This accumula-tion of cholesteryl esters and modified protein from glycated low-density lipoprotein may contribute to cellular dysfunction and the increased atherosclerosis observed in pe
Trang 1Glycation of low-density lipoprotein results in the
time-dependent accumulation of cholesteryl esters
and apolipoprotein B-100 protein in primary human
monocyte-derived macrophages
Bronwyn E Brown1, Imran Rashid1, David M van Reyk2and Michael J Davies1,3
1 Free Radical Group, The Heart Research Institute, Camperdown, Sydney, NSW, Australia
2 Department of Health Sciences, University of Technology Sydney, NSW, Australia
3 Faculty of Medicine, University of Sydney, NSW, Australia
Complications associated with diabetes are the major
cause of mortality and morbidity in people with this
disease These include microvascular complications that
induce damage to the retina, nephrons and peripheral nerves, and macrovascular disease that is associated with accelerated atherosclerosis (deposition of lipids in
Keywords
aldehydes; atherosclerosis; foam cells;
human monocyte-derived macrophages;
low-density lipoproteins
Correspondence
M J Davies, 114 Pyrmont Bridge Road,
Camperdown, Sydney, NSW 2050, Australia
Fax: +61 2 95655584
Tel: +61 2 82088900
E-mail: daviesm@hri.org.au
(Received 12 December 2006, accepted
15 January 2007)
doi:10.1111/j.1742-4658.2007.05699.x
Nonenzymatic covalent binding (glycation) of reactive aldehydes (from glu-cose or metabolic processes) to low-density lipoproteins has been previ-ously shown to result in lipid accumulation in a murine macrophage cell line The formation of such lipid-laden cells is a hallmark of atheroscler-osis In this study, we characterize lipid accumulation in primary human monocyte-derived macrophages, which are cells of immediate relevance to human atherosclerosis, on exposure to low-density lipoprotein glycated using methylglyoxal or glycolaldehyde The time course of cellular uptake
of low-density lipoprotein-derived lipids and protein has been character-ized, together with the subsequent turnover of the modified apolipoprotein B-100 (apoB) protein Cholesterol and cholesteryl ester accumulation occurs within 24 h of exposure to glycated low-density lipoprotein, and increases in a time-dependent manner Higher cellular cholesteryl ester lev-els were detected with glycolaldehyde-modified low-density lipoprotein than with methylglyoxal-modified low-density lipoprotein Uptake was signifi-cantly decreased by fucoidin (an inhibitor of scavenger receptor SR-A) and
a mAb to CD36 Human monocyte-derived macrophages endocytosed and degraded significantly more125I-labeled apoB from glycolaldehyde-modified than from methylglyoxal-modified, or control, low-density lipoprotein Dif-ferences in the endocytic and degradation rates resulted in net intracellular accumulation of modified apoB from glycolaldehyde-modified low-density lipoprotein Accumulation of lipid therefore parallels increased endocytosis and, to a lesser extent, degradation of apoB in human macrophages exposed to glycolaldehyde-modified low-density lipoprotein This accumula-tion of cholesteryl esters and modified protein from glycated low-density lipoprotein may contribute to cellular dysfunction and the increased atherosclerosis observed in people with diabetes, and other pathologies linked to exposure to reactive carbonyls
Abbreviations
AGE, advanced glycation end-products; apoB, apolipoprotein B-100; HBSS, Hank’s balanced salt solution; HMDM, human monocyte-derived macrophage; HSA, human serum albumin; LDL, low-density lipoprotein.
Trang 2the artery wall) in the coronary, peripheral and carotid
arteries [1] Factors that may contribute to this
acceler-ated atherosclerosis include chronic elevacceler-ated glucose
levels (hyperglycemia) and insulin resistance,
dyslipide-mias, and abnormalities of homeostasis [2]
Macrovas-cular disease has been reported to appear in people
with type 2 diabetes at, or near the time of, first
diag-nosis of diabetes, consistent with a shared underlying
pathogenesis [2] An early and persistent feature of the
atherosclerotic lesion is the presence of lipid-laden
(foam) cells in the intima of the artery wall, arising
from cholesterol and cholesteryl ester accumulation by
macrophage cells present in the artery wall [3]
Low-density lipoproteins (LDLs) are the likely source of this
lipid, with unregulated LDL uptake occurring via
receptors other than the native LDL receptor, including
CD36 and class A scavenger receptors [4,5] These
receptors recognize abnormal LDL species, including
those modified by oxidation, aggregation, chemical
modification and formation of immune complexes [4,6]
Elevated glucose levels are strongly linked to the
incidence and severity of atherosclerosis [7,8] Of
par-ticular relevance is the potential role of glucose (or
species derived from glucose) in LDL modification
[9,10] Previous studies have identified multiple
poten-tial mechanisms of LDL modification, including
gly-cation and glycoxidation [9] Glygly-cation involves the
covalent adduction of an aldehyde (from glucose or
related species) to a reactive amine (e.g Lys and Arg
side chains, N-terminus [11–13]) or thiol (Cys) groups
on proteins [14], such as those of the single protein
molecule of LDL, apolipoprotein B-100 (apoB) The
initial Schiff base undergoes subsequent rearrangement
to yield Amadori products (e.g fructose-lysine)
Glyc-oxidation consists of two related processes) oxidation
of protein-bound sugars (from glycation), and
oxida-tion of free glucose and its products Both processes
can generate radicals that modify LDL, and hence
potentially contribute to the enhanced uptake of such
particles by macrophages [12,15–17]
The species formed by glycation and glycoxidation
undergo subsequent reactions to give a heterogeneous
and complex mixture of materials often called
advanced glycation end-products (AGEs) [9,12]
Eleva-ted levels of AGEs have been reporEleva-ted in people with
diabetes compared to controls [18], with some of these
materials (e.g Ne-carboxymethyl-lysines and Ne
-carboxy-ethyl-lysines and pentosidine) being known to
accumu-late with age on tissue proteins, and at an increased
rate in LDL and atherosclerotic lesions in people
with diabetes [16,19–21] Ne-carboxymethyl-lysine and
Ne-carboxyethyl-lysine can arise from reaction of Lys
residues with reactive aldehydes (glyoxal⁄
glycolalde-hyde and methylglyoxal, respectively) [22], providing strong evidence for the formation and subsequent reac-tions of these aldehydes in atherosclerotic lesions The plasma concentrations of these aldehydes are elevated
in people with diabetes [23,24], although the concentra-tions of these materials present in the artery wall, and
in atherosclerotic lesions, are unknown
The role of glycation and the two facets of glycoxida-tion in generating modified LDLs and lipid-laden (foam) cells, in vitro or in vivo, is incompletely under-stood Most studies have employed conditions under which both processes have occurred, or where the nat-ure and extent of modifications have not been quanti-fied adequately [15,25] It is therefore unclear as to whether glycation of LDL, in the absence of oxidation, results in foam cell formation in cell types of direct rele-vance to human atherosclerosis It is also not known whether the protein and lipid components of modified LDL accumulate in synchrony, or to similar levels, due
to differences in the rates of cellular proteolysis and lipolysis Furthermore, the cellular handling of the resulting glycated apoB has not been well characterized Modified proteins have been shown to have different susceptibilities to proteolysis than native proteins, with both enhanced and decreased rates having been charac-terized [26,27] The latter may result in the accumula-tion of modified proteins within cells, and subsequent perturbation of cellular metabolism [16,19–21]
Previously, we have characterized conditions that yield glycated, but nonoxidized, LDL [28], and have shown that such particles give rise to lipid accumula-tion in cultured mouse macrophage-like cells [29] In the current study, we have determined whether lipid accumulation also occurs in a more relevant cell type) human monocyte-derived macrophages (HMDMs))
on exposure to LDL glycated using methylglyoxal or glycolaldehyde The time course of cellular uptake of LDL-derived lipid and protein has been characterized,
as well as the subsequent turnover of the apoB protein
It is shown that both lipid and protein are taken up,
in a time-dependent manner, via scavenger receptor SR-A- and CD36-mediated processes, and that the uptake of lipid and protein occurs in synchrony Fur-thermore, it is shown that both lipid and modified pro-tein accumulate in cells, despite significant proteolytic degradation of the modified protein
Results
LDL characterization Glycated LDL particles were prepared using meth-ylglyoxal, glycolaldehyde and glucose, as described
Trang 3previously [28,29] This method results in minimal
oxidation of apoB, cholesterol, cholesteryl esters, or
a-tocopherol [28,29], and does not affect the relative
cholesterol, cholesteryl ester, phospholipid or
triglycer-ide composition of the particles (Table 1) In contrast,
significant time- and concentration-dependent
glyca-tion of apoB occurs with methylglyoxal or
glycolal-dehyde when compared to control or glucose-modified
particles, as indicated by particle charge, aggregation,
and amino acid modification [28,29] The relative
elec-trophoretic mobility of the particles used in the current
study was not significantly different to that reported
previously [29], irrespective of LDL iodination (data
not shown)
Lipid accumulation in HMDMs
Lipid accumulation was quantified after exposure of
HMDMs (1· 106cells per well) at 37C, for up to
96 h, to LDL (0 or 100 lgÆmL)1) previously modified
by methylglyoxal (100 mm), glycolaldehyde (100 mm)
or glucose (100 mm ± 1 lm Cu2+), or control LDL
incubated with EDTA (50 lm) No change in cell
viab-ility or protein was detected in comparison to control
cells not exposed to LDL LDL chemically modified
by acetylation was employed as a positive control
Cells exposed to glucose (± Cu2+)-modified LDL did
not contain significantly elevated cellular cholesterol or
cholesteryl ester levels in comparison to control cells
incubated with LDL exposed to EDTA (data not
shown) No increase in cellular free cholesterol levels
was observed on exposure of HMDMs to
methylgly-oxal- or glycolaldehyde-modified LDL for 0–96 h in
comparison to incubation controls (LDL incubated
with EDTA; Fig 1A), although these values were
sig-nificantly higher than in cells exposed to no LDL No
significant difference was observed in free cholesterol
levels of HMDMs incubated with unmodified LDL,
compared to no LDL, except at the 96 h time point
(Fig 1A)
In contrast to the above, significant time-dependent accumulation of cholesteryl esters in HMDMs was observed on incubation with glycolaldehyde- or methyl-glyoxal-modified LDL (Fig 1B) Glycolaldehyde-modi-fied LDL induced the greatest accumulation, with this being significantly higher than for methylglyoxal-modi-fied LDL, or control LDL, at all time points Methyl-glyoxal-modified LDL induced significantly greater cholesterol ester accumulation than control LDL at the
48 h and 96 h time points, with the majority of this accumulation occurring over the first 24 h There was
no significant difference in cellular cholesterol ester content between cells incubated with unmodified LDL and cells not incubated with LDL, at all time points Glycolaldehyde-modified LDL induced a steady increase in the percentage of total cholesterol present
as esters over the 96 h period, reaching a value of
53 ± 7% (Fig 1C) Similar levels were detected with acetylated LDL (data not shown) Methylglyoxal-modified LDL also induced a significant increase in the percentage of cholesterol esters when compared to control LDL at all time points, with this reaching
23 ± 6% at 96 h There was no significant difference
in the percentage of cholesterol esters between HMDMs incubated with unmodified LDL and those not incubated with LDL, indicating an absolute requirement for LDL modification for significant lipid accumulation in these cells
Accumulation and turnover of apoB in HMDMs HMDMs were exposed to glycated 125I-labeled LDL, with the levels of cell surface, endocytosed, degraded and intracellular accumulated apoB being determined
by radioactive counting 125I-Labeled acetylated LDL was used as a positive control (data not shown), and gave similar results to those observed for glycol-aldehyde-modified LDL The extent of endocytosis, degradation and intracellular accumulation of apoB increased over time in HMDMs exposed to control
Table 1 Lipid composition (nmol lipidÆmg)1apo B) of native, control and glycated LDL LDL (1 mg proteinÆmL)1) was incubated with 50 l M EDTA (control LDL) or 100 m M modifying agent ± 1 l M Cu2+, in NaCl⁄ P i (pH 7.4), for 7 days at 37 C Values are means ± SEM from three experiments, each with triplicate samples None of the treatments resulted in significantly different values compared to the native LDL (P > 0.05).
Total cholesterol Free cholesterol Cholesteryl ester Triglyceride Phospholipid
Trang 4LDL (Fig 2A), methylglyoxal-modified LDL (Fig 2B), glycolaldehyde-modified LDL (Fig 2C) and LDL modified by glucose ± Cu2+(similar to control LDL; data not shown) However, the absolute amount of apoB endocytosed, degraded and accumulated was dependent upon the nature of the LDL modification; these were quantified at the 96 h time point (Fig 2D) The extent of protein endocytosis, degradation and intracellular accumulation of apoB was increased in HMDMs exposed to glycolaldehyde-modified LDL in comparison to those exposed to control LDL HMDMs exposed to methylglyoxal-modified LDL showed signifi-cantly increased endocytosis and degradation when compared to those exposed to control LDL, although this was less marked than with glycolaldehyde-modified LDL These parameters were not elevated for HMDMs exposed to glucose (± Cu2+)-modified LDL when compared to those exposed to control LDL In each case, amounts of cell surface (bound) apoB were min-imal, remained constant over time, and did not vary between conditions (Fig 2A–D)
The turnover of intracellular (accumulated) apoB was examined over a 24 h chase period using LDL-free medium following exposure of HMDMs to labeled gly-colaldehyde- and methylglyoxal-modified LDL, and control LDL, for 96 h The use of LDL-free medium during the chase period allows the turnover of preaccu-mulated protein to be studied in the absence of further cellular uptake In these studies, cell death was < 12%
as measured by the appearance of nondegraded apoB
in the medium In each case, a time-dependent decrease
in (previously nondegraded) intracellular apoB concen-trations was detected, and was matched by an increase
in the concentration of degraded apoB (i.e peptides) in the medium (Fig 3A–C) In all cases, only 20–30% of the apoB present at the start of the chase period was degraded The absolute concentration of apoB turned over decreased in the order glycolaldehyde-modi-fied > methylglyoxal-modiglycolaldehyde-modi-fied > control (P < 0.05) With a 24 h loading period with125I-labeled LDL prior
to a 24 h chase period, a greater turnover of intracellu-lar apoB was observed, with 35–55% of the
nondegrad-ed intracellular apoB being turnnondegrad-ed over (data not shown) The absolute concentration of apoB turned over was lower under these conditions, due to the lower initial accumulation of nondegraded intracellular apoB (data not shown)
Investigation of the nature of the receptors responsible for uptake of glycated LDL HMDMs were exposed to methylglyoxal- or glycol-aldehyde-modified LDL in the absence or presence of
Fig 1 Cellular free cholesterol (A), total cholesteryl esters (B) and
percentage cholesteryl esters of total cholesterol (sum of free
cho-lesterol plus total cholesteryl esters) (C) present in HMDMs after
exposure to no LDL (circles), incubation control LDL (LDL + EDTA;
triangles), methylglyoxal-modified LDL (squares), or
glycolaldehyde-modified LDL (diamonds) HMDMs (1.0 · 10 6
cells per well) were exposed to 100 lgÆmL)1modified LDL (1 mg proteinÆmL)1,
incuba-ted with 100 m M modifying agent or 50 l M EDTA, in NaCl ⁄ P i ,
pH 7.4, for 7 days at 37 C) for up to 96 h in medium containing
10% lipoprotein-deficient serum (with fresh medium and LDL
added at 48 h) before extraction and analysis by HPLC with UV
detection Values are means ± SEM from three or more
experi-ments, each with triplicate samples *, # and + indicate statistically
elevated values (P < 0.05) compared to the control cells (no LDL),
LDL plus EDTA-treated cells, and methylglyoxal-modified
LDL-treated cells, respectively, at each time point.
Trang 5mAb to CD36, fucoidin or AGE–human serum albu-min (HSA) for 48 h, and changes in total cellular cho-lesteryl esters were determined using HPLC Exposure
of cells to methylglyoxal-modified LDL (Fig 4A) or glycolaldehyde-modified LDL (Fig 4B) and the mAb
to CD36 or fucoidin resulted in significantly decreased cellular cholesteryl ester accumulation in comparison
to cells exposed only to modified LDL Cells exposed
to modified LDL in the presence of AGE–HSA had lower cholesteryl ester levels, but this decrease was not significantly different in comparison to cells exposed to modified LDL alone
Fig 3 Turnover of accumulated apoB in HMDMs after exposure to
50 lgÆmL)1incubation control [ 125 I]LDL (A) or 50 lgÆmL)1[ 125 I]LDL modified by methylglyoxal (B) or glycolaldehyde (C) for 96 h The preparation and cellular exposure to [ 125 I]LDL were performed as described in Fig 1, after iodination of the LDL, and were followed
by cell washing and exposure to LDL-free chase medium (DMEM containing 1 mgÆmL)1 BSA in place of serum) At the appropriate chase times, cells were lysed and processed to determine nonde-graded intracellular apoB (triangles), denonde-graded intracellular apoB (open circles), and degraded extracellular apoB (squares) Values are means ± SEM from three experiments, each with triplicate samples Note different axis scales * and # indicate statisti-cally elevated values (P < 0.05) compared to 0 h chase time for nondegraded intracellular apoB and extracellular degraded apoB, respectively.
Fig 2 Time course of endocytosis (open circles), surface binding
(triangles), degradation (diamonds) and intracellular accumulation
(squares) of apoB from incubation control [125I]LDL (A) and
[ 125 I]LDL modified by methylglyoxal (B) or glycolaldehyde (C) in
HMDMs (D) compares the data obtained at the 96 h time point on
the cellular handling of apoB in HMDMs exposed to 50 lgÆmL)1
incubation control [ 125 I]LDL (white) or 50 lgÆmL)1[ 125 I]LDL
modi-fied by methylglyoxal (black), glycolaldehyde (horizontal stripes) or
glucose in the absence (dots) or presence (vertical stripes) of Cu2+.
HMDMs (1.0 · 10 6 cells per well) were exposed to 50 lgÆmL)1
modified [ 125 I]LDL for up to 96 h (with fresh medium and [ 125 I]LDL
added at 48 h) before analyses The preparation and cellular
expo-sure to [ 125 I]LDL were performed as described in Fig 1, after
iodi-nation of the LDL Endocytosed material is the sum of degraded
and intracellular measurements Values are means ± SEM from
three experiments, each with triplicate samples Note different axis
scales *, # and + (A–C) indicate statistically elevated values
(P < 0.05) compared to the 0 h time point for apoB endocytosis,
degradation and intracellular accumulation, respectively Columns
(D) with different letters above them are significantly different by
one-way ANOVA (P < 0.05) for that apoB measurement.
Trang 6The present study has shown that incubation of
pri-mary HMDMs with glycated, but nonoxidized, LDL
can give rise to time-dependent lipid loading, with this
lipid accumulation occurring in parallel with the
endo-cytosis and degradation of the protein (apoB)
compo-nent of the LDL This uptake of glycated LDL occurs
primarily via scavenger receptor SR-A and CD36
endocytosis, as demonstrated by receptor-blocking
experiments The rates of uptake of both the lipid and
protein components are not matched by the rate of
cel-lular metabolism of these species, resulting in the
accu-mulation of both unmodified cholesteryl esters and
glycated apoB in the cells The rate of removal of the
latter species is slow, with only 20–30% of the glycated
protein being degraded over a 24 h chase period
In contrast to the rapid and extensive lipid
accumu-lation induced by LDL modified by glycolaldehyde or
methylglyoxal, incubation of HMDMs with LDL modified by glucose, or glucose plus Cu2+(with a con-centration of Cu2+ similar to that detected in advan-ced atherosclerotic lesions [30]), did not result in significant cellular sterol accumulation This is in con-trast to the results of a previous study, in which a two-fold increase in cholesteryl ester synthesis was observed
in HMDMs exposed to glucose-modified LDL [25]
No characterization data were presented for the LDL used in this previous study, so this discrepancy may arise from the nature of the modified LDL used, with oxidation being a potential confounding factor Uptake of oxidized LDL has been previously shown to result in foam cell formation [17] The lack of choleste-ryl ester accumulation resulting from LDL being incu-bated with glucose, in the presence or absence of
Cu2+, is consistent with our previous studies using murine macrophage-like cells [29]
Exposure of LDL to 100 mm glycolaldehyde has been shown previously to result in extensive modifica-tion of the Lys residues present on the apoB protein [29] Such modification has been reported to result in recognition by macrophage scavenger receptors [15,29,31] The cellular accumulation of cholesteryl esters (approximately 50% of total sterol levels) observed in the present study is consistent with that reported with cultured murine macrophage-like cells [29] The proportion of total cholesterol present as cholesterol esters in these HMDMs is of a similar mag-nitude to that detected in human atherosclerotic lesions [32]
It has been reported that LDL modified by 10 mm methylglyoxal for 3 days is recognized by macrophage scavenger receptors, but results in decreased intracellu-lar cholesteryl ester synthesis in comparison to controls [33] This is in contrast to the situation with cultured murine cells, where exposure to LDL modified with methylglyoxal for 14 days (with approximately 80% of Lys residues modified) resulted in significant choleste-ryl ester accumulation, with approximately 25% of the total cellular sterol being present as esters [29] In the present study, HMDMs exposed to LDL modified by methylglyoxal for 7 days accumulated significant levels
of cholesteryl ester within 24 h, with approximately 25% of total sterols being present as cholesteryl esters
by 96 h Thus, modification of LDL by methylglyoxal appears to result in macrophage scavenger receptor recognition, and significant cholesteryl ester accumula-tion, in human macrophages It has been reported that LDL isolated from people with diabetes can stimulate cholesteryl ester synthesis in HMDMs, although the level of modification reported (approximately 5% of Lys residues [34]) is lower than that used in the current
Fig 4 Cholesteryl ester changes in HMDMs exposed to
100 lgÆmL)1LDL modified by methylglyoxal (A) or glycolaldehyde
(B) for 48 h in the absence (control) or presence of mAb to CD36
(2 lgÆmL)1), fucoidin (200 lgÆmL)1), or AGE–HSA (200 lgÆmL)1).
Modified LDL was prepared and incubated with cells as described
in Fig 1 Values are means ± SEM from three experiments, each
with triplicate samples *Significantly decreased (P < 0.05) cellular
cholesteryl esters levels compared to cells incubated with the
modified LDL in the absence of any receptor inhibitors.
Trang 7study However, direct comparison between these two
sets of data is not possible, as the extent of other
mod-ifications present on these in vivo-modified particles is
not known We have suggested that it may be the
nat-ure of the products arising from glycation, rather than
purely the loss of the parent amino acid, which is the
key factor in terms of receptor recognition [29] There
are no data available on the extent of Lys (and other
amino acid) modification, arising from glycation, in
LDL isolated from human atherosclerotic lesions, so it
is not possible to judge the extent, or type, of amino
acid modification on LDL to which macrophage cells
might be exposed in vivo Further studies are required
to fully elucidate this point
The increased rates of endocytosis and
intracellu-lar degradation of methylglyoxal- and
glycolaldehyde-modified apoB protein from LDL, in HMDMs, is
consistent with particle recognition by macrophage
scavenger [15,29,31,33], or other receptors [35] These
data are in agreement with previous, more limited,
stud-ies with glycolaldehyde-modified LDL [15,31] The
cel-lular uptake and turnover of apoB in macrophages
exposed to methylglyoxal-modified LDL has not been
examined previously, although increased endocytosis
and degradation of apoB modified by other aldehydes
(e.g 4-hydroxynonenal, malondialdehyde) has been
reported [36] The pattern of uptake and degradation of
apoB from the various types of modified LDL examined
here mirrors cholesterol ester accumulation, with
glycol-aldehyde inducing the largest changes, glucose (with or
without Cu2+) the least, and methylglyoxal showing
intermediate behavior Previous studies have reported
both decreased [15] and increased [25] degradation of
apoB from glucose-modified LDL when compared to
native LDL; however, the nature and extent of
modifi-cation (or oxidation) of these particles are not known
Interestingly, apoB from glycolaldehyde-modified
LDL accumulated in HMDMs over time
Accumula-tion of modified proteins has been previously
implica-ted in diseases such as atherosclerosis and diabetes
[16,19–21], and reported to have a variety of cellular
effects It has been shown that moderately oxidized
proteins are more sensitive to proteolysis [37], and are
endocytosed more quickly than native proteins, which
in turn are more rapidly removed than heavily
oxid-ized proteins [27,37] Previous studies have shown that
some proteins that contain AGEs (e.g
pyrraline-modi-fied albumin) accumulate in macrophages because of
decreased cellular degradation rates and a reduced
sus-ceptibility of this glycated protein to lysosomal
proteo-lytic enzymes [38] Thus, glycation alone appears to be
sufficient to inhibit lysosomal degradation of modified
proteins Interestingly, apoB from oxidized LDL has
been shown to accumulate in secondary lysosomes in macrophages because of inefficient degradation [39], although the extent of (labeled) apoB turnover in the chase period (i.e after the cessation of loading) observed in the current study with glycated LDL is much lower than that observed previously for some forms of oxidized LDL (e.g that generated on expo-sure to 10 lm Cu2+ for 4 h [36]), consistent with poor cellular handling of the glycated apoB protein This may be partly explained by the resistance of the modi-fied apoB to degradation by lysosomal cathepsins [40]
In addition we have also shown that glycated⁄ glycoxi-dized proteins can inhibit thiol-dependent lysosomal cathpesins [41], as well as other intracellular enzymes, including lactate dehydrogenase, glyceraldehyde-3-phosphate dehydrogenase, and glutathione reductase [42] The inhibition of the thiol-dependent lysosomal cathepsins by glycated proteins may be of particular importance in apoB turnover
The accumulation of glycated apoB within HMDMs may be related to that of the cholesteryl esters observed under identical conditions, as a result of an interdependence of proteolysis and lipolysis Jessup
et al have postulated, on the basis of studies with oxidized LDL, that failure of macrophages to degrade oxidized apoB may protect LDL cholesteryl esters in the core of the particle from lysosomal esterases, or that impaired lipolysis of LDL lipids may block pro-teolysis of apoB [36] This may arise as a result of the failure of hydrophobic regions of apoB, which have been reported to be recognition signals for proteolysis,
to become exposed [43,44] The accumulation of such AGE-modified proteins may have significant cellular and atherogenic effects, and requires further study SR-A and CD36 have previously been reported to account for 75–90% of the uptake and degradation of acetylated or oxidized LDL [45] Glycated⁄ glycoxi-dized LDL has also been previously reported to
be recognized by macrophage scavenger receptors, although data on which specific scavenger receptors were involved have not been reported [15,31,33]; the current data are consistent with SR-A and CD36 being key species Greater than 60% modification of parent apoB Lys residues has been reported to result in macro-phage scavenger receptor recognition for both glycated and acetylated LDL [31,46] Lys data previously repor-ted by our group [28,29] show that that greater than 60% Lys modification is observed for methylglyoxal-and glycolaldehyde-modified LDL under the condi-tions used in these studies, consistent with this previous conclusion To investigate the types of recep-tor responsible for the uptake of glycated LDL observed in the current study, cells were incubated
Trang 8with LDL glycated using glycolaldehyde or
methylgly-oxal, and either a mAb to CD36 [47], the SR-A
inhib-itor fucoidin, or AGE–HSA, which is known to bind
to RAGE [48] The inhibition of uptake observed with
the mAb to CD36 or fucoidin indicates that SR-A and
CD36 are responsible for most of the observed uptake
Inhibition of RAGE by AGE–HSA did not decrease
uptake significantly Although RAGE is not an
endo-cytotic receptor, binding of AGE ligands to RAGE
has been shown to activate signaling pathways [49]
that potentially could have affected LDL uptake
The aldehyde concentrations utilized in this study
are higher than those reported for plasma from both
healthy controls and people with diabetes [23,24,50,51]
These plasma values (up to 0.5 mm [51]) are, however,
potentially misleading, as they represent only the
(small) fraction of these highly reactive species that has
not undergone reaction with plasma proteins, a process
that is known to be extremely rapid and efficient [52]
The true flux of these compounds is therefore likely to
be considerably higher Irrespective of this, it is clear
that the levels of these aldehydes are elevated in people
with diabetes [24] Furthermore, the levels of these
aldehydes may be substantially greater in the artery
wall than in plasma, as a result of cell-mediated
forma-tion of these species, with the major route to such
aldehydes being via the intracellular decomposition of
triose phosphates [53], the concentrations of which are
markedly elevated in hyperglycemia [54] It has also
been shown that the heme enzyme myeloperoxidase,
which is present at elevated levels at sites of
inflamma-tion (such as atherosclerotic lesions [55]) as a result of
the influx and activation of neutrophils and
mono-cytes, can oxidize free amino acids to reactive
alde-hydes, including methylglyoxal [56] Both these
processes might therefore be expected to give higher
levels of reactive aldehydes within tissues, and
partic-ularly at sites of inflammation, than would be present
in plasma Subendothelial entrapment of LDL [57–59]
may also result in more extensive LDL modification
than observed in the circulation, as a result of longer
exposure times
Overall, these studies have established that LDL
gly-cation, in the absence of significant oxidation, is
suffi-cient to induce lipid loading in primary human
macrophages, primarily via the scavenger receptors
SR-A and CD36 The accumulation of lipid in these
macrophages is accompanied by increased endocytosis
and degradation of apoB, with the difference in the
rates of the latter two processes resulting in
accumula-tion of modified apoB in HMDMs exposed to
glycolal-dehyde-modified LDL Thus, alglycolal-dehyde-modified LDL
may contribute to the increased atherosclerosis and
accumulation of glycated proteins observed in people with diabetes
Experimental procedures
Materials
Reagents were obtained from the following sources Sigma-Aldrich (Castle Hill, NSW, Australia): methylglyoxal, glycolaldehyde, fatty acid-free BSA, HSA, fucoidin, tryp-sin [type I, Na-benzoyl-l-arginine ethyl esters, 10 000 unitsÆ(mg protein))1], EDTA, Hank’s balanced salt solution (HBSS), PenStrep (100 unitsÆmL)1 penicillin, 0.1 mgÆmL)1 streptomycin), and Dulbecco’s NaCl⁄ Pi, (pH 7.4) BDH (Merck, Kilsyth, VIC, Australia): glucose Bio-Rad (Regents Park, NSW, Australia): Chelex-100 resin ICN (Seven Hills, NSW, Australia): CuSO4 Amersham Biosciences (Castle Hill, NSW, Australia): PD10 columns and Na125I (‡ 15 CiÆmg)1iodide) JRH Biosciences (CSL, North Ryde, NSW, Australia): RPMI-1640 medium Trace Scientific (Mel-bourne, VC, Australia): glutamine Australian Red Cross, Clarence St Blood Bank: human serum Axis-Shield (Oslo, Norway): Lymphoprep BD Biosciences-Pharmingen (San Diego, CA, USA): purified mouse anti-(human CD36) mAb All other chemicals were of analytical grade, and all solvents were of HPLC grade
Solutions were prepared with nanopure water (Milli Q system, Millipore-Waters, Lane Cove, NSW, Australia) treated with washed Chelex-100 resin to remove trace trans-ition metal ions, with the exception of tissue culture rea-gents, for which Baxter (Old Toongabbie, NSW, Australia) sterile, endotoxin-free, water, NaCl⁄ Pior HBSS were used
LDL modification
LDL was isolated as reported previously from multiple healthy male and female donors (four males, five females, aged 22–42 years) [29] 125I-Labeling of LDL was per-formed, prior to other modification, using iodine mono-chloride [36,60] Specific activity (typically 50–100 c.p.m.Æng)1 apoB protein) was determined by c-counting (Cobra II; Packard, Downers Grove, IL, USA) Acetylation
of LDL was performed as reported previously [29] Modifi-cation of LDL was performed as described previously [28] Briefly, sterile LDL (1 mg proteinÆmL)1) was incubated with 100 mm glycolaldehyde, methylglyoxal or glucose (± 1 lm CuSO4) in Chelex-treated NaCl⁄ Pi at 37C for
7 days Incubation controls contained 50 lm EDTA in place of glucose or aldehyde Excess reagents were removed
by elution of the LDL through PD10 columns before use Modification was confirmed by changes in relative elec-trophoretic mobility [29] LDL lipid composition (total cho-lesterol, free chocho-lesterol, triglycerides and phospholipids) was determined using a Roche Diagnostics⁄ Hitachi 902
Trang 9autoanalyzer (Roche Diagnostics GmbH, Mannheim,
Ger-many) [61,62] Cholesteryl ester concentrations were
calcu-lated as the difference between total and free cholesterol
concentrations
Isolation and culture of HMDMs
Monocytes were isolated by countercurrent elutriation
[63,64], using HBSS (with phenol red and 0.01% EDTA,
but without Ca2+ and Mg2+) White cell concentrates
were diluted 1 : 2 in HBSS, and 30 mL samples were
underlaid with 15 mL of Lymphoprep and centrifuged
using a Beckman (Palo Alto, CA, USA) GS–6KR
centri-fuge with a GH3Æ8 rotor (2060 g, 40 min, 22C)
Periph-eral mononuclear cells were isolated from the interface,
washed, and resuspended in 30 mL The cells were then
loaded into a Beckman Avanti J–20XPI centrifuge
equipped with a JE 5.0 elutriation rotor (770 g, flow rate
9 mLÆmin)1) The flow rate was increased by 1 mLÆmin)1
every 10 min, and the monocyte cell fractions collected
with flow rates of 15, 16, 17, 18 and, finally, 40 mLÆmin)1
were collected and combined The presence of monocytes
was confirmed by cytospinning and staining (Diff Quik,
Narrabeen, NSW, Australia) Cells were diluted (1.0· 106
cellsÆmL)1 in RPMI-1640, no serum), added to 12-well
plates (1 mL per well; Costar, Corning, NY, USA), and
left to adhere for 1–2 h Cells were then washed, and
RPMI medium [containing 10% heat-inactivated human
serum, 4 mm glutamine and 1% (v⁄ v) PenStrep] was
added; this was followed by incubation (5% CO2, 37C)
for 9–11 days, with the medium being changed every
3 days, to give matured HMDMs
Cellular cholesterol and cholesteryl ester analysis
HMDMs were exposed to 0 or 100 lgÆmL)1modified LDL
for 0–96 h in medium containing 10% lipoprotein-deficient
serum (prepared as reported previously [29]) Fresh LDL
and medium were added at 48 h Cell medium samples were
collected at the stated times, and the cells were washed and
lysed in water Cell viability was determined by assaying
lactate dehydrogenase release [29] Cellular cholesterol and
cholesteryl ester content was quantified using HPLC, as
described previously [29]
Cellular apoB accumulation and turnover
HMDMs were incubated with modified [125I]LDL (50 lg
proteinÆmL)1) as described above Cell medium (0.5 mL)
and cells (after being washed twice with cold NaCl⁄ Pi)
were sampled at the indicated times For turnover studies,
the [125I]LDL-containing medium was removed after the
accumulation phase The cells were then washed with
warm NaCl⁄ Pi, and medium containing 1 mgÆmL)1 BSA
was added; this was followed by incubation for 0–24 h
At the indicated times, medium (0.5 mL) was collected, and the cells were washed with cold NaCl⁄ Pi For both the accumulation and turnover studies, after the medium was collected, trypsin (1 mL, 0.01% w⁄ v) was added to the wells (60 min, 4C) to remove surface-bound ligand [36] This medium was retained to quantify cell surface-bound apoB Triton X-100 (1 mL, 0.1% v⁄ v) was then added (30 min, 4C) Of the resulting lysate, 0.5 mL was used to measure total intracellular radioactivity BSA (0.1 mL, 30 mgÆmL)1) and trichloroacetic acid (1 mL,
3 m) were added to the remaining lysate, and medium samples; this was followed by incubation (20 min, 4C) and centrifugation using a Sorvall (Sorvall Instruments, Newtown, CT, USA) RT600B centrifuge and a H1000B rotor (10 min, 1500 g, 4C) to precipitate proteins The supernatant (1 mL) was added to AgNO3 (0.25 mL, 0.7 m) and respun to precipitate free iodide One milliliter
of the iodide-free, trichloroacetic acid-soluble, supernatant from the medium or lysate was counted to quantify extra-cellular and intraextra-cellular degraded apoB, respectively [36] The medium and cell protein pellets were washed (3· 5%
w⁄ v trichloroacetic acid), and then counted to determine extracellular and intracellular nondegraded apoB, respect-ively
Receptor blocking
HMDMs were exposed to 0 or 100 lgÆmL)1 control or modified LDL for 48 h in medium containing 10% lipopro-tein-deficient serum with 200 lgÆmL)1 fucoidin [48],
2 lgÆmL)1 mAb to CD36 [47], or 200 lgÆmL)1 AGE–HSA [48] AGE–HSA was prepared by incubation of
20 mgÆmL)1 HSA with 1 m glucose for 4 weeks at 37C, followed by dialysis to remove unreacted glucose [65] At the end of 48 h, cell medium samples were collected, and the cells were washed and lysed in water Cell viability was determined by assaying lactate dehydrogenase release, and cellular cholesterol and cholesteryl ester content was quanti-fied by HPLC, as described above
Protein assay
Protein concentrations were quantified using the bicinchoni-nic acid assay (Pierce, Rockford, IL, USA) with 60 min of incubation at 60C, using BSA as a standard
Data analysis
Data are expressed as mean ± SEM from three or more separate experiments with triplicate samples One-way or two-way analysis of variance (anova) was used with Bon-ferroni’s post hoc analysis, with P < 0.05 taken as signifi-cant
Trang 10This work was supported by grants from the Diabetes
Australia Research Trust and the Australian Research
Council B E Brown and I Rashid gratefully
acknow-ledge receipt of Australian Postgraduate Awards
administered through the University of Sydney The
authors thank Professor Roger T Dean and Associate
Professor Wendy Jessup for helpful discussions, and
Mr Pat Pisansarakit for the isolation of HMDMs
References
1 Brownlee M (2001) Biochemistry and molecular cell
bio-logy of diabetic complications Nature 414, 813–820
2 Eaton JW & Dean RT (2000) Diabetes and
athero-sclerosis In Atherosclerosis (Dean RT & Kelly DT,
eds), pp 24–45 Oxford University Press, Oxford
3 Ross R & Glomset JA (1976) The pathogenesis of
atherosclerosis N Engl J Med 295, 369–416
4 Steinberg D (1997) Oxidative modification of LDL and
atherogenesis Circulation 95, 1062–1071
5 Li AC & Glass CK (2002) The macrophage foam cell as
a target for therapeutic intervention Nat Med 8, 1235–
1242
6 Goldstein JL, Ho YK, Basu SK & Brown MS (1979)
Binding site on macrophages that mediates uptake and
degradation of acetylated low density lipoprotein,
pro-ducing massive cholesterol deposition Proc Natl Acad
Sci USA 76, 333–337
7 Khaw KT, Wareham N, Bingham S, Luben R, Welch
A & Day N (2004) Association of hemoglobin A1c with
cardiovascular disease and mortality in adults: the
Eur-opean prospective investigation into cancer in Norfolk
Ann Intern Med 141, 413–420
8 Nathan DM, Lachin J, Cleary P, Orchard T, Brillon
DJ, Backlund JY, O’Leary DH & Genuth S (2003)
Intensive diabetes therapy and carotid intima-media
thickness in type 1 diabetes mellitus N Engl J Med 348,
2294–2303
9 Lopes-Virella MF, Klein RL & Virella G (1996)
Modifi-cation of lipoproteins in diabetes Diabetes Metab Rev
12, 69–90
10 Tomkin GH & Owens D (2001) Abnormalities in apo
B-containing lipoproteins in diabetes and
atherosclero-sis Diabetes Metab Rev 17, 27–43
11 Lyons TJ & Jenkins AJ (1997) Lipoprotein glycation and
its metabolic consequences Curr Opin Lipidol 8, 174–180
12 Baynes JW & Thorpe SR (2000) Glycoxidation and
lipoxidation in atherogenesis Free Radic Biol Med 28,
1708–1716
13 Thorpe SR & Baynes JW (2003) Maillard reaction
pro-ducts in tisue proteins: new propro-ducts and new
perspec-tives Amino Acids 25, 275–281
14 Zeng J & Davies MJ (2005) Evidence for the formation
of adducts and S-(carboxymethyl) cysteine on reaction
of alpha-dicarbonyl compounds with thiol groups on amino acids, peptides, and proteins Chem Res Toxicol
18, 1232–1241
15 Kawamura M, Heinecke JW & Chait A (1994) Pathophy-siological concentrations of glucose promote oxidative modification of low density lipoprotein by a superoxide-dependent pathway J Clin Invest 94, 771–778
16 Baynes JW & Thorpe SR (1999) Role of oxidative stress
in diabetic complications) a new perspective on an old paradigm Diabetes 48, 1–9
17 Chisolm GM & Steinburg D (2000) The oxidative modi-fication hypothesis of atherogenesis: an overview Free Radic Biol Med 28, 1815–1826
18 Thornalley PJ, Battah S, Ahmed N, Karachalias N, Agalou S & Babaei-Jadidi R (2003) Quantitative screen-ing of advanced glycation endproducts in cellular and extracellular proteins by tandem mass spectrometry Biochem J 375, 581–592
19 Nagai R, Hayashi CM, Xia L, Takeya M & Horiuchi S (2002) Identification in human atherosclerotic lesions of GA-pyridine, a novel structure derived from glyco-laldehyde-modified proteins J Biol Chem 348, 2294–2303
20 Schleicher ED, Wagner E & Nerlich AG (1997) Increased accumulation of the glycoxidation product N(epsilon)-(carboxymethyl) lysine in human tissues in diabetes and aging J Clin Invest 99, 457–468
21 Sakata N, Imanaga Y, Meng J, Tachikawa Y, Take-bayashi S, Nagai R & Horiuchi S (1999) Increased advanced glycation end products in atherosclerotic lesions of patients with end-stage renal disease Athero-sclerosis 142, 67–77
22 Thornalley PJ, Langborg A & Minhas HS (1999) Formation of glyoxal, methylglyoxal and 3-deoxygluco-sone in the glycation of proteins by glucose Biochem J
344, 109–116
23 Atkins TW & Thornalley PJ (1989) Erythrocyte gly-oxalase activity in genetically obese (ob⁄ ob) and strepto-zotocin diabetic mice Diabetes Res 11, 125–129
24 Odani H, Shinzato T, Matsumoto Y, Usami J & Maeda
K (1999) Increase in three a,b-dicarbonyl compound levels in human uremic plasma: specific in vivo deter-mination of intermediates in advanced Maillard reac-tion Biochem Biophys Res Commun 256, 89–93
25 Lopes-Virella MF, Klein RL, Lyons TJ, Stevenson HC
& Witztum JL (1988) Glycosylation of low-density lipo-protein enhances cholesteryl ester synthesis in human monocyte-derived macrophages Diabetes 37, 550–557
26 Dean RT, Thomas SM, Vince G & Wolff SP (1986) Oxidation induced proteolysis and its possible restric-tion by some secondary protein modificarestric-tions Biomed Biochim Acta 45, 1563–1573