© 1995 American Heart Association, Inc.
Articles |
Aminoguanidine Has Both Pro-oxidant and Antioxidant Activity Toward LDL
From the Department of Medicine, University of California San Diego, La Jolla.
Correspondence to Joseph L. Witztum, Department of Medicine 0682, Basic Science Bldg Rm 1080, University of California at San Diego, 9500 Gilman Dr, La Jolla, CA 92093.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract We previously demonstrated that aminoguanidine (AMGN) was able to prevent oxidative modification of LDL. Initially, we thought that this occurred solely because AMGN trapped reactive breakdown products of lipid peroxidation and prevented apoB modification, similar to AMGN's proposed ability to trap reactive glucose intermediates and prevent advanced glycosylation end-product formation. We now demonstrate that AMGN also displays dose-dependent pro-oxidant and antioxidant activity toward LDL. Moderate doses of AMGN (0.05 to 1.0 mmol/L) prevented lipid peroxidation in LDL exposed to copper. AMGN prevented the loss of polyunsaturated fatty acids and delayed or prevented conjugated-diene formation, both of which are sensitive indicators of lipid peroxidation. The same doses of AMGN also prevented apoB modification, a step distal to lipid peroxidation, as evidenced by the ability to (1) prevent fluorescence at 420 nm, (2) block enhanced electrophoretic mobility, and (3) prevent changes leading to enhanced macrophage uptake. Thus, AMGN inhibits LDL modification both by inhibiting lipid peroxidation as well as by trapping reactive breakdown products of lipid peroxidation. It was also demonstrated that for every LDL, there was also a very low dose of AMGN (about 0.01 mmol/L) that actually promoted lipid oxidation and subsequent protein modification. This activity of AMGN could be enhanced by increasing the content of lipid hydroperoxide in the LDL, eg, by aging or radioiodinating the LDL. Conversely, the pro-oxidant activity could be reduced by pretreatment of LDL with ebselen or vitamin E. We propose a mechanism by which AMGN effects pro-oxidant activity toward LDL at very low concentrations and antioxidant activity at higher concentrations and discuss the practical implications of these observations.
Key Words: oxidized LDL • atherosclerosis • diabetes • aminoguanidine
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Aminoguanidine (AMGN) is a small, water-soluble compound that is currently being tested for its ability to prevent the formation of advanced glycosylation end products (AGEs). AGE is the term coined by Brownlee et al1 to describe the Maillard products that are formed when organic compounds are incubated with high concentrations of glucose for prolonged periods. For example, when proteins are exposed to high glucose concentrations, available free amino groups, such as the
-amino group of lysine, undergo
nonenzymatic glycosylation to form readily reversible Schiff bases,
which then undergo Amadori rearrangement to yield stable products. In
turn, these undergo further complicated rearrangement reactions, in
part, secondary to oxidative decomposition of Amadori products, to
yield a variety of AGEs.1 2 3 4 Such AGEs lead to
cross-linking of proteins and alterations in the structure and function
of a wide variety of proteins.1 4 Evidence has been
accumulating that many of the complications of hyperglycemia may be due
to the formation of such AGEs. AMGN is a small, hydrazine compound
structurally identical to the amino-terminal group of arginine. Because
AMGN is such a strong nucleophile, it should preferentially bind
aldehydes and other breakdown products of Amadori reactions. As
originally shown by Brownlee and colleagues,5 AMGN can be
used to block reactive intermediates of AGEs and consequently prevent
AGE formation.1 6 Under oxidative stress, the polyunsaturated fatty acids (PUFAs) of LDL undergo peroxidation reactions and subsequent decomposition to yield highly reactive aldehydes and other breakdown products (reviewed in References 7 and 87 8 ). In turn, these products modify the lysine residues of apoB, thereby creating the epitopes on oxidized LDL (Ox-LDL) that form the ligand(s) leading to enhanced uptake by one or more "scavenger" receptors on macrophages.9 By analogy to the protective effect of AMGN in inhibiting AGE formation, we previously reasoned that AMGN could also trap highly reactive aldehydes (and similar products) and thus prevent apoB modification and consequent macrophage uptake. Indeed, in a recent report, we demonstrated that AMGN was capable of preventing the protein modification of LDL that leads to macrophage uptake when LDL is oxidized, whether caused by copper or endothelial cells.10 Initially we thought that this inhibition was due solely to the ability of AMGN to trap reactive aldehyde intermediates, as noted above. From the structure of AMGN NH2 || NH2NHC=NH one would not predict that it had any direct "antioxidant" activity. Nevertheless, when LDL was oxidized by exposure to copper and conjugated-diene (CD) formation was measured, we were surprised to find that at concentrations of 5 to 10 mmol/L, CD formation was greatly inhibited,10 suggesting inhibition of lipid peroxidation. In addition, we noted that at very low doses of AMGN, the lag time until initiation of CD formation was actually shortened,10 suggesting an apparent promotion of lipid peroxidation.
The unexpected observation that AMGN appeared to inhibit lipid peroxidation when added at high concentrations but actually promote lipid peroxidation when used at low concentrations led to detailed studies of the pro-oxidant and antioxidant effects of AMGN on LDL described in this report. On the basis of these results, we propose a hypothesis that would explain the opposing effects of AMGN toward LDL.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Materials
AMGN HCl (a gift from Dr Michael Yamin) was dissolved in distilled water. Vitamin E (D-
-tocopherol) (a gift
from Henkel Corp) and ebselen
(2-phenyl-1,2-benzoisoselenazol-3[2H]-one) (a gift from Ciba-Geigy)
were diluted in 100% ethanol to the concentrations indicated in each
experiment. Linoleic acid, soybean lipoxygenase (type V), superoxide
dismutase, and cytochrome c (cyt c) were
purchased from Sigma Chemical Co.
LDL Isolation
Plasma was obtained from healthy donors or hyperlipidemic
patients who were not taking antioxidant supplements. Blood was
collected into tubes containing EDTA (final concentration, 0.27
mmol/L), and final concentrations of 0.22 mmol/L gentamicin, 0.15
mmol/L chloramphenicol, 1 µmol/L
D-phenylalanyl-L-propyl-L-arginine
chloromethyl ketone, and 2 mmol/L benzamidine were added to plasma and
all solutions during subsequent LDL isolation, as previously
described.11 EDTA (0.27 mmol/L) was also present in
all solutions throughout the LDL isolation and subsequent dialysis
procedures. LDL samples isolated from individual donors were used for
most experiments except as indicated. For these samples, plasma density
was adjusted to d=1.19 g/mL with KBr, overlayered with
d=1.006 g/mL solution, and spun in an SW41 rotor at 38 000
rpm for 40 hours at 7°C as described.12 A single LDL
band was isolated and dialyzed against phosphate-buffered saline with
EDTA (0.27 mmol/L). In several experiments, LDL was isolated from
pooled plasma by sequential ultracentrifugation as previously
described.11 All samples were sterile-filtered, and
protein content was determined by the method of Lowry et
al.13 LDL was radioiodinated with 125I by the
method of Salacsinski et al14 as
described.11
Oxidation of LDL
For most of these experiments, oxidation was achieved by
incubating LDL (100 µg protein per milliliter) with CuSO4
(5 µmol/L) for varying times in the absence or presence of AMGN
and/or other reagents.15 In selected experiments, the LDL
was dialyzed at 4°C for 20 hours against phosphate-buffered saline
without EDTA just before use in oxidation experiments. In other cases,
the EDTA content of the LDL samples was adjusted (by dilution) to 2 to
3 µmol/L before use in oxidation experiments.
Measurement of Extent of LDL Oxidation
Assays of Extent of Lipid Peroxidation
In these studies we did not measure thiobarbituric
acid–reactive substances, a classic index of lipid peroxidation,
because in the presence of AMGN, malondialdehyde (or related compounds)
that form as a result of lipid decomposition react preferentially with
AMGN instead of with thiobarbituric acid (see Reference 1010 ). Thus, we
used the loss of unsaturated fatty acids in LDL as one sensitive index
of lipid peroxidation16 and CD formation17 as
another.
Fatty Acid Analysis. Fatty acids were extracted by a modification of the method of Folch et al.18 The fatty acids were transmethylated and the resulting methyl esters analyzed on a Varian gas chromatograph (model 3700) equipped with a column of 10% Silar 5CP on Gas Chrom QII, 100/200 mesh (Alltech Associates, Inc) as previously described.19 A C-15 internal standard was added to each sample before extraction, and calculation of absolute fatty acid content was determined for each LDL sample.19 Data are expressed as a percentage of total fatty acids analyzed.
CD Formation. CD formation was measured by the method of Esterbauer et al17 as previously described.20 LDL (100 µg/mL) was placed in cuvettes in phosphate-buffered saline, to which was added CuSO4 (5 µmol/L) in the presence or absence of AMGN and/or other reagents. CD formation was measured as absorbance at an optical density of 234 nm at 15-minute intervals in a continuous-recording spectrophotometer maintained at 30°C (Uvikon 810, Kontron).
Assays of Protein Modification
As a result of oxidation, the protein moiety of LDL is
profoundly modified. This was monitored by (1) changes in mobility on
agarose gel electrophoresis, (2) generation of fluorescent products,
and (3) enhanced uptake of macrophages by scavenger receptors.
Agarose Gel Electrophoresis. Electrophoresis was performed on agarose gels (Universal Gel/8) supplied by Ciba Corning Diagnostic Corp and stained with fat red 2B as described.15
Generation of Fluorescent Products. Generation of fluorescent products was assayed with a continuously recording spectrofluorometer (model L550, Perkin-Elmer) as previously described.21 An excitation wavelength of 320 nm was used and results were recorded at emission wavelengths of 350 to 500 nm.
Macrophage Degradation. In previous experiments, we directly assayed the effects of AMGN for its ability to inhibit modification of LDL leading to enhanced macrophage uptake.10 In those experiments we used iodinated LDL samples, but as described in "Results," such iodinated LDL samples are already oxidized.22 Therefore, in these experiments we prepared an iodinated Ox-LDL sample and determined its rate of uptake by macrophages. We then indirectly tested the ability of various modified LDL preparations to compete with the 125I-Ox-LDL tracer for uptake and degradation by mouse peritoneal macrophages. 125I-Ox-LDL was prepared by incubating 125I-LDL with 5 µmol/L copper for 14 hours. Unlabeled LDL samples to be used as competitors were incubated with copper for 14 hours in the absence or presence of varying doses of AMGN. Resident mouse peritoneal macrophages were prepared as described.15 125I-Ox-LDL (2.5 µg) was incubated with macrophages (5 to 8x105 cells in 0.6 mL of Dulbecco's modified Eagle's medium) for 5 hours at 37°C in the absence or presence of 25 µg of unlabeled competitor (either native LDL or LDL oxidized in the absence or presence of AMGN). The extent of degradation of the 125I-Ox-LDL tracer was determined as described.15 19
Measurement of Superoxide Anion Generation
To test the ability of AMGN to generate a superoxide anion–like
free radical from fatty acid hydroperoxide, we incubated AMGN with
linoleoyl hydroperoxide as the substrate and tested the products'
ability to reduce cyt c. Peroxidation of linoleic acid (1
mmol/L) was performed by incubation with soybean lipoxygenase (type V,
1500 U) in phosphate-buffered solution for 30 minutes as
described.21 Completion of reactions was monitored by
recording the optical density at 234 nm. Lipid hydroperoxide (LOOH, 250
nmol) was then combined with cyt c (0.58 mg) in the absence
or presence of increasing doses of AMGN. The resultant reduction of cyt
c was measured spectrophotometrically (Spectronic 1201) by
recording the change in absorbance at 549 nm as measured at 15-second
intervals for a total of 180 seconds. To document the presence of a
superoxide anion–like compound as being responsible for the reduction
of cyt c, superoxide dismutase was added in some
experiments.21
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Evidence That AMGN Inhibits Lipid Peroxidation of LDL
Loss of PUFAs is a sensitive indicator of LDL lipid peroxidation. As shown in Table 1
, when a freshly isolated LDL
preparation was exposed to copper for 5 hours under the conditions
used, the linoleic and arachidonic fatty acid contents were reduced,
from 39% and 7.5% of total LDL fatty acids, respectively, to 10% and
0%. However, when 0.05 or 0.1 mmol/L AMGN was present during the
pro-oxidant stimulus, there was near-total preservation of the linoleic
and arachidonic fatty acid content.
|
We next reexamined the ability of AMGN to prevent CD formation when LDL
was exposed to copper by using many different LDL preparations. A
representative experiment is shown in Fig 1.
Note that when this LDL preparation was exposed to copper in the
absence of AMGN, there was a lag phase of approximately 120 minutes
until initiation of the rapid propagation phase, as measured by CD
formation. However, when as little as 0.1 mmol/L AMGN was present,
no rapid propagation phase was seen, even after 16 hours. This degree
of antioxidant protection is similar to that seen with the classic
antioxidant probucol.23 Note, however, that when a very
low concentration (0.01 mmol/L) of AMGN was used in the presence of
copper, the lag phase was actually shortened, suggesting enhancement of
lipid peroxidation. In separate control experiments, addition of AMGN
to LDL—in the absence of copper—did not cause any rise in CD
formation (data not shown). When many different LDL preparations were
studied, we saw a similar pattern. However, the absolute amounts of
AMGN needed to provide complete protection (or propagation, at low
doses) varied with different LDL preparations. We noted that LDL
samples for which low doses of AMGN were very effective in preventing
CD formation were those that were used immediately after isolation,
whereas "older" LDL preparations required larger doses for
similar degrees of protection. A dramatic and exaggerated example of
this phenomenon is shown in Fig 2, in which the same LDL
sample was tested in a similar manner 6, 11, and 93 days after
isolation and storage in the dark in EDTA. Presumably, the
hydroperoxide content of LDL increased as the LDL "aged," and the
presence of preexisting hydroperoxides in the LDL affected its response
to the added AMGN.
|
|
To further test this idea, we divided a freshly isolated LDL sample
into six aliquots, some of which were "preoxidized" with copper
for varying periods before the addition of 1 mmol/L AMGN. We then
analyzed these preparations for CD formation (Fig 3).
Note that the aliquots preoxidized for 0 or 30 minutes (those that
presumably had the fewest LOOHs) were fully protected from oxidation by
the dose of AMGN used, whereas those aliquots preoxidized for 60, 90,
or 120 minutes had lag times much shorter than controls. These data are
consistent with the hypothesis that the concentration of AMGN required
to protect a given LDL preparation is proportional to the preexisting
concentrations of LOOHs.
|
Effects of AMGN on ApoB Modification
If lipid peroxidation is inhibited, then the subsequent
modification of apoB should not occur. Modification of apoB can be
monitored by the appearance of fluorescence at an emission wavelength
of approximately 420 nm.16 24 Fig 4 shows
the emission fluorescence pattern for aliquots of freshly prepared LDL
after 14 hours of exposure to copper in the presence or absence of
increasing amounts of AMGN. Immediately after addition of copper (t=0),
there was no fluorescence peak detected in either the presence or
absence of AMGN (data not shown). However, after 14-hour exposure to
copper, there was a large fluorescence peak at 420 nm characteristic of
Ox-LDL. As little as 0.1 mmol/L AMGN was able to completely inhibit
protein modification. However, at very low concentrations of AMGN (0.01
mmol/L), there actually was enhancement of emission. These data suggest
that low doses, such as 0.1 and certainly 1 mmol/L, ought to totally
inhibit protein modification and subsequent macrophage uptake. However,
in our previous studies it took 5 or even 10 mmol/L AMGN to prevent
macrophage uptake of modified LDL.10 The reason for this
discrepancy was suggested from other studies in our laboratory, which
have recently demonstrated that iodination of LDL increases its
susceptibility to oxidation by radiation-induced initiation of lipid
peroxidation.22 Thus, iodinated LDL samples would have
greater amounts of endogenous hydroperoxides and would be predicted to
require greater amounts of AMGN to prevent copper-induced
modification.
|
To test this hypothesis, we divided an LDL sample into two aliquots
(Fig 5). Fig 5A shows CD formation in freshly isolated,
noniodinated LDL and the ability of AMGN to inhibit lipid peroxidation.
Note that with this preparation, as little as 0.1 mmol/L AMGN was
highly effective in preventing CD formation, even on prolonged exposure
to copper. The other LDL aliquot, which had previously been iodinated,
was also exposed to copper under the same conditions as in Fig 5A.
