© 1999 American Heart Association, Inc.
Atherosclerosis and Lipoproteins |
Human Serum Paraoxonase/Arylesterase's Retained Hydrophobic N-Terminal Leader Sequence Associates With HDLs by Binding Phospholipids
Apolipoprotein A-I Stabilizes Activity
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract—In serum, human paraoxonase/arylesterase (PON1) is found exclusively associated with high density lipoprotein (HDL) and contributes to its antiatherogenic properties by inhibiting low density lipoprotein (LDL) oxidation. Difficulties in purifying PON1 from apolipoprotein A-I (apoA-I) suggested that PON1's association with HDL may occur through a direct binding between these 2 proteins. An unusual property of PON1 is that the mature protein retains its hydrophobic N-terminal signal sequence. By expressing in vitro a mutant PON1 with a cleavable N-terminus, we demonstrate that PON1 associates with lipoproteins through its N-terminus by binding phospholipids directly rather than binding apoA-I. Nonetheless, apoA-I stabilized arylesterase activity more than did phospholipid alone, apoA-II, or apoE. Consequently, we studied the role of apoA-I in PON1 expression and HDL association in mice genetically deficient in apoA-I. Though present in HDL fractions at decreased levels, PON1 arylesterase activity was less stable than in control mice. Furthermore, PON1 could be competitively removed from HDL by phospholipids, suggesting that PON1's retained N-terminal peptide allows transfer of the enzyme between phospholipid surfaces. Thus, our data suggest that PON1 is stabilized by apoA-I, and its binding to HDL and physiological distribution are dependent on the direct binding of the retained hydrophobic N-terminus to phospholipids optimally presented in association with apoA-I.
Key Words: atherogenesis • lipid peroxidation • HDL • LDL oxidation • paraoxonase
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Human serum paraoxonase/arylesterase (PON1) is exclusively associated with HDLs and is a genetically polymorphic enzyme that has the ability to hydrolyze a wide range of organophosphates and carboxyesters.1 In addition, PON1 is 1 of the few enzymes or modifying transfer proteins that interacts with serum lipoproteins. Nevertheless, surprisingly little is known about the requirements for PON1 to associate with HDL or the structure of its endogenous substrates. Epidemiological and biochemical evidence is accumulating, however, which indicates that PON1 catalyzes the removal of the biologically active products of lipid peroxidation2 3 4 5 6 and that the aforementioned genetic polymorphism may result in phenotypes with different susceptibilities to coronary heart disease.7 8 9 10 11 12 13 14 15 In humans, a well-established inverse relationship exists between plasma HDL levels and cardiovascular disease.16 17 The antiatherogenic qualities of HDL may in part be due to the activities of PON1. Indeed, PON1-deficient mice are more susceptible to organophosphate toxicity as well as to diet-induced atherosclerosis than are their wild-type littermates.18
The exclusive association of circulating PON1 with HDL also implies that PON1's expression and biological role are affected by other HDL-associated proteins. For example, PON1 may only encounter its endogenous substrates and have biological significance in vivo if the enzyme is distributed with HDL. Therefore, the structural features of PON1 that determine its association with HDL may be as important as its catalytic components. Other apolipoproteins may contribute to PON's association with HDL. In this regard, immunopurified PON1 is found to be associated with 2 major proteins, apoA-I and apoJ.19 20 Furthermore, Blatter and coworkers20 demonstrated that 90% of PON1 coimmunopurifies with anti–apoA-I antibodies, 30% coimmunopurifies with anti-apoJ antibodies, and 10% coimmunopurifies with anti–apoA-II antibodies. In all cases, apoA-I coimmunopurified with PON1, regardless of which antibodies were used. Thus, these studies suggest that virtually all PON1 is associated with apoA-I. It is notable, however, that PON1 does not coimmunopurify with any apolipoproteins in the presence of nonionic detergents.21 These findings are consistent with the fact that separating the enzyme from apoA-I was a significant advance in the purification of PON1. Interestingly, the isolation of PON1 in an active form was found to be possible only through the addition of a nonionic detergent.22 23
The close association between PON1 and apoA-I suggests direct binding
of the 2 proteins. Direct sequencing of PON1 purified from human and
rabbit sera has demonstrated an unusual property: in both species, the
mature, circulating form retains its N-terminal hydrophobic
signal sequence.20 24 25 26 This observation has led to
the suggestion that PON1's hydrophobic N-terminal sequence
may facilitate its association with HDL. Retention of this signal
peptide has been postulated to be the consequence of an evolutionary
substitution of large, polar residues at the signal-peptide cleavage
site where smaller residues are typically present, thereby eluding
cleavage by signal peptidases.25 26 The presence of
similar large, polar residues in the deduced N-terminal
sequence from mouse PON1 suggests that it is also retained (Figure 1
) and that this is a generalized feature
of mammalian PON1s.27 28 29 We postulated that changing
the histidine (H) and glutamine (Q) residues in positions 20 and 21,
respectively, to alanines by mutagenesis would allow the in vitro
expression of a normally processed and secreted PON1 enzyme. The
expressed protein would have its N-terminal hydrophobic
signal sequence cleaved and serve as a model to determine whether this
region is required for PON1's association with lipoproteins. In the
current study, we present evidence that the retained
N-terminal hydrophobic signal peptide is the structural
requirement for PON1 to associate with HDL. We also show that the
N-terminal peptide binds to phospholipids directly, in the
absence of any apolipoproteins. Furthermore, using apoA-I–deficient
mice, we define a role for this protein in stabilizing PON1. The
association of PON1 and apoA-I with HDL appears to be largely mediated
through their common binding to lipids rather than a direct binding
between them. Hence, PON1 is able to circulate with HDL by directly
binding to HDL-associated phospholipids through its retained
N-terminal hydrophobic peptide.
|
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Site-Directed Mutagenesis
Creation of wild-type PON1 (type Q) and its sequence verification have been described previously.27 To allow the recombinant, mutant PON1 enzyme to be normally targeted to the endoplasmic reticulum for appropriate folding, glycosylation, and signal-sequence cleavage, we chose to mutate both residues H20 and Q21 to create a theoretically cleavable site. These specific residues were chosen because they were in the -3 position from the predicted cleavage site if small, nonpolar residues were in their place.26 30 31 Site-directed mutagenesis to replace H20 and Q21 simultaneously with alanines was accomplished as described previously by using wild-type PON1 (type Q) as a template with a mutagenic polymerase chain reaction primer to create the cDNA for PON1 (type Q) A20A21 (5'-CTGGCACTCTTCAGGAACGCCGCGTCTTCTTA-CCAAACACGA-3'). The nucleotide sequence of the entire coding region of PON1 (type Q) A20A21 was determined by the University of Michigan DNA Sequencing Core Facility by automated fluorescent sequencing and manually using dideoxy chemistry (fmol Cycle Sequencing Kit from Promega). The mutant cDNA was subcