AN ABSTRACT OF THE DISSERTATION OF
Richard C. Scheri for the degree of Doctor of Philosophy in Toxicology presented on
May 30, 2008
Title: Chlordecone Pretreatment Promoted Subcellular Distribution of Scavenger
Receptor Class B Type II to Murine Hepatic Microsomes; Mass Spectrometry
Detected Hepatic Soluble Cholesterol Binding Proteins and Comparison of Protein
iTRAQ Ratios Using ESI QTOF and MALDI TOF/TOF
Abstract approved:
__________________________________________________________________
Lawrence R. Curtis
Chlordecone belongs to the class of persistent organochlorine pesticides that
are remarkably resistant to environmental degradation. Even though their use was
banned in the United States in 1978, these compounds can still be detected in both
humans and wildlife throughout the world. Previous work has shown that the
pretreatment of male C57BL/6 mice with low doses of the persistent organochlorine
(OC) pesticide, chlordecone (CD) stimulated biliary excretion of exogenous CH up to
3-fold, and further, that increased biliary excretion was not associated with changes in
ATP-binding cassette transporter G8 (ABCG8) or scavenger receptor class B type I
(SR-BI). In rodents, hepatic basolateral SR-BI is important in controlling plasma
lipoprotein levels and cholesterol (CH) homeostasis, with major roles in reverse CH
transport (RCT) and biliary excretion. The hepatic ABCG5/G8 heterodimer is a
membrane transporter present on the apical surfaces of hepatocytes, and also plays a
key role in biliary CH secretion. Scavenger receptor class B type II (SR-BII) was
identified as a splice variant from the SR-BI gene and is expressed in a variety of
tissues. Although the function of SR-BII is not clear it was proposed to play a role in
CH homeostasis and trafficking that was distinctly different than SR-BI.
In the present study, western blotting was used to show that a single dose of
CD promotes subcellular distribution of SR-BII to murine hepatic microsomes while
having no effect on liver crude membrane SR-BII. Western blotting also indicated CD
pretreatment had no effect on the levels of liver fatty acid binding protein (L-FABP) in
cytosol, but an increase in myosin-9 was observed with mass spectrometry. Myosin-9
may play a role in intracellular vesicular transport. This may at least partially explain
the previously observed alterations in CH homeostasis produced by CD pretreatment.
Changes in relative protein levels using a CH binding protein enriched fraction
prepared from hepatic cytosol were detected with mass spectrometry using isobaric
tagging for relative and absolute quantification (iTRAQ) methodology. Many factors
can affect the accuracy of quantification with this technique. It has been observed that
the low collision energies normally used in ESI QTOF can result in low iTRAQ
reporter ion abundances. After evaluating protein changes in response to CD
pretreatment, two-way ANOVA was used to compare the mean protein iTRAQ results
from mass spectrometers using different collision energies in the same samples. It
appears that iTRAQ analyses performed on an ESI QTOF without any special
modifications in instrumental parameters produce essentially the same protein ratios as
those obtained on a MALDI TOF/TOF.
© Copyright by Richard C. Scheri
May 30, 2008
All Rights Reserved
Chlordecone Pretreatment Promoted Subcellular Distribution of Scavenger Receptor
Class B Type II to Murine Hepatic Microsomes; Mass Spectrometry Detected Hepatic
Soluble Cholesterol Binding Proteins and Comparison of Protein iTRAQ Ratios Using
ESI QTOF and MALDI TOF/TOF
by
Richard C. Scheri
A DISSERTATION
submitted to
Oregon State University
in partial fulfillment of
the requirement for the
degree of
Doctor of Philosophy
Presented May 30, 2008
Commencement June 2009
Doctor of Philosophy dissertation of Richard C. Scheri presented on May 30, 2008
APPROVED:
Major Professor, representing Toxicology
Head of the Department of Environmental and Molecular Toxicology
Dean of the Graduate School
I understand that my dissertation will become part of the permanent collection of
Oregon State University libraries. My signature below authorizes release of my
dissertation to any reader upon request.
Richard C. Scheri, Author
ACKNOWLEDGEMENTS
I would like to thank Professors Lawrence Curtis and Douglas Barofsky for
their guidance, encouragement and patience during my tenure at Oregon State
University. Without their consistency and humor, especially the latter, this sojourn
would have been far less rewarding and several orders of magnitude more painful.
Many thanks are also due to the individuals that served on my committee: Dr. Michael
Schimerlik, Dr. David Williams and Dr. Mark Harmon.
Many thanks to Dr. Dave Williams lab and especially to Beth Siddens for
equipment, lab space in which to work, antibodies, patience and advice.
I also need to sincerely thank Lilo Barosky and Brain Arbogast for getting me
up to speed with HPLC and mass spectrometry . Thanks to Jeff Morre, Mike Hare and
the rest of the mass spectrometry group for their support.
Dr. Sonia Anderson’s lab was extremely helpful and provided equipment and
advice and I sincerely thank Dean Malencik for what little I know about protein
chemistry.
Thanks also go to Eddie O’Donnell from Dr. Siva Koluri’s lab, and Nathan
Lopez from Dr Joseph Beckman’s lab for teaching me how to do westerns and to Dr
Samuel Bennett for the use of his imaging equipment and stimulating conversations.
Finally I would like to thank my good friends Jing Wang and Ben Bythell for
their friendship and support.
CONTRIBUTION OF AUTHORS
Dr. Junga Lee assisted with sample collection and sample preparation for the
work presented Chapter 2 and to a limited extent in Chapter 3.
The author helped with sample preparation and data analysis in the study
presented in Appendix A and with sample preparation in the work presented in
Appendix B.
Dr. David Mangelsdorf and Dr. Yuan Zhang performed the reporter gene
assays presented in Appendix 2.
Brain Arboghast, Lilo Barofsky and Mike Hare, although not coauthors, made
technical contributions to the work presented in this dissertation. Brian Arboghast ran
the samples and generated the pkl files for all samples run on the QTOF. Lilo
Barofsky and Mike Hare performed MALDI TOF/TOF analysis on some of the
samples.
Dr. David William’s laboratory generously provided antibodies to CYP4A.
Dr. Samuel Bennett kindly allowed the use of his imaging instrumentation.
TABLE OF CONTENTS
Page
Chapter 1
Introduction……………………………………………
1
Chapter 2
Chlordecone Pretreatment Promoted Subcellular
Distribution of Scavenger Receptor Class B Type II to
Murine Hepatic Microsomes…………………………..
13
A Comparison of Relative Quantification with Isobaric
Tags on a Subset of the Murine Hepatic Proteome using
ESI QTOF and MALDI TOF/…………………………
39
Conclusions……………………………………………
63
Bibliography…………………………………………………………...
66
Appendices…………………………………………………………….
77
Chapter 3
Chapter4
Appendix A
Appendix B
Chlordecone Altered Hepatic Disposition of
14
[ C]Cholesterol and Plasma Cholesterol
Distribution but not SR-BI or ABCG8 Proteins in
Livers of C57BL/6 Mice…………….…………
78
Chlordecone, a Mixed Pregnane X Receptor (PXR)
and Estrogen Receptor Alpha (ERα) Agonist, Alters
Cholesterol Homeostasis and Lipoprotein
Metabolism in C57BL/6 Mice…………………
107
LIST OF FIGURES
Figure
Page
1-1
Structure of Chlordecone………………………………………… 10
1-2
ICAT Reagent Structure…………………………………………. 11
1-3
iTRAQ Reagent Chemistry………………………………………
2-1
Process flow chart for isotope labeling and mass spectrometry
analysis…………………………………………………………… 33
2-2
Up regulation of hepatic microsomal SR-BII by CD
pretreatment……………………………………………………… 34
2-3
CD pretreatment has no effect on SR-BII expression in the
hepatic crude membrane fraction………………………………… 35
2-4
Up regulation of hepatic crude membrane fraction Na+/K+ATPase
by CD pretreatment………………………………………………. 36
2-5
CD pretreatment had no effect on cytosolic L-FABP……………. 37
3-1
Flow chart showing the sequence of iTRAQ labeling and
mass spectrometry analysis………………………………………. 56
3-2
Flow chart showing how iTRAQ samples analyzed……………... 57
3-3
Enrichment of [14C] cholesterol and decrease in protein
observed in the supernatant and streptomycin precipitate
of hepatic cytosol…………………………………………………. 58
3-4
Mass spectra of the iTRAQ labeled peptide VFIEDVSK………… 59
12
LIST OF TABLES
Table
2-1
Page
Proteins showing a significantly different expression ratio
and L-FABP in the CH binding protein enriched fraction……….
38
Proteins detected in all four animal pairs and with both
instruments………………………………………………………
60
3-2
Mean protein iTRAQ values from both instruments………..….
61
3-3
P-values associated with the factors sample and instrument…...
62
3-1
Chlordecone Pretreatment Promoted Subcellular Distribution of Scavenger Receptor
Class B Type II to Murine Hepatic Microsomes; Mass Spectrometry Detected Hepatic
Soluble Cholesterol Binding Proteins and Comparison of Protein iTRAQ Ratios Using
ESI QTOF and MALDI TOF/TOF
Chapter 1
Introduction
Chlordecone (CD), trade name Kepone®, is a member of the persistent
organochlorine (OC) pollutants (POPs) that was first synthesized and registered as a
pesticide during the early 1950s (ATSDR 1995). The structure of CD is shown in
Figure 1-1. Its manufacture was banned in 1976 following the “Kepone episode” in
Hopewell, Virginia where poor manufacturing practices resulted in symptoms of CD
toxicity in a majority of plant workers and contamination of the nearby James River
estuary (Raloff 1976; Guzelin 1982). The majority of CD produced in the United
States was exported to West Germany, where it was used in the production of the
pesticide Kelevan, which was then exported for use primarily against the banana root
borer weevil in Central and South America (Cannon et. al., 1978). Like the other OC
pesticides, CD is lipophilic and resistant to bio/photodegradation. Since it is lipophilic
and so resistant to degradation, transfer through food webs is the primary route of
exposure where it tends to accumulate in the tissues of higher level predators
throughout the world.
In addition to concerns about the long term toxicity of CD itself, several
studies have shown that low doses perturb lipid homeostasis in a variety of species.
Carpenter and Curtis (1989; 1991) showed that pretreatment of mice with an acute,
2
nontoxic dose of CD altered the subsequent distribution of an exogenous dose of
[14C]CD or [14C] cholesterol (CH). Subsequent studies using CD to perturb lipid
homeostasis indicated, consistent with previous results, that acute CD pretreatment (15
mg/kg) resulted in increased biliary excretion of an intraperitoneal (ip) lipid bolus of
[14C]CH and reduced deposition of the labeled CH in liver without affecting either
scavenger receptor class B type 1 (SR-BI) or ATP cassette binding protein G8
(ABBG8) (Appendix 1). The same study showed that CD pretreatment altered
lipoprotein metabolism and, further, that CD (10 μM) strongly suppressed liver X
receptor beta (LXRβ) activation in cell reporter assays (Appendix 2). Additionally, the
same dose of CD was a strong agonist for a human homolog of the pregnane X
receptor (PXR), estrogen receptor alpha (ERα) while suppressing the activation of
estrogen receptor beta (ERβ) by β-estradiol (E2) (Appendix 2).
The tissue distribution of CD differs from other lipophilic OC pesticides. Its
tissue distribution is not proportional to tissue lipid content in both humans and
rodents and it tends to concentrate in liver (Cohn et. al., 1978; Egle et. al., 1978).
Other OC pesticides, e.g., dichlorodiphenyltrichloroethane (DDT) and dieldrin, are
associated primarily with low density lipoproteins (LDLs) and very low density
lipoproteins (VLDLs) in the blood (Micket et. al., 1971). In contrast, CD is associated
primarily with albumin and high density lipoprotein (HDL), which may account for
the different tissue distribution profiles that are observed (Soine et. al., 1982). While
VLDL and LDL are primarily involved in the delivery of lipids to peripheral tissues,
HDL is primarily involved in lipid transport from peripheral tissues to the liver for
biliary excretion, a process termed reverse CH transport (RCT) (Lewis and Rader
3
2005; Attie 2007). Additionally, CD is refractory to metabolism in mice, therefore the
production of metabolites is not an issue in this species (Guzelian 1982).
CH serves many important functions in mammalian cells. It is a constituent of
membranes and used for the synthesis of bile acids and steroid hormones. Its
association with sphingomyelin allows for the formation of lipid rafts within
membranes which may play a role in cell signaling pathways and intracellular
transport (Brown and London 2000; Prinz 2007). Although several mechanisms
contribute to the carefully controlled intracellular regulation of CH homeostasis,
primary regulation is accomplished by the cells ability to monitor sterol levels within
the endoplasmic reticulum (ER). This system utilizes sterol controlled transport of the
membrane bound transcription factor, sterol response element binding protein
(SREBP), from the ER to the Golgi where it is cleaved by proteases. After cleavage,
the homodimer migrates to the nucleus where it interacts with sterol response elements
in a variety of genes which control CH synthesis and uptake (Brown and Goldstein,
1997). CH is packaged for transport in the plasma as lipoproteins, the levels of which
are dictated by a variety of nuclear receptors and the diet (Nomiyama and Breummer
2008; Li and Glass 2004; Wang 2007). Nuclear receptors also mediate the metabolism
of CH to bile acids for excretion into the bile, a process that involves the transfer of
CH from HDL to SR-BI, at the plasma membrane, the transport of CH through the cell
and finally the transfer of CH across the canalicular membrane by the ABCG5/G8
heterodimer (Kosters et. al., 2005). Several studies have indicated that HDL CH is the
sole source for biliary CH (Wustner et. al., 2003; Robins and Fasulo 1997). However,
4
the sequence of events whereby the CH is actually transported from the basolateral to
the apical membrane has not yet been fully elucidated.
SR-BI mediates selective lipid uptake (SLU) from HDLs and reducing its
expression increases plasma HDL-CH and decreases biliary CH in rodents (Rhainds
and Brissette 2004; Rigotti et. al., 2003). Scavenger receptor class B type II (SR-BII)
is an alternative splice variant of the SR-BI gene that also mediates SLU from HDL,
but the role it plays in lipid homeostasis is not clear (Webb et. al., 1998). In an effort
to explain the altered lipid homeostasis resulting from CD pretreatment, earlier work
had shown that SR-BI or ABCG8 protein levels were not altered (Appendix 1). The
present study examines whether SR-BII protein levels in hepatic plasma membrane or
microsomal membrane fractions are affected by CD pretreatment. This work is
presented in Chapter 2.
The intracellular transport of CH is complex and both vesicular and nonvesicular transport mechanisms are involved (Ikonen 2008; Prinz 2007). Their relative
roles have not been fully elucidated - the non-vesicular transport mechanisms are
especially not well understood. Another possible explanation for the altered lipid
homeostasis observed with CD pretreatment that does not involve membrane bound
transporters might involve soluble (cytosolic) CH binding proteins. A proteomic
approach using isobaric tagging for relative and absolute quantification (iTRAQ)
methodology (described below) was used in this work because it appeared to have the
greatest potential for quickly identifying and characterizing the relative changes in
these soluble CH binding proteins. The details of this approach are presented in
5
Chapter 3, and the changes in protein expression that resulted from CD pretreatment
are presented in Chapter 2.
Proteomics is the general term coined for the analysis of the proteome, i.e., the
total complement of proteins produced at a given time, under a given set of conditions
in a cell. Advances in chromatographic and mass spectrometry techniques over the last
decade or so has made proteomics a reality (Abersold and Mann 2003). Although the
complete identification of the entire complement of the proteins in a proteome has not
yet been achieved, great strides have been made in this direction. A common method
of analysis is termed shotgun, or bottom-up proteomics. In this technique, the proteins
in a sample are enzymatically or chemically digested into peptides (increasing the
complexity of an already complex sample) that are analyzed on a mass spectrometer
after chromatographic separation. Because of the aforementioned complexity of the
proteome and the limited dynamic range of a mass spectrometer, several means of
reducing the complexity of the sample have been developed. Two-dimension gels
were first utilized. Gel spots were cut out and the proteins in the spots digested. The
resulting mixture of peptides could then be run on an HPLC and analyzed on a mass
spectrometer. Inherent problems included variability between gels, their inability to
separate membrane proteins or proteins with extreme pIs, and the difficultly of
imaging low abundance proteins on a gel (Gorg et. al., 2000; Gygi et. al., 2000). The
2D gel methodologies have slowly given way to multidimensional chromatographic
techniques (Washburn et al., 2001; Chen el al., 2006). In this technique the peptide
digest is first separated with some type of chromatography; a common first stage
chromatographic step uses a strong cation exchange resin. The fractions eluted from
6
this column are then run on another column, e.g., a reverse phase C18, and the effluent
is analyzed on a mass spectrometer.
As a further means of simplifying the cytosolic proteome in this study,
enrichment of the soluble CH binding fraction was explored as a means to make the
mass spectrometry more tenable. Streptomycin precipitation of cytosol proved to be
effective in concentrating [14C] labeled CH while at the same time decreasing the total
amount of protein present (Chapter 3).
In order to perform mass spectrometry, biomolecules first have to be ionized
into the gas phase. The two most commonly used techniques for ionization of
biomolecules are electrospray ionization (ESI) and matrix-assisted laser
desorption/ionization (MALDI) (Yamishita and Fenn 1984; Karas et. al., 1987). Since
ESI ionizes molecules directly from the liquid phase by solvent
desolvation/evaporation, it can easily be coupled to LC. The peptide ions produced by
this technique tend to be multiply charged. In MALDI the analytes are mixed into a
matrix that is then crystallized on a sample plate. The dried matrix absorbs photons
provided by a laser resulting in the matrix and analytes being transferred to the gas
phase and ionized in the process. MALDI tends to produce singly charged ions, which
makes the mass spectra easier to interpret. In this study, ion separation in the ESI
analyses was accomplished by a quadrapole/ time-of-flight (TOF) (QTOF) and in the
MALDI analyses by sequential TOF analyzers (TOF/TOF). Both of these instruments
are capable of tandem mass spectrometry (MS/MS), a technique wherein the first
analyzer is used to choose a precursor ion; a collision cell located between the
analyzers fragments the precursor; and the fragment ions produced are then analyzed.
7
One of the differences between the two instruments used in this study was the MS/MS
fragmentation energies used in the collision cells. The MALDI TOF/TOF fragments
precursors at an energy of about 1.0 keV while the ESI QTOF fragments them
anywhere from 25-65 eV, depending on the precursor charge state.
Once MS/MS spectra are generated they are searched against a protein
database in order to identify the proteins present. Peptides from proteins in the
database are fragmented in silico, and the resulting theoretical spectra are compared
with the experimental spectra. The matches are scored differently by different search
engines. The Mascot search engine was used in this study; it uses a probability based
implementation of the molecular weight search (MOWSE) algorithm (Matrix Science,
UK). It is essentially a probabilistic threshold model that sorts scores by some
threshold. The problem is that different search engines have different scoring schemes
that result in different thresholds. Additionally with each search engine, several values
are usually reported and the question of which value(s) to use or some means of
combining this information must be utilized. For example, Mascot reports homology
and identity scores, as well as ion scores, but most people just take ions above the
identity score as those being significant. This means that results using different search
engines cannot be directly compared. Other problems with threshold type algorithms
arise because different thresholds may be needed for different charge states,
instruments, databases, or sample complexities. Peptide Prophet is an algorithm
developed to address these issues (Keller et. al., 2002; Nesvizhskii and Aebersold
2004). It uses Bayesian statistics to compute the true probability that a match is correct.
A single discriminant score is calculated for each peptide from all of the threshold
8
scores from a particular search engine. A histogram is plotted of the number of spectra
versus the discriminant score; curves are fit to the bimodal distribution that results
(which consists of incorrect and correct distributions); and the probabilities are
calculated (Keller et. al., 2002; Nesvizhskii et. al., 2004).The distributions also allow
the estimation of errors. Probabilities generated by this software can be compared
between search engines, instrument platforms, charge states, etc. The commercial
product (Proteome Software, Portland, OR) that incorporates Peptide Prophet is called
Scaffold. As mentioned previously, Mascot was used for peptide identification, but the
Mascot results were further processed in Scaffold before protein identification
(Chapter 3).
The ability to accurately compare tissue protein concentrations under different
conditions is vital for understanding the underlying biological process(es). A variety
of methods have been developed using isotopic labeling in order to quantify proteins
identified with mass spectrometry (Julka and Regnier 2004). These techniques can be
MS or MS/MS based depending on the properties of the label itself. A popular
approach that is MS based uses an isotope coded affinity tag (ICAT) (Gygi, et. al.,
1999) This method uses light and heavy cysteine specific tags (Figure 2) that are
linked to a biotin moiety so the cysteine containing peptides can be enriched using
affinity chromatography. However, not all proteins contain cysteine, and since only
two tags (light and heavy) were available, multiplexing was impossible. An MS/MS
methodology, known as iTRAQ was developed by Ross et al., (2004) and
commercialized by Applied Biosystems Inc. It was designed to overcome some of the
problems associated with the ICAT labeling methodology. The molecule consists of a
9
charged N-methylpiperazine based reporter group, a carbonyl balance group, and an
N-hydroxysuccinimide (NHS) peptide reactive group (Figure 3). The total mass of the
reporter and balance group is kept constant at 145.1 Da though use of 13C, 15N and 18O
atoms. Currently eight reporters are available so multiplexing up to eight samples,
time points, or treatments is possible. After protein digestion, all primary amines in the
peptides are labeled. Isobaric tagging implies that peptides from different experimental
groups containing different labels elute at the same time and appear in MS spectra at
the same m/z, significantly simplifying the spectra. Charged reporter ions are then
released during collision induced dissociation (CID) with the carbonyl undergoing a
neutral loss. Relative quantification is achieved using either the peak area or intensity
of the reporter ions.
It has been observed that the low collision energies normally used in ESI
QTOF can result in low iTRAQ reporter ion abundances (Wiese et al., 2007).Since I
had access to both ESI and MALDI instruments I decided to compare the protein
iTRAQ ratios obtained on both instruments. I used two-way analysis of variance
(ANOVA) to compare the iTRAQ ratios that were generated on an ESI QTOF and a
MALDI TOF/TOF. This work in presented in Chapter 3.
10
Figure 1-1. Structure of Chlordecone
11
Biotin Moiety
Iodoacetamide
Labeled Region
Figure 1-2. ICAT reagent structure; modified from Liebler, D.C., 2002,
Introduction to Proteomics: Tools for the New Biology, Humana Press
12
13
Chapter 2
Chlordecone Pretreatment Promoted Subcellular Distribution of Scavenger Receptor
Class B Type II to Murine Hepatic Microsomes
Richard C. Scheri, Junga Lee, Douglas F. Barofsky and Lawrence R. Curtis
14
ABSTRACT
The intracellular transport of CH is complex and both vesicular and non-vesicular
transport mechanisms are involved. Their relative roles have not been fully elucidated
and the non-vesicular transport mechanisms are not well characterized. Previous work
has shown that the pretreatment of male C57BL/6 mice with low doses of the
persistent organochlorine (OC) pesticide, chlordecone (CD) stimulated biliary
excretion of exogenous CH up to 3-fold. Increased biliary excretion was not associated
with changes in ATP-binding cassette transporter G8 (ABCG8) or scavenger receptor
class B type I (SR-BI). Scavenger receptor class B type II (SR-BII) was identified as a
splice variant from the SR-BI gene and is expressed in a variety of tissues. Although
the function of SR-BII is not clear it was proposed to play a role in CH homeostasis
and trafficking that was distinctly different than SR-BI. Here we use western blotting
to show that a single dose of CD promotes subcellular distribution of SR-BII to
murine hepatic microsomes about 3 fold while having no effect on liver crude
membrane SR-BII. Western blotting also showed CD pretreatment had no effect on
the levels of liver fatty acid binding protein (L-FABP) in cytosol, but an increase in
myosin-9 was observed with mass spectrometry. This may at least partially explain the
previously observed alterations in CH homeostasis produced by CD pretreatment.
INTRODUCTION
CD is a member of the group of persistent OC pesticides whose manufacture in the
United States was banned in 1976. Although western manufacture of CD was halted
over 30 years ago, it is still present in the biosphere and detected in the tissues of
15
wildlife and humans throughout the world (Bocquene and Franco 2005; Luellen, et al.,
2006). It is highly hydrophobic, generally extremely resistant to biodegradation and
undergoes negligible photodegradation in the atmosphere (ATSDR 1995). The major
route of exposure for CD and OC pesticides in general is trophic transfer through the
food web with maximal accumulations in top predators (Christensen et al., 2005). OC
pesticides, e.g., CD, dichlorodiphenyltrichloroethane (DDT) and dieldrin, differ from
the polychlorinated biphenyls and the chlorinated furans and dioxins in that they
exhibit only slight affinity for the aryl hydrocarbon receptor (Poland and Knutson
1982). CD is unusual in that a large body of toxicological data is available for humans
as a result of a well publicized poisoning epidemic in chemical workers that occurred
in the mid-seventies (Cannon et al., 1978; Guzelian 1982) Like most OC pesticides,
neurotoxicity is a prominent mode of action and tremors were a common symptom in
the exposed human population (Cannon et al., 1978). CD has been observed to affect
ion channels and inhibit Na+/K+ ATPases in a variety of tissues (Narahashi et al., 1998;
Desaiah 1982). It is also a liver carcinogen in both rats and mice (Reuber 1978). CD
has been shown to alter lipid homeostasis in a variety of species. Ishikawa showed
single doses of CD and dieldrin decreased total plasma CH and triglycerides in rats
(Ishikawa et al., 1978). Several studies showed altered lipid homeostasis and tissue
deposition of CD, dieldrin and CH in rainbow trout, rats and mice (Carpenter and
Curtis 1989, 1991; Donohoe et. al, 1998; Gilroy et al., 1994). Single doses (5 mg/kg)
of CD altered the tissue distribution of an exogenous lipid bolus containing 14[C]CD
or 14[C]CH (Carpenter and Curtis 1991). Pretreatment with single doses of CD altered
lipoprotein metabolism in fasted mice (Lee et al., 2007). High density lipoprotein
16
(HDL) CH was not affected by CD pretreatment, but the non HDL-CH decreased and
the HDL-CH:total plasma CH ratio increased (Lee et al., 2007).The same authors
showed CD pretreatment increased the biliary excretion of exogenous 14[C]CH 4 h
after administration while having no effect on expression of hepatic ABCG8 or SR-BI
(Lee et.al., 2008). Whole livers of C57BL/6 mice contained approximately 85 μM CD
16 h after administration of 5 mg CD/kg (Carpenter and Curtis, 1989). Similar
concentrations of CD in cell reporter assays indicated that CD was a partial agonist for
the farnesoid X receptor (FXR) and peroxisome proliferator activated receptor alpha
(PPARα). CD was also a strong agonist for the human homolog of the pregnane X
receptor (PXR), and estrogen receptor alpha (ERα). It was a weak antagonist for liver
X receptor alpha (LXRα), a strong antagonist for LXRβ, and would inhibit the binding
of β-estradiol (E2) to ERβ, but it had no agonist activity on ERβ itself. In vivo CD
exposures increased hepatic microsomal cytochrome P450 (CYP) protein and enzyme
activities associated with these nuclear receptors. CD was a strong PXR agonist
(CYP3A11) with little effect on CYP7A1 (LXR/FXR) or CYP4A1 (PPARα) activity.
Since most OC pesticides are believed to be PXR agonists and CD has been
demonstrated to have estrogenic properties these results were not surprising (Guzelian
1982; Coumoul et. al., 2002).