However, as shown in Fig 5B, even 10 mmol/L AMGN, while prolonging CD
formation, did not totally prevent oxidation as it had with the
noniodinated LDL sample. In fact, the concentrations of 0.1 and 1.0
mmol/L AMGN, which had prevented oxidation of the nonradiolabeled LDL,
actually promoted lipid peroxidation in this iodinated sample. A
similar pattern was seen when another pair of LDL preparations were
examined for appearance of fluorescence at 420 nm, indicative of
protein modification. As shown in Fig 6, 1 mmol/L AMGN
fully protected the noniodinated LDL preparation, whereas the 0.1
mmol/L dose promoted protein modification. However, for the iodinated
LDL aliquot, all tested doses of AMGN enhanced protein modification and
none of the tested doses prevented it (Fig 6).
|
|
)
and labeled (
) LDL preparations were also studied for development
of fluorescence under identical copper-mediated conditions as described
in the legend for Fig 5
To further address the impact of iodination, we divided the same LDL
preparation into two aliquots, one of which was iodinated. Then both
aliquots were subjected to copper-mediated oxidation in the absence or
presence of increasing concentrations of AMGN, and their
electrophoretic mobility on agarose gel electrophoresis was determined
(Fig 7). Whereas as little as 0.1 mmol/L AMGN almost
completely abolished the enhanced electrophoretic mobility (due to
oxidation of noniodinated LDL), even 10 mmol/L AMGN could not
completely prevent the enhanced migration of the 125I-LDL
preparation exposed to copper.
|
In previous experiments, in which we used macrophage uptake of
125I-LDL after exposure to copper as an index of the
ability of AMGN to protect LDL, we were unable to demonstrate
protection unless concentrations as high as 10 or 25 mmol/L were
used.10 To bypass the effect of oxidation produced by
radiation from 125I, the macrophage degradation studies
were repeated with 125I-Ox-LDL as a tracer, and the ability
of unlabeled LDL preparations to compete for uptake was used as a
measure of the extent of oxidation. Unlabeled LDL was oxidized by
exposure to copper in the absence or presence of AMGN and then
incubated with macrophages together with the 125I-Ox-LDL
tracer. Using this protocol, we observed that much lower doses of AMGN
prevented changes that led to enhanced macrophage uptake (Table 2). In this experiment, LDL oxidized in the absence of
AMGN competed for 56% of the uptake of Ox-LDL tracer. However, when
unlabeled LDL was oxidized in the presence of as little as 0.1 mmol/L
AMGN, LDL was protected from oxidation and was unable to compete for
uptake.
|
Potential Mechanisms by Which Low Doses of AMGN Promote Lipid
Peroxidation
In the presence of copper, very low doses of AMGN result in the
propagation of oxidation, especially when LDL with large numbers of
preexisting hydroperoxides are used. As noted in greater detail in the
"Discussion," we propose that such propagation may be due to a
direct, nucleophilic attack by AMGN on the carbon containing an
existing peroxy radical, with consequent release of a superoxide-like
anion, which then acts on another PUFA to promote lipid peroxidation.
To directly evaluate this possibility, we determined whether AMGN could
cause the generation of superoxide-like radicals when incubated with
linoleoyl hydroperoxide. Reduction of cyt c was used as an
index of generation of superoxide-like anions. Linoleoyl hydroperoxide
was added to cyt c in the absence or presence of various
doses of AMGN, and reduction of cyt c was measured
spectrophotometrically over 15-second intervals. To obtain measurable
levels of superoxide-like radicals, 250 nmol of LOOH and higher
concentrations of AMGN were needed. In the absence of AMGN, linoleoyl
hydroperoxide did not result in reduction of cyt c. In
contrast, in the presence of AMGN, reduction of cyt c
occurred in a dose-dependent manner (Fig 8). In separate
experiments, we demonstrated that cyt c reduction was
inhibitable by increasing doses of superoxide dismutase (data not
shown), suggesting that the product generated was a superoxide
anion–like radical.
|
These data are compatible with the suggestion that AMGN is capable of
initiating lipid peroxidation by nucleophilic displacement of peroxy
radicals, generated from fatty acid hydroperoxides, to yield a
superoxide anion–like radical capable of promoting lipid peroxidation
(see the "Discussion" for further details of this hypothesis). If
this is true, then removing or greatly diminishing the content of fatty
acid hydroperoxides from LDL before adding AMGN should abolish the
ability of AMGN to promote lipid peroxidation. Ebselen
(2-phenyl-1,2-benzoisoselenazol-3[2H]-one) is a synthetic
selenium-containing heterocyclic compound that can reduce fatty acid
hydroperoxides to their corresponding (unreactive) alcohols, but
ebselen does not act as a radical-scavenging antioxidant (at least at
low doses).25 26 Thus, pretreatment of LDL with ebselen
should reduce the content of peroxy radicals (derived from
hydroperoxides) in LDL and diminish or prevent enhancement of lipid
peroxidation produced by low doses of AMGN and copper. An "old"
LDL preparation was selected for this experiment to enrich the LDL with
preexisting hydroperoxides. A low dose of ebselen was selected (2.5
µmol/L) to minimize any possibility that it might act as a
chain-breaking antioxidant (Fig 9, top). In separate
experiments, we found that pretreatment of LDL with similar doses of
ebselen alone (in the absence of reduced glutathione) reduced the
content of LOOHs by 50% to 60% (M. Ezaki and J.L. Witztum,
unpublished data, 1994), a finding similar to that in a recent
report.27 Pretreatment of LDL for 15 minutes with 2.5
µmol/L ebselen prolonged the lag phase for CD formation by 40 minutes
but did not prevent lipid peroxidation. With this LDL preparation,
addition of AMGN in the absence of ebselen—at doses of either 0.1 or
even 1.0 mmol/L—led to shortening of the lag time, consistent with the
data shown earlier for older LDL preparations. However, when the LDL
was pretreated with 2.5 µmol/L ebselen, not only was the enhanced
oxidation noted previously with both doses of AMGN prevented but also
the same doses of AMGN now completely prevented lipid peroxidation,
even after 10 hours of exposure to copper.
|
If low doses of AMGN promote lipid peroxidation by generation of
superoxide anion–like radicals, then addition of a free-radical
scavenger like -tocopherol should both block the pro-oxidant
activity of low doses of AMGN and synergistically enhance the
antioxidant activity of higher doses (Fig 9
, bottom). In this
experiment, the same older LDL preparation was used again, and 0.1
mmol/L AMGN again promoted lipid peroxidation. The addition of 20
µmol/L -tocopherol alone delayed the lag time twofold, as
expected,16 28 but did not prevent lipid peroxidation.
However, combined use of 0.1 mmol/L AMGN and 20 µmol/L vitamin E
completely prevented lipid peroxidation for as long as 16 hours.
Inherent Differences in Susceptibility to Oxidation of Different
LDL
As demonstrated above, the degree of protection given by any
particular dose of AMGN varied with different LDL preparations. We
demonstrated that this was partly due to different degrees of ex vivo
oxidation that had occurred due to differences in handling of the LDL
preparation. To test the idea that some of this variability might also
be due to different inherent susceptibility to oxidation of different
LDL samples,8 29 we simultaneously prepared four different
LDL preparations under identical conditions and subjected each to
copper-mediated oxidation in the absence or presence of varying doses
of AMGN (Fig 10). Note that the lag time of CD
formation varied among individuals and in turn the ability of any given
dose of AMGN to protect against or enhance lipid peroxidation also
varied. Thus, the varying effects of AMGN depend both on the inherent
susceptibility to oxidation of a given LDL sample as well as on changes
that may occur ex vivo, such as those due to "aging" or
iodination.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We previously suggested that AMGN might be useful in preventing oxidative modification of LDL.10 Our original suggestion was based on the idea that similar to the manner in which AMGN presumably traps reactive glucose intermediates during AGE formation and thus inhibits AGE formation,1 it would also bind aldehydes and other reactive breakdown products resulting from fatty acid peroxidation in LDL and prevent the covalent modification of apoB.7 8 Because such lipid-protein adducts are thought to form the epitopes on modified LDL that are recognized by the scavenger receptor of macrophages,8 30 AMGN could prevent the changes in Ox- LDL that lead to enhanced and unregulated macrophage uptake and foam cell formation. Indeed, we demonstrated that AMGN was capable of achieving this aim, whether oxidation of LDL was initiated by exposure to copper or to endothelial cells.10 Because of the structure of AMGN, we did not expect it to have any antioxidant activity. However, when tested for its ability to inhibit CD formation in LDL exposed to copper, we observed a prolongation of the lag time of CD formation when larger doses of AMGN were used, but paradoxically there was a shortening of the lag time when very low doses of AMGN were added.
In this article, we present multiple lines of evidence that AMGN does indeed display antioxidant activity toward LDL and prevents lipid peroxidation in LDL exposed to copper. Moderate concentrations of AMGN prevented the loss of PUFAs and delayed or prevented CD formation, both processes that are sensitive indicators of lipid peroxidation. (We did not retest the ability of AMGN to inhibit the formation of thiobarbituric acid–reactive substances, as the avidity of malondialdehyde for AMGN is so much greater than that for thiobarbituric acid that the assay cannot be used in the presence of AMGN.10 ) We demonstrated that LDL protein modification, a step distal to lipid peroxidation, was also inhibited by AMGN. This evidence included (1) the ability of AMGN to inhibit formation of fluorescence at 420 nm, typical for Ox-LDL; (2) the ability to prevent enhanced mobility on agarose gel electrophoresis; and (3) the ability to prevent changes leading to enhanced macrophage uptake. Additionally, AMGN inhibited the fragmentation of apoB that also occurs as a consequence of oxidative modification (data not shown). Thus, it is likely that AMGN inhibits apoB modification by both inhibiting lipid peroxidation per se as well as by trapping reactive breakdown products that result from lipid peroxidation of unsaturated fatty acids.
For every LDL sample there appeared to be a given dose-response curve
that described the ability of AMGN to protect against lipid
peroxidation. For most LDL samples, concentrations of 0.1 to 1 mmol/L
or greater were protective, but concentrations of 0.05 mmol/L were also
protective for some. We also consistently noted that for any given LDL,
there was a concentration of AMGN lower than that needed for protection
that conversely promoted lipid peroxidation. In general, levels of 0.01
mmol/L fell into this category, but this dose appeared to depend on the
preexisting state of oxidation of the LDL. Presumably, LDL containing
increased levels of LOOHs were more susceptible to AMGN-induced
enhancement of lipid peroxidation (and therefore, required even larger
concentrations for antioxidant protection). Thus, LDL samples that were
aged ex vivo or were iodinated22 required much larger
doses of AMGN for protection and in turn, exhibited enhanced lipid
peroxidation to lower doses of AMGN. However, some variability also
appeared to be due to intrinsic differences in the susceptibility of
LDL to oxidation (Fig 10). In fact, the differing profiles of
enhancement of CD formation in the presence of low concentrations of
AMGN (eg, 0.01 mmol/L) might be an extremely sensitive index of the
endogenous context of LOOHs in a given LDL sample.
As noted above, the structure of AMGN does not resemble that of
traditional antioxidants. In what way then does AMGN affect antioxidant
activity at some concentrations and actually act as a pro-oxidant at
lower concentrations? Although the reactions involved are likely to be
very complex, we propose the following working hypothesis to explain
these effects (Fig 11). Previously, we provided
evidence to support the idea that reactive breakdown products of lipid
peroxidation, such as malondialdehyde and 4-hydroxynonenal, form
adducts with lysine residues of apoB.31 32 Work from our
laboratory has also suggested an additional pathway by which
peroxidized fatty acids can directly react with apoB lysine residues to
generate the lipid-protein adducts that yield the fluorescent products
found with LDL oxidation (Fig 11A). As proposed by Fruebis et
al,21 there is a concerted reaction between the -amino
group of lysine and the carbon containing a peroxy radical in a
modified PUFA. In this scheme, the -amino group makes a nucleophilic
attack, thereby liberating a protonated superoxide anion–like radical,
(H)O2
- (also called a
perhydroxyl radical33 ), from LOO, which in
turn propagates the lipid peroxidation reactions (eg, conversion of LH
to LOO). In a similar manner we propose that the much
stronger nucleophile AMGN makes a concerted attack on a PUFA carbon
containing a peroxy radical. This results in a transient (or possibly
stable) AMGN–fatty acid adduct (Fig 11B) and subsequent liberation of
an (H)O2- radical (Fig 11B). In
the presence of low concentrations of AMGN, the liberated radicals can
proceed to promote lipid peroxidation.21 33 However, in
the presence of a relative excess of AMGN, we propose that the radicals
released from the initial nucleophilic displacement react with AMGN,
possibly forming a nitrone-like compound that, in turn, could decompose
to yield a variety of products, possibly including nitric oxide (Fig 11C). In support of this hypothesis we demonstrated the ability of AMGN
to react with a fatty acid hydroperoxide to generate a superoxide
anion–like radical capable of reducing cyt c. With older
LDL preparations or iodinated LDL samples that may contain more
endogenous peroxy radicals, much higher concentrations of AMGN were
needed to inhibit lipid oxidation, probably because insufficient AMGN
was available to react with the liberated superoxide-like radicals.
(Conversely, higher doses of AMGN actually promoted lipid
peroxidation.) Consistent with this hypothesis, pretreatment of LDL
with ebselen, a compound that reduces LOOH content,25 27
abolished the ability of AMGN to promote lipid peroxidation.
Furthermore, -tocopherol, a potent chain-breaking free-radical
scavenger, also prevented the pro-oxidant activity of AMGN and in fact
acted synergistically to greatly augment AMGN's antioxidant activity
(Fig 9, bottom). This likely occurred as vitamin E reduced the content
of LOO, and consequently fewer superoxide-like radicals
were generated by AMGN. It is also possible that AMGN reduced the
vitamin E radical in a manner similar to that for
ascorbate.34 35 Whatever the exact mechanism(s) involved,
the synergy between vitamin E and AMGN may have important therapeutic
consequences.
|
Obviously, the reactions involved in oxidation of LDL are complex and numerous, and it is possible that AMGN acts as an "antioxidant" by other mechanisms as well. However, our data suggest that AMGN is capable of inhibiting oxidative modification of LDL by at least two general mechanisms, one relating to its ability to trap reactive breakdown products of lipid peroxidation, such as aldehydes, and the other by its ability to act as an antioxidant.
After this manuscript was completed, Scaccini et al36 reported that 20 mmol/L AMGN was required to prevent oxidation of LDL, as measured by several different indices. This contrasts with the present data, which show that doses of 0.10 mmol/L or less completely inhibited lipid peroxidation for as long as 18 hours. We suspect that the reasons for the high doses of AMGN required by Scaccini et al were that the LDL they used was already oxidized to some extent and that many of their experiments used iodinated LDL.
AMGN is currently undergoing clinical trials in humans to test its
safety and efficacy in inhibiting complications of AGE formation in
diabetic subjects. Because AGE formation is widespread and occurs even
in circulating LDL37 where it may promote oxidation, AMGN
could have an important therapeutic impact. However, the data
presented here suggest that AMGN has a bimodal effect on lipid
peroxidation, at least for LDL, by promoting lipid peroxidation at low
concentrations and inhibiting it at higher levels. In rabbits, plasma
concentrations of 0.03 to 0.075 mmol/L AMGN can be achieved in vivo
(W.P. et al, unpublished data). Because concentrations of 0.05 to 0.10
mmol/L AMGN are capable of inhibiting oxidation of most freshly
isolated LDL, which most likely contains hydroperoxides generated at
least partly ex vivo, it seems conceivable that AMGN might act as an
antioxidant in vivo. However, the actual concentrations of AMGN in
various tissues and microenvironments (as in the intima of the artery)
where oxidation might occur are unknown. Thus, whether AMGN might have
localized antioxidant or even pro-oxidant effects in vivo is unknown.
The observation that vitamin E, a potent free-radical scavenger,
greatly potentiated the antioxidant activity of AMGN and blocked its
pro-oxidant activity (Fig 9, bottom) suggests that combination therapy
with AMGN and vitamin E might provide potent, synergistic antioxidant
activity in vivo. This may be particularly relevant, since AMGN has
also been reported to inhibit catalase, as least in
vitro.38 These data suggest that consideration should be
given to the augmented intake of vitamin E (or other antioxidants) if
AMGN therapy is to be used as a long-term therapeutic option in human
subjects.
|
Acknowledgments |
|---|
This work was supported in part by National Institutes of Health grant HL-14197 (SCOR). S. Picard was supported by an International Lilly Fellowship, a Lavoisier Fellowship (French Government), and a fellowship of the International Institute Lipha for Medical Research. The authors thank Dr Masanori Ezaki for measuring LOOHs and Drs Daniel Steinberg, Peter Reaven, and Joachim Fruebis for helpful advice and discussion; Elizabeth Miller and Felicidad Almazan for expert technical assistance; and Lisa Gallo for assistance in preparation of this manuscript.
