Title Differential Gene Expression in Human
Macrophages During Foam Cell Formation
Adviser Prof. Dr. Erwin Arno Galinski, Institute of Biochemistry,
University of Münster
Thesis Committee Priv.-Doz. Dr. Dr. Paul Cullen, Institute
of Arteriosclerosis Research, Prof. Dr. Erwin Arno Galinski (adviser);
Prof. Dr. Bernt Krebs, FRSC, Institute of Inorganic Chemistry, Prof.
Dr. Ernst-Ulrich Würthwein, Institute of Organic Chemistry, University
Atherosclerosis is a very common chronic inflammatory disease of the
subintimal space of the medium-sized and large arteries. Sites of
predilection are the coronary arteries supplying the heart, the carotid
vessels supplying the brain, and the arteries of the legs. The full-blown
atherosclerotic plaque takes many years to develop. Characteristically,
it takes the form of an eccentric lesion consisting of a lipid-rich
core containing necrotic cellular debris that is surrounded by areas
of smooth muscle cell proliferation and inflammatory cellular infiltrates.
This lesion may partially obstruct the artery, causing a painful lack
of oxygen in downstream tissue with the symptoms of angina pectoris
(tightness and pain in the chest) or intermittent claudication (leg
pain). The thin fibrous cap overlying the lesion may also rupture,
triggering formation of a blood clot that totally obstructs the artery.
If this occurs in a coronary artery, the result is infarction (death)
of the cardiac muscle or myocardium, a condition that, despite advances
in medical care, is still fatal in half of all cases. If the clot
occurs in a carotid artery, the result is stroke that may be fatal
or lead to a life of profound disability. Overall, atherosclerosis
and its complications are responsible for about 40% of deaths in the
developed world and for about 30% of fatalities in developing nations.
A main feature of the atherosclerotic plaque is the
presence of large numbers of macrophages. These macrophages ingest
in unregulated fashion chemically and/or enzymatically altered low-density
lipoproteins (LDL) that have become trapped beneath the arterial intima.
Storage of the cholesterol from this LDL as lipid droplets imparts
a foamy appearance to the cytoplasm of the macrophages that are then
termed foam cells. Much in vitro data and pathological findings
suggest that the process of macrophage foam cell formation has an
important influence on the genesis and progression of atherosclerosis.
Aim of thesis project
The aim of my research was to determine how the process of foam cell
formation affects the pattern of gene expression within the macrophage.
We hoped that this would enable us to understand more about how atherosclerosis
develops and maybe even point the way towards the development of new
Methods and results
The first task was to secure a supply of good-quality human macrophages.
To do this, monocytes, the precursors of macrophages, were isolated
by a process called countercurrent cell elutriation from the blood
of healthy volunteers, usually medical students. All donors had normal
blood lipids and all were selected to be homozygous for so-called
E3 variant of a macrophage protein called apolipoprotein E. This was
because previous work by our group had shown that the way macrophages
handle cholesterol is profoundly influenced by what variant of apolipoprotein
E they produce.
The monocytes were allowed to differentiate into macrophages
by growing them in culture medium supplement with 20% human serum
for 14 days. To transform them into foam cells, we added chemically
modified LDL to the culture medium for two days. This modified LDL
is taken up by macrophages in large amounts. To monitor foam cell
formation, we measured the accumulation of cholesterol within the
cells using a high-performance liquid chromatography (HPLC) technique
developed in our laboratory.
To analyse gene expression, we used the method of differential
display reverse transcription polymerase chain reaction (DDRT-PCR).
In this technique, very short stretches of DNA are used as so-called
primers to amplify the messenger ribonucleic acid (mRNA) molecules
within the cell. Because these stretches occur commonly throughout
the genome, it is reckoned that DDRT-PCR amplifies over 90% of all
the mRNAs in a cell. Unfortunately, DDRT-PCR is a difficult and unforgiving
technique, and is beset by the problem of false-positive results,
i.e. differences in gene expression are found where none exist.
We therefore began by improving the technique. We developed novel
DDRT-PCR protocols that increased signal intensity, reduced non-specific
results, and isolated longer stretches of mRNAs that were easier to
Using DDRT-PCR, we identified 36 mRNAs that appeared
to be differentially expressed during foam cell formation. In three
cases, this regulation was confirmed by independent techniques. Further
studies showed that these three mRNAs coded respectively for an acidic
calcium-independent type A2 phospholipase, the adenosine triphosphate
(ATP)-binding cassette transporter G1 (ABCG1), and a novel gene product
that had not previously been identified.
ABC proteins form a huge family that occurs in all known
organisms. Most are membrane-spanning transporters that actively move
molecules between cell compartments or cells using energy derived
from the breakdown of ATP.
Several mutations in ABC proteins have been identified
as the cause of such diverse diseases as adrenoleukodystrophy, a disorder
of the nervous system, and cystic fibrosis, a disease that affects
the lungs, the pancreas and the liver and that renders affected men
infertile. In 1999, Rust et al. from our institute together
with two other groups showed that mutations in the human ABCA1 gene
cause the rare disorder of lipid disorder known as Tangier disease,
which is characterized by the accumulation of cholesterol within macrophages.
Based on this finding, and on the results of DDRT-PCR experiments,
we concluded that ABC proteins may play a role in the lipid metabolism
of cells and might also contribute to the development of atherosclerosis.
A second hallmark of Tangier disease is the lack of
high-density lipoprotein (HDL) particles in the circulation. HDL is
thought to be the major particle responsible for transport of excess
cholesterol from peripheral tissues back to the liver where it can
be excreted. This is important because mammals have no way of breaking
down the cholesterol molecule once it has been formed. The only way
it can be removed is to excrete it intact from the liver in the form
of bile salts. In tissue culture, HDL removes cholesterol from cells.
In population studies, high levels of HDL in the blood were associated
with a decreased risk of atherosclerosis. Conversely, the risk of
atherosclerosis is increased in persons with Tangier disease.
While we were performing our experiments, other groups
showed foam cell formation in macrophages switches on the gene for
ABCA1. What we found, however, was that increase in expression of
the ABCG1 gene that occurs the foam cell formation is far greater
than that of ABCA1. At the time, however, the function of ABCG1 was
unknown, and even today no inherited disease caused by mutations in
the ABCG1 gene has been identified.
We therefore decided to dig a little deeper. First,
we investigated how the expression of ABCG1 mRNA was affected by different
techniques of foam cell formation and found that it increased to about
the same extent irrespective of whether foam cells were produced using
oxidized or acetylated LDL or just so-called liposomes as vesicles
containing cholesterol. Conversely, we found that expression of ABCG1
decreased when we removed cholesterol from the macrophages by adding
HDL. In addition we found that the expression of ABCG1 differed depending
on the variant of apolipoprotein E produced by the cells.
All these findings were clues that like ABCA1, AGCG1
was somehow involved in how the cell dealt with cholesterol. This
suspicion increased when we found that the gene for ABCG1 was constitutively
switched on in the macrophages of patients with Tangier disease, which
contain more cholesterol than normal macrophages.
At this stage we formulated the hypothesis that like
its cousin ABCA1, ABCG1 also plays a role in removing cholesterol
from cells. To further investigate the function of ABCG1 we made antibodies
directed towards a loop of the ABCG1 protein we knew from computer
modelling was likely to project outside the cell. Using these antibodies,
we found out a number of things. First, we showed that the changes
in the mRNA for ABCG1 were mirrored by changes in the amount of ABCG1
protein. Second, we used the antibodies to track down ABCG1 within
the cell and found that in macrophages it was concentrated mainly
in the outer cell membrane and in membrane structures around the cell
nucleus. In fibroblasts, the principal cells of connective tissue,
by contrast, ABCG1 seemed to be diffusely scattered about within the
cell. Third, we found that in atherosclerotic arteries, ABCG1 is present
mainly in foamy macrophages and in bundles of nerves coursing in the
outermost layer of the artery, the adventitia.
So what exactly was ABCG1 doing? We knew from studies
in the fruit fly, Drosophila melanogaster, that a related protein
called "white" was involved in the transport of the amino acid
tryptophan. Tryptophan is the raw material for the pigment in the
eye of the fly, and the gene derives its name from the fact that its
deficiency leads to a lack of pigment and an eye that is white in
color. We therefore measured the uptake of radioactive tryptophan
in relation to ABCG1 expression in macrophages, but found no relationship,
although foamy macrophages did accumulate 5-hydroxyindoleacetic acid
(5-HIAA), a breakdown product of tryptophan. However the accumulation
of 5-HIAA was not influenced by inhibition of ABCG1 protein formation
using so-called antisense oligonucleotides that prevented the translation
of ABCG1 mRNA.
A second avenue to investigating the role of ABCG1 was
to determine the structure of the ABCG1 gene and in particular to
characterize the promoter, the region that regulates gene activity.
To do this, we used advanced PCR-based techniques to identify the
stretches of DNA upstream (5') and downstream (3') of the mRNA. We
were surprised to uncover an unusual degree of complexity. We found
that the gene contains five coding regions or exons that had not previously
been described, and that it codes for a very large number of so-called
splice variants with different amino acid sequences. In addition,
we found that these alternatively spliced ABCG1 transcripts are controlled
by different promoters.
To investigate the function of these promoters, we exposed
the cells to compounds which are known to bind to the liver X receptor
(LXR) and the retinoid X receptor (RXR), proteins regulating gene
expression that reside within the cell nucleus. Both types of compound
increased the activity of the ABCG1 gene.
Taken together, my results indicate that ABCG1 is involved both in
the transport of cholesterol within macrophages and in foam cell formation.
The strong expression of ABCG1 in macrophages in the arterial wall
indicates that it may also play a role in the development of atherosclerosis.
Results have also been obtained by our collaborator, Prof. Dr. Arnold
von Eckardstein, that lend further support to this hypothesis. Dr.
von Eckardstein's group showed that reducing the production of ABCG1
protein by means of antisense oligonucleotides decreased the secretion
of apolipoprotein E, which, as noted above, is a very important player
in removing cholesterol from cells and may well protect against atherosclerosis.
Therefore, ABCG1 might actually turn out to be an interesting target
for the development of new anti-atherosclerotic therapies.
Since completion of my PhD thesis, further work by our group has led
to the identification of a new ABC transporter, ABCG4, that is highly
homologous to ABCG1 and that is also expressed in human monocyte-derived
macrophages. ABCG4 gene expression seems to be regulated in the same
fashion as ABCG1 gene expression.