FXR, LXR, PPAR and PXR are members of the nuclear receptor superfamily and can
act at several points to influence lipid homeostasis. They are ligand-activated
transcription factors that bind to specific response elements after heterodimerization
with the retinoid X receptor (RXR) (McKenna et al., 1999; Chalwa et al., 2001).
17
Physiological ligands for FXR are bile acids and FXR response elements are located in
genes involved in bile acid homeostasis (Kalaany and Magelsdorf, 2006). CYP7A1 is
the rate limiting step in the neutral bile acid synthesis pathway. Excess bile acids
activate FXR, which acts indirectly through small heterodimer partner (SHP) to
decrease the expression of CYP7A1 (Kuipers, et al., 2007). Oxysterols are
physiological ligands for LXR and high concentrations of oxysterols act through LXR
to downregulate CYP7A1 (Kalaany and Magelsdorf, 2006). Similarly, the three
isoforms of PPAR play key roles in the synthesis and metabolism of fatty acids, which
are their physiological ligands (Smith 2002; Michalik and Wahli 1999). They also
regulate the expression of sterol 12α hydroxylase (CYP4A1), a step in bile acid
synthesis. A variety of xenobiotics are ligands for PXR, including bile acids that can
induce CYP7A1 through this nuclear receptor (Staudinger et. al., 2001a; 2001b).
SR-BI mediates selective lipid uptake (SLU) from HDL and reducing its expression
increases plasma HDL-CH and decreases biliary CH in rodents (Rhainds and Brissette
2004; Rigotti et al., 2003). It can also efflux free CH to HDL (Mardones et al., 2001).
SR-BII is an alternative splice variant of the SR-BI gene that also mediates SLU from
HDL but is expressed at much lower levels in liver (Webb et al., 1998). The role it
plays in lipid homeostasis is not clear.
One possible explanation for the altered lipid homeostasis observed with CD
pretreatment that does not involve membrane bound transporters might involve
changing levels of cytosolic CH binding proteins. The intracellular transport of CH is
18
complex and both vesicular and non-vesicular transport mechanisms are involved
(Ikonen 2008; Prinz 2007). Their relative roles have not been fully elucidated and the
non-vesicular transport mechanisms are especially not well understood. Candidates for
soluble sterol transporters include, but are not limited to, FABPs, caveolins, sterol
carrier protein 2 (SCP-2), oxysterol binding proteins (OSBPs) and OSBP-related
proteins (ORPs), Niemann Pick type C (NPC) proteins and the steroidogenic acute
regulatory protein (StAR) and the StAR related lipid transfer (START) proteins
(Ikonen 2008; Prinz 2007). A proteomic approach was applied here because it had the
greatest potential for quickly identifying and characterizing the relative changes in
hepatic cytosolic proteins.
In this work, we examined the effect of CD pretreatment on murine hepatic SR-BII,
Na+/K+ATPase, liver fatty acid binding protein (L-FABP), and an enriched cytosolic
fraction of CH binding proteins after the intraperitoneal (ip) injection of 5 mL/kg of
corn oil containing either 5mg/kg or 15 mg/kg CD. SR-BII was analyzed by western
blotting hepatic microsomal fractions from CD treated mice (5 mg/kg and 15 mg/kg)
and controls. Na+/K+ATPase was examined by western blotting crude hepatic
membrane fractions from CD treated mice (5 mg/kg and 15 mg/kg) and controls. LFABP was analyzed by western blotting hepatic cytosol fractions from CD treated
mice (5 mg/kg and 15 mg/kg) and controls. Relative quantification of a cytosolic
enriched CH binding fraction was performed with mass spectrometry using stable
isotope labeling.
19
Our results demonstrated that CD pretreatment increased hepatic microsomal SR-BII
and hepatic crude membrane Na+/K+ATPase while having no effect on cytosolic LFABP. An increase in myosin-9 was noted in the CH binding protein enriched fraction,
whereas decreases in peroxiredoxin 1 (PRX-1), selenium binding protein 1 (SBP-1),
and α-enolase were observed.
MATERIALS AND METHODS
Chemicals
CD was purchased from Chem Service (West Chester, PA). Purity (99%) was
confirmed by gas chromatography-electron ionization/mass spectrometry.
Bicinchoninic acid (BCA) protein assay reagents were purchased from Pierce
Biotechnology Inc. (Rockford, IL). iTRAQ reagents were purchased from Applied
Biosystems (ABI, Foster City, CA). Burdick and Jackson solvents (VWR, West
Chester, PA) were used for all high performance liquid chromatography (HPLC). All
other chemicals were from Sigma (St. Louis, MO).
Treatment of mice
Six-seven week old male C57BL/6 mice were purchased from Simenson Laboratory
(Gilroy, CA). They were randomly assigned to one of three groups: control, low dose
(5 mg/kg) CD, or high dose (15 mg/kg) CD. They were individually housed in a
temperature (22±1ºC) and light controlled (12 h light/12 h dark daily) facility with free
access to water and fed AIN93 diet (Dyets, Inc., Bethlehem, PA) ad libitum. Seven
days were allowed for acclimatization before treatments were initiated. Corn oil was
20
used as a vehicle (5 mL/kg). Mice received a single dose of either corn oil or CD (5
mg/kg or 15 mg/kg) via intraperitoneal (IP) injection. After 72 h, they were fasted for
4 h and killed by carbon dioxide anesthesia and exsanguination. Livers were harvested.
The procedures for animal use were approved by the Oregon State University
Institutional Animal Care and Use Committee (IACUC).
Tissue Preparation
Liver was homogenized with a Polytron PT 3000 PT (Brinkmann Instruments,
Westbury CT) in buffer consisting of 0.01 M potassium phosphate, pH 7.4, 0.15 M
potassium chloride (KCL), 1.0 mM ethylenediaminetetraacetic acid (EDTA), 0.1 mM
butylated hydroxytoluene (BHT) and 0.1 mM phenylmethlysulfonyl fluoride (PMSF).
All centrifugation was performed at 4ºC. The homogenate was first centrifuged at
1200 g for 30 minutes. The pellet produced consisted of plasma membrane/cellular
debris and was designated the hepatic crude membrane fraction. The supernatant was
then centrifuged at 12,000 g for 30 minutes to eliminate mitochondria. The
supernatant from this run was centrifuged at 100,000 g for 90 minutes in a Ti 70 rotor.
The resulting supernatant and pellet were the cytosolic and microsomal fractions
respectively. Microsomes were resuspended in 0.1 M potassium phosphate, pH 7.25,
1.0 mM EDTA, 30% glycerol, 20 μM BHT, 1.0 mM dithiothreitol (DTT), and 0.1 mM
PMSF. BCA assay determined protein content.
PAGE and Western Blot Analysis
21
Cytosol, debris and microsome fractions were prepared as described above. Proteins
(10 μg) were separated on either 10% or 4-12% Bis-Tris gels (Invitrogen, Carlsbad,
CA) at 100V for 2 h. Proteins were stained with Biosafe Coomassie (Biorad, Hercules,
CA) for PAGE gel imaging. For Western blots the unstained proteins were
electrophoretically transferred to a polyvinylidene fluoride (PVDF) membrane using a
XCell II Blot module (Invitrogen, Carlsbad, CA) at 30 V for 1 h. Debris, microsome,
and cytosol fractions were then blotted with antibodies: Rabbit anti-SR-BII was
purchased from Abcam (Cambridge, MA) or GeneTex (SanAntonio, TX). Rabbit antiβ-Actin was also purchased from Abcam. Rabbit anti-Na+/K+ ATPase, and rabbit antiL-FABP were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Goat
anti-CYP4A was a generous gift from the laboratory of Professor David Williams.
Debris was blotted with antibodies against SR-BII and Na+/K+ATPase. Microsomes
were blotted with SR-BII and CYP 4A. Cytosol was blotted with L-FABP and β-Actin.
Blots were then probed with either donkey anti-goat or goat anti-rabbit conjugated to
horse radish peroxidase (HRP) as the secondary antibody (Santa Cruz Biotechnology,
Inc.). Proteins were detected after development with SuperSignal® West Pico
Chemiluminescent Substrate (Pierce Biotechnology, Rockford, IL). Images were
acquired on a Syngene Chemigenius 2 bioimaging system and analyzed with
Genesnap/Genetools software (Syngene, Frederick, MD).
Tissue Preparation for iTRAQ Experiments
Cytosolic fractions enriched in CH binding proteins were prepared by streptomycin
precipitation (Scheri et al., 2008). The BCA assay determined protein concentrations.
22
Protein (100 μg) samples were precipitated with six volumes of ice cold acetone and
labeled with iTRAQ reagents as detailed below.
iTRAQ Labeling
Samples were labeled with iTRAQ reagents according to the manufacturers’ protocol
(Figure 2-1). Control samples were labeled with iTRAQ 114.1 or 116.1 m/z, while CD
treated samples (15 mg/kg) were labeled with iTRAQ 115.1 or 117.1 m/z. An iTRAQ
sample consisted of 100 μg of protein from a control animal pooled with 100 μg of
protein from a CD treated animal. One animal pair was used for each iTRAQ sample,
and four pairs were analyzed. Two technical replicates were run on each instrument
for each sample pair. The iTRAQ ratios from the replicates were then averaged. Mean
iTRAQ protein ratios were generated using the average ratio from both instruments.
Ratios were then averaged across the biological sample pairs. A complete description
this protocol is given elsewhere (Scheri et al., 2008).
Mass Spectrometry
After nano high pressure liquid chromatography (HPLC) (Waters nanoAcquity Ultra
Performance LC, Milford, MA), mass spectrometry (MS) was performed on a
quadrapole orthogonal time-of-flight mass spectrometer (QTOF Ultima Global,
Micromass/Waters, Manchester UK) and on an ABI 4700 Proteomics Analyzer
(Foster City, CA). The QTOF incorporated an electro-spray ionization (ESI) source,
while the 4700 used a matrix assisted laser enhanced desorption ionization (MALDI)
source. Tandem mass spectra (MS/MS) were generated with each instrument.
23
Data was searched against the mammalian Swiss-Prot database (V55.0) using the
Mascot (Matrix Science, London, UK) search engine. iTRAQ ratios were generated in
Mascot and median normalized. Data sets were further analyzed with Scaffold
(Proteome Software, Inc., Portland, OR). Only peptides with a probability of 80% or
greater as determined by Scaffold were utilized for quantitative purposes. In the case
of duplicate peptides or different charge states, the iTRAQ ratio with the highest
Mascot score was used. A minimum of 2 peptides was required for protein
identification.
A complete description this protocol is given elsewhere (Scheri et al., 2008).
Statistical Methods
Statistical analyses were conducted with either StatGraphics (StatPoint, Herndon, VA)
or S-Plus (Insightful Corp, Seattle, WA). Normality of data was accessed with
standardized kurtosis, standardized skewness and homogeneity of variance. Western
blot data is expressed as the mean ± the standard error (SE). Following validation of
normality, comparisons between two groups was performed with a Student’s t-test.
Multiple groups were compared with a one-way analysis of variance (ANOVA). In all
cases, a 95% confidence level was used as the criterion for significance.
Data for iTRAQ ratios was analyzed in logarithm (natural, ln) space. Means
were geometric. The Student’s t-test was used to determine whether the mean iTRAQ
ratio for a particular protein was different than 1 (0 in ln space), both within a
24
biological sample pair and across the four biological sample pairs. In all cases, a 95%
confidence level was used as the criterion for significance.
RESULTS
Previous work has shown that the stimulated biliary excretion of exogenous CH
associated with CD pretreatment was not associated with changes in ABCG8 or SR-BI
(Lee et al., 2008). SR-BII was identified as a splice variant from the SR-BI gene and is
expressed in a variety of tissues. Although the function of SR-BII is not clear it was
proposed to play a role in CH homeostasis and trafficking that was distinctly different
than SR-BI (Eckhardt et al., 2004; 2006). This work suggested that analyzing SR-BII
levels after CD pretreatment might offer a partial explanation for the observed
alteration in CH homeostasis. As mentioned earlier, CD pretreatment had no effect on
the protein levels or activity of hepatic microsomal CYP4A1, which supported their
work showing that CD negligibly activated the nuclear receptor PPARα (Lee, el. al.,
2007). We therefore normalized the microsomal western SR-BII intensities to that of
CYP4A (Figure 2-2). SR-BII proteins levels increased at least 2-fold (p<0.05) as a
result of CD pretreatment.
SR-BII is present on the plasma membrane and undergoes rapid endocytosis (Eckhardt
et al., 2004), therefore, western blots were performed on hepatic crude membrane
fractions. Originally, normalization with respect to Na+/K+ ATPase was planned, but
the protein content of this enzyme changed with CD pretreatment. Both SR-BII and
Na+/K+ ATPase were therefore normalized with respect to lane band intensity (Fig. 3
25
and 4). CD pretreatment had no significant effect on SR-BII protein levels but resulted
in a statistically significant (p<0.05) increase in Na+/K+ATPase.
CD pretreatment had no effect on the levels of cytosolic L-FABP (normalized to βactin) (Figure 2-5).
Another possible explanation for the altered lipid homeostasis observed with CD
pretreatment was altered concentration of some kind of soluble sterol binding
protein(s). Mass spectrometry coupled with iTRAQ methodology for relative
quantification addressed this issue. Mass spectrometry was limited by dynamic range,
and the cytosolic proteome wascomplex. An enrichment of the fraction was of interest
to simplify matters and make the mass spectrometry more tenable. Original
enrichment experiments were performed with [14C] CH and streptomycin precipitation
of cytosol enriched the label per unit protein and left most of the cytosolic protein
(>85%) in the supernatant (Scheri et al., 2008).
Approximately 80 proteins were detected with mass spectrometry under the conditions
described in the experimental section (data not shown). Four proteins exhibited mean
iTRAQ protein ratios significantly different (p<0.05) than one (Table 2-1).
DISCUSSION
The underlying mechanisms for the CD pretreatment induced perturbation of CH
homeostasis and lipoprotein metabolism remain to be elucidated. As indicated earlier,
26
no change was observed in either SR-BI or ABCG8 in crude hepatic membrane
fractions after CD pretreatment (Lee et. al., 2008b). Although the ABCG5/G8
heterodimer has been shown to be the rate limiting step in biliary CH secretion, its
expression does not always correlate with this function which may indicate other
processes are involved (Yu el. al., 2005; Geuken et al., 2005).
SR-BI is a promiscuous receptor for a variety of ligands including HDL and modified
and unmodified forms of LDL. It mediates selective lipid uptake, which does not
require the uptake and degradation of lipoprotein, and CH efflux from cells (Rhainds
and Brisette 2004; Rigotti et al., 2003). In rodents, it plays a primary role in
controlling plasma lipoprotein levels and CH homeostasis, with major roles in reverse
CH transport (RCT) and biliary excretion (Rigotti et al., 2003; Mardones et al., 2001).
SR-BII was identified as a splice variant from the SR-BI gene and is expressed in a
variety of tissues (Webb et al., 1998). SR-BII also mediates selective lipid uptake from
HDL but at a lower efficiency than SR-BI (Webb et al., 1998). However, SR-BII is
able to efflux about twice as much free CH as SR-BI (Mulcahy et al., 2004). The
cellular distribution and internalization of these receptors also differs. In primary
hepatocytes, SR-BI is predominantly located on the plasma membrane although a
distinct pool appears to be associated with late endosomes and lysosomes (Ahras et al.,
2008; Eckhardt, et., al., 2004). The cytoplasmic tail of SR-BI in hepatic and intestinal
tissue interacts with the C-terminal linking and modulating protein (CLAMP), and this
association appears to be required for distribution to the plasma membrane because
knock-out of CLAMP prevents the surface expression of this receptor (Rhainds and
27
Brissette 2004). SR-BI undergoes endocytosis but at a much lower rate than SR-BII in
CHO cells (Eckhardt et al., 2004). Endocytosis is not required for selective lipid
uptake with SR-BI (Nieland et al., 2005; Wustner et al., 2003). The endocytosis of
these receptors may serve to transport lipids to different pools in different locations
within the cell, and it has been suggested that SR-BII endocytosis plays a role in
altering the CH content of the endocytotic compartment (Eckhardt et al., 2006). SRBII is located primarily intracellularly and is rapidly endocytosed in transferrin
containing vesicles in CHO cells (Eckhardt et al., 2004). A dileucine motif in the
carboxy terminal appears to be responsible for the rapid endocytosis and incorporation
of this motif into SR-BI resulted in endocytosis of SR-BI (Eckhardt et al., 2006).
Although mRNA levels of these receptors in murine liver are similar, SR-BII is
expressed 10-15% of the levels at which SR-BI is expressed (Webb et al., 1998).
The function of SR-BII is not clear but it was proposed to play a role in CH
homeostasis and trafficking that was distinctly different than SR-BI (Eckhardt et al.,
2004). Wustner et al. (2003) proposed that an endocytotic/retroendocytotic pathway
might be involved in biliary CH secretion. As mentioned previously, SR-BII effluxed
free CH at a greater rate than SR-BI (Mulcahy et al., 2004). Based on a number of
observations (by themselves as well as others) Lopez and McLean (2005) proposed
that increased levels of SR-BII potentially increased excretion of CH in the bile. Our
results showed no change in total biliary CH with CD pretreatment in 4h fasted mice
(Lee et. al., 2008a). However, this was perhaps a question of limited detection
sensitivity of the change in total bile CH after CD pretreatment in fasted mice without
28
administration of exogenous CH. Free CH from HDL was preferentially excreted in
bile, as opposed to de novo free CH (Robins and Fasulo, 1997). CD pretreatment
stimulated biliary excretion of exogenous [14C]CH (2 μCi/animal, 10 mg CH/kg) and
this resulted in bile concentrations of approximately 120 nmol [14C]CH equivalents/ml
(Lee et al., 2008). Average bile CH concentrations in wild type mice range from about
2-5 μmol/ml (Wang et al, 2006; Yu et al., 2005; Tang et al, 2006), so that even with a
2-fold increase in SR-BII, a change of nmol amounts of CH in bile would not have
been detected. The total amount of SR-BII on the plasma membrane was not altered
by CD (Figure 2-3) and yet hepatic microsomal content of SR-BII was increased
(Figure 2-2). Perhaps this represented an increased capacity for biliary excretion of
CH in hepatocytes and upon the administration of exogenous CH, its excretion in the
bile was enhanced. The intracellular mechanism for this was unknown. Perhaps a
fraction of SR-BII was redistributed to the apical/plasma membrane upon the
administration of an exogenous lipid bolus. Increased turnover of plasma membrane
SR-BII was also a possibility. Since endocytosis of HDL appeared unimportant in
selective lipid uptake, perhaps increased turnover of plasma/apical membrane SR-BII
played a role in faster uptake of and excretion of CH into bile.
Apolipoprotein A-I (ApoA-I) is the major lipoprotein found in HDL and CD
pretreatment significantly increased its concentration in the hepatic lipoprotein rich
fraction and in microsomes (Lee et. al., 2007). LXR activation inhibited the synthesis
and secretion of apoA-I in Hep3B and HepG2 cells (Huuskonen et al., 2006), but,
based on the negligible effects CD had on CYP7A1, the involvement of LXR seems
29
unlikely. Bachman et al. (2004) showed PXR increased HDL-CH and apoA-I and
down regulated ABCA1 in vivo and in vitro in rats. ABCA1 is associated with the
transport of free CH to ApoA-1 and lipid poor ApoA1. ABCA1 was located on the
plasma membrane, with high concentrations in liver and intestine, and plays an
important role in RCT (Attie 2007). PXR agonists down regulated ABCA-1 and SRBI (Sporstal et al., 2005). The total and free CH concentrations in liver were not
changed, but there was a significant decrease in CH ester associated with CD
pretreatment (Lee et al., 2007). There was no evidence that SRB-II mobilized CH
esters in CHO cells (Mulcahy et al., 2004). Plasma Apo-AI and apolipoprotein B
(Apo-B), the latter being the major lipoprotein of VLDL and LDL, were unaffected by
CD pretreatment (Lee et. al., 2008a). In humans, estrogen increased biliary lipid
output and upregulated both Apo-B and the LDL receptor (Brown and Goldstein 1986;
Wang 2007). However, LDL receptors were not upregulated by estrogen in mice
(Srivastava et al., 1997). Estrogen decreased SR-BI and upregulated SR-BII in rat liver
(Lopez and McLean 2006; Zhang et al., 2007). Additionally, estrogen increased the
secretion of apoA-I in HepG2 cells (Lamon-Fava et al., 1999).
In the present study CD pretreatment increased the expression of SR-BII in the
microsomal fraction by at least 3-fold (Figure 2-2) but there was no effect on SR-BII
expression in the crude membrane fraction (Figure 2-3). Although CD was shown to
be an agonist for human ERα, no change in SR-BI we observed and there was no
change in total biliary CH after CD pretreatment in fasted mice (Lee et al, 2008a).
Elevated CYP3A11 in hepatic microsomes 14 days after treatment with CD was
30
consistent with PXR signaling, but no increase in HDL-CH or plasma apoA-1 was
detected (Lee 2008a). The constitutive androstane receptor (CAR) and PXR increased
Insig-1 gene expression in mice (Roth et al., 2008). Increased Insig 1 in the
endoplasmic reticulum (ER) would result in increased retention of sterol response
element binding protein 1 (SREBP-1) in the ER and the down regulation of target
genes including those involved in triglyceride synthesis. Triglycerides were not
analyzed in this experiment but an earlier study using the same CD pretreatment
regimen detected no changes in plasma triglycerides (Lee 2002). Additionally, toxic
doses of CD in humans resulted in no change in plasma triglycerides (Guzelain 1982).
Nuclear factor (erythroid-derived 2)-related factor 2 (Nrf-2) is a transcription factor
that binds to antioxidant response elements and regulates genes involved in cellular
response to oxidative stress (Tirona and Kim 2005). A significant reduction in prx1
gene expression in Nrf2 knock-outs has been observed (Kim el. al., 2007). The effect
of CD pretreatment on the expression of other nuclear receptors, e.g., CAR and Nrf2
was of future interest but the mechanism for the observed increase in SR-BII remained
unknown. Never the less, these results were compatible with a role for SR-BII in CD
pretreatment induced changes in CH homeostasis.
Na+/K+ ATPase protein content increased with CD pretreatment in hepatic crude
membrane fractions (Figure 2-4). Neurotoxicity was a prominent mode of action for
OC pesticides and several groups showed CD inhibited Na+/K+ ATPase (reviewed in
Desaiah, 1982). The increase observed was perhaps a compensatory response of this
inhibited transport capacity per unit protein.
31
Mass spectrometry has a limited dynamic range, and the cytosolic proteome is
complex. Selective protein enrichment of the fraction of interest, i.e., that containing
CH/sterol binding proteins, would help to simplify matters and make the mass
spectrometry more tenable. Original enrichment experiments were performed with
[14C] CH, and streptomycin precipitation of cytosol proved to be effective in
concentrating the label while at the same time decreasing the amount of protein
present (RCMS). Clearly, L-FABP protein levels did not change in cytosol with CD
pretreatment (Figure 2-5); a similar result was obtained for the streptomycin enriched
fraction (Table 2-1).
Approximately 80 proteins were detected (data not shown) in the streptomycin
enriched fraction under the conditions described in the methods section. Since the
iTRAQ technique was based on a peak ratio between the CD pretreated samples and
controls, a ratio different than one indicated a changed protein concentration.
Statistically significant differences from one were observed with four proteins.
Although both SCP-2 and L-FABP were detected, SCP-2 did not meet the minimum
criteria for protein identification. Additionally, with this technique, differences less
than 20% were viewed with caution. Decreased expression was observed with αenolase, PRX-1, and SBP-1. Increased ratio was observed with myosin-9. It has been
suggested that myosin-9 and SBP-1 played different roles in intracellular vesicular
transport (van den Boom et al., 2007; Porat et al., 2000). While PRX-1 was a
peroxide/ROS scavenger, the α-enolase gene encoded both a glycolytic enzyme and a
32
transcription factor, c-MYC binding protein (MBP-1) (Egler et al., 2005; Subramanian
and Miller 2000). An interesting side note was that both proteins appeared to play
important regulatory roles for the c-Myc transcription factor, a product of the c-MYC
oncogene. The c-MYC oncogene was associated with a variety of cell processes,
including apoptosis, cell growth, differentiation and cell cycle progression (Fernandez
et al., 2003; Egler et al., 2005). While MBP-1 binds the c-MYC promoter and
downregulates c-Myc transcription, prx1 interacts with the c-Myc transcription factor
itself (Subramanian and Miller 2000; Egler et al., 2005). This was perhaps related to
the carcinogenicity of long-term CD exposures in rodents.
In summary, CD pretreatment increased the SR-BII protein content of hepatic
microsomes but not crude membrane fractions. Increased myosin-9 was consistent
with an increased capacity for vesicular transport. These observations supported a
model where SR-BII contributed to CD stimulated biliary excretion of CH. Decreased
SBP-1, α-enolase and PRX-1in a CH enriched fraction of cytosol was observed, but
whatever this contributed to CD induced perturbations in CH homeostasis was unclear.
There was no evidence for a role of cytosolic binding proteins although additional
mass spectrometry experiments were needed to adequately address this issue. Overall,
these results provide new insights potentially explained how CD pretreatment induced
alterations in CH and lipoprotein homeostasis.
33
Control (corn oil)
Streptomycin
Precipitate (100 mg)
( 1 Animal Pair)
Treated (15 mg/kg CD)
Streptomycin
Precipitate (100 mg)
Reduction
Alkylation
Trypsin Digestion
Reduction
Alkylation
Trypsin Digestion
Label with 114 or 116
iTRAQ
Label with 115 or 117
iTRAQ
Pool Samples
SCX Spin Columns
- 4 KCl fractions, 75 mM, 150 mM, 350 mM and 700 mM
C18 HPLC followed by MALDI MS/MS and
ESI MS/MS.
Data Analysis by GPS Explorer, Mascot and Scaffold
Figure 2-1. Process flow chart for isotope labeling and mass spectrometry
analysis. A total of four animal pairs were analyzed.
34
A
Control
CD (5 mg/kg)
CD (15 mg/kg)
100 kDa
SR-BII
75 kDa
75 kDa
CYP 4A
50 kDa
B
0.40
0.35
*
*
CD (5 mg/kg)
CD (15 mg/kg)
SR-BII:Cyp4a
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Control
Treatment
Figure 2-2. Up regulation of hepatic microsomal SR-BII by CD pretreatment.
Animals were treated with corn oil, 5 mg/kg or 15 mg/kg CD by ip injection. Hepatic
microsomes were prepared from individual animals 3 days after CD pretreatment: (A)
western blot of SR-BII and CYP4A proteins, (10 μg protein/lane was separated by
SDS-PAGE and blotted with antibodies against SR-BII and CYP4A; (B) variation in
SR-BII with CD pretreatment, data normalized to CYP4A, (error bars are ± standard
error (SE), asterisks indicate different than control (p<0.05) after one-way ANOVA).
35
A
Control
CD (5 mg/kg)
100 kDa
75 kDa
CD (15 mg/kg)
SR-BII
B
SR-BII Intensity Ratio
0.60
0.50
0.40
0.30
0.20
0.10
0.00
Control
CD (5 mg/kg)
CD (15 mg/kg)
Treatment
Figure 2-3. No significant effect of CD pretreatment has no effect on SR-BII
expression in the hepatic crude membrane fraction. Animals were treated with
corn oil, 5 mg/kg or 15 mg/kg CD by ip injection and crude hepatic membrane
fractions were prepared from individual animals 3 days after CD pretreatment: (A)
western blot of SR-BII, (10 μg protein/lane was separated by SDS-PAGE and blotted
with antibodies against SR-BII); (B) Variation in SR-BII with CD pretreatment, data
normalized to lane band intensity, (error bars are ± SE).
36
A
Control
CD (5 mg/kg)
CD (15 mg/kg)
100 kDa
75 kDa
Na/K ATPase
B
Na+/K+ATPase Intensity Ratio
1.00
0.90
*
*
CD (5 mg/kg)
CD (15 mg/kg)
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
Control
Treatment
Figure 2-4. Increased hepatic crude membrane fraction content of Na+/K+ATPase
after CD pretreatment. Animals were treated with corn oil, 5 mg/kg or 15 mg/kg CD
by ip injection and crude hepatic membrane fractions were prepared from individual
animals 3 days after CD pretreatment: (A) western blot of Na+/K+ATPase, (10 μg
protein/lane was separated by SDS-PAGE and blotted with antibodies against
Na+/K+ATPase); (B) Variation in Na+/K+ATPase with CD pretreatment, data
normalized to lane band intensity, (error bars are ± SE, asterisks indicate different than
control (p<0.05) after one-way ANOVA).
37
Control
CD (15 mg/kg)
A
50 kD
b-Actin
37 kD
15 kD
L-FABP
10 kD
B
2.0
1.8
L-FABP:B-Action
1.5
1.3
1.0
0.8
0.5
0.3
0.0
Control
CD (15mg/kg)
Treatment
Figure 2-5. CD pretreatment had no effect on cytosolic L-FABP.
Animals were treated with corn oil or 15 mg/kg CD by ip injection and hepatic cytosol
was prepared from individual animals 3 days after CD pretreatment: (A) western blot
of L-FABP and β-actin proteins, ( μg protein/lane was separated by SDS-PAGE and
blotted with antibodies against L-FABP and β-actin); (B) L-FABP data normalized to
β-actin, (error bars are ± SEM, asterisks indicate different than control (p<0.05) after a
Student’s t-test).
38
Number of
Protein
Pairs (n)
Alpha-enolase
4
Fatty acid-binding protein (L-FABP)
4
Myosin-9
4
Peroxiredoxin-1
4
Selenium-binding protein 1
4
Mean iTRAQ Ratio
(Treated:Control)
0.81
0.89
1.35
0.87
0.72
Confidence
Interval (95%)
0.78 - 0.84
0.63 - 1.24
1.03 -1.75
0.77 - 0.99
0.58 - 0.90
Table 2-1. Proteins showing a significantly different expression ratio and LFABP in the CH binding protein enriched fraction. L-FABP was included as an
example of a potential sterol carrier protein whose mean iTRAQ ratio was not
significantly different than one. Confidence intervals were determined using the
Student’s T-test, p<0.05.
p
0.000
0.334
0.037
0.044
0.019
39
Chapter 3
A Comparison of Relative Quantification with Isobaric Tags on a Subset of the
Murine Hepatic Proteome using ESI QTOF and MALDI TOF/TOF
Richard C. Scheri+, Junga Lee1, Lawrence R. Curtis1 and Douglas F. Barofsky2
40
ABSTRACT
The use of isobaric tagging for relative and absolute quantification (iTRAQ) has
increased dramatically over the past few years. Many factors can affect the accuracy
of quantification. Some of these include the number of biological/technical replicates,
sample complexity, instrumentation, method of peptide/protein identification and the
statistical techniques used for data analysis. It has been observed that the low collision
energies normally used in ESI QTOF can result in low iTRAQ reporter ion
abundances. We used two-way analysis of variance (ANOVA) to compare the iTRAQ
ratios that were generated on an ESI QTOF and a MALDI TOF/TOF. It appears that
iTRAQ analyses performed on an ESI QTOF without any special modifications in
instrumental parameters produce essentially the same protein ratios as those obtained
on a MALDI TOF/TOF.
INTRODUCTION
The ability to detect protein expression changes is vital for understanding the
underlying biology of cellular processes at different times or in different states. For
example, a single low dose of the organochlorine pesticide chlordecone (CD) has been
shown to alter CH homeostasis in mice (Carpenter and Curtis, 1989;1991). Although
membrane transport proteins like ATP-binding cassette protein G8 and scavenger
receptor class B type I were not affected, the possibility of alterations in soluble CH
binding proteins is currently under investigation (Lee et al, 2008). A great deal of
effort has been expended in the development of a variety of stable isotope methods for
the relative quantification of changes in protein expression (Julka and Regnier, 2004).
41
Stable isotopes or isotopically labeled amino acids can be added to media and
incorporated into proteins during growth (Kolkman et al., 2005; Ong et al., 2003), but
this method is not easily adapted for use outside of the cell culture arena. Chemical
labeling methods are required for tissue analysis and many have been developed that
react with a variety of functional groups in proteins and peptides (Julka and Regnier,
2004; Schmidt et al., 2005). These techniques can be MS or MS/MS based depending
on the properties of the label itself. Isotope coded affinity tag (ICAT) technology is
MS based and was first developed by Gygi et al., (1999). It uses light and heavy
cysteine specific tags bound to a biotin moiety so the cysteine containing peptides can
be enriched using affinity chromatography. Not all proteins contain cysteine however,
and with the early versions of the ICAT reagents, the different masses of the light and
heavy tags resulted in different chromatographic performance. Additionally, only two
tags (light and heavy) were available so multiplexing (running more that two samples
at the same time) was impossible. The iTRAQ methodology was developed by Ross et
al., (2004), and commercialized by Applied Biosystems Inc. It was designed to
overcome some of the liabilities of the ICAT labeling methodology, and its use has
increased dramatically over the past few years. The molecule consists of a charged Nmethylpiperazine based reporter group, a carbonyl balance group and an Nhydroxysuccinimide (NHS) peptide reactive group. The mass of the reporter and
balance groups sum to 145.1 m/z and this is kept constant though use of 13C, 15N and
18
O atoms. Originally, only four different reporter groups were available, m/z 114.1,
115.1, 116.1 and 117.1, but the current version is available with eight reporters, which
would allow the comparison of up to eight states or time points. The NHS moiety
42
reacts with primary amines, and labeling is conventionally performed after trypsin
digestion. Consequently, the N-terminal and any lysines present in the peptide are
labeled. Since the tags are isobaric, the peptides from different experimental groups
containing different labels elute at the same time and appear in MS spectra at the same
m/z, significantly simplifying the spectra. N-methylpiperazine is a basic moiety and
accepts protons readily, which enhances the overall ionization efficiency of the
peptides (Ross et al., 2004). iTRAQ is MS/MS based in the sense that the charged
reporter ions are released during collision induced dissociation (CID) with the
carbonyl undergoing a neutral loss. Relative quantification is achieved using either the
peak area or intensity of the reporter ions. Successful use of this technique has been
demonstrated in many organisms and tissues (Ross et al., 2004; Aggarwal et al., 2005;
Hu et al., 2006; Glen et al., 2008; DeSouza et al., 2006; Reinders et al, 2006).
There are numerous factors that can affect the accuracy of iTRAQ quantification.
Some of these include the number of biological/technical replicates, sample
complexity, instrumentation, method of peptide/protein identification and the
statistical techniques used for data analysis. It has been observed that the low collision
energies normally used in ESI QTOF can result in low iTRAQ reporter ion
abundances (Wiese et al., 2007). We used multifactor ANOVA to compare the iTRAQ
ratios that were generated on an ESI QTOF and a MALDI TOF/TOF from a modest
set of samples.
EXPERIMENTAL
43
Materials
Chlordecone (CD) was purchased from Chem Service (West Chester, PA). Purity
(99%) was confirmed by gas chromatography-electron ionization/mass spectrometry.
Radiolabeled CH ([4-14C] CH, 55 mCi/mmol) was obtained from Radiolabeled
Chemicals, Inc. (St. Louis, MO). Bicinchoninic acid (BCA) protein assay reagents
were purchased from Pierce Biotechnology Inc. (Rockford, IL). iTRAQ reagents were
purchased from Applied Biosystems (ABI, Foster City, CA). Burdick and Jackson
solvents (VWR, West Chester, PA) were used for all high performance liquid
chromatography (HPLC). All other chemicals were purchased from Sigma (St. Louis,
MO).
Animals (CH Binding Protein Enrichment)
Previous dose response and time course experiments (Carpenter and Curtis, 1989,
1991) provided optimal doses and sampling times for CD pretreatment and the
exogenously administered lipid bolus containing radiolabeled CH ([14C]CH). Sixseven week old male C57BL/6 mice were purchased from Simenson Laboratory
(Gilroy, CA). They were randomly assigned to one of three groups: Control, Low dose
(5 mg/kg) CD, or High dose (15 mg/kg) CD. They were individually housed in a
temperature (22±1ºC) and light controlled (12 hours light/12 hours dark daily) facility
with free access to water and fed AIN93 diet (Dyets, Inc., Bethlehem, PA) ad libitum.
Seven days were allowed for acclimatization before treatments were initiated. Corn oil
was used as a vehicle (5 mL/kg). Mice received a single dose of either corn oil or CD
(5 mg/kg or 15 mg/kg) via intraperitoneal (IP) injection. After 72 hours a lipid bolus
44
of corn oil (5 mL/kg) containing 10 mg/kg CH (2 μCi/mouse) was administered IP.
After injection the mice continued to be housed individually and were allowed free
access to water. They were killed 16 hours later by carbon dioxide anesthesia and
exsanguination. Livers were removed for analysis. The procedures for animal use were
approved by the Oregon State University Institutional Animal Care and Use
Committee (IACUC).
Animals (iTRAQ Experiment)
Six-seven week old male C57BL/6 mice were purchased from Simenson Laboratory
(Gilroy, CA). They were randomly assigned to one of two groups; 1) Control or 2)
High dose (15 mg/kg) CD. They were individually housed in a temperature (22±1ºC)
and light controlled (12 hours light/12 hours dark daily) facility with free access to
water and fed AIN93 diet (Dyets, Inc., Bethlehem, PA) ad libitum. Seven days were
allowed for acclimatization before treatments were initiated. Corn oil was used as a
vehicle (5 mL/kg). Mice received a single dose of either corn oil or CD (5 mg/kg or 15
mg/kg) via intraperitoneal (IP) injection. After 72 hours they were fasted for four
hours and killed by carbon dioxide anesthesia and exsanguination. Liver, gall bladder
and blood were harvested. Here again, the procedures for animal use were approved
by the Oregon State University Institutional Animal Care and Use Committee
(IACUC).
Tissue Preparation
45
Liver was homogenized with a Polytron PT 3000 PT (Brinkmann Instruments,
Westbury CT) in buffer consisting of 0.01 M potassium phosphate, pH 7.4, 0.15 M
potassium chloride (KCL), 1.0 mM ethylenediaminetetraacetic acid (EDTA), 0.1 mM
butylated hydroxytoluene (BHT) and 0.1 mM phenylmethlysulfonyl fluoride (PMSF).
All centrifugation was performed at 4ºC. The homogenate was first centrifuged at
1200 g for 30 minutes. The pellet produced consisted of plasma membrane/cellular
debris. The supernatant was then centrifuged at 12000 g for 30 minutes to eliminate
mitochondria. The supernatant from this run was then centrifuged at 100,000 g for 90
minutes in a Ti 70 rotor. The resulting supernatant and pellet were the cytosolic and
microsomal fractions respectively. Microsomes were resuspended in 0.1 M potassium
phosphate, pH 7.25, 1.0 mM EDTA, 30% glycerol, 20 μM BHT, 1.0 mM
dithiothreitol (DTT) and 0.1 mM PMSF. BCA assay was used to determine protein
content.
Radioactivity Quantification Assessed Fraction Enrichment of CH Binding
Proteins
Subcellular fractions were prepared as described under tissue preparation, but in this
experiment, the mice were injected with [14C] CH. Fractionation of [14C] CH was
applied in order to evaluate the concentration of CH binding proteins. Fifty μL/mL of
30% streptomycin sulfate was added to cold hepatic cytosol and gently shaken
overnight at 4ºC. The resulting precipitate was dissolved in 8.0 M urea, 10 mM
potassium phosphate, pH 7.4 and dialyzed overnight in 1.0 M urea, 10 mM potassium
phosphate, pH 7.4. Aliquots of the streptomycin precipitated cytosolic fractions
46
(supernatant and solubilized precipitate) were added to 10 mL of Hionic-Fluor
(Perkin-Elmer, Boston, MA). The samples were mixed and then placed in the dark
overnight to allow the chemiluminescence to decay. Samples were counted in a
Beckman LS7500 liquid scintillation counter. The BCA assay was used to determine
protein concentrations.
Fluorescent Spectroscopy
The fluorescent analog NBD CH (22-(N-(7-nitrobenz-2-oxa-1,3- diazol-4-yl)amino)23,24-bisnor-5- cholen-3 -ol) was purchased from Invitrogen/molecular Probes
(Carlsbad, CA). Experiments were performed on a Perkin Elmer LS50 luminescence
spectrometer (Waltham, MA).
Tissue Preparation for iTRAQ Experiments
CH enriched fractions were prepared by streptomycin precipitation as outlined above.
BCA assay was used to determined protein concentrations. 100-μg samples of the
streptomycin precipitate fraction were precipitated with six volumes of ice cold
acetone and labeled with iTRAQ reagents as detailed below.
iTRAQ Labeling
Following the manufacturer’s protocol, samples were labeled with iTRAQ reagents
according to the scheme shown in Figure 3-1. Briefly, 100-μg samples were dissolved
in 20 μL of 0.5 M tetraethyl ammonium bicarbonate (TEAB), denatured with 0.1%
47
SDS and reduced with 5 mM tris-(2-carboxyethyl) phosphine (TCEP). After a one
hour incubation at 60ºC, cysteines were blocked with 10 mM methyl methane–
thiosulfonate (MMTS) in isopropanol, and the samples were incubated at room
temperature for 10 minutes. Promega (Madison, WI) sequencing grade trypsin was
added in a trypsin:protein ratio of 1:20, and the samples were incubated at 37ºC for 16
hours. The samples were vacuum concentrated, resolubilized in 30 μL TEAB and
labeled with 70 μl iTRAQ reagent in isopropanol. Control samples were labeled with
iTRAQ 114.1 or 116.1, while CD treated samples (15 mg/kg) were labeled with
iTRAQ 115.1 or 117.1 and incubated at room temperature for 2 hours. One control
mouse and 1 treated mouse composed one biological sample. Four samples, i.e., a total
of eight mice, were prepared. Control and CD treated samples were randomly
combined and vacuum concentrated to ~30 μL. Samples were prepared for analysis by
chromatography on strong cation exchange (SCX) spin columns (The Nest Group, Inc.,
Southboro, MA) and eluted with four salt fractions; 75 mM, 150 mM, 350 mM and
750 mM KCl in 25% acetonitrile, 10 mM potassium phosphate monobasic (KH2PO4),
pH 3.0. The eluted fractions were vacuum concentrated to ~30 μL, and 300 μL of
0.1% trifluoroacetic acid (TFA) was added. Samples were desalted on C-18 spin
columns (The Nest Group), eluted with 80% acetonitrile, 0.1% TFA and vacuum
concentrated to dryness. Samples were resuspended in 20 μL 0.1% TFA before
performing nano HPLC mass spectrometry.
HPLC-Electrospray Ionization (ESI)-MS/MS
48
Nano HPLC ESI-MS/MS was performed on a nanoAcquity Ultra Performance LC
(Waters, Milford, MA) coupled to a quadrapole orthogonal time-of-flight mass
spectrometer (QTOF Ultima Global, Micromass/Waters, Manchester UK). Four-μL
samples were injected. Peptides were trapped and washed on a Symmetry C18 column
(5 μm, 180 μm × 20 mm, Waters) using 2% acetonitrile in 0.1% formic acid at a flow
rate of 4 μL/minute for 6 minutes. Peptides were separated using an ethylene bridged
hybrid (BEH) column (1.7 μm, 75 μm × 150 mm, Waters) with a binary gradient
consisting of 2% acetonitrile in 0.1% forming acid (Solvent A) and acetonitrile in
0.1% formic acid (Solvent B) over 70 minutes at a flow rate of 260 nL/minute.
The ion source was operated with a spray voltage of 2.2 kV in the positive ion mode.
Data dependent analysis (DDA) MS/MS mode was utilized with a 0.6 sec survey scan
and a 2.4 sec MS/MS scan on the base peak in the MS survey scan. Previously
selected m/z values were excluded from DDA for 60 sec. The MS/MS collision energy
ranged from 25-65 eV and was based on the precursor charge state selected by the
quadrapole. No effort was made to optimize the transmission of low m/z ions through
the instrument, i.e., the instrument was run in its standard operating mode. Calibration
was performed with [Glu1]-fibrinopeptide B (MH+ 1570.6774 Da, monoisotopic)
fragment ions. Lock mass correction was also performed using [Glu1]-fibrinopeptide B
([M+2H]2+ 785.8426 Th) every 30 sec for one scan. Peak list (pkl) files were
generated using Protein Lynx Global Server (PLGS, Waters, Manchester, UK).
HPLC-Matrix Assisted Laser Desorption Ionization (MALDI)-MS/MS
49
Peptide separation was performed on a on a nanoAcquity Ultra Performance LC
(Waters, Milford, MA) coupled to a Probot (Dionex/LC Packings, Amsterdam, The
Netherlands) MALDI spotting device. Two-μL samples were injected. Peptides were
trapped and washed on a Symmetry C18 column (5 μm, 180 μm × 20 mm, Waters)
using 0.1% TFA at a flow rate of 4 μl/minute for 6 minutes. Peptides were separated
using an Atlantis C18 column (3 μm, 75 μm × 150 mm, Waters) with a binary
gradient consisting of 0.1% TFA (Solvent A) and acetonitrile in 0.1% TFA (Solvent B)
over 60 minutes at a flow rate of 300 nl/minute.
Column effluent was mixed in a 1:4 ratio with MALDI matrix (3 mg/mL α-cyano-4hydroxycinnamic acid in water/acetonitrile/TFA/ammonium phosphate dibasic (6
mg/mL), 37.6/56.4/0.1/6, vol:vol) via a tee-junction. Fractions (20 s/spot, 144
spots/plate) were spotted onto an Opti-TOF MALDI plate system (ABI) using
stainless steel plates.
MALDI plates were analyzed on an ABI 4700 Proteomics Analyzer equipped with a
200 Hz Nd:YAG laser operating at a wavelength of 355 nm. Data were acquired in the
MALDI reflector mode over a mass range of 800-4000 m/z using five spots of ABI
calibration mixture as an external standard. ABI calibration mixture consists of the
following peptides (monoisotopic mass of the singly protonated ion is given in
parenthesis in Da); [Glu1]-fibrinopeptide B (1570.6774), adrenocorticotropic hormone
fragment 18-39 (2465.1989), angiotensin I (1296.6853) and [des-Arg1]-bradykinin
(904.6774). Mass spectra were obtained from each spot using 1000 laser shots per
50
spectrum. Peaks with a signal:noise (S/N) ratio greater than 30 were selected for
MS/MS. Up to 10 MS/MS spectra could be obtained from each spot. MS/MS spectra
were acquired with 1500 laser shots, starting with the least intense peak and ending
with the most intense peak. Constant laser fluence was used and kept low enough to
prevent signal saturation. Ions formed in the source were accelerated to 8 keV.
Precursors were selected using timed ion selection (resolving power = 100 full width
half maximum [FWHM]) and then decelerated in the collision cell to 1 keV for CID.
Air was used as the collision gas at a pressure of about 5.5 × 10-7 torr. Fragment ions
were accelerated to 14 keV before entering the reflector.
Data Processing and Database Searching
MALDI-MS/MS data were processed with GPS Explorer (Applied Biosystems Inc.,
Framington MA) to create Mascot searchable files. MS/MS peak list files for the four
salt fractions, which constituted one biological sample pair (control and treated), were
combined using Mascot Daemon (Matrix Science, UK) and searched against the
mammalian Swiss-Prot database (V55.0) using the Mascot (Matrix Science, UK)
search engine. β–Methylthiolation [methylthiocysteine] was set as a fixed peptide
modification. Variable peptide modifications included methionine (M) oxidation and
iTRAQ labeling of the N-terminal, lysine (K) and tyrosine (Y). One missed cleavage
site was selected with Trypsin/P as the digestion enzyme. ESI-MS/MS precursor and
fragment ion mass tolerances were set to ±0.1 Da and ±0.12 Da respectively; MALDIMS/MS precursor and fragment ion mass tolerances were set to ±0.2 Da and ±0. 2 Da
respectively.
51
iTRAQ ratios were generated in Mascot and median normalized. Tyrosine labeled data
was excluded from the iTRAQ ratio calculations. Data sets were further analyzed with
Scaffold (Proteome Softare, Portland, OR). Only peptides with a probability of 80% or
greater as determined by Scaffold were utilized. In the case of duplicate peptides or
different charge states, the iTRAQ ratio with the highest Mascot score was used. A
minimum of 2 peptides was required for protein identification.
Statistical Methods
An iTRAQ sample consisted of 100 μg of protein from a control animal pooled with
100 μg of protein from a CD treated animal. One animal pair was used for each
iTRAQ sample, and four pairs in all were analyzed. Each animal pair was analyzed
twice on each instrument to generate technical replicates A and B (Figure 3-2A). The
iTRAQ ratios from the A and B replicates were averaged for every protein with 2 or
more peptides detected in each replicate. If two or more peptides of a protein were
detected in only one replicate, that iTRAQ ratio was used in subsequent calculations.
Mean iTRAQ protein ratios were generated for each biological pair on the basis of
each instrument alone and the average ratios of pairs from both instruments; these
mean ratios were then averaged across the biological sample pairs (Figure 3-2B).
Statistical analyses were conducted with either StatGraphics (Statistical Graphics
Corp., Herndon, VA) or S-Plus (Insightful Corp., Seattle, WA). Normality of data was
accessed based on standardized kurtosis, standardized skewness and homogeneity of
52
variance. Data for iTRAQ ratios was analyzed in logarithm (natural), ln, space. Means
and standard deviations are geometric. The Student’s t-test was used to access whether
the mean iTRAQ ratio for a particular protein was different than 1 (0 in ln space), both
within a biological sample pair and across the four biological sample pairs. Two-way
ANOVA was used to compare the effect of the factors sample (pair) and instrument
(ESI QTOF, MALDI TOF/TOF) on the dependent variable, the ln(iTRAQ ratio). In all
cases, a 95% confidence level was the criterion for significance. Grubbs' Test
(assumes normality) was applied for outlier identification.
RESULTS AND DISCUSSION
NBD CH showed little change (<10%) in emission characteristics (500 – 600 nm)
over a range of streptomycin concentrations ranging from 0.05 μM – 15 μM in
methanol (data not shown). This indicated little interaction between CH and
streptomycin in this solvent.
Mass spectrometry has a limited dynamic range, and the cytosolic proteome is
complex. Enrichment of the fraction of interest was explored as a means to simplify
matters and make the mass spectrometry more tenable. Original enrichment
experiments were performed with [14C]CH as described under methods. Streptomycin
precipitation of cytosol proved to be effective in concentrating the label while at the
same time decreasing the amount of protein present (Figure 3-3). Clearly the label is
enriched in the precipitate (Figure 3-3A), most of the protein remains in the
53
supernatant (Figure 3-3B), thus reducing the number of proteins that need to be
detected by mass spectrometry.
Approximately 75 proteins were detected with the ESI QTOF and 40 with the MALDI
TOF/TOF under the conditions described in the experimental section (data not shown).
The observed difference between the two techniques was probably due to the
differences in HPLC methodology. The overall relatively low number of proteins
detected probably was due to the limited number (4) of isocratic KCl elutions from the
SCX spin columns.
Wiese et al., (2007) showed that higher collision energies were needed for efficient
liberation of iTRAQ reporter ions from lysine residues on multiply charged peptides.
Peptide VFIEDVSK was chosen to illustrate this effect (Figure 3-4). Their group also
reported a loss of peptide sequence information under the same conditions. They
reported good iTRAQ quantification with single charged peptides and MALDI
MS/MS analysis. Figure 3-4 shows the product ion mass spectra of the iTRAQ labeled
peptide VFIEDVSK generated via ESI (Figure 3-4A) and MALDI (Figure 3-4B). The
higher collision energy of the MALDI instrument resulted in the 115.1 m/z reporter
iTRAQ ion being the base peak. The lower collision energy of the ESI instrument
coupled with the fact that the transmission of low m/z ions was not optimized through
the instrument resulted in iTRAQ reporter ion intensities well below the base peak (in
this case the y2 ion). Both spectra show extensive b and y ion ladders that indicated
efficient fragmentation. Intact iTRAQ ions resulting from the cleavage of the amide
54
bond between the label and the primary amine group without neutral loss of the
carbonyl moiety were observed at m/z 145.1.
The question of whether or not iTRAQ data generated using multiply charged peptides
and lower collision energies (ESI QTOF) would be comparable to that of singly
charged peptides and higher collision energies (MALDI TOF/TOF) (Wiese et al.,
2007). was addressed in the present study by performing two-way ANOVA on
proteins that were detected in all four biological sample pairs and with both
instruments. Specifically, ANOVA was used to determine the extent to which factors
sample and instrument affected the ln(iTRAQ ratio). Proteins detected in all four
animal pairs and with both instruments are listed in Table 3-1. The mean and standard
deviation (SD) of the iTRAQ ratio for each protein in each pair are geometric and
were generated from the two technical replicates generated for each pair (Figure 3-2A).
The standard deviations could clearly be improved with more technical replicates;
however, cost considerations precluded generation of larger data sets for this study.
Table 3-2 shows the mean iTRAQ values and the 95% confidence intervals after the
QTOF and TOF/TOF data were combined and then averaged across all four pairs.
Myosin 9 was the only protein with an observed mean iTRAQ ratio significantly
different than 1. Its possible role in vesicular intracellular transport (van der Boom et
al., 2007) will be addressed in future work. Table 3-3 shows the results of performing
ANOVA on the data presented in Table 3-1; the p-values associated with each factor
for each protein are given at the 95% confidence level. Clearly, the sample has a
significant effect on the ln(iTRAQ) ratio for more than half of the proteins. This could
55
be due to the biological variability inherent in the samples themselves. In contrast, the
p-values for the instrument factor are greater than 0.05 in all cases, indicating that the
instruments had no significant effect on the ln(iTRAQ ratio).
It appears, therefore, that iTRAQ analyses performed on an ESI TOF without any
special modifications of the instrumental parameters produce essentially the same
protein ratios as obtained on a MALDI TOF/TOF. Although this observation needs to
be tested with a larger data set produced with other instruments, it indicates that it
should be possible in general to compare iTRAQ data obtained from these two types
of instruments
56
Control (corn oil)
Streptomycin
Precipitate (100 μg)
( 1 Animal Pair)
Treated (15 mg/kg CD)
Streptomycin
Precipitate (100 μg)
Reduction
Alkylation
Trypsin Digestion
Reduction
Alkylation
Trypsin Digestion
Label with 114 or 116
iTRAQ
Label with 115 or 117
iTRAQ
Pool Samples
SCX Spin Columns
- 4 KCl fractions, 75 mM, 150 mM, 350 mM and 700 mM
C18 HPLC followed by MALDI MS/MS and
ESI MS/MS.
Data Analysis by GPS Explorer, Mascot and Scaffold
Figure 3-1. Flow chart showing the sequence of iTRAQ labeling and mass
spectrometry analysis. A total of four animal pairs were analyzed.
57
A
Pair 1
QTOF
A
B
TOFTOF
B
A
B
Mean Protein iTRAQ Ratio
Pair 1
Pair 2
Pair 3
Pair 4
QTOF
QTOF
QTOF
QTOF
Mean Protein iTRAQ Ratio
Pair 1
Pair 2
Pair 3
Pair 4
TOFTOF
TOFTOF
TOFTOF
TOFTOF
Mean Protein iTRAQ Ratio
Pair 1
QTOF
TOFTOF
Pair 2
QTOF
TOFTOF
Pair 3
QTOF
TOFTOF
Pair 4
QTOF
TOFTOF
Figure 3-2. Flow chart showing how iTRAQ samples were analyzed: (A)
breakdown for one biological sample: (B) three procedures used for averaging the
iTRAQ protein ratios over all four biological samples.
[14C]CH Equivalents (pmol/mg Protein)
58
120
100
80
60
40
20
0
Control
100
Percent Protein Recovery
A
Supernatant
Precipitate
90
Chlordecone
(5 mg/kg)
Chlordecone
(15 mg/kg)
B
Supernatant
Precipitate
80
70
60
50
40
30
20
10
0
Control
Chlordecone
(5 mg/kg)
Chlordecone
(15 mg/kg)
Figure 3-3. Enrichment of [14C] CH and decrease in protein observed in the
supernatant and streptomycin precipitate of hepatic cytosol: (A) CH
equivalents/mg protein versus chlordecone treatment and streptomycin fraction (n=3
pooled samples, bar heights=mean of three liquid scintillation runs) ± SD; (B)
cytosolic protein recovery versus chlordecone treatment and streptomycin fraction n=3
pooled samples, one analysis was done for each pooled sample therefore no error bars
are included).
59
4A
V
y2
378.25
100
b1
(iTRAQ)
b2
b3
b4
b5
b6
b7
FIEDVSK
y7
y5
y6
y4
y3
y2
(iTRAQ)
y1
115.12
115.12
iTRAQ Reporter Ions
% Intensity
y3
y1
477.32
291.22
%
a1
115.12
117.13
117.13
216.18
b
b2
1
244.18
116.13
391.25
145.11
358.18
y5
721.39
(M+2H)2+
b3
476.35
613.38
504.35
b
834.48
749.43
835.48
633.40
506.25
b4
187.11
115
116
117
118
y7
y6
5
748.41
612.85
478.34
292.22
114
y4
592.35
981.61
982.56
b6
1080.67
983.56
0
100
200
300
400
500
600
115.11
b2
(iTRAQ)
b3
900
1000
1100
b4
b5
b6
1200
b7
FIEDVSK
y7
115.11
V
y6
y4
y3
y2
(iTRAQ)
y1
115.11
90
80
iTRAQ Reporter Ions
117.12
117.12
70
113.82
114.13
60
558.6
803.4
m/z
1048.2
1225.96
y7
1080.64
b7
934.49
y6
847.48
b6
981.54
748.42
b4
685.46
719.41
592.37
b3
633.40
378.26
391.27
291.22
313.8
b2
b5
y4
772.42
y3
575.32
0
69.0
216.21
10
y1
145.12
20
72.07
30
118
y2
117.12
40
116
477.34
114
504.36
50
107.25
% Intensity
800
m/z
4B
100
700
1
Figure 3-4. Mass spectra of the iTRAQ labeled peptide VFIEDVSK: (A) spectrum
generated on ESI TOF from doubly charged precursor; (B) spectrum generated on
MALDI TOF/TOF. The fragment ions were labeled according to the nomenclature of
Beimann (1988). The insets in the spectra show enlargements of the reporter ion
region.
60
A
NCBI
Accession
Number
P60710
P00329
Q91Y97
Q61176
P16460
O35490
P12710
P16858
P30115
P10649
P19157
P01942
Q8VDD5
P35700
P08228
P68369
P68372
B
NCBI
Accession
Number
P60710
P00329
Q91Y97
Q61176
P16460
O35490
P12710
P16858
P30115
P10649
P19157
P01942
Q8VDD5
P35700
P08228
P68369
P68372
QTOF Data
Protein
Actin, cytoplasmic 1
Alcohol dehydrogenase 1
Fructose-bisphosphate aldolase B
Arginase-1
Argininosuccinate synthase
Betaine--homocysteine S-methyltransferase 1
Fatty acid-binding protein, liver
Glyceraldehyde-3-phosphate dehydrogenase
Glutathione S-transferase A3
Glutathione S-transferase Mu 1
Glutathione S-transferase P 1
Hemoglobin subunit alpha
Myosin-9
Peroxiredoxin-1
Superoxide dismutase [Cu-Zn]
Tubulin alpha-1A chain
Tubulin beta chain
Pair 1
Pair 2
Pair 3
Pair 4
Mean
Mean
Mean
Mean
iTRAQ
iTRAQ
iTRAQ
iTRAQ
Ratio SD Ratio SD Ratio SD Ratio SD
0.73 1.01 0.47 1.27 1.28 1.17 0.99 1.38
0.64 1.13 0.62 1.25 1.14 1.23 0.92 1.23
0.62 1.43 0.76 1.25 0.68 1.06 1.00 1.17
0.81 1.10 1.15 1.34 0.79 1.15 0.80 1.02
1.37 1.01 1.25 1.49 0.99 1.22 1.08 1.41
1.20 1.02 0.96 1.44 0.93 1.26 1.08 1.11
0.66 1.01 1.35 1.48 0.73 1.33 1.05 1.03
0.82 1.03 1.06 1.03 0.93 1.03 0.96 1.25
1.05 1.08 1.13 1.17 0.71 1.52 0.70 1.20
1.23 1.05 1.00 1.89 0.65 1.24 0.72 1.62
1.24 1.16 1.71 1.08 0.62 1.37 0.74 1.51
0.83 1.07 1.35 1.15 1.33 1.17 1.44 1.19
1.59 1.10 1.32 1.09 1.12 1.12 1.37 1.05
0.70 1.34 1.03 1.07 0.88 1.06 1.06 1.12
0.66 1.04 1.14 1.10 0.74 1.30 1.52 1.92
0.76 1.98 0.48 1.25 1.61 1.13 1.07 1.37
0.92 1.19 0.46 1.31 1.87 1.07 1.01 1.22
TOF/TOF Data
Protein
Actin, cytoplasmic 1
Alcohol dehydrogenase 1
Fructose-bisphosphate aldolase B
Arginase-1
Argininosuccinate synthase
Betaine--homocysteine S-methyltransferase 1
Fatty acid-binding protein, liver
Glyceraldehyde-3-phosphate dehydrogenase
Glutathione S-transferase A3
Glutathione S-transferase Mu 1
Glutathione S-transferase P 1
Hemoglobin subunit alpha
Myosin-9
Peroxiredoxin-1
Superoxide dismutase [Cu-Zn]
Tubulin alpha-1A chain
Tubulin beta chain
Pair 1
Pair 2
Pair 3
Pair 4
Mean
Mean
Mean
Mean
iTRAQ
iTRAQ
iTRAQ
iTRAQ
Ratio SD Ratio SD Ratio SD Ratio SD
0.88 1.07 0.62 1.02 1.22 1.02 1.10 1.05
0.77 1.17 0.60 1.01 1.18 1.09 1.26 1.29
0.94 1.03 0.77 1.04 0.83 1.02 1.03 1.03
0.91 1.38 1.11 1.21 0.85 1.15 0.96 1.34
1.36 1.21 1.46 1.11 0.96 1.10 1.02 1.18
1.26 1.10 1.05 1.22 0.83 1.06 1.18 1.08
0.78 1.04 0.96 1.01 0.81 1.51 0.95 1.05
0.97 1.03 1.13 1.18 0.86 1.02 0.93 1.03
0.94 1.17 0.87 1.15 0.75 1.11 0.77 1.07
1.09 1.12 1.00 1.24 0.92 1.09 0.82 1.01
1.04 1.30 1.27 1.25 0.64 1.05 0.68 1.10
0.89 1.06 1.40 1.12 1.20 1.10 1.44 1.04
1.71 1.34 1.21 1.04 1.19 1.13 1.29 1.01
1.04 1.02 0.68 1.02 0.86 1.02 1.10 1.07
0.78 1.15 1.06 1.49 0.98 1.55 1.00 1.01
0.84 1.31 0.44 1.46 1.27 1.10 1.09 1.11
0.92 1.19 0.49 1.06 1.52 1.00 0.89 1.05
Table 3-1. Proteins detected in all four animal pairs and with both instruments:
means and standard deviations (SDs) are geometric and were generated from the two
technical replicates run for each pair; (A) data generated by ESI QTOF; (B) data
generated by MALDI TOF/TOF.
61
QTOF + TOFTOF
NCBI Accession
Number
P60710
P00329
Q91Y97
Q61176
P16460
O35490
P12710
P16858
P30115
P10649
P19157
P01942
Q8VDD5
P35700
P08228
P68369
P68372
Protein
Actin, cytoplasmic 1
Alcohol dehydrogenase 1
Fructose-bisphosphate aldolase B
Arginase-1
Argininosuccinate synthase
Betaine--homocysteine S-methyltransferase 1
Fatty acid-binding protein, liver
Glyceraldehyde-3-phosphate dehydrogenase
Glutathione S-transferase A3
Glutathione S-transferase Mu 1
Glutathione S-transferase P 1
Hemoglobin subunit alpha
Myosin-9
Peroxiredoxin-1
Superoxide dismutase [Cu-Zn]
Tubulin alpha-1A chain
Tubulin beta chain
n
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Confidence
Mean Interval (95%)
0.87
0.48-1.56
0.86
0.52-1.42
0.83
0.66-1.04
0.91
0.72-1.15
1.17
0.89-1.54
1.05
0.83-1.33
0.89
0.63-1.24
0.95
0.82-1.11
0.85
0.64-1.13
0.91
0.66-1.25
0.93
0.49-1.75
1.20
0.84-1.72
1.34
1.05-1.70
0.90
0.77-0.99
0.96
0.65-1.41
0.87
0.40-1.88
0.92
0.40-2.08
p
0.490
0.397
0.075
0.295
0.164
0.533
0.334
0.401
0.167
0.421
0.741
0.205
0.032
0.044
0.738
0.600
0.754
Table 3-2. Mean iTRAQ values and the 95% confidence interval after the
instrument data were combined and then averaged across all four pairs. Ratios
significantly different than one are shown in bold font and (n) is the number of sample
pairs used in the calculation of the geometric mean.
62
ANOVA (Multifactor)
NCBI Accession
Number
P60710
P00329
Q91Y97
Q61176
P16460
O35490
P12710
P16858
P30115
P10649
P19157
P01942
Q8VDD5
P35700
P08228
P68369
P68372
Protein
Actin, cytoplasmic 1
Alcohol dehydrogenase 1
Fructose-bisphosphate aldolase B
Arginase-1
Argininosuccinate synthase
Betaine--homocysteine S-methyltransferase 1
Fatty acid-binding protein, liver
Glyceraldehyde-3-phosphate dehydrogenase
Glutathione S-transferase A3
Glutathione S-transferase Mu 1
Glutathione S-transferase P 1
Hemoglobin subunit alpha
Myosin-9
Peroxiredoxin-1
Superoxide dismutase [Cu-Zn]
Tubulin alpha-1A chain
Tubulin beta chain
p -Value
(Sample)
0.011*
0.022*
0.073
0.049*
0.030*
0.025*
0.199
0.195
0.120
0.152
0.009*
0.006*
0.019*
0.863
0.239
0.006*
0.003*
p -Value
(Instrument)
0.162
0.261
0.060
0.166
0.747
0.953
0.715
0.629
0.550
0.443
0.157
0.994
0.905
0.707
0.950
0.556
0.355
Table 3-3. P-values associated with the factors sample and instrument at the 95%
confidence level generated from the ANOVA analysis on the data shown in Table
3-1. The asterisks indicate p-values less than 0.050.
63
Chapter 4
Conclusions
The principle toxicological finding of this study is that CD pretreatment
increased the SR-BII protein content of hepatic microsomes but not crude membrane
fractions. This increase in microsomal SR-BII may represent an increase in transport
capacity for the hepatocytes to handle CH. If this is the case, then CD pretreated
animals provided with an exogenous lipid bolus of CH should be able to transport CH
through the liver and into the bile more effectively than control animals. The
mechanics of this transport remains illusive. SLU with SR-BI occurs without
endocytosis and SR-BII is twice as efficient as SR-BI when it comes to CH efflux
(Nieland et al., 2005; Mulcahy et al., 2004). Additionally, as mentioned previously,
SR-BII is endocytosed much more rapidly than SR-BI. The fact that endocytosis is not
needed for SLU excretion into bile provides strong evidence for the involvement of a
non-vesicular CH transport system. There was no evidence for a role of cytosolic
binding proteins although additional mass spectrometry experiments are needed to
adequately address this issue. No changes in plasma membrane SR-BII were observed
but it was not possible to distinguish between basolateral and apical membranes.
Therefore, if a differential increase and corresponding decrease on one or the other of
these membranes is occurring, it would be unobservable. If the number of SR-BII
transporter proteins on the cell surface remains constant with CD pretreatment, the
increase in microsomal SR-BII may indicate an increased flux between the SR-BII
transporters on the cell surface and intracellular membrane. The specific mechanism
of SLU has not been elucidated, but it appears to be diffusion driven between the
64
lipoprotein and the plasmas membrane. After the action of hepatic lipase on a
lipoprotein, e.g., HDL, the phospholipids on the HDL surface will be broken down to
release fatty acids. This would make the CH in that layer less stable and, thereby,
easier to transport down the concentration gradient to the plasma membrane of the cell.
At this point the CH is moved across the cell by some still unknown mechanism
presumably involving non-vesicular transport mechanisms and excreted into the bile.
The efficiency of transfer of CH from the HDL to the plasma membrane would
depend on how long the concentration gradient near the transporter could be
maintained. Each transporter would be expected to be associated with a specific, albeit,
transient lipid microdomain. Perhaps rotating transporters out of the plasma membrane
rapidly would allow for more efficient CH transfer. The data from this study supports
a model wherein SR-BII could contribute to CD induced stimulation of biliary
excretion of CH.
A statistically significant increase in the crude membrane content of
Na+/K+ATPase was also observed. Since CD is known to inhibit membrane Na+/K+
ATPase, the observed increase may reflect a compensatory response.
Proteomic analysis of a CH enriched fraction of hepatic cytosol detected
significant changes in four proteins. Increased myosin-9 was consistent with an
increased capacity for vesicular transport. Decreases in SBP-1, α-enolase, and PRX1in a CH enriched fraction of cytosol was observed but how this fits into CD induced
perturbations in CH homeostasis is not clear.
The proteomic analysis detected about 80 proteins overall (data not shown).
Although both cytosolic sterol carriers L-FABP and SCP-2 were detected, SCP-2 did
65
not meet the minimum criteria for protein identification. However, many soluble sterol
carrier proteins were not detected. Approximately 80 proteins were present in the
streptomycin enriched fraction; using more effective separation techniques should
allow the detection of many more proteins. Simply increasing the number of salt cuts
used in the first chromatographic dimension by a factor of 10 would be a good first
step. A more thorough examination of the cytosolic proteome is required before the
role of cytosolic transport proteins can be definitively dismissed.
Overall these results may at least partially explain the CD pretreatment induced
alterations in CH homeostasis.
The principle bioanalytical finding of this study is that iTRAQ analyses
performed on an ESI TOF without any special modifications of the instrumental
parameters produce essentially the same protein ratios as obtained on a MALDI
TOF/TOF. With the caveat that this observation needs to be tested with a larger data
set produced with other instruments, this observation indicates that it should be
possible in general to compare iTRAQ data obtained from these two types of
instruments.
66
Bibliography
Aebersold, R. and Mann. M., 2003, Mass spectrometry-based proteomics. Nature, 422,
198-207.
Aggarwal, K., Choe, L.H. and Lee, K.L., 2005, Quantitative analysis of protein
expression using amine-specific isobaric tags in Escherichia coli cells expressing rhsA
elements, Proteomics 5, 2297-2308.
Ahras, M., Naing, T. and McPherson, R., 2008, Scavenger receptor-BI localizes to a
late endosomal compartment. J. Lipid Res. Mar 29, [Epub ahead of print].
Applied Biosystems Inc., Version 3.6, Framington, MA.
ATSDR, Agency for Toxic Substances and Disease Registry, Toxicological Profile for
Mirex and Chlordecone, (Final Report), Atlanta, GA:ATSDR, Public Health Service,
U.S. Department of Health and Human Services, 1995, NTIS Accession No. PB95264354.
Attie, A.A., 2007, ABCA1:at the nexus of cholesterol, HDL and atherosclerosis.
Trends Biochem. Sci. 32, 172-179.
Bachmann, K., Patel, H., Batayneh, Z., Slama, J., White, D., Posey, J., Ekins, S., Gold,
D., and Sambucetti, L., 2004, PXR and the regulation of apoA1 and HDL-cholesterol
in rodents. Pharmacol Res. 50, 237-246.
Biemann, K., 1988, Contributions of mass spectrometry to peptide and protein
structure. Biomed. Environ. Mass Spectrom. 16, 99-111.
Bocquene, G. and Franco, A., 2006, Pesticide contamination of the coastline of
Martinique, Environ Toxicol Chem. 25(1):112-9.
Brown, D.A. and London, E, 2000, Structure and Function of Sphingolipid and
Cholesterol Rich Membrane Rafts, J. Biol. Chem. 275, 17221-17224.
Brown, M. S., and Goldstein, J. L., 1986, A receptor-mediated pathway for cholesterol
homeostasis. Science 232, 34-47
67
Brown, M.S. and Goldstein, J.L., 1997, The SREBP pathway: regulation of
cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell,
89, 331-340.
Brown, D.A. and London, E, 2000, Structure and Function of Sphingolipid and
Cholesterol Rich Membrane Rafts, J. Biol. Chem. 275, 17221-17224.
Bocquene, G. and Franco, A., 2005, Pesticide Contamination of the Coastline of
Martinique, Mar. Pollut. Bull. 51, 612-619.
Cannon, C.B., Veazey, J.M., Jr., Jackson, R.S., Burse, V.W., Hayes, C., Straub, W.E.,
Landrigan, P.J. and Liddle, J.A., 1978, Epidemic kepone poisoning in chemical
workers. Am. J. Epidemiol. 107, 529-537.
Carpenter, H.M., and Curtis, L.R., 1989, A characterization of chlordecone
pretreatment-altered pharmacokinetics in mice. Drug Met Disp. 17, 131-138.
Carpenter, H.M., and Curtis, L.R., 1991, Low dose chlordecone pretreatment altered
cholesterol disposition without induction of cytochrome P-450. Drug Met Disp. 19,
673-678.
Chalwa, A., Repa, J.J., Evans, R.M. and Mangelsdorf, D. J., 2001, Nuclear Receptors
and Lipid Physiology:Opening the X-Files, Science, 294, 1866-1870.
Chen, E.I., Hewel, J., Felding-Habermann, B. and Yates J.R., 2006, Large scale
protein profiling by combination of protein fractionation and multidimensional protein
identification technology (MudPIT), Mol. Cell Prot. 5, 53-56.
Christensen, J.B., Macduffee, M., Macdonald, R.W., Whitcar, M. and Ross, P.S., 2005,
Persistent organic pollutants in British Columbia grizzly bears. Consequence of
divergent diets. Environ. Sci. Technol. 39, 6952-6960.
Cohn, W.J., Boylan, J.J., Blanke, R.V., Fariss, M.W., Howell, J.R. and Guzelian, P.S.,
1978, Treatment of chlordecone (Kepone) toxicity with cholestyramine: results of a
controlled clinical trail. N. Engl. J. Med. 298, 243-248.
68
Coumoul, X., Diry, M. and Barouki, R., 2002, PXR-dependent induction of human
CYP3A4 gene expression by organochlorine pesticides. Biochem. Pharmacol. 64,
1513-1519.
Desaiah, D., 1982, Biochemical Mechanisms of Chlordecone Toxicity: A Review,
Neurotoxicology 3, 103-110.
DeSouza, L, Diehl, G., Rodrigues, M.J., Guo, J., Romaschin, A.D., Colgan, T.J. and
Siu. K.W.M., 2006, Search for cancer markers from endometrial tissues using
differentially labeled tags iTRAQ and cICAT with multidimensional liquid
chromatography and tandem mass spectrometry, J. Proteome Res. 4, 377-386.
Donohoe, R.M., Zhang, Q., Siddens, L.K., Carpenter, H.M., Hendricks, J.D. and
Curtis, L.R., 1998, Modulation of 7,12-dimethylbenz[a]anthracene disposition and
hepatocarcinogenesis by dieldrin and chlordecone in rainbow trout. J.
Toxicol.Enriron.Health, Part A, 54, 227-242.
Eckhardt, E.R.M, Cai, L., Sun, S., Webb, N.R. and van der Westhuyzen, D.R., 2004,
High density lipoprotein uptake by scavenger receptor SR-BII, J. Biol. Chem. 279,
14372-14381.
Eckhardt, E. R. M., Cai, L., Shetty, S., Zhao, Z., Szanto, A., Webb, N.R. and van der
Westhuyzen, D.R., 2006, High Density Lipoprotein Endocytosis by Scavenger
Receptor SR-BII Is Clathrin-dependent and Requires a Carboxyl-terminal Dileucine
Motif, J. Biol. Chem., 281, 4348-4353.
Egle, J.L., Fernandez, J.B., Guzelian, P.S and Borzelleca, J.F., 1978, Distribution and
excretion of chlordecone (Kepone) in the rat. Drug Met Disp. 6, 91-95.
Egler, R.A., Fernandes, E., Rothermund, K., Sereika, S., de Sousa-Pinto, N., Jaruga, P.,
Dizdaroglu, M. and Prochownik, 2005, Regulation of reactive oxygen species, DNA
damage, and c-Myc function by peroxiredoxin 1, Oncogene, 24, 8038-8050.
Fernandez, P.C., Frank, S.R., Wang, L., Schroeder, M., Liu, S., Greene, J., Cocito, A
and Amati, B., 2003, Genomic targets of human c-Myc protein, Genes Dev. 17, 11151129.
69
Gilroy, D.J., Carpenter, H.M. and Curtis L.R., 1994, Chlordecone pretreatment alters
[14C]chlordecone and [14C]cholesterol transport kinetics in the perfused rat liver. Fund.
Appl. Toxicol. 22, 286-292.
Glen, A., Gan, C.S., Hamdy, F.C., Eaton, C.L., Cross, S.S., Catto, J.W.F., Wright, P.C.
and Rehman, I., 2008, iTRAQ-facilitated proteomic analysis of human prostate cancer
cells identifies proteins associated with progression, J. of Proteome Res. 7, 897-907.
Geuken, E., Visser, D.S., Leuvenink, H.G.D., de Jong, K.P., Peeters, P.M.J.G., Slooff,
M.J.H., Kiupers, F. and Porte, R.J., 2005, Hepatic expression of ABC transporters G5
and G8 does not correlate with biliary cholesterol secretion in liver transplant patients.
Hepatology, 42, 1166-1174.
Gorg, A., Obermaier, C., Boguth, G., Harder, A., Scheibe, B., Wildgruber, R and
Weiss, W., 2000, The current state of two-dimensional electrophoresis with
immobilized pH gradients, Electrophoresis, 21, 1037-1053.
Guzelian, P.S., 1982, Comparative toxicology of chlordecone (kepone) in humans and
experimental animals. Ann. Rev. Toxicol. 22, 286-292.
Gygi, S.P., Corthals, G.L., Zhang, Y., Rochon, Y. and Aebersold, R., 2000, Evaluation
of two-dimensional gel electrophoresis-based proteome analysis technology. PNAS, 97,
9390-9395.
Gygi, S.P., Rist, B., Gerber, S.A., et. al., 1999, Quantitative analysis of complex
protein mixtures using isotope-coded affinity tags. Nat. Biotechnol., 17, 994-999.
Harder, C.J., Meng, A., Rippstein, P., McBride, H.M. and McPherson, R., 2007, SRBI undergoes cholesterol-stimulated transcytosis to the bile canaliculus in polarized
WIF-B cells, J. Biol. Chem. 282, 1445-1455.
Hu, J., Qian, J., Borisov, O., Pan, S., Li, Y., Liu, T., Deng, L., Wannemacher, K.,
Kurnellas, M., Patterson, C., Elkabes, S., Li, H., 2006, Optimized proteomic analysis
of a mouse model of cerebellar dysfunction using amine-specific isobaric tags.
Proteomics 6, 4321-4334.
70
Huuskonen J, Vishnu M, Chau P, Fielding PE and Fielding CJ, 2006, Liver X receptor
inhibits the synthesis and secretion of apolipoprotein A1 by human liver-derived cells.
Biochemistry, 45(50):15068-74.
Ikonen, E., 2008, Cellular cholesterol trafficking and compartmentalization. Nat. Rev.
Mol. Cell Biol. 9 125-138.
Insightful Corp, Version 8.0, Seattle, WA.
Ishikawa, T., McNeeley, S., Steiner, P.M., Glueck, C.K., Mellies, M., Gartside, P.S.
and McMillin, C., 1978, Effects of chlorinated hydrocarbon on plasma alphalipoprotein in rats. Metabolism, 27, 89-96.
Julka, S. and Regnier, F., 2004, Quantification in proteomics through stable isotope
coding: a review, J. Proteome Res., 3,350-363.
Kalaany, N.Y. and Magelsdorf, D.J., 2006, LXRs and FXR:the yin and yang of
cholesterol and fat metabolism, Annu. Rev. Physiol. 68, 159-191.
Karas, M., Bachmann, D., Bahr, U. and Hillenkamp, F., 1987, Matrix- assisted
ultraviolet laser desorption ionization of non-volitile compounds, Int. J. Mass
spectrum. Ion Proc. 78, 53-68.
Keller, A., Nesvizhskii, A.I., Kolker, E. and Aebersold, R., 2002, Empirical statistical
model used to estimate the accuracy of peptide identifications made by MS/MS and
database search. Anal. Chem. 74, 5383-5392.
Kim, Y.J., Ahn, J.Y, Liang, P., Ip, C., Zhang, Y. and Park, Y.M., 2007, Human prx1
gene is a target for Nrf2 and is upregulated by hypoxia/reoxygenation:implication to
tumor biology, Cancer Res. 67, 546-554.
Kolkman, A, Dirksen, E.H.C., Slijper, M. and Heck, A.J., 2005, Double standards in
quantitative proteomics: direct comparative assessment of difference in gel
electrophoresis and metabolic stable isotope labeling, Mol. Cell Proteomics, 4,255-266.
71
Kosters, A., Kunne, C., Looije, N., Patel, S.B., Oude Elferink, R.P. and Groen, S.K.,
2005, Diosgenin-induced biliary cholesterol secretion in mice requires ABCG8.
Hepatology, 41, 141-150.
Kuipers, F, Stroeve, J.H.M., Caron, S. and Staels, B., 2007, Bile acids, farnesoid X
receptor, atherosclerosis and metabolic control. Curr. Opinion Lipid. 18, 289-297.
Lamon-Fava, S., Ordovas, J. M., and Schaefer, E. J., 1999, Estrogen increases
apolipoprotein (apo) A-I secretion in hep G2 cells by modulating transcription of the
Apo A-I gene promoter. Arterioscler. Thromb. Vasc. Biol. 19, 2960-2965.
Lee, J., 2002, Alteration of cholesterol disposition by chlordecone is not explained by
induction of Cyp7a or Cyp4a1, M.S. Thesis, Oregon State University.
Lee, J., 2007, Chlordecone (CD) A mixed steroid X receptor (SXR) and estrogen
receptor alpha (ERα) agonist, altered hepatic cholesterol (CH) homeostasis and
lipoprotein metabolism, Ph.D. Dissertation, Oregon State University.
Lee, J., Scheri, R.C. and Curtis, L.R., 2008, Chlordecone altered hepatic disposition of
[14C] cholesterol and plasma cholesterol distribution but not SR-BII or ABCG8
proteins in livers of C57BL/6 mice. Toxicol. Appl. Pharmacol., 2008, in Press.
Lewis, G.F. and Rader, D.J., 2005, New insight into the regulation of HDL
metabolism and reverse cholesterol transport. Circ. Res. 96, 1221-1232.
Li, A.C. and Glass, C., 2004, PPAR and LXR dependent pathways controlling lipid
metabolism and the development of atherosclerosis, J. Lipid Res. 45, 2161-2173.
Lopez, D. and McLean, M.P., 2006, Estrogen regulation of the scavenger receptor
class B gene: Anti-atherogenic or steroidogenic, is there a priority? Mol. Cell.
Endocrinol. 247, 22-33
Luellen, D.R., Vadas, G.G. and Unger, M.A., 2006, Kepone in James river fish:19762006, Sci. Total Environ. 358, 286-297.
72
Mardones, P., Quinones, V., Amigo, L., Moreno, M., Miguel, J.F., Schwarz, M,
Miettinen, H.E., Trigatti, B., Krieger, M., VanPatten, S, Cohen, D.E. and Rigotti, A.,
2001, Hepatic cholesterol and bile acid metabolism and intestinal cholesterol
absorption in scavenger receptor class B type I deficient mice. J. Lipid Res. 42, 170180.
Matrix Science, Version 2.2, London, UK.
McKenna, N.J., Lanz, R.B. and O’Malley, B.W., 1999, Nuclear receptor
coregulators:cellular and molecular biology, Endocrine Rev. 20, 321-344.
Michalik, L. and Wahli, W., 1999, Peroxisome proliferator-activated receptors:three
isoforms for a multitude of functions. Curr. Opin. Biotech. 10, 564-570.
Mick, D.L., Long, K.R., Dretchen, J.S. and Bonderman, D.P., 1971, Aldrin and
diledrin in human blood components, Arch. Environ. Health, 23, 177-180.
Mulcahy, J.V., Riddell, D.R. and Owen, J. S., 2004, Human scavenger receptor class
B type II (SR-BII) and cellular cholesterol efflux. Biochem. J. 377, 741-747.
Narahashi, T., Ginsbury, K.S., Nagata, K., Song, J.H. and Tatebayashi, H., 1998, Ion
channels as targets for insecticides. Neurotoxicology 19, 581-590.
Nesvizhskii, A.I. and Aebersold, R., 2004, Analysis, statistical validation and
dissemination of large-scale proteomics datasets generated by tandem MS. Drug Dis.
Today, 9, 173-181.
Nieland, T.J.F., Ehrlich, M., Krieger, M. and Kirchhausen, T., 2005, Endocytosis is
not required for the selective lipid uptake mediated by murine SR-BI. Biochim.
Biophys. Acta. 1734, 44-51.
Nomiyama, T. and Breummer, D., 2008, Liver X receptors as therapeutic targets in
metabolism and atherosclerosis, Curr. Atherosclerosis Rep. 10, 88-95.
73
Ong, S.E., Foster, L.J. and Mann, M., 2003, Mass spectrometric-based approaches in
quantitative proteomics, Methods 29,124-130.
Poland, A. and Knutson, J.C., 1982, 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related
halogenated aromatic hydrocarbons:examination of the mechanism of toxicity. Ann.
Rev. Pharmacol. Toxicol. 22, 517-554.
Porat, P., Sagiv, Y. and Elazar, Z., 2000, A 56-kDa selenium-binding protein
participates in intra-golgi protein transport, J. Biol. Chem.275, 14457-14465.
Prinz, W.A., 2007, Non-vesicular sterol transport in cells. Prog. Lipid Res. 46, 297314.
Proteome Software, Inc., Version 1.6, Portland, OR.
Raloff, J., 1976, The Kepone incident. Chemistry, 49, 20-21.
Reinders, Y., Schulz, I., Graf, R. and Sickmann, A., 2006, Identification of novel
centrosomal proteins in dictyostelium discoideum by comparative proteomic
approaches, J. Proteome Res. 5, 589-598.
Reuber, M.D., 1978, Carcinogenicity of Kepone, J. Toxicol. Environ. Health 4, 895911.
Rhainds, D. and Brissette, L., 2004, The role of scavenger receptor class B type I (SRBI) in lipid trafficking: Defining the rules for lipid traders. Intl. J. Biochem. Cell Biol.
36, 39-77.
Rigotti, A., Miettinen, H.E. and Krieger, M., 2003, The role of the high density
lipoprotein receptor SR-BI in the lipid metabolism of endocrine and other tissues.
Endocrine Rev. 24 357-387.
Robins, S.J. and Fasulo, J.M., 1997, High density lipoproteins, but no other
lipoproteins, provide a vehicle for sterol transport to bile. J. Clin. Invest. 99, 380-384.
74
Ross, P.L., Huang, Y.N., Marchese, J.N., Williamson, B., Parker, K., Hattan, S.,
Khainovski, N., Pillai, S., Dey, S., Daniels, S., Purkayastha, S., Juhasz, P., Martin, S.,
Bartlet-Jones, M., He, F., Jacobson, A. and Pappin. D.J., 2004, Multiplexed protein
quantitation in saccharomyces cerevisiae; using amine-reactive isobaric tagging
reagents, Mol. Cell Proteomics 3,1154-1169.
Roth, A., Looser, R., Kaufmann, M., Blaettler, S., Rencurel, F., Huang, W, Moore,
D.D. and Meyer, Y.A., 2008, Regulatory crosstalk between drug metabolism and lipid
homeostasis: CAR and PXR increase Insg-1 expression, Mol Pharmacol. 2008 73,
1282-1289.
Scheri, R.C., Junga, L., Barofsky, D.F. and Curtis, L.R, A comparison of relative
quantification with isobaric tags on a subset of the murine hepatic proteome using ESI
QTOF and MALDI TOF/TOF. Submitted, Rapid Comm. Mass Spec., May 2008.
Schmidt, A., Kellerman, J. and Lottspeich, F., 2005,A novel strategy for quantitative
proteomics using isotope-coded protein labels., Proteomics 5,4-15.
Smith, S.A., 2002, Peroxisome proliferator-activated receptors and the regulation of
mammalian lipid metabolism. Biochem. Soc. Trans. 30, 1086-1090.
Soine, P.J., Blanke, R.V. and Schwartz, C.C., 1982, Preferential binding of
chlordecone to the protein and high density lipoprotein fractions of plasma from
humans and other species. J. Toxicol. Environ. Health, 14, 319-355.
Sporstol, M., Tapia, G., Malerod, L., Mousavi, S. A., and Berg, T., 2005, Pregnane X
receptor-agonists down-regulate hepatic ATP-binding cassette transporter A1 and
scavenger receptor class B type I. Biochem Biophys Res Commun 331, 1533-1541.
Srivastava, R. A., Srivastava, N., Averna, M., Lin, R. C., Korach, K. S., Lubahn, D.
B., and Schonfeld, G., 1997, Estrogen up-regulates apolipoprotein E (ApoE) gene
expression by increasing ApoE mRNA in the translating pool via the estrogen receptor
alpha-mediated pathway. J Biol Chem 272, 33360-33366.
Statistical Graphics Corp., Version 5.1, Herndon, VA.
75
Staudinger, J., Liu, Y., Madan, A., Habeebu, S., and Klaassen, C. D. (2001a).
Coordinate regulation of xenobiotic and bile acid homeostasis by pregnane X receptor.
Drug Metab Disp 29, 1467-1472.
Staudinger, J. L., Goodwin, B., Jones, S. A., Hawkins-Brown, D., MacKenzie, K. I.,
LaTour, A., Liu, Y., Klaassen, C. D., Brown, K. K., Reinhard, J., Willson, T. M.,
Koller, B. H., and Kliewer, S. A. (2001b). The nuclear receptor PXR is a lithocholic
acid sensor that protects against liver toxicity. PNAS, U S A 98, 3369-3374.
Subramanian A, Miller D.M., 2000, Structural analysis of alpha-enolase; mapping the
functional domains involved in down-regulation of the c-myc protooncogene. J Biol
Chem. 275, 5958-65.
Tang, W., Ma, Y. and Yu, L., 2006, Plasma cholesterol is hyper-responsive to statin in
ABCG5/ABCG8 transgenic mice. Hepatology, 44, 1259-1266.
Tirona, R.G. and Kim, R.B., 2005, Nuclear Receptors and Drug Disposition Gene
Regulation, J. Pharm. Sci. 94, 1169-1186.
Van den Boom, F., Düssmann, H., Uhlenbrock, K., Abouhamed, M. and Bähler, M.,
2007, The myosin IXb motor activity targets the myosin IXb RhoGAP domain as
cargo to sites of actin polymerization, Mol Biol Cell. 18, 1507–1518.
Wang, D.Q.H., 2007, Regulation of intestinal cholesterol uptake. Annu. Rev. Physiol.
69, 221-248.
Wang, J., Einarsson, C., Murphy, C., Parini, P., Bjorkhem, I., Gafvels, M. and
Eggertsen, G., 2006, Studies on LXR and FXR mediated effects on cholesterol
homeostasis in normal and cholic acid-depleted mice. J. Lipid Res. 47, 421-430.
Washburn, M.P., Wolters, D. and Yates, J.R., 2001, Large-scale analysis of the yeast
proteome by multidimensional protein identification technology. Nat. Biotechnol. 19,
242-247.
Webb, N.R., Connell, P. M., Graf, G.A., Smart, E.J., de Villiers, W.J.S., de Beer, F.C.
and van der Westhuyzen, D.R., 1998, SR-BII, an isoform of the scavenger receptor BI
76
containing an alternate cytoplasmic tail, mediates lipid transfer between high density
lipoprotein and cells. J. Biol. Chem. 273, 15241-15248.
Wiese, S., Reidegeld, K.A., Meyer, H.E. and Warscheid, B., 2007, Protein labeling by
iTRAQ: a new tool for quantitative mass spectrometry in proteome research.
Proteomics, 7:340-50.
Wustner, D., Mondal, M., Huang, A. and Maxfield, F.R., 2003, Different transport
routes for high density lipoprotein and its associated free sterol in polarized hepatic
cells, J. Lipid Res. 45,427-436.
Yamashita, M. and Fenn, J.B., 1984, Electrospray ion source; another variation on the
free-jet theme, J. Phys. Chem. 88, 4451-4459.
Yong, J., Wang, N., Ramakrishnan, R., Sehayek, E., Huzar, D., Breslow, J.L. and Tall,
A.R., 1999, Hepatic Scavenger Receptor BI Promotes Rapid Clearance of High
Density Lipoprotein Free Cholesterol and Its Transport into Bile, J. Biol. Chem. 274,
33398-33402.
Yu, L., Gupta, S., Xu, F., Liverman, A.D., Moschetta, A., Mangelsdorf, D.F., Repa,
J.J., Hobbs, H.H. and Cohen, J.C., 2005, Expression of ABCG5 and ABCG8 is
required for regulation of biliary cholesterol secretion. J. Biol. Chem. 280, 8742-8747.
Zhang, X., Moor, A.N., Merkler, K.A., Liu, Q and McLean, M.P., 2007, Regulation of
alternative splicing of liver scavenger receptor class B gene by estrogen and the
involved regulatory splicing factors, Endocrinology, 148, 5295-5304.
77
Appendices
78
Appendix A
14
Chlordecone Altered Hepatic Disposition of [ C]Cholesterol and Plasma Cholesterol
Distribution but not SR-BI or ABCG8 Proteins in Livers of C57BL/6 Mice
Junga Lee, Richard C. Scheri and Lawrence R. Curtis+
79
ABSTRACT
Organochlorine (OC) insecticides continue to be found in tissues of humans and
wildlife throughout the world although they were banned in the United States a few
decades ago. Low doses of the OC insecticide chlordecone (CD) alter hepatic
disposition of lipophilic xenobiotics and perturb lipid homeostasis in rainbow trout,
mice and rats. CD pretreatment altered tissue and hepatic subcellular distribution of
14
exogenous [ C]cholesterol (CH) equivalents 4 and 16 h after a bolus intraperitoneal
(ip) injection of 5 ml corn oil/kg that contained 10 mg CH/kg. CD pretreatment altered
14
tissue distribution of exogenously administered [ C]CH by decreased hepatic and
renal accumulation, and increased biliary excretion up to 300%. Biliary excretion of
14
polar [ C]CH metabolites was not altered by CD. CD pretreatment decreased
14
subcellular distribution of [ C]CH equivalents in hepatic cytosol and microsomes and
lipoprotein-rich fraction-to-homogenate ratio. CD pretreatment increased the ratio of
14
[ C]CH equivalents in high density lipoprotein (HDL) to that in plasma and reduced
14
[ C]CH equivalents in the non-HDL fraction 4 h after a bolus lipid dose. CD
pretreatment increased plasma non-HDL total CH by 80% 4 h after a bolus lipid dose.
Scavenger receptor class B type I (SR-BI) and ATP-binding cassette transporter G8
(ABCG8) proteins were quantified by western blotting in hepatic membranes from
control and CD treated mice. Liver membrane contents of SR-BI and ABCG8 proteins
were unchanged by CD pretreatment. The data demonstrated that a single dose of CD
altered CH homeostasis and lipoprotein metabolism.
Keywords: organochlorines, chlordecone, cholesterol disposition, liver, biliary
excretion, lipoproteins.
80
INTRODUCTION
Organochlorines (OC) are a major group of persistent organic pollutants including
chlorinated dioxins and furans, polychlorinated biphenyls (PCBs), and the OC
insecticides. Despite bans on most uses of OCs in the United States and Europe, they
continue to occur in tissues of humans and wildlife throughout the world (Bocquene
and Franco 2005; Luellen et al. 2006). Because they are highly lipophilic and
generally very resistant to degradation, trophic transfer though food webs is the
principle route of environmental exposure to these agents. Modes of action for OC
insecticides are distinct from that prototypical for chlorinated dioxins and planar PCB
congeners in that they exhibit negligible affinity for the aryl hydrocarbon receptor
(Poland and Knutson, 1982). As is typical for OC insecticides, neurotoxicity is a
prominent mode of action for CD: it is unusual in that there is extensive epidemiologic
documentation of a major human poisoning following poor manufacturing and
disposal practices (reviewed by Guzelian, 1982). Prominent weight loss despite
normal appetite is a common sign of occupational CD poisoning in humans (10 of 23
cases).
The OC insecticides dieldrin and chlordecone (CD) alter hepatic disposition of
lipophilic xenobiotics and perturb lipid homeostasis in rainbow trout, mice and rats
(Barnhill et al., 2003; Carpenter and Curtis 1989, 1991; Carpenter et al. 1996;
Donohoe et al. 1998; Gilroy et al. 1994). These effects of OC insecticides occur at
liver concentrations in the 1-75 parts per million (ppm) range. Single, doses of CD and
dieldrin decrease plasma triglycerides and total CH in rats 21 and 60 days after
81
administration, respectively (Ishikawa et al. 1978). A single dose of 5 mg CD/kg to
14
C57BL/6 mice also significantly alters tissue distribution of exogenous [ C]CD or
14
[ C]CH, decreasing hepatic disposition and increasing distribution to other tissues
(Carpenter and Curtis 1991).
CH is an important constituent of cell membranes and a precursor of steroid hormones
and bile acids (Chen and Raymond 2006; Tabas 2002). However accumulation of
excess CH contributes to several diseases especially heart attacks and stroke by
promoting atherosclerosis (Krieger 1999). CH metabolism is regulated by plasma
lipoprotein-mediated extracellular CH transport, the sterol regulatory element binding
protein (SREBP) system involved CH synthesis (Brown and Goldstein 1997), nuclear
receptor-mediated CH catabolism to bile acids (Peet et al. 1998), and plasma
membrane transporter-mediated excretion into bile (Yu et al. 2002).
Transport kinetic studies indicated that the rate of uptake of high density lipoprotein
14
14
(HDL)-bound [ C]CD or [ C]CH was significantly lower in perfused livers from
Sprague-Dawley rats pretreated with a single dose of 15 mg/kg CD (Gilroy et al.
1994). Gilory et al. (1994) reported a significantly increased rate of efflux of HDL14
14
bound [ C]CD to perfusate and cumulative biliary excretion of HDL-bound [ C]CH in
these same preparations. Plasma HDL transfers CH from peripheral tissue to the liver
for biliary excretion by the reverse CH transport pathway (Fielding and Fielding
1995). Therefore HDL is an important source of CH for secretion into the bile in
rodents (Fielding and Fielding 1995; Lee and Parks 2005). Plasma low-density
82
lipoproteins (LDL) and very low-density lipoproteins (VLDL) deliver CH to
peripheral tissues such as adipose. In contrast to other OC insecticides, such as
dieldrin and DDT in which tissue distribution is proportional to tissue lipid contents
(Lindstrom et al. 1974), CD distributes preferentially to the liver in both humans
(Cohn et al. 1978) and rodents (Egle et al. 1978). Dieldrin and DDT are primarily
associated with LDL and VLDL (Mick et al. 1971). CD binds preferentially with
albumin and HDL in vitro, consistent with in vivo studies of human, rat, and pig
plasma (Soine et al. 1984a; 1982) and this may account for its predominant
distribution to the liver. These previous works indicate that a change in hepatic uptake
of HDL-complexed CH is important in alteration of lipid homeostasis by CD. Low
doses of CD pretreatment may modulate membrane proteins which transport CH
thereby disturbing the distribution of the exogenous CH to liver and increasing
delivery of CH into bile.
Scavenger receptor class B type I (SR-BI), best known as a physiological HDL
receptor, mediates selective uptake of both CH esters and other lipids from HDL
(Hobbs and Rader 1999; Krieger 1999). SR-BI deficiency reduces the rate of the
clearance of HDL-CH from plasma (Out et al. 2004). ATP-binding cassette (ABC)
transporter G5 (ABCG5) and G8 (ABCG8) are members of the G family of ABC
transporters. Co-expression of G5 and G8 which are obligate heterodimers promotes
transport of CH from hepatocytes to bile and probably from enterocytes into the
intestinal lumen (Yu et al. 2005; 2002).
83
The present studies examined the effect of CD pretreatment on the tissue distribution
14
of exogenous [ C]CH 4 and 16 h after intraperitoneal (ip) injection of 5 ml corn oil/kg
14
that contained 10 mg CH/kg. Hepatic subcellular distribution of [ C]CH was also
determined. Blood plasma, hepatic and biliary total CH were examined in control and
CD pretreated mice. This investigation determined whether the reduced disposition of
14
[ C]CH to liver was associated with decreased SR-BI protein, SR-BI was analyzed by
immunoblotting with hepatic membranes from CD treated mice (15 mg/kg) or
controls. Immunoblot analysis for ABCG8 was conducted with hepatic membranes
from CD treated mice (15 mg/kg) or controls to determine whether the stimulation of
biliary CH efflux by CD was coupled to the increased CH efflux pump
(ABCG5/ABCG8).
Our results demonstrated that CD pretreatment altered plasma, tissue and hepatic
14
subcellular distribution of exogenously administered [ C]CH. CD pretreatment also
altered the amount of total CH in blood plasma and gallbladder. The reduced hepatic
14
14
[ C]CH equivalents and increased biliary [ C]CH equivalents were not associated
with altered hepatic membrane SR-BI or ABCG8 contents.
MATERIALS AND METHODS
Chemicals
CD (99 % purity) was purchased from Chem Service (West Chester, PA) and purity
14
was confirmed by GC-EI/MS. [4- C]CH (55 mCi/mmol) was obtained from American
Radiolabeled Chemicals. Inc. (St. Louis, MO). Solvable tissue solubilizer and Hionic-
84
fluor scintillation cocktail were purchased from PerkinElmer (Boston, MA). All other
chemicals and general reagents were purchased from Sigma (St. Louis, MO).
Treatment of mice
14
The CD and lipid bolus including [ C]CH doses and times of sampling were based on
previous dose-response and time-course experiments (Carpenter and Curtis, 1989;
1991). Male C57BL/6 mice, 6 to 7-weeks-old, were obtained from Simenson
Laboratory (Gilroy, CA). The mice were randomly divided into three groups; control,
5 mg or 15 mg CD/kg body weight treatment. The animals were housed in a
o
temperature controlled room (22 ± 1 C) with a daily cycle of 12 h of light and 12 h
darkness and fed ad libitum with AIN93 diet (Dyets, Inc., Bethlehem, PA) with a free
access to water. Treatments were initiated after seven days acclimatization.
14
[ C]CH tissue distribution experiments
CD was dissolved in corn oil. Mice received CD (5 or 15 mg/kg body weight) or corn
oil alone by intraperitoneal injection (ip, 5 ml/kg body weight). After 3 days, a bolus
14
lipid dose of corn oil (5 ml/kg) containing [ C]CH was administered (10 mg CH/kg
14
body weight, ip). [ C]CH injected mice were housed individually and allowed free
access to water. After 4 or 16 h animals were killed by CO2 anesthesia and
exsanguination. Blood samples and tissues (liver, kidneys, gallbladder, adrenal gland
and adipose tissue) were removed and prepared as required for each analysis. The
procedures for animal use were approved by the Oregon State University Institutional
Animal Care and Use Committee (IACUC).
85
Quantification of radioactivity in tissues, blood and hepatic subcellular fractions
Blood was collected by cardiac puncture in heparinized syringes. Plasma was isolated
o
by centrifugation at 3,000 g for 25 min at 4 C. The HDL fraction was separated after
apolipoprotein-B containing lipoproteins were precipitated with dextran sulfate and
magnesium (Rachem, San Diego, CA). Kidneys, fat, and adrenal glands were
homogenized in homogenization buffer (0.01 M potassium phosphate, pH 7.5; 0.15 M
KCl; 1.0 mM EDTA; 0.1 mM BHT (butylated hydroxytoluene); 0.1 mM PMSF
(phenylmethylsulfonyl fluoride)) with a Dounce homogenizer. Liver was
homogenized in homogenization buffer with a polytron PT 3000 (Brinkmann
Instruments, Westbury, NY). Liver homogenate was centrifuged at 1200 g for 30 min.
The pellet was cell debris. The supernatant was centrifuged at 9,000 g for 20 min to
prepare mitochondria. The supernatant was again centrifuged at 100,000 g for 90 min
o
at 4 C with Ti 70 rotor. The remaining supernatant and pellet were cytosolic and
microsomal fractions, respectively. Subsamples of each tissue homogenate and each of
hepatic subcellular fraction (cell debris, mitochondria, cytosol and microsomes) were
o
solubilized with Solvable (PerkinElmer, Boston, MA) at 50 to 60 C for 1.5 – 2 h and
decolorized by adding 30% hydrogen peroxide. Plasma and HDL fraction were
analyzed by liquid scintillation counting (LSC) without further treatment. Ten ml of
Hionic-Fluor (PerkinElmer, Boston, MA) liquid scintillation cocktail was added. After
decay of chemiluminescence liquid scintillation counting was conducted using a
Beckman LS7500 liquid scintillation counter. Protein concentration was determined
by BCA assay (Pierce Biotechnology, Inc. Rockford, IL).
Gallbladders plus bile from mice killed 4 hr after lipid bolus injections were digested,
86
14
decolorized, and analyzed by LSC for total [ C]CH equivalents as described above.
14
Lipid and polar fractions of gallbladders plus bile from mice killed 16 hr after [ C]CH
injections were separated by liquid-liquid extraction. Neutral steroids were extracted
with slight modifications of the method described by Miettinen et al (1965). A 2.9 ml
volume of basic (1.5 N KOH) ethanol/water (2:1) was added to each gallbladder plus
bile sample which was heated at 70oC for 30 min. One ml of water was added to
cooled samples which were then extracted with 2 ml of hexane. The hexane layer was
removed and dried under a nitrogen stream. The residue was resuspended in 1 ml 2propanol that contained 10% Triton X-100. Duplicate 0.1 ml samples of the
ethanol/water phase (polar metabolites) and the hexane residues containing neutral
steroids were analyzed by LSC.
Plasma lipid analyses
Total plasma CH concentrations were measured enzymatically by using 2 μL of
plasma and 200 μL of CH reagent from Sigma Diagnostic (St. Louis, MO). HDL-CH
concentration was measured after apolipoprotein-B containing particles precipitation
with dextran sulfate and magnesium (Rachem, San Diego, CA). Non-HDL-CH
concentrations were taken as the difference between the two.
87
Western blot analysis
Liver crude membrane fractions were prepared as described (Voshol et al. 2001)
from control or 15 mg CD/kg treated animals. Proteins were separated on
Tris/glycine gels (Bio-Rad, Richmond, CA) under reducing conditions. Following
gel electrophoresis at 100 V for 2 h, proteins were electrophoretically transferred
onto PVDF membranes utilizing a Mini trans-blot transfer cell (Bio-Rad,
Richmond, CA). Membranes were then blotted with rabbit anti-SR-BI (Abcam,
Cambridge, MA) and probed with horseradish peroxidase conjugated goat antirabbit IgG. Membranes were also blotted with mice anti-ABCG8 (a generous gift
from Professor Helen H. Hobbs, Univ. of Texas Southwestern Medical Center) and
then probed with horseradish peroxidase conjugated rat anti-mouse IgG as the
secondary antibody (Bio-Rad, Richmond, CA). The proteins were detected after
development with enhanced chemiluminescence (ECL) detection (Amersham,
Piscataway, NJ). Quantification of the intensity of the protein bands was performed
with NIH-image software.
Biliary lipid analyses
Total CH was determined in the hexane extracts of bile treated with basic
ethanol/water for 30 min at 70oC (detailed above). Duplicate subsamples in 2propanol that contained 10% Triton X-100 were assayed with a CH Quantitation
KitR BioVision, Mt. View, CA
Statistical Methods
Statistical analyses were conducted with StatGraphics software (StatPoint,
Herndon, VA). Normality of data sets were assessed with standardized kurtosis,
88
standardized skewness, and homogeneity of variance. Data that failed one or more
test for normality were log base 10 transformed. Transformed data were assessed
for normality with the criteria specified above. Data were expressed as means ±
SE. Comparisons among groups validated for normality were submitted to a oneway analysis of variance (ANOVA). For comparison between two groups,
Student’s t-test with equal-variance was applied. In all analyses, a 95%
confidential level was used as the criterion for significance. For outlier
identification, Grubbs' Test (assumes normality) was applied.
RESULTS
The mean body weights of the control, 5, and 15 mg/kg CD pretreated mice at 4 or
14
16 h after 10 mg [ C]CH/kg in 5 ml corn oil/kg (ip bolus lipid) were similar as
were the liver weights (Table 1-1). No significant differences in gallbladder
weights between control and CD pretreated mice were observed (Table 1).
However, 16 h following ip bolus lipid dose, gallbladder weights were
significantly smaller (~9 to 10 mg) than those at 4 h following ip bolus lipid dose
(~15 to 17 mg) (Table 1).
14
Plasma [ C]CH equivalents were approximately 3-fold higher at 16 h than at 4 h
after the ip bolus lipid dose (p < 0.05, Fig. 1A, 1B), although plasma total CH
concentration at 16 h was not different from that at 4 h after the ip bolus lipid (Fig.
14
1C, 1D). Plasma [ C]CH equivalents from CD (5 or 15 mg CD/kg) pretreated
mice significantly decreased 4 h following the bolus lipid dose (31 %, p <
0.05)(Fig. 1A). CD pretreatment (5 or 15 mg CD/kg) significantly decreased
14
[ C]CH equivalents in the non-HDL fraction compared with controls (45%, p <
89
0.05)(Fig. 1A). On the contrary, CD pretreatment increased total CH in non-HDL
fraction 4 h following ip bolus lipid (about 1.8-fold, p < 0.05)(Fig. 1C). At 16 h
14
after the ip bolus lipid, neither [ C]CH equivalents nor total CH concentrations in
plasma were altered by CD pretreatment (Fig. 1B, 1D). No significant differences
14
in the [ C]CH equivalents and total CH in the either HDL or non-HDL fractions
14
were observed 16 h following ip bolus lipid (Fig. 1B, 1D). Although the [ C]CH
14
equivalents in the HDL fraction were unchanged, the ratio of [ C]CH equivalents
in HDL fraction to that in plasma was mildly but statistically significantly
increased by CD pretreatment 4 h after ip bolus lipid in a dose-dependent manner
(Fig. 2A).
SR-BI was analyzed by immunoblotting with hepatic membranes from CD treated
mice (15 mg/kg) or controls. Liver membrane contents of SR-BI protein in CDtreated mice were not different from control animals (Fig. 2B).
14
Consistent with previous results, CD decreased [ C]CH equivalents distribution to
the liver at 16 h after the bolus lipid dose (43 %, p < 0.05)(Fig. 3A). Altered
14
[ C]CH disposition in the liver of CD pretreated mice was further characterized in
hepatic subcellular fractions (homogenate, lipoprotein-rich fraction, cytosol,
microsomes, mitochondria and cell debris/nuclei). At 4 h after the bolus lipid dose,
14
[ C]CH equivalents significantly decreased about 50 % in the microsomal fraction
from CD (15 mg/kg) pretreated mice compared to control animals (10.1 ± 1.3 vs
14
5.9 ± 0.4 pmol/mg protein, p < 0.05). At 16 h after ip bolus lipid, [ C]CH
equivalents significantly decreased about 30 % in hepatic cytosolic fraction from
CD (15 mg/kg) pretreated mice compared to control animals ( 5.7 ± 0.4 vs 4.0 ±
90
0.3 pmol/mg protein, p < 0.05). To assess the hepatic compartmental re14
14
distribution of C]CH, [ C]CH equivalents from each hepatic fraction were
14
14
normalized with [ C]CH equivalents in homogenate. The ratios of [ C]CH
lipoprotein-rich fraction-to-homogenate were lower at 16 h than at 4 h after the ip
bolus lipid in both control and CD pretreated mice. Interestingly, 15 mg CD/kg
14
pretreatment significantly reduced [ C]CH equivalents lipoprotein-rich fraction-tohomogenate ratio 16 h after ip bolus lipid, compared with control (about 50%, p <
14
0.05)(Fig. 3B). A trend for decreased [ C]CH in lipoprotein-rich and microsomal
fraction-to-homogenate ratios at 4 and cytosolic fraction-to-homogenate ratios at
16 h after the ip bolus lipid in CD pretreated animals was observed (Fig. 3B).
14
CD pretreatment increased [ C]CH equivalents of gallbladder 4 and 16 h after the
ip bolus lipid dose, in a dose-dependent manner (Fig. 4A). A statistically
14
significant 300% increase total [ C]CH equivalents was observed 4 h after the
14
tracer in 15 mg CD/kg pretreated animals (p < 0.05). A trend for increased [ C]CH
equivalents in the neutral steroid fraction from gallbladders at 16 h after the ip
bolus lipid in 15 mg CD/kg pretreated animals was observed (p < 0.056, Fig. 4A).
14
Polar metabolites constituted 5.6-10.7% of total biliary [ C]CH equivalents at 16
hr after lipid bolus and there were no significant differences between treatment
14
groups. The absolute polar biliary [ C]CH equivalents for control, 5 and 15 mg
CD/kg pretreatments were 1.6, 1.7 and 0.9 nmol/g. CD pretreatment increased total
CH concentration in gallbladder 16 h after lipid bolus compared with control (Fig.
4B). Immunoblot analysis for ABCG8 was conducted with hepatic membrane from
CD treated mice (15 mg/kg) or controls. Liver membrane content of ABCG8
protein was unchanged by CD treatment (Fig. 4C). Statistically significant
91
14
reductions of [ C]CH distribution to the kidney (48 %, p < 0.05) was observed in
15 mg CD/kg pretreated animals (data not shown). No significant change was
observed in other tissues (data not shown).
DISCUSSION
The present results demonstrate that pretreatment with a single, dose of CD alters
plasma (Figs. 1, 2), tissue and hepatic subcellular distribution of exogenously
14
administered [ C]CH (Figs. 3, 4). There were a number of limitations in
14
interpretation of plasma [ C]CH equivalents for explanation of CD-altered
14
disposition of the exogenous lipid bolus. Total plasma [ C]CH equivalents
included material: (1) newly absorbed from the peritoneal cavity associated with
chylomicrons; (2) subjected to intraplasmic exchange between lipoprotein classes;
and (3) secreted from the liver as nascent HDL or as VLDL. Nonetheless
14
comparisons of data for plasma [ C]CH equivalents and plasma total CH in
fractions treated with dextran sulfate and magnesium demonstrated CD-dependent
14
differences (Fig. 1A, 1C). CD pretreatment decreased plasma non-HDL [ C]CH
and increased plasma total non-HDL-CH 4 h after the ip lipid bolus dose (Fig. 1A,
1C). There were at least two mutually nonexclusive explanations. (1) CD delayed
14
[ C]CH incorporation into non-HDL plasma lipoproteins, perhaps into hepatic
VLDL synthesized for secretion. Exogenous CH occurred in a hepatic pool distinct
from that synthesized in the liver (Oram and Vaughan 2006). Consistent with this,
14
in control mice 4 h after the ip bolus lipid dose about 50 % of plasma [ C]CH
equivalents were in the non-HDL fraction, while about 17 % of plasma total CH
14
appeared in this fraction. (2) CD perhaps increased clearance of non-HDL-[ C]CH.
14
Stimulated plasma clearance of non-HDL-[ C]CH seemed unlikely since total non-
92
HDL-CH was elevated in CD pretreated mice 4 h after the bolus lipid (Fig. 1C).
14
Plasma non-HDL-[ C]CH equivalents in CD pretreated mice were lower than
controls 4 but not 16 h after the bolus lipid dose (Figs. 3A, 3B). This indicated that
14
CD altered a CH exchange pathway or pool that delayed [ C]CH incorporation into
14
non-HDL lipoproteins. At 16 h after ip lipid bolus dose [ C]CH appeared
integrated into the total plasma CH pool (Fig. 1B). In fact, it appeared
14
preferentially retained or secreted in the non-HDL pool since 32 to 35% of [ C]CH
equivalents were in that fraction compared 18 to 20% of total CH at 16 h. Elevated
total plasma non-HDL CH in CD pretreated mice 4 h after the ip lipid bolus (fig.
1C) was of specific interest. This reflected approximately doubled “bad CH”
associated with promotion of atherosclerosis (Krieger, 1999). Plasma HDL
contained 80-90% of total plasma CH in control mice (fig. 1C, 1D), typical for this
species. Since non-HDL lipoproteins predominate in humans, the shift in plasma
CH distribution after lipid bolus in CD pretreated mice was not directly applicable
to human risk assessment. However, the results suggested work in animals with
lipoprotein profiles similar to humans was warranted.
14
Hepatic [ C]CH equivalents were not different between control and CD pretreated
mice at 4 h after ip lipid bolus dose. CD exhibited specific binding affinity with
albumin and HDL and was preferentially accumulated in the liver (Soine et al.
1982). Gilroy et al. (1994) suggested a competitive interaction between CH and
14
CD by the observation of decreased uptake of HDL-bound [ C]CH in the CD
treated perfused rat liver. CD was an agonist for the human pregnane x receptor
(PXR) in a reporter gene assay and this was supported by CYP3A protein
induction in mouse liver by CD (Lee et al., unpublished data). PXR agonists but
93
not a selective constitutive androstane receptor agonist increased expression of the
apoA-I gene in mice (Bachmann et al. 2004), the principal component of HDL.
Sporstol et al. (2005) reported the down regulation of SR-BI by a PXR agonist in
vitro. The HDL receptor SR-BI plays critical roles in the uptake of plasma CH by
the liver. Therefore, hepatic SR-BI protein content was determined, this assessed
whether hepatic uptake of HDL-CH was important in alteration of plasma CH by
CD. SR-BI content in hepatic plasma membrane was not changed by CD treatment
(Fig. 2B). Even though SR-BI was a physiological HDL receptor, it exhibited
binding affinity for HDL, LDL, and VLDL (Acton et al. 1996; Calvo et al. 1998).
Over expression of SRBI in the liver increased the clearance of LDL and VLDL
(Wang et al. 1998). More non-spherical and heterogeneous HDL particles were
observed in plasma from CD treated animals compared with controls (Lee et al.,
unpublished data). Therefore it was possible that CD binding to HDL particles
perturbed the SR-BI mediated selective CH ester uptake from HDL while
clearance of LDL and VLDL was not affected. However it seemed unlikely since
total non-HDL-CH was elevated in CD pretreated mice 4 h after the bolus lipid
14
(Fig. 1C). Lower lipoprotein-rich fraction-to-liver homogenate [ C]CH equivalents
ratios in 15 mg CD/kg pretreated mice were consistent with suppression of
incorporation of exogenous CH into lipoprotein complexes (Fig. 2-3B). Markedly
14
reduced [ C]CH lipoprotein-rich fraction-to-homogenate ratio from 4 to 16 h after
the bolus lipid dose in both control and CD pretreated mice suggested the secretory
phase peaked before 16 h (Fig. 3B). Elevation in plasma non-HDL-total CH at 4 h
compared to 16 h after the lipid dose (Figs. 3C, 3D) in CD pretreated mice
supported this interpretation.
94
Gallbladder bile of mice pretreated with 15 mg CD/kg contained 3-fold more
14
[ C]CH equivalents than controls 4 h after the ip lipid bolus. A trend for this
persisted after 16 h but was not statistically significant. Since gallbladder weight
was lower at 16 than 4 h after the lipid bolus (Table 1) leakage or ejection of bile
probably occurred more extensively before the later than the earlier sampling time.
14
This likely reduced accuracy of gallbladder bile [ C]CH equivalents for estimation
14
of biliary secretion at 16 compared to 4 h after the ip lipid bolus. Polar [ C]CH
equivalents in gallbladder bile 16 hr after ip lipid bolus were unchanged (Results).
This indicated CD stimulated biliary CH excretion itself, not CH metabolism to
bile salts or other polar metabolites. Biliary CH secretion is one of the major
pathways to eliminate excess CH from body. Heterodimers of ABCG5 and
ABCG8 regulate and the whole-body retention of plant sterols and promote
hepatobiliary secretion of CH (Kosters et al. 2006; Yu et al. 2002). CH feeding
upregulates expression of ABCG5/ABCG8 coordinately through activation of
LXR (liver X receptor) although the binding sites of LXR to these genes have not
been mapped (Repa et al. 2002). The mRNA level of ABCG5/ABCG8 was higher
in the liver of mice lacking LXRα/LXRβ (Repa et al. 2002). CD moderately
inhibited LXRα activation and strongly suppressed LXRβ activation in a reporter
gene assay (Lee et al., unpublished data). We hypothesized that increased basal
activity of ABCG5/ABCG8 by CD stimulated biliary CH excretion. Hepatic
membrane ABCG8 protein contents were not different between controls and 15 mg
14
CD/kg treatment (Fig. 2-4C). CD pretreatment increased not only [ C]CH
equivalents but also total CH in gallbladder 4 and 16 h after the ip lipid bolus dose,
respectively. This indicated that the lipid bolus dose affected biliary CH secretion
in CD pretreated mice compared with controls. Gallbladder bile total CH was not
95
increased in CD pretreated mice compared to controls in animals that received no
ip lipid bolus dose (Lee et al., unpublished data). Therefore it was possible that CD
pretreatment and ip bolus lipid dose somehow synergistically stimulated an
ABCG5/8-dependent or ABCG5/8-independent pathway of biliary CH secretion.
Another possible explanation was SRBII mediated stimulation of biliary CH
excretion. SR-BII is an alternatively spliced form of SR-BI. SR-BII is expressed
intracellularly and suggested to mediate the rapid internalization of HDL for biliary
excretion (Eckhardt et al. 2004). Although HDL was an important source of CH
for biliary excretion in rodents, HDL-CH was the preferred substrate for bile acid
synthesis. CH is secreted into bile after arrival at the canalicular membrane. An
endocytic/retroendocytic pathway of HDL has been suggested to play a important
role for the delivery of CH to the apical membrane for release into the bile
(Wustner et al. 2004). In addition to membrane transporter, cytoplasmic proteins
such as liver fatty acid-binding protein and sterol carrier protein 2 were suggested
to play a role in the intracellular transport and biliary CH secretion (Kosters et al.
2005). Essentially all CD existed in pig liver cytosol in a protein-bound form
14
(Soine et al. 1982). [ C]CH equivalents significantly decreased in the hepatic
microsomal (~ 50 %) and cytosolic (~30%) fractions from CD pretreated mice
compared to control animals. Soine et al. (1982) suggested that CD and CH shared
a common transport pathway in liver cytosol and CD interacted with CH transport
and metabolism since isolated-CD binding proteins bind both CD and CH (Soine et
14
al. 1984b). Decreased hepatic disposition of [ C]CH equivalents in the microsomal
fraction suggested that binding of CD to cytosolic proteins perhaps not only
inhibited CH transport to microsomes but also modulated biliary CH secretion.
However, a definitive explanation for increased biliary CH excretion in CD
96
pretreated mice remains elusive.
Chronic exposures of humans and wildlife via trophic transfer through food webs
is the most widespread pathway of exposure to persistent organic pollutants,
including OC insecticides. Residues occur as complex mixtures (Lordo et al, 1996)
and direct extrapolation of results after ip administration to a single compound is
tenuous. There is value in elucidation of potential modes of action, however. The
ip route of exposure in mice yielded a hepatic CD concentration quite similar to
that in rats after dietary exposure. Ten mg CD/kg ip in mice yielded 75 ppm in
liver 3 days after dosing (Carpenter and Curtis, 1989). Livers of rats fed a diet that
contained 10 pm CD for 15 days accumulated 52 ppm CD (Curtis and Mehendale,
1981). To the extent that liver OC concentration mediates pathophysiological
changes, ip administration provides a convenient alternative to long-term feeding
studies.
In summary, CD pretreatment altered the plasma, tissue and hepatic subcellular
14
disposition of subsequently administered [ C]CH. Plasma non-HDL CH almost
doubled 4 h after a bolus lipid dose without a significant change in HDL total CH.
Our data demonstrated that SR-BI and ABCG8 protein contents in hepatic plasma
membranes were unchanged by CD treatment indicating that mechanisms other
than SR-BI or ABCG8 may be involved in modulating CD induced CH
homeostasis and lipoprotein metabolism.
97
Acknowledgements:
This work was supported by the Oregon Agricultural Experiment Station
(ORE00871) and grant number T32 ES007060 from the National Institutes of
Health.
98
References
Acton, S., Rigotti, A., Landschulz, K. T., Xu, S., Hobbs, H. H., and Krieger, M.
(1996). Identification of scavenger receptor SR-BI as a high density lipoprotein
receptor. Science 271, 518-520.
Bachmann, K., Patel, H., Batayneh, Z., Slama, J., White, D., Posey, J., Ekins, S.,
Gold, D., and Sambucetti, L. (2004). PXR and the regulation of apoA1 and HDLcholesterol in rodents. Pharmacol Res 50, 237-246.
Barnhill, M. L., Rosemond, M. V. M. and Curtis, L.R. (2003). Dieldrin stimulates
14
biliary excretion of C-benzo [a] pyrene polar metabolites but does not change the
biliary metabolite profile in rainbow trout (Oncorhynicus mykiss) Toxicol. Sci 75:
249-259.
Bocquene, G., and Franco, A. (2005). Pesticide contamination of the coastline of
Martinique. Mar Pollut Bull 51, 612-619.
Brown, M. S., and Goldstein, J. L. (1997). The SREBP pathway: regulation of
cholesterol metabolism by proteolysis of a membrane-bound transcription factor.
Cell 89, 331-340.
Calvo, D., Gomez-Coronado, D., Suarez, Y., Lasuncion, M. A., and Vega, M. A.
(1998). Human CD36 is a high affinity receptor for the native lipoproteins HDL,
LDL, and VLDL. J Lipid Res 39, 777-788.
Carpenter H.M., Hedstrom O.R., Siddens L.K., Duimstra J.R., Cai Z.W., Fisher
K.A., Curtis L.R.., 1996, Ultrastructural, protein, and lipid changes in liver
associated with chlordecone treatment of mice, Fundam Appl Toxicol. 34, 157-64.
Carpenter, H. M., and Curtis, L. R. (1989). A characterization of chlordecone
pretreatment-altered pharmacokinetics in mice. Drug Metab Dispos 17, 131-138.
Carpenter, H. M., and Curtis, L. R. (1991). Low dose chlordecone pretreatment
altered cholesterol disposition without induction of cytochrome P-450. Drug
Metab Dispos 19, 673-678.
Chen, J., and Raymond, K. (2006). Nuclear receptors, bile-acid detoxification, and
cholestasis. Lancet 367, 454-456.
Cohn, W. J., Boylan, J. J., Blanke, R. V., Fariss, M. W., Howell, J. R., and
Guzelian, P. S. (1978). Treatment of chlordecone (Kepone) toxicity with
cholestyramine. Results of a controlled clinical trial. N Engl J Med 298, 243-248.
Curtis, L. R. and Mehendale, H. M. (1981). Hepatobiliary dysfunction and
inhibition of adenosine triphosphatase activity of bile canaliculi enriched fraction
following in vivo mirex, photomirex and chlordecone exposures. Toxicol Appl
Pharmacol 61, 429-440.
Donohoe, R. M., Zhang, Q., Siddens, L. K., Carpenter, H. M., Hendricks, J. D.,
and Curtis, L. R. (1998). Modulation of 7,12-dimethylbenz[a]anthracene
disposition and hepatocarcinogenesis by dieldrin and chlordecone in rainbow trout.
99
J Toxicol Environ Health A 54, 227-242.
Eckhardt, E. R., Cai, L., Sun, B., Webb, N. R., and van der Westhuyzen, D. R.
(2004). High density lipoprotein uptake by scavenger receptor SR-BII. J Biol
Chem 279, 14372-14381.
Egle, J. L., Fernandez, J. B., Guzelian, P. S., and Borzelleca, J. F. (1978).
Distribution and excretion of chlordecone (Kepone) in the rat. Drug Metab Dispos
6, 91-95.
Fielding, C. J., and Fielding, P. E. (1995). Role of an N-ethylmaleimide-sensitive
factor in the selective cellular uptake of low-density lipoprotein free cholesterol.
Biochemistry 34, 14237-14244.
Gilroy, D. J., Carpenter, H. M., and Curtis, L. R. (1994). Chlordecone pretreatment
alters [14C]chlordecone and [14C]cholesterol transport kinetics in the perfused rat
liver. Fundam Appl Toxicol 22, 286-292.
Guzelian, P.S. (1982). Comparative toxicology of chlordecone (kepone) in humans
and experimental animals. Ann. Rev. Pharmacol. Toxicol. 22, 89-113.
Hobbs, H. H., and Rader, D. J. (1999). ABC1: connecting yellow tonsils,
neuropathy, and very low HDL. J Clin Invest 104, 1015-1017.
Ishikawa, T., McNeeley, S., Steiner, P. M., Glueck, C. J., Mellies, M., Gartside, P.
S., and McMillin, C. (1978). Effects of chlorinated hydrocarbon on plasma alphalipoprotein cholesterol in rats. Metabolism 27, 89-96.
Kosters, A., Frijters, R. J., Kunne, C., Vink, E., Schneiders, M. S., Schaap, F. G.,
Nibbering, C. P., Patel, S. B., and Groen, A. K. (2005). Diosgenin-induced biliary
cholesterol secretion in mice requires Abcg8. Hepatology 41, 141-150.
Kosters, A., Kunne, C., Looije, N., Patel, S. B., Oude Elferink, R. P., and Groen,
A. K. (2006). The mechanism of ABCG5/ABCG8 in biliary cholesterol secretion
in mice. J Lipid Res 47, 1959-1966.
Krieger, M. (1999). Charting the fate of the "good cholesterol": identification and
characterization of the high-density lipoprotein receptor SR-BI. Annu Rev Biochem
68, 523-558.
Lee, J. Y., and Parks, J. S. (2005). ATP-binding cassette transporter AI and its role
in HDL formation. Curr Opin Lipidol 16, 19-25.
Lindstrom, F. T., Gillett, J. W., and Rodecap, S. E. (1974). Distribution of HEOD
(dieldrin) in mammals. I. Preliminary model. Arch Environ Contam Toxicol 2, 942.
Lordo, R. A., Dinh, K. T., and Schwemberger, J. G. (1996). Semivolatile organic
compounds in adipose tissue: estimated averages for the US population and
selected subpopulations. Amer J Pb Hlth 86, 1253-1259.
100
Luellen, D. R., Vadas, G. G., and Unger, M. A. (2006). Kepone in James River
fish: 1976-2002. Sci Total Environ 358, 286-297.
Mick, D. L., Long, K. R., Dretchen, J. S., and Bonderman, D. P. (1971). Aldrin
and Dieldrin in human blood components. Arch Environ Health 23, 177-180.
Miettinen, T. A., Ahrens, E. H., and Grundy, S. M. (1965). Quantitative isolation
and gas-liquid chromatographic analysis of total dietary and fecal neutral steroids.
J. Lipid Res. 6, 411-424.
Oram, J. F., and Vaughan, A. M. (2006). ATP-Binding cassette cholesterol
transporters and cardiovascular disease. Circ Res 99, 1031-1043.
Out, R., Hoekstra, M., Spijkers, J. A., Kruijt, J. K., van Eck, M., Bos, I. S., Twisk,
J., and Van Berkel, T. J. (2004). Scavenger receptor class B type I is solely
responsible for the selective uptake of cholesterol esters from HDL by the liver and
the adrenals in mice. J Lipid Res 45, 2088-2095.
Peet D.J., Janowski B.A., Mangelsdorf D.J., 1998, The LXRs: a new class of
oxysterol receptors, Curr Opin Genet Dev. 5, 571-5.
Poland, A. and Knutson, J. C. (1982). 2, 3, 7, 8- Tetrachlordibenzo-p-dioxin and
related halogenated aromatic hydrocarbons: examination of the mechanism of
toxicity. Ann. Rev Pharmacol Toxicol 22, 517-554.
Repa, J. J., Berge, K. E., Pomajzl, C., Richardson, J. A., Hobbs, H., and
Mangelsdorf, D. J. (2002). Regulation of ATP-binding cassette sterol transporters
ABCG5 and ABCG8 by the liver X receptors alpha and beta. J Biol Chem 277,
18793-18800.
Soine, P. J., Blanke, R. V., Chinchilli, V. M., and Schwartz, C. C. (1984a). Highdensity lipoproteins decrease the biliary concentration of chlordecone in isolated
perfused pig liver. J Toxicol Environ Health 14, 319-335.
Soine, P. J., Blanke, R. V., Guzelian, P. S., and Schwartz, C. C. (1982).
Preferential binding of chlordecone to the protein and high density lipoprotein
fractions of plasma from humans and other species. J Toxicol Environ Health 9,
107-118.
Soine, P. J., Blanke, R. V., and Schwartz, C. C. (1984b). Isolation of chlordecone
binding proteins from pig liver cytosol. J Toxicol Environ Health 14, 305-317.
Sporstol, M., Tapia, G., Malerod, L., Mousavi, S. A., and Berg, T. (2005).
Pregnane X receptor-agonists down-regulate hepatic ATP-binding cassette
transporter A1 and scavenger receptor class B type I. Biochem Biophys Res
Commun 331, 1533-1541.
Tabas, I. (2002). Cholesterol in health and disease. J Clin Invest 110, 583-590.
Voshol, P. J., Schwarz, M., Rigotti, A., Krieger, M., Groen, A. K., and Kuipers, F.
(2001). Down-regulation of intestinal scavenger receptor class B, type I (SR-BI)
101
expression in rodents under conditions of deficient bile delivery to the intestine.
Biochem J 356, 317-325.
Wang, N., Arai, T., Ji, Y., Rinninger, F., and Tall, A. R. (1998). Liver-specific
over expression of scavenger receptor BI decreases levels of very low density
lipoprotein ApoB, low density lipoprotein ApoB, and high density lipoprotein in
transgenic mice. J Biol Chem 273, 32920-32926.
Wustner, D., Mondal, M., Huang, A., and Maxfield, F. R. (2004). Different
transport routes for high density lipoprotein and its associated free sterol in
polarized hepatic cells. J Lipid Res 45, 427-437.
Yu, L., Gupta, S., Xu, F., Liverman, A. D., Moschetta, A., Mangelsdorf, D. J.,
Repa, J. J., Hobbs, H. H., and Cohen, J. C. (2005). Expression of ABCG5 and
ABCG8 is required for regulation of biliary cholesterol secretion. J Biol Chem 280,
8742-8747.
Yu, L., Hammer, R. E., Li-Hawkins, J., Von Bergmann, K., Lutjohann, D., Cohen,
J. C., and Hobbs, H. H. (2002). Disruption of Abcg5 and Abcg8 in mice reveals
their crucial role in biliary cholesterol secretion. Proc Natl Acad Sci U S A 99,
16237-16242.
102
Table 1. Body and organ weight of vehicle control and CD pretreated
14
mice that received a challenge dose of [ C]CH.
Body
Control
5.0 mg
CD/kg
15.0 mg
CD/kg
mass (g)
4 hr
16 hr
22.2 ± 0.8
21.9 ± 0.4
22.5 ± 0.7
21.8 ± 0.4
22.9 ± 0.7
22.3 ± 0.3
Liver (g)
4 hr
1.21 ±
0.03
1.23 ±
0.05
1.32 ±
0.05
16 hr
1.19 ±
0.05
1.22 ±
0.04
1.22 ±
0.02
Gallbladder (mg)
4 hr
16 hr
15.0 ± 0.9
9.2 ± 0.5*
17.4 ± 0.7
10.7 ± 1.1*
16.2 ± 0.7
8.9 ± 0.5*
There were 6-7 mice in each group. Values are expressed mean ± SE.
*Significantly different from the 4 hr (p < 0.05).
103
14
Figure 1. [ C]CH equivalents and total CH concentration in plasma of vehicle
14
control and CD pretreated mice that received a challenge dose of [ C]CH.
14
Male C57BL16 mice were pretreated with corn oil, 5, or 15 mg CD/kg. [ C]CH
(10 mg/kg in 5 ml/kg corn oil) was administered 3 days following pretreatment.
14
Determinations were performed 4 and 16 h after ip administration of [ C]CH.
14
[ C]CH equivalents of plasma and HDL were determined as described under
“Material and methods”. Plasma total CH and HDL-CH was determined
enzymatically as described under “Material and Methods”. Non-HDL-CH content
14
was taken as the difference between the total plasma CH and HDL-CH. A. [ C]CH
14
equivalents of plasma 4 h after ip lipid bolus. B. [ C]CH equivalents of plasma 16
h after ip lipid bolus. C. Plasma total CH 4 h after ip lipid bolus. D. Plasma total
CH 16 h after ip lipid bolus. Values are expressed mean ± SE(n=6 or 7).
*Significantly different from the 4 h (p < 0.05).
104
14
Figure 2. The ratio of [ C]CH equivalents in HDL to that in plasma and
hepatic SR-BI contents. Treatments were as described in Figure 1. A. The ratio of
14
[ C]CH equivalents in HDL to that in plasma. Values are expressed as mean ± SE.
B. Hepatic SR-BI contents. Corn oil or 15 mg CD/kg was treated to C57BL/6
mice. Hepatic plasma membrane fractions were prepared from animals (6 mice in
each group) as described under “material and methods”. Hepatic SR-BI content
was measured by western blotting with SR-BI antibody. There were 6-7 mice in
*
each group. Indicates a statistically significant difference (p < 0.05) when
compared with control.
105
A
14
Figure 3. Hepatic disposition of [ C]CH equivalents of vehicle control and CD
14
pretreated mice that received a challenge dose of [ C]CH. Treatments were as
14
described in Figure 1. A. Hepatic [ C]CH equivalents 4 and 16 h after ip bolus
14
lipid of [ C]CH. B. Hepatic subcellular fraction to homogenate ratios. Hepatic
subcellular fractions were prepared and analyzed from animals (6 mice in each
group) as described under “material and methods”. Data are normalized by the
14
amount of [ C]CH equivalents per mg protein in the liver homogenate. Values are
*
expressed as mean ± SE. There were 6-7 mice in each group. Indicates a
statistically significant difference (p < 0.05) when compared with control.
Indicates a statistically difference (p< 0.1) when compared with control.
106
14
Figure 4. [ C]CH equivalents and total CH concentration in gallbladder and
hepatic ABCG8 content. Treatments were as described in Figure 1. A. Biliary
14
14
[ C]CH equivalents 4 and 16 h after ip administration of [ C]CH. B. Biliary total
14
CH concentrations were determined 16 h after ip administration of [ C]CH using
an enzymatic method. C. Hepatic ABCG8 contents. Hepatic plasma membrane
fractions were prepared from animals (6 mice in each group) as described under
“material and methods”. Hepatic ABCG8 level was measured by western blotting
with ABCG8 antibody. Values are expressed as mean ± SE. There were 6-7 mice
*
in each group. The asterisk indicates a statistically significant difference (p <
0.05) when compared with control. +Indicates a statistically difference (p< 0.1)
when compared with control.
107
Appendix B
Chlordecone, a Mixed Pregnane X Receptor (PXR) and Estrogen Receptor
Alpha (ERα) Agonist, Alters Cholesterol Homeostasis and Lipoprotein
Metabolism in C57BL/6Mice
Junga Lee, Richard C. Scheri, Yuan Zhang, and Lawrence R. Curtis
108
ABSTRACT
Chlordecone (CD) is one of many banned organochlorine (OC) insecticides that
are widespread persistent organic pollutants. OC insecticides alter lipid
homeostasis in rodents at doses that are not neurotoxic or carcinogenic.
Pretreatment of mice or rats with CD altered tissue distribution of a subsequent
dose of [14C]CD or [14C]cholesterol (CH). Nuclear receptors regulate expression of
genes important in the homeostasis of CH and other lipids. In this study, we report
that CD suppresses in vitro reporter systems for human liver X receptor (LXRs)
and activates those for human farnesoid X receptor (FXR), pregnane X receptor
(PXR) and estrogen receptor α (ERα) in a concentration-dependent manner (0-50
μM). Consistent with human PXR activation in vitro, three days after a single dose
of CD (15 mg/kg) hepatic microsomal CYP3A11 protein increases in C57BL/6
mice. CD decreases hepatic CH ester content without altering total CH
concentration. Apolipoprotein A-I (apoA-I) contents of hepatic lipoprotein-rich
and microsomal fractions of CD treated mice are higher than controls. There is a
significant reduction in non-high density lipoprotein CH but not apolipoprotein B48/100 (apoB-48/100) in plasma from CD treated mice after a 4 hr fast.. At 14 days
after 15 mg CD/kg apoA-I and apoB-100 proteins but not CYP3A11 protein in
hepatic microsomes are similar to controls. This work indicates altered CH
homeostasis is a mode of OC insecticide action of relevance after a single dose.
This at least partially explains altered CH tissue distribution in CD-pretreated mice.
Keywords: apolipoproteins, CH, pregnane X receptor, estrogen receptor alpha,
liver x receptors, organochlorine insecticides, chlordecone.
109
INTRODUCTION
Banned organochlorine (OC) insecticides are widespread persistent organic
pollutants that bioaccumulate in wildlife and humans. While OC insecticides are
structurally diverse, they share a number of modes of toxic action. They are
neurotoxic through their interactions with ion channels (Narahashi et al. 1998).
Some OC insecticides, chlordecone (CD) for example, are carcinogenic in rodent
bioassays (Reuber 1978). Others, such as dieldrin, are tumor promoters in rodent
models (Kolaja et al. 1998). Alterations in lipid homeostasis occur in rodents at
OC insecticide doses that are not overtly neurotoxic or carcinogenic. Single doses
of CD and dieldrin decrease plasma triglycerides and total CH in rats 21 and 60
days after administration, respectively (Ishikawa et al. 1978). Carpenter et al.
(1996) report dose-dependent loss of lipid droplets in hepatocytes, but not Ito cells,
of mice after 5-40 mg CD/kg. A single dose of 5 mg CD/kg also alters tissue
distribution of exogenous [14C]CH (Carpenter and Curtis 1991; Lee et. at., 2008).
The mechanisms that underlie OC insecticide alterations in lipid homeostasis are
unclear. The work below employs CD since its toxicities are representative of the
OC insecticide class. It is novel in that it is refractory to metabolism in rodents
(Guzelian 1982); therefore contributions of metabolites of this OC to altering CH
regulatory pathways are highly unlikely.
Liver X receptors (LXRs), farnesoid X receptor (FXR), peroxisome proliferatoractivated receptors (PPARs), and pregnane X receptor (PXR) are nuclear receptors
which function as ligand activated transcription factors (Makishima et al. 1999;
Parks et al. 1999; Peet et al. 1998). These receptors bind specific response
elements in 5 ′ upstream promoter regions of genes after heterodimerization with
110
retinoid X receptor (RXR). LXRs respond to elevated sterol concentration and
increase the transcription of the Cyp7a1 gene that governs oxidation of CH to bile
acids. Physiological concentrations of bile acids activate FXR and thereby repress
the transcription of the Cyp7a1 gene (Makishima et al. 1999). Therefore, LXR and
FXR coordinate CH homeostasis by regulating feed forward and feed back
pathways, respectively. PPARs play a major role as lipid metabolism regulators.
They are also suggested as controllers of a CH oxidation pathway. PPARα
modulates expression of Cyp8b1 (sterol 12α-hydroxylase), a hepatic microsomal
enzyme involved in the biosynthesis of bile acids (Hunt et al. 2000). Recent work
suggests a potential role of PXR in CH metabolism. Bile acids activate PXR,
which reduces their production through repression of the Cyp7a1 gene (Staudinger
et al. 2001a; 2001b; Xie et al. 2001). Most banned OC insecticides bind and
activate PXR (Goodwin et al. 2002) but exhibit negligible affinity for the aryl
hydrocarbon receptor (Poland and Knutson, 1982). Some, including p,o-DDT, CD,
and dieldrin are also estrogenic in whole animals or cell based reporter gene assays
(Charles et al. 2002; Donohoe and Curtis 1996). CD competes for high affinity
estrogen binding sites in rainbow trout liver and increases plasma concentration of
the estrogen receptor (ERα)-dependent lipoprotein, vitellogenin (Donohoe and
Curtis 1996). Estrogen increases apolipoprotein A-I (apoAI) secretion in Hep G2
cells, but ERα probably does not directly mediate this (Lamon- Fava et al. 1999).
Modulating transcription of the apoA-I gene promoter appears more likely. PXR
agonists, but not a selective constitutive androstane receptor (CAR) agonist,
increase expression of the apoA-I gene in mice (Bachmann et al. 2004). ApoA-I
protein is a principal component of high density lipoprotein (HDL). HDL is the
lipoprotein central to CH transport from peripheral tissues to the liver (reverse CH
111
transport)(Lee and Parks 2005). This research addresses the hypothesis that
modulation of lipoprotein metabolism at least partially explains CD alterations in
CH homeostasis in male C57BL/6 mice.
MATERIALS AND METHODS
Chemicals
Chlordecone (CD, 99 % purity) was purchased from Chem Service (West Chester,
PA) and purity was verified by GC-EI/MS. LG268 and T0901317 were acquired
from Ligand Pharmaceuticals (San Diego, CA) and Cayman Chemical (Chicago,
IL), respectively. [1-14C] Lauric acid (55 mCi/mmol) was obtained from American
Radiolabeled Chemicals. Inc. (St. Louis, MO). Other chemicals including 17βestradiol (E2) were obtained from Sigma (St. Louis, MO).
Animals
Male C57BL/6 mice, 6 to 7-weeks-old (20-25 g in weight), were obtained from
Simeonson Laboratory (Gilroy, CA). The animals were housed in a temperature
controlled room (22 ± 1ºC) with a daily cycle of 12 hr of light and 12 hr darkness
and fed ad libitum with AIN93 diet (Dyets, Inc., Bethlehem, PA) with free access
to water. Treatments were initiated after seven days of acclimatization. CD was
dissolved in corn oil. Mice received CD (5 or 15 mg/kg body weight) or corn oil
alone by intraperitoneal injection (5 ml/kg body weight). After 3 or 14 days,
animals were fasted for 4 hr and killed by CO2 anesthesia and exsanguinations.
Blood, liver, and gallbladder were collected. The procedures for animal use were
approved by the Oregon State University Institutional Animal Care and Use
Committee.
112
Cell culture and Reporter gene assay
The human embryonic kidney cell line, HEK293, was maintained at 37 0C, 95% O2
and 5 % CO2 in DMEM containing 10 % fetal bovine serum. In some experiments
HEK293 cells were cotransfected with a plasmid encoding a fusion protein of the
Gal4 DNA binding domain and a nuclear receptor ligand-binding domain along
with a luciferase reporter plasmid-containing response element. In other
experiments HEK293 cells were contransfected with plasmids encoding full length
RXRα with LXRα, LBRβ, or FXR along with a luciferase reporter containing the
LXR response element (LXRE) or FXR response element (FXRE). CMX-β-gal was
contransfected as an internal control. Luciferase cotransfection assays were
performed as described (Makishima et al. 2002). Nuclear receptor ligand binding
domains were from human LXRα, LXRβ, FXR, RXR, ERα and ERβ, or murine
PPARα, PPARδ, PPARγ. Cells were treated with vehicle (ethanol) alone or 1 to 50
μM CD and/or 1 nM T0901317 (LXR agonist), 100 nM LG268 (RXR agonist), and
1 nM E2 (ER agonist). Cell morphology by light microscopy and β-galactosidase
activity assessed potential cytotoxicity. Activation of nuclear receptors was
assessed by measuring luciferase activity. Individual assays were repeated at least
three times.
Tissue preparation
Blood was collected by cardiac puncture in heparinized syringes and kept on ice
until plasma was isolated by centrifugation at 3000 g for 25 min at 4 0C. Plasma
was stored at -80 0C. Liver was homogenized in buffer (0.01 M potassium
phosphate, pH 7.5; 0.15 M KCl; 1.0 mM ethylenediaminetetraacetic acid (EDTA);
113
0.1 mM butylated hydroxytoluene (BHT); 0.1 mM phenylmethylsulfonyl fluoride
(PMSF)) with a polytron PT 3000 (Brinkmann, Westbury, NY). Liver homogenate
was centrifuged at 12,000 g for 30 min. The supernatant was again centrifuged at
100,000 g for 90 min at 4 0C with Ti 70 rotor (Beckman, Fullerton, CA). The
floating fatty layer in the supernatant was regarded as lipoprotein-rich fraction. The
remaining supernatant and resulting pellet were cytosolic and microsomal fractions,
respectively. Microsomes were resuspended in microsome resuspension buffer (0.1
M potassium phosphate, pH 7.25; 1.0 mM EDTA; 30% glycerol; 0.1 mM PMSF,
1.0 mM DTT (dithiothreitol), and 20 μM BHT). Protein concentration was
determined by the BCA assay (Pierce Biotechnology, Inc. Rockford, IL).
Western blot analysis
Mouse liver subcellular fractions (lipoprotein-rich, cytosol and microsomes) were
prepared from control, 5 or 15 mg CD/kg treated animals as above. Proteins from
each preparation were separated on Tris/glycine gels (Bio-Rad, Richmond, CA)
under reducing conditions. Following gel electrophoresis at 100 V for 2 h, proteins
were electrophoretically transferred onto PVDF membranes utilizing a Mini transblot Rabbit anti-CYP7A antibody was a generous gift from Professor John Y. L.
Chiang (Northwestern Ohio University) or purchased from Santa Cruz
Biotechnology (Santa Cruz, CA). Rabbit anti-CYP3A antibody was a generous gift
from Professor Donald Buhler (Oregon State University). Goat anti-CYP4A
antibody was purchased from BD Gentest (Woburn, MA). ApoAl and ApoB
antibodies were purchased from Biodesign (Saco, ME). The blots were then probed
with horseradish peroxidase conjugated goat anti-rabbit IgG or rabbit anti-goat IgG
as the secondary antibody (Bio-Rad, Richmond, CA). The proteins were detected
114
after development with enhanced chemiluminescence (ECL) detection (Amersham,
Piscataway, NJ). Quantification of the intensity of the protein bands was performed
with NIH-image software.
Enzyme activities
Liver enzyme activities were analyzed with a HPLC (Waters 2690) and a reverse
phase C18 HPLC column (4.6 mm × 25 cm, 5 μm, ultrasphere, Beckman). In the
assay of CH 7α-hydroxylase, reaction products after the enzymatic conversion of
7α- hydroxylase-CH to 7α-hydroxy-4-cholesten-3-one (7α-HCO) by CH oxidase
(Sigma, St. Louis, MO) were detected by a UV spectrophotometer (Chiang 1991).
[1-14C] Laurate hydroxylation was measured using slight modifications of the
procedure described by Williams et al. (1984). Samples were analyzed using 62 %
methanol containing 0.2 % acetic acid to elute ω- and ω-1 hydroxylaurate at a flow
rate of 1 ml/min and the mobile phase was switched to 100 % methanol to elute
parent compound and detected by Flo-One TR505 radioactivity flow monitor.
Plasma lipid analyses
Total plasma CH concentrations were determined after a 4 h fast using enzymatic
kits (Sigma Diagnostic, St. Louis, MO). Plasma HDL-CH concentrations were
measured after precipitation of apolipoprotein-B containing particles with dextran
sulfate and magnesium (Rachem, San Diego, CA). Plasma contents of apoA-I and
apoB-48/100 were quantified by densitometric scanning following electrophoresis
on 4-15% Tris/glycine gels (Bio-Rad, Richmond, CA) and transfer onto PVDF
membranes (Bio-Rad, Richmond, CA). The membranes were incubated with rabbit
anti-apoA-I or rabbit anti-apoB-48/100 polyclonal antibodies (Biodesign, Saco,
115
ME) and probed with horseradish peroxidase conjugated goat anti-rabbit IgG as the
secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA). The proteins
were visualized by enhanced chemiluminescence (ECL) detection (Amersham,
Piscataway, NJ).
Hepatic CH content
For extraction of lipids, ~100 mg homogenized liver was incubated in 1.5 N KOH
in ethanol/water (2:1) at 700C for 30 min. Following incubation, 2 ml hexane was
added and the solution was inverted three times and centrifuged for 3 min at 3,000
g. The lipid layer (upper phase) of each sample was collected and dried under
nitrogen. Dried lipids were dissolved in 2-propanol containing 10 % Triton-X-100
as assay samples. The CH concentration was determined using a CH/Cholesteryl
ester Quantitation KitR (BioVision, Mt. View, CA). Hepatic esterified CH
concentration was taken as the difference between the total CH and free CH.
Statistical Methods
Statistical analyses were conducted with StatGraphics software (StatPoint,
Herndon, VA). Normality of data sets were assessed with standardized kurtosis,
standardized skewness, and homogeneity of variance. Data that failed one or more
tests for normality were log base 10 transformed. Transformed data were assessed
for normality with the criteria specified above. Data were expressed as means ± SE.
Comparisons among groups validated for normality were submitted to a one-way
analysis of variance (ANOVA). For comparison between two groups, Student’s ttest with equal-variance was applied. In all analyses, a 95% confidence level was
116
used as the criterion for significance. For outlier identification, Grubbs' Test
(assumes normality) was applied
RESULTS
Cell morphology and β-galactosidase activity were not significantly altered by CD
(data not shown). A series of reporter gene cotransfection assays in HEK 293 cells
characterized potential interactions of CD with nuclear receptors involved with
lipid homeostasis. CD activated or inhibited transactivation by several nuclear
receptors (Fig. 1). At 10 μM CD there was mild but statistically significant
suppression of reporter construct alone and RXRα (about 30 %)(Fig. 1A). CD
more substantially inhibited LXRα and strongly suppressed activation for LXRβ.
CD modestly but significantly activated (about 1.3-fold) FXR and PPARα and
strongly activated PXR about 3.5-fold over background (Fig. 1A, 1B). PPARδ and
PPARγ were not affected by CD treatment (Fig. 1B).
Activation of human ERs by CD was also assessed. CD significantly activated a
chimera consisting of DNA binding domain of Gal4 and the ligand binding domain
of ERα slightly but significantly at 1 μM and almost 10-fold at 10 μM and 50 μM
(Fig. 2A). There was no CD significant effect on human ERβ activity (Fig. 2A) but
CD repressed E2-induced ERβ activation (40 %) (Fig. 2B). There was no
significant difference in activation of ERα by E2 in the presence or absence of 10
μM CD (Fig. 2B).To further characterize CD (Fig. 3A) activation of nuclear
receptors involved in lipid homeostasis not previously assessed with regard to OC
insecticides, a transient transfection assay using a synthetic LXR or FXR
responsive reporter plasmid was assessed. CD alone strongly suppressed activities
117
of the murine LXRα-RXR and LXRβ-RXR heterodimers in a concentration
dependent manner (Fig. 3B and 3C). CD (10 μM) strongly inhibited LXRα-RXR
activation by LG268 (RXR agonist) (Fig. 3B), but insignificantly inhibited the
receptor activation by T0901317 (LXR agonist)(data not shown). Fifty μM CD
moderately but significantly repressed activation of LXRβ-RXR activated by
T0901317 (Fig. 3C). CD (50 μM) activated FXR approximately 50 % as
effectively as the specific RXR agonist LG268 (Fig. 3D). In the presence of LG268,
CD increased FXR transactivation cooperatively (Fig. 3D).
The relevance of CD activation of nuclear receptors in vitro was assessed with in
vivo assays in C57BL/6 mice. Interactions with LXRα and FXR were assessed by
hepatic microsomal CYP7A1 protein immunoquantitation and CH 7α-hydroxylase
activity. There was a trend for decreased CYP7A1 protein content and its enzyme
activity but it was not statistically significant (Fig. 4A, 4B). Interaction with
PPARα was assessed by hepatic microsomal CYP4A1 protein immunoquantitation
and lauric acid ω-1 hydroxylase activity. There was no evidence for activation of
this receptor by the CD doses administered to mice (Fig. 4A, 4B). Finally,
interaction with PXR was assessed by hepatic microsomal CYP3A11 protein
immunoquantitation. CYP3A11 protein level was highly increased (4-fold) in
livers from CD treated animals (Fig. 4A). Increased hepatic microsomal CYP3A11
protein after CD treatment indicated PXR activation.
The effect of CD on total plasma CH and HDL-CH concentration were measured
using enzymatic analyses. Plasma non-HDL-CH concentration was significantly
decreased by CD treatment (Fig. 5A). Plasma HDL-CH was unchanged but the
118
ratio of HDL-CH to plasma CH was mildly but significantly increased (Fig. 5A,
5B).
Immunoblot analysis of plasma revealed no significant difference in apoB-48/100
or apoA-I in CD treated mice (Fig. 5C). The effect of CD on apolipoprotein
content was further characterized in liver subcellular fractions (lipoprotein-rich,
cytosol and microsomes). Interestingly apoA-I significantly increased in the
hepatic lipoprotein-rich and microsomal fractions from CD treated mice (1.5-fold,
1.8-fold, respectively) (Fig. 6A). There was no significant difference in hepatic
microsomal apoB-100 between CD treated and control mice (Fig. 6B).
Hepatic total, free and esterified CH content was measured enzymatically. About
90 % of hepatic CH was in the unesterified form (Fig. 7). Hepatic total and free
CH contents from CD-treated mice were not different from control mice (Fig.7).
However, CD significantly decreased hepatic esterified CH concentration (about
40 % of control) (Fig. 7).
An ancillary experiment examined persistence of CD induced protein changes
in liver. Hepatic microsomal CYP3A11 was significantly higher in 15 mg CD/kg
treated mice than controls 14 days after injection (Fig. 8). Hepatic microsomal
CYP7A1 and CYP 4A1 were not different from controls in the same mice (Fig. 8).
Plasma and hepatic microsomal apoA-I were similar to controls 14 days after 15
mg CD/kg (Fig. 8). Hepatic microsomal apoB-100 was not different from controls
in the same mice (Fig. 8).
119
DISCUSSION
Humans and wildlife are exposed to complex mixtures of xenobiotics including
pesticides and other environmental contaminants. In the body, xenobiotics, dietary
nutrients, and endogenous substances may interact to modulate regulatory
pathways for intermediary metabolism. Inhibition or activation of nutrient
metabolism or transport pathways may yield beneficial or adverse effects. Prior
reports indicate that a single dose of CD, an OC insecticide, alters tissue
distribution of exogenous CH in mice and rats (Carpenter and Curtis 1991; Gilroy
et al. 1994; Lee et al. 2008). Subsequent experiments and other work suggests that
CD also modulates other aspects of lipid homeostasis (Carpenter et al. 1996;
Chetty et al. 1993). However, molecular mechanisms underlying these
observations are unclear.
Excessive CH contributes to heart attacks and stroke by promoting atherosclerosis
(Krieger 1999). Therefore, CH levels are precisely controlled in the body. The liver
is the major organ that controls CH homeostasis in mammals. The role of the liver
in CH homeostasis involves coordinate regulation of biosynthesis of CH, its uptake
from plasma and catabolism to bile acid (Brown and Goldstein 1999). . A large
body of literature shows that nuclear receptors play important roles in the
regulation of metabolism of CH and other lipids through a coordinated network of
transcriptional programs (Chawla et al. 2001; Lobaccaro et al. 2001; Makishima
2003). To increase understanding of how low doses of CD alter CH homeostasis,
we assessed CD interactions with several nuclear receptors in vitro and in vivo.
Whole livers of C57BL/6 mice contained 84 μM CD 16 h after 5 mg/kg (Carpenter
and Curtis 1989). Hepatic CD residues were persistent in livers with less than a
120
50% reduction from 3 to 14 days after exposure. Therefore, in vitro CD activation
or inhibition of nuclear receptors occurred at concentrations relevant to those
detected in livers of mice treated with doses examined herein. CD significantly
affected transactivation by a chimera consisting of the DNA binding domain of
Gal4 and the ligand-binding domains of LXRs and FXR (Fig. 1A). CD minimally
increased FXR activation and inhibited LXRs to different extents (Fig. 2B, 2C, 2D,
Fig. 1A). CD (50 μM) minimally increased FXR activation (Fig. 1A). Inhibition of
LXRβ was about twice as marked as that of LXRα. Additional in vitro experiments
and in vivo work further assessed CD interaction with LXRs and FXR (detailed
below).
PPARα, was slightly but significantly activated by 10 μM CD (Fig 1B). PPARδ
and PPARγ were not activated or inhibited by CD treatment (Fig. 1B). The CYP4A
proteins were generally recognized as up-regulated by PPARα agonists (Hunt et.
al., 2000). CD altered neither the CYP4A protein nor lauric acid ω-1 hydroxylase
activity (Fig. 4). Therefore, physiological relevance of PPAR contributions to CD
interaction with lipid homeostasis in mice seemed unlikely.
PXR was identified as a key regulator of CYP3A expression since it was activated
by diverse compounds that induced CYP3A expression and bound xenobiotic
response elements in CYP3A promoters (Xie et al. 2000). Because of evolutionary
divergence of the PXR ligand-binding domain in humans and mice, striking
species-specific differences were evident between two species (Kliewer 2003; Xie
et al. 2000). The human specific CYP3A inducer, rifampicin, had little effect on
CYP3A gene expression in mice, as did the mouse specific CYP3A inducer, PCN,
121
in human hepatocytes (Bachmann et al. 2004). Here CD not only transactivated
human PXR in an in vitro assay (Fig. 1B) but also increased hepatic microsomal
CYP3A11 protein in hepatic microsomes from mice (Fig. 4A). The data indicated
that CD activated PXR in both species. Recent results suggested that PXR strongly
induced the CYP3A4 gene by inhibiting small heterodimer partner (SHP) gene
transcription in human livers (Li and Chiang 2006).
Activation of ERα was required for estrogen-mediated protection against vascular
injury (Pare et al. 2002). CD was an ERα but not an ERβ agonist in the micromolar
range (Fig. 2A). CD activated human ERα cooperatively with E2, although the
activity was 10,000 times lower than E2 (Fig. 2B). This difference in potency was
similar to that reported for competition assays for estrogen binding sites in rainbow
trout liver (Donohoe and Curtis 1996). CD inhibited E2-induced ERβ activation
effectively (Fig. 2B). A role for ERα in CD altered CH homeostasis was of potential
importance since estrogens increased apoB expression (Srivastava et al. 1993; 1997).
Plasma non-HDL CH was elevated in CD treated compared to control mice that
received 5 ml corn oil plus 10 mg CH/kg 4 hr before sampling (Lee et al. 2008). The
reverse was observed in CD treated mice fasted 4 hr before sampling (Fig. 5A). CD
depleted lipid droplets and increased putative VLDL particles in mouse liver
(Carpenter et al. 1996). Perhaps low hepatic lipid masked CD-stimulated non-HDL
secretory capacity in fasted mice and this was reversed when exogenous lipid was
provided by ip corn oil. Increased clearance of apoB containing lipoproteins and/or
reduced VLDL/LDL-CH secretion potentially contributed to the decreased plasma
non-HDL-CH in CD treated fasted mice. E2 treatment increased low density
lipoprotein (LDL) receptor protein in livers of rats but not mice. (Srivastava et al.
1993). Activation of LXRs induced hepatic LDL-receptor expression (Masson et al.
122
2004). Since LXRα activation was inhibited by CD in vitro (Fig. 1A, 3B) and
LDL receptors were not up-regulated by E2 in mice, reduced VLDL/LDL-CH
production was the most plausible explanation for decreased plasma non-HDL-CH
concentration in CD treated fasted mice.
Presence or absence of exogenous lipid was a key experimental variable since
plasma non-HDL-CH decreased in fasted CD-treated mice compared to fasted
controls (Fig. 5A). ApoB is known to be synthesized constitutively and regulated
primarily by co and post-transitional mechanisms in the secretory pathway
(Avramoglu and Adeli 2004). ApoB secretion is regulated by ubiquitin-mediated
proteasomal degradation or non-proteasomal degradation pathways. There is a
positive correlation between hepatic availability of neutral lipids proximal to the
site of apoB synthesis and amount of hepatic apoB secretion (Avramoglu and
Adeli 2004). Reduced hepatic CH ester, but not free CH, inhibited VLDL apoB
production (Telford et al. 2005). Previous work demonstrated that CD decreased
cytosolic lipid droplets in hepatocytes in C57BL/6 mice (Carpenter et al. 1996).
CD decreased hepatic CH ester, not total or free CH (Fig. 8) but there was no
difference in apoB content in the liver microsomal fraction from CD treated
animals. Thus, insufficient lipid perhaps increased apoB degradation by ubiquitinmediated proteasomal degradation and reduced VLDL secretion.
Co-expression of FXR with exogenous RXR resulted in a significant activation
enhancement (about 5-fold over background) (Fig. 3D). This finding indicated the
requirement of RXR (heterodimer partner) for CD-induced FXR activation. These
observations demonstrated that the activation or inhibition required the ligand
123
binding domain of these receptors. CH 7α-hydroxylase (CYP7A) was the first and
a rate limiting enzyme in bile acid synthesis, the major CH elimination pathway
(Russell 1999). LXRα and FXR regulated the expression of CH 7α-hydroxylase in
opposite ways (Makishima 2003). Therefore, bile acid synthesis was regulated by
feed forward and feedback mechanisms. Although CD activated FXR and inhibited
LXRα activation in vitro, there was no statistically significant reduction of hepatic
CYP7A1 and its enzyme activity by CD treatment in C57BL/6 mice (Fig. 4A, 4B).
LXRα induced CYP7A1 by binding to the 5’-flanking region and increased
expression of mouse and rat 7α-hydroxlase. FXR repressed the Cyp7a1gene
indirectly through the induction of the transcriptional repressor small heterodimer
partner (SHP) (Staudinger et al. 2001a; 2001b). Hepatic content of other proteins
in CD treated mice indicated little role for LXRs or FXR in CD-altered CH
homeostasis. ATP binding cassette (ABC) G5 and scavenger receptor B1 (SRBI)
genes were positively regulated by LXRs (Kalaany and Mangelsdorf, 2006), but
hepatic contents of these proteins were not altered by CD (Lee et al., 2008).
Activation of FXR negatively regulated expression of apoA1 and apoB genes
(Kalaany and Mangelsdorf, 2006). There was no reduction in hepatic microsomal
apoB 100 and significantly increased apoA-I in that fraction for CD treated mice
(Fig. 6).
LXRβ knockout mice eliminated excess CH as effectively as wild type (Alberti et
al. 2001). However, recent studies demonstrated that the LXRβ isoform also
contributed to CH homeostasis through activation of ABCAI in macrophages, liver,
and intestine (Quinet et al. 2006; Repa et al. 2000). ABCAI was classified as a
member of a large family of ABC transporters (Langmann et al. 1999). The
124
ABCAI transporter increased efflux of intracellular CH to lipid-poor apoA-I to
form preβ−HDL or nascent HDL (Wang et al. 2000). While results of the reporter
construct assays (Fig. 2A) suggested potential for CD inhibition of LXRβ signaling
and disrupted ABCAI activation, there was no reduction in plasma HDL-CH in CD
treated mice (Fig. 5A).
HDL is an important source of CH for biliary excretion in rodents. HDL transports
CH from peripheral tissues to the liver by the reverse CH transport (RCT) pathway
(Srivastava 2003). ApoA-I is a major apolipoprotein in HDL particles. SRBI, a
physiological HDL receptor, facilitates the selective uptake of HDL-CH esters in
the RCT pathway (Rigotti et al. 1996; Ueda et al. 1999).
PXR activation increased HDL-CH and expression of apoA-I in vitro and in vivo
(Bachmann et al. 2004; Sporstol et al. 2005). Estrogens were suggested as
antiatherogenic since they reduced LDL-CH and elevated HDL-CH (Hargrove et
al.1999). While CD was a mixed agonist for human PXR and ERα, HDL-CH and
plasma apoA-I contents were not changed in fasted mice (Fig. 5A, 5C). Instead, we
observed that CD increased apoA-I content in hepatic lipoprotein-rich and
microsomal fractions (Fig. 6A). Gilroy et al. (1994) reported circumstantial
evidence that CD stimulated hepatic HDL secretion. CD was strongly associated
with HDL rather than other lipoproteins (Guzelian 1982). In situ rat livers were
perfused with HDL complexed [14C]CD for one hour, which was then replaced
with fresh solution. Efflux of [14C]CD into the perfusate from livers of CD
pretreated rats about doubled over one hr. compared to controls. Plasma HDL-tototal [14C]CH ratio increased in CD treated compared to control mice that received
125
5ml corn oil/kg 4 h prior to sampling (Lee et al, 2008). Therefore, lipid availability
was likely a key factor in secretion of apoAl-I containing lipoprotein by liver.
Since HDL-CH was widely considered as the good CH that provided some
protection from heart disease, the potential role of PXR agonists in raising HDLCH was an important area of current interest. However not all HDL-subclasses
were equally protective. Large HDL2b was believed more important for protection
from coronary heart disease in clinical observational studies (Pascot et al. 2001).
Studies demonstrated that PXR regulated expression of ABCAI negatively and
apoA-I positively in vitro and in vivo (Bachmann et al. 2004; Sporstol et al. 2005).
Clearance of [14C]CH complexed with HDL from perfusate was reduced in situ
liver preparations of CD-treated rats (Gilroy et al., 1994). These observations
indicated that CD perhaps alters lipoprotein metabolism by perturbing the RCT
pathway.
Hepatic CD concentrations declined slowly in C57BL/6 mice (Carpenter andCurtis
1989). Liver residues declined less than 50% from 3 to 14 days after
intraperitoneal doses similar to those used in this study. Elevated CYP3A11content
in hepatic microsomes 14 days after injection was consistent with persistent PXR
signaling (Fig. 8). Hepatic content of apoA-I and apoB were not elevated at that
time (Fig. 8). Perhaps differential adaptation of signaling pathways contributed to
this phenomenon.
Even though the United States bans most uses of persistent OC pesticides, they
continue to occur in tissues of humans and wildlife. Trophic transfer through food
webs that provide dietary fish, meat and dairy products is the principal route of
126
environmental exposure to these agents. Residues of OC insecticides occur in
tissues of human and wildlife as complex mixtures. In human fat from the United
States, the sum of OC insecticides averages in the nM range (Lordo et al 1996).
Concentrations for wildlife that are top predators are higher. For example, fat of
grizzly bears that feed heavily on Pacific salmon contain 25-fold higher maximum
concentrations (Christensen, et al, 2005) than humans. Since many OC insecticides
or their metabolites are PXR (Goodwin et al, 2002) and ER agonists (Charles et al,
2002; Donohoe and Curtis, 1996) their potential interactions with endocrine
regulation deserves attention (Rajapakse et al 2002). Perturbation of CH
homeostasis by a low dose of CD may represent an environmentally relevant mode
of endocrine disruption. Therefore, it is necessary to continue studies to explain
altered CH and lipid homeostasis by banned OC insecticides. In summary: (1) CD
was a mixed agonist for human PXR and ERα and an effective antagonist for
LXRβ. (2) CD probably modulated lipoprotein metabolism by multiple interactions
with nuclear receptors. (3) This work indicated altered CH homeostasis and
lipoproteins were a mode of OC insecticide action of potential environmental
relevance.
127
Acknowledgements:
This work was supported by the Oregon Agricultural Experiment Station
(ORE00871) and grant number T32 ES007060 from the National Institutes of
Health.
128
References
Alberti, S., Schuster, G., Parini, P., Feltkamp, D., Diczfalusy, U., Rudling, M.,
Angelin, B., Bjorkhem, I., Pettersson, S., and Gustafsson, J. A. (2001). Hepatic
cholesterol metabolism and resistance to dietary cholesterol in LXRbeta-deficient
mice. J Clin Invest 107, 565-573.
Avramoglu, R. K., and Adeli, K. (2004). Hepatic regulation of apolipoprotein B.
Rev. Endocrine Metabolic Disorders 5, 293-301.
Bachmann, K., Patel, H., Batayneh, Z., Slama, J., White, D., Posey, J., Ekins, S.,
Gold, D., and Sambucetti, L. (2004). PXR and the regulation of apoA1 and HDLcholesterol in rodents. Pharmacol Res 50, 237-246.
Brown, M. S., and Goldstein, J. L. (1999). A proteolytic pathway that controls the
cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci U S A 96,
11041-11048.
Carpenter, H. M., and Curtis, L. R. (1989). A characterization of chlordecone
pretreatment-altered pharmacokinetics in mice. Drug Metab Dispos 17, 131-138.
Carpenter, H. M., and Curtis, L. R. (1991). Low dose chlordecone pretreatment
altered cholesterol disposition without induction of cytochrome P-450. Drug
Metab Dispos 19, 673-678.
Carpenter, H. M., Hedstrom, O. R., Siddens, L. K., Duimstra, J. R., Cai, Z. W.,
Fisher, K. A., and Curtis, L. R. (1996). Ultrastructural, protein, and lipid changes
in liver associated with chlordecone treatment of mice. Fundam Appl Toxicol 34,
157-164.
Charles, G. D., Gennings, C., Zacharewski, T. R., Gollapudi, B. B., and Carney, E.
W. (2002). Assessment of interactions of diverse ternary mixtures in an estrogen
receptor-alpha reporter assay. Toxicol Appl Pharmacol 180, 11-21.
Chawla, A., Repa, J. J., Evans, R. M., and Mangelsdorf, D. J. (2001). Nuclear
receptors and lipid physiology: opening the X-files. Science 294, 1866-1870.
Chetty, K. N., Walker, J., Brown, K., and Ivie, G. W. (1993). The effects of dietary
calcium and chlordecone on cholinesterase, triglycerides, low density lipoproteins,
and cholesterol in serum of rat. Arch Environ Cont Toxicol, 24, 365-367.
Chiang, J. Y. (1991). Reversed-phase high-performance liquid chromatography
assay of cholesterol 7 alpha-hydroxylase. Methods Enzymol 206, 483-491.
Christensen, J.R., Mac Duffee, M., Mac Donald, R. W., Whiticar, M., and Ross, P.
S. (2005). Persistent organic pollutants in British Columbia grizzly bears:
consequence of divergent diets. Environ Sci Technol 39, 6952-6960.
Donohoe, R., and Curtis, L. (1996). Estrogenic activity of chlordecone, o,p'-DDT,
and o,p'-DDE in juvenile rainbow trout: induction of vitellogenesis and interaction
129
with hepatic estrogen binding sites. Aquat Toxicol 36, 31-52
Gilroy, D. J., Carpenter, H. M., and Curtis, L. R. (1994). Chlordecone pretreatment
alters [14C]chlordecone and [14C]cholesterol transport kinetics in the perfused rat
liver. Fundam Appl Toxicol 22, 286-292.
Goodwin, B., Redinbo, M. R., and Kliewer, S. A. (2002). Regulation of cyp3a
gene transcription by the pregnane x receptor. Annu Rev Pharmacol Toxicol 42, 123.
Guzelian, P. S. (1982). Comparative toxicology of chlordecone (Kepone) in
humans and experimental animals. Annu Rev Pharmacol Toxicol 22, 89-113.
Hargrove, G. M., Junco, A., and Wong, N. C. (1999). Hormonal regulation of
apolipoprotein AI. J Mol Endocrinol 22, 103-111.
Hunt, M. C., Yang, Y. Z., Eggertsen, G., Carneheim, C. M., Gafvels, M.,
Einarsson, C., and Alexson, S. E. (2000). The peroxisome proliferator-activated
receptor alpha (PPARalpha) regulates bile acid biosynthesis. J Biol Chem 275,
28947-28953.
Ishikawa, T., McNeeley, S., Steiner, P. M., Glueck, C. J., Mellies, M., Gartside, P.
S., and McMillin, C. (1978). Effects of chlorinated hydrocarbon on plasma alphalipoprotein cholesterol in rats. Metabolism 27, 89-96.
Kliewer, S. A. (2003). The nuclear pregnane X receptor regulates xenobiotic
detoxification. J Nutr 133, 2444S-2447S.
Kalaany, N. Y. and Mangelsdorf, D. J. (2006). LXRs and FXR: The yin and yang
of cholesterol and fat metabolism. Ann Rev Physiol 68: 159-191.
Kolaja, K. L., Xu, Y., Walborg, E. F., Jr., Stevenson, D. E., and Klaunig, J. E.
(1998). Vitamin E modulation of dieldrin-induced hepatic focal lesion growth in
mice. J Toxicol Environ Health A 53, 479-492.
Krieger, M. (1999). Charting the fate of the "good cholesterol": identification and
characterization of the high-density lipoprotein receptor SR-BI. Annu Rev Biochem
68, 523-558.
Lamon-Fava, S., Ordovas, J. M., and Schaefer, E. J. (1999). Estrogen increases
apolipoprotein (apo) A-I secretion in hep G2 cells by modulating transcription of
the apo A-I gene promoter. Arterioscler Thromb Vasc Biol 19, 2960-2965.
Langmann, T., Klucken, J., Reil, M., Liebisch, G., Luciani, M. F., Chimini, G.,
Kaminski, W. E., and Schmitz, G. (1999). Molecular cloning of the human ATPbinding cassette transporter 1 (hABC1): evidence for sterol-dependent regulation
in macrophages. Biochem Biophys Res Commun 257, 29-33.
Lee, J., Scheri, R. C., and Curtis, L. R. (2008). Chlordecone altered hepatic
disposition of [14C] cholesterol and plasma cholesterol distribution but not SR-B1
or ABCG8 proteins in livers of C57BL/6 mice. Toxicol Appl Pharmacol In press.
130
Lee, J. Y., and Parks, J. S. (2005). ATP-binding cassette transporter AI and its role
in HDL formation. Curr Opin Lipidol 16, 19-25.
Li, T., and Chiang, J. Y. (2006). Rifampicin induction of CYP3A4 requires
pregnane X receptor cross talk with hepatocyte nuclear factor 4 alpha and
coactivators, and suppression of small heterodimer partner gene expression. Drug
Metab Dispos 34, 756-764.
Lobaccaro, J. M., Repa, J. J., Lu, T. T., Caira, F., Henry-Berger, J., Volle, D. H.,
and Mangelsdorf, D. J. (2001). [Regulation of lipid metabolism by the orphan
nuclear receptors]. Ann Endocrinol (Paris) 62, 239-247.
Lordo, R. A., Dinh, K. T., Schwemberger, J. G. (1996). Semivolatile organic
compounds in adipose tissue: estimated averages for the U.S. population and
selected subpopulations. Amer J Pub Hlth 86, 1253-1259.
Makishima, M. (2003). Lipid metabolism and nuclear receptors, Seikagaku 75,
391-395.
Makishima, M., Lu, T. T., Xie, W., Whitfield, G. K., Domoto, H., Evans, R. M.,
Haussler, M. R., and Mangelsdorf, D. J. (2002). Vitamin D receptor as an intestinal
bile acid sensor. Science 296, 1313-1316.
Makishima, M., Okamoto, A. Y., Repa, J. J., Tu, H., Learned, R. M., Luk, A.,
Hull, M. V., Lustig, K. D., Mangelsdorf, D. J., and Shan, B. (1999). Identification
of a nuclear receptor for bile acids. Science 284, 1362-1365.
Masson, D., Staels, B., Gautier, T., Desrumaux, C., Athias, A., Le Guern, N.,
Schneider, M., Zak, Z., Dumont, L., Deckert, V., Tall, A., Jiang, X. C., and
Lagrost, L. (2004). Cholesteryl ester transfer protein modulates the effect of liver
X receptor agonists on cholesterol transport and excretion in the mouse. J Lipid
Res 45, 543-550.
Narahashi, T., Ginsburg, K. S., Nagata, K., Song, J. H., and Tatebayashi, H.
(1998). Ion channels as targets for insecticides. Neurotoxicology 19, 581-590.
Pare, G., Krust, A., Karas, R. H., Dupont, S., Aronovitz, M., Chambon, P., and
Mendelsohn, M. E. (2002). Estrogen receptor-alpha mediates the protective effects
of estrogen against vascular injury. Circ Res 90, 1087-1092.
Parks, D. J., Blanchard, S. G., Bledsoe, R. K., Chandra, G., Consler, T. G.,
Kliewer, S. A., Stimmel, J. B., Willson, T. M., Zavacki, A. M., Moore, D. D., and
Lehmann, J. M. (1999). Bile acids: natural ligands for an orphan nuclear receptor.
Science 284, 1365-1368.
Pascot, A., Lemieux, I., Prud'homme, D., Tremblay, A., Nadeau, A., Couillard, C.,
Bergeron, J., Lamarche, B., and Despres, J. P. (2001). Reduced HDL particle size
as an additional feature of the atherogenic dyslipidemia of abdominal obesity. J
131
Lipid Res 42, 2007-2014.
Peet, D. J., Turley, S. D., Ma, W., Janowski, B. A., Lobaccaro, J. M., Hammer, R.
E., and Mangelsdorf, D. J. (1998). Cholesterol and bile acid metabolism are
impaired in mice lacking the nuclear oxysterol receptor LXR alpha. Cell 93, 693704.
Poland, A. and Knutson, J. C. (1982). 2, 3, 7, 8- Tetrachlordibenzo-p-dioxin and
related halogenated aromatic hydrocarbons: examination of the mechanism of
toxicity. Ann. Rev Pharmacol Toxicol 22, 517-554.
Quinet, E. M., Savio, D. A., Halpern, A. R., Chen, L., Schuster, G. U., Gustafsson,
J. A., Basso, M. D., and Nambi, P. (2006). Liver X receptor (LXR)-beta regulation
in LXRalpha-deficient mice: implications for therapeutic targeting. Mol Pharmacol
70, 1340-1349.
Rajapakse, N., Silva, E., and Kortenkamp, A. (2002). Combining xenoestrogens at
levels below individual no-observed-effect concentrations dramatically enhances
steroid hormone action. Environ Hlth Perspec 110, 917-921.
Repa, J. J., Liang, G., Ou, J., Bashmakov, Y., Lobaccaro, J. M., Shimomura, I.,
Shan, B., Brown, M. S., Goldstein, J. L., and Mangelsdorf, D. J. (2000).
Regulation of mouse sterol regulatory element-binding protein-1c gene (SREBP1c) by oxysterol receptors, LXRalpha and LXRbeta. Genes Dev 14, 2819-2830.
Reuber, M. D. (1978). Carcinogenicity of kepone. J Toxicol Environ Health 4,
895-911.
Rigotti, A., Edelman, E. R., Seifert, P., Iqbal, S. N., DeMattos, R. B., Temel, R. E.,
Krieger, M., and Williams, D. L. (1996). Regulation by adrenocorticotropic
hormone of the in vivo expression of scavenger receptor class B type I (SR-BI), a
high density lipoprotein receptor, in steroidogenic cells of the murine adrenal
gland. J Biol Chem 271, 33545-33549.
Russell, D. W. (1999). Nuclear orphan receptors control cholesterol catabolism.
Cell 97, 539-542.
Sporstol, M., Tapia, G., Malerod, L., Mousavi, S. A., and Berg, T. (2005).
Pregnane X receptor-agonists down-regulate hepatic ATP-binding cassette
transporter A1 and scavenger receptor class B type I. Biochem Biophys Res
Commun 331, 1533-1541.
Srivastava, R. A. (2003). Scavenger receptor class B type I expression in murine
brain and regulation by estrogen and dietary cholesterol. J Neurol Sci 210, 11-18.
Srivastava, R. A., Baumann, D., and Schonfeld, G. (1993). In vivo regulation of
low-density lipoprotein receptors by estrogen differs at the post-transcriptional
level in rat and mouse. Eur J Biochem 216, 527-538.
Srivastava, R. A., Srivastava, N., Averna, M., Lin, R. C., Korach, K. S., Lubahn,
132
D. B., and Schonfeld, G. (1997). Estrogen up-regulates apolipoprotein E (ApoE)
gene expression by increasing ApoE mRNA in the translating pool via the estrogen
receptor alpha-mediated pathway. J Biol Chem 272, 33360-33366.
Staudinger, J., Liu, Y., Madan, A., Habeebu, S., and Klaassen, C. D. (2001a).
Coordinate regulation of xenobiotic and bile acid homeostasis by pregnane X
receptor. Drug Metab Dispos 29, 1467-1472.
Staudinger, J. L., Goodwin, B., Jones, S. A., Hawkins-Brown, D., MacKenzie, K.
I., LaTour, A., Liu, Y., Klaassen, C. D., Brown, K. K., Reinhard, J., Willson, T.
M., Koller, B. H., and Kliewer, S. A. (2001b). The nuclear receptor PXR is a
lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci U S
A 98, 3369-3374.
Telford, D. E., Lipson, S. M., Barrett, P. H., Sutherland, B. G., Edwards, J. Y.,
Aebi, J. D., Dehmlow, H., Morand, O. H., and Huff, M. W. (2005). A novel
inhibitor of oxidosqualene:lanosterol cyclase inhibits very low-density lipoprotein
apolipoprotein B100 (apoB100) production and enhances low-density lipoprotein
apoB100 catabolism through marked reduction in hepatic cholesterol content.
Arterioscler Thromb Vasc Biol 25, 2608-2614.
Ueda, Y., Royer, L., Gong, E., Zhang, J., Cooper, P. N., Francone, O., and Rubin,
E.
M. (1999). Lower plasma levels and accelerated clearance of high density
lipoprotein (HDL) and non-HDL cholesterol in scavenger receptor class B type I
transgenic mice. J Biol Chem 274, 7165-7171.
Wang, N., Silver, D. L., Costet, P., and Tall, A. R. (2000). Specific binding of
ApoA-I, enhanced cholesterol efflux, and altered plasma membrane morphology in
cells expressing ABC1. J Biol Chem 275, 33053-33058.
Williams, D. E., Okita, R. T., Buhler, D. R., and Masters, B. S. (1984).
Regiospecific hydroxylation of lauric acid at the (omega-1) position by hepatic and
kidney microsomal cytochromes P-450 from rainbow trout. Arch Biochem Biophys
231, 503-510.
Xie, W., Radominska-Pandya, A., Shi, Y., Simon, C. M., Nelson, M. C., Ong, E.
S., Waxman, D. J., and Evans, R. M. (2001). An essential role for nuclear receptors
SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci U S A 98,
3375-3380.
133
Figure 1. Chlordecone activates Pregnane X Receptor (human homolog) HEK
293 cells were cotransfected as in Fig. 3-1B with the Gal 4 luciferase reporter
and a series of chimeras in which Gal4 DNA binding domain is fused to the
indicated nuclear receptor ligand binding domain. Cells were treated with ethanol
or CD (0.1, 1, or 10 μM). Activation of nuclear receptors was assessed by
measuring luciferase activity. Values are presented as relative luciferase induction
from triplicate assays ± SD.
134
Figure 2. Chlordecone activates ERα A. HEK 293 cells were cotransfected with
the Gal 4 luciferase reporter and chimeras in which Gal4 DNA binding domain is
fused to the indicated nuclear receptor ligand binding domain. B. Expression
vectors for human ERα or ERβ were cotransfected with luciferase report construct
containing ER response element [ERE(EFP)Tkluc] plasmid along with the CMXβ-gal internal control. Cells were treated with ethanol as vehicle, 10 μM CD or 1
nM E2 (ER agonist). Values are presented as relative luciferase induction from
triplicate assays ± SD.
135
Figure 3. Chlordecone inhibits LXRs and activates FXR. A. Structure of
chlordecone (CD). B. HEK 293 cells were cotransfected with a
luciferase reporter plasmid [(LXRE)3 Tklc] containing three copies of the LXR
response element upstream of the thymidine kinase (TK) promoter and expression
vector for murine LXRα and RXR along with the CMX-β-gal internal control. C.
HEK 293 cells were cotransfected as in Fig. 3-1B for LXRβ and RXR. Cells were
treated with vehicle (ethanol) alone or 0.1 to 50 μM CD and/or 1 μM T0901317
(LXR agonist) or 100nM LG268 (RXR agonist) as indicated. D. HEK 293 cells
were cotransfected with a luciferase reporter plasmid [(FXRE) Tklc] construct
containing an FXR response element upstream of the TK promoter and expression
vector for FXR and RXR. Cells were treated with vehicle (ethanol) alone or 50 μM
CD and/or 100 nM LG268 (RXR agonist) as indicated. Values are presented as
relative luciferase induction from triplicate assays ± SD.
136
Figure 4. In vivo evidence for chlordecone activation of PXR. Animals were
treated with corn oil, 5 mg CD/kg or 15 mg CD/kg body weight by ip injection.
Hepatic microsomes were prepared from individual animals (5-6 mice in each
group) after 3 days as described under “Materials and Methods”. A. Immunoblot
analyses of CYP7A1, CYP4A1, and CYP3A11 proteins in liver microsomes. Total
20 μg protein samples were separated by SDS-PAGE and blotted with antibodies
against the specific CYP isoforms as described under “Materials and Methods”.
Values are expressed as the mean relative protein expression (A.U) ± standard error
(SEM) compared with the controls. B. Enzyme activity analyses of CH 7 αhydroxylase for CYP7A1 and lauric acid ω-1 hydroxylase for CYP4A1. Values are
expressed as the mean relative enzyme activity (%) ± standard error (SEM)
compared with the controls. *Significantly different from the control (p < 0.05).
+Different from the control (p=0.056).
137
Figure 5. Chlordecone reduces non-HDL plasma CH after a 4 hr fast.
Individual plasma was collected from animals (5-6 mice in each group) treated
with either corn oil or 15 mg/kg body weight by ip injection. A. Plasma total CH
and HDLCH was determined enzymatically as described under “Material and
Methods”. Non- HDL-CH content was taken as the difference between the total
plasma total CH and HDL-CH. B. Ratio of HDL-CH to plasma CH. C. Western
blot analyses of plasma apoA-I and apoB48/100 protein. Plasma apolipoprotein
content was measured by western blotting with either apoA-I or apoB100
antibodies. Values are expressed as mean ± standard error (SEM). +Different from
the control (p = 0.059).
138
Figure 6. Effect of Chlordecone on hepatic apolipoprotein content. Animals
were treated with corn oil, 5 mg CD/kg or 15 mg CD/kg body weight by ip
injection. Hepatic subcellular fractions were prepared from individual animals (5-6
mice in each group) after 3 days as described under “Materials and Methods”. A.
Hepatic apolipoprotein A-I or B. apolipoprotein B100 was measured by
immunoblotting with either apoA-I or apoB100 antibodies. Values are expressed
as the mean relative protein expression (A.U) ± standard error (SEM) compared
with the controls. ** Plasma CH concentration was 55.3 mg/dl. Grubbs' test was
applied for outlier identification. +Different from the control (p < 0.1).
139
Figure 7. Chlordecone reduces hepatic CH ester content. Animals were treated
with corn oil, 5 mg CD/kg or 15 mg CD/kg body weight by ip injection. Liver was
collected from individual animals (5-6 mice in each group) after 3 days. After lipid
extraction with hexane, liver total and free CH content were determined using an
enzymatic method as in under “Material and Methods. Hepatic esterified CH
content was taken as the difference between the total CH and free CH. Values are
expressed as mean CH level ± standard error (SEM). *Significantly different from
the control (p < 0.05).
140
Figure 8. Apolipoproteins and cytochrome P450 contents at 14 days after 15
mg CD/kg treatment. Animals were treated with corn oil or 15 mg CD/kg body
weight by ip injection. Plasma and hepatic microsomal fractions were prepared
from individual animals (5-6 mice in each group) after 14 days as described under
“Materials and Methods”. Immunoblotting was performed with apoA-I, apoB100,
CYP7A, CYP4A or CYP3A antibodies.