Received October 21, 1994; accepted December 16, 1994.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Brownlee M, Cerami A, Vlassara H. Advanced glycosylation end products in tissue and the biochemical basis of diabetic complications. N Engl J Med. 1988;318:1315-1321. [Medline] [Order article via Infotrieve]
-
Pongor S, Ulrich PC, Bencsath FA, Cerami A. Aging of
proteins: isolation and identification of a fluorescent chromophore
from the reaction of polypeptides with glucose. Proc Natl Acad
Sci U S A. 1984;81:2684-2688.
[Abstract/Free Full Text] - Njoroge FG, Monnier VM. The chemistry of the Maillard reaction under physiological conditions: a review. Prog Clin Biol Rev. 1989;304:85-107.
- Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes. 1991;40:405-412. [Abstract]
-
Brownlee M, Vlassara H, Kooney A, Cerami A. Aminoguanidine
prevents diabetes-induced arterial wall protein cross-linking.
Science. 1986;232:1629-1632.
[Abstract/Free Full Text] - Edelstein D, Brownlee M. Mechanistic studies of advanced glycosylation end product inhibition by aminoguanidine. Diabetes. 1992;41:26-29. [Abstract]
- Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest. 1991;88:1785-1792.
-
Witztum JL. Role of oxidised low density lipoprotein in
atherogenesis. Br Heart J. 1993;69:512-518.
[Abstract/Free Full Text] - Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915-924. [Medline] [Order article via Infotrieve]
-
Picard S, Parthasarathy S, Fruebis J, Witztum JL.
Aminoguanidine inhibits the oxidative modification of low density
lipoprotein and the subsequent increase in uptake by macrophage
scavenger receptors. Proc Natl Acad Sci U S A. 1992;89:6876-6880.
[Abstract/Free Full Text] -
Reaven PD, Khouw A, Beltz WF, Parthasarathy S, Witztum JL.
Effect of dietary antioxidant combinations in humans: protection of low
density lipoprotein by vitamin E but not by ß-carotene.
Arterioscler Thromb. 1993;13:590-600.
[Abstract/Free Full Text] -
Reaven PD, Ferguson E, Navab M, Powell F. Susceptibility of
human LDL to oxidative modification: effects of variations in
ß-carotene concentration and oxygen tension.
Arterioscler Thromb. 1994;14:1162-1169.
[Abstract/Free Full Text] -
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein
measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.
[Free Full Text] - Salacsinski PR, McLean C, Sykes JE, Clement-Jones VV, Lowry PJ. Iodination of proteins, glycoproteins and peptides using a solid-phase oxidizing agent, 1,3,4,6 tetrachloro-3 alpha,6-alpha-diphenyl glycoluril (Iodogen). Anal Biochem. 1981;117:136-146. [Medline] [Order article via Infotrieve]
-
Steinbrecher UP, Parthasarathy S, Leake DS, Witztum JL,
Steinberg D. Modification of low density lipoprotein by endothelial
cells involves lipid peroxidation and degradation of low density
lipoprotein phospholipids. Proc Natl Acad Sci U S A. 1984;81:3883-3887.
[Abstract/Free Full Text] - Esterbauer H, Gebicki J, Puhl H, Jürgens G. The role of lipid peroxidation and antioxidants in oxidative modification of LDL. Free Radic Biol Med. 1992;13:341-382. [Medline] [Order article via Infotrieve]
- Esterbauer H, Striegl G, Puhl H, Rotheneder M. Continuous monitoring of in vitro oxidation of human low density lipoprotein. Free Radic Res Commun. 1989;6:67-75. [Medline] [Order article via Infotrieve]
-
Folch J, Lees JM, Stanley GHS. A simple method for the
isolation and purification of total lipids from animal tissues. J
Biol Chem. 1957;226:497-509.
[Free Full Text] - Reaven P, Parthasarathy S, Grasse B, Miller E, Steinberg D, Witztum JL. Effects of oleate rich and linoleate rich diets on the susceptibility of low density lipoprotein to oxidative modification in mildly hypercholesterolemic subjects. J Clin Invest. 1993;91:668-676.
-
Parthasarathy S, Khoo JC, Miller E, Barnett J, Witztum JL,
Steinberg D. Low density lipoprotein rich in oleic acid is protected
against oxidative modification: implications for dietary prevention of
atherosclerosis. Proc Natl Acad Sci U S A. 1990;87:3894-3898.
[Abstract/Free Full Text] -
Fruebis J, Parthasarathy S, Steinberg D. Evidence for a
concerted reaction between lipid hydroperoxides and polypeptides.
Proc Natl Acad Sci U S A. 1992;89:10588-10592.
[Abstract/Free Full Text] - Khouw AS, Parthasarathy S, Witztum JL. Radioiodination of low density lipoprotein initiates lipid peroxidation: protection by use of antioxidants. J Lipid Res. 1993;34:1483-1496. [Abstract]
- Parthasarathy S, Young SG, Witztum JL, Pittman RC, Steinberg D. Probucol inhibits oxidative modification of low density lipoprotein. J Clin Invest. 1986;77:641-644.
-
Steinbrecher UP. Oxidation of human low density lipoprotein
results in derivatization of lysine residues of apolipoprotein B by
lipid peroxide decomposition products. J Biol Chem. 1987;262:3603-3608.
