- and extrahepatic bile ducts. In the bile and
the gut, bile salts form water-solvable aggregates, so called micelles, together with the fatty
acids originating from the dietary fats. The formation of micelles is essential to transport the
dietary fats towards the enterocytes across the aqueous intestinal lumen. Absence of bile
salts in the gut results in severe intestinal fat malabsorption. In recent years, it has become
clear that bile salts are not only involved in dietary food digestion. It has been shown that bile
salts also play vital roles a variety of systemic metabolic regulatory processes, which are,
however, outside the scoop of this thesis (79).
Chapter 1
18
1
THE ENTEROHEPATIC CIRCULATION OF BILE SALTS
Bile salts are efficiently recycled via the portal system back to the liver in the so called
enterohepatic circulation (80). Bile salts are to a large extent (>95%), absorbed in the terminal
ileum, the final section of the small intestine. The total amount of bile salts in the body is
balanced and is kept in a tight, steady state, (81). Under steady state conditions, the fecal loss
of bile salts is entirely compensated by de novo bile salt synthesis of primary bile salts in the
liver. The primary bile salts in humans are cholate and chenodeoxycholate. The primary bile
salts are excreted via de bile into the intestine. In the intestinal lumen, the bile salts can be
metabolized by the gut flora. Bacteria are capable of deconjugating bile salts and
transforming them into a variety of different secondary bile salts. Bile salt species are
amphipathic molecules with a hydrophilic and a hydrophobic domain. Bile salts differ in their
water solubility and their hydrophobic-hydrophilic balance.
Hydrophobic bile salts have a high capability for solubilizing fats and lipids (82). As a result,
hydrophobic bile salts also have the ability to solubilize the lipid structures of cell membranes.
Therefore, the more hydrophobic a bile salt, the higher their detergent cytotoxicity for cells
and tissues exposed to them.
Under physiological conditions, the hydrophobic cytotoxicity of biliary bile salts is limited by
co-secretion of phospholipids (and cholesterol), leading to the formation of mixed micelles
(83). Phospholipids are excreted into bile by the membrane transporter enzyme MDR3 (Mdr2
in mice). A genetic incapacity to excrete phospholipids into the bile results in severe bile duct
destruction and biliary liver disease. In humans, an inactivating MDR3 mutation is the basis
for the disease “progressive familiar cholestatic disease type 3 (PFIC 3)” (84). The disease
presents with severe biliary cirrhosis and cholestasis, usually at child age. A genetic mouse
model without functional Mdr 2 protein expression (Mdr2 knockout mice) spontaneously
develops biliary disease, similar to human PFIC 3 patients (85). The bile duct destruction in
Mdr2 knockout mice can be even augmented by increasing the hydrophobicity of the bile salt
pool via administration of the hydrophobic bile salt cholate to the mice (86).
BILE PRODUCTION AND BILE FLOW
The magnitude of bile flow is determined at two levels in the biliary tract. The first level is the
so called canalicular lumen, the smallest intercellular biliary domain between hepatocytes.
Canaliculi form the start of the bile ducts and are surrounded by hepatocytes. When
hepatocytes actively transport bile salt into the canalicular lumen, water passively follows as a
result of the osmotic activity of the bile salts. The amount of water transport that is generated
via bile salt osmosis is called the bile salt dependent bile flow (87).
General introduction
19
1 1
Various bile salt species differ in their choleretic capacity, i.e the capability to induce bile flow
(88). The choleretic capacity of bile salt depends on the molecular structure and their
hydrophobicity. Several hydrophilic bile salts are capable of inducing an exceptionally high
bile salt dependent bile flow. For this choleretic capacity some bile salts, like for instance
UDCA, are used in clinical situations were decreased bile flow or bile duct obstruction is
hypothesized to be contributing to the biliary disease (89).
The second level at which bile flow is determined are the bile ducts. Cholangiocytes form the
lining of the bile ducts. Cholangiocytes are capable of secreting water and e.g. bicarbonate
into the bile, thereby increasing bile flow and diluting the (canalicular) bile content (90). This
portion of the bile production is called the ductular bile flow. Different ion and anion
membrane transporter proteins are involved in the complex mechanism of water secretion
over the apical membrane by cholangiocytes (91).
The actual driving force behind this process of water secretion of cholangiocytes is the active
Cl- transport into the bile duct lumen (92). The most important Cl
- transporter is CFTR, but
other Cl- channels have also been described (93-95). The active Cl
- transport creates a Cl
-
gradient over the apical membrane. The luminal Cl- is then exchanged for bicarbonate (HCO3
-)
via a protein called Anion exchanger-2. In this way, the Cl- gradient is replaced by a HCO3
--
osmotic gradient. Water leaves the cholangiocytes, via water channels or so called aquaporins
on the basis of this osmotic HCO3- gradient. In this manner, the water secretion by
cholangiocytes is directly related to the Cl- transport capacity and CFTR function (96)
The chloride transport of CFTR is an active ATP consuming, process (97). CFTR belongs to the
extensive family of ATP binding cassette membrane proteins (98). The CFTR chloride channel
only opens after binding and hydrolysis of the intracellular energy source ATP (99). The ATP
binding necessary for CFTR activation is regulated via the protein cycle AMP (c-AMP). In the
cholangiocytes, c-AMP activity is controlled via different pathways. The most important
stimulating pathway of c-AMP is the hormone secretin (100). This gastro-intestinal hormone
is released during feeding and in this manner influences the bile flow rate. c-AMP can also be
induced via intraluminal factors like bile salts or intraluminal ATP (94, 101).
BILE SALT SYNTHESIS
Bile salts are synthesized in the hepatocytes from cholesterol (figure 4). The synthesis
requires several sequential enzymatic steps (78). The synthesis of the bile acids is
quantitatively the predominant pathway of cholesterol catabolism in mammals. The major
pathway for the synthesis of the bile acids is initiated via hydroxylation of cholesterol at the 7
position via the action of cholesterol 7α-hydroxylase (CYP7A1), an ER localized enzyme.
CYP7A1 is a member of the cytochrome P450 family of metabolic enzymes. This pathway
Chapter 1
20
1
initiated by CYP7A1 is referred to as the "classic" pathway of bile acid synthesis. There is an
alternative pathway via the mitochondrial enzyme sterol 27-hydroxylase (CYP27A1).
Although, in rodents the alternative pathway can account for up to 25% of total bile acid
synthesis, in humans it has been suggested to account for no more than 6% (102).
Figure 4. Schematic representation of the bile salt synthesis pathways. (Illustration from Lipid research 2009;50:s120-
s125)(102)
General introduction
21
1 1
THE REGULATION OF BILE SALT SYNTHESIS
Bile salts have an important physiological contribution to the intestinal fat digestion, besides
elimination of cholesterol and other endogenic and exogenic catabolytes and metabolytes
from the body. Bile salt synthesis is under the control of at least two different negative
feedback pathways (figure 5.). In the first pathway, the sensing of bile salt takes place in the
hepatocytes (103). In the other pathway, bile salt sensing takes place at the level of the
enterocytes of the gut. In both pathways binding of bile salt to Farnesoid X receptor (FXR)
plays a pivotal role in the initiation of the negative feedback regulation. FXR is a nuclear
receptor. Nuclear receptors are transcription factors, whose ligands can determine their DNA
binding and transcription modulating activity. The ligands for the nuclear receptors are,
among others, steroids. In response, these nuclear receptors work with other proteins to
regulate expression of specific genes, thereby controlling and regulating homeostasis in an
organism. In the case of the nuclear receptor FXR bile salt can function as ligands.
Figure 5. Schematic representation of the enterohepatic circulation of bile salts in the context of CFTR. Several bile
salt involving feedback loops regulated bile salt synthesis and liver growth. The bile salt induced nuclear receptor FXR
plays a pivotal role in intestine and in the liver.
Chapter 1
22
1
However different bile salt species vary in their efficacy as FXR ligands (104). In general
hydrophobic bile salts are stronger ligands for FXR compared to hydrophilic bile salts.
Bile salt/FXR interaction results in different physiological responses, in the different cell types,
for example, hepatocytes and intestinal cells. In hepatocytes FXR activation down regulates
the expression of a protein called small heterodimer protein or SHP. SHP can directly down
regulates the expression of CYP7A the rate limiting step in the bile salt synthesis from
cholesterol (105).
The intestinal regulation of bile salt synthesis is more indirect: FXR stimulates the expression
of FGF19 (fibroblast growth factor 19, equivalent to Fgf15 in rodents) that is subsequently
released into the portal blood (106). At the basolateral membrane of the hepatocytes FGF19
can bind to the FGFR4 receptor (107). This binding leads to activation of an intracellular
pathway that, via the so called ERK system, down regulates CYP7A expression and
subsequently reduces bile salt synthesis (108). The hepatic and intestinal sensing of bile salts
by FXR and their respectively negative feedback pathways are functionally independent.
3) THE ROLE OF CFTR IN GASTROINTESTINAL DISEASE.
Different from other organs like lungs and sweat gland, the contribution of CFTR in the
intestine to the clinical phenotype is not entirely clear. In the intestine CFTR is expressed on
the apical membrane of the enterocytes (109). Also in the gut, CFTR functions as a Cl- channel.
The Cl- channel activity of the gut can actually be quantified in ex vivo electrophysiological
studies of gut tissue (110). It is hypothesized that CFTR in the intestine plays a direct role in
the lubrication of the gut contents. Reduced water content of intestinal content may lead to
thickened stools (111). This principle of decreased water content probably lies at the basis of
intestinal pathology in CF like meconium ileus and constipation. Based on studies in mice we
know that in the intestine bile salts are capable of stimulating Cftr via the intestinal
membrane receptor protein Asbt (apical sodium binding protein) (112). The latter indicating
that in Cystic fibrosis intestinal disease, besides the intrinsic malfunction of the CFTR protein,
additionally disease related alternation in bile salt metabolism can be responsible for
decreasing lubrication of the gut contents
FAT MALABSORPTION
One of the most striking features of the gastrointestinal phenotypes in CF is the intestinal fat
malabsorption (113). The fatty stools or steatorrhea is a clinical sign of fat malabsorption. Fat
malabsorption in patients causes several severe problems. Due to the high energy content of
General introduction
23
1 1
fat in general, fecal fat malabsorption causes malnutrition and poor growth. A secondary
effect of the intestinal fat malabsorption is the reduced absorption of the fat soluble vitamins
A, D, E and K. Vitamin malabsorption can lead to hypovitaminosis and vitamin K dependent
coagulopathy (114-116). Therefore, CF patients with intestinal fat malabsorption are usually
dependent on oral vitamin ADEK supplementation.
The main reason the intestinal malabsorption is exocrine pancreatic dysfunction (113). As a
result of the fibrotic destruction, already during pregnancy and infancy, the pancreas loses
the capacity to excrete sufficient amounts of digestive enzymes into the intestine. Pancreatic
enzymes form the bulk of the available intestinal digestive enzymes. The pancreas excretes
lipases for digestion of nutritional fat and proteases for the digestion of nutritional proteins
(117).
The majority of dietary fats consist of triglycerides. These triglycerides are hydrolyzed, by
lipases, into smaller and more polar fatty acids. To overcome the intestinal fat malabsorption,
CF patients with pancreatic insufficiency, are prescribed pancreas enzyme replacement
therapy (PERT) (33). This supplementation is administered simultaneously with every dietary
fat containing meal. PERT dosing is primarily related to the amount of fat calculated from the
diet and adjusted based on clinical effects on symptoms of steatorrhea and growth (118).
INTESTINAL BILE SALT MALABSORPTION
CF patients have a persistently elevated fecal BS excretion compared to healthy controls (26).
The origin of the increased bile salt excretion is not known. First it was suggested to be
related to the intestinal fat malabsorption (119). However increased fecal bile alt excretion is
still present during adequate PERT treatment. Furthermore, increased fecal bile salt excretion
is present in patients with still sufficient exocrine pancreas function, i.e. without fat
malabsorption (120). The latter observation was confirmed in experimental CF mice models.
Also in “mild” CF mouse models, with normal fat absorption, fecal bile salt loss is increased
(121).
Based on the observation that increased fecal bile salt excretion is independent of fecal fat
malabsorption it is assumed, that increased fecal bile excretion in CF is related to the
dysfunctional Cftr, located in the enterocytes of the intestine itself (122, 123). The majority of
bile salts are re-absorbed in the distal part of the small intestine, the terminal ileum. Bile salt
re-absorption into the enterocytes is facilitated by a membrane protein called apical sodium
dependent bile salt transporter (ASBT). In CF mouse models, it has been shown that the
absence or Cftr function in the enterocytes down regulates the transport capacity of bile salt
by the co-located membrane protein ASBT (112). This observation is suggesting that the
Chapter 1
24
1
intestinal re-absorption of bile salt maybe hindered by dysfunctional CFTR. This may in turn
contribute to the observed increased fecal bile salt loss.
Finally, there are suggestions that the increased fecal bile salt excretion could be associated
with CFTR dependent differences in the intestinal bacterial flora (124). Small intestinal
bacterial overgrowth (SIBO) is common in CF. SIBO can result in steatorrhea, abdominal pain,
bloating, flatulence, nausea, and anorexia (125). Also based on results in CF mice models, it
has become known that absence of Cftr function causes significant changes in the intestinal
bacterial composition (126). In mice, differences in intestinal bacterial composition are
associated with changes in bile salt metabolism and with increased fecal bile salt loss(127).
Therefore, it could well be that the reported changes in intestinal bacterial flora in CF
condition contribute to the increased fecal bile salt excretion.
INTESTINAL INFLAMMATION
A more recently recognized GI condition in CF is intestinal inflammation. Based on fecal
calprotectin levels (a marker for intestinal inflammation), intestinal inflammation is a frequent
event in CF patients (128). Evaluation via capsule endoscopy has indicated that macroscopic
inflammatory lesions are frequently present in CF patients (129). It is not known whether the
inflammation is a direct effect of dysfunctional CFTR in the gut or otherwise. Another
hypothesis is that Cftr function interferes with inflammatory regulatory mechanism and
proteins like the peroxisome proliferator activated receptors (PPARs)(130, 131). Finally, it is
also possible that the intestinal inflammation is related to the combination complex
interactions of the bacterial flora of the gut, bile salts metabolism and the immunological
inflammatory response in the context of CFTR dysfunction.
4) CURRENT HEPATIC AND GASTROINTESTINAL ISSUES OF CYSTIC
FIBROSIS DISEASE
Recent developments in the pathophysiology of CF have elucidated many aspects of the
hepatic and gastrointestinal consequences and pathology of cystic fibrosis. However, many
questions have remained unanswered in this research field. To optimize the care for and
prognosis of CF patients in the future, we aimed to gain more insights in the role of CFTR in
the pathology of hepatic and gastrointestinal CF. It is to be expected that insights in the
pathophysiology of CF in the GI tract will also provide knowledge concerning the physiological
roles of CFTR in other tissues. Furthermore, we are facing a new era of therapeutic options in
CF, when CFTR correctors and potentiatiors are currently coming to the market (132). These
General introduction
25
1 1
promising new treatments have to be monitored for their efficacy. New insights in the role of
CF and CFTR in the liver and gastrointestinal tract can offer growing opportunities to test and
evaluated these new therapeutic options (133).
CURRENT ISSUES FOR CFLD
To improve prevention and treatment possibilities of CFLD, a better knowledge of the etiology
and pathophysiology of liver disease in CF is essential. Several potential leads for the etiology
of CFLD have been reported in the literature (134-136). One of the suggestions for the
pathogenesis of CFLD has been the involvement of increased viscosity of bile, due to CFTR
dependent changes in bile composition. Bile duct obstruction then, sequentially, leads to the
development of an obstructive biliary cirrhosis (137)
. However, no firm experimental evidence
is available to support this hypothesis.
Another potentially contributing factor is bile salt cytotoxicity. In the liver CFTR is located at
the apical membrane of the cholangiocytes lining the bile ducts. In cholangiocytes, CFTR
functions, as a chloride channel, indirectly in water and bicarbonate secretion into the bile
(91). Therefore, CFTR plays an important role in the magnitude of bile production and bile
composition (96). Bile salt cytotoxicity can theoretically be enhanced in CF conditions. The
overall bile salt concentration can be increased due to reduced water secretion and reduced
dilution of the bile.
Furthermore, bile cytotoxicity can be increased due to changes in bile composition. Normally
the detergent effects bile salts are reduced by the biliary phospholipids (75, 83). CFTR related
changes in bile composition of the biliary lipids and/or bile salts could play a role in enhanced
susceptibility to bile salt cytotoxicity and the development of biliary liver disease in cystic
fibrosis.
Bile salt cytotoxicity could be enhanced by increasing the contribution of hydrophobic bile
salts and/or by decreasing the biliary secretion of “protective” phospholipids. The biliary bile
salt profile is determined by de novo synthesis of bile salts and by the secondary bile salt
conversion by the intestinal bacterial flora. A change towards a more cytotoxic bile salt
composition could be the result of either altered primary bile salt synthesis or secondary bile
salt conversion.
The treatment of CFLD remains a controversy in the care for CF patients. The only medication
currently used is ursodeoxycholate (UDCA) (64). This relatively hydrophilic bile salt is
administered orally. Its working mechanism is supposed to be related to its choleretic and
anti-inflammatory/anti-fibrotic capacities. In several clinical trials, it has been shown that
UDCA is capable to recover elevated liver functions tests like AST, ALT and GGT to normal
Chapter 1
26
1
values (138, 139). However, there are no long-term follow up studies of the therapeutic
effects of UDCA with respect to clinical endpoints, such as mortality, need for liver
transplantation, or fibrosis/cirrhosis complications. It has never been shown that UDCA is able
to prevent cirrhosis, nor its complications. One of the problems for clinically studies is that we
are not able to identify the subgroup of the CF patients (~10%) that are at risk for cirrhotic
disease.
It has been assumed that the therapeutic effect of UDCA functions via its capacity to increase
bile flow (61). However, based on fundamental studies, it has not been demonstrated that
UDCA indeed is actually capable of increasing bile flow in CF conditions and if so, to what
extent. Measuring bile flow and bile production in humans is invasive and difficult. Therefore,
there is a need for experimental support that clarifies this issue.
Recent studies on cirrhosis by other etiologies indicate that, even in progressive fibrotic liver
diseases, the cirrhosis can be stopped or even reverse on removal of the causative agent
and/or on treatment of the underlying disease (140). In particular antiviral treatment has
been shown to be able to reverse the severity of Hepatitis B virus cirrhosis (141, 142). New
developments are evolving concerning the use of anti-fibrotic therapies in liver fibrosis and
cirrhosis. Although theoretically promising, to date there not yet anti-fibrotic therapies
available in humans (143). However, the scope of these positive developments indicates the
rising opportunity and potential profit for preemptive treatment in CCFLD.
These promising scientific advances indicate the need for reliable and relevant markers to
identify patients at risk for CCFLD. Therefore, one of the most urgent issues in CFLD is the
possibility of early detection of patients at risk for, or in an early phase of, the disease.
Progress in this diagnostic field of CF would offer opportunities for evaluation of current en
new interventional therapies to prevent or treat CFLD. As stated above, no tools for early
detection or recognition are available. As a consequence, CFLD is frequently only recognized
clinically in an advanced stage of severe fibrosis or even cirrhosis. However, cirrhosis typically
becomes manifest around the age of 10 years and it is not present at birth. This observation
indicates that cirrhosis in CFLD develops progressively over a period of years during
childhood.
At this time, cirrhosis is usually diagnosed based on physical examination and abdominal
ultrasound studies of the liver and spleen (53, 144). On physical examination, the most
prominent clinical feature of cirrhosis is splenomegaly. This is often accompanied by
hematological signs of hypersplenism, like thrombo- and leucocytopenia. Ultrasound of the
liver, in case of cirrhosis, can show irregular liver edges and an inhomogeneous, nodular,
pattern of the parenchyma (43, 145, 146). The spleen span can be enlarged for age. Since
portal hypertension is often a characteristic symptom in CF related cirrhosis, abdominal
ultrasound may reveal signs vascular collaterals originating from the portal system. Upper GI
General introduction
27
1 1
endoscopy may reveal evidence for esophageal or gastric variceal disease based on the often
severe portal hypertension (147).
Liver function test (LFTs), like AST, ALT and GGT, are often used to diagnose CFLD. However,
their clinical value for diagnosing (development of) liver fibrosis or even cirrhosis has not been
established. CF patients are routinely checked for elevation LFTs. Frequently patients have
(even persistent) elevations of LFTs above the upper limit of normal (148). By some authors,
this elevation of LFTs has been defined as one of the characteristics of CFLD (52). There is no
proven correlation, however, between elevation of LFTs and development of liver fibrosis or
cirrhosis (60). The current way of evaluating LFTs elevation is thus not validated and useable
to identify patient at risk or actually developing CFLD.
The gold standard for validation of the stage of liver fibrosis in general is histology, normally
obtained via percutaneous liver biopsy. Several studies have shown that liver histology can
classify milder forms of fibrosis in CF (149). However, liver biopsy has several practical and
important drawbacks. Based on autopsy studies in CF patients it has been reported that
fibrosis and cirrhosis may not be evenly spread throughout the liver. This observation
highlights the risk of a sampling error in the case of liver biopsy (60, 150). Another problem
with liver biopsy in CF is the patient selection. It is known that (“only”) around 10% of CF
patients with severe mutations will develop cirrhosis. Since liver biopsy is an invasive
procedure, it is not ethical to use it, with a low threshold, in CF patients. Also in the case of a
liver biopsy, the reliable identification of patients seriously at risk for cirrhosis, could justify
the use of this invasive diagnostic procedure.
Several studies have reported on the possibilities to use ultrasound of the liver as a screening
tool for staging liver fibrosis and cirrhosis in CF patients (43, 145, 150). Ultrasound of the liver
seems promising as a screening measure. It is non-invasive, widely available and already
frequently used in standard clinical CF care. Apart from the inhomogeneity and nodular
abnormalities consistent with cirrhosis in general, several other ultrasound particularities can
be found in CF (146). Often on ultrasound the echogenicity of liver parenchyma is increased.
Increased echogenicity of the liver, however, is not specific for the development of cirrhosis:
rather, it may be related to an increased fat content or steatosis of the liver. There are no
indications that steatosis is preceding the development of fibrosis in CFLD. In general, it is
concluded that ultrasound studies of the liver are a reliable tool for the diagnosis of cirrhosis
in CF, i.e. the end-stage, but are not sensitive enough to recognize or classify CFLD in an
earlier phase.
A more recent development in the diagnosis for liver fibrosis in CF is the use of transient
elastography(151, 152). This technology is based on a method to determine tissue stiffness by
measuring sound wave reflection produced by a special probe. The method is non-invasive
and could potentially be used a screening tool in all pediatric CF patients. Problem with this
method is that it has not been satisfactorily validated for the pediatric (CF) population (153,
Chapter 1
28
1
154). Another difficulty with elastography is that, in CF, it remains to be validated against the
gold standard for liver fibrosis; histology from liver biopsy.
CURRENT ISSUES FOR INTESTINAL CF
Better nutritional status is related to a better survival of CF patients (28). Therefore,
improvement and maintenance of the nutritional status is one of the cornerstones in the care
for CF patients (118). To achieve this goal optimizing dietary intake and improving intestinal
absorption of nutrients are priorities in the research agenda and treatment of CF. Since
dietary fats provide the largest caloric share in the normal dietary energy intake, optimizing
dietary fat intake and absorption provides the greatest potential benefit for improvement of
the nutritional status.
Despite adequate and optimal pancreatic enzyme replacement therapy for exocrine
pancreatic insufficiency, in many CF patients, a persistent fat malabsorption remains (35). This
PERT resistant fat malabsorption is probably associated with other, non-pancreatic, CF related
factors involved in the intestinal fat absorption (36). To further optimize the diagnosis and the
treatment possibilities of PERT resistant fat malabsorption we need more insight in the
factors involved.
Basically, intestinal fat absorption proceeds in two sequential phases. First lipolysis is
facilitated by the digestive enzymes mainly produced by the pancreas. Secondly there are the
processes involved in post-lipolytic fat absorption like the solubilization of fats by bile salts. To
date there is no clear answer to the exact cause of the malfunction of the postlipolytic phase
in CF conditions. However, several factors are associated with PERT resistant fat
malabsorption like changes in bile salt metabolism and the enterohepatic circulation and
differences in bacterial flora. It could also possible that the PERT resistant fat malabsorption
in CF is a direct consequence of the dysfunctional CFTR protein in the apical membrane of the
enterocytes.
5) OUTLINE OF METHODOLOGY USED FOR THIS THESIS
For the studies in this thesis, we applied a variety of methods to obtain more
pathophysiological insights in CF disease in the liver and gastrointestinal tract. Broadly, our
methods could be divided into 3 major groups. First, we aimed to carry out a literature review
study concerning the reported and theoretical background of PERT resistant intestinal fat
malabsorption in the context of CF and CFTR malfunction. Secondly, we performed a
retrospective study in CF patients, to determine the potential value of specific LFTs in
General introduction
29
1 1
identifying CF patients at risk for cirrhosis. Thirdly, we performed fundamental studies in a
variety of CF mice models to determine the role of bile salt metabolism in CF conditions and
the role of bile salts in the development of CFLD. In the CF mice models, we performed
experiments to assess the role of bile salts, their metabolism and enterohepatic circulation.
To aid the reader with limited knowledge on CF mice models, the background of these models
and the methodologies used will first be discussed in more detail.
CYSTIC FIBROSIS MOUSE MODELS
Cystic fibrosis was the first monogenic genetic disease in which the disease causing gene
mutation was identified (5, 6). As a result of this breakthrough in 1989, the genetic and
molecular insights into the functions of the CFTR protein increased rapidly. Based on this
knowledge, several genetically engineered experimental mouse models were developed, in
which the genotype of human CF was mimicked (155, 156). These mice models have been
used successfully to understand CF pathophysiology in different organs. Also in the present
thesis, we used specific CF mouse models to delineate the role of Cftr in the pathophysiology
of cystic fibrosis.
Globally the CF mice models can be divided into two categories. In the first category the Cftr
gene is no longer functionally expressed, the so called complete Cftr knockout mice or Cftr-/-
mice. The knockout mice are not able to produce any functional Cftr protein. The second
category of CF mouse models is engineered in a way that they express a mutated form of the
Cftr gene, in accordance with the most frequent CFTR gene mutations in human CF patients.
Theoretically these Cftr mutated mice resemble the genetic and functional situation of CF
patients more closely than the knock-out mouse models. For example, in parallel to human
CF, mouse models have been constructed that carry the homozygous 508del mutations in
their (murine) CF gene. The delta F508 mutation consists of a deletion of the three
nucleotides that comprise the codon for phenylalanine (F) at position 508. Having two copies
of this mutation (one inherited from each parent) is the leading cause of CF (157).
The phenotype of CF mouse models does not resemble the human CF disease in every aspect.
Some CF disease traits are found similarly in CF mice and humans, including increased fecal
bile salt excretion (121). Others, however, like for instance lung disease, are hardly or not
present in CF mice. Furthermore, strong mouse strain background effects are present among
the different CF gene modifications, irrespective of complete or incomplete Cftr inactivation.
The strain effects are probably related to the presence of disease modifying genes in the
genome of the different genetic strain backgrounds.
Chapter 1
30
1
HEPATIC PHENOTYPE OF CF MOUSE MODELS
CF mouse models display varying types and degrees of hepatic histological phenotypes. Most
mouse models show a normal liver histology. However, there are several reports of CF mice
models with spontaneous developed histopathology. The most severe hepatic phenotype is
described in the University of North Carolina (UNC) Cftrunc/unc
knockout mice. Even without a
further challenge, these Cftr-/-
animals develop focal and progressive hepatobiliary disease
(158). By 3 months of age, they have varying degrees of periportal and bridging fibrosis, which
then progresses with age. Freudenberg et Al. describe histopathology in the homozygous
del508 (Cftr508/508
) mice (135). Liver histopathology of CF and controls livers displays variable,
mild, patchy cholangiopathy characterized by reactive changes in the biliary epithelium, bile
ductular proliferation, and mild portal fibrosis. These findings were only rarely present in WT
mice. Notwithstanding the hepatic phenotype, none of these mice manifested any advanced
degree of liver fibrosis or cirrhosis. The mice did not exhibit any bile duct lesions even though
some animals were more than 12 month old.
Besides spontaneous development of liver histopathology, there are also CF mouse models
with increased susceptibility to an exogenous challenge to induce a liver phenotype. There is
an existing clinical and experimental relationship between colitis and another biliary disease
namely primary sclerosing cholangitis. In experimental animals, a chemical colitis can be
induced via oral dextran sulfate sodium (DSS) administration (136). Based on this concept it
was hypothesized that loss of Cftr function in the setting of a DSS induced colitis can lead to
(enhanced) bile duct injury. Blanco et al. demonstrated that mice homozygous for Cftr
mutations developed bile duct injury following the DSS induction of colitis. In an additional
study from the same research group, the bile duct susceptibility of the Cftr-/-
mouse could be
related to decrease Ppar α expression in Cftr-/-
mice (159). These results seem to suggest that
changes in the systemic inflammatory regulation could be involved in the etiology of CFLD.
THE INTESTINAL PHENOTYPE OF CF MICE MODELS
The most marked phenotypical symptomatology of CF mice is intestinal obstruction (160).
This phenotypical trait resembles in some aspects the clinical picture of intestinal obstruction
in meconium ileus or DIOS in CF patients. In mice, it typically presents when mouse pups are
weaned from breast feeding to solid foods (usually at age ~18-19 days). The obstruction can
be severe and even lethal for the mice. Some mice models have to be treated with oral
laxatives to restore normal bowel movements and/or to prevent mortality. Other mice with
an even more severe intestinal phenotype need persistent liquid feeding to prevent
(re)occurrence of intestinal obstruction (161).
General introduction
31
1 1
Histology of the pancreas does show some preliminary signs of pancreatic fibrosis. However,
based on evaluation of pancreatic function tests, CF mice are pancreatic sufficient. Despite
this normal exocrine pancreatic function, CF mice models do present an increased fecal fat
excretion compared to non CF normal control mice. The latter, indicating other, non-lipolytic,
causes for intestinal fat malabsorption in CF conditions (158).
Another intestinal phenotype of the CF mouse models is intestinal inflammation in
combination with small intestine bacterial overgrowth (SIBO). Oxana et al. describe the novel
finding that a specific innate immune response in the CF mouse small intestine has a role in
inflammation and contributes to the failure to thrive in this mouse model of CF (162). When
then treated with broad-spectrum antibiotics for 3 weeks, the Cftr-/-
mice increased in body
weight when compared with controls (163). Wouthuyzen et al. reported that this positive
effect of antibiotics on body weight was not mediated by increasing the absorption of long-
chain triglycerides (127). Antibiotic treatment of homozygous ΔF508 mice without SIBO
neither augmented body weight nor increased fat absorption. Collectively, the data indicate
that the positive effect of antibiotics on body weight in CFTR-knockout mice may be
attributable to the treatment of SIBO and not necessarily to enhanced absorption of long-
chain triglycerides.
Like human CF patients, CF mice also have an increased fecal bile salt excretion compared to
control mice. To date there is no conclusive explanation for this observation. Since CF mouse
models display normal exocrine pancreatic function, the increased bile salt excretion seems
not related to the pancreatic enzyme dependent lipolysis of the dietary fats (164). Another
possibility is a decreased bile salt uptake in the ileum by the apical sodium dependent bile salt
transporter protein (ASBT). Bijvelds et al. demonstrated that ileal TC uptake was reduced by
17% in Cftr-null mice (112). Because the distal ileum is the discrete site of active BS uptake
and has a pivotal role in the near-complete recovery (∼95%) of the intestinal BS load, this
reduction may well explain the high level of fecal BS excretion reported previously for our
Cftr-null mice (165). However, for homozygous F508del mice, we did not find a reduction in
ileal BS uptake, despite the fact that fecal BS loss is increased to a similar extent as in null
mice. Another explanation may involve interaction between intestinal bile salt metabolism
and Cftr related changes in intestinal bacterial flora. Even under physiological conditions
there is an active and complex interaction between bacterial flora and bile salt metabolism.
For instance, mice with a deficiency in microbiota (germ free mice or mice treated with
antibiotics), display decreased fecal bile acid excretion and an increased bile acid pool size
(166, 167). These studies indicate that bacteria of the gut microflora are interrelated with the
modulation of host bile salts. Since bacterial flora is altered in CF conditions, this could lead
changes in fecal bile salt excretion.
Chapter 1
32
1
GENERAL DESCRIPTION SPECIFIC METHODOLOGY USED IN CF MOUSE
MODELS
There are several methods available to study the hepatic and intestinal consequences of Cftr
function in experimental mice models. There are anthropometry methods like measuring
body weight, liver weight and the determining the liver weight to body weight ratio. These
methods provided the possibility to assess nutritional status and possible trophic factors of
liver growth
FECAL SAMPLE COLLECTION
Fecal sample collection can be used for different purposes. In combination with 72 hour
registration of the oral dietary intake, we used 72 hours stools collection to estimate the
percentage of fat absorption. Fecal fat content was further evaluated to determine the
differential absorbability of the various fatty acids species. Fecal samples were also used to
measure the total bile salt excretion and determine the profile of the different bile salts
species.
BILE DUCT CANNULATION
We performed bile duct cannulation to determine bile production and bile composition. In
this method, the mice are operated via microsurgery. The abdomen is explored and the
gallbladder identified. The choledochal duct is identified and ligated, thereby blocking the bile
flow to the intestine. A hole is punctured in the gallbladder with a needle, after which it is
canulated and fixated. Subsequently all the bile produced by the liver flow freely via the tube
and can be collected in a tube for gravimetric analysis to measure bile production. The bile is
used for biochemical analysis like bile salt concentration and composition, and biliary lipids
(phospholipids and cholesterol).
BILE SALT KINETICS
Additionally we applied the determination of bile salt kinetic parameters, i.e., pool size
(amount of bile salts in the body), fractional turnover rate (the portion of the pool that is
newly synthesized per day), and synthesis rate. These parameters reflect hepatocellular
function, the metabolism of bile salts and the efficiency of enterohepatic cycling. To measure
General introduction
33
1 1
these kinetic parameters we used a previously developed and validated stable isotope
dilution technique without the need to interrupt the enterohepatic circulation. This method
allows simultaneous determination of kinetic parameters. We used the stable isotope
technique with [2H4]-cholate as labeled bile salt. Cholate is a major primary bile salt species
and comprises 50 to 80% (rodents) of the total bile salt pool. Therefore, cholate pool size,
fractional turnover rate (FTR) and synthesis rate are kinetic parameters that allow description
of bile salt kinetics of the quantitatively most important bile salt.
EXPERIMENTAL BILE SALT SUPPLEMENTED DIETS
For several of our mouse experiments, we used bile salt supplemented diets. The bile salt
were added and mixed to regular chow mice feeding. We used ursodeoxycholate (UDCA)
supplemented diets to mimic the clinical situation of UDCA treatment in CF patients. In other
experiments, we used supplementation of the hydrophobic bile salt cholate to mimic the
human hydrophobic bile composition (humans have a higher hydrophobicity than mice). The
purpose of the latter approach was to evaluate if a higher hydrophobicity would contribute to
the cytotoxic effect of bile on the development of CFLD like pathology in the CF mice models.
6) SCOPE OF THIS THESIS:
Despite great progress in the care and treatment of CF, the disease still is severe, lifelong and
lifespan limiting. Besides the clinically progressive pulmonary disease, the CF phenotype
includes a variety of gastro-intestinal and hepatic manifestations. Some of these CF
manifestations, like intestinal fat malabsorption and the development of cirrhosis, give rise to
severe and life threatening complications. Better understanding, concerning the
pathogenesis, the mutual interrelationships and the potential treatment options, of the
gastro-intestinal and hepatic manifestation of CF will have profound positive effects on the
prognosis of CF patients.
The enterohepatic circulation of bile salts connects the physiology of the liver with the
physiology of the intestine. Bile salts are crucial functional and regulatory molecules in a
variety of essential physiological processes of the hepatobiliary and gastrointestinal tract. For
example, bile salt secretion is the driving force behind bile formation and bile salts are
indispensable for intestinal fat, and thus energy, absorption. On the other hand, bile salts are
strong detergents, capable of inflicting serious and permanent cell and tissue injury. The
balance between the essential physiologic functions and the destructive forces of bile salts
requires a precise regulation of this delicate system. Disturbances of the equilibrium of the
enterohepatic circulation can potently lead to physiological malfunction and disease.
Chapter 1
34
1
CF is a multi-organ disease including, among others the hepatobiliary and gastrointestinal
tract. Therefore, it is plausible that primary or secondary consequences of CFTR protein
dysfunction lead to alteration or disturbances in the enterohepatic circulation of bile salts.
The specific aim of the research described in this thesis was to determine the role of bile salts
and their enterohepatic circulation in the hepatic and intestinal phenotype in CF.
Many patients with exocrine pancreatic insufficiency suffer from persistent intestinal fat
malabsorption despite adequate PERT therapy. In chapter 2, we describe, in a comprehensive
literature review, all factors potentially involved in PERT resistant fat absorption including the
enterohepatic circulation of bile salts. In the review, we focus in particularly on the role of bile
salts and enterohepatic circulation.
In the following chapters we switch gear towards CF related liver disease. First we turn to the
CF mouse models to answer basic questions concerning the pathogenesis and development of
CFLD. In chapter 3, we test the hypothesis that biliary bile salt cytotoxicity lays at the basis of
the development of CFLD in CF a specific mice model with spontaneous liver disease. This
initial experimental chapter is followed by chapter 4 in which we study if CFLD can be
induced, by feeding a hydrophobic bile salt-containing diet to CF mice without spontaneous
liver disease. Since CFLD is often treated with UDCA, in chapter 5, we tested the effects of
UDCA on bile production and bile composition in a CF mouse model.
To prevent or treat CFLD, it is essential to be able to recognize CF patient at risk or in an early
phase of the disease. To address this issue, in chapter 6, we return to the clinical aspects of
CCFLD. In this chapter, we aim to identify CF patients at risk for cirrhosis in an early clinical
phase. In a retrospective approach, we tested the hypothesis that the development of
cirrhotic CFLD can be predicted on the basis of follow up of biochemical liver function tests. In
chapter 7, we discuss our overall results, place these in a clinical and experimental
framework, and discuss future perspectives.
General introduction
35
1 1
REFERENCE LIST
1. Ratjen F, Döring G. Cystic fibrosis. Lancet. 2003;361(9358):681-9.
2. Vernooij-van Langen AM, Loeber JG, Elvers B, Triepels RH, Gille JJ, Van der Ploeg, Catharina
PB, et al. Novel strategies in newborn screening for cystic fibrosis: A prospective controlled
study. Thorax. 2012;67(4):289-95.
3. Slieker MG, Uiterwaal CS, Sinaasappel M, Heijerman HG, van der Laag J, van der Ent,
Cornelis K. Birth prevalence and survival in cystic FibrosisA national cohort study in the
netherlands. CHEST Journal. 2005;128(4):2309-15.
4. Cystic fibrosis foundation patient registry 2011. 2012.
5. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.
Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989
09/08;245(4922):1073-80.
6. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of
the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.
1989;245(4922):1066.
7. Collins FS. Cystic fibrosis: Molecular biology and therapeutic implications. Science. 1992
05/08;256(5058):774-9.
8. Cystic fibrosis mutation database (CFTR1) [Internet]. Available from:
http://www.genet.sickkids.on.ca/cftr/Home.html.
9. The Clinical and Functional Translation of CFTR (CFTR2) website [Internet]. Available from:
http://www.cftr2.org/index.php.
10. Cystic fibrosis manifestion .png [Internet]. Available from:
http://commons.wikimedia.org/wiki/Main_Page.
11. Kerem E. Pharmacological induction of CFTR function in patients with cystic fibrosis:
Mutation-specific therapy. Pediatr Pulmonol. 2005;40(3):183-96.
12. Hoffman LR, Ramsey BW. Cystic fibrosis TherapeuticsCurrent and future cystic fibrosis
TherapiesThe road ahead. CHEST Journal. 2013;143(1):207-13.
13. Dalemans W, Barbry P, Champigny G, Jallat S, Dott K, Dreyer D, et al. Altered chloride ion
channel kinetics associated with the ΔF508 cystic fibrosis mutation. Nature.
1991;354(6354):526-8.
14. Kerem E, Corey M, Kerem B, Rommens J, Markiewicz D, Levison H, et al. The relation
between genotype and phenotype in cystic fibrosis—analysis of the most common mutation
(ΔF508). N Engl J Med. 1990;323(22):1517-22.
Chapter 1
36
1
15. Trezíse AE, Buchwald M. In vivo cell-specific expression of the cystic fibrosis
transmembrane conductance regulator. . 1991.
16. Wilschanski M, Zielenski J, Markiewicz D, Tsui L, Corey M, Levison H, et al. Correlation of
sweat chloride concentration with classes of the cystic fibrosis transmembrane conductance
regulator gene mutations. J Pediatr. 1995;127(5):705-10.
17. Quinton P, Reddy M. Control of CFTR chloride conductance by ATP levels through non-
hydrolytic binding. . 1992.
18. Wine JJ. The genesis of cystic fibrosis lung disease. J Clin Invest. 1999;103(3):309.
19. Quinton PM. Cystic fibrosis: Impaired bicarbonate secretion and mucoviscidosis. The
Lancet. 2008;372(9636):415-7.
20. Littlewood J. Gastrointestinal complications in cystic fibrosis. J R Soc Med. 1992;85(Suppl
19):13.
21. Zielenski J. Genotype and phenotype in cystic fibrosis. Respiration. 2000;67(2):117-33.
22. Shwachman H, Dooley RR, Guilmette F, Patterson PR, Weil C, Leubner H. Cystic fibrosis of
the pancreas with varying degrees of pancreatic insufficiency. AMA journal of diseases of
children. 1956;92(4):347-68.
23. Oppenheimer EH, Esterly JR. Hepatic changes in young infants with cystic fibrosis: Possible
relation to focal biliary cirrhosis. J Pediatr. 1975;86(5):683-9.
24. Shwachman H, Kulczycki LL. Long-term study of one hundred five patients with cystic
fibrosis: Studies made over a five-to fourteen-year period. Arch Pediatr Adolesc Med.
1958;96(1):6.
25. Rosenstein BJ, Langbaum TS. Incidence of distal intestinal obstruction syndrome in cystic
fibrosis. J Pediatr Gastroenterol Nutr. 1983;2(2):299-301.
26. Weber AM, Roy CC, Morin CL, Lasalle R. Malabsorption of bile acids in children with cystic
fibrosis. N Engl J Med. 1973 11/08;289(19):1001-5.
27. Corey M, McLaughlin F, Williams M, Levison H. A comparison of survival, growth, and
pulmonary function in patients with cystic fibrosis in boston and toronto. J Clin Epidemiol.
1988;41(6):583-91.
28. Yen EH, Quinton H, Borowitz D. Better nutritional status in early childhood is associated
with improved clinical outcomes and survival in patients with cystic fibrosis. J Pediatr. 2012.
29. Borowitz D, Robinson KA, Rosenfeld M, Davis SD, Sabadosa KA, Spear SL, et al. Cystic
fibrosis foundation evidence-based guidelines for management of infants with cystic fibrosis. J
Pediatr. 2009;155(6):S73-93.
General introduction
37
1 1
30. Marino CR, Matovcik LM, Gorelick FS, Cohn JA. Localization of the cystic fibrosis
transmembrane conductance regulator in pancreas. J Clin Invest. 1991;88(2):712.
31. Couper R, Corey M, Moore D, Fisher L, Forstner G, Durie P. Decline of exocrine pancreatic
function in cystic fibrosis patients with pancreatic sufficiency. Pediatr Res. 1992;32(2):179-82.
32. Harper A, Raper HS. Pancreozymin, a stimulant of the secretion of pancreatic enzymes in
extracts of the small intestine. J Physiol (Lond ). 1943;102(1):115-25.
33. Fieker A, Philpott J, Armand M. Enzyme replacement therapy for pancreatic insufficiency:
Present and future. Clinical and experimental gastroenterology. 2011;4:55.
34. Borowitz D, Stevens C, Brettman LR, Campion M, Chatfield B, Cipolli M. International phase
III trial of liprotamase efficacy and safety in pancreatic-insufficient cystic fibrosis patients.
Journal of Cystic Fibrosis. 2011.
35. Kalivianakis M, Minich DM, Bijleveld CMA, van Aalderen WMC, Stellaard F, Laseur M, et al.
Fat malabsorption in cystic fibrosis patients receiving enzyme replacement therapy is due to
impaired intestinal uptake of long-chain fatty acids. Am J Clin Nutr. 1999 01;69(1):127-34.
36. Wouthuyzen-Bakker M, Bodewes F, Verkade H. Persistent fat malabsorption in cystic
fibrosis; lessons from patients and mice. Journal of Cystic Fibrosis. 2011.
37. Valman H, France N, Wallis P. Prolonged neonatal jaundice in cystic fibrosis. Arch Dis Child.
1971;46(250):805-9.
38. Lykavieris P, Bernard O, Hadchouel M. Neonatal cholestasis as the presenting feature in
cystic fibrosis. Arch Dis Child. 1996;75(1):67-70.
39. Farrell PM, Kosorok MR, Rock MJ, Laxova A, Zeng L, Lai H, et al. Early diagnosis of cystic
fibrosis through neonatal screening prevents severe malnutrition and improves long-term
growth. Pediatrics. 2001;107(1):1-13.
40. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis
Child. 1991 06;66(6):698-701.
41. Fishbein MH, Gardner KG, Potter CJ, Schmalbrock P, Smith MA. Introduction of fast MR
imaging in the assessment of hepatic steatosis. Magn Reson Imaging. 1997;15(3):287-93.
42. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.
Hepatology. 1999 11;30(5):1151-8.
43. Patriquin H, Lenaerts C, Smith L, Perreault G, Grignon A, Filiatrault D, et al. Liver disease in
children with cystic fibrosis: US-biochemical comparison in 195 Patients1. Radiology.
1999;211(1):229-32.
44. Dodge JA, Turck D. Cystic fibrosis: Nutritional consequences and management. Best
Practice & Research Clinical Gastroenterology. 2006;20(3):531-46.
Chapter 1
38
1
45. Strandvik B, Hultcrantz R. Liver function and morphology during long‐term fatty acid
supplementation in cystic fibrosis. Liver. 1994;14(1):32-6.
46. Stern RC, Rothstein FC, Doershuk CF. Treatment and prognosis of symptomatic gallbladder
disease in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr. 1986;5(1):35-40.
47. Willi UV, Reddish JM, Teele RL. Cystic fibrosis: Its characteristic appearance on abdominal
sonography. AJR Am J Roentgenol. 1980 May;134(5):1005-10.
48. Dobson RL, Johnson MA, Hennig RC, Brown NE. Sonography of the gallbladder, biliary tree,
and pancreas in adults with cystic fibrosis. Can Assoc Radiol J. 1988 Dec;39(4):257-9.
49. Colombo C, Bertolini E, Assaisso ML, Bettinardi N, Giunta A, Podda M. Failure of
ursodeoxycholic acid to dissolve radiolucent gallstones in patients with cystic fibrosis. Acta
Paediatrica. 1993;82(6‐7):562-5.
50. Freudenberg F, Leonard MR, Liu SA, Glickman JN, Carey MC. Pathophysiological
preconditions promoting mixed “black” pigment plus cholesterol gallstones in a ΔF508 mouse
model of cystic fibrosis. Am J Physiol Gastrointest Liver Physiol. 2010;299(1):G205-14.
51. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of
cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr
Gastroenterol Nutr. 1997;25(1):32-6.
52. Colombo C, Battezzati PM, Crosignani A, Morabito A, Costantini D, Padoan R, et al. Liver
disease in cystic fibrosis: A prospective study on incidence, risk factors, and outcome.
Hepatology. 2002;36(6):1374-82.
53. Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the
diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros.
2011;10:S29-36.
54. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-
associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.
55. Giniès JL, Couetil JP, Houssin D, Guillemain R, Champion G, Bernard O. Hepatopulmonary
syndrome in a child with cystic fibrosis. J Pediatr Gastroenterol Nutr. 1996;23(4):497-500.
56. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre
experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008
05;7(3):252-7.
57. Bartlett JR, Friedman KJ, Ling SC, Pace RG, Bell SC, Bourke B, et al. Genetic modifiers of
liver disease in cystic fibrosis. JAMA: the journal of the American Medical Association.
2009;302(10):1076-83.
General introduction
39
1 1
58. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic
fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.
1999;31(1):77-83.
59. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy
findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis.
2008;7(3):215-21.
60. Lewindon PJ, Shepherd RW, Walsh MJ, Greer RM, Williamson R, Pereira TN, et al.
Importance of hepatic fibrosis in cystic fibrosis and the predictive value of liver biopsy.
Hepatology. 2011;53(1):193-201.
61. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of
action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.
62. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for liver
disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.
2003;23(6):1484-90.
63. Nousia-Arvanitakis S, Fotoulaki M, Economou H, Xefteri M, Galli-Tsinopoulou A. Long-term
prospective study of the effect of ursodeoxycholic acid on cystic fibrosis-related liver disease. J
Clin Gastroenterol. 2001;32(4):324-8.
64. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.
Cochrane Database Syst Rev. 2012;10.
65. Hultcrantz R, Mengarelli S, Strandvik B. Morphological findings in the liver of children with
cystic fibrosis: A light and electron microscopical study. Hepatology. 1986 09;6(5):881-9.
66. Shier KJ, Horn Jr RC. The pathology of liver cirrhosis in patients with cystic fibrosis of the
pancreas. Can Med Assoc J. 1963;89(13):645.
67. Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr
Scand. 1988;363:45,50; discussion 50-1.
68. Sokol RJ, Durie PR. Recommendations for management of liver and biliary tract disease in
cystic fibrosis. J Pediatr Gastroenterol Nutr. 1999;28:S1.
69. Cotting J, Lentze M, Reichen J. Effects of ursodeoxycholic acid treatment on nutrition and
liver function in patients with cystic fibrosis and longstanding cholestasis. Gut.
1990;31(8):918-21.
70. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in
cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.
71. Salvatore F, Scudiero O, Castaldo G. Genotype–phenotype correlation in cystic fibrosis: The
role of modifier genes. Am J Med Genet. 2002;111(1):88-95.
Chapter 1
40
1
72. Vawter GF, Shwachman H. Cystic fibrosis in adults: An autopsy study. Pathol Annu.
1979;14 Pt 2:357-82.
73. Gooding I, Dondos V, Gyi KM, Hodson M, Westaby D. Variceal hemorrhage and cystic
fibrosis: Outcomes and implications for liver transplantation. Liver transplantation.
2005;11(12):1522-6.
74. Nash K, Collier J, French J, McKeon D, Gimson A, Jamieson N, et al. Cystic fibrosis liver
disease: To transplant or not to transplant? American Journal of Transplantation.
2008;8(1):162-9.
75. Hofmann AF. Bile acid secretion, bile flow and biliary lipid secretion in humans.
Hepatology. 1990 09;12(3):17S-22S.
76. Hofmann A. Fat digestion: The interaction of lipid digestion products with micellar bile acid
solutions. In: Lipid Absorption: Biochemical and Clinical Aspects. Springer; 1976. p. 3-21.
77. Hofmann AF, Roda A. Physicochemical properties of bile acids and their relationship to
biological properties: An overview of the problem. J Lipid Res. 1984;25(13):1477-89.
78. Chiang JY. Bile acids: Regulation of synthesis. J Lipid Res. 2009;50(10):1955-66.
79. Trauner M, Claudel T, Fickert P, Moustafa T, Wagner M. Bile acids as regulators of hepatic
lipid and glucose metabolism. Digestive Diseases. 2010;28(1):220-4.
80. Hofmann AF. Enterohepatic circulation of bile acids. . 1969.
81. Small DM, Dowling RH, Redinger RN. The enterohepatic circulation of bile salts. Arch Intern
Med. 1972;130(4):552.
82. Hofmann A, Small D. Detergent properties of bile salts: Correlation with physiological
function. Annu Rev Med. 1967;18(1):333-76.
83. Hofmann AF. Bile acids: The good, the bad, and the ugly. Physiology. 1999;14(1):24-9.
84. De Vree, J Marleen L, Jacquemin E, Sturm E, Cresteil D, Bosma PJ, Aten J, et al. Mutations
in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proceedings of the
National Academy of Sciences. 1998;95(1):282-7.
85. Mauad TH, van Nieuwkerk CM, Dingemans KP, Smit JJ, Schinkel AH, Notenboom RG, et al.
Mice with homozygous disruption of the mdr2 P-glycoprotein gene a novel animal model for
studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis. The American
journal of pathology. 1994;145(5):1237.
86. van Nieuwerk CM, Groen AK, Ottenhoff R, van Wijland M, Van Den Bergh Weerman MA,
Tytgat GN, et al. The role of bile salt composition in liver pathology of mdr2 (-/-) mice:
Differences between males and females. J Hepatol. 1997 01;26(1):138-45.
General introduction
41
1 1
87. Boyer JL, Klatskin G. Canalicular bile flow and bile secretory pressure. evidence for a non-
bile salt dependent fraction in the isolated perfused rat liver. Gastroenterology. 1970
Dec;59(6):853-9.
88. Drew R, Priestly B. Choleretic and cholestatic effects of infused bile salts in the rat.
Experientia. 1979;35(6):809-11.
89. Hofmann AF. Biliary secretion and excretion in health and disease: Current concepts. Ann
Hepatol. 2007 01;6(1):15-27.
90. Roberts SK, Kuntz SM, Gores GJ, LaRusso NF. Regulation of bicarbonate-dependent
ductular bile secretion assessed by lumenal micropuncture of isolated rodent intrahepatic bile
ducts. Proceedings of the National Academy of Sciences. 1993;90(19):9080-4.
91. Fitz JG. Regulation of cholangiocyte secretion. Semin Liver Dis. 2002 08;22(3):241-9.
92. Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride
currents in rat bile duct epithelial cells. J Clin Invest. 1993 01;91(1):319-28.
93. Schlenker T, Romac JM, Sharara AI, Roman RM, Kim SJ, Larusso N, et al. Regulation of
biliary secretion through apical purinergic receptors in cultured rat cholangiocytes. American
Journal of Physiology-Gastrointestinal and Liver Physiology. 1997;273(5):G1108-17.
94. Minagawa N, Nagata J, Shibao K, Masyuk AI, Gomes DA, Rodrigues MA, et al. Cyclic AMP
regulates bicarbonate secretion in cholangiocytes through release of ATP into bile.
Gastroenterology. 2007;133(5):1592-602.
95. Chen B, Nicol G, Cho WK. Role of calcium in volume-activated chloride currents in a mouse
cholangiocyte cell line. J Membr Biol. 2007;215(1):1-13.
96. Feranchak AP, Sokol RJ. Cholangiocyte biology and cystic fibrosis liver disease. Seminars in
liver disease; ; 2001.
97. Gunderson KL, Kopito RR. Conformational states of CFTR associated with channel gating:
The role of ATP binding and hydrolysis. Cell. 1995;82(2):231-9.
98. Schinkel AH, Jonker JW. Mammalian drug efflux transporters of the ATP binding cassette
(ABC) family: An overview. Adv Drug Deliv Rev. 2012.
99. Schwiebert EM, Egan ME, Hwang T, Fulmer SB, Allen SS, Cutting GR, et al. CFTR regulates
outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell.
1995;81(7):1063-73.
100. Banales JM, Arenas F, Rodríguez‐Ortigosa CM, Sáez E, Uriarte I, Doctor RB, et al.
Bicarbonate‐rich choleresis induced by secretin in normal rat is taurocholate‐dependent and
involves AE2 anion exchanger. Hepatology. 2006;43(2):266-75.
Chapter 1
42
1
101. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic
acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.
Gastroenterology. 2007 11;133(5):1603-13.
102. Russell DW. Fifty years of advances in bile acid synthesis and metabolism. J Lipid Res.
2009;50(Supplement):S120-5.
103. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, et al. Molecular basis for
feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell. 2000;6(3):507-15.
104. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for
the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.
105. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory
cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol
Cell. 2000;6(3):517-26.
106. Lundåsen T, Gälman C, Angelin B, Rudling M. Circulating intestinal fibroblast growth
factor 19 has a pronounced diurnal variation and modulates hepatic bile acid synthesis in
man. J Intern Med. 2006;260(6):530-6.
107. Xie M, Holcomb I, Deuel B, Dowd P, Huang A, Vagts A, et al. FGF-19, a novel fibroblast
growth factor with unique specificity for FGFR4. Cytokine. 1999;11(10):729-35.
108. Rao Y, Studer EJ, Stravitz RT, Gupta S, Qiao L, Dent P, et al. Activation of the
Raf‐1/MEK/ERK cascade by bile acids occurs via the epidermal growth factor receptor in
primary rat hepatocytes. Hepatology. 2002;35(2):307-14.
109. Trezíse AE, Buchwald M. In vivo cell-specific expression of the cystic fibrosis
transmembrane conductance regulator. . 1991.
110. De Jonge HR, Ballmann M, Veeze H, Bronsveld I, Stanke F, Tümmler B, et al. Ex vivo CF
diagnosis by intestinal current measurements (ICM) in small aperture, circulating ussing
chambers. Journal of Cystic Fibrosis. 2004;3:159-63.
111. Wilschanski M, Durie P. Pathology of pancreatic and intestinal disorders in cystic fibrosis.
J R Soc Med. 1998;91(Suppl 34):40.
112. Bijvelds MJC, Jorna H, Verkade HJ, Bot AGM, Hofmann F, Agellon LB, et al. Activation of
CFTR by ASBT-mediated bile salt absorption. American Journal of Physiology-Gastrointestinal
and Liver Physiology. 2005;289(5):G870.
113. Andersen DH. Cystic fibrosis of the pancreas and its relation to celiac diseasea clinical and
pathologic study. American journal of Diseases of Children. 1938;56(2):344-99.
114. Hahn TJ, Squires AE, Halstead LR, Strominger DB. Reduced serum 25-hydroxyvitamin D
concentration and disordered mineral metabolism in patients with cystic fibrosis. J Pediatr.
1979;94(1):38-42.
General introduction
43
1 1
115. Walters TR, Koch CHF. Hemorrhagic diathesis and cystic fibrosis in infancy. Arch Pediatr
Adolesc Med. 1972;124(5):641.
116. Farrell PM, Bieri JG, Fratantoni JF, Wood RE, di Sant'Agnese PA. The occurrence and
effects of human vitamin E deficiency: A study in patients with cystic fibrosis. J Clin Invest.
1977;60(1):233.
117. Domínguez‐Muñoz JE. Pancreatic exocrine insufficiency: Diagnosis and treatment. J
Gastroenterol Hepatol. 2011;26(s2):12-6.
118. Stallings VA, Stark LJ, Robinson KA, Feranchak AP, Quinton H. Evidence-based practice
recommendations for nutrition-related management of children and adults with cystic fibrosis
and pancreatic insufficiency: Results of a systematic review. J Am Diet Assoc. 2008;108(5):832-
9.
119. Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in
cystic fibrosis. Gut. 1976;17(4):295.
120. Watkins J, Tercyak A, Szczepanik P, Klein P. Bile salt kinetics in cystic fibrosis: Influence of
pancreatic enzyme replacement. Gastroenterology. 1977;73(5):1023.
121. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.
Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption
in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.
122. Fondacaro JD, Heubi JE, Kellogg FW. Intestinal bile acid malabsorption in cystic fibrosis: A
primary mucosal cell defect. Pediatr Res. 1982;16(6):494-8.
123. O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid
malabsorption in cystic fibrosis. Gut. 1993;34(8):1137.
124. Ridlon JM, Kang D, Hylemon PB. Bile salt biotransformations by human intestinal
bacteria. J Lipid Res. 2006;47(2):241-59.
125. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in
cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.
Acta Biochim Pol. 2009;56(4):631.
126. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.
2006;42(1):46-52.
127. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade
HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.
2011;71(1):4-12.
Chapter 1
44
1
128. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal
inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.
Aliment Pharmacol Ther. 2004;20(7):813-9.
129. Werlin SL, Benuri-Silbiger I, Kerem E, Adler SN, Goldin E, Zimmerman J, et al. Evidence of
intestinal inflammation in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr.
2010;51(3):304.
130. Ollero M, Junaidi O, Zaman MM, Tzameli I, Ferrando AA, Andersson C, et al. Decreased
expression of peroxisome proliferator activated receptor γ in CFTR−/− mice. J Cell Physiol.
2004;200(2):235-44.
131. Dekkers JF, van der Ent, Cornelis K, Kalkhoven E, Beekman JM. PPARγ as a therapeutic
target in cystic fibrosis. Trends Mol Med. 2012;18(5):283-91.
132. Accurso FJ, Rowe SM, Clancy J, Boyle MP, Dunitz JM, Durie PR, et al. Effect of VX-770 in
persons with cystic fibrosis and the G551D-CFTR mutation. N Engl J Med. 2010;363(21):1991-
2003.
133. Dekkers JF, Wiegerinck CL, de Jonge HR, Bronsveld I, Janssens HM, de Winter-de Groot,
Karin M, et al. A functional CFTR assay using primary cystic fibrosis intestinal organoids. Nat
Med. 2013.
134. Pereira TN, Lewindon PJ, Greer RM, Hoskins AC, Williamson RM, Shepherd RW, et al.
Transcriptional basis for hepatic fibrosis in cystic fibrosis–associated liver disease. J Pediatr
Gastroenterol Nutr. 2012;54:328-35.
135. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.
Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.
Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.
136. Blanco PG, Zaman MM, Junaidi O, Sheth S, Yantiss RK, Nasser IA, et al. Induction of colitis
in cftr–/–mice results in bile duct injury. American Journal of Physiology-Gastrointestinal and
Liver Physiology. 2004;287(2):G491.
137. Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin
Gastroenterol. 2009;25(3):272-8.
138. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for
liver disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.
1996;23(6):1484-90.
139. Lindblad A, Glaumann H, Strandvik B. A two‐year prospective study of the effect of
ursodeoxycholic acid on urinary bile acid excretion and liver morphology in cystic fibrosis–
associated liver disease. Hepatology. 1998;27(1):166-74.
General introduction
45
1 1
140. Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. J Hepatol. 2012
5;56(5):1171-80.
141. Brown A, Goodman Z. Hepatitis B-associated fibrosis and fibrosis/cirrhosis regression
with nucleoside and nucleotide analogs. Expert review of gastroenterology & hepatology.
2012;6(2):187-98.
142. D'Ambrosio R, Aghemo A, Rumi MG, Ronchi G, Donato MF, Paradis V, et al. A
morphometric and immunohistochemical study to assess the benefit of a sustained virological
response in hepatitis C virus patients with cirrhosis. Hepatology. 2012;56(2):532-43.
143. Friedman SL. Evolving challenges in hepatic fibrosis. Nature Reviews Gastroenterology
and Hepatology. 2010;7(8):425-36.
144. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory
Reviews. 2013.
145. Lenaerts C, Lapierre C, Patriquin H, Bureau N, Lepage G, Harel F, et al. Surveillance for
cystic fibrosis-associated hepatobiliary disease: Early ultrasound changes and predisposing
factors. J Pediatr. 2003 09;143(3):343-50.
146. Williams SGJ, Evanson JE, Barrett N, Hodson ME, Boultbee JE, Westaby D. An ultrasound
scoring system for the diagnosis of liver disease in cystic fibrosis. J Hepatol. 1995;22(5):513-21.
147. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic
fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.
1999;31(1):77-83.
148. Mayer-Hamblett N, Kloster M, Ramsey BW, Narkewicz MR, Saiman L, Goss CH. Incidence
and clinical significance of elevated liver function tests in cystic fibrosis clinical trials.
Contemporary clinical trials. 2012.
149. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of
cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr
Gastroenterol Nutr. 1997;25(1):32-6.
150. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy
findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis. 2008
5;7(3):215-21.
151. de Lédinghen V, Le Bail B, Rebouissoux L, Fournier C, Foucher J, Miette V, et al. Liver
stiffness measurement in children using FibroScan: Feasibility study and comparison with
fibrotest, aspartate transaminase to platelets ratio index, and liver biopsy. J Pediatr
Gastroenterol Nutr. 2007;45(4):443-50.
152. Foucher J, Chanteloup E, Vergniol J, Castera L, Le Bail B, Adhoute X, et al. Diagnosis of
cirrhosis by transient elastography (FibroScan): A prospective study. Gut. 2006;55(3):403-8.
Chapter 1
46
1
153. Witters P, De Boeck K, Dupont L, Proesmans M, Vermeulen F, Servaes R, et al. Non-
invasive liver elastography (fibroscan) for detection of cystic fibrosis-associated liver disease.
Journal of Cystic Fibrosis. 2009;8(6):392-9.
154. Malbrunot-Wagner A, Bridoux L, Nousbaum J, Riou C, Dirou A, Ginies J, et al. Transient
elastography and portal hypertension in pediatric patients with cystic fibrosis: Transient
elastography and cystic fibrosis. Journal of Cystic Fibrosis. 2011;10(5):338-42.
155. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al.
Mouse models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.
2011;10:S152-71.
156. Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst
Fibros. 2004;3:183-90.
157. Bobadilla JL, Macek M, Fine JP, Farrell PM. Cystic fibrosis: A worldwide analysis of CFTR
mutations—correlation with incidence data and application to screening. Hum Mutat.
2002;19(6):575-606.
158. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol. 2004 04;164(4):1481-93.
159. Pall H, Zaman MM, Andersson C, Freedman SD. Decreased peroxisome proliferator
activated receptor [alpha] is associated with bile duct injury in cystic fibrosis transmembrane
conductance regulator-/-mice. J Pediatr Gastroenterol Nutr. 2006;42(3):275.
160. van Doorninck JH, French PJ, Verbeek E, Peters RH, Morreau H, Bijman J, et al. A mouse
model for the cystic fibrosis delta F508 mutation. EMBO J. 1995 09/15;14(18):4403-11.
161. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production
of a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.
162. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun.
2004;72(10):6040-9.
163. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.
2006;42(1):46.
164. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat
absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am
J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.
General introduction
47
1 1
165. Dawson PA, Haywood J, Craddock AL, Wilson M, Tietjen M, Kluckman K, et al. Targeted
deletion of the ileal bile acid transporter eliminates enterohepatic cycling of bile acids in mice.
J Biol Chem. 2003;278(36):33920-7.
166. Miyata M, Takamatsu Y, Kuribayashi H, Yamazoe Y. Administration of ampicillin elevates
hepatic primary bile acid synthesis through suppression of ileal fibroblast growth factor 15
expression. J Pharmacol Exp Ther. 2009;331(3):1079-85.
167. Islam K, Fukiya S, Hagio M, Fujii N, Ishizuka S, Ooka T, et al. Bile acid is a host factor that
regulates the composition of the cecal microbiota in rats. Gastroenterology.
2011;141(5):1773-81.
Chapter 1
48
1
CHAPTER 2
PERSISTENT FAT MALABSORPTION
IN CYSTIC FIBROSIS; LESSONS
FROM PATIENTS AND MICE
Adapted from: Journal of Cystic Fibrosis 10.3 (2011): 150-158.
Marjan Wouthuyzen-Bakker
Frank A. J. A. Bodewes,
Henkjan J. Verkade
University Medical Center Groningen, the Netherlands
Chapter 3
50
2
ABSTRACT
Fat malabsorption in pancreatic insufficient cystic fibrosis (CF) patients is classically treated
with pancreatic enzyme replacement therapy (PERT). Despite PERT, intestinal fat absorption
remains insufficient in most CF patients. Several factors have been suggested to contribute to
the persistent fat malabsorption in CF (CFPFM). We reviewed the current insights concerning
the proposed causes of CFPFM and the corresponding intervention studies. Most data are
obtained from studies in CF patients and CF mice. Based on the reviewed literature, we
conclude that alterations in intestinal pH and intestinal mucosal abnormalities are most likely
to contribute to CFPFM. The presently available data indicate that acid suppressive drugs and
broad spectrum antibiotics could be helpful in individual CF patients for optimizing fat
absorption and/or nutritional status.
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
51
2
INTRODUCTION
Most patients with cystic fibrosis (CF) have a suboptimal nutritional status. In order to
minimize morbidity and mortality, obtaining and maintaining a good nutritional status
remains one of the important challenges in CF (1-3). Next to preventing and controlling
pulmonary infections, a high dietary energy intake diet combined with pancreatic enzyme
replacement therapy (PERT) are classic cornerstones in CF care.
Despite optimizing and maximizing PERT, fat malabsorption in most CF patients persist
(CFPFM). Rather than the physiological absorption above 95%, most CF patients show a
decreased intestinal fat absorption of ± 85-90% of dietary fat intake (4,5). It is reasonable to
assume that at least part of the failure to approach a physiological degree of fat absorption
could still be attributed to inefficiency and inaccurate dosing of PERT. Also, a non-
synchronous entrance of PERT with dietary energy intake into the duodenum may play a role
in CFPFM (6). However, besides PERT related issues, other factors have been implied to be
involved in CFPFM (7,8).
During the last decades, several of these potential contributing factors to CFPFM have been
addressed and investigated in patient studies as well as animal studies. In this review, we will
provide an overview of the different mechanisms that have been implied to contribute to
CFPFM and describe the results of corresponding interventional studies on fat absorption
and/or nutritional status. We will mainly describe studies performed in CF patients and CF
mice. CF mice are especially suitable to exclusively investigate CFPFM, as most CF mouse
models do not suffer from pancreatic insufficiency but do exhibit impaired absorption of fat
(9). In order to fully recognize and understand the CF-related factors that might be involved in
CFPFM, we start to describe the physiological process of fat digestion and absorption.
PHYSIOLOGY OF FAT DIGESTION AND ABSORPTION
Dietary fat intake mainly consists of long-, medium- and short-chain triglycerides. The
mechanism of digestion and absorption of these lipids can be dissected into several
sequential steps. In contrast to long-chain triglycerides, medium- and short-chain triglycerides
are known to circumvent several steps in lipid digestion and absorption. For example, unlike
long-chain triglycerides, medium- and short chain triglycerides are less dependent upon
solubilization, escape the re-esterification into triglycerides and are directly absorbed into the
portal system without being assembled into chylomicrons (10). Since the majority of dietary
fat and energy intake consists of long-chain triglycerides (92-96%), we exclusively focused on
their mechanism of intestinal digestion and absorption. The mechanism of digestion and
absorption of long-chain triglycerides can be dissected into several sequential processes
(figure 1A).
Chapter 3
52
2
Figure 1. Simplified scheme of fat digestion and absorption in health and in CF conditions. A) Simplified
representation of the process of long-chain triglyceride digestion and absorption in health as described
in the text. B) Overview of the various steps in fat digestion and absorption in relation to findings in CF
patients and CF mice. +; disturbed, -; not disturbed, +/-; likely to be disturbed. ND: not determined.
References are described in the text.
.
CF
patients
CFTR knockout
mice
Homozygous delta
F508 mice
1. Emulsification
2. Lipolysis
3. Solubilization
4. Translocation
5. Intracellular processing
6. Chylomicron production
+
+
+
±
±
±
-
+
+
±
ND
ND
-
-
+
±
ND
ND
A
B
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
53
2
1) Emulsification. Emulsification of triglycerides is a process in which (water-insoluble) fat
droplets are suspended in an aqueous environment. Emulsification can be achieved by
mechanical as well as chemical means. Mechanical emulsification is attained by chewing and
forcing dietary fat with high pressure through a small opening (e.g. the pylorus) and
dispersing large fat droplets into smaller droplets. Chemical emulsification is attained by the
action of bile and prevents the emulsion from re-coalescing. Emulsification results in a fine,
relatively stable oil-in-water emulsion with an increased surface area. The water soluble
digestive / lipolytic enzymes are active at the site of the water-oil interface.
2) Lipolysis. During lipolysis, the breakdown (hydrolysis) of lipids into smaller particles is
continued. Around 10-30% of the triglycerides are hydrolyzed in the stomach into diglycerides
and free fatty acids, catalyzed by lingual and gastric lipase (a process also known as
biochemical/lipolytic emulsification (10)). Under physiological conditions, pancreatic lipase
completes the process in the proximal part of the small intestine, by hydrolyzing the
remaining triglycerides and diglycerides into monoglycerides and free fatty acid molecules.
Bile salts are amphipathic molecules, i.e. they possess a hydrophilic and a hydrophobic site.
Upon exposure to dietary lipid emulsion at the level of the proximal small intestine, the
hydrophobic site orients itself towards the hydrophobic lipid droplets, whereas their
hydrophilic site exposes itself to the aqueous (water) phase. Lipids droplets whose surface is
thus covered bile salts are not accessible to pancreatic lipase. Pancreatic co-lipase allows the
anchoring of the pancreas lipase to the oil-water interface and is essential for lipolytic activity.
Under physiological conditions, pancreatic lipase is present in vast excess. Thus, in CF, fat
maldigestion only occurs during severe pancreatic malfunction (11). During exocrine
pancreatic insufficiency when pancreatic lipase is severely reduced, lipolysis is more
dependent upon the activity of lingual and gastric lipase. In CF conditions, lingual lipase
remains fully active in the intestine, where it accounts for more than 90% of lipase activity,
even when the intestinal pH drop below the optimal level for bile salt dependent lipolysis
(12).
3) Solubilization. The process in which fat molecules are dissolved in water in the form of
(mixed) micelles is called solubilization. Solubilization enhances the aqueous solubility of fatty
acids by several orders of magnitude (100 to 1000 fold) (10). Solubilization is necessary for
monoglycerides and free fatty acids to efficiently overcome the diffusion barrier of the so
called unstirred water layer of the enterocytes (10). The unstirred water layer separates the
enterocytes from the luminal contents of the intestine. Solubilization is achieved by the
formation of mixed micelles; mainly consisting of phospholipids and bile salts derived from
biliary secretion. The diameter of the lipid droplets ranges from 100 – 1000 nm, whereas that
of mixed micelles ranges 3-5 nm. The hydrophobic part of the lipid molecules, such as the
acyl-chains, will be oriented inwards, whereas the hydrophilic parts, (such as the carboxylic
headgroups) orients towards the aqueous outside of the micelle. Saturated fatty acids are
more dependent on solubilization than unsaturated fatty acids, due to the higher
Chapter 3
54
2
hydrophobicity of the former. The difference in bile salt dependency can be derived from
absorption studies in rats with chronic bile diversion: in such an intestinal bile deficient
condition, absorption of saturated fatty acids is below 30% of the ingested amount, while the
absorption of unsaturated fatty acids is relatively maintained (~80%) (13). Just like the
digestion of fat, the process of micelle formation is also pH sensitive. Low intestinal pH levels
can severely inhibit micelle formation or induce premature release of lipolytic products out of
micelles (10).
4) Translocation. Once the mixed micelles diffuse through the unstirred water layer and arrive
at the proximity of the (apical) enterocyte membrane, the free fatty acids and monoglycerides
dissociate from the micelles. It has been postulated that the acidic microclimate near the
apical membrane of intestinal mucosal cells, induces micelle- disintegration and favours the
translocation of fatty acid molecules across the enterocyte membrane (10). It has remained
unclear whether intestinal membrane transporters are essential for the translocation of fatty
acids and monoglycerides. It is suggested that free fatty acid uptake is concentration-
dependent, in which a high intra-luminal concentration drives passive diffusion. Two putative
intestinal transporters have been proposed to be involved fatty acid uptake; the fatty acid
binding protein (FABP) and the fatty acid translocase / cluster determinant 36 (FAT/CD36)
(14). However, both transporters are not likely to be involved in fatty acid uptake into the
enterocytes. FABP is only expressed in a small area of the crypt-villus of the intestine and
knockout mice for FABP and CD36 do not exhibit impairments in fatty acid uptake (15-17).
5) Intracellular processing. After being absorbed into the enterocytes, the lipolytic products
migrate to the endoplasmic reticulum, possibly mediated via the fatty acid binding proteins
(14,15). At the cytoplasmic surface of the endoplasmic reticulum, the fatty acids and
monoglycerides are re-esterified into triglycerides. Under physiological conditions, re-
esterification mainly occurs via the monoacylglycerol pathway, i.e. the sequential acylation of
monoacylglycerol by acyl-CoA (10).
6) Chylomicron production. Newly synthesized triglycerides are transferred into the smooth
endoplasmic reticulum and are assembled into lipoprotein particles called chylomicrons.
Intestinal phospholipids are required for chylomicron production in order to prevent
accumulation in the enterocyte (18,19). Maturation of chylomicrons (i.e. assembly of fat
particles with a phospholipid-cholesterol-apolipoprotein surface) takes place in the Golgi
apparatus. Chylomicron formation is followed by exocytosis via the secretory pathway at the
basolateral surface of the enterocyte. The chylomicrons are released into the circulation via
the mesenteric lymph system, via the thoracic duct into the venous system, after which their
contents are systemically delivered.
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
55
2
FACTORS PROPOSED TO BE ASSOCIATED WITH PERSISTENT FAT
MALABSORPTION IN CF
Several factors have been proposed to contribute to CFPFM. For each of these factors we will
describe; (1) how these factors are affected in CF conditions; (2) what is known about the
relation of these factors with CFPFM; and, (3) what the result of intervention studies are in CF
patients and CF mice when improving or altering these factors (figure 1B).
3.1 INTESTINAL PH
1) CF conditions. CFTR dysfunction reduces pancreatic and duodenal bicarbonate secretion
(20,21). Gastric bicarbonate secretion in CF patients is not affected (22). Hence, the duodenal
pH is (on average) 1-2 units lower in CF patients compared with healthy controls. Also, CF
patients have significantly longer periods (postprandial) in which the duodenal pH drops
below 4 (23). The pH values in jejunal and ileal contents from CF patients vary from lower to
similar pH values compared with healthy controls (23). CFTRtm1Cam
knockout mice have
adequate pancreatic bicarbonate secretion, but probably do have reduced secretions of
duodenal bicarbonate (9). This might explain the observation in CFTRtm1Unc
knockout mice,
where duodenal pH is only minimally reduced (~0.25) (21).
2) Relation with CFPFM. A normal intestinal pH is essential for adequate digestion and
absorption of fat (10). Recently, Quinton et al. showed that adequate bicarbonate secretion is
essential for mucin expansion in the intestine (24,25). When mucin expansion is disturbed by
inadequate release of bicarbonate in CF conditions, viscous and sticky mucus might impair
translocation of lipids to the enterocyte. A strong relation has been described between
postprandial duodenal pH levels and the degree of fat malabsorption in CF patients treated
with pancreatic enzymes (26). These data indicate that interventions to increase intestinal pH
values might improve CFPFM.
3) Intervention studies. Two double-blind crossover studies evaluated fat absorption in
pediatric and adult CF patients after the addition of bicarbonate to enteric coated pancreatic
enzymes. In the first study, the average fat absorption increased from 75% to 82% of the
ingested amount (27). In this study, a beneficial effect on fat absorption was observed in
seventy-five percent of the patients. In the second study, the average fat absorption did not
increase, but fifty percent of patients did show an improvement in fat absorption (>5%) after
bicarbonate supplementation (28) . Due to the overall minimal observed changes in fat
absorption, pancreatic enzymes buffered with bicarbonate are not implemented in CF care.
In a systematic review, Jones A. evaluated the effect of acid suppressant therapy on fat
absorption and/or faecal fat excretion and nutritional status (29). While multiple studies
reported improved fat absorption or reduced faecal fat excretion after acid suppressive
Chapter 3
56
2
therapy, other studies reported no improvements. Improved fat absorption was not
accompanied with an improvement in nutritional status. The performed studies showed high
variability in the used dosage of pancreatic enzymes, choice of acidic suppressive medication
and differed in the inclusion criteria for the degree of fat malabsorption. Due to these large
inter-study differences, it remains difficult to draw an overall definite conclusion about the
effectiveness of acid suppressive drugs. Evaluating individual data in these intervention
studies, a subgroup of patients clearly showed improved fat absorption. The results indicate
that some CF patients do benefit from acid suppressive therapy. Why some patients respond,
but others show no improvement in fat absorption, illustrates that an altered pH is not the
only factor responsible CFPFM. Next to this, it is known that acid suppressive drugs can induce
small intestinal bacterial overgrowth (SIBO) and can alter bile salt metabolism, with a
potentially negative effect on fat absorption (30). Therefore, acid suppressive therapy should
be imposed with caution, especially in CF patients who are relatively more prone to SIBO. A
trial of proton pump inhibitors might be useful to evaluate its effectiveness in the individual
CF patient.
Bijvelds et al. investigated the effect of the acidic suppressive drug omeprazole on lipolysis
and uptake of lipolytic products in CFTRtm1CAM
knockout mice by using radio-isotope labelled
triglycerides and fatty acids. The researchers showed that lipolytic activity and lipid
absorption improved after omeprazole treatment (9). It is likely that increasing the intestinal
pH has a general positive effect on intestinal fat absorption, since lipolytic and post-lipolytic
activity also partially improved in the omeprazole treated wild type mice (9).
3.2 INTRA-LUMINAL BILE SALTS
1) CF conditions. CF patients have an increased loss of faecal bile salts (31,32). It is speculated
that the loss of bile salts is due to impaired bile salt uptake by intestinal mucosal alterations;
like thickening of the mucus barrier and/or SIBO (33). Because the biosynthesis of taurine is
limited in humans, the faecal loss of bile salts induces an increased glycine/taurine ratio of
conjugated bile salts in CF patients (34). In addition, CF patients have an altered bile salt
composition in gallbladder bile. Due to a quantitative increase in biliary cholate, the percent
contribution of cholate is higher, and the percent contribution of chenodeoxycholate and
deoxycholate is lower in the bile of CF patients (35).
2) Relation with CFPFM. It has been proposed that excessive faecal loss of bile salts diminishes
the bile salt pool in CF patients and consequently impairs fat absorption by reducing the
solubilization capacity of bile (31). However, a subsequent study indicated that the amount of
faecal bile salt excretion was not related to the degree of fat malabsorption in CF patients
(32). Strandvik et al. showed that adult CF patients have normal to large bile salt pool sizes
and similar amounts of duodenal bile salts as healthy controls (35). Bile salt synthesis was
normal or even increased, indicating that CF patients adequately compensate for the faecal
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
57
2
bile salt loss. We confirmed in CF mouse models that faecal loss of bile salts does not
influence the absorption of fat (9). Homozygous ∆F508 mice and CFTRtm1CAM
knockout mice
both exhibit, to the same extent, an increased faecal loss of bile salts, but only the CFTRtm1CAM
knockout mice had fat malabsorption (9). In conclusion, bile salt malabsorption in itself, at the
levels observed in CF patients and CF mice, does not seem to contribute to CFPFM.
Theoretically, the increased glycine/taurine ratio may impair fat absorption in an acidic
intestinal lumen. Due to the higher pKa of glycine, glycine conjugated bile salts are less able to
remain in micellar solution (36). In addition, part of the glycoconjuates are passively absorbed
in the in the proximal part of the intestine and are, in comparison to tauroconjugates, less
resistant to bacterial degradation (37,38).
Deoxycholate and chenodeoxycholate are relatively hydrophobic in comparison to other bile
salts. Therefore, a reduction in these bile salts may theoretically impair fat absorption by
diminishing the solubilization capacity of bile. In general, the percent contribution, and not
the quantitative amount, of deoxycholate and of chenodeoxycholate is diminished in CF
patients. Thus, the altered bile salt composition is not likely to contribute to CFPFM. This
hypothesis is supported by the fact that both homozygous ∆F508 mice and CFTR tm1CAM
knockout mice show these alterations, while only the CFTR tm1CAM
knockout mice exhibit fat
malabsorption (9).
3) Intervention studies. Multiple studies showed that taurine supplementation reduces faecal
fat excretion and improves the nutritional status of CF patients, particularly in patients with
severe steatorrhoea (39-41). However, the beneficial effect of taurine supplementation on fat
absorption is not unequivocally demonstrated (42-45) . Furthermore, the degree of fat
absorption did not relate to changes in the serum glycine/taurine ratio in CF children (45).
Altogether, the use of taurine supplementation remains controversial and is not implemented
in nutritional CF care.
3.3 INTESTINAL MUCOSAL ABNORMALITIES
1) CF conditions. Several intestinal mucosal abnormalities are described in CF patients. These
abnormalities include accumulation of viscous and sticky mucus, SIBO, ileal hyperthrophy and
villous atrophy, increased intestinal permeability and (chronic) inflammation of the small
intestine (46-53).
2) Relation with CFPFM. All the above mentioned factors may theoretically contribute to fat
malabsorption in CF patients, as they might impair adequate translocation of fatty acids in(to)
the enterocyte. In addition, SIBO might impair micelle formation by the bacterial
deconjugation of bile salts (54). However, no studies evaluated the actual contribution of
intestinal mucosal abnormalities to CFPFM.
Chapter 3
58
2
3) Intervention studies. The group of De Lisle et al. evaluated the effect of antibiotic treatment
on intestinal inflammation, mucus accumulation and SIBO in CFTRtm1Unc
knockout mice (55).
CFTRtm1Unc
knockout mice display a phenotype of SIBO with a 400 fold increase in small
intestinal bacterial content compared to wild type littermates (56). Broad spectrum antibiotic
treatment with ciprofloxacine and metronidazole did not only reduce the bacterial load in the
small intestine, but also decreased intestinal mucus accumulation and inflammation. More
importantly, three weeks of treatment substantially improved the body weight of these mice.
The same effect on the bacterial load, intestinal mucus accumulation, inflammation and body
weight was observed after laxative treatment (57). It remains to be elucidated whether the
growth benefit was due to improved fat absorption, reduced competition for nutrients by
intestinal bacteria or due to reduced intestinal inflammation. As O’Brien et al. showed that a
seven-day treatment with metronidazole reduced faecal fat excretion in four CF patients (32),
it is reasonable to assume that the increased weight is due to improved fat absorption
Although the effect on fat absorption was not assessed in either of these studies, the results
suggest that metronidazole might be a potential therapeutic option for increasing fat
absorption or nutritional status in CF patients.
One study evaluated the effect of probiotics on intestinal inflammation in CF patients (49). It
has been suggested that probiotics improve intestinal barrier function and modify the
immune response (58). Bruzzese et al. reported that a four week treatment with Lactobacillus
rhamnosus GG, reduced faecal calprotectin levels (marker for intestinal inflammation) in 8
out of 10 treated CF patients (49). The long term effect of probiotics on intestinal
inflammation was not evaluated, nor the effect on fat absorption or nutritional status. We
found only one mice study which evaluated fat absorption after probiotics; lactobacillus
supplementation increased intestinal absorption of dietary lipids in germ-free mice colonized
with human baby flora (59). Until now, the clinical value of probiotics and its relation with fat
absorption in CF patients is not clear.
3.4 SMALL INTESTINAL TRANSIT TIME
1) CF conditions. No studies exclusively investigated small intestinal transit in CF patients, but
several groups have evaluated the oro-cecal transit time (OCTT) (23). The lactulose/hydrogen
breath test was most commonly applied to for determination of the OCTT. Eighty five percent
of CF patients had a prolonged OCTT (at least +50%) in the fasted state (60,61). Another
method to evaluate intestinal activity is by measuring the intestinal muscle activity during the
inter-digestive state, also known as the migrating motor complex. The migrating motor
complex indirectly reflects the ability to transit food through the intestine. Hallberg et al.
showed that duodenal motility during fasting was normal in 8 out of 10 CF patients (22). This
contradiction with transit studies underlines the necessity to evaluate small intestinal transit
in CF patients. Especially, since intestinal smooth muscle activity shows an erratic pattern and
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
59
2
is unresponsive to cholinergic stimulation in CFTRtm1Unc
knockout mice (62). It has been
repeatedly shown, that these mice have a slower small intestinal transit time in comparison
to wild type littermates on the same diet (57-64).
2) Relation with CFPFM. Fat absorption partly depends on the time fat is in contact with the
absorptive epithelium of the intestine. A classical thought is that during fat malabsorption the
intestine prolongs the intestinal transit time (as a compensatory mechanism), in order to
enhance its ability to absorb fat; a phenomenon also known as the ‘ileal brake’ (65). It is
reasonable to assume that this compensatory mechanism also occurs in CFPFM as intestinal
transit time is prolonged in CF. However, prolonged intestinal transit time is also a risk factor
for the occurrence of SIBO in CF conditions (8). Therefore, one might ask, if prolonged
intestinal transit time might actually induce or worsen CFPFM. The relation between intestinal
transit time and CFPFM is not yet evaluated.
3) Intervention studies. The research group of De Lisle et al. performed several (intervention)
studies on gastro-intestinal transit in CFTRtm1Unc
knockout mice (57,62,64). Oral laxative
treatment improved circular smooth muscle function in the small intestine and normalized
intestinal transit time (62). Moreover, as earlier described, laxative treatment reduced
intestinal mucus accumulation, eradicated overgrowth of bacteria in the small intestine and
improved body weight (57). Unfortunately, fat absorption was not evaluated in these studies.
It is tempting to speculate that laxative treatment may also improve the nutritional status in
CF patients, but before recommendations could be made, more clinical data are needed.
3.5 ESSENTIAL FATTY ACID DEFICIENCY
1) CF conditions. Despite high fat diets and PERT, CF patients can still have a deficiency in
essential fatty acids (EFA) (14). In general, many CF patients have subnormal levels of serum
and tissue linoleic acid (LA) and cervonic acid (DHA), often with increased levels of arachidonic
acid (AA) (14,65,66). The reduction in EFA levels is, apart from fat malabsorption, due to many
other CF-related factors. The exact mechanism behind the alterations in EFA’s still requires
further exploration, but increased lipid turnover in cellular membranes and increased
oxidation of EFA’s are considered to play the most prominent role (14,67-70). The prevalence
of EFA deficiency in CF patients strongly depends on which biochemical marker is applied as
diagnostic criterion. Magbool et al. recently demonstrated that the LA status in serum
phospholipids is more relevant than the triene/tetraene ratio as clinical indicator for EFA
deficiency in CF patients (71). When serum LA values are applied as a diagnostic criterion for
EFA-deficiency, around forty percent of pediatric CF patients are classified as EFA deficient .
The literature is not consistent on the occurrence of EFA deficiency in CF mice. Werner et al.
reported that the LA and DHA content in serum and jejunum was not altered in CFTRtm1Cam
knock-out mice and homozygous delta F508 mice (72), while Freedman et al. reported
decreased levels of DHA in the ileum of CFTRtm1Unc
knock-out mice (73).
Chapter 3
60
2
2) Relation with CFPFM. In relation to fat malabsorption, the EFA content in the intestine is
very important. EFA’s provide fluidity and flexibility to a cell membrane and play an important
role in the maintenance of cell membrane functionality of the enterocyte. EFA deficiency in
itself can cause fat malabsorption by affecting the absorption capacity of the intestine (74).
Apart from reduced LA and DHA levels, increased levels of AA in the CF intestine could also
(indirectly) affect fat absorption, by contributing to intestinal inflammation, mucus secretion
and intestinal smooth muscle relaxation (75) . It has been shown that EFA deficiency affects
both the intra-luminal and intra-cellular process of fat absorption, as exemplified in (non-CF)
animal models (69). While the degree of EFA deficiency in CF patients is generally mild, we
cannot exclude the possibility that it might interact with fat absorption. Especially since serum
EFA levels do not necessarily correlate with EFA levels in the intestine (76,77). Ex vivo
experiments on intestinal biopsies from CF patients indicated that abnormal intracellular lipid
processing occurs in the enterocyte (76). A low incorporation of [14
C]palmitic acid in
triglycerides was found in cultures of duodenal biopsies after 18 hours of incubation. The
secretion of triglyceride-rich chylomicrons was reduced in comparison to intestinal biopsies
from healthy controls (76). Whether this impaired lipid transport is due to EFA deficiency, to
aberrant CFTR itself or to a complex relation between the two, requires further investigation
(78,79). To what extent reduced intestinal EFA levels in CF patients interfere with the actual
percent absorption of dietary fat intake is unknown. An association has been described
between the degree of EFA deficiency and the nutritional status of CF patients (80), but EFA
deficiency can also be present in well-nourished CF patients (81). Thus, the clinical relevance
of the reduced EFA levels in CF patients on fat malabsorption remains to be determined.
3) Intervention studies. Despite the abundance of (intervention) studies (66,82-85), no study
determined the effect of EFA supplementation on CFPFM. Improved DHA levels in duodenal
mucosa of CF patients after 70 mg of DHA/kg body weight for six weeks has been reported,
but whether this improved fat absorption or nutritional status was not described. Reviews on
omega 3 fatty acid supplementation or DHA supplementation alone, conclude that there is
still insufficient evidence for the therapeutic effect of omega 3 fatty acids to recommend the
routine use in CF patients (82,85). Other studies showed that energy supplements rich in LA
improved the body weight in CF patients, but it is hard to differentiate between the effects of
LA from that of a high energy intake (83,84). Accordingly, we do not recommend the routine
use of EFA supplements in CF for reasons to enhance fat absorption, but it could be
considered to improve nutritional status in individual cases.
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
61
2
CF patients CF mice
(primarily Cftr-/- mice)
Treatment effect CF patients
(fat absorption+weight)
Treatment effect CF mice
(fat absorption+weight)
Intestinal pH Duodenum: ↓ Duodenum: minor ↓ Bicarbonate enteric coated pancreatic enzymes: fat absorption: ↔ or minor ↑ weight: not known
Acid suppressant drugs:triglyceride uptake: ↑ fatty acid uptake: ↑
weight: not known
Jejunum: ↓ or ↔ Jejunum: not known
Ileum: ↔ Ileum: not known Acid suppressant drugs: fat absorption: ↔ or ↑ weight: ↔
Intraluminal bile salts
Fecal bile salt excretion: ↑ Fecal bile salt excretion: ↑ Taurine supplementation: fat absorption: ↔ or minor ↑ weight: ↔ or minor ↑
No interventional studies
Glycine/taurine ratio: ↑ Altered bile salt composition
Altered bile salt composition
Intestinal wall abnormalities
SIBO (~ 40%) SIBO No interventional studies Antibiotic treatment: fat absorption: not known weight: ↑ Intestinal mucus accumulation Intestinal mucus accumulation
Intestinal inflammation Intestinal inflammation
Intestinal permeability Intestinal permeability
Small intestinal transit time
Not known Prolonged No interventional studies Laxative treatment: fat absorption: not known weight: ↑
Essential fatty acid deficiency
Serum/tissue LA and DHA: ↓ or ↔
Serum/tissue LA and DHA: ↓ or ↔
Energy supplements with LA: fat absorption: not known weight: ↑
No interventional studies
Serum/tissue AA: ↑ or ↔ Serum/tissue AA: ↑ or ↔
Table 1. Factors that are suggested to contribute to fat malabsorption in cystic fibrosis (CF). Table presents an
overview of the described alterations observed in CF patients and CF mice and, interventional studies that evaluated
fat absorption and/or nutritional status. ↔ : not affected, ↑: increased, ↓: decreased . The described data on CF
mice are mainly based on studies in CFTR knockout mice. References are described in the text. SIBO: small intestinal
bacterial overgrowth. LA: linoleic acid. DHA: cervonic acid. AA: arachidonic acid
SUMMARY, CLINICAL IMPLICATIONS AND FUTURE PERSPECTIVES
Several factors have been proposed to contribute to CFPFM (table 1). Based on the available
data we consider it not likely that bile salt alterations do contribute to CFPFM. The
quantitative contribution of intestinal mucosal abnormalities, small intestinal transit time and
EFA deficiency to CFPFM remains to be determined. Based on the findings in CF mouse
models, CF patients with signs of small bacterial overgrowth might benefit from treatment
with broad-spectrum antibiotics. Antibiotic treatment should preferably be evaluated with
changes in nutritional status or fat absorption, using three day fat balances. However, we
should realize that three day fat balances show high intra-individual variability, which
exemplifies the need for novel methodologies to assess fat absorption in CF patients. So far,
the influence of reduced duodenal pH levels seems most convincingly demonstrated to play a
role in CFPFM. Although not necessarily beneficial in all CF patients, a trial of acid suppressive
drugs might be helpful to optimize fat absorption in individual cases.
Chapter 3
62
2
REFERENCE LIST
(1) Corey M, McLaughlin F, Williams M, Levison H. A comparison of survival, growth, and
pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol
1988;41(6):583-591.
(2) Beker LT, Russek-Cohen E, Fink RJ. Stature as a prognostic factor in cystic fibrosis survival. J
Am Diet Assoc 2001;101(4):438-442.
(3) Pencharz P, Durie P. Pathogenesis of malnutrition in cystic fibrosis, and its treatment.
Clinical Nutrition 2000;19(6):387-394.
(4) Littlewood JM, Wolfe SP, Conway SP. Diagnosis and treatment of intestinal malabsorption
in cystic fibrosis. Pediatr Pulmonol 2006;41(1):35-49.
(5) FitzSimmons SC, Burkhart GA, Borowitz D, Grand RJ, Hammerstrom T, Durie PR, et al. High-
dose pancreatic-enzyme supplements and fibrosing colonopathy in children with cystic
fibrosis. N Engl J Med 1997;336(18):1283-1289.
(6) Taylor C, Hillel P, Ghosal S, Frier M, Senior S, Tindale W, et al. Gastric emptying and
intestinal transit of pancreatic enzyme supplements in cystic fibrosis. Arch Dis Child
1999;80(2):149-152.
(7) Kalivianakis M, Minich DM, Bijleveld CMA, van Aalderen WMC, Stellaard F, Laseur M, et al.
Fat malabsorption in cystic fibrosis patients receiving enzyme replacement therapy is due to
impaired intestinal uptake of long-chain fatty acids. Am J Clin Nutr 1999 01;69(1):127-134.
(8) Borowitz D, Durie PR, Clarke LL, Werlin SL, Taylor CJ, Semler J, et al. Gastrointestinal
outcomes and confounders in cystic fibrosis. J Pediatr Gastroenterol Nutr 2005;41(3):273-285.
(9) Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat
absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am
J Physiol Gastrointest Liver Physiol 2005 04;288(4):G646-G653.
(10) Verkade HJ, Tso P. Biophysics of intestinal luminal lipids. Intestinal Lipid Metabolism:
Springer; 2001. p. 1-18.
(11) Zoppi G, Andreotti G, Pajno-Ferrara F, Njai D, Gaburro D. Exocrine pancreas function in
premature and full term neonates. Pediatr Res 1972;6(12):880-886.
(12) Abrams CK, Hamosh M, Hubbard VS, Dutta SK, Hamosh P. Lingual lipase in cystic fibrosis.
Quantitation of enzyme activity in the upper small intestine of patients with exocrine
pancreatic insufficiency. J Clin Invest 1984;73(2):374-382.
(13) Minich DM, Voshol PJ, Havinga R, Stellaard F, Kuipers F, Vonk RJ, et al. Biliary
phospholipid secretion is not required for intestinal absorption and plasma status of linoleic
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
63
2
acid in mice. Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids
1999;1441(1):14-22.
(14) Peretti N, Marcil V, Drouin E, Levy E. Mechanisms of lipid malabsorption in Cystic Fibrosis:
the impact of essential fatty acids deficiency. Nutr Metab (Lond) 2005;2(1):11-19.
(15) Vassileva G, Huwyler L, Poirier K, Agellon LB, Toth MJ. The intestinal fatty acid binding
protein is not essential for dietary fat absorption in mice. The FASEB Journal
2000;14(13):2040-2046.
(16) Levy E, Ménard D, Delvin E, Montoudis A, Beaulieu J, Mailhot G, et al. Localization,
function and regulation of the two intestinal fatty acid-binding protein types. Histochem Cell
Biol 2009;132(3):351-367.
(17) Goudriaan JR, Dahlmans VE, Febbraio M, Teusink B, Romijn JA, Havekes LM, et al.
Intestinal lipid absorption is not affected in CD36 deficient mice. Cellular Lipid Binding
Proteins: Springer; 2002. p. 199-202.
(18) Lightfoot FG, Lauretta Grau MS S, Satchitanandum M. Lipid accumulation in jejunal and
colonic mucosa following chronic cholestyramine (Questran) feeding. Dig Dis Sci
1985;30(5):468-476.
(19) Werner A, Havinga R, Perton F, Kuipers F, Verkade HJ. Lymphatic chylomicron size is
inversely related to biliary phospholipid secretion in mice. American Journal of Physiology-
Gastrointestinal and Liver Physiology 2006;290(6):G1177-G1185.
(20) Tang L, Fatehi M, Linsdell P. Mechanism of direct bicarbonate transport by the CFTR anion
channel. Journal of Cystic Fibrosis 2009;8(2):115-121.
(21) Kaur S, Norkina O, Ziemer D, Samuelson LC, De Lisle RC. Acidic duodenal pH alters gene
expression in the cystic fibrosis mouse pancreas. American Journal of Physiology-
Gastrointestinal and Liver Physiology 2004;287(2):G480-G490.
(22) Hallberg K, Abrahamsson H, Dalenbäck J, Fändriks L, Strandvik B. Gastric secretion in
cystic fibrosis in relation to the migrating motor complex. Scand J Gastroenterol
2001;36(2):121-127.
(23) Gregory P. Gastrointestinal pH, motility/transit and permeability in cystic fibrosis. J
Pediatr Gastroenterol Nutr 1996;23(5):513-523.
(24) Garcia MAS, Yang N, Quinton PM. Normal mouse intestinal mucus release requires cystic
fibrosis transmembrane regulator–dependent bicarbonate secretion. J Clin Invest
2009;119(9):2613-2622.
(25) Quinton PM. Birth of mucus. American Journal of Physiology-Lung Cellular and Molecular
Physiology 2010;298(1):L13-L14.
Chapter 3
64
2
(26) PJ Robinson MB B. Duodenal pH in cystic fibrosis and its relationship to fat malabsorption.
Dig Dis Sci 1990;35(10):1299-1304.
(27) Brady MS, Garson JL, Krug SK, Kaul A, Rickard KA, Caffrey HH, et al. An enteric-coated
high-buffered pancrelipase reduces steatorrhea in patients with cystic fibrosis: a prospective,
randomized study. J Am Diet Assoc 2006;106(8):1181-1186.
(28) Kalnins D, Ellis L, Corey M, Pencharz PB, Stewart C, Tullis E, et al. Enteric-coated
pancreatic enzyme with bicarbonate is equal to standard enteric-coated enzyme in treating
malabsorption in cystic fibrosis. J Pediatr Gastroenterol Nutr 2006;42(3):256-261.
(29) Jones DE, Palmer JM, Kirby JA, De Cruz DJ, McCaughan GW, Sedgwick JD, et al.
Experimental autoimmune cholangitis: a mouse model of immune-mediated cholangiopathy.
Liver 0226 10;20(5):351-356.
(30) Shindo K, Machida M, Fukumura M, Koide K, Yamazaki R. Omeprazole induces altered bile
acid metabolism. Gut 1998;42(2):266-271.
(31) Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in
cystic fibrosis. Gut 1976;17(4):295.
(32) O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid
malabsorption in cystic fibrosis. Gut 1993;34(8):1137.
(33) Fondacaro JD, Heubi JE, Kellogg FW. Intestinal bile acid malabsorption in cystic fibrosis: a
primary mucosal cell defect. Pediatr Res 1982;16(6):494-498.
(34) Roy CC, Weber AM, Morin CL, Combes JC, Nusslé D, Mégevand A, et al. Abnormal biliary
lipid composition in cystic fibrosis. N Engl J Med 1977;297(24):1301-1305.
(35) Strandvik B, Einarsson K, Lindblad A, Angelin B. Bile acid kinetics and biliary lipid
composition in cystic fibrosis. J Hepatol 1996 07;25(1):43-48.
(36) Regan PT, Malagelada J-, Dimagno EP, Go VLW. Reduced intraluminal bile acid
concentrations and fat maldigestion in pancreatic insufficiency: Correction by treatment.
Gastroenterology 1979 08/01;77(2):285-289.
(37) Krag E, Phillips SF. Active and passive bile acid absorption in man. Perfusion studies of the
ileum and jejunum. J Clin Invest 1974;53(6):1686.
(38) Hepner GW, Sturman JA, Hofmann AF, Thomas PJ. Metabolism of Steroid and Amino Acid
Moieties of Conjugated Bile Acids in Man III. CHOLYLTAURINE (TAUROCHOLIC ACID). J Clin
Invest 1973;52(2):433-440.
(39) Belli DC, Levy E, Darling P, Leroy C, Lepage G, Giguère R, et al. Taurine improves the
absorption of a fat meal in patients with cystic fibrosis. Pediatrics 1987;80(4):517-523.
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
65
2
(40) Darling PB, Lepage G, Leroy C, Masson P, Roy CC. Effect of taurine supplements on fat
absorption in cystic fibrosis. Pediatr Res 1985;19(6):578-582.
(41) Smith LJ, Lacaille F, Lepage G, Ronco N, Lamarre A, Roy CC. Taurine Decreases Fecal Fatty
Acid and Sterol Excretion in Cystic Fibrosis: A Randomized Double-blind Trial. Arch Pediatr
Adolesc Med 1991;145(12):1401-1404.
(42) Colombo C, Arlati S, Curcio L, Maiavacca R, Garatti M, Ronchi M, et al. Effect of taurine
supplementation on fat and bile acid absorption in patients with cystic fibrosis. Scand J
Gastroenterol 1988;23(S143):151-156.
(43) Merli M, Bertasi S, Servi R, Diamanti S, Martino F, De Santis A, et al. Effect of a medium
dose of ursodeoxycholic acid with or without taurine supplementation on the nutritional status
of patients with cystic fibrosis: a randomized, placebo-controlled, crossover trial. J Pediatr
Gastroenterol Nutr 1994;19(2):198-203.
(44) De Curtis M, Santamaria F, Ercolini P, Vittoria L, De Ritis G, Garofalo V, et al. Effect of
taurine supplementation on fat and energy absorption in cystic fibrosis. Arch Dis Child
1992;67(9):1082-1085.
(45) Thompson GN, Robb TA, Davidson GP. Taurine supplementation, fat absorption, and
growth in cystic fibrosis. J Pediatr 1987;111(4):501-506.
(46) Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal
permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.
Pediatr Res 1996;39(6):985-991.
(47) Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic
fibrosis: importance of breath testing. J Paediatr Child Health 1998;34(1):79-82.
(48) Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in
cystic fibrosis: combined hydrogen and methane measurements are required for its detection.
Acta Biochim Pol 2009;56(4):631.
(49) Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal
inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.
Aliment Pharmacol Ther 2004;20(7):813-819.
(50) Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic
inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res
2000;47(3):344-350.
(51) Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the
small bowel of patients with cystic fibrosis. Pediatr Res 1998;43(2):234-239.
(52) Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic
fibrosis. Arch Dis Child 2000;82(5):394-399.
Chapter 3
66
2
(53) Quigley EM, Quera R. Small intestinal bacterial overgrowth: roles of antibiotics,
prebiotics, and probiotics. Gastroenterology 2006;130(2):S78-S90.
(54) Martin FJ, Dumas M, Wang Y, Legido-Quigley C, Yap IK, Tang H, et al. A top-down systems
biology view of microbiome-mammalian metabolic interactions in a mouse model. Molecular
systems biology 2007;3(1):1-16.
(55) De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr
2006;42(1):46-52.
(56) Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun
2004;72(10):6040.
(57) De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus
accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small
intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology
2007;293(3):G577-G584.
(58) Perdigón G, Fuller R, Raya R. Lactic acid bacteria and their effect on the immune system.
Curr Issues Intestinal Microbiol 2001;2(1):27-42.
(59) Martin FJ, Wang Y, Sprenger N, Yap IK, Lundstedt T, Lek P, et al. Probiotic modulation of
symbiotic gut microbial–host metabolic interactions in a humanized microbiome mouse
model. Molecular systems biology 2008;4(1):1-14.
(60) Dalzell A, Freestone N, Billington D, Heaf D. Small intestinal permeability and orocaecal
transit time in cystic fibrosis. Arch Dis Child 1990;65(6):585-588.
(61) Bali A, Stableforth DE, Asquith P. Prolonged small-intestinal transit time in cystic fibrosis.
Br Med J (Clin Res Ed) 1983;287(6398):1011-1013.
(62) De Lisle R, Sewell R, Meldi L. Enteric circular muscle dysfunction in the cystic fibrosis
mouse small intestine. Neurogastroenterology & Motility 2010;22(3):341-e87.
(63) Lin HC. Ileal brake: neuropeptidergic control of intestinal transit. Curr Gastroenterol Rep
2006;8(5):367-373.
(64) De Lisle RC. Altered transit and bacterial overgrowth in the cystic fibrosis mouse small
intestine. Am J Physiol Gastrointest Liver Physiol 2007;293(1):G104-G111.
(65) Freedman SD, Blanco PG, Zaman MM, Shea JC, Ollero M, Hopper IK, et al. Association of
cystic fibrosis with abnormalities in fatty acid metabolism. N Engl J Med 2004;350(6):560-569.
(66) Jumpsen JA, Brown NE, Thomson AB, Paul Man S, Goh YK, Ma D, et al. Fatty acids in blood
and intestine following docosahexaenoic acid supplementation in adults with cystic fibrosis.
Journal of Cystic Fibrosis 2006;5(2):77-84.
Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice
67
2
(67) Andersson C, Al-Turkmani MR, Savaille JE, Alturkmani R, Katrangi W, Cluette-Brown JE, et
al. Cell culture models demonstrate that CFTR dysfunction leads to defective fatty acid
composition and metabolism. J Lipid Res 2008;49(8):1692-1700.
(68) Strandvik B. Fatty acid metabolism in cystic fibrosis. Prostaglandins, Leukotrienes and
Essential Fatty Acids 2010;83(3):121-129.
(69) Levy E, Garofalo C, Thibault L, Dionne S, Daoust L, Lepage G, et al. Intraluminal and
intracellular phases of fat absorption are impaired in essential fatty acid deficiency. American
Journal of Physiology-Gastrointestinal and Liver Physiology 1992;262(2):G319-G326.
(70) Rogiers V, Dab I, Michotte Y, Vercruysse A, Crokaert R, Vis HL. Abnormal Fatty Acid
Turnover in the Phospholipids of the Red Blood Cell Membranes of Cystic Fibrosis Patients
( in Vitro Study&rpar. Pediatr Res 1984;18(8):704-709.
(71) Maqbool A, Schall JI, Garcia-Espana JF, Zemel BS, Strandvik B, Stallings VA. Serum linoleic
acid status as a clinical indicator of essential fatty acid status in children with cystic fibrosis. J
Pediatr Gastroenterol Nutr 2008;47(5):635-644.
(72) Werner A, Bongers ME, Bijvelds MJ, de Jonge HR, Verkade HJ. No indications for altered
essential fatty acid metabolism in two murine models for cystic fibrosis. J Lipid Res
2004;45(12):2277-2286.
(73) Freedman SD, Katz MH, Parker EM, Laposata M, Urman MY, Alvarez JG. A membrane lipid
imbalance plays a role in the phenotypic expression of cystic fibrosis in cftr−/− mice.
Proceedings of the National Academy of Sciences 1999;96(24):13995-14000.
(74) Snipes RL. The effects of essential fatty acid deficiency on the ultrastructure and
functional capacity of the jejunal epithelium. Lab Invest 1968 Feb;18(2):179-189.
(75) De Lisle RC, Meldi L, Flynn M, Jansson K. Altered eicosanoid metabolism in the cystic
fibrosis mouse small intestine. J Pediatr Gastroenterol Nutr 2008;47(4):406-416.
(76) Peretti N, Roy CC, Drouin E, Seidman E, Brochu P, Casimir G, et al. Abnormal intracellular
lipid processing contributes to fat malabsorption in cystic fibrosis patients. American Journal
of Physiology-Gastrointestinal and Liver Physiology 2006;290(4):G609-G615.
(77) Korotkova M, Strandvik B. Essential fatty acid deficiency affects the fatty acid composition
of the rat small intestinal and colonic mucosa differently. Biochimica et Biophysica Acta (BBA)-
Molecular and Cell Biology of Lipids 2000;1487(2):319-325.
(78) Mailhot G, Ravid Z, Barchi S, Moreau A, Rabasa-Lhoret R, Levy E. CFTR knockdown
stimulates lipid synthesis and transport in intestinal Caco-2/15 cells. American Journal of
Physiology-Gastrointestinal and Liver Physiology 2009;297(6):G1239-G1249.
Chapter 3
68
2
(79) Mailhot G, Rabasa-Lhoret R, Moreau A, Berthiaume Y, Levy E. CFTR depletion results in
changes in fatty acid composition and promotes lipogenesis in intestinal Caco 2/15 cells. PloS
one 2010;5(5):e10446.
(80) Shoff SM, Ahn H, Davis L, Lai H. Temporal associations among energy intake, plasma
linoleic acid, and growth improvement in response to treatment initiation after diagnosis of
cystic fibrosis. Pediatrics 2006;117(2):391-400.
(81) Roulet M, Frascarolo P, Rappaz I, Pilet M. Essential fatty acid deficiency in well nourished
young cystic fibrosis patients. Eur J Pediatr 1997;156(12):952-956.
(82) Van Biervliet S, Van Biervliet J, Robberecht E, Christophe A. Docosahexaenoic acid trials in
cystic fibrosis: a review of the rationale behind the clinical trials. Journal of Cystic Fibrosis
2005;4(1):27-34.
(83) Steinkamp G, Demmelmair H, Rühl-Bagheri I, von der Hardt H, Koletzko B. Energy
supplements rich in linoleic acid improve body weight and essential fatty acid status of cystic
fibrosis patients. J Pediatr Gastroenterol Nutr 2000;31(4):418-423.
(84) Van Egmond A, Kosorok MR, Koscik R, Laxova A, Farrell PM. Effect of linoleic acid intake
on growth of infants with cystic fibrosis. Am J Clin Nutr 1996;63(5):746-752.
(85) Coste TC, Armand M, Lebacq J, Lebecque P, Wallemacq P, Leal T. An overview of
monitoring and supplementation of omega 3 fatty acids in cystic fibrosis. Clin Biochem
2007;40(8):511-520.
CHAPTER 3
INCREASE OF SERUM GAMMA
GLUTAMYLTRANSFERASE (GGT)
ASSOCIATED WITH THE
DEVELOPMENT OF CIRRHOTIC
CYSTIC FIBROSIS LIVER DISEASE PREDICTION OF CIRRHOTIC CYSTIC FIBROSIS LIVER DISEASE.
Submitted
Frank A.J.A. Bodewes1
Hubert. P.J. van der Doef2
Roderick H.J. Houwen2
Henkjan J Verkade1
1Department of Pediactric Gastroentrology and Hepatology, Beatrix Children’s Hospital, University
Medical Center Groningen, University of Groningen, the Netherlands
2Department of Pediactric Gastroentrology and Hepatology University Medical Center Utrecht, the
Netherlands
Chapter 3
70
3
ABSTRACT
Background:
Cirrhotic liver disease (CCFLD) develops in 5-10% of CF patients. Identification of patients at
risk on CCFLD is potentially beneficial for preventive treatment. We studied the evolution of
liver enzymes (ALT, AST and GGT) in years preceding the diagnosis of CCFLD.
Methods:
We analyzed medical records of 277 pediatric CF patients. The time point of CCFLD diagnosis
was defined as the date when the ultrasonographic macronodular liver and the clinical
splenomegaly were first recorded. We analyzed liver enzymes in the 2 years preceding the
diagnosis of CCFLD. We compared these results with the annual liver enzymes of no-CCFLD
controls (>15 years of age and no ultrasonographic or physical signs of CCFLD).
Results:
At group level the median GGT, and not AST or ALT, of CCFLD patients, in the 2 years
preceding their diagnosis, significantly higher than the median of all GGT results of no-CCFLD
controls (45 vs. 17 U/l, respectively, P<0.001). The value of a single GGT result, for predicting
future CCFLD, was low. However, for CF patients with a mean GGT>35 U/L, based on repeated
measurements, the Odds ratio to develop CCFLD was 39 (95% CI: [9-175], sensitivity: 95%,
specificity: 64%, positive predictive value: 50%).
Conclusion:
In pediatric CF patients a persistent, high-normal, serum GGT is strongly associated with the
diagnosis of CCFLD within 2 years. The prognostic value of a single GGT measurement remains
limited. Our results indicate that groups of patients at increased risk for CCFLD can be
identified on the basis of repeated GGT measurements.
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
71
3
INTRODUCTION
Cirrhotic cystic fibrosis liver disease (CCFLD) develops in 5-10% of cystic fibrosis patients (1). It
is a serious complication of CF and patients with CCFLD tend to have a more severe CF
phenotype than CF patients without liver disease (2, 3).CCFLD is characterized by extensive
and often inhomogeneous cirrhosis. Clinically, CCFLD patients frequently have splenomegaly,
hypersplenism and complications of portal hypertension, including variceal bleeding and
ascites (4, 5). Liver synthesis and detoxification functions are often spared, and the need for
primary liver transplantation for CF liver disease has remained rather rare (6, 7).
CCFLD is an acquired complication of CF and is not yet present in infancy. Most CCFLD
patients develop the disease during childhood and have established disease before puberty
(8). The diagnosis of CCFLD is mostly based on clinical, imaging and biochemical (liver
enzymes) parameters. CCFLD patients develop splenomegaly and thrombocytopenia (9). The
specific findings of hepatic nodularity and splenomegaly on ultrasound (US) are reported as
reliable markers for advanced fibrosis with only limited discrepancy with liver biopsy (10).
Histology of liver biopsies specimens typically shows a severe bridging type of portal fibrosis
and proliferation and destruction of bile ducts (11).
The recognition CCFLD in an early phase and/or the identification of patients at risks may
have relevant clinical and therapeutic consequences (12). To date the only treatment option
used for CCFLD in clinical practice is ursodeoxycholic acid (UDCA)(13). It has been
hypothesized that UDCA improves or recovers the compromised or obstructed bile flow in CF
conditions. Theoretically, it would seem favorable to start treatment of CCFLD either in
patients at risk (prevention) or an early phase of the disease. Not only UDCA, but also other
bile salt analogues and anti-fibrogenic agents are currently under development and evaluated
(14). Identification of patients at risk for CCFLD or in an early, preclinical stage of the disease
would be of value to test the preventive and/or disease course modifying capacity of these
agents.
Diagnosis of CCFLD is presently only possible in the cirrhotic phase of the disease. To date no
reliable diagnostic tool or measurement are established to predict patients at risk for CCFLD
or to recognize the early phase of the disease. Different clinical signs are suggested to be
related to the development of CCFLD. However, no strong relations are established between,
for example, hepatic US findings and the risk for the development of CCFLD (15, 16).
Liver enzymes such as the transaminases AST (aspartate transaminase) and ALT (alanine
transaminases) and gamma-glutamyl transpeptidase (GGT) are frequently evaluated during
routine clinical checkups of CF patients (17). Elevated AST, ALT, and GGT are sometimes
regarded as indicators for the presence or development of CFLD. Colombo et al. reported that
UDCA treatment in CF patients frequently corrects the elevation of transaminases (18). It
Chapter 3
72
3
needs to be realized, however, that elevation of liver enzymes occurs frequently and
transiently in CF (19). These elevations of transaminases could be induced by hepatotoxic
therapies like antibiotics or be a para-infectious phenomenon associated with pulmonary
exacerbation.
The relation between liver enzymes and the presence of CCFLD was first studied by Potter et
al.(17). These scientists related liver biopsy findings of 43 CF patients to their (ALT, AST and
GGT levels at the time of biopsy. The biopsies were performed based on clinical indications
for liver disease including hepatomegaly, abnormal liver function tests, splenomegaly or
esophageal varices. Potter et al. found liver fibrosis (grade≤3) in 37% of their biopsies and that
GGT, at the time of the biopsy, did not correlate with the presence of fibrosis. Unfortunately,
no historical biochemical results were available to include in the analysis. Therefore, the
results of this study could not address the role of liver enzymes during the development of
CCFLD. Lindblad et al. reported on a historical cohort of liver biopsy results in 41 CF patients
(4). Nine out of these 41 patients had histology proven cirrhosis, of which 5 indeed had
clinical signs of cirrhosis. In this cohort the sensitivity of liver enzymes was 100% and the
specificity 41% for the presence of moderate or severe fibrosis and cirrhosis.
In this retrospective, controlled, study we aimed to identify potential biochemical risk factors
for the future development of CCFLD. Therefore we focused evolution of liver enzymes (ALT,
AST and GGT) in years preceding the diagnosis of CCFLD in patients with already established
cirrhosis.
METHODS
Study cohort
We performed a retrospective analysis in a cohort that contained all pediatric CF patients (2 –
18 years) from the Cystic Fibrosis centers of the University Medical Center Utrecht and the
Beatrix Children’s Hospital, University Medical Center Groningen, The Netherlands (reference
date January 1st 2007). According to the clinical protocols patients were seen and
evaluated at least yearly in our CF centers. The annual medical checkup included blood testing
for ALT, AST and GGT and ultrasonography of liver and spleen.
Study method
We defined the existence of cirrhosis as the ultrasonographic appearance of multilobular
macronodularity of the liver and the presence of splenomegaly (10) For this purpose we
reviewed the annual radiology reports for the ultrasonographic description of
macronodularity. Additionally we reviewed medical records for reported an enlarged spleen
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
73
3
on physical examination or ultrasonographic enlarged spleen compared to the maximum
references spleen size according to age (20). We reviewed all biochemistry results for ALT,
AST and GGT. We defined the upper limit of normal for AST, ALT and GGT of 50 U/l (21). Since
liver enzymes could be transiently elevated due to neonatal cholestasis in CF patients,
unrelated to the development of CCFLD, we excluded liver enzymes results obtained before
the age of 2 years. Based on these results we categorized patients into 4 groups:
A) Macronodularity and splenomegaly
B) Macronodularity without splenomegaly
C) Splenomegaly without macronodularity
D) No established macronodularity or splenomegaly
CCFLD study group
We considered group A (macronodularity and splenomegaly) as patients with an established
diagnosis of CCFLD. We defined the date of CCFLD diagnosis as the first evaluation date on
which the patient met the CCFLD criteria of ultrasonographic liver macronodularity and
splenomegaly.
No-CCFLD control group
From group D (no macronodularity and no splenomegaly) we selected a no-CCFLD control
group. This group consisted of CF patients that, at the age of 15 years, had never developed
any signs of CCFLD (e.g. both normal liver ultrasounds and no reported splenomegaly). Of this
group we use all the annually collected liver enzymes results after the second year of life.
Statistics
For statistical analysis, we used IBM SPPS version 20. For comparison of the nominal
variables, we use the Chi square testing or Fischer exact, when appropriate. For comparison
of the continuous variables, we used the Mann–Whitney U test. We use receiver operating
characteristic (ROC) to determine AOC and the cutoff value. We used contingency tables to
analyze frequency distribution and risk ratios.
Chapter 3
74
3
Figure 1. Schematic representation of the composition of the studies population. Group A are patients
with cirrhotic cystic fibrosis liver disease (CCFLD) defined as ultrasonographic macronodularity and
splenomegaly, Group B are patients with macronodularity and no splenomegaly, group C are patients
with splenomegaly without macro nodular disturbances of the liver on US and group D with no signs of
CCFLD. Group D is referred to as the reference population. The study group contained all patients of
group A of whom biochemistry results were available in period 2 years prior to the date of diagnosis of
CCFLD. The control group was defined as CF patients out of the reference population that had not
developed any signs of CCFLD (e.g. normal liver ultra sound and no splenomegaly) at the age of 15 years.
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
75
3
RESULTS
Study cohort
The total study cohort consisted of 277 pediatric CF patients (Figure 1). Nineteen (7%)
patients met the criteria for established CCFLD (Group A). Additionally we identified patients
with either ultrasonographic signs of macronodularity but no splenomegaly [group B: N=10
(4%)], or presence of splenomegaly but no macronodularity on liver ultra sound [group C:
N=12 (4%)]. Two hundred thirty six (85%) patients had not displayed any signs of
ultrasonographic macronodularity or splenomegaly (group D, reference group). In the current,
retrospective, study we found a relatively high and variable age of diagnosis for CF. This is
explained by the fact that in the Netherlands a nationwide neonatal CF screening program
was introduced only in 2011 (Table 1).
We evaluated different potential risk factors or factors reported being related with the
development of cystic fibrosis related liver disease (Table 1). Almost all patients with signs of
liver involvement had started UDCA treatment. We found a significantly increased prevalence
of DIOS in the two patients groups with macro-nodular abnormalities on liver ultrasound
(group A and B). In patients with only splenomegaly (group C), on the other hand, the
prevalence of DIOS was low.
We analyzed the relationships between liver enzymes and prevalence of liver involvement in
CF (Table 1). We did this by evaluating the proportion of patients per study group, in which
any liver enzymes results had been above the upper limit of normal (AST, ALT and GGT> 50
U/l). We found no significant higher proportion of patients with increased liver transaminases
(AST or ALT) in the study groups with any signs of liver disease (group A, B and C) compared to
group D. However, in the CCFLD patient (group A) we did observe a significantly higher
proportion of patients with a GGT above the upper limit of normal of 50 U/l compared to the
patients without any signs of liver disease (group D). The latter indicative for a potential
relation between GGT elevations and the development or presence of CCFLD.
CCFLD study group
We evaluated the age of presentation of CCFLD (group A) in the study population. We
determined the date of diagnosis as the first day patients met the defined criteria of CCLFD.
The peak incidence of CF liver disease was around the age of 10 years. We found no patients
who developed CCFLD before the age of 5 or after the age of 15 years. The median age at
presentation of CCFLD in our population was 10 years. Almost all (95%) CCFLD patients from
group A were treated with UDCA.
Chapter 3
76
3
Table 1. Description of clinical symptoms and biochemistry data of total studied pediatric CF
population.
A B C D1
macronodularity
and
splenomegaly
macronodularity
and
no splenomegaly
splenomegaly
and
no macronodularity
no macronodularity
and
no splenomegaly
Total
Patients (N) 19 (7%) 10 (4%) 12 (4%) 236 (85%) 277
Median age at
evaluation date
(years)2
16* (8-17) 11 (6-15) 13* (8-17) 10 (2-17)
Median age at
diagnosis CF
(days)2
61 (4-4894) 84 (9-1009) 195 (16-3511) 126(0-3440)
Clinical symptoms N % P value2 Odds ratio (CI) N % N % N % N %
Male3 13 (68%) 0.092 7 (70%) 8 (67%) 114 (48%) 142 (51%)
UDCA use3 18 (95%) >0.001 72 (9-556) 10 (100%) 9 (75%) 47 (20%) 84 (30%)
DIOS3 4 (21%) 0.009 5 (1-16) 4* (40%) 1 (8%) 13 (6%) 22 (8%)
Meconium ileus3 5 (26%) 0.112 3 (30%) 3 (25%) 31 (13%) 42 (15%)
Pancreatic
insufficient (PERT)3 19 (100%) 0.621 10 (100%) 12 (100%) 233 (99%) 274 (99%)
Severe genotype3# 19 (100%) 0.062 8 (80%) 11 (92%) 199 (84%) 237 (86%)
DF508/DF5083 11 (58%) 0.622 6 (60%) 9 (75%) 150 (64%) 176 (64%)
Biochemistry
ASAT>50 U/l3 2 (14%) 0.285 3 (33%) 3 (30%) 33 (17%) 41 (18%)
ALAT>50U/l3 2 (14%) 0.543 0 (0%) 2 (20%) 24 (12%) 28 (12%)
GGT>50U/l3 6 (43%) <0.001 16 (5-52) 2 (22%) 2 (20%) 6 (3%) 16 (7%)
1 Reference population
2Mann–Whitney U (either group A, B or C individually compared to group D)
3Chi square test, group (either group A, B or C individually compared to group D)
*P<0.05
#Severe genotype defined as CFTR class1-3 mutations
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
77
3
Table 2. Comparisons of study groups for GGT sub analysis.
CCFLD
Study group
N=14, 33 data sets
No-CCFLD
Control group
N=36, 205 data sets
N % N %
Male 9 64 22 61
Severe genotype# 14 100 26 72
DF508/DF508 7 50 23 63.9
Meconium ileus 3 21 3 8.3
Pancreatic insufficient 14 100 36 100
DIOS 2 14 1 3
UDCA use 13 93 8 22
#Severe genotype defined as CFTR class1-3 mutations
No-CCFLD control group
For an additional analysis, we established a no- CCFLD control group out of the patient group
D, of which we assumed that they had and would not develop CCFLD (Figure 1). By definition,
they were all older than 15 years of age at the time of study evaluation. The no-CCFLD control
group consisted of 36 patients (Table 2). The control group contained a slightly higher
proportion of males. Meconium ileus and DIOS had also occurred in the control group.
Remarkably 22% of no-CCFLD control group patients had reported use of UDCA. Since in this
group no splenomegaly or macronodularity has been found at any time, the indication for
UDCA most likely was increased liver enzymes, then considered as an indication for UDCA in
our centers.
Chapter 3
78
3
Controls CCFLD0
10
20
30
40
50
A
GG
T (
U/L
)
0 100 200 3000
50
100
150Controls
CCFLD
B
ALT (U/L)
GG
T (
U/l
)
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
C
1- Specificity
Sen
sit
ivit
y
Figure 2. GGT in CCFLD compared to no-CCFLD CF patients. The mean gamma-glutamyl transpeptidase
(GGT) results of the period 2 years prior to the diagnosis cirrhotic cystic fibrosis liver disease (CCFLD) and
GGT results of no-CCFLD CF (A). Relation between alanine aminotransferase (ALT) results and
corresponding GGT results of CCFLD patient (closed black dots) 2 years prior to the diagnosis CCFLD
compared and all corresponding GTT results(open gray dots, obtained between age 2-15 years, from CF
patients without any signs of liver disease at the age of 15 years) (B). Receiver operating curve (ROC)
analysis for the relation between presence of CCFLD and GGT results within 2 years prior to the diagnosis
CCFLD or no-CCFLD controls shows a AOC of 0.9±0.1; P<0.001, at a cutoff value for GGT of 21 U/l (C)
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
79
3
Table 3. Overview of effects, of different cut off values for GGT, on diagnostic testing for risk of
development of CCFLD
Cutoff value Specificity Sensitivity Odds ratio (CI) PPV NPV NNT
GGT > 20U/l 82 93 57 (7-1206) 26 99 4
GGT > 25U/l 87 71 16 (4-68) 27 98 4
GGT > 30U/l 92 71 28 (7-119) 37 98 3
GGT > 35U/l 95 64 39 (9-175) 50 98 2
GGT > 40U/l 96 50 28 (7-127) 50 96 2
CCFLD patients vs. no-CCFLD control patients
We compared GGT results of the CCFLD patients in the 2 years preceding the diagnosis CCFLD
(N=28 GGT measurements) with those of the no-CCFLD control group (N=205 GGT
measurements). We found that the mean GGT in the CCFLD patient group was significantly
higher than in the controls (45 vs. 17 U/l, respectively, P<0.001) (Figure 2A). However the
mean GGT of the CCFLD group was still within the normal-high region. Elevation of GGT in
CCFLD patients appeared, in most cases, an isolated event, i.e. increase in GGT was almost
never accompanied by simultaneous elevation in ALT (Figure 2B). In the CCFLD group 82% of
the patients had a repeated GGT level over 30 U/l in consecutive repeated measurements
versus 18% in the no-CCFLD controls. Only 14% of the patients in the no-CCFLD control group
ever showed a GGT over 35 U/l, however, in the CFFLD group this was 79%. In 2 (9%) of no-
CCFLD controls the GGT was never below 30 U/l (N=5/5 measurements, range 30-141U/l)
despite the fact that there were no ultrasonographic or physical signs of CCFLD in these
patients. In the CCFLD group 58% of patients never had a GGT below 30 U/l.
To establish a cutoff value of the GGT above which the risk for developing CCFLD would be
significantly and relevantly increased, we applied a ROC analysis. We used all the GGT results
of the CCFLD study group and the no-CCFLD control group. If in the CCFLD group, more than
one GGT results per patients was available in the period 2 years before the diagnosis CCFLD,
we used the mean value of all available results. In this manner each CCFLD patient only
contributed one GGT value to the analysis. We compared these mean GGT results of CCFLD
patients to all the available GGT results of the control group. We found an AOC of 0.9±0.1
(P<0.001), and determined a cutoff value for GGT of 21 U/l (Figure 3C). To establish an
indication for potential clinical relevance of cutoff values, we performed cross tabulations
analysis (Table 3). This analysis shows the different consequences for Odds ratio, specificity,
Chapter 3
80
3
sensitivity, positive predictive value and negative predictive value for different, arbitrarily,
GGT cutoff values. In a potential interventional trail for prevention of CCFLD the numbers
need to treat based on the applied cut of values would be highly relevant. Based on the
different chosen cutoff values for GGT, we calculated the theoretical numbers needed to
treat, and found them to vary from 2 to 4 (Table 3).
DISCUSSION
Cirrhotic liver disease is a severe and chronic complication of cystic fibrosis. CF patients in
early stage of disease could benefit from preventive treatment strategies. However, to date,
we do not have the possibility to test and identify patients at risk for CCFLD. In this study we
found that, at group level, in pediatric CF patients in the period 2 years preceding their
diagnosis CCFLD the serum GGT levels are significantly higher than in CF patients who never
develop CCFLD. It is notable that although generally higher than in no-CCFLD controls, also in
the CCFLD group GGT results often remain within the normal and high-normal range.
However, for CF patients with a mean GGT>35 U/L, based on repeated measurements, the
Odds ratio to develop CCFLD was 39. These results indicate that in pediatric CF patients a
persistent, high-normal, serum GGT is strongly associated with the diagnosis of CCFLD within
2 years. However the prognostic value of a single GGT measurement remains limited. Our
results do indicate that groups of patients at increased risk for CCFLD can be identified on the
basis of repeated GGT measurements.
Recent studies in cirrhosis research indicate that progression of fibrotic liver diseases can be
stopped or even reversed by removal of the causative agent or treatment of the underlying
disease (22). In particular in the field of Hepatitis B and C disease, antiviral treatment is shown
to be able to reverse the severity of fibrosis (23, 24). In addition, new developments are
evolving concerning the use of anti-fibrotic therapies in forms of liver disease. Although
theoretically promising, to date there have not yet anti-fibrotic therapies become available
for humans (25). However the scope of these positive developments indicates the rising
opportunity and potential profit for preemptive treatment in CCFLD. These scientific advances
also indicate the need for reliable and relevant markers to identify patients at risk for CCFLD.
For future treatment strategies, aimed at preventing CCFLD, the number of patients that need
to be treated, based on the used risk factors, would be relevant. To date, CF patients often
start with UDCA treatment based on persistent elevation of liver enzyme above the upper
limit of normal. Liver enzymes elevations, in particular AST and ALT, are rather frequent
events in CF patient. Additionally AST and ALT elevation seem not to be strongly associated
with the development of the cirrhotic form of CF liver disease. Therefore many patients are
probably treated with UDCA even though they do not actually carry an increased risk for the
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
81
3
development of CCFLD. Our results indicate that a cutoff value for GGT of 35 U/l could be
clinically relevant to identify patient at risks for the diagnosis CCFLD. When this cutoff value is
used in patient selection for preemptive treatment strategies the numbers needed to treat
would be two with high sensitivity, positive and negative predictive value and a relative
acceptable specificity.
For most of our CCFLD patients, GGT results of repeated measurements were available, in the
2 years preceding the diagnosis of CCFLD. In our current analysis, we used the mean of
available GGT measurements as an indicator for the course of GGT in this period. However, in
the ROC analysis for the no-CCFLD group we used all available GGT results per patients from
the age 2-15 years including outliers. We did not correct the no-CCFLD group for any isolated
single GGT elevations. Still, only two no-CCFLD patients had a mean GGT, of all their GGT
results, over 30 U/l. Therefore we feel that this methodology, that involved all available GGT
results of the control group, strengthens our conclusions. Based on these findings we
hypothesize that in a prospective study design that includes only persistent GGT elevation in
the analysis; an even more distinct difference between CCFLD and no-CCFLD can be obtained.
The current study provides additional information on the presence of (isolated) elevated liver
transaminases in CF patients. In the no-CCFLD control group we found that respectively 18%
and 12% of patients had recorded episodes of AST and ALT elevation above the upper limited
of normal. This observation indicates that increased transaminases are rather frequent events
in pediatric CF patients, not related to future development of CCFLD. Based on this
observation we conclude that elevation of transaminases are not a sensitive predictive
biochemical markers for development of CCFLD. It is possible that elevation of transaminases
is associated with other forms of CF related liver disease like for example, non-alcoholic
steatohepatitis. Elevation of transaminases can also result from secondary hepatotoxic effects
of medication or clinical infectious episodes..
We are aware that our retrospective study design has potential inherent weaknesses. First a
retrospective analysis of the clinical follow up may not be as accurate and complete as
desired. Secondly the date of the CCFLD diagnosis cannot be pinpointed to a specific day.
However the date of diagnosis used in this study is based, on the rather random date, of the
annual clinical checkup. Thirdly, for the no-CCFLD control group, we used a cut-off age of 15
years. This cut-off is based on the reasonable assumption that CCFLD does not develop
anymore beyond this age. However, we do realize that it cannot be ruled out that new cases
of CCFLD could still develop after the age of 15 years. Based on the above-listed limitations of
the current study we realize that the reported data, ideally, need confirmation in a
prospective study design.
We defined CCFLD as ultrasonographic macronodularity of the liver in combination with
splenomegaly. However in our analysis, we also encounter patients with either
macronodularity or splenomegaly and not the combination of both. We cannot exclude that
Chapter 3
82
3
these observations indicate an earlier phase of CCFLD development. For ultrasonographic
macronodularity, without splenomegaly, the only likely explanation is an advanced stage of
fibrosis, however still in the absence of relevant portal hypertension. For isolated
splenomegaly the situation is different and less clear. For example, splenomegaly has been
described in cystic fibrosis patients, in the absence of any signs of liver disease (26). On the
other hand Mueller-Abt et al. described a case of one patient with splenomegaly as the only
pathologic ultrasonographic finding in a patient with liver histology that matched cirrhosis
(10).
In summary, we found evidence that in pediatric CF patients, a persistent, high-normal,
serum GGT is strongly associated with the diagnosis of CCFLD within 2 years. The prognostic
value of a single GGT measurement however, remains limited. Our results indicate that
groups of patients at increased risk for CCFLD can be identified on the basis of repeated GGT
measurements.
Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease
83
3
REFERENCE LIST
1. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis
Child. 1991 06;66(6):698-701.
2. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in
cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.
3. Corbett K, Kelleher S, Rowland M, Daly L, Drumm B, Canny G, et al. Cystic fibrosis–
associated liver disease: A population-based study. J Pediatr. 2004 9;145(3):327-32.
4. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.
Hepatology. 1999 11;30(5):1151-8.
5. Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the
diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros.
2011;10:S29-36.
6. Cystic fibrosis foundation patient registry 2011. 2012.
7. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre
experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008
05;7(3):252-7.
8. Bartlett JR, Friedman KJ, Ling SC, Pace RG, Bell SC, Bourke B, et al. Genetic modifiers of liver
disease in cystic fibrosis. JAMA: the journal of the American Medical Association.
2009;302(10):1076-83.
9. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic
fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.
1999;31(1):77-83.
10. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy
findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis.
2008;7(3):215-21.
11. Lindblad A, Hultcrantz R, Strandvik B. Bile-duct destruction and collagen deposition: A
prominent ultrastructural feature of the liver in cystic fibrosis. Hepatology. 1992 08;16(2):372-
81.
12. Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin
Gastroenterol. 2009;25(3):272-8.
13. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.
Cochrane Database Syst Rev. 2012;10.
14. Fallowfield JA. Therapeutic targets in liver fibrosis. American Journal of Physiology-
Gastrointestinal and Liver Physiology. 2011;300(5):G709-15.
Chapter 3
84
3
15. Lenaerts C, Lapierre C, Patriquin H, Bureau N, Lepage G, Harel F, et al. Surveillance for
cystic fibrosis-associated hepatobiliary disease: Early ultrasound changes and predisposing
factors. J Pediatr. 2003 09;143(3):343-50.
16. Colombo C, Apostolo MG, Ferrari M, Seia M, Genoni S, Giunta A, et al. Analysis of risk
factors for the development of liver disease associated with cystic fibrosis. J Pediatr. 1994
3;124(3):393-9.
17. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of
cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr
Gastroenterol Nutr. 1997;25(1):32-6.
18. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for liver
disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.
2003;23(6):1484-90.
19. Mayer-Hamblett N, Kloster M, Ramsey BW, Narkewicz MR, Saiman L, Goss CH. Incidence
and clinical significance of elevated liver function tests in cystic fibrosis clinical trials.
Contemporary Clinical Trials;34(2):232-8.
20. Loftus WK, Metreweli C. Ultrasound assessment of mild splenomegaly: Spleen/kidney
ratio. Pediatr Radiol. 1998;28(2):98-100.
21. Schwimmer JB, Dunn W, Norman GJ, Pardee PE, Middleton MS, Kerkar N, et al. SAFETY
study: Alanine aminotransferase cutoff values are set too high for reliable detection of
pediatric chronic liver disease. Gastroenterology. 2010 4;138(4):1357,1364.e2.
22. Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. J Hepatol. 2012
5;56(5):1171-80.
23. Brown A, Goodman Z. Hepatitis B-associated fibrosis and fibrosis/cirrhosis regression with
nucleoside and nucleotide analogs. Expert review of gastroenterology & hepatology.
2012;6(2):187-98.
24. D'Ambrosio R, Aghemo A, Rumi MG, Ronchi G, Donato MF, Paradis V, et al. A
morphometric and immunohistochemical study to assess the benefit of a sustained virological
response in hepatitis C virus patients with cirrhosis. Hepatology. 2012;56(2):532-43.
25. Friedman SL. Evolving challenges in hepatic fibrosis. Nature Reviews Gastroenterology and
Hepatology. 2010;7(8):425-36.
26. Penketh A, Wise A, Mearns M, Hodson M, Batten J. Cystic fibrosis in adolescents and
adults. Thorax. 1987;42(7):526-32.
CHAPTER 4
CHOLIC ACID INDUCES A CFTR
DEPENDENT BILIARY SECRETION
AND LIVER GROWTH RESPONSE IN
MICE Submitted
Frank A. J. A. Bodewes1
Marcel J. Bijvelds2
Willemien de Vries1
Julius F.W. Baller1
Annet S.H. Gouw3
Hugo R. de Jonge2
Henkjan J. Verkade1
1 Department of Pediatrics, University of Groningen, Beatrix Children’s Hospital - University
Medical Center, Groningen, The Netherlands
2 Department of Gastroenterology & Hepatology, Erasmus University Medical Center,
Rotterdam, The Netherlands
3 Department of Pathology, University Medical Center, Groningen, The Netherlands
Chapter 4
86
4 4
ABSTRACT
Introduction
Cystic fibrosis (CF) mouse models do not display cirrhotic CF liver disease. We hypothesized
that a cirrhotic CF liver disease equivalent may be provoked in CF mice by hydrophobic bile
salt exposure.
Methods
We studied the effect of acute (intravenous) and chronic (dietary) cholate administration on
bile formation, bile composition and histology in a CF mouse model.
Results
Basal bile production, i.e. without bile salt administration, was similar in Cftr-/-
mice and
controls. Intravenous taurocholate stimulated bile salt secretion and bile flow similarly in both
genotypes. Cholate administration for 3 weeks also increased bile salt secretion similarly in
Cftr-/-
mice and controls. However, cholate administration for 3 weeks increased bile flow in
Cftr-/-
mice to a lower extend, resulting in a significantly higher biliary bile salt concentration,
compared with controls. In controls, but not in Cftr-/-
mice, chronic cholate increased both the
biliary phospholipid-to-bile salt (molar) ratio and bile hydrophobicity. Cholate administration
for 3 weeks did not induce liver histology analogous to cirrhotic cystic fibrosis liver disease in
Cftr-/-
mice. However, chronic cholate administration induced a liver growth (+52%, p<0.001)
and parenchymal proliferation response (Ki67) in controls that was lacking in CF mice.
Conclusion
We conclude that chronic CA administration induces a Cftr dependent biliary secretion and
liver growth and proliferation response in mice. Chronic CA exposure did not induce cirrhotic
cystic fibrosis liver disease in CF mice. This response includes Cftr dependent cytoprotective
changes in the bile composition. These findings point to a, potentially protective, Cftr
dependent, hepatic response to prolonged hydrophobic bile salt exposure.
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
87
4 4 4
INTRODUCTION
Cystic fibrosis (CF) is caused by mutations in the CFTR gene (1,2). Cystic fibrosis liver disease
(CFLD) develops in 5-10% of cystic fibrosis patients (3). It is a serious complication of CF (4,5).
CFLD is characterized by cirrhosis and patients often present with splenomegaly,
hypersplenism and complications of portal hypertension, including variceal bleeding and
ascites (6,7). Although the synthesis and excretory liver function are usually spared, in some
CFLD patients liver transplantation can be indicated (8,9). Despite major advances in basic
knowledge about cystic fibrosis, the pathogenesis of CFLD is still unexplained.
In the liver, CFTR is expressed exclusively at the apical membrane of the cholangiocytes lining
the bile ducts (10,11). CFTR in cholangiocytes mediates water secretion, and controls the pH
of bile (12,13). Consequently, loss of CFTR function may increase the concentration of
potentially cytotoxic bile salts in bile and increase its viscosity, leading to occlusion of small
bile ducts and, ultimately, obstructive biliary cirrhosis (14-16).
CF mice may show minor gallbladder and liver abnormalities that do not lead to loss of liver
function (17-20). In general, mice produce bile that is relatively hydrophilic (21). Hydrophilic
bile salts are considered less cytotoxic than hydrophobic ones (22,23). The hydrophilic profile
of murine bile could prevent or mitigate liver disease in CF mice.
To test this hypothesis we sought to change the composition of murine bile to one containing
more hydrophobic bile salts. To this end, we either acutely or chronically challenged CF mice
with cholic acid (CA), a bile salt with strong detergent action or monitored effects on bile
production, composition, serum alanine transaminase (ALT), liver morphology, and histology
(23)
MATERIALS AND METHODS
Animals.We used C57Bl/6;129 Cftr-/-tm1CAM
mice and Cftr +/+tm1CAM
littermate controls. All mice
were bred and accommodated at the Animal Experimental Center of the Erasmus Medical
Center in Rotterdam, The Netherlands (17). Mice were housed in a light-controlled (lights on
6 AM to 6 PM) and temperature-controlled (21oC) facility, and had free access to tap water
and a semi-synthetic diet (SRM-A; Hope Farms BV Woerden, The Netherlands) from the time
of weaning. All experiments were performed with mice of 10-20 weeks of age. Group size
varied per experiment from 5-7 mice per genotype. Experimental protocols were approved by
the Ethical Committee for Animal Experiments of Erasmus MC. All surgery was performed
under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.
Experimental procedures. To test our hypothesis we used a mouse gall bladder cannulation
model (24,25). We first applied an acute bile salt infusion, as previously described (25). We
infused taurine-conjugated cholate (TCA) (43 mM dissolved in phosphate-buffered saline, pH
Chapter 4
88
4 4
7.4) via the jugular vein using an infusion pump. The bile salt dosage was increased every 30
minutes in a stepwise manner (dosage steps 0, 50, 300, 450 and 600 nmol.min-1
). During bile
salt infusion, bile was collected in 15 minutes fractions.
Different groups of Cftr-/-
and control mice were either fed a control diet consisting of
standard chow or the same diet enriched with cholate (CA 0.5% wt/wt) for 3 weeks.
Additional experimental groups of Cftr-/-
mice and controls were fed the diet for 3 months.
After gallbladder cannulation as described above, genuine bile production was determined by
bile collection for 30 minutes. The bile was collected for analysis. Mice were then sacrificed by
cardiac puncture, blood was drawn by cardiac puncture and livers samples were immersed in
neutral buffered formalin.
Analytical techniques. Biliary bile salt concentrations were determined by an enzymatic
fluorometric assay (26). Lipids were extracted from the bile (27). The phospholipid
concentrations were determined using a spectrophotometric assay (28). Biliary bile salt
composition was determined by capillary gas chromatography (29). The hydrophobicity of bile
salts in bile was calculated according to the Heuman index based on the fractional
contribution of the mayor bile salt species (23). ALT was determined in plasma samples.
For histological analysis of the liver samples, hematoxylin/eosin staining was applied on
paraffin embedded sections. To quantify mitotic active cells we used Ki-67 immunostaining
using a rabbit anti Ki-67p monoclonal antibody (Novocastra Laboratories Ltd., Newcastle upon
Tyne, UK) (30).
An experienced hepato-pathologist (A.S.H.G.) assessed the liver samples in a blinded fashion.
Histology of the liver parenchyma and portal tracts were evaluated separately for
inflammation, fibrosis and Ki-67 mitotic activity. The score was performed in a semi-
quantitative approach on a scale from 0 to 3 (0: absent, 1: sporadic, 2: regular, 3: frequent)
(31).
Statistical analysis. Analyses were performed using SPSS version 18.0 for Windows (SPSS Inc.,
Chicago, IL). All nominal results are reported as means ± SEM. The ordinal histology results
are reported as median and range. Differences between study groups were evaluated using
the Mann-Whitney U test. The level of significance was set at a P value of less than 0.05.
RESULTS
Bile production and bile salt secretion after interruption of the enterohepatic circulation. Bile
production, assessed over 30 minutes immediately after interruption of the enterohepatic
circulation, did not differ significantly between Cftr-/-
mice and controls (mean of 2 time points
in first 30 minutes: 6.3±0.7 vs. 5.8±0.5 μl.min-1
.100 g -1, respectively; Figure 1A). Bile salt
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
89
4 4 4
concentrations and bile salt secretion rates were also similar between Cftr-/-
mice and controls
(Figure 1B and C).
Bile production and bile salt secretion during acute IV taurocholic acid (TCA) administration.
Administration of TCA (iv) dose-dependently increased bile production (Figure 1A). The dose-
response relationship was similar for Cftr-/-
mice and controls, indicating that bile production
was not impaired in CF mice. A concomitant dose-dependent increase in bile salt
concentrations and bile salt secretion rates was observed upon TCA infusion, in both Cftr-/-
mice and controls (Figure 1B and C). A linear correlation was observed between the bile salt
secretion rate and bile flow (Figure 1D). The slopes derived after linear regression analysis,
were similar for Cftr-/-
mice and controls, indicating similar levels of bile salt dependent bile
flow. These observations strongly suggest that the bile flow in mice, both under basal
conditions and after acute IV TCA administration, is (predominantly) generated via CFTR
independent mechanisms.
0
5
10
15
20
Cftr+/+
ACftr
-/-
150300
450600 TCA infusion rate
(nmol.min -1)
30
Time (minutes)
60 90 120 1650
Bile f
low
( l.m
in-1
.100 g
ram
s B
W-1
)
0
50
100
150
Cftr+/+
BCftr
-/-
150300
450600 TCA infusion rate
(nmol.min -1)
30
Time (minutes)
60 90 120 1650
Bil
e s
alt
co
ncen
trati
on
(mM
)
0
500
1000
1500
2000
Cftr+/+
CCftr
-/-
150300
450600 TCA infusion rate
(nmol.min -1)
30
Time (minutes)
60 90 120 1650
Bil
e s
alt
secre
ton
rate
(nm
ol.m
in-1
.100 g
ram
s B
W-1
)
0 500 1000 1500 20000
5
10
15
Cftr+/+
DCftr
-/-
Bile salt secreton rate
(nmol.min-1
.100 grams BW-1
)
Bile f
low
( l.m
in-1
.100 g
ram
s B
W-1
)
Figure 1. Biliary parameters during intravenous taurocholic acid (TCA) administration. Biliary bile flow (A), bile salt
concentration (B), bile salt secretion rate (C) and relationship between bile salt secretion rate and bile flow (D) in Cftr
knockout mice (Cftr-/) and control littermates (Cftr+/+) during intravenous infusion with TCA in stepwise increasing
dosage of. The grey symbols in (D) represent baseline values before the start of TCA infusion. Data are presented as
means ± SEM of N=5-7 mice per group. There was no significant difference between Cftr-/- and Cftr+/+ mice, at any of
the individual time points, for bile flow, bile salt concentration and bile salt secretion rate.
Chapter 4
90
4 4
Bile production and bile salt composition during chronic cholic acid feeding. Feeding a CA
containing diet for 3 weeks increased bile flow, both in Cftr-/-
mice and controls. However, the
increase was larger in controls than in Cftr-/-
mice (Figure 2A). The CA diet increased the biliary
bile salt concentration in Cftr-/-
animals but, not in controls (Figure 2B). Chronic CA feeding
increased the bile salt secretion rate in both Cftr-/-
mice and controls (Figure 2C).
To assess cytotoxicity, we measured the biliary phospholipid concentration and calculated the
phospholipid -to-bile salt ratio. During regular chow feeding, biliary phospholipid
concentrations were similar in Cftr-/-
mice and controls (Figure 2D), as were the phospholipid -
to- bile salt ratios (Figure 2E). CA feeding appeared to increase phospholipid concentrations to
a similar magnitude in Cftr-/-
and controls although the apparent differences between means
did not reach statistical significance (Figure 2D). CA administration increased the phospholipid
-to- bile salt ratio in controls. Because CA feeding increased the biliary bile salt concentration
in Cftr-/-
mice (but not in controls; see above) the phospholipid -to- bile salt ratio, under these
conditions, was significantly lower in Cftr-/-
mice than in controls (Figure 2E). We found no
significant difference in serum ALT in Cftr-/-
mice after CA diet compared to normal diet
(Figure 2F). The CA diet, however, significantly increased ALT, compared to normal in controls
and not in Cftr-/-
mice.
CA feeding markedly increased the fractional contribution of CA to the total bile salt pool of
the bile. In control mice, it rose to 90%, whereas in Cftr-/-
mice it reached near 100%. In the
control animals, the remaining 10% was attributed to deoxycholic acid (DCA), which is formed
from CA by intestinal bacteria (Figure 2G). Based on this result, after CA diet, the Heuman
hydrophobicity index of the bile was higher in controls than in Cftr-/-
mice (0.061 vs. 0.003,
respectively, P<0.01, Figure 2H).
On regular chow, serum ALT levels were similar in Cftr-/-
mice and controls (Figure 2F). CA
feeding increased serum ALT had in controls, suggesting that, despite the concomitant
increase in the phospholipid-to-bile salt ratio, the bile produced was more cytotoxic (Figure
2F). In Cftr-/-
mice, the CA diet induced a similar mean increase in serum ALT levels, despite a
markedly lower hydrophobicity index. However, the variation between individual animals was
larger than for controls, suggesting that a subset of Cftr-/-
animals may be more prone to
develop lesions of the biliary tract.
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
91
4 4 4
Normal diet CA diet0
5
10
15
20
25
Cftr+/+
Cftr-/-
A**
*
Bile f
low
(l.m
in-1
.100 g
ram
s B
W-1
)
Normal diet CA diet0
50
100
150
Cftr+/+
Cftr-/-
B*
*
Bile s
alt
co
ncen
trati
on
(mM
)Normal diet CA diet
0
500
1000
1500
2000
Cftr+/+
Cftr-/-
C*
*
Bile s
alt
secre
ton
rate
(nm
ol.m
in-1
.100 g
ram
s B
W-1
)
Normal diet CA diet0
5
10
15
Cftr+/+
Cftr-/-
D
Ph
osp
ho
lip
id c
on
cen
trati
on
(mM
)
Normal diet CA diet0.00
0.05
0.10
0.15
0.20
0.25
Cftr+/+
Cftr-/-
E
*
Ph
osp
ho
lip
id
to b
ile s
alt
rati
o
Normal diet CA diet0
50
100
150
Cftr+/+
Cftr-/-
F
*
AL
T (
U/l)
Cftr+/+
Cftr-/-
Cftr+/+
Cftr-/-
0
25
50
75
100
CholateDCA
Normal diet CA diet
G
Others
%
-0.10
-0.05
0.00
0.05
0.10
Cftr+/+
Cftr-/-*
H
Normal diet CA diet
*
Hyd
rop
ho
bic
ity i
nd
ex
Figure 2. Biliary parameters and bile composition before and after chronic dietary cholic acid (CA) administration.
Biliary bile flow (A), bile salt concentration (B), bile salt secretion rate (C), phospholipid concentration (D),
phospholipid-to- bile salt ratio (E) Serum ALT levels (F), percent contribution of the bile salts cholate, deoxycholate and
others (chenodeoxycholate, deoxycholate, ursodeoxcholate, α-muri cholate and β-muricholate) in bile (G) and
Heuman index of total bile salts in bile representing the hydrophobicity of bile salts (H) in Cftr knockout mice (Cftr-/-)
and control littermates (Cftr+/+) after a regular or 0.5%-CA (wt/wt) chow diet for 3 weeks. Data are presented as
means ± SEM or percentage if appropriate of N=6-7 mice per group. *P-value<0.05.
Chapter 4
92
4 4
Cftr dependent effects of chronic CA exposure on liver weight. Similar to previous observations
in other CF mouse models, our Cftr-/-
mice tended to be smaller than littermate controls
(Figure 3A) (32). CA feeding did not markedly affect body weight. On regular chow, Cftr-/-
mice
and controls had similar absolute and relative liver weights (Figure 3C and 3E). Chronic CA
administration increased the absolute and relative liver weight in controls, but not in Cftr-/-
mice (+52%, p<0.001), indicating that chronic CA administration induces a hepatic, Cftr-
dependent adaptive mechanism to stimulate liver enlargement (Figure 3C and 3E). The CA-
induced liver growth response (+33%, p<0.01) was also observed in control mice with a
genetic background (FVB;129Sv) different from the one of the Cftr-/-
littermates (C57Bl6;129).
The liver growth response to CA-diet was lost in mice expressing mutant Cftr (homozygous
F508del Cftr mice; Cftrtm1EUR
), indicating that it is largely independent of subtle differences
between mouse strains and requires functional CFTR, Figure 3 B, 3D and 3F) (33).
Cftr dependent effects of chronic CA exposure on liver histology. On a regular chow diet,
parenchymal cell proliferation was enhanced in Cftr-/-
mice, relative to controls (Figure 4C),
but no signs of parenchymal inflammation were evident. Chronic CA administration increased
parenchymal mitotic activity in controls to a level similar to that found in Cftr-/-
mice on either
regular chow or after chronic CA exposure. Concomitantly, 3 week CA feeding significantly
increased the parenchymal inflammation score in controls (Figure 4D). Prolonged (3 month)
CA exposure exceeds the 3 weeks findings, showing increased parenchymal mitotic activity
and parenchymal inflammation, particularly observed in controls. CA feeding did not generate
pathological changes in the portal tract areas of the Cftr-/-
mice, or controls (Figure 4 A and B).
DISCUSSION
In mice, both acute and chronic hydrophilic bile salt administration markedly increased biliary
bile salt production and flow rates. Cftr-/-
mice showed a reduced capacity to adapt to chronic
bile salt administration. Although biliary bile salt production was similar the increase in bile
flow was less pronounced in Cftr-/-
mice compared controls. This reduced capacity, of Cftr-/-
mice, to increase bile flow resulted in significantly higher biliary bile salt concentrations. No
such defect was observed upon acute intravenous bile salt loading, suggesting that only long-
term adaptations to bile salt loading, such as hepatic cell proliferation (discussed below), is
perturbed in CF. These results imply that in mice, in contrast to humans, either Cftr does not
significantly contribute to canalicular or ductular fluid secretion or that, in the absence of Cftr
is substituted by other anion channels.
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
93
4 4 4
Normal diet CA diet0
10
20
30
40
Cftr+/+
Cftr-/-
A*
Bo
dy w
eig
ht
(gra
ms)
Normal diet CA diet0
10
20
30
Cftrwt/wt
Cftrdf508/df508
B
Bo
dy w
eig
ht
(gra
ms)
Normal diet CA diet0
1
2
3
Cftr+/+
Cftr-/-
C**
Liv
er
weig
ht
(gra
ms)
Normal diet CA diet0
1
2
Cftrwt/wt
Cftrdf508/df508
D**
Liv
er
weig
ht
(gra
ms)
Normal diet CA diet0.00
0.02
0.04
0.06
0.08
Cftr+/+
Cftr-/-
E**
Liv
er
to b
od
y w
eig
ht
rati
o
Normal diet CA diet0.00
0.02
0.04
0.06
0.08
Cftrwt/wt
Cftrdf508/df508
F**
Liv
er
to b
od
y w
eig
ht
rati
o
Figure 3. Body and liver weight before and after chronic dietary cholic acid (CA) or UDCA administration. Body weight
(A/B), absolute liver weight (C/D) and relative liver weight reported as a percentage of the body weight (E/F) in Cftr
knockout mice (Cftr-/-) or ΔF508 Cftr mice (Cftrtm1EUR) and their respective control littermates (Cftr+/+) after a
regular or 0.5%-CA (wt/wt) chow diet for 3 weeks. Data are presented as means ± SEM of N=5-7 mice per group. *P-
value<0.05.
Chapter 4
94
4 4
0
1
2
3 Cftr-/-
Normaldiet
CA diet3 weeks
CA diet3 months
Cftr+/+
AP
ort
al
Ki6
7
sco
re
0
1
2
3Cftr
-/-
Normaldiet
CA diet3 weeks
CA diet3 months
Cftr+/+
B
Po
rtal
infl
am
tio
n
sco
re
0
1
2
3
Normaldiet
CA diet3 weeks
CA diet3 months
*
*
C
Cftr-/-
Cftr+/+
Pare
nch
ym
al
Ki6
7
sco
re
0
1
2
3Cftr
-/-
Normaldiet
CA diet3 weeks
CA diet3 months
Cftr+/+
D *
*
*
Pare
nch
ym
al
infl
am
tio
n
sco
re
Figure 4. Liver histology before and after chronic cholic acid (CA) administration. Histological evaluation of liver
parenchymal inflammation (A) parenchymal Ki67 activity (B), portal inflammation (C) and portal Ki67 activity (D) in
Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+) after a regular, a 0.5%-CA (wt/wt) chow diet for 3 weeks or
0.5%-CA (wt/wt) chow diet for 3 months. Histology was scored in a blinded and semiquantitative approach on an
ordinal scale: 0: absent, 1: sporadic, 2: regular, 3 frequent. All individual results are presented; the horizontal black
line represents the median of N=4-6 mice per group
We had expected that in CF conditions prolonged CA feeding would the increase in biliary bile
salt concentration together with increased level of secondary hydrophobic bile salts. We
anticipated that this combination of hydrophilic bile salts would result in more cytotoxic bile.
However, we found that under these conditions, bile of Cftr-/-
mice contained little of the
secondary bile salt deoxycholate (DCA), whereas in controls its level in bile was markedly
enhanced by CA feeding. Since DCA is produced exclusively by intestinal bacteria, these
results suggest that alterations in the composition of this micro flora, which are typical of CF,
reduce DCA production (34-36). Alternatively, absorption of DCA may be reduced in CF. This
low availability of DCA may reduce the cytotoxicity of bile. Therefore, this hitherto unknown
mechanism may protect CF mice against bile salt-induced biliary disease.
A previous study reported the development of spontaneous liver disease in CF mice (18). The
mice used in these studies had a genetic background (C57BL/6J) different from the animals
used by us. This difference between strains indicates that genetic modifiers might affect the
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
95
4 4 4
hepatobiliary phenotype. Considering the potential effect of the intestinal micro flora on bile
composition, environmental factors may also play a role (36). Peptamen, a liquid diet
formulation used to prevent intestinal obstruction (used in ref. 14, but not by us), has been
shown to promote small intestinal bacterial overgrowth in CF mice (37). This may lead to
enhanced production of secondary bile salts, in turn leading to the production of a more
cytotoxic bile and, consequently, liver disease.
We found that CA feeding significantly increased liver mass in control mice. Similar findings
have been reported previously (38). This earlier study showed that CA feeding also enhanced
liver regeneration after partial hepatectomy. It was proposed that this response is mediated
through the Farnesoid X receptor (FXR) nuclear receptor that is directly activated by specific
bile salt, including CA and DCA. In accordance, we have found earlier that UDCA, which is a
limited FXR ligand, does not resemble the effect of CA on liver growth (39). In this study, oral
feeding of UDCA (0.5% wt/wt) to Cftr-WT mice for 3 weeks, did not result in a significant
increase in relative liver weight (0.048 ±0.003 vs. 0.044±0.001, respectively; unpublished
data).
Presently, we found no increase in liver mass in CA-fed Cftr-/-
and CftrΔF508/ΔF508
mice. This
indicates that FXR signaling may be perturbed in CF mice, a notion, which is reinforced by a
previous study showing that FXR-controlled genes are down-regulated in the ileum of CF
mice, possibly, because of reduced bile salt uptake in ileocytes (40).
In summary, we show that, in mice, CFTR plays a role in the hepatic adaptive response to high
bile salt exposure. The CF condition is characterized by a reduction in bile salt-induced hepatic
cell proliferation, increase in biliary bile salt levels and difference in secondary bile salt
composition. These changes suggest that the CF liver may be more prone to develop hepato-
biliary disease upon (chronic) exposure to hepatotoxic substances.
ACKNOWLEDGMENTS:
Rick Havinga for the technical assistance and the mice micro-surgery procedures
REFERENCE LIST
(1) Collins FS. Cystic fibrosis: molecular biology and therapeutic implications. Science 1992
05/08;256(5058):774-779.
(2) Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.
Identification of the cystic fibrosis gene: genetic analysis. Science 1989 09/08;245(4922):1073-
1080.
Chapter 4
96
4 4
(3) Scottjupp R, Lama M, Tanner MS. Prevalence of Liver-Disease in Cystic-Fibrosis. Arch Dis
Child 1991 06;66(6):698-701.
(4) Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in
Cystic Fibrosis Liver Disease. Am J Gastroenterol 2010;106(1):104-109.
(5) Corbett K, Kelleher S, Rowland M, Daly L, Drumm B, Canny G, et al. Cystic fibrosis–
associated liver disease: a population-based study. J Pediatr 2004 9;145(3):327-332.
(6) Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.
Hepatology 1999 11;30(5):1151-1158.
(7) Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the
diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros
2011;10:S29-S36.
(8) Cystic Fibrosis Foundation Patient Registry 2011. Annual Data Report Cystic Fibrosis
Foundation Bethesda, Maryland 2012.
(9) Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre
experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros 2008
05;7(3):252-257.
(10) Cohn JA, Strong TV, Picciotto MR, Nairn AC, Collins FS, Fitz JG. Localization of the cystic
fibrosis transmembrane conductance regulator in human bile duct epithelial cells.
Gastroenterology 1993 12;105(6):1857-1864.
(11) Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride
currents in rat bile duct epithelial cells. J Clin Invest 1993 01;91(1):319-328.
(12) Beuers U, Hohenester S, de Buy Wenniger LJM, Kremer AE, Jansen PLM, Elferink RPJO. The
biliary HCO3- umbrella: A unifying hypothesis on pathogenetic and therapeutic aspects of
fibrosing cholangiopathies. Hepatology 2010;52(4):1489-1496.
(13) Fitz JG. Regulation of cholangiocyte secretion. Semin Liver Dis 2002 08;22(3):241-249.
(14) Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr
Scand 1988;363:45-50; discussion 50-1.
(15) Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin
Gastroenterol 2009;25(3):272-278.
(16) Feranchak AP, Sokol RJ. Cholangiocyte biology and cystic fibrosis liver disease. Semin Liver
Dis 2001 11;21(4):471-488.
(17) Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production
of a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet 1993 05;4(1):35-41.
Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice
97
4 4 4
(18) Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol 2004 04;164(4):1481-1493.
(19) Freudenberg F, Leonard MR, Liu SA, Glickman JN, Carey MC. Pathophysiological
preconditions promoting mixed “black” pigment plus cholesterol gallstones in a ΔF508 mouse
model of cystic fibrosis. Am J Physiol Gastrointest Liver Physiol 2010;299(1):G205-G214.
(20) Fiorotto R, Scirpo R, Trauner M, Fabris L, Hoque R, Spirli C, et al. Loss of CFTR Affects
Biliary Epithelium Innate Immunity and Causes TLR4-NF-kB-Mediated Inflammatory Response
in Mice. Gastroenterology 2011;141(4):1498-1508.e5.
(21) van Nieuwerk CM, Groen AK, Ottenhoff R, van Wijland M, Van Den Bergh Weerman MA,
Tytgat GN, et al. The role of bile salt composition in liver pathology of mdr2 (-/-) mice:
differences between males and females. J Hepatol 1997 01;26(1):138-145.
(22) Benedetti A, Alvaro D, Bassotti C, Gigliozzi A, Ferretti G, La Rosa T, et al. Cytotoxicity of
bile salts against biliary epithelium: a study in isolated bile ductule fragments and isolated
perfused rat liver. Hepatology 1997;26(1):9-21.
(23) Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. Correlation
between biliary bile salt hydrophobicity index and the activities of enzymes regulating
cholesterol and bile acid synthesis in the rat. J Lipid Res 1989;30(8):1161-1171.
(24) Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat
absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am
J Physiol Gastrointest Liver Physiol 2005 04;288(4):G646-G653.
(25) Verkade HJ, Havinga R, Shields DJ, Wolters H, Bloks VW, Kuipers F, et al. The
phosphatidylethanolamine N-methyltransferase pathway is quantitatively not essential for
biliary phosphatidylcholine secretion. J Lipid Res 2007;48(9):2058-2064.
(26) Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin
Chim Acta 1976 07/01;70(1):79-86.
(27) Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Am J Physiol
Gastrointest Liver Physiol 1959;37(8):911-917.
(28) Böttcher C, Van Gent C. Changes in the composition of phospholipids and of phospholipid
fatty acids associated with atherosclerosis in the human aortic wall. J Atheroscler Res
1961;1(1):36-46.
(29) Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic
circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the
absence of ileal bile acid-binding protein (Ibabp). J Biol Chem 2003;278(43):41930-41937.
Chapter 4
98
4 4
(30) Mauad TH, Van Nieuwkerk CM, Dingemans KP, Smit JJ, Schinkel AH, Notenboom RG, et al.
Mice with homozygous disruption of the mdr2 P-glycoprotein gene. A novel animal model for
studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis. Am J Pathol
1994 11;145(5):1237-1245.
(31) Van Nieuwkerk CM, Elferink RP, Groen AK, Ottenhoff R, Tytgat GN, Dingemans KP, et al.
Effects of Ursodeoxycholate and cholate feeding on liver disease in FVB mice with a disrupted
mdr2 P-glycoprotein gene. Gastroenterology 1996 07;111(1):165-171.
(32) Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst
Fibros 2004;3:183-190.
(33) van Doorninck JH, French PJ, Verbeek E, Peters RH, Morreau H, Bijman J, et al. A mouse
model for the cystic fibrosis delta F508 mutation. EMBO J 1995 09/15;14(18):4403-4411.
(34) De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr
2006;42(1):46-52.
(35) De Lisle RC, Mueller R, Boyd M. Impaired Mucosal Barrier Function in the Small Intestine
of the Cystic Fibrosis Mouse. J Pediatr Gastroenterol Nutr 2011;53(4):371-379.
(36) Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade
HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res
2011;71(1):4-12.
(37) De Lisle RC. Altered transit and bacterial overgrowth in the cystic fibrosis mouse small
intestine. Am J Physiol Gastrointest Liver Physiol 2007;293(1):G104-G111.
(38) Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent
bile acid signaling is required for normal liver regeneration. Science 2006
04/14;312(5771):233-236.
(39) Bodewes FAJA, Wouthuyzen-Bakker M, Bijvelds MJC, Havinga R, de Jonge HR, Verkade HJ.
Ursodeoxycholate modulates bile flow and bile salt pool independently from the cystic fibrosis
transmembrane regulator (Cftr) in mice. Am J Physiol Gastrointest Liver Physiol
2012;302(9):G1035-G1042.
(40) Debray D, Rainteau D, Barbu V, Rouahi M, El Mourabit H, Lerondel S, et al. Defects in
gallbladder emptying and bile acid homeostasis in mice with cystic fibrosis transmembrane
conductance regulator deficiencies. Gastroenterology 2012;142(7):1581-1591. e6.
CHAPTER 5
BILE SALT METABOLISM IN A CF
MICE MODEL WITH
SPONTANEOUS LIVER DISEASE
Submitted
Frank A.J.A. Bodewes1
Mariëtte Y.M. van der Wulp 1
Satti Beharry2
Rick Havinga1
Renze Boverhof1
M. James Phillips2
Peter R. Durie2
Henkjan J. Verkade1
Affiliations
1Department of Pediatric Gastroenterology and Hepatology, University Medical Center
Groningen, The Netherlands,
2 Department of Gastroenterology, Hepatology and Nutrition, The Hospital for Sick Children,
Toronto, Canada
Chapter 5
100
5
ABSTRACT
Introduction:
Long-living congenic C57BL/6J cystic fibrosis transmembrane regulator (Cftr)-/-
mice are
reported to spontaneously develop liver pathology in an age-dependent fashion. Alterations
in bile production and composition have been postulated to be involved in the development
of CFLD. To address this hypothesis, we determined bile production and composition in this
CF mouse model and controls.
Methods:
C57BL/6J Cftr-/-
mice and control littermates (N=10per genotype), mean age 74±25 days, were
fed a liquid diet (Peptamen Nestle) ad libitum from the time of weaning. We determined bile
production and composition after collection via gall bladder cannulation and sampled stools
for bile salt secretion and bile salt composition.
Results:
Cftr-/-
and control mice did not differ in bile production rate, biliary bile salt and phospholipid
concentration and secretion rate, and phospoholipid to bile salt ratio. Bile salt secretion– bile
flow graphs indicated that Cftr-/-
mice had a significantly higher bile salt dependent bile flow
compared with controls (Cftr+/+
, +130%, p=0.01). The biliary bile salt profile had a similar bile
hydrophobicity index in Cftr-/-
and control mice. However, the proportion of secondary bile
salts was significantly higher in Cftr-/-
mice than in controls (17±3 vs. 29±10%, resp.; p< 0.001).
In particular, the secondary, hydrophilic bile salt ursocholate was ~7 fold higher in Cftr-/-
mice
than in controls. (25% vs. 3%of total bile salts resp.; p=0.004).
Conclusion:
The reported development of liver disease in the C57BL/6J Cftr-/-
mice is not related to
alterations in bile production or biliary lipid composition. Cftr-/-
mice, on a liquid diet, have
increased amounts of secondary bile salts, which may be due to specific changes in bile salt
metabolism in the intestinal lumen of this CF mouse model.
Bile salt metabolism in a CF mice model with spontaneous liver disease
101
5
INTRODUCTION
Cystic fibrosis can be accompanied by liver disease (CFLD), which in ~10 % of all patients can
results in cirrhosis (1). Liver histology of these patients shows a severe biliary type of fibrosis
and cirrhosis (2).
To this day the pathogenesis and developmental pathways of CFLD remains obscure. It has
long been suggested that, in parallel to the suggested development of the obstructive CF lung
disease, local biliary obstruction by thickened, sticky bile formed the pathological basis for the
liver disease (3). However, this hypothesis is not well supported by experimental evidence.
Most information concerning the pathology of CFLD in humans has been derived from post
mortem evaluations (4). Prospective histological studies in humans are scarce, mainly because
of the need for invasive diagnostic procedures in asymptomatic patients (5).
Different CF mice models, with a wide variation in genetic background, are available (6). Most
CF mice models do not show liver pathology. Durie et al., however, described, age related,
histological abnormalities in the liver of C578L/6J Cftr−/−
mice (7). Additionally Freudenberg et
al. reported that older ΔF508 CF mice (100-200 days) had more liver fibrosis than their
controls (8).
Bile salts play an important role in intestinal lipid absorption and in biliary lipid excretion. Bile
salts can act as detergents and have cytotoxic properties. Bile salt composition and in
particular biliary bile hydrophobicity is related to development of liver disease in general (9).
The primary bile salts are synthesized in the liver and secreted into the intestine via the bile
ducts. In the intestine they serve mainly as detergents to solubilise water insoluble
components such as cholesterol and fats. In the intestine primary bile salts are subject to
biotransformation by the intestinal flora. In this way a variety of different secondary bile salts
are formed. All bile acids are absorbed in the terminal ileum and return to the liver, thus
undergoing enterohepatic circulation (10).
We studies the potential role of bile salt metabolism and biliary bile salt cytotoxicity in the
pathogenesis of the liver pathology the C578L/6J Cftr−/−
mice model. To address this question
we measured bile production and determined biliary and fecal bile salt excretion and
composition.
Chapter 5
102
5
METHODS
Animals. All experimental protocols were conducted in the Hospital for Sick children, the
Research Institute, Toronto, Canada, after approval by the institutional Animal Care
Committee. Mice were bred to wild-type C57BL/ 6J mice, obtained from the Jackson
Laboratories (Bar Harbor, ME). Mice were genotyped at 14 days of age using polymerase
chain reaction analysis of tail clip DNA. To minimize bowel obstruction and optimize long-
term viability, 20- to 23-day-old congenic C57BL/ 6J Cftr-/-tm1Unc
mice and their Cftr+/+
littermates were weaned to a liquid diet (Peptamen, Nestlé Nutrition, Canada) using glass
liquid mouse feeders, prepared in sterile water according to the manufacturer’s instructions.
Fresh diet and feeders, sterilized by autoclave, were replaced daily. Mice were housed in a
non-sterile conventional housing unit in micro-isolators cages, with corncob bedding changed
daily, and provided with sterile water in addition to the liquid diet. The colony was
maintained at a pathogen-free status by serological screening at a commercial laboratory.
Mice were kept in a 12-hour light-dark cycle. The animals used for the experiments were
approximately 1.5-4 months old
Experimental procedures. The total feces of 3 consecutive days was collected, dried and
homogenized. Bile was collected after surgical ligation of the common bile duct and gall
bladder cannulation using silicone tubing (size 0.020" x 0.037", Degania Silicone Ltd.)
under anesthesia. The anesthetic mixture consisted of: 0.75 ml ketamine (100 mg/ ml),
0.25 ml xylazine (20 mg/ ml) and 4 ml saline. The mixture (0.1 ml per 10 grams of body
weight) was intraperitoneal administered. Body temperature was maintained by placement of
the animal in a temperature and humidity controlled incubator. Bile secretions were collected
in 30 min. fractions, after a 10 equilibration period. Bile flow rate was assessed
gravimetrically, assuming that 1 g of secretion corresponds to 1 ml.
Analytical techniques. Biliary BS concentrations in bile and feces were determined by an
enzymatic fluorimetric assay (11). Lipids were extracted from the bile (12).The phospholipids
and cholesterol concentrations were determined using a spectrophotometric assay. Biliary BS
composition in bile and feces was determined by capillary gas chromatography (13). The
hydrophobicity of BS in bile was calculated according to the Heuman index based on the
fractional contribution of the mayor BS species (14).
Liver histology. Hemotoxylin and eosin (H&E) sections of the liver were assessed blindly by an
expert hepato-pathologist (JP) at The Hospital for Sick Children (Toronto, Canada) for the
degree of biliary duct obstruction and inflammation. We applied a previously published
numerical scoring system for each parameter, using an arbitrary 0 to 4 scale, in which zero
represents normal histology and four represents the most severe pathology for each
parameter (7). A minimum of five portal tracts was assessed for each mouse.
Bile salt metabolism in a CF mice model with spontaneous liver disease
103
5
RNA isolation from whole livers. Quantitative real time polymerase chain reactions (qRT-PCR,
combined for in vivo and in vitro experiments PCR was performed on a 7900HT Fast Real-Time
PCR system (Applied Biosystems).
Statistical analysis. Analyses were performed using SPSS version 18.0 for Windows (SPSS Inc.,
Chicago, IL). All nominal results are reported as means ± SEM. Differences between study
groups were evaluated using the Mann-Whitney U test. Bile flow was analyzed by correlation
and regression analysis. The categorized histology results were analyzed using cross tabs and
Fischer’s exact test. The level of significance was set at a P value of less than 0.05.
RESULTS
Body and liver weights. Figure 1 shows body weight and liver weight of CF and control mice.
C57BL/ 6J Cftr-/-
mice bodyweight was significantly lower compared to control littermates. The
liver weight of the Cftr-/-
was significantly less compared to controls, but the difference was
less pronounced than for the body weight. Accordingly, the relative liver weight compared to
the total bodyweight was significantly higher in the Cftr-/-
mice.
0
10
20
30
Cftr+/+
Cftr-/-
A*
Body w
eig
ht
(gra
ms)
0.0
0.5
1.0
1.5
Cftr+/+
Cftr-/-
B*
Liv
er
weig
ht
(gra
ms)
0.0
2.5
5.0
7.5
Cftr+/+
Cftr-/-
C*
Liv
er
to b
ody w
eig
ht ra
tio
Figure 1. Body and liver weight of Cftr+/+ and Cftr-/- mice. Body weight (A), absolute liver weight (B) and relative liver
weight reported as a percentage of the BW (C) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are
presented as means ± SEM of N=9-9 mice per group. *P-value<0.05.
Bile production and bile salt secretion. To determine if Cftr plays a role in the magnitude of
the bile production in C57BL/ 6J CF mice in vivo we measured bile production after
interruption of the enterohepatic circulation via gall bladder cannulation (Figure 2).
Interestingly, we found that the total bile production, was ~20-30% higher in Cftr-/-
mice
compared to control animals (P< 0.05 in the last time period, Figure 2A)
Chapter 5
104
5
Bile production is largely determined by the bile salt concentration and bile salt secretion
rate. The total bile biliary salt concentration was not different between Cftr-/-
versus control
mice (Figure 2B). The total biliary bile salt secretion rate was not significantly different
between Cftr-/-
and control mice (Figure 2C), but a consistent tendency towards higher values
was observed in the former. To evaluate the magnitude by which bile salts determine bile
production in CF conditions we related the bile flow to the bile salt secretion rate. (Figure 2D)
In this figure the BA-dependent flow (slope) and BA independent flow (intercept y-axis) can
be distinguished. Regression analysis showed a similar BA-independent bile flow between
both groups (P= 0.193). In contrast, the BA-dependent flow on the contrary, was higher in
Cftr-/-
vs. Cftr+/+
mice (P= 0.014).
30 min 60 min 90 min0
2
4
6
8
10
Cftr+/+
Cftr-/-
*
A
Bile
flo
w
(l.m
in-1
.100 g
ram
s B
W-1
)
30 min 60 min 90 min0
20
40
60
B
Cftr+/+
Cftr-/-
Bile
salt
concentr
atio
n
(mM
)
30 min 60 min 90 min0
100
200
300
Cftr+/+
Cftr-/-
C
Bile
salt s
ecr
etion r
ate
(nm
ol.m
in-1
.100 g
ram
BW
-1)
0 100 200 300 4000
2
4
6
8
10
Cftr+/+
Cftr-/-
D
Bile salt secreton rate
(nmol.min-1
.100 grams BW-1
)
Bile
flo
w
(l.m
in-1
.100 g
ram
s B
W-1
)
Figure 2. Biliary parameters of Cftr+/+ and Cftr-/- mice. Biliary bile flow (A), bile salt flow (B) bile salt secretion rate (C)
and bile secretion vs. bile flow (D) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). In panel D there is a
significant correlation between BASR and bile flow in Cftr+/+ (Spearman’s Rho= 0.394, P=0.042), as well as in Cftr-/-
mice (Spearman’s Rho=0.762, P=0.000).Data are presented as means ± SEM of N=9-9 mice per group. *P-value<0.05.
Bile salt metabolism in a CF mice model with spontaneous liver disease
105
5
Biliary bile salt composition. Bile salt composition and in particular biliary bile hydrophobicity
is related to development of liver disease in general (9). To address the hypothesis that bile
salt hydrophobicity plays a role in CFLD we evaluated biliary and fecal bile salt composition.
Significant differences in bile salt composition were observed between the genotypes (Figure
3A). In Cftr-/-
compared to controls there were significant lowered biliary proportion of CA
(40.3±8.5 vs. 53.0±7.5% respectively; p<0.05), DCA (1.6±2.0 vs. 5.3±2.7% respectively; p<0.05)
and ω-MCA (0.9±0.5 vs. 6.1± 2.6% respectively; p<0.05) content in bile. Primary BA (produced
in the liver: CA, CDCA, α- and β-MCA) content was decreased in Cftr-/-
bile compared to Cftr+/+
(70% vs. 84% respectively; p<0.05) with a replacement by secondary BA (HDCA, UDCA, DCA,
ω-MCA). (Figure 3C) Gas chromatographic analysis of bile showed an unexpectedly, ~7-fold
higher, UCA content in Cftr-/-
mice compared to controls (24.6 vs. 3.3% respectively; p<0.01).
Mass spectrometry confirmed the identification and abundance of UCA in CF mice. As a result
of increased proportion of the hydrophilic UCA the biliary bile salt hydrophobicity index of
Cftr-/-
mice compared to Cftr+/+
mice was significantly lower vs. (-0.2.5±0.09 vs. -0.13±0.05
respectively; P<0.05; Figure 3C).
-MCA
DCA
CA
CDCA
HDCA
UCA
-MCA
-MCA
0
20
40
60
80
100
Cftr+/+
Cftr-/-
*
*
*
A
% o
f to
tal b
ile a
cid
s
Cftr+/+ Cftr-/-0
25
50
75
100
Primary bile salts
Secondary bile salts
B
%
-0.4
-0.2
-0.0
0.2
0.4
Cftr+/+
Cftr-/-
C
*H
ydro
phobic
ity in
dex
Figure 3. Biliary bile composition in bile of Cftr+/+ and Cftr-/- mice. BA profiles (A), data are presented as percentage of
different BA species present in bile. Abbreviations: CA = cholic acid, DCA = deoxycholic acid, CDCA = chenodeoxycholic
acid, HDCA = hyodeoxycholic acid, UDCA= ursodeoxycholic acid, UCA= ursocholic acid, MCA= muricholic acid. Relative
distribution between primary BA content and secondary BA (B). Primary BA content is decreased in Cftr-/- mice,
whereas secondary BA content is increased (P= 0.004). Heuman index (C) of total BS in bile representing the
hydrophobicity of bile salts. Data are presented as means ± SEM or percentage if appropriate of N=6-7 mice per
group. *P-value<0.05.
Fecal bile salt composition. Increased total fecal bile salt loss has been frequently observed in
CF conditions, including by ourselves. In the C57BL/ 6J CF mice model we found that fecal bile
salt loss was ~80% higher compared to control animals. (Figure 4A) Corresponding with the
biliary bile salt composition results, Cftr-/-
mice had a profoundly higher UCA content (27%) in
feces than the Cftr+/+
mice (1%, mass spectrometry analysis; Figure 4B) Cftr-/-
mice had a
Chapter 5
106
5
significantly lower DCA (11.4 ± 10.3 vs 42.0 ± 10.7% respectively; P<0.01), HDCA (0.3 ± 0.2 vs.
2.2 ± 0.9% respectively; P<0.01) and ω-MCA (1.2 ± 0.9 vs. 18.1 ± 8.3 respectively; P<0.01)
content in their feces. Cftr-/-
mice had significantly more CA in their feces, compared to
controls (26.5 ± 9.1% vs. 8.4 ± 7.4% respectively; P<0.01).
To address whether the observes differences were genotype or (microbial) environment
related, we performed fecal bile salt analysis in mice of the same CF and control genotypes
(N=5/5) from a different colony at another institution (courtesy of Dr R.C. De Lisle University
of Kansas, Kansas City, Missouri). The pattern and of the bile salt composition, in particular
the presence of UCA, was virtually identical, supporting a genotype related effect. (Data not
shown)
0
10
20
30
40
A
Cftr+/+
Cftr-/-
*
Fecal b
ile s
alt
excre
tion
(um
ol/1
00gB
W/d
ay)
-MCA
DCA
CA
CDCA
HDCA
UCA
-MCA
-MCA
0
20
40
60
80
100
Cftr+/+
Cftr-/-
**
*
B%
of to
tal b
ile a
cid
s
Figure 4. Fecal bile salt excretion and fecal bile salt composition of Cftr+/+ and Cftr-/- mice. Total bile salt excretion (A),
presented as means ± SEM. BA profiles (B), data are presented as percentage of different BA species present in bile.
Abbreviations: CA = cholic acid, DCA = deoxycholic acid, CDCA = chenodeoxycholic acid, HDCA = hyodeoxycholic acid,
UDCA= ursodeoxycholic acid, UCA= ursocholic acid, MCA= muricholic acid. N=9-9 mice per group. *P-value<0.05.
Bile salt metabolism in a CF mice model with spontaneous liver disease
107
5
30 min 60 min 90 min0
2
4
6
8
10
Cftr+/+
Cftr-/-
AP
ho
sp
ho
lip
s (
mM
)
0 100 200 300 4000
20
40
60Cftr+/+
Cftr-/-
B
Bile salt secretion rate
(nmol.min-1
.100 gram BW-1
)
PL s
ecre
tion r
ate
30 min 60 min 90 min0.00
0.05
0.10
0.15
0.20
Cftr+/+
Cftr-/-
C
Ph
os
po
ho
lip
id t
o b
ile
sa
lt r
ati
o
Figure 5. Phospholipid secretion in Cftr+/+ and Cftr-/- mice. PL secretion rate (a) and the PL secretion vs. BS secretion
rate (B) and PL to BS ratio in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are presented as means
± SEM of N=9-9 mice per group.
Biliary lipids composition. Apart from bile salt hydrophobicity the biliary cytotoxicity and
detergent activity is determined by the biliary lipid composition (15). We found that the
phospholipid secretion was not significantly different between Cftr-/-
mice and controls
(Figure 5A). In both Cftr-/-
and controls the phospholipid secretion rate was linearly related to
the bile salt secretion rate, despite the significant differences in biliary bile salt
hydrophobicity. (Figure 5B).We found no difference in the biliary phospholipid to bile salt
ratio between Cftr-/-
and control mice (Figure 5C). Additionally we found no difference in
biliary cholesterol concentration and cholesterol secretion rate between Cftr-/-
and Cftr+/+
mice (data not shown).
Bile salt synthesis genes. Different bile salts have distinct influence in the induction of bile salt
synthesis. In particular DCA and CDCA are regarded as strong ligands for Fxr. We found a
significantly difference in CA and DCA proportions (Figure 6). This could imply difference in
bile synthesis activation between genotypes. To address the issue of potential difference we
determined the expression of different bile salt synthesis genes. (Figure 6) We found no
difference in expression of Fxr between Cftr-/-
and control mice, but the expression of the
orphan receptor Shp was significantly lower in the former. We did not observe significant
differences in the hepatic expression of Cyp 7a, encoding the rate limiting protein in the bile
salt synthesis process.
Chapter 5
108
5
*
FxrShp
Cyp
7a1
Cyp
8b1
BSEP
0
1
2
3
Cftr+/+
Cftr-/-
Rela
tive m
RN
A le
vel
norm
aliz
ed to b
eta
-actin
Figure 6. Gene expression in isolated livers in Cftr+/+ and Cftr-/- mice. mRNA levels were normalized to a housekeeping
gene (β-actin) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are presented as means ± SEM of
N=9-9 mice per group. *P-value<0.05.
Histology. Durie et al described in their initial report concerning liver pathology in Unc mice
that all Cftr-/-
animals showed focal and progressive hepatobiliary disease (7). The mean age
of the mice in the current experiment was 75±25 days in both the Cftr-/-
and control group.
Although, in general the current histo-pathological findings are consistent with the previous
report they are less pronounced and the global duct proliferation score, in this cohort,
revealed no significant difference between Cftr-/-
and control mice. (Figure 7)
0
1
2
3
4Cftr-/-
Cftr+/+
Port
al i
nfla
matio
n s
core
0
1
2
3
4Cftr-/-
Cftr+/+
Ducta
l change s
core
0
1
2
3
4Cftr-/-
Cftr+/+
NS (P=0.09)
Bridgin
g fib
rosis
score
Figure 7. Histological bile duct proliferation score. Bile duct proliferation score base on the observation of 10 portal
tracts per mouse graded: normal: 1 point, mild proliferation: 2 points, severe proliferation: 3 points. The bile duct
proliferation represents the mean of al scored points per mouse. Data are presented as means ± SEM of N=10-10 mice
per group.
DISCUSSION
We evaluated bile salt metabolism and bile salt composition and in the C57BL/ 6J CF mice
model that has been reported to spontaneously develop CFLD like liver disease (7). We
Bile salt metabolism in a CF mice model with spontaneous liver disease
109
5
hypothesized that this Cftr-/-
mice develop liver disease in the context of a cytotoxic biliary bile
salt profile. However in contrast of what could be expected we found that in Cftr-/-
mice on a
liquid diet, the biliary bile salt profile is substantially more hydrophilic and the bile flow is
increased compared to control mice. This increase in bile production is explained by the
increased presence of the secondary hydrophilic and highly choleretic bile salt ursocholate in
this mice model. Since UCA metabolism is converted form cholate by intestinal microbiota our
results emphasize the critical role of intestinal microflora in modifying the enterohepatic
circulation of bile salt in CF conditions.
Based on hydrophobicity and Pl content we did not find an increased cytotoxic profile of the
biliary bile of Cftr-/-
mice that could explain the reported spontaneous development of biliary
liver disease in this model. Others have suggested different potential mechanism for the
development of liver disease in the CF mouse models. Blanco et al. reported that chemical
induction of colitis, by dextran, in Cftr-/-
Unc-mice results in a increase in bile duct injury (16).
In an additional report from the same group they describe that decreased peroxisome
proliferator activated receptor alpha is associated with bile duct Injury in Cftr-/-
mice report
(17). These findings point in the direction of a Cftr related difference in immune response to
development of liver disease in CF mice. Beharry et al. reported a highly significant protective
effect of long-term docosahexaenoic acid therapy on the progression of CF-like hepatobiliary
disease in C57BL/ 6J CF Cftr-/-
mice (18). They suggested that the inhibition of inflammation by
DHA may account for the anti-inflammatory effects in the liver. The latter report again
endorsing a potentially immunologic/inflammatory genesis of CF related liver disease.
UCA, the 7α-hydroxy epimer of cholic acid, is a natural bile acid characterized by sixfold and
threefold greater hydrophilicity than chenodeoxycholic and ursodeoxycholic acid, respectively
(19). The contribution of UCA to the bile salt pool is usually rather low, due to rapid
biotransformation of UCA during enterohepatic circulation into DCA by intestinal micoflora. In
our Cftr-/-
mice, however, DCA contribution was significantly lower than in control mice,
suggesting a decreased biotransformation of UCA into DCA in the CF intestine. Due to it high
choleretic potency the enrichment of UCA in our model leads to a significant increased bile
salt dependent flow and low hydrophobicity index in Cftr-/-
mice.
It cannot be excluded that the exclusively liquid feeding could have influenced biliary bile salt
composition in our model. The C57BL/ 6J CF Cftr-/-
mice are dependent on liquid feeding after
weaning to prevent lethal intestinal obstruction (20). Therefore Peptamen liquid feeding is
used. Peptamen is a commercial infant formula use for instance for patients with short bowel
syndrome, intractable malabsorption, patients with proven inflammatory bowel disease and
bowel fistulae. Peptamen is a semi-elementary formula with a high proportion of protein and
medium chain triglycerides. Different from regular chow, Peptamen does not contain dietary
fibre or grain products. Since Cftr-/-
mice and controls were both fed the liquid Peptamen, diet
effects cannot explain the differences in bile salt composition between the genotypes.
Chapter 5
110
5
In our model we found a distinct increased fecal bile salt excretion. Increased fecal bile salt
loss has reported before in CF patients and animals models (21, 22). It has been suggested
that the underlying mechanism involves increased fecal fat excretion in CF conditions (23),
but previously we reported that increased fecal bile salt loss was also observed in CF mice
without fat malabsorption. Increased hepatic bile salt synthesis compensates for the higher
bile salt loss in Cftr-/-
mice (24). However, in the current study, the proportion of primary bile
salt in the biliary bile was actually lower in Cftr-/-
mice, acutely indicating a decreased
synthesis of bile salt in these CF mice.
Several publications in recent years have revealed the complex regulatory interaction of bile
salt metabolism between intestine and liver (25, 26). Bile salt synthesis in rodents is regulated
in the intestine via the Fxr-Fgf15 axes as well as more directly in the hepatocyte, via the Fxr-
Shp axis. Both routes eventually target the expression of Cyp7a, the rate limiting enzyme in
the bile salt synthesis. UCA is reported to have no or limited influence on bile salt synthesis, in
contrast to the hydrophobic DCA, which is one of the most potent FXR ligands to suppress bile
salt synthesis (27). The reduction of intestinal transformation of UCA to DCA may have lead to
a reduced bile salt synthesis as represented by the reduction in the proportion of primary
biliary bile salt. However this finding did not correspond to the Cyp7a expression levels that
were not different in Cftr-/-
mice compared to controls. On the other hand the reported
reduction in Shp expression would imply a partial down regulation of bile salt synthesis could
be induced via the alternative intestinal Fxr-Fgf15 pathway.
We found that the reported biliary disease in the Unc Cftr-/-
mice is not related to increased
biliary cytotoxicity. However Cftr-/-
mice shows distinctive changes in the metabolism and
enterohepatic circulation of the bile salts. Recent discoveries in the intestinal regulation of
bile salt synthesis shed a new light on the importance and role of the intestinal bile salt
composition. The reported difference in bile salt metabolism could be origin of the intestinal
bile salt loss and in CF condition. However additional research is needed to further determine
the role of the enterohepatic circulation in CF.
REFERENCE LIST
1. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis
Child. 1991 06;66(6):698-701.
2. Hultcrantz R, Mengarelli S, Strandvik B. Morphological findings in the liver of children with
cystic fibrosis: A light and electron microscopical study. Hepatology. 1986 09;6(5):881-9.
3. Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr Scand.
1988;363:45,50; discussion 50-1.
Bile salt metabolism in a CF mice model with spontaneous liver disease
111
5
4. Shier KJ, Horn Jr RC. The pathology of liver cirrhosis in patients with cystic fibrosis of the
pancreas. Can Med Assoc J. 1963;89(13):645.
5. Lewindon PJ, Shepherd RW, Walsh MJ, Greer RM, Williamson R, Pereira TN, et al.
Importance of hepatic fibrosis in cystic fibrosis and the predictive value of liver biopsy.
Hepatology. 2011;53(1):193-201.
6. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse
models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.
2011;10:S152-71.
7. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol. 2004 04;164(4):1481-93.
8. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.
Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.
Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.
9. Attili AF, Angelico M, Cantafora A, Alvaro D, Capocaccia L. Bile acid-induced liver toxicity:
Relation to the hydrophobic-hydrophilic balance of bile acids. Med Hypotheses. 1986
1;19(1):57-69.
10. Hofmann AF. Enterohepatic circulation of bile acids. In: Comprehensive Physiology. John
Wiley & Sons, Inc.; 2010.
11. Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin
Chim Acta. 1976 07/01;70(1):79-86.
12. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Am J Physiol
Gastrointest Liver Physiol. 1959;37(8):911-7.
13. Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic
circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the
absence of ileal bile acid-binding protein (ibabp). J Biol Chem. 2003;278(43):41930-7.
14. Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. correlation
between biliary bile salt hydrophobicity index and the activities of enzymes regulating
cholesterol and bile acid synthesis in the rat. J Lipid Res. 1989;30(8):1161-71.
15. Loria P, Carulli N, Medici G, Menozzi D, Salvioli G, Bertolotti M, et al. Effect of ursocholic
acid on bile lipid secretion and composition. Gastroenterology. 1986 Apr;90(4):865-74.
16. Blanco PG, Zaman MM, Junaidi O, Sheth S, Yantiss RK, Nasser IA, et al. Induction of colitis
in cftr–/–mice results in bile duct injury. American Journal of Physiology-Gastrointestinal and
Liver Physiology. 2004;287(2):G491.
Chapter 5
112
5
17. Pall H, Zaman MM, Andersson C, Freedman SD. Decreased peroxisome proliferator
activated receptor [alpha] is associated with bile duct injury in cystic fibrosis transmembrane
conductance regulator-/-mice. J Pediatr Gastroenterol Nutr. 2006;42(3):275.
18. Beharry S, Ackerley C, Corey M, Kent G, Heng YM, Christensen H, et al. Long-term
docosahexaenoic acid therapy in a congenic murine model of cystic fibrosis. American Journal
of Physiology-Gastrointestinal and Liver Physiology. 2007;292(3):G839-48.
19. Roda A, Hofmann AF, Mysels KJ. The influence of bile salt structure on self-association in
aqueous solutions. Journal of Biological Chemistry. 1983 May 25;258(10):6362-70.
20. Snouwaert JN. An animal model for cystic fibrosis made by gene targeting. Science.
1992;257(5073):1083.
21. O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid
malabsorption in cystic fibrosis. Gut. 1993;34(8):1137.
22. Bijvelds MJC, Jorna H, Verkade HJ, Bot AGM, Hofmann F, Agellon LB, et al. Activation of
CFTR by ASBT-mediated bile salt absorption. American Journal of Physiology-Gastrointestinal
and Liver Physiology. 2005;289(5):G870.
23. Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in
cystic fibrosis. Gut. 1976;17(4):295.
24. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat
absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am
J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.
25. Inagaki T, Moschetta A, Lee YK, Peng L, Zhao G, Downes M, et al. Regulation of
antibacterial defense in the small intestine by the nuclear bile acid receptor. Proc Natl Acad Sci
U S A. 2006;103(10):3920.
26. Lundåsen T, Gälman C, Angelin B, Rudling M. Circulating intestinal fibroblast growth factor
19 has a pronounced diurnal variation and modulates hepatic bile acid synthesis in man. J
Intern Med. 2006;260(6):530-6.
27. Heuman DM, Hylemon PB, Vlahcevic ZR. Regulation of bile acid synthesis. III. correlation
between biliary bile salt hydrophobicity index and the activities of enzymes regulating
cholesterol and bile acid synthesis in the rat. Journal of Lipid Research. 1989 August
01;30(8):1161-71.
CHAPTER 6
URSODEOXYCHOLATE MODULATES
BILE FLOW AND BILE SALT POOL
INDEPENDENTLY FROM CFTR IN
MICE Based on: American Journal of Physiology-Gastrointestinal and Liver Physiology302.9 (2012):
G1035-G1042.
Frank A.J.A. Bodewes1
Marjan Wouthuyzen-Bakker1
Marcel J. Bijvelds2
Rick Havinga1
Hugo R. de Jonge2
Henkjan J. Verkade1
1 Department of Pediatrics, University of Groningen, Beatrix Children’s Hospital - University
Medical Center, Groningen, The Netherlands
2 Department of Gastroenterology & Hepatology, Erasmus University Medical Center,
Rotterdam, The Netherlands
Chapter 6
114
6
ABSTRACT
Introduction
Cystic fibrosis liver disease (CFLD) is treated with ursodeoxycholate (UDCA). Our aim was to
evaluate, in Cftr-/- mice and wild type controls, if the supposed therapeutic action of UDCA is
mediated via choleretic activity or effects on bile salt metabolism.
Methods
Cftr-/- mice and controls, under general anesthesia, were IV infused with TUDCA in increasing
dosage or were fed either standard or UDCA enriched chow (0.5%wt/wt) for 3 weeks. Bile
flow and bile composition were characterized. In chow fed mice, we analyzed bile salt
synthesis and pool size of cholate (CA).
Results
In both Cftr-/- and controls IV TUDCA stimulated bile flow by ~250% and dietary UDCA by
~500%, compared with untreated animals (p<0.05). In non-UDCA treated Cftr-/- mice, the
proportion of CA in bile was higher compared to controls (61±4% vs. 46±4%, resp.; p<0.05),
accompanied by an increased CA synthesis (16±1 vs. 10±2 μmol/hour/100gramBW resp.;
p<0.05) and CA pool size (28±3 vs. 19±1 μmol/100gramBW, resp.; p<0.05). In both Cftr-/- and
controls UDCA treatment drastically reduced the proportion of CA in bile below 5% and
diminished CA synthesis (2.3±0.3 vs. 2.2±0.4 μmol/day/100gramBW, resp.; NS) and CA pool
size (3.6±0.6 vs. 1.5±0.3 µmol/100gramsBW, resp.; p<0.05).
Conclusion
Acute TUDCA infusion and chronic UDCA treatment both stimulate bile flow in CF conditions
independently from Cftr function. Chronic UDCA treatment reduces the hydrophobicity of the
bile salt pool in Cftr-/- mice. These results support a potential beneficial effect of UDCA on
bile flow and bile salt metabolism in CF conditions.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
115
6
INTRODUCTION
Cystic Fibrosis (CF) is caused by mutations in the Cystic Fibrosis Transmembrane conductance
Regulator (Cftr) gene which result in dysfunction of the CFTR protein (1, 2). CFTR is a cAMP
activated chloride channel and is present in the apical membrane of various epithelial and
non-epithelial tissues (3-5). Cirrhotic Cystic Fibrosis Liver Disease (CFLD) develops in ~5-10% of
CF patients and is the second leading cause of mortality in CF patients (6-8). Absence of
functional CFTR in biliary epithelium is thought to initiate abnormal secretin/cAMP-stimulated
chloride and bicarbonate secretion, leading to decreased bile flow and bile duct plugging by
thickened secretions. Secondary cholangiocyte and hepatocyte injury can ultimately lead to
the development of cirrhosis (9).
Treatment with ursodeoxycholate (UDCA) is applied for different cholangiopathies, including
CFLD. Although routinely applied in CF patients with increased levels of liver function
parameters in serum, the therapeutic action of UDCA in CF conditions remains unclear. It has
been hypothesized that UDCA is beneficial by its choleretic activity and/or its capacity to
correct aberrant bile salt metabolism (10). Indeed, it has been shown that UDCA stimulates
biliary secretion of bile acids in patients with primary biliary cirrhosis and primary sclerosing
cholangitis (11, 12). However, there are only a few trials assessing the efficacy of UDCA for
the treatment of CFLD and currently, there is insufficient evidence to justify the routine use of
UDCA in CF (13).
Shimokura et al. demonstrated in biliary cells that UDCA, in pharmacological concentrations,
increased intracellular calcium and induced chloride efflux. These scientists speculated that
UDCA increased bile flow by direct stimulation of ductular secretion, what could be of
therapeutic benefit for patients with CF who have impaired cyclic AMP-dependent biliary
secretion (14). However, Fiorotto et al. suggested that UDCA induced bile flow in a Cftr
dependent manner (15). They showed that UDCA stimulated cholangiocytic fluid secretion in
vitro in bile duct units and in isolated perfused livers from wild type, but not from Cftr
knockout mice. UDCA induced a Cftr-dependent ATP release, which, by activating the
purinergic signaling pathway, induced cholangiocyte secretion by stimulation of calcium-
activated chloride channels. These in vitro and ex vivo observations clearly supported the
concept that UDCA-induced fluid secretion is Cftr dependent and that, in CF conditions, UDCA
induced cholangiocytic choleresis can’t be achieved.
An alternative beneficial effect of UDCA in CF conditions could be related to changes in bile
salt metabolism. It is known, that the size and composition of the bile salt pool is critical for
adequate bile formation(16). In addition, a more hydrophobic bile salt composition is
suggested to be toxic to the hepatobiliary tract and might contribute to the development of
CFLD (17). Freudenberg et. al reported an increased biliary hydrophobicity index in Cftr508/508
mice compared to controls. Different effects are published on the influence of UDCA on bile
salt metabolism. Frenkiel et al. reported that UDCA reduced the cholate pool size and cholate
Chapter 6
116
6
synthesis rate in patients with gallstones and decreased the hydrophobicity of bile (18). In
contrast, Beuers et al. described that UDCA did not decrease the bile salt pool size in patients
with cholestatic liver disease (19). In UDCA treated patients with CFLD, duodenal bile became
enriched with the conjugated species of UDCA (accounting for 12% to 3% of the total biliary
bile salts), indicating an increased hydrophilic bile composition (20). Thus far, the effect of
UDCA on bile salt metabolism under CF conditions has not been conclusive.
We have reasoned that insights in the effects of UDCA on bile production, cholate
biosynthesis rate and cholate pool size under controlled in vivo CF conditions have been
lacking. In the present study we aimed to overcome this knowledge gap by exploiting recently
developed techniques allowing the study of these parameters in small experimental animals
(21, 22). We determined the effect of UDCA on bile flow and bile salt metabolism in Cftr
knockout mice and wild type littermates during acute intravenous tauroursodeoxycholate
(TUDCA) infusion and after chronic dietary administration of UDCA. Although biliary
phenotypes have been described in different CF mouse models the C57Bl/6;129 Cftr-/- tm1CAM
mouse model we use does not exhibit CF related gallbladder or liver abnormalities (23-25).
The lack of hepatobiliary abnormalities allows us to exclusively investigate the action UDCA
under complete Cftr null conditions, without the secondary interference of cholestasis on
biliary parameters.
MATERIAL AND METHODS
Animals and diets. C57Bl/6;129 Cftr-/- tm1CAM
mice and Cftr+/+tm1CAM
littermates were bred and
accommodated at the Animal Experimental Center of the Erasmus Medical Center in
Rotterdam, The Netherlands. Southern blotting of tail-clip DNA was performed to verify the
genotype of individual animals (25). In accordance with previous studies, the Cftr-/-
mice did
not exhibit CF liver or gallbladder abnormalities (data not shown). Mice were housed in a
light-controlled (lights on 6 AM to 6 PM) and temperature-controlled (21oC) facility, and were
allowed access to tap water and a semi-synthetic chow diet (SRM-A; Hope Farms BV
Woerden, The Netherlands) from the time of weaning. All experiments were performed on
female and male animals of 10-20 weeks of age. Group size varied per experiment from 5-9
animals per genotype. Experimental protocols were approved by the Ethical Committee for
Animal Experiments of the Erasmus Medical Center.
Experimental procedures. To evaluate the effect of UDCA on bile production and bile
composition, we used a mouse bile duct cannulation experimental model. This model
provides the option to measure bile production over an extended period of time. The animals
were placed in a temperature and humidity controlled incubator. Bile was collected after
surgical ligation of the common bile duct and cannulation of the gallbladder using
polyethylene tubing under intraperitoneal anesthesia with hypnorm (fentanyl/fluanisone 1
µl/gBW) and diazepam (10 µg/gram BW), as previously described (26). Bile secretions were
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
117
6
collected in 15 minute fractions during the stepwise dose increase phase and in 10 minutes
fraction at the highest dose (600 nmol.min-1
) for 60 minutes. Bile samples were used for bile
salt composition analysis. Bile flow rate was assessed gravimetrically, assuming a density of
1g/ml.
Acute intravenous TUDCA administration. We performed an acute bile salt infusion
experiment to evaluate the dose-response effect of acute UDCA supplementation on bile flow
and bile salt composition in Cftr-/-
and control mice. Acute TUDCA administration provides the
possibility to measure the direct choleretic effects of bile salts without the possible interfering
effects of adaptations to long-term bile salt administration. We infused taurine-conjugated
UDCA (TUDCA) for our acute infusion experiment, to closely mimic the physiological condition
in which bile salts are secreted into bile almost exclusively in conjugated form. An intravenous
line was placed in the jugular vein and the gallbladder was cannulated. After equilibration of
the bile flow for 5-10 minutes, spontaneous bile production was assessed for 30 minutes, i.e.
without TUDCA infusion. Subsequently, TUDCA solution (43 mM dissolved in phosphate-
buffered saline, pH 7.4) was administered using an IV pump(27). The TUDCA dosage was
increased every 30 minutes in a stepwise manner (dosage steps 150, 300, 450 and 600
nmol/min). The maximal dosage was given for 60 minutes. During TUDCA infusion, bile was
collected in 15 minute- fractions.
Chronic dietary UDCA administration. In CF patients, UDCA supplementation is given
chronically via the enteral route (28). Analogous to the human situation, we evaluated the
effect of chronic enteral (dietary) UDCA treatment in Cftr-/-
and control mice compared to
untreated animals. Mice were fed either a control diet consisting of standard chow or the
same diet enriched with UDCA (0.5% wt/wt) for 3 weeks. Body weight was measured after the
diet period. After gallbladder cannulation, spontaneous bile production was determined by
bile collection for 30 minutes. We additionally determined cholate synthesis and pool size,
using a previously developed and validated stable isotope dilution technique (22). In short,
3.0 mg of [2H4]-cholate in a solution of 0.5% NaHCO3 in phosphate-buffered saline was slowly
injected via the penile vein during isoflurane anesthesia. Blood samples were taken before
injection and at 12, 24, 36, 48, 60, and 72 h after injection. Blood samples (100 µl) were
collected by tail bleeding. Blood was collected in EDTA tubes and centrifuged to obtain
plasma. After centrifuging (3,000 rpm for 10 min at 4°C), plasma was stored at –20°C until
analysis. At the last day of the experiment (72 h), mice were anesthetized and equipped with
a catheter in the bile duct as described above. Subsequently, bile was collected for 30 min,
after an initial equilibration period of 5-10 min. Animals were euthanized by heart puncture.
Analytic procedures. Biliary bile salt concentrations were determined by an enzymatic
fluorimetric assay (29). Biliary bile salt composition was determined by capillary gas
chromatography (30). The hydrophobicity of bile salts in bile was calculated according to the
Heuman index based on the fractional contribution of the major murine bile salt species
cholate, chenodeoxycholate, deoxycholate, ursodeoxcholate, α-muricholate and β-
Chapter 6
118
6
muricholate (31). Alanine transaminase (ALT) was determined in plasma samples. Plasma and
bile samples were prepared for GLC-MS analysis on a Finnigan SSQ7000 Quadrupole GC-MS
machine as described previously by Stellaard et al. (32). The isotope dilution technique is
based on the dilution of a labeled tracer into the pool of the metabolite of interest. It has
been demonstrated to result in virtual identical cholate synthesis rates as obtained with a 14
C-
cholesterol bile salt synthesis measuring methods (33). The tracer is administered as a bolus,
which mixes into the pool. Shortly after mixing, the isotopic enrichment is highest. Thereafter,
the enrichment decreases due to dilution with unlabelled molecules introduced by de novo
synthesis. Enrichment of 2H4-cholate in plasma was determined as the increase of the M4-/M0-
cholate, relative to baseline measurements and was expressed as the natural logarithm of
atom percent excess (ln APE). From the decay curve of ln APE (calculated by linear regression
analysis), daily fractional turnover rate (FTR; equals the slope of the regression line) and pool
size ([administered amount of label x isotopic purity x 100] / intercept of the y-axis of the ln
APE curve) of cholate were calculated. Multiplying the pool size with the FTR results in a value
for the absolute turnover rate. In the steady-state situation, the absolute turnover rate equals
the synthesis rate. (22, 33, 34).
Statistical analysis. Statistical analysis was performed using SPSS version 16.0 for Windows
(SPSS Inc., Chicago, IL). All results are reported as means±SEM. Differences between
genotypes or diet groups were evaluated using the Mann-Whitney U test. The level of
significance was set at a P value of less than 0.05.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
119
6
RESULTS
The physiological state of the enterohepatic circulation in Cftr-/-
mice. We evaluated bile
production during the first 30 min after acute interruption of the enterohepatic circulation,
i.e. closely reflecting the physiological state (represented by time points 15 and 30 minutes in
Figure 1A-C). We found that bile flow (4.5±0.6 vs. 4.0±0.4 μl/min/100gram BW, NS), biliary
bile salt concentration (67±12 vs. 54±5 mM, NS) and biliary bile salt output (313±72 vs.
208±25 nmol/min/100gram BW, NS) were similar in Cftr-/-
and control mice respectively,
indicating no quantitative differences in te choleretic capacity between Cftr-/-
mice and their
controls at baseline.
0
5
10
15
Cftr+/+
ACftr-/-
150300
450
30
Time (minutes)
60 90 120 180
0
600 TUDCA infusion rate
(nmol/min)
Bile f
low
( l/
min
/100 g
ram
BW
)
0
500
1000
1500
2000
2500
Cftr+/+
CCftr
-/-
150300
450
30
Time (minutes)
60 90 120 180
600 TUDCA infusion rate
(nmol/min)
Bil
e s
alt
secre
tio
n r
ate
(nm
ol/
min
/100 g
ram
BW
)
0 500 1000 1500 2000 25000
5
10
15
D Cftr+/+
Cftr-/-
Bile salt secretion rate(nmol/min/100 gram BW)
Bil
e f
low
( l/
min
/100 g
ram
BW
)
0
50
100
150
200
Cftr+/+
BCftr-/-
150300
450
30
Time (minutes)
60 90 120 180
600 TUDCA infusion rate
(nmol/min)
Bil
e s
alt
co
ncen
trati
on
(mM
)
Figure 1 Biliary parameters during acute intravenous TUDCA administration. Biliary bile flow (A), bile salt
concentration (B), bile salt secretion rate (C) and relationship between bile salt secretion rate and bile flow (D) in Cftr
knockout mice (Cftr-/-) and control littermates (Cftr+/+) during acute intravenous infusion with TUDCA. The dosage of
TUDCA was increased at indicated intervals. The grey symbols in Fig. D represent baseline values without TUDCA
infusion. Data are presented as means ± SEM of N=5 mice per group. There was no significant difference between Cftr-
/- and Cftr+/+ mice, at any of the individual time points, for bile flow, bile salt concentration and bile salt secretion rate.
TUDCA administration increased bile flow equally in Cftr-/-
and control mice in a dose-
dependent manner. Infusions with TUDCA increased bile flow to a similar extend in Cftr-/-
mice
Chapter 6
120
6
and controls (+ ~250%; Figure 1A). The results indicate that TUDCA is capable of generating a
bile salt induced bile flow independent of the expression of CFTR. The biliary bile salt
concentration increased in parallel to the infused bile salt dosage (Fig. 1B). The bile salt
secretion rates, i.e. the product of flow (Figure 1A) and concentration (Figure 1B), increased in
equal pace with an increased TUDCA IV dosage and was not different between Cftr-/-
and
control mice (Figure 1C). We performed Mann–Whitney U tests for all individual time point
during the TUDCA infusion experiments. There was no significant difference between Cftr-/-
and Cftr+/+
mice at any of the individual time points for bile flow, bile salt concentration and
bile salt secretion rate. To determine the choleretic capacity of TUDCA in both Cftr-/-
and
control mice we related the bile flow to the bile salt secretion rate (Figure 1D) The choleretic
capacity of TUDCA was similar in Cftr-/-
and control mice, as was the estimated bile salt
dependent bile flow, based on the Y-intercepts (4.6 vs. 5.3 μl/min/100gram BW, respectively,
NS). The increase in bile flow was linearly related to the bile salt output (control: R2 0.9, slope
0.003 µmol/nmol; Cftr-/-
R2 0.9, slope 0.003 µmol/nmol).
Normal diet UCDA diet0
10
20
30
40
50
Cftr+/+
Cftr-/-
A
Bil
e f
low
( l/
min
/100 g
ram
BW
)
Normal diet UCDA diet0
25
50
75
100
Cftr+/+
Cftr-/-
*
BB
A c
on
cen
trati
on
(mM
)
Normal diet UCDA diet0
500
1000
1500
2000
Cftr+/+
Cftr-/-
*
C
BS
SR
(nM
/min
/100 g
ram
s B
W)
Normal diet UCDA diet0
25
50
Cftr+/+
Cftr-/-
*
D
Bo
dy w
eig
ht
(gra
m)
Figure 2 Biliary parameters after chronic dietary UDCA administration. Biliary bile flow (A), bile salt secretion (B) bile
salt secretion rate (C) body weight (D) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+) after a normal or
0.5%-UDCA chow diet for 3 weeks. Data are presented as means ± SEM of N=5-6 mice per group. *P-value<0.05.
UDCA treatment increased bile flow equally in Cftr-/-
and control mice. Chronic UDCA
treatment for 3 weeks increased bile flow with ~500% when compared to untreated mice
(p<0.05). This increase was comparable for Cftr-/-
and control mice (29.0±2.6 vs. 31.0±1.9
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
121
6
μl/min/100gram BW, resp.; NS; Figure 2A). Chronic UDCA treatment decreased the total
biliary bile salt concentration in both genotypes, but to a lower extent in Cftr-/-
mice (Cftr+/+
-
51% vs. Cftr-/-
mice -37%), resulting in significantly higher BS concentration in Cftr-/-
mice
compared to controls (42±3 vs. 26±3 mM, resp., p<0.05, Figure 2B). The bile salt output was
significantly higher in Cftr-/-
mice compared with controls during chronic UDCA treatment
(Cftr-/-
, 1232±147 vs. control, 827±113 μmol/min/100gram BW, resp.; p<0.05, Figure 2C). A
significant reduced growth of the Cambridge Cftr-/-
mice compared to wild type animals has
been reported (25). Since the findings of the differences in the bile composition on treatment
with UDCA may be affected by the nutritional status we measured body weight of the treated
and untreated mice (Figure 2D). In our current experiment the phenotype of the Cftr-/-
mice
includes a decreased body weight compared to controls, however this is not affected by
UDCA treatment.
Increased hydrophobic bile salt composition of Cftr-/-
mice compared to control mice. In non-
UDCA treated Cftr-/-
mice, the fractional biliary cholate content was higher compared with
control mice (61.4±1.5% vs. 46.5±3.8%, resp.; p<0.05; Table 1.) The natural biliary UDCA
enrichment was ~50% lower in Cftr-/-
mice compared with controls (2.7±0.4 vs. 6.0±0.5% resp.;
p<0.05; Fig. 2A). In non UDCA treated mice β-muricholate is the major hydrophilic biliary bile
salt in both Cftr-/-
and control mice (42±3,2 vs. 33±1,6% resp.; NS). Based on the fractional
contribution of the bile salt we calculated the biliary hydrophobicity index, according to
Heuman et al (14). The bile salt composition of non-UDCA treated Cftr-/-
mice was significantly
more hydrophobic than the control mice (0.05±0.002 vs. -0.07±0.005 resp.; p<0.01; Figure
3B).
Chapter 6
122
6
Biliary bile salt profile: proportional contribution of the major murine
bile salt species
Non-UDCA treatment UDCA treatment
Cftr+/+
Cftr-/-
Cftr+/+
Cftr-/-
CA 46.5 ± 3.8 61.5 ± 1.5* 2.7 ± 0.9† 4.0 ± 1.3
†
CDC 3.4 ± 0.4 2.1 ± 0.3 2.8 ± 0.5 1.3 ± 0.2*
DC 1.7 ± 0.4 0.7 ± 0.4 ND ND
UDCA 6.0 ± 0.5 2.7 ± 0.4* 83.0 ± 2.7† 82.0 ± 1.9
†
α-M ND ND 3.3 ± 0.3† 2.5 ± 0.3
†
β-M 42.0 ± 3.2 33.0 ± 1.6 6.5 ± 1.8† 8.7 ± 1.6
†
Table 1. Values are means ±SE in percent; N 4–6 mice per group. Biliary bile salt composition is shown of the mayor
bile salt species [cholate (CA), chenodeoxycholate (CDC), deoxycholate (DC), ursodeoxycholate (UDCA),α-muricholate
(α-M), and β-muricholate (β-M)] in cystic fibrosis transmembrane regulator (Cftr) knockout mice (Cftr_/_) and control
littermates (Cftr_/_) after a normal or 0.5% UDCA chow diet for 3 wk. ND, not detectable. Mann-Whitney U-test: *P<
0.05, difference between genotype (Cftr_/_ vs. Cftr_/_); †P <0.05, difference between treatment group (non-UDCA vs.
UDCA).
UDCA treatment decreased the biliary hydrophobicity of Cftr-/-
mice. After UDCA treatment,
the biliary bile salt composition changes extensively (Table 1.) After treatment the bile salt
composition consisted for more than ~80% of UDCA in both Cftr-/-
mice and controls (Figure
3A). UDCA treatment drastically reduced the fraction of cholate in the bile to below 5% in
both Cftr-/-
mice and controls. However, the fraction of β-muricholate was also reduced in Cftr-
/- and controls mice compared to the untreated animals (6.5±1.8 vs. 8.7±1.6% resp.; NS).
Taken together, UDCA treatment decreased the bile salt hydrophobicity index of Cftr-/-
mice to
wild type levels (-0.08±0.003 vs. -0.08±0.002 resp.; NS; Figure 3B). To evaluate the potential
hepatotoxic effects of UDCA and differences in biliary bile hydrophobicity we measured
plasma ALT levels in during normal diet and in UDCA treated animals (Figure 3C). Under
control diet conditions, although not significant, ALT was higher in CF than in control mice
(40±8 vs. 60±13 U/l resp.; NS). UDCA decreased ALT both in CF and control mice (31±9 vs.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
123
6
15±1 U/l resp.; NS). UDCA treatment significantly lowered plasma ALT levels in Cftr-/-
mice
compared to Cftr-/-
on control diet (15±1 vs. 60±13 U/l resp.; p<0.05).
Cftr+/+
Cftr-/-
Cftr+/+
Cftr-/-
0
25
50
75
100
Cholate
OthersUDCA
Normal diet UDCA Diet
A
%
-0.10
-0.05
0.00
0.05
0.10
Cftr+/+
Cftr-/-
*
B
Normal diet UDCA diet
Hyd
rop
ho
bic
ity i
nd
ex
Normal diet UCDA diet0
25
50
75
100
Cftr+/+
Cftr-/-
*
C
AL
T (
U/l
)
Figure 3 Bile salt composition after chronic dietary UDCA administration. (A) Percent contribution of the bile salts
cholate, ursodeoxcholate and others (chenodeoxycholate, deoxycholate, α-muri cholate and β-muricholate) in bile, (B)
Heuman index of total bile salts in bile representing the hydrophobicity of bile salts, (C) Alanine aminotransferase
(ALT) in plasma of Cftr-/- and control littermates (Cftr+/+) after a normal or 0.5%-UDCA chow diet for 3 weeks. Data are
presented as means ± SEM of 4-9 mice per group. *P-value<0.05.
Chapter 6
124
6
*
Normal diet UCDA diet0
10
20
30
40
Cftr+/+
Cftr-/-
*
A
Ch
ola
te p
oo
l siz
e
( m
ol/
100 g
ram
s B
W)
Normal diet UCDA diet0
5
10
15
20
Cftr+/+
Cftr-/-
*
B
Ch
ola
te s
yn
thesis
rate
( m
ol/
day/1
00 g
ram
BW
)
Figure 4 Cholate pool size and cholate synthesis after chronic dietary UDCA administration. Cftr knockout mice (Cftr-/-)
and control littermates (Cftr+/+) were intravenously injected with 2H4-cholate after a normal or 0.5%-UDCA chow diet
for 3 weeks Enrichment of the administered 2H4-cholate was determined in plasma until 72 hours after the
administered label. From the plasma decay curve, cholate pool size (A) and cholate synthesis rate (B) were calculated,
as detailed in the Materials and Methods. Data are presented as means ± SEM of 4-6 mice per group. *P-value<0.05.
UDCA treatment decreased cholate synthesis and pool size in Cftr-/-
and control mice. Non-
UDCA treated Cftr-/-
mice had a higher cholate synthesis rate (16±1 vs. 10±2 μmol/day/100
gram BW resp.; p<0.05) and larger cholate pool size (28±3 vs. 19±1 μmol/100 gram BW, resp.;
p<0.05) compared to controls (Figure 4).
Chronic UDCA treatment reduced the cholate pool size by ~90% in both Cftr-/-
and control
mice (p<0.05). Nevertheless, the pool size of Cftr-/-
mice remained higher compared with
controls (3.6±0.6 vs. 1.5±0.3 µmol/100 grams BW, resp.; p<0.05). Under the assumption that
steady state conditions of the bile salt kinetics were obtained after 3 weeks of treatment we
found that UDCA decreased the cholate synthesis rate by ~85% in both Cftr-/-
mice and
controls compared to untreated mice (p<0.05; Fig. 4B). Interestingly, UDCA treatment
straightened out the difference in synthesis rate between Cftr-/-
and control mice (2.3±0.3 vs.
2.2±0.4 μmol/hour/100 gram BW, resp.; NS).
DISCUSSION
The major physiologic effect of UDCA is its capacity to increase bile flow. This property
supports the therapeutic use of UDCA in a variety of cholangiopathies, including CFLD (35). In
vitro and ex vivo studies indicated that the stimulatory of UDCA on cholangiocyte secretion
depends on the presence of CFTR (14, 15). In the present study, we demonstrated that UDCA,
in vivo, either during acute or chronic administration, induced a significant Cftr independent
increase of bile flow in mice. Therefore, our results indicate that a positive choleretic effect of
UDCA can also to be expected in CF conditions. UDCA treatment reduced the relative
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
125
6
hydrophobic biliary bile salt composition in Cftr-/-
mice by replacing the high percent
contribution of the bile salt cholate by UDCA and by the quantitative reduction of the cholate
pool size. These properties could contribute to the assumed beneficial effects of UDCA in
CFLD.
Our present in vivo results are in apparent contrast with in vitro and ex vivo studies, which
report on the interaction between Cftr and bile salt stimulated biliary secretion (14, 15). In
these studies, an important role is ascribed to the function of calcium induced chloride
channels in UDCA stimulated bile flow, through the induction of purinergic signaling via Cftr
dependent ATP release. There can be several possibilities underlying the divergence of our in
vivo results from the in vitro and ex vivo reported studies. First, the choleretic effect found in
our in vivo studies probably predominantly reflects an osmotic, bile salt induced canalicular
bile flow, rather than a major ductular bile flow. The canalicular bile flow may thereby
predominate the Cftr dependent secretion effect of UDCA at the level of the cholangiocytes.
Second, the activation of Cftr independent routes may differ between the in vitro and in vivo
conditions. In vivo, UDCA may stimulate hepatocytes to secrete ATP into bile in a CFTR-
independent manner and subsequently induce bicarbonate secretion via paracrine purinergic
pathways linked to calcium-activated chloride channels (CaCC) in the cholangiocytes (36).
CaCCs have been suggested to play a more prominent role in epithelial fluid secretion in mice
than in other species, including pigs (37). This might therefore explain the relatively mild
phenotype in CF mouse models (38). Recently, Beuers et al. postulated the “bicarbonate
umbrella hypothesis”, by suggesting that biliary bicarbonate secretion in humans serves to
maintain an alkaline pH near the apical surface of hepatocytes and cholangiocytes in order to
prevent the uncontrolled membrane permeation of protonated glycine-conjugated bile(39).
In this concept, the bile acid receptor TGR5 (GPBAR-1), localized on the tip of the cilia of
apical cholangiocyte membranes in mice and humans could trigger a Cftr independent
signaling cascade resulting in cholangiocyte secretion.
During chronic UDCA administration, a CF biliary phenotype became apparent: an increased
bile salt secretion rate (Fig. 2C, p<0.05). The difference in secretion rate is based on the
product of two measured parameters. Since the majority of biliary bile salts are derived from
enterohepatic circulation, the most plausible explanation is an increased total BS pool size,
which during treatment is predominantly accounted for by UDCA (Fig 3A). It seems therefore
logical to assume that the total amount of bile salts in the enterohepatic circulation is
increased in CF mice. Indeed, this explanation seems to be supported by the ~50% higher
cholate pool size in CF mice fed a normal diet (Figure 5).
During bile salt treatment, however, CF and control mice were administered the same,
supraphysiological dosages of TUDCA and UDCA in the acute and chronic experiment,
respectively. We previously reported an increased fecal loss of bile salts in the Cftr-/-
mice
(40). Although UDCA could potentially influence intestinal fat malabsorption we could not find
an effect of UDCA treatment on the body weight in Cftr-/-
or control mice. Furthermore we
Chapter 6
126
6
recently reported in rats that UDCA does not influence fecal fat excretion(41). Therefore, our
results suggest a different bile salt kinetic steady state in CF mice, in which overcompensation
of bile salt synthesis results in an enlargement of the bile salt pool. Since the higher bile salt
secretion rate in CF mice was seen most prominently during chronic treatment, we speculate
that CFTR is involved in adaptation of the enterohepatic circulation to chronic bile salt
treatment, either at the level of the intestine, of the liver, or both.
Although differences exist in bile salt metabolism between mice and man (42), we found clear
similarities between CF patients and Cftr-/-
mice in bile composition. Untreated Cftr-/-
mice (i.e.
without UDCA treatment) have a more hydrophobic bile salt pool composition and an
increased cholate synthesis rate and pool size compared to control mice. This is in line with
the report by Freudenberg et al of the increased biliary hydrophobicity in their Cftr508/508
mice
model. Similar results have been found in CF patients: Strandvik et al. reported an increased
proportion of primary bile salts in serum and bile of CF, which has been attributed to an
enhanced bile salt biosynthesis in response to increased fecal bile salt disposal (43). The Cftr-/-
mice used in the present study are therefore comparable to human CF patients in this
respect.
The relatively hydrophobic bile salt composition in CF has been implied in the development of
liver disease in CF conditions, but definitive proof for this concept is (still) lacking (44).
Nevertheless, chronic UDCA treatment is apparently capable to partially correct the
hydrophobic bile salt profile in CF mice. Additionally UDCA treatment normalized the initially
increased liver function tests in CF mice, in agreement with our hypothesis.
We investigated the effect of UDCA under CF conditions in the absence of CF-related liver
disease (CLFD). The advantage of this approach is the ability to exclusively examine UDCA
effects on several bile salt parameters during CFTR deficient conditions, without possible
interference of secondary changes due to liver disease. The major induction of bile flow and
alterations in bile salt composition could contribute to a preventive role of UDCA on the
development of CLFD. It is unclear whether effects on bile flow and bile salt composition can
be found under conditions of CFLD. Nakagawa et. al. evaluated duodenal bile salt composition
after a two month UDCA-treatment in nine CF patients with CFLD (45). Similar to our results,
the percent contribution of UDCA increased, resulting in a more hydrophilic bile salt pool. The
contribution of UDCA was not as high as in our study (25%, compared with ~80% in the
present study), possibly due to the relatively low dosage of UDCA that was used (10-15
mg/kg/BW/day). In patients with primary sclerosing cholangitis UDCA treatment dosage of
25-30 mg/kg/day resulted in UDCA comprising 74% or the bile salt pool, with a corresponding
cholate reduction from 29% to 6% (46). In addition, a reduction in cholate pool size and
cholate synthesis is also described in gallstone patients treated with 750 mg UDCA (47).
Therefore, regarding the effect of UDCA on bile composition, we have no indication that the
observed effects will be absent during CF-related cholestasis or other signs of CFLD.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
127
6
Figure 5 Schematic representation of Cftr dependent effects of UDCA treatment on bile salt kinetics. Schematic
representation of the enterohepatic circulation of the primary bile salt cholate in Cftr knockout mice (Cftr-/-) and
control littermates, either untreated (top) or treated for 3 weeks with a 0.5%-UDCA chow diet for 3 weeks. Cholate
undergoes enterohepatic cycling (dotted lines). In the liver bile salt are excreted via the bile into the intestine, and
then almost completely reabsorbed at the level of the terminal ileum. Under steady state conditions the fecal cholate
loss is compensated for by de novo cholate synthesis in the liver as represented by the width of the black arrows
coming from the liver (µmol/day/100 gram BW), maintaining the total cholate pool size in equilibrium. The diameters
of the circles represent the magnitude of the pool size (μmol/100 gram BW). Cholate synthesis rate and pool size are
determined as detailed in the Materials and Methods. Data are presented as means ± SEM of 4-6 mice per group.
In conclusion, our results in mice in vivo indicate that UDCA exerts a choleretic effect and
influences the bile salt profile and synthesis, independent of the presence of functional CFTR.
When extrapolated to the human situation, this might imply that UDCA treatment results
both in CF and in non-CF individuals in increased choleresis, reduced bile salt synthesis and a
more hydrophilic bile salt pool. Interpretation of the present results for the human (CF)
condition needs to take into account the possibility of species specificity of the observed
effects. However the outcome of this study does provides a firm experimental basis to explain
the beneficial effects of UDCA observed in CF patients
Chapter 6
128
6
REFERENCE LIST
1. Collins FS. Cystic fibrosis: Molecular biology and therapeutic implications. Science. 1992
05/08;256(5058):774-9.
2. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.
Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989
09/08;245(4922):1073-80.
3. Stutts MJ, Canessa CM, Olsen JC, Hamrick M, Cohn JA, Rossier BC, et al. CFTR as a cAMP-
dependent regulator of sodium channels. Science. 1995 08/11;269(5225):847-50.
4. Lamhonwah AM, Bear CE, Huan LJ, Chiaw PK, Ackerley CA, Tein I. Cystic fibrosis
transmembrane conductance regulator in human muscle: Dysfunction causes abnormal
metabolic recovery in exercise. Ann Neurol. 2010;67(6):802-8.
5. Li H, Ganta S, Fong P. Altered ion transport by thyroid epithelia from CFTR−/− pigs suggests
mechanisms for hypothyroidism in cystic fibrosis. Exp Physiol. 2010;95(12):1132-44.
6. Lindblad A, Hultcrantz R, Strandvik B. Bile-duct destruction and collagen deposition: A
prominent ultrastructural feature of the liver in cystic fibrosis. Hepatology. 1992 08;16(2):372-
81.
7. Colombo C, Battezzati PM, Crosignani A, Morabito A, Costantini D, Padoan R, et al. Liver
disease in cystic fibrosis: A prospective study on incidence, risk factors, and outcome.
Hepatology. 2002 12;36(6):1374-82.
8. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre
experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008
05;7(3):252-7.
9. Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride
currents in rat bile duct epithelial cells. J Clin Invest. 1993 01;91(1):319-28.
10. Hofmann AF. Biliary secretion and excretion in health and disease: Current concepts. Ann
Hepatol. 2007 01;6(1):15-27.
11. Lindor K, Therneau T, Jorgensen R, Malinchoc M, Dickson E. Effects of ursodeoxycholic acid
on survival in patients with primary biliary cirrhosis. Gastroenterology. 1996;110(5):1515-8.
12. Lindor K, Jorgensen R, Anderson M, Gores G, Hofmann A, LaRusso N. Ursodeoxycholic acid
and methotrexate for primary sclerosing cholangitis: A pilot study. Am J Gastroenterol.
1996;91(3):511-5.
13. Cheng K, Ashby D, Smyth R. Ursodeoxycholic acid for cystic fibrosis-related liver disease.
Cochrane Database Syst Rev. 2000(2):CD000222.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
129
6
14. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic
calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995
09;109(3):965-72.
15. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic
acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.
Gastroenterology. 2007 11;133(5):1603-13.
16. Hofmann AF. Chemistry and enterohepatic circulation of bile acids. Hepatology.
1984;4(S2):4S-14S.
17. Benedetti A, Alvaro D, Bassotti C, Gigliozzi A, Ferretti G, La Rosa T, et al. Cytotoxicity of bile
salts against biliary epithelium: A study in isolated bile ductule fragments and isolated
perfused rat liver. Hepatology. 1997;26(1):9-21.
18. Frenkiel PG, Lee D, Cohen H, Gilmore C, Resser K, Bonorris G, et al. The effect of diet on bile
acid kinetics and biliary lipid secretion in gallstone patients treated with ursodeoxycholic acid.
Am J Clin Nutr. 1986;43(2):239.
19. Beuers U, Spengler U, Zwiebel FM, Pauletzki J, Fischer S, Paumgartner G. Effect of
ursodeoxycholic acid on the kinetics of the major hydrophobic bile acids in health and in
chronic cholestatic liver disease. Hepatology. 1992;15(4):603-8.
20. Colombo C, Setchell KDR, Podda M, Crosignani A, Roda A. Effects of ursodeoxycholic acid
therapy for liver disease associated with cystic fibrosis*. J Pediatr. 1990;117(3):482-9.
21. Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst
Fibros. 2004 08;3 Suppl 2:183-90.
22. Hulzebos CV, Renfurm L, Bandsma RH, Verkade HJ, Boer T, Boverhof R, et al. Measurement
of parameters of cholic acid kinetics in plasma using a microscale stable isotope dilution
technique: Application to rodents and humans. J Lipid Res. 2001;42(11):1923.
23. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol. 2004 04;164(4):1481-93.
24. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.
Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.
Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.
25. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of
a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.
26. Kuipers F, Van Ree J, Hofker MH, Wolters H, Veld G, Havinga R, et al. Altered lipid
metabolism in apolipoprotein e–deficient mice does not affect cholesterol balance across the
liver. Hepatology. 1996;24(1):241-7.
Chapter 6
130
6
27. Verkade HJ, Havinga R, Shields DJ, Wolters H, Bloks VW, Kuipers F, et al. The
phosphatidylethanolamine N-methyltransferase pathway is quantitatively not essential for
biliary phosphatidylcholine secretion. J Lipid Res. 2007 09;48(9):2058-64.
28. Colombo C, Setchell KDR, Podda M, Crosignani A, Roda A. Effects of ursodeoxycholic acid
therapy for liver disease associated with cystic fibrosis*. J Pediatr. 1990;117(3):482-9.
29. Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin
Chim Acta. 1976 07/01;70(1):79-86.
30. Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic
circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the
absence of ileal bile acid-binding protein (ibabp). J Biol Chem. 2003;278(43):41930-7.
31. Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. correlation
between biliary bile salt hydrophobicity index and the activities of enzymes regulating
cholesterol and bile acid synthesis in the rat. J Lipid Res. 1989;30(8):1161-71.
32. Stellaard F, Langelaar S, Kok R, Jakobs C. Determination of plasma bile acids by capillary
gas-liquid chromatography-electron capture negative chemical ionization mass
fragmentography. J Lipid Res. 1989;30(10):1647.
33. Mitchell JC, Stone BG, Duane WC. Measurement of bile acid synthesis in man by release of
14 CO 2 from [26-14 C] cholesterol: Comparison to isotope dilution and assessment of
optimum reference cholesterol specific activity. Lipids. 1992;27(1):68-71.
34. Stellaard F, Brufau G, Boverhof R, Jonkers EZ, Boer T, Kuipers F. Developments in bile acid
kinetic measurements using 13C and 2H: 105 times improved sensitivity during the last 40
years†. Isotopes Environ Health Stud. 2009;45(4):275-88.
35. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of
action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.
36. Minagawa N, Nagata J, Shibao K, Masyuk AI, Gomes DA, Rodrigues MA, et al. Cyclic AMP
regulates bicarbonate secretion in cholangiocytes through release of ATP into bile.
Gastroenterology. 2007;133(5):1592-602.
37. Clarke LL, Grubb BR, Yankaskas JR, Cotton CU, McKenzie A, Boucher RC. Relationship of a
non-cystic fibrosis transmembrane conductance regulator-mediated chloride conductance to
organ-level disease in cftr(-/-) mice. Proc Natl Acad Sci U S A. 1994 01/18;91(2):479-83.
38. Liu X, Luo M, Zhang L, Ding W, Yan Z, Engelhardt JF. Bioelectric properties of chloride
channels in human, pig, ferret, and mouse airway epithelia. American journal of respiratory
cell and molecular biology. 2007;36(3):313.
Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice
131
6
39. Beuers U, Hohenester S, de Buy Wenniger LJM, Kremer AE, Jansen PLM, Elferink RPJO. The
biliary HCO3- umbrella: A unifying hypothesis on pathogenetic and therapeutic aspects of
fibrosing cholangiopathies. Hepatology. 2010;52(4):1489-96.
40. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat
absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am
J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.
41. Cuperus FJC, Hafkamp AM, Havinga R, Vitek L, Zelenka J, Tiribelli C, et al. Effective
treatment of unconjugated hyperbilirubinemia with oral bile salts in gunn rats.
Gastroenterology. 2009;136(2):673,682. e1.
42. Hofmann AF, Hagey LR, Krasowski MD. Bile salts of vertebrates: Structural variation and
possible evolutionary significance. J Lipid Res. 2010;51(2):226.
43. Strandvik B, Einarsson K, Lindblad A, Angelin B. Bile acid kinetics and biliary lipid
composition in cystic fibrosis. J Hepatol. 1996 07;25(1):43-8.
44. Smith JL, Lewindon PJ, Hoskins AC, Pereira TN, Setchell KDR, O'Connell NC, et al.
Endogenous ursodeoxycholic acid and cholic acid in liver disease due to cystic fibrosis.
Hepatology. 2004;39(6):1673-82.
45. Nakagawa M, Colombo C, Setchell KDR. Comprehensive study of the biliary bile acid
composition of patients with cystic fibrosis and associated liver disease before and after UDCA
administration. Hepatology. 1990;12(2):322-34.
46. Sinakos E, Marschall HU, Kowdley KV, Befeler A, Keach J, Lindor K. Bile acid changes after
high‐dose ursodeoxycholic acid treatment in primary sclerosing cholangitis: Relation to
disease progression. Hepatology. 2010;52(1):197-203.
47. Hofmann AF. Medical treatment of cholesterol gallstones by bile desaturating agents.
Hepatology. 1984;4(S2):199S-208S.
Chapter 6
132
6
CHAPTER 7
SUMMARY AND GENERAL
DISCUSSION
Chapter 7
134
7
GENERAL SUMMARY AND DISCUSSION
In this thesis, we studied clinical and basic aspects of the hepatic and gastrointestinal
involvement in cystic fibrosis. We focused on the physiological and pathological role of bile
salts and the enterohepatic circulation in Cystic fibrosis conditions. From our clinical study,
we learned that, based on the involvement of gamma-GT (GGT), cirrhotic Cystic fibrosis liver
disease (CCFLD) is primarily a biliary disease since GGT elevation is a predictor for the
development of CCFLD. From our experiments in CF mice models we conclude that cystic
fibrosis related liver disease is not related to biliary bile salt cytotoxicity. Nevertheless, in
different CF mice models we could establish a CFTR dependent and bile salt specific, biliary
phenotype. We provided proof that this biliary phenotype can, partially, be corrected by the
hydrophilic bile salt UDCA. Finally, we demonstrated that the CF phenotype in mice includes
alterations in the metabolism and enterohepatic circulation of bile salts. These changes are
due to interaction between the CFTR protein and the bacterial flora of the gut. Our findings
provide new experimental evidence for the significant role of CFTR in the equilibrium of the
enterohepatic circulation. Impaired CFTR function, like in Cystic fibrosis disease conditions,
can therefore give rise to negative alterations in the enterohepatic circulation and bile salt
metabolism. Our results provide new understanding in the gastro-intestinal and hepatic role
of CFTR that provides new insights for development of future treatment options in Cystic
fibrosis.
CCFLD is a life-threatening, hepatic complication of CF (1, 2). The clinical presentation of
CCFLD is often characterized by severe portal hypertension and gastrointestinal variceal
bleeding (3). In patients with end stage CCFLD liver transplantation might be indicated (4, 5).
Therefore patients at risk for or still in early stages of the disease could benefit from
preventive treatment for CCFLD. However, to date no predictors to identify patients at risk for
CCFLD were available. In chapter 2, we report that, in the time period 2 years prior to the
diagnosis of CCFLD, an increase of GGT, even if the level remains within the normal range,
indicates a significantly increased risk for the development CCFLD. This finding provides new
opportunities for identifying high risk patient for CCFLD and more targeted preventive or
preemptive therapy strategies like UDCA.
Within a few years after the CFTR gene was identified in humans, the first genetically
engineered CF mouse models were developed (6-9). These models have been of great
importance for increasing the understanding Cftr physiology and pathology in various organ
systems. However, the phenotype of the developed CF mice models differs in several aspects
from the human CF disease equivalent. For example lung disease, the most prominent CF
disease feature in CF patients, is barely found in mice models (10). On the other hand, CF
mouse models exhibit a wide range of gastrointestinal phenotypes that closely resemble the
human CF condition (11). In chapter 3, we reviewed current knowledge on the hepatic and
Summary and general discussion
135
7
gastrointestinal CF phenotypes in mouse models and in human patients. We focused on
chronic intestinal fat malabsorption, present in many CF patients. We indentified useful
similarities in the phenotype between mice and man. For example, the comparable
phenotype of increased fecal bile salt loss and alterations in bile salt composition in both CF
mice and CF patients (12). Also, the intestinal mucosal abnormalities like small intestinal
bacterial overgrowth (SIBO), intestinal mucus accumulation, intestinal inflammation and
increased intestinal permeability have corresponding phenotypes in CF humans and mice(13,
14). From these findings we concluded that the CF mouse, despite differences in phenotypic
expression among different genetic mice backgrounds, can well serve as useful experimental
model for gastrointestinal CF phenomena. Based on this analysis, our studies in CF mice
models could provide an educated basis for future translational research.
We investigated to what extent CF mouse models can provide pathophysiological and
mechanistic understanding in the development of CF related hepatic disease (chapter 4-6).
Our main focus was to determine whether biliary bile salt cytotoxicity is involved in the
development of CFLD like disease in CF mice. To answer this question we analyzed the liver
involvement in various CF mouse models. In the Cftr-/-tm1Unc
CF mice, a model characterized by
spontaneous development of CFLD like disease, we could not establish a contribution of
biliary bile salt cytotoxicity to its hepatic histo-pathologic phenotype. On the other hand, in
other CF mouse models we did find a significant increase in the hydrophobic profile of the
biliary bile. However, even in these mice models, this increased biliary cytotoxicity could not
be related to CFLD like disease. Finally, we could not prove that CF mice, when challenged
with and prolonged hydrophobic bile salt exposure, were more susceptible to develop CFLD
like disease. Based on these combined results we conclude that the development of CFLD like
disease, at least in mice, is not related to Cftr related changes in biliary cytotoxicity. We
postulate that this conclusion can be extrapolated to the development of liver involvement in
humans with CF.
Although we could not establish a specific histo-pathological phenotype in mice that was
corresponding with CCFLD in humans, we did find a unique hepatic and biliary phenotype.
(chapter 4). Cftr-/-
mice, when exposed to the hydrophobic bile salt cholate, over an extended
period of time, display an increased hydrophobicity of bile and a decreased capacity to
increase bile flow. Cftr-/-
mice are apparently less able to dilute the bile during chronic cholate
administration. This is clearly illustrated by a significantly increased biliary BS concentration
but unaffected BS secretion rates. We observed a distinct Cftr dependent liver growth and
proliferation response, on prolonged hydrophobic BS exposure: in wild type mice the liver
weight increased by ~50% upon a chronic BS administration, whereas liver weight was stable
in different CF mouse models. These observations point to a, hitherto unrecognized, role of
CFTR in the hepatic adaptive response to prolonged hydrophobic BS exposure. We speculate
that, this impaired regenerative capability of the CF phenotype, could be related to an
increased hepatic vulnerability, for example in cholestatic conditions. This deficiency in
regenerative hepatic capacity may play a role in liver disease development.
Chapter 7
136
7
Our results in Cftr-/-tm1Unc
mice (chapter 5) demonstrated Cftr dependent differences in
intestinal bile salt metabolism. We found, Cftr specific, alterations in the bacterial formation
of the secondary bile salts and accordingly in the fecal and biliary bile salt composition. As a
result we found that in Cftr-/-tm1Unc
mice, exclusively fed a liquid diet, the biliary bile salt profile
was considerably more hydrophilic and the bile flow was increased compared to wild type
mice. The increase in bile production, in the Cftr-/-tm1Unc
, could be explained by increased
concentrations of ursocholate (UCA). The contribution of UCA to the bile salt pool is usually
relatively small, due to its rapid biotransformation into DCA by the intestinal bacterial flora.
Therefore, the reported increased UCA enrichment of the Cftr-/-tm1Unc
mice can only be
explained by, specific Cftr related, changes in the functioning of intestinal bacterial flora. UCA
is a very hydrophilic and highly choleretic bile salt. Therefore, UCA enrichment in the CftrUnc-/-
mice, results in a significantly increased bile salt dependent bile flow and in a significantly
decreased biliary bile hydrophobicity. Our results provide a clear example of the way in which
alterations in the interaction between intestinal flora and bile salt formation can significantly
affect the enterohepatic circulation of bile salt. In CF conditions these changes, in bile salt
metabolism, could be related to its phonotypical characteristics or even result in induction or
prevention of liver disease development. Several intestinal mucosal abnormalities are
described in CF patients and CF mice models that could influence the composition of the
bacterial micro flora. These abnormalities include accumulation of viscous and sticky mucus,
small intestinal bacterial over growth, increased intestinal permeability and inflammation of
the small intestine (15-24).
The importance of the intestinal bacterial flora for the CF gastrointestinal phenotype was also
illustrated by other experiments. We treated control and CF mice with either the hydrophilic
bile salt UDCA or the hydrophobic bile salt cholate (CA). We found that control mice, on CA
diet, displayed a slightly more hydrophobic bile composition compared to the Cftr-/-
mice. This
effect, in wild type mice, can, in its entirety, be explained by the presence of the secondary
bile salt deoxycholate (DCA), not found in bile of Cftr-/-
mice. DCA is produced exclusively, via
biotransformation, by the intestinal micro flora. Recently, we and others have reported that
intestinal micro flora and the intestinal mucosal permeability are changed in CF (25-28). Our
results emphasize the critical role of the intestinal micro flora, for CF conditions, in modifying
the enterohepatic circulation of bile salts. We believe that, Cftr specific, genotypic changes in
intestinal flora are primary responsible for a majority of the, CF related, alterations in bile salt
composition and production. These observations open the perspective for targeted
manipulation of the intestinal micro flora in CF mice and patients.
As stated, the Cftr dependent changes in bile salt composition and metabolism can have
profound effects on the enterohepatic circulation in CF conditions. Yet, to some extent, the
observed changes in bile salt formation can also result from, Cftr related, alterations in the
enterohepatic circulation itself. The reported difference in bacterial BS biotransformation, in
particular the conversion CA to DCA, can indirectly affect hepatic bile salt synthesis. For
example DCA is a strong activating ligand of the farnesoid X receptor (FXR) (29). This nuclear
Summary and general discussion
137
7
receptor is highly expressed in the liver and ileum (30). In the liver, an important function of
FXR activation is the suppression of bile acid synthesis from cholesterol (31). Activation of
intestinal FXR induces the expression of FGF19 (fibroblast growth factor 19, equivalent to
Fgf15 in rodents) by the enterocytes (32). FGF19 is excreted and transported via the portal
circulation to the liver. Here it can induce the cascade of bile salt synthesis from cholesterol.
Additionally, recent publications describe the central role of FGF15/19 in hepatocyte
proliferation (33, 34). Normally FGF15/19 is only expressed in the terminal ileum and not in
normal liver. However, hepatic expression of FGF15/19 can occur in cholestatic conditions
and is associated with hepatic adaptation aimed at protecting against BS injury (35).
In our chronic hydrophobic bile salt (cholate) exposure model, expression of Fgf15 could
potentially be induced both in the intestine and in the liver. The increased biliary fraction of
DCA, found in the controls after chronic CA diet, can induce Fgf15 expression via FXR in the
enterocytes or directly by inducing Fgf15 expression in the liver. In either situation, induction
of Fgf15 could induce liver proliferation and liver growth (36). There could also be a primary
role for the intestinal flora. A recent study in germ free mice showed that gut microbiota do
regulate bile salt metabolism by influencing naturally occurring FXR antagonists like tauro-
beta-muricholic acid (37). On the other hand cholate induced liver growth in mice, similar to
our current findings in controls, was described previously. Huang et al. reported that CA
feeding (0.2%) for 5 days stimulated normal liver growth and increased the relative liver
weight by approximately 30% in wild type mice (38). These investigators also observed that
CA supplementation enhanced liver regeneration after partial hepatectomy. They concluded
that normal liver regeneration depends on BS activation via nuclear receptor-dependent
signaling pathways, in particular FXR.
The potential role of FXR is underlined by our finding that liver growth is not observed during
chronic UDCA exposure. UDCA does not lead to increased presence of strong FXR ligands like
DCA. Current findings offer opportunities for future research. To differentiate the flora effect
from the bile salt effect in regulating FXR it would be of value to develop germ free CF mice to
repeat the experiments. Furthermore it would be of great interest to perform a cholate diet
experiment in FXR intestinal and liver tissue specific knockout mice models. These models
provide the opportunity to separate the intestinal FXR stimulation from the hepatic FXR
effects on liver growth.
UDCA as a preventive treatment option for cirrhotic CFLD remains controversial (39). An
important factor in this argument is the lack of support by basic experimental evidence.
Additionally there or no well-designed clinical study showing unequivocally a long term
clinical benefit on relevant clinical outcome (40). A major physiological effect of UDCA is its
capacity to increase bile flow. This property has supported the therapeutic use of UDCA in a
variety of cholangiopathies, including CFLD (41). In vitro and ex vivo studies indicated,
however that the stimulatory role of UDCA on cholangiocyte secretion depends on the
presence of CFTR (42, 43). In the present thesis (chapter 6), we studied the effects of UDCA in
Chapter 7
138
7
vivo in CF mice. We demonstrated that UDCA, in vivo, either during acute or chronic
administration, induced a significant, Cftr independent, increase of bile flow in mice.
Therefore, our results suggest that a positive choleretic effect of UDCA can also be expected
in human CF conditions. Furthermore, UDCA treatment reduced the relative hydrophobic
biliary bile salt composition in Cftr-/-
mice. In bile UDCA replaced cholate and reduced the
quantitative total pool size of cholate. These properties could contribute to the assumed
beneficial effects of UDCA in CFLD. However, we would like to underline that our results do
not provide evidence for preventive effects of UDCA on the development of liver fibrosis.
We feel that the results of our studies lead to an improved understanding of the development
of Cystic fibrosis liver disease. Our results provide new opportunities for future clinical and
basic research. This is especially relevant given the current groundbreaking therapeutic
developments in CF. These new therapies are target to directly correct the basic genetic CFTR
defect (44). To explore these new therapeutic options further, early intervention in young
children, who have not yet developed severe pulmonary disease, would be most favorable
(45). Classic lung function measurements, the traditional clinical endpoint in CF clinical
research, may not be the most indicative in young patients still without a measurable
decrease pulmonary function. To perform future therapeutic studies new clinical endpoints
and intermediate endpoints for Cystic fibrosis disease need to be developed (46).
The results of his thesis offer additional options for the development of new gasto-intestinal
and hepatic outcome parameters for future clinical trials (47). For example measuring
differences in fecal bile secretion or changes in bile salt metabolism might have high potential
as clinical outcome variables. Total bile salt secretions, as well as differences in fecal bile salt
formation, are relatively easily obtainable results and do not require an invasive approach.
These measurements are feasible to achieve even in young children. Another example that
open new research opportunities could be the ever growing possible study methods to
identify and quantify the intestinal bacterial micro flora. The CFTR specific changes in the
interaction of the intestinal bacterial flora with intestinal bile salt metabolism and absorption
as reported in this thesis can serve as the basis for further research in this direction.
Moreover, our results proved new opportunities for treatment options for CCFLD. Based on
our results, it appears justifiable to evaluate the prognostic value of GGT to identify patients
at risk for severe cirrhotic CFLD in a large prospective cohort study. This study can than
provide the necessary validation and support for clinical use and interventional therapeutic
strategies. Simultaneously, therapeutic interventional and observational (registry) studies
could provide additional candidate parameters on gastrointestinal function. These could
serve as new parameters to evaluate either the severity of the gastrointestinal phenotype in
general or for the development of CCFLD in particular.
In addition to clinical, patient-oriented studies, we feel that continuation of the basic studies
in relevant model systems is essential. Concerning the search for the pathogenesis of CCFLD
we believe the focus should switch from bile salt cytotoxicity to the liver-intestine-micro flora
Summary and general discussion
139
7
interaction. The results presented in this thesis provide strong support for the concept that
CFTR dependent changes in the intestinal bacterial flora composition are strongly related to
intestinal inflammation and intestinal bile salt metabolism. Focus on pathophysiology of small
intestinal bacterial overgrowth (SIBO) and of the intestinal flora in general in CF seems
indicated. The role of intestinal paneth cells, secreting anti-bacterial proteins into the
intestinal lumen, may also be relevant to delineate in CF conditions.
Another line of logical continuation of our current research involves the hepatic-intestinal axis
and the enterohepatic circulation of bile salt in CF conditions. Our results show a distinct CFTR
dependent phenotype of bile salt synthesis, intestinal bile salt biotransformation and fecal
bile salt loss. This interaction of CFTR function with the enterohepatic circulation of bile salts
including the interaction with the intestinal bacterial flora needs further exploration.
Finally there are new developments in possibilities to study different CF animal models.
Recently CF ferrets and CF pigs were generated and characterized in experimental studies (48,
49). The firsts result of these studies show that, in particular the CF pigs, demonstrate a
hepatic CFLD like phenotype (50). These new models provide opportunities to further study
CFLD development and treatment.
In conclusion we believe that our research provides key additional expertise in the field of
gastro-intestinal and hepatic consequence of Cystic fibrosis disease. We feel that our present
results, in combination with the current promising developments in CF drug development, the
expansion of global patient registries and the development of new experimental animal
models, add to improve the understanding and the treatment for CF disease in the near
future.
Chapter 7
140
7
REFERENCE LIST
1. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory
Reviews. 2013.
2. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.
Hepatology. 1999 11;30(5):1151-8.
3. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic
fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.
1999;31(1):77-83.
4. Chryssostalis A, Hubert D, Coste J, Kanaan R, Burgel P, Desmazes-Dufeu N, et al. Liver
disease in adult patients with cystic fibrosis: A frequent and independent prognostic factor
associated with death or lung transplantation. J Hepatol. 2011;55(6):1377-82.
5. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in
cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.
6. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.
Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989
09/08;245(4922):1073-80.
7. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of
the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.
1989;245(4922):1066.
8. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of
a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.
9. van Doorninck JH, French PJ, Verbeek E, Peters R, Morreau H, Bijman J, et al. A mouse model
for the cystic fibrosis delta F508 mutation. EMBO J. 1995;14(18):4403.
10. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse
models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.
2011;10:S152-71.
11. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol. 2004 04;164(4):1481-93.
12. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.
Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption
in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.
13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial
overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr. 2007;44(2):212-8.
Summary and general discussion
141
7
14. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun.
2004;72(10):6040-9.
15. Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal
permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.
Pediatr Res. 1996;39(6):985-91.
16. Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic
fibrosis: Importance of breath testing. J Paediatr Child Health. 1998;34(1):79-82.
17. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in
cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.
Acta Biochim Pol. 2009;56(4):631.
18. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal
inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.
Aliment Pharmacol Ther. 2004;20(7):813-9.
19. Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic
inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res.
2000;47(3):344-50.
20. Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the
small bowel of patients with cystic fibrosis. Pediatr Res. 1998;43(2):234-9.
21. Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic
fibrosis. Arch Dis Child. 2000;82(5):394-9.
22. Quigley EM, Quera R. Small intestinal bacterial overgrowth: Roles of antibiotics, prebiotics,
and probiotics. Gastroenterology. 2006;130(2):S78-90.
23. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun.
2004;72(10):6040.
24. De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus
accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small
intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology.
2007;293(3):G577-84.
25. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.
2006;42(1):46-52.
26. De Lisle RC, Mueller R, Boyd M. Impaired mucosal barrier function in the small intestine of
the cystic fibrosis mouse. J Pediatr Gastroenterol Nutr. 2011;53(4):371-9.
Chapter 7
142
7
27. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade
HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.
2011;71(1):4-12.
28. Martin CR, Zaman MM, Ketwaroo GA, Bhutta AQ, Coronel E, Popov Y, et al. CFTR
dysfunction predisposes to fibrotic liver disease in a murine model. American Journal of
Physiology-Gastrointestinal and Liver Physiology. 2012;303(4):G474-81.
29. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for
the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.
30. Vaquero J, Monte MJ, Dominguez M, Muntané J, Marin JJG. Differential activation of the
human farnesoid X receptor depends on the pattern of expressed isoforms and the bile acid
pool composition. Biochem Pharmacol. 2013 10/1;86(7):926-39.
31. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory
cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol
Cell. 2000;6(3):517-26.
32. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, et al. Differential regulation of bile acid
homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res. 2007
12;48(12):2664-72.
33. Wu X, Ge H, Lemon B, Vonderfecht S, Weiszmann J, Hecht R, et al. FGF19-induced
hepatocyte proliferation is mediated through FGFR4 activation. J Biol Chem.
2010;285(8):5165-70.
34. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, et al. FGF19
regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and
independent pathways. PloS one. 2011;6(3):e17868.
35. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM. High expression of the bile
salt‐homeostatic hormone fibroblast growth factor 19 in the liver of patients with extrahepatic
cholestasis. Hepatology. 2009;49(4):1228-35.
36. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, et al.
Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its
application in the prevention of post-resection liver failure in mice. Gut. 2013;62(6):899-910.
37. Sayin S, Wahlström A, Felin J, Jäntti S, Marschall H, Bamberg K, et al. Gut microbiota
regulates bile acid metabolism by reducing the levels of tauro-beta-muricholic acid, a naturally
occurring FXR antagonist. Cell Metabolism. 2013 2/5;17(2):225-35.
38. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent
bile acid signaling is required for normal liver regeneration. Science. 2006
04/14;312(5771):233-6.
Summary and general discussion
143
7
39. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-
associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.
40. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.
Cochrane Database Syst Rev. 2012;10.
41. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of
action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.
42. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic
acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.
Gastroenterology. 2007 11;133(5):1603-13.
43. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic
calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995
09;109(3):965-72.
44. Ramsey BW, Davies J, McElvaney NG, Tullis E, Bell SC, Dřevínek P, et al. A CFTR potentiator
in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011 11/03;
2012/09;365(18):1663-72.
45. Ashlock MA, Olson ER. Therapeutics development for cystic fibrosis: A successful model for
a multisystem genetic disease. Annu Rev Med. 2011;62:107-25.
46. De Boeck K, Bulteel V, Tiddens H, Wagner T, Fajac I, Conway S, et al. Guideline on the
design and conduct of cystic fibrosis clinical trials: The european cystic fibrosis Society–Clinical
trials network (ECFS-CTN). Journal of Cystic Fibrosis. 2011;10:S67-74.
47. Rowe SM, Borowitz DS, Burns JL, Clancy JP, Donaldson SH, Retsch-Bogart G, et al. Progress
in cystic fibrosis and the CF therapeutics development network. Thorax. 2012;67(10):882-90.
48. Rogers CS, Stoltz DA, Meyerholz DK, Ostedgaard LS, Rokhlina T, Taft PJ, et al. Disruption of
the CFTR gene produces a model of cystic fibrosis in newborn pigs. Science.
2008;321(5897):1837-41.
49. Sun X, Sui H, Fisher JT, Yan Z, Liu X, Cho H, et al. Disease phenotype of a ferret CFTR-
knockout model of cystic fibrosis. J Clin Invest. 2010;120(9):3149.
50. Stoltz DA, Rokhlina T, Ernst SE, Pezzulo AA, Ostedgaard LS, Karp PH, et al. Intestinal CFTR
expression alleviates meconium ileus in cystic fibrosis pigs. J Clin Invest. 2013;123(6):2685.
Chapter 7
144
7
CHAPTER 8
NEDERLANDSE DISCUSSIE EN
SAMENVATTING
Chapter 8
146
8
NEDERLANDSE SAMENVATTING
Dit proefschrift beschrijft onze klinische en basale experimentele onderzoeken naar de
hepatologische- en gastro-intestinale aspecten van de ziekte Cystic fibrosis (CF). We hebben
ons hierbij in het bijzonder gericht op de fysiologische en pathologische rol van galzouten en
de enterohepatische circulatie. In onze klinische studie beschrijven we de vroege verhoging
van serum gamma-GT (GGT) bij de ontwikkeling van de ernstige cirrotische vorm van Cystic
fibrosis leverziekte (CCFLD). Deze resultaten ondersteunen een rol voor het meten en
vervolgen van GGT verhoging als een voorspeller voor de ontwikkeling van CCFLD. Uit onze
experimenten in verschillende CF muismodellen kunnen we concluderen dat Cystic fibrosis
gerelateerde leverziekte niet is geassocieerd met een veranderde cytotoxiciteit van biliaire
galzoutsamenstelling. Toch konden we, op basis van experimenten in CF muismodellen, wel
een Cftr specifiek, biliair fenotype vaststellen. We tonen aan dat dit biliaire fenotype,
gedeeltelijk, kan worden gecorrigeerd door het toedienen van het hydrofiele galzouten
UDCA. Tot slot hebben we aangetoond dat het CF fenotype bij muizen wordt gekenmerkt
door specifiek veranderingen in het metabolisme en de enterohepatische circulatie van
galzouten. Deze variaties worden veroorzaakt door de interactie van het Cystic fibrosis
transmembrane conductance regulator (CFTR) eiwit en de intestinale bacteriële flora. Onze
bevindingen bieden nieuwe experimenteel bewijs voor de belangrijke rol van CFTR in het
equilibrium van de enterohepatische kringloop van galzouten. Een verminderde of afwezig
CFTR functie, zoals bij de ziekte Cystic fibrosis, kan dan daardoor aanleiding geven
tot pathologische veranderingen in de enterohepatische circulatie en het galzouten
metabolisme. Onze resultaten geven nieuwe inzichten in belangrijke de gastro-intestinale en
hepatologisch rol van CFTR. Deze nieuwe inzichten bieden een perspectief voor het verdere
begrip en behandeling van de ziekte Cystic fibrosis.
CCFLD is een levensbedreigende complicatie van CF (1, 2). De kliniek van CCFLD wordt vaak
gekenmerkt door ernstige portale hypertensie en hoge kans op gastro-intesintale
varicesbloedingen (3). In sommige patiënten met CCFLD is levertransplantatie als therapie
noodzakelijk (4, 5). De piekleeftijd voor de presentatie van CCFLD ligt rond de leeftijd van 10
jaar. CCFLD kent een “stille” preklinische fase waarin de litteken vorming (fibrose) in de lever
zich geleidelijk, in een periode van jaren, ontwikkelt tot cirrose. Patiënten met een verhoogd
risico voor CCFLD of patiënten in een vroege fase van de ziekte zouden daarom theoretische
kunnen profiteren van preventieve behandeling. Echter op dit moment zijn er geen testen
beschikbaar om patiënten met een risico voor CCFLD te identificeren. In hoofdstuk 2,
beschrijven wij dat een toename in het serum GGT, vaak nog binnen de grenzen van normaal,
een indicator kan zijn voor de presentatie van CCFLD binnen 2 jaar. Deze bevingen biedt de
Nederlandse discussie en samenvatting
147
8
mogelijkheid om risico patiënten voor de ontwikkeling van CCFLD te identificeren en bij hen
preventieve behandeling, bijvoorbeeld met ursodeoxy cholzuur (UDCA) te overwegen.
Binnen een paar jaar nadat de het CFTR gen bij de mens was geïdentificeerd werden de eerste
genetische gemodificeerde muis modellen ontwikkeld. (6-9). Deze experimentele
diermodellen zijn van groot belang geweest bij de totstandkoming van het groeiende inzicht
in de fysiologie en pathologie van het functioneren van CFTR in verschillende orgaan
systemen. Echter het CF fenotype van de muis modellen verschillen op een aantal kenmerken
met de humane uitingsvormen van Cystic fibrosis. Zo wordt bijvoorbeeld longziekte of
chronische longschade, de belangrijkste ziekte uiting van CF bij de mens, bij muizen
nauwelijks gezien (10). Echter CF muis modellen presenteren wel verschillende gastro-
intestinale fenotypes die goed overeenkomen met het humane gastro-intestinale
ziektebeelden van CF (11). In hoofdstuk 3, geven wij een overzicht van de kennis in de
literatuur op het gebied van hepatologische en gastro-intestinale fenotypes van CF muis
modellen in vergelijking met het humane ziekte beeld van CF. We hebben ons hierbij
geconcentreerd op klinisch beeld van de persisterende intestinale vetmalabsorptie dat bij de
veel CF patiënten aanwezig is.
Galzouten spelen als emulgatoren een zeer belangrijke bij de intestinale vetabsorptie.
Opvallend hierbij is dat bij zowel CF patiënten als CF muizen op vergelijkbare manier sprake is
van een toegenomen fecaal verlies van galzouten en veranderingen in het
galzoutmetabolisme (12). Deze CF gerelateerde veranderingen op het gebied van
galzoutmetabolisme en galzouttransport zouden goed gerelateerd kunnen zijn aan de
beschreven persisterende vetmalabsorptie in CF. Andere gastro-intestinale fenotypische
overeenkomsten tussen CF patiënten en muismodellen zijn bijvoorbeeld het voorkomen van
intestinale bacteriële overgroei (“small intestinal bacterial overgrowth” of SIBO), intestinale
inflammatie en een toegenomen intestinale mucosale permeabiliteit (13, 14). Uit onze studie
concluderen we dan ook dat CF muizen een geschikt basaal experimenteel model kunnen
vormen voor studies naar gastro-intestinale CFTR gerelateerde, fenotypes.
Wij hebben onderzocht in welke mate CF muismodellen kunnen bijdragen aan het begrip over
de pathofysiologische oorzaak van CF gerelateerde leverziekte. (hoofdstuk 4-6). Ons
hoofddoel hierbij was om vast te stellen in welke mate de cytotoxiciteit van de biliaire
galzouten oorzakelijk betrokken is bij het ontstaan van CFLD. Om deze vraag te beantwoorden
onderzochten wij de pathologie en functioneren van de lever in verschillende type CF
muismodellen. Bij één van deze muis modellen (Cftr-/-tm1Unc)
, dat zich kenmerkt door de
spontane ontwikkeling van een op humaan CFLD lijkende leverziekte, konden wij geen relatie
vaststellen tussen de biliaire galzouttoxiciteit en het beschreven patho-histologisch beeld van
de lever. In andere CF muismodellen konden wij een wel significante toename vaststellen van
Chapter 8
148
8
het hydrofobe profiel van de biliaire gal samenstelling. Echter ook in deze muismodellen
konden wij deze toegenomen cytotoxiciteit niet relateren aan patho-histolgische
veranderingen in de lever. Ten slotte konden wij niet vaststellen dat, CF muismodellen, ook
niet na een langdurige blootstelling aan hydrofobe galzouten, meer gevoelig waren om een
vorm van CF gerelateerde leverziekte te ontwikkelen. Gebaseerd op deze resultaten
concluderen wij dat de ontwikkeling van CFLD, in ieder geval in muizen, niet gerelateerd is
aan CFTR gerelateerde veranderingen in de biliaire galzoutcytotoxiciteit. We veronderstellen
dat deze conclusie ook doorgetrokken kan worden naar de humane situatie en de
ontwikkeling van CFLD bij patiënten.
Hoewel wij niet de een direct overeenkomstig lever fenotype tussen mens en muis hebben
kunnen vaststellen, hebben wij wel een uniek CFTR afhankelijk hepatisch en biliair fenotype in
CF muizen kunnen aantonen (hoofdstuk 4). Wij vonden dat als Cftr-/-
muizen gedurende een
langere periode worden blootgesteld aan het hydrofobe galzout cholaat, zij een duidelijk
toegenomen hydrofobiciteit van de gal en een verminderd vermogen om de galproductie te
verhogen ontwikkelen. Hier uit blijkt dat Cftr-/-
muizen, in tegenstelling tot wild type muizen,
veel minder goed in staat zijn om de gal te verdunnen tijdens de blootstelling aan cholaat. Dit
blijkt ook duidelijk uit te toegenomen totale biliaire galzoutconcentratie bij basaal gelijk
blijvende galzoutsecretiesnelheden. Verder vonden wij een specifiek, Cftr gerelateerd, groei
effect van cholaat toediening op de levergroei en leverproliferatie. Hierbij zagen wij dat bij
wild type muizen de het levergewicht globaal 50% toenam tijdens de langdurige toediening
van cholaat. Dit in tegenstelling tot de verschillende type CF muizen waarbij, na langdurige
cholaat toediening het levergewicht gelijk bleef. Deze bevinding wijzen op, een tot nu toe
onbekende, rol van CFTR in the normale adaptieve leverrespons op langdurige blootstelling
aan hydrofobe galzouten. Wij veronderstellen dat de verminderde regeneratieve capaciteit
van het CF fenotype en rol kan spelen bij de ontwikkeling van CFLD als gevolg van een
toegenomen kwetsbaarheid van de lever in cholestatische condities.
Onze resultaten van de studies in de Cftr-/-tm1Unc
muizen (hoofdstuk 5) laten duidelijke, Cftr
afhankelijke verschillen zien in het intestinale galzoutmetabolisme. Wij vonden, onder
andere, Cftr specifieke, veranderingen in bacteriële vorming van secundaire galzouten en de
daarbij passende veranderingen in de biliaire en fecale galzoutsamenstelling. Bijvoorbeeld
zagen wij dat in Cftr-/-tm1Unc
muizen dat de biliaire galzoutsamenstelling veel hydrofieler was en
de galproductie veel hoger was in vergelijking met wild type muizen. Deze toename in
galproductie bij de CF muizen kon worden verklaard uit de aanwezigheid van verhoogde
concentraties van het zeer hydrofiele galzout ursocholaat (UCA). Normaal gesproken is de
proportie van UCA in de totale galzoutpool erg klein als gevolg van de zeer snelle omzetting
naar deoxycholaat (DCA) door de intestinale bacteriële flora. De, door ons gevonden, hoge
concentratie van UCA kan dan ook alleen ontstaan als gevolg van Cftr gerelateerde
Nederlandse discussie en samenvatting
149
8
veranderingen in het functioneren van de intestinale bacteriële flora. De hoge concentratie
en hoge hydrofobiciteit van UCA zijn de aanleiding voor de gevonden sterke toename in
galproductie en de lage biliaire hydrofobiciteit bij de Cftr-/-tm1Unc
muizen. Onze resultaten
vormen een duidelijk voorbeeld van de manier waarop, door de interactie tussen intestinale
bacteriële flora en het galzoutmetabolisme, de gehele enterohepatische kringloop van
galzouten kan worden beïnvloed. In de situatie van CF kunnen de veranderingen in het
galzoutmetabolisme gekoppeld zijn aan het induceren van leverziekte of misschien wel het
voorkomen hiervan. Bij CF patiënten en CF muismodellen worden diverse intestinale
mucosale veranderingen beschreven die van invloed zijn op de samenstelling van de
intestinale bacteriële flora. Hierbij kan bijvoorbeeld gedacht worden aan de beschreven
verandering in de accumulatie van viskeus en taai mucus secreet, SIBO, toegenomen
intestinale permeabiliteit en de aanwezigheid van intestinale inflammatie in CF condities(15-
24).
Het grote belang van de intestinale flora op het CF fenotype werd verder nog aangetoond in
één van onze andere experimenten. In deze studie behandelden wij CF muizen en wild type
muizen met of het hydrofiele galzout UDCA of het hydrofobe galzout cholaat. We vonden in
de wild type muizen met de cholaat behandeling er sprake was van een iets toegenomen
hydrofobiciteit in vergelijking met de Cftr-/-
muizen. Deze toename kon in zijn geheel worden
verklaard door de toegenomen aanwezigheid van het hydrofobe secundaire galzout
deoxycholaat (DCA), wat niet werd aangetoond in Cftr-/-
muizen. DCA wordt uitsluitend
geproduceerd door de intestinale bacteriële flora. Recent hebben wij en andere
gerapporteerd over de veranderde intestinale bacteriële flora en intestinale permeabiliteit in
CF condities (25-28). Onze resultaten onderstrepen de grote invloed van intestinale bacteriële
flora op de totale enteroheaptische kringloop in het bijzonder in situatie van CF. Wij zijn van
mening dat CF genotype specifieke veranderingen in de intestinale bacteriële flora primair
verantwoordelijk zijn voor meerderheid van de waargenomen variatie in de samenstelling van
de galzoutpool en de galproductie bij CF muizen. Onze resultaten bieden de mogelijkheden
om de functionele relevantie van de waarnemingen verder te beoordelen met behulp van
interventies gericht op de intestinale bacteriële flora in CF muismodellen en CF patiënten.
Zoals hiervoor aangegeven kunnen, Cftr afhankelijke, veranderingen in de galzoutpool
samenstelling gevolgen hebben op de enterohepatische circulatie in CF condities. Echter, tot
op zeker hoogte, kunnen de veranderingen in galzoutformatie ook het gevolg zijn van CFTR
afhankelijke veranderingen in de enterohepatische circulatie zelf. De gevonden
veranderingen in bacteriële biotransformatie van primaire naar secundaire galzouten, in het
bijzonder van cholaat naar deoxycholaat, kunnen indirect de galzoutsynthese beïnvloeden.
DCA is bijvoorbeeld een zeer krachtige activerende ligand van de farnesoid X receptor (FXR)
(29). Deze nucleaire receptor wordt in een hoge mate tot expressie gebracht in de lever en
Chapter 8
150
8
het ileum (30). In de lever speelt FXR een belangrijke rol bij de activatie van de
galzoutsynthese uit cholesterol(31). Activatie van intestinaal FXR leid tot expressie van
fibroblast growth factor 19 (FGF19) (32). FGF15/19 kan vervolgens vanuit de darm, via het
portale bloed worden getransporteerd naar de lever. Recente publicatie beschrijven de
centrale rol van Fgf15 (Fgf15 is het muis equivalent van het humane FGF19) bij hepatocyten
proliferatie en de regulatie van de systemische galzoutcompositie (33, 34). FGF15/19 worden
alleen tot expressie gebracht in de darm en normale gesproken niet in de lever. Echter
expressie van FGF15/19 in de lever kan voorkomen in cholestatische condities en is dan
geassocieerd met de hepatische adaptatie gericht op de bescherming tegen galzouttoxiciteit
(35).
In ons chronisch hydrofoob galzoutbelastingsmodel zou expressie van Fgf15 in zowel darm als
lever een rol kunnen spelen bij het toegenomen van lever gewicht in wild type muizen. De
verhoogde biliaire fractie van DCA, zoals gevonden in de wild type muizen, na langdurige
cholaat blootstelling, kan zo expressie van Fgf15, via intestinaal FXR kunnen induceren. Het
zou ook kunnen dat DCA direct de expressie van Fgf15 in de lever induceert. In beide situatie
kan Fgf15 expressie aanleiding geven tot hepatocyten proliferatie en levergroei (36). Er zou
hierbij ook nog een essentiële rol kunnen zijn voor de intestinale bacteriële flora. Een recente
studie in bacterie vrije (“germ free”) muismodel toonde aan darm bacteriën mede het
galzoutmetabolisme reguleren door beïnvloeding het natuurlijk voorkomende FXR
antagonistsen zoals tauro beta-muricholaat (37).
Cholaat gestimuleerde levergroei in muizen, vergelijkbaar met onze resultaten bij wild type
muizen, is eerder gerapporteerd. Huang et al. beschreven dat voeren van cholaat (0,2%)
gedurende 5 dagen, bij wild type muizen, de normale levergroei stimuleert en het relatieve
levergewicht verhoogd met gemiddeld 30% (38). Deze onderzoeker zagen verder dat cholaat
supplementatie leverregeneratie versnelde naar een partiële hepatectomie. Zij concluderen
hieruit dat de normale lever regeneratie afhankelijk is van galzoutstimulatie via nucleaire
receptoren zoals een aan FXR gekoppeld feedback mechanisme. De cruciale rol van FXR wordt
nog eens bevestigd door het feit dat langdurige blootstelling aan UDCA niet leidt tot
levergroei in wild type muizen. Dit is zeer waarschijnlijk gerelateerd aan het feit dat UDCA, in
tegenstelling tot DCA, niet functioneert als een ligand voor FXR. Onze huidige bevindingen
openen voor ons mogelijkheden voor toekomstig onderzoek. Om onderscheid te kunnen
maken tussen het rol van galzouten versus intestinale bacteriële flora zou het zeer waarde vol
kunnen zijn om een “germ free” CF muismodel te ontwikkelen en hierbij onze huidige
experimenten te herhalen. Tevens zou het zeer interessant kunnen zijn cholaat
blootstellingsexperimenten te herhalen in lever en darm specifiek FXR knockout muis
modellen. Deze modellen zouden de mogelijkheid bieden om de onderscheid te maken
tussen de rol van intestinaal en lever FXR bij de stimulatie van levergroei.
Nederlandse discussie en samenvatting
151
8
De behandeling van CFLD patiënten door middel van UDCA is een controversieel punt (39).
Een belangrijk aspect hierbij is de het ontbreken van basaal experimenteel bewijs over de
mogelijke werking van UDCA bij CFLD. Tevens ontbreken er goed ontworpen prospectieve
studies die onomstotelijk een gunstige lange termijn klinisch effect bewijzen en onderbouwen
(40). Een belangrijkste fysiologische eigenschappen van UDCA is dat de galproductie kan
verhogen. Deze eigenschap vormt de basis voor het therapeutisch gebruik van UDCA bij
diverse cholangiopathien, zo ook CFLD (41). In vitro en ex vivo studies hebben echter laten
zien dat het choleretisch effect van UDCA afhankelijk is van de aanwezigheid en de functie
van CFTR (42, 43). In onze huidige studie (hoofdstuk 6) onderzochten wij het effect van UDCA,
in vivo, in een CF (Cftr-/-
) muismodel. Wij toonden hierbij aan dat, in vivo, zowel bij acute als
bij langdurige toediening van UDCA er sprake is van een significante, Cftr onafhankelijke,
toename van de galproductie. Hiermee laten onze resultaten zien dat, ook in CF condities,
bijvoorbeeld bij patiënten, er ten minste een positieve choleretisch effect van UCA kan
worden verwacht. Verder zagen wij dat langdurige UDCA ook de hydrofobiciteit van de
galsamenstelling in CF muizen verlaagt. UDCA neemt hierbij de plaats in van cholaat en tevens
verlaagt UDCA de kwantitatieve grootte van de cholaat galzoutpool. Deze additionele
eigenschappen zouden nog verder kunnen bijdragen aan het vermeende positieve effect van
UDCA in de behandeling van CFLD. Wij willen echter onderstrepen dat onze resultaten geen
onderbouwing leveren voor eventuele gunstige effecten op de voorkoming van lever fibrose
of andere lange termijn gunstige effecten van UDCA in CF condities.
Wij zijn van mening dat de resultaten van onze studies leiden tot een verbeterd begrip van de
ontwikkeling van CF gerelateerde leverziekte. Onze studies bieden nieuwe mogelijkheden
voor toekomstig klinische en basaal experimenteel onderzoek. Dit positieve vooruitzicht met
betrekking tot mogelijkheden voor onderzoek is juist nu essentieel, in het licht van de huidige
baanbrekende ontwikkelingen op het gebied van de behandeling van CF (44). Om het effect
van deze nieuwe therapieën te optimaliseren is het theoretisch te verkiezen om de
behandeling al te starten bij jonge kinderen, nog zonder tekenen van chronische longschade
(45). Echter klassieke outcome parameters voor effect meting bij CF, zoals longfunctie
metingen, zullen echter weinig richtinggevend zijn bij patiënten nog zonder significante
longziekte. Om in de toekomst toch relevante klinische effectiviteitsstudie te kunnen
uitvoeren moeten er nieuwe klinische resultaten metingen ontwikkeld worden (46). De
gevonden en beschreven resultaten in dit proefschrift bieden aanvullende mogelijkheden om
nieuwe outcome gastro-intestinale en hepatologische outcome parameters te ontwikkelen
welke al toepasbaar zijn vanaf jonge leeftijd (47).
Zo hebben bijvoorbeeld het meten van de fecale galzoutexcretie en het meten van
veranderingen in het galzoutmetabolisme duidelijke potentie om te kunnen functioneren als
potentiële resultaat variabelen. De totale fecale galzoutsecretie en verandering in de
Chapter 8
152
8
galzoutcompositie zijn relatief eenvoudig te meten. Het zijn weinig invasieve onderzoeken,
die potentieel goed uitvoerbaar en haalbaar zijn, ook bij jonge kinderen. Verder bieden ook
de steeds maar in aantal toenemende onderzoeksmethodes voor het kwantificeren en
kwalificeren van de intestinale bacteriële flora grote potentie als toekomstige
onderzoeksmethodes. De door ons beschreven CFTR specifieke verandering met betrekking
tot de veranderde interactie tussen van de intestinale bacteriële flora met het
galzoutmetabolisme kunnen goed dienen als basis voor verder onderzoek in deze richting.
De resultaten zoals beschreven in dit proefschrift bieden nieuwe mogelijkheden voor de
behandeling van de cirrotische vorm van CFLD (CCFLD). Gebaseerd op onze resultaten is goed
te verantwoorden om, de prognostische waarde van het serum GGT als voorspeller van de
ontwikkeling van CCFLD, te onderzoeken in een groot prospectief cohort. Een dergelijke
studie bied de mogelijkheid om preventieve therapeutische interventies te evalueren op
effectiviteit. Gelijktijdig kunnen observationele (registratie) studie de mogelijkheden bieden
om nieuwe potentiele gasto-intestinale en hepatologische outcome resultaten te evalueren
en te valideren.
In aanvulling op klinische georiënteerde studies is het van het grootste belang is om ook
basaal experimenteel onderzoek in relevante experimentele modellen te continueren. Met
betrekking tot de ontwikkeling en de pathogenesis van CFLD zijn wij van mening dat de focus
van onderzoek moet worden verlegt van galzoutcytotoxiciteit naar de lever/intestinale
bacteriële flora interactie. De resultaten beschreven in dit proefschrift bieden een duidelijke
onderbouwing voor de relatie tussen de CFTR afhankelijke verandering in de intestinale
bacteriele flora, galzout metabolisme en daarmee ook potentieel met intestinale inflammatie.
Hiermee lijkt het goed te rechtvaardigen om de focus van onderzoek te richten op intestinale
bacteriële flora en bacteriële overgroei in CF condities. Hierbij zouden bijvoorbeeld de
intestinale panethcellen, betrokken van de secretie van anti-bacteriel eiwitten in het
intestinale lumen, ook bij CF een duidelijke rol kunnen spelen.
Gebaseerd op onze bevindingen zou een andere logische onderzoekslijn de intestinale-lever
as, in de regulatie van de enterohepatische kringloop en synthese van galzouten, in CF
condities zijn. Onze resultaten tonen een duidelijk CFTR afhankelijk fenotype van de
galzoutsynthese, de intestinale galzouttransformatie en het fecale galzoutverlies zien.
Tenslotte zijn er de mogelijkheden met de recent nieuw ontwikkelde CF diermodellen. Recent
werden het CF varken en de CF fretten beschreven als experimentele diermodellen voor
Cystic fibrosis. (48, 49). De eerste resultaten van deze studies, in het bijzonder bij het CF
varken, laten een duidelijk CFLD achtige fenotype zien(50). Deze bevindingen bieden
duidelijke mogelijkheden om nieuwe studies te doen naar de ontwikkeling van CFLD.
Nederlandse discussie en samenvatting
153
8
Concluderen kunnen wij stellen dat, onderzoeksresultaten in dit proefschrift, aanvullende
kennis bieden op het gebied van de gastro-intestinale en hepatologische aspecten van Cystic
fibrosis. Wij zijn van mening dat deze resultaten, in combinatie met de huidige ontwikkeling
van nieuwe curatieve therapieën voor CF, het gebruik van CF patiënten registratie data en de
ontwikkeling van de nieuwe diermodellen zullen leiden tot een toegenomen begrip van de
pathofysiologische mechanismen van CF en de behandeling van CF patiënten.
Chapter 8
154
8
REFERENTIELIJST
1. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory
Reviews. 2013.
2. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.
Hepatology. 1999 11;30(5):1151-8.
3. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic
fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.
1999;31(1):77-83.
4. Chryssostalis A, Hubert D, Coste J, Kanaan R, Burgel P, Desmazes-Dufeu N, et al. Liver
disease in adult patients with cystic fibrosis: A frequent and independent prognostic factor
associated with death or lung transplantation. J Hepatol. 2011;55(6):1377-82.
5. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in
cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.
6. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.
Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989
09/08;245(4922):1073-80.
7. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of
the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.
1989;245(4922):1066.
8. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of
a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.
9. van Doorninck JH, French PJ, Verbeek E, Peters R, Morreau H, Bijman J, et al. A mouse model
for the cystic fibrosis delta F508 mutation. EMBO J. 1995;14(18):4403.
10. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse
models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.
2011;10:S152-71.
11. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic
fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J
Pathol. 2004 04;164(4):1481-93.
12. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.
Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption
in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.
Nederlandse discussie en samenvatting
155
8
13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial
overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr. 2007;44(2):212-8.
14. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun.
2004;72(10):6040-9.
15. Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal
permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.
Pediatr Res. 1996;39(6):985-91.
16. Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic
fibrosis: Importance of breath testing. J Paediatr Child Health. 1998;34(1):79-82.
17. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in
cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.
Acta Biochim Pol. 2009;56(4):631.
18. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal
inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.
Aliment Pharmacol Ther. 2004;20(7):813-9.
19. Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic
inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res.
2000;47(3):344-50.
20. Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the
small bowel of patients with cystic fibrosis. Pediatr Res. 1998;43(2):234-9.
21. Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic
fibrosis. Arch Dis Child. 2000;82(5):394-9.
22. Quigley EM, Quera R. Small intestinal bacterial overgrowth: Roles of antibiotics, prebiotics,
and probiotics. Gastroenterology. 2006;130(2):S78-90.
23. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis
transmembrane conductance regulator null mouse small intestine. Infect Immun.
2004;72(10):6040.
24. De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus
accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small
intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology.
2007;293(3):G577-84.
Chapter 8
156
8
25. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in
the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.
2006;42(1):46-52.
26. De Lisle RC, Mueller R, Boyd M. Impaired mucosal barrier function in the small intestine of
the cystic fibrosis mouse. J Pediatr Gastroenterol Nutr. 2011;53(4):371-9.
27. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade
HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.
2011;71(1):4-12.
28. Martin CR, Zaman MM, Ketwaroo GA, Bhutta AQ, Coronel E, Popov Y, et al. CFTR
dysfunction predisposes to fibrotic liver disease in a murine model. American Journal of
Physiology-Gastrointestinal and Liver Physiology. 2012;303(4):G474-81.
29. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for
the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.
30. Vaquero J, Monte MJ, Dominguez M, Muntané J, Marin JJG. Differential activation of the
human farnesoid X receptor depends on the pattern of expressed isoforms and the bile acid
pool composition. Biochem Pharmacol. 2013 10/1;86(7):926-39.
31. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory
cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol
Cell. 2000;6(3):517-26.
32. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, et al. Differential regulation of bile acid
homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res. 2007
12;48(12):2664-72.
33. Wu X, Ge H, Lemon B, Vonderfecht S, Weiszmann J, Hecht R, et al. FGF19-induced
hepatocyte proliferation is mediated through FGFR4 activation. J Biol Chem.
2010;285(8):5165-70.
34. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, et al. FGF19
regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and
independent pathways. PloS one. 2011;6(3):e17868.
35. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM. High expression of the bile
salt‐homeostatic hormone fibroblast growth factor 19 in the liver of patients with extrahepatic
cholestasis. Hepatology. 2009;49(4):1228-35.
Nederlandse discussie en samenvatting
157
8
36. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, et al.
Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its
application in the prevention of post-resection liver failure in mice. Gut. 2013;62(6):899-910.
37. Sayin S, Wahlström A, Felin J, Jäntti S, Marschall H, Bamberg K, et al. Gut microbiota
regulates bile acid metabolism by reducing the levels of tauro-beta-muricholic acid, a naturally
occurring FXR antagonist. Cell Metabolism. 2013 2/5;17(2):225-35.
38. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent
bile acid signaling is required for normal liver regeneration. Science. 2006
04/14;312(5771):233-6.
39. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-
associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.
40. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.
Cochrane Database Syst Rev. 2012;10.
41. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of
action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.
42. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic
acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.
Gastroenterology. 2007 11;133(5):1603-13.
43. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic
calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995
09;109(3):965-72.
44. Ramsey BW, Davies J, McElvaney NG, Tullis E, Bell SC, Dřevínek P, et al. A CFTR potentiator
in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011 11/03;
2012/09;365(18):1663-72.
45. Ashlock MA, Olson ER. Therapeutics development for cystic fibrosis: A successful model for
a multisystem genetic disease. Annu Rev Med. 2011;62:107-25.
46. De Boeck K, Bulteel V, Tiddens H, Wagner T, Fajac I, Conway S, et al. Guideline on the
design and conduct of cystic fibrosis clinical trials: The european cystic fibrosis Society–Clinical
trials network (ECFS-CTN). Journal of Cystic Fibrosis. 2011;10:S67-74.
47. Rowe SM, Borowitz DS, Burns JL, Clancy JP, Donaldson SH, Retsch-Bogart G, et al. Progress
in cystic fibrosis and the CF therapeutics development network. Thorax. 2012;67(10):882-90.
Chapter 8
158
8
48. Rogers CS, Stoltz DA, Meyerholz DK, Ostedgaard LS, Rokhlina T, Taft PJ, et al. Disruption of
the CFTR gene produces a model of cystic fibrosis in newborn pigs. Science.
2008;321(5897):1837-41.
49. Sun X, Sui H, Fisher JT, Yan Z, Liu X, Cho H, et al. Disease phenotype of a ferret CFTR-
knockout model of cystic fibrosis. J Clin Invest. 2010;120(9):3149.
50. Stoltz DA, Rokhlina T, Ernst SE, Pezzulo AA, Ostedgaard LS, Karp PH, et al. Intestinal CFTR
expression alleviates meconium ileus in cystic fibrosis pigs. J Clin Invest. 2013;123(6):2685.
Curriculum vitae
159
CURRICULUM VITAE
Franciscus Albertus Johannes Andreas Bodewes was born on February 14th, 1966 in Eelde in
the Netherlands. He was third in a family of 3 children, with an older brother and sister. He
grew up and went to the Maria primary school in Eelde. After graduating from the
Gemeentelijke MAVO in Eelde, he finished his secondary school period at the Sint
Maartenscollege in Haren.
He went on to study Medicine at the University of Groningen. In his second year he was
offered the opportunity to join the Groningen liver transplant program student team headed
by pioneer transplant surgeon Professor Maarten Slooff.
After he had finished his Medicine study he started to work as a pediatric resident in the
Medical center in Leeuwarden and later on in the Sophia hospital in Zwolle. In 1996 he started
the Pediatrics residency program in University Hospital Groningen lead by, respectively,
Professor Jan Kimpen and Pieter Sauer. As part of this training he worked for two years in the
St Elisabeth Hospital, on the island of Curaçao in the Dutch Antilles. During his residency years
he worked for 3 years as a medical volunteer in holyday camps for Cystic fibrosis patients.
Here he learned about the severity and impacted of Cystic fibrosis diseaseon patients and
their families. After he had become a pediatrician he started his Pediatric gastroenterology,
hepatology and nutrition training in the Beatrix Children’s hospital of the University of
Groningen in de group of Charles Bijleveld.
Since he finished his pediatric GI training program in 2005 he is working as a senior staff
member/consultant in the department of Pediatric gastroenterology and hepatology of the
University medical center Groningen. His clinical focus is gastroenterology, hepatology, organ
transplantation and Cystic fibrosis. His main research topics concern the role of the CFTR
protein in the physiology and patho-physiology of bile formation and the development of
Cystic fibrosis related liver disease as well as the role of the CFTR protein in the physiology
and pathophysiology of intestinal fat absorption in CF patients and experimental animals. He
started this research project, concerning the role of bile salt in the development of Cystic
fibrosis liver disease, in the research group of Professor Henkjan Verkade.
Frank Bodewes married to Els van den Berg in 1997. She is program director for the 400th
anniversary of the University of Groningen. Together they have 3 children: Jelle, Silke and
Douwe.
Dankwoord/acknowledgments
161
DANKWOORD/ACKNOWLEDGMENTS
De totstandkoming van mijn promotieonderzoek heeft zich uitgestrekt over meerdere jaren.
In deze periode hebben vele mensen een bijdrage geleverd aan de studies, aan het schrijven
van de manuscripten en aan alle andere taken die komen kijken bij een promotie. Ik wil
hierbij alle mensen, die mij hebben gesteund en geholpen, van harte bedanken. Een aantal
mensen wil ik persoonlijk noemen vanwege de bijzondere bijdrage die zij hebben geleverd.
Henkjan, Prof. H.J. Verkade. Wat een weg hebben we samen afgelegd! De eerste keer dat ik
je ontmoette was, denk ik, 20 jaar geleden. Ik was student en deed een onderzoeksproject bij
Klary Niezen op Bloemsingel 1 en jij werkte daar op het lab als promovendus. De echte basis
van onze samenwerking werd gelegd in onze opleidingstijd in Groningen. Jij hoorde, toen ik in
1997 begon, al bij de meer senior arts assistenten. Je leerde mij dat je je naast het gewone
werk ook breder moest ontwikkelen en profileren. Als onderzoeker was ik je in Zwolle, waar ik
werkte als agnio, ook al tegengekomen. Je deed daar, samen met Edmond, onderzoek naar de
vetuitscheiding van premature neonaten. Je maakte grote indruk op mij met je verhaal dat je
complete luiers met babypoep oploste in chloroform/methanol om zo de fecale lipiden in
oplossing te brengen. Toen ik, in 2001, begon als staflid bij de kindergeneeskunde wilde ik mij
graag verder gaan specialiseren tot “Kinderarts-gastroenteroloog”. Daarvoor was het doen
van onderzoek een pre. Ik had interesse in Cystic fibrosis en jij wilde graag je lopende Cystic
fibrosis onderzoekslijn van Mini Kalivianakis continueren. Onze onderzoekssamenwerking was
geboren. Ik denk niet dat je toen wist wat je je op de hals haalde. Ik had geen
onderzoekservaring, zeker niet met basaal wetenschappelijk onderzoek en proefdieren.
Daarnaast ben ik een klein beetje dyslectisch. Eerst was het gewoon “onderzoek doen”, later
toen we wat langer bezig waren werd het toch steeds meer een onderzoekslijn en uiteindelijk
een promotietraject. Ik weet wel zeker dat ik jouw langstlopende promotietraject ben
geweest. Henkjan, ik ben je echt dankbaar. Schijnbaar ben je nooit het vertrouwen in mij en
mijn onderzoeksactiviteiten kwijt geraakt. Je behield altijd je positieve houding en bleef mij
voorhouden dat het eindpunt haalbaar was. In de loop van de jaren is onze band als collega’s
steeds hechter geworden. We kennen elkaars sterke punten en elkaars gebruiksaanwijzingen.
Maar het meest belangrijke is dat we respect voor elkaar hebben en eerlijk tegen elkaar
durven zijn. Ik denk met plezier terug aan de vele gezellige en leuke momenten tijdens de
buitenlandse congressen en andere leuke reisjes die we samen over de wereld hebben
gemaakt. De vorm en inhoud van onze samenwerking is in de loop der jaren steeds mee
veranderd met onze veranderende omstandigheden en verantwoordelijkheden. Het
vertrouwen is gebleven.
Hugo en Marcel. Het is in grote mate aan de samenwerking met jullie te danken dat we het
merendeel van de onderzoeken dit van dit proefschrift hebben kunnen uitvoeren. Dankzij
162
jullie wereldwijd befaamde goedlopende onderzoeksprogramma met de CF muismodellen en
jullie bijzondere kennis op dit vakgebied werden onze plannen mogelijk. Hugo volgens mij was
ik 4 jaar geleden al op jouw afscheidssymposium maar “you’re still going strong” met volgens
mij 10 keer meer energie als voor je “pensioen”. Dank voor je zeer deskundige, zeer
inhoudelijke en gedetailleerde commentaren, en dank dat je mijn copromotor wilt zijn.
Marcel bedankt voor de samenwerking, je adviezen en het meedenken. Ik waardeer je
nauwkeurigheid en je doorzettingsvermogen. Ik hoop dat we ook in de komende jaren nog
meer gezamenlijk projecten kunnen uitvoeren.
Willemien, Mariëtte en Marjan. Alle drie kwamen jullie als nieuwe jonge student-
onderzoekers een tijd meelopen op het CF project. Soms was het een beetje onduidelijk
situatie voor jullie als je “bij-begeleider” zelf nog met zijn promotieonderzoek bezig is. Alle
drie hebben jullie mij zeer geholpen met mijn onderzoek. Ieder voor zich hebben jullie een
groot stuk van de puzzel in dit CF onderzoeksproject gelegd. Het waren juist deze essentiële
stukken die maakten dat voor mij het eindpunt van promotie zichtbaar en haalbaar bleef.
Jullie zijn alle drie doorgegaan met je jullie eigen projecten. Allemaal zijn jullie al bij Henkjan
gepromoveerd met prachtige proefschriften. Jullie waren voor mij voorbeelden en
motivatoren om door te gaan.
Peter, Prof. Durie., You’re an Inspirator. I admire the way you make difficult thing sound easy
and the way you provide feedback without a single unkind word. I love your humor and your
generous smile. You’re the global pioneer and expert in Cystic fibrosis gastro-intestinal and
hepatic disease. For me it was an honor to have collaborated with you in our research project.
Thank you and get well soon.
Satti. Actually we met only once, at the NACFC meeting in Anaheim. I appreciate all the work
that you did for our collaborative “Toronto” research project. Thank you for your help and
effort.
Juul. Dank voor de zeer goede samenwerking en het vele werk dat je voor mij hebt gedaan.
Altijd was je bereid mijn vragen te beantwoorden of om mij te helpen. Ik bewonder je
positieve levenshouding en de manier waarop je andere mensen altijd met respect
behandeld.
Annette, Prof. Gouw. Onze samenwerking in dit project was kort maar essentieel. Ik
bewonder de manier waarop jij op en zeer ontspannende manier zeer nauwkeurige en hoog
kwalitatieve beoordeling gaf van de histologie van de muizenlevers. Ik herinner mij nog goed
je uitspraak: “wat heb je met deze muizen gedaan, die lever zien er echt niet normaal uit
hoor!”
Dankwoord/acknowledgments
163
Hans en Marcella. Het is echt geweldig dat jullie nu met Henkjan en mij samenwerken om dit
CF project te continueren en verder te brengen. Met zoveel extra intellect en talent moet dit
wel goed komen.
Alle collega’s van de Kinder-MDL. Renée, Patrick, Nicolette, Hubert, Henkjan en Edmond,
Robert, Hester, Els, Ekkehard en Charles. Bedankt voor de tijd en ruimte die ik gekregen heb
om aan mijn onderzoek en promotie te kunnen werken. In het bijzonder wil ik Hester noemen
met wie ik, als kamer- en lotgenoot, altijd met veel lol en humor, alle verzuchtingen van het
onderzoek en promotie heb mogen delen. Dank voor de steun, het delen van jullie
persoonlijke ervaringen op het vlak van promoveren en de bemoedigende woorden. Ik ben er
trots op om in Groningen bij de KMDL te werken. Ik ben echt een gelukkig mens als ik naar
jullie illustere rijtje namen kijk en ik oprecht kan zeggen dat ik, vooral dankzij jullie, elke dag
met plezier naar mijn werk ging en ga.
Medewerkers van het laboratorium Kindergeneeskunde. In het bijzonder de
“oudgedienden” Renze, Vincent, Henk en Rik. Ik voelde mij vaak als een vis op het droge in
het lab. Het was echt niet mijn natuurlijke habitat. Hoezeer ik veel zaken inhoudelijk zeer
interessant vond, is het werken op het lab en het doen van experimenten andere koek. Jullie
zijn allemaal bijzondere getalenteerde professionals op jullie eigen terrein. Ik wil jullie van
harte bedanken voor jullie tijd en de belangrijke bijdrages die jullie hebben geleverd aan mijn
experimenten.
Onderzoekers van het laboratorium Kindergeneeskunde. Vele van jullie heb ik in de loop der
jaren ontmoet en velen van jullie hebben mij geholpen. Vaak met praktische dingen en zeer
goede tips. De sfeer van collegialiteit en behulpzaamheid van de onderzoekers op het lab is
echt bijzonder.
Leiding en stafleden Beatrix kinderziekenhuis en het UMCG. Velen van jullie hebben
interesse getoond in mijn onderzoeks- en promotieactiviteiten. Harma en Leonie, mijn oud
kamergenoten die mochten meegenieten van de meelwormen die uit muizenvoer op mijn
bureau vielen. Bart en Marc voor het delen van hun promotie ervaringen. De “club van
promoverende stafleden” die, onder leiding van Pieter Sauer, samen kwam in tapas
restaurant “Cervantes”. De meesten van deze club zijn nu klaar of volgen op korte termijn.
Toch een geweldig resultaat.
Bertjan en Bas. Bedankt dat jullie, zonder een moment van twijfel en zonder ook maar een
idee te hebben van de taak en inhoud van jullie functie, mijn paranimfen willen zijn.
Familie vrienden en kennissen. Dank voor jullie ondersteuning en in het bijzonder jullie
oprechte interesse. Het delen van mijn activiteiten en plannen met andere gaf altijd weer
motivatie (en een stok achter de deur) om verder te gaan.
164
Els en Jelle, Silke, Douwe. Zonder jullie geen proefschrift. Zo leuk als ik mijn werk vind zoveel
te leuker vind ik het om naar huis te gaan en bij jullie te zijn. Hoe dynamisch ons gezinsleven
en huishouden vaak ook zijn, er is voor mij geen fijnere plek als bij jullie. Lieve Els dank voor je
onvoorwaardelijke steun, motivatie en begrip voor mijn promotie project. Ik ben er trots op
dat ik mijn promotie nog net binnen de eerste 400 jaar van het bestaan van de
Rijksuniversiteit Groningen heb kunnen afronden. Verder ben ik blij dat jouw en niet mijn
project de naam “For infinity” heeft gekregen. Heel veel succes met het lustrum. Jongens,
papa heeft nu weer lekker veel tijd voor “huiswerkbegeleiding” ;-) en andere leuke dingen
natuurlijk.
Frank Bodewes