[Abstract/Free Full Text] - Noguchi N, Yoshida V, Kaneda H, Yamamoto Y, Niki E. Action of ebselen as an antioxidant against lipid peroxidation. Biochem Pharmacol. 1992;44:39-44. [Medline] [Order article via Infotrieve]
-
Thomas CE, Jackson RL. Lipid hydroperoxide involvement in
copper-dependent and independent oxidation of low density lipoproteins.
J Pharmacol Exp Ther. 1991;256:1182-1188.
[Abstract/Free Full Text] - Sattler W, Maiorino M, Stocker R. Reduction of HDL- and LDL-associated cholesterylester and phospholipid hydroperoxides by phospholipid hydroperoxide glutathione peroxidase and Ebselen (PZ 51). Arch Biochem Biophys. 1994;309:214-221. [Medline] [Order article via Infotrieve]
-
Reaven PD, Witztum JL. Comparison of supplementation of
RRR--tocopherol and racemic -tocopherol in humans.
Arterioscler Thromb. 1993;13:601-608.
[Abstract/Free Full Text] -
Jialal I, Freeman DA, Grundy SM. Varying susceptibility of
different low density lipoproteins to oxidative modification.
Arterioscler Thromb. 1991;11:482-488.
[Abstract/Free Full Text] -
Parthasarathy S, Fong LG, Otero D, Steinberg D. Recognition of
resolubilized apoproteins from delipidated, oxidatively modified low
density lipoprotein (LDL) by the acetyl-LDL receptor. Proc Natl
Acad Sci U S A. 1987;84:537-540.
[Abstract/Free Full Text] -
Steinbrecher UP, Witztum JL, Parthasarathy S, Steinberg D.
Decrease in reactive amino groups during oxidation or endothelial cell
modification of LDL: correlation with changes in receptor-mediated
catabolism. Arteriosclerosis. 1987;7:135-143.
[Abstract/Free Full Text] -
Palinski W, Ylä-Herttuala S, Rosenfeld ME, Butler S,
Socher SA, Parthasarathy S, Curtiss LK, Witztum JL. Antisera and
monoclonal antibodies specific for epitopes generated during the
oxidative modification of low density lipoprotein.
Arteriosclerosis. 1990;10:325-335.
[Abstract/Free Full Text] -
Aikens J, Dix TA. Perhydroxyl radical (HOO·) initiated
lipid
peroxidation: the role of fatty acid hydroperoxides. J Biol
Chem. 1991;266:15091-15098.
[Abstract/Free Full Text] - Kagan VE, Serbinova EA, Forte T, Scita G, Packer L. Recycling of vitamin E in human low density lipoproteins. J Lipid Res. 1992; 33:385-397.
- Jialal I, Grundy SM. Preservation of the endogenous antioxidants in low density lipoproteins by ascorbate but not probucol during oxidative modification. J Clin Invest. 1991;87:597-601.
- Scaccini C, Chiesa G, Jialal I. A critical assessment of the effects of aminoguanidine and ascorbate on the oxidative modification of LDL: evidence for interference with some assays of lipoprotein oxidation by aminoguanidine. J Lipid Res. 1994;35:1085-1092.[Abstract]
-
Bucala R, Makita Z, Koschinsky T, Cerami A, Vlassara H. Lipid
advanced glycosylation: pathway for lipid oxidation in vivo. Proc
Natl Acad Sci U S A. 1993;90:6434-6438.
[Abstract/Free Full Text] - Ou P, Wolff SP. Aminoguanidine: a drug proposed for prophylaxis in diabetes inhibits catalase and generates hydrogen peroxide in vitro. Biochem Pharmacol. 1993;46:1139-1144.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  R. A. Patterson, E. T. M. Horsley, and D. S. Leake Prooxidant and antioxidant properties of human serum ultrafiltrates toward LDL: important role of uric acid J. Lipid Res., March 1, 2003; 44(3): 512 - 521. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Stitt, T. A. Gardiner, N. L. Anderson, P. Canning, N. Frizzell, N. Duffy, C. Boyle, A. S. Januszewski, M. Chachich, J. W. Baynes, et al. The AGE Inhibitor Pyridoxamine Inhibits Development of Retinopathy in Experimental Diabetes Diabetes, September 1, 2002; 51(9): 2826 - 2832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Exner, M. Hermann, R. Hofbauer, S. Kapiotis, P. Quehenberger, W. Speiser, I. Held, and B. M.K Gmeiner Semicarbazide-sensitive amine oxidase catalyzes endothelial cell-mediated low density lipoprotein oxidation Cardiovasc Res, June 1, 2001; 50(3): 583 - 588. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. F. Wu, L. Pewe, and S. Perlman Coronavirus-Induced Demyelination Occurs in the Absence of Inducible Nitric Oxide Synthase J. Virol., August 15, 2000; 74(16): 7683 - 7686. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. J. Lyons, W. Li, B. Wojciechowski, M. C. Wells–Knecht, K. J. Wells–Knecht, and A. J. Jenkins Aminoguanidine and the Effects of Modified LDL on Cultured Retinal Capillary Cells Invest. Ophthalmol. Vis. Sci., April 1, 2000; 41(5): 1176 - 1180. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. Zhang, S. Ren, D. Sun, and G. X. Shen Influence of Glycation on LDL-Induced Generation of Fibrinolytic Regulators in Vascular Endothelial Cells Arterioscler. Thromb. Vasc. Biol., July 1, 1998; 18(7): 1140 - 1148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-X. Fu, J.ús R. Requena, A. J. Jenkins, T. J. Lyons, J. W. Baynes, and S. R. Thorpe The Advanced Glycation End Product, N[IMAGE]-(Carboxymethyl)lysine, Is a Product of both Lipid Peroxidation and Glycoxidation Reactions J. Biol. Chem., April 26, 1996; 271(17): 9982 - 9986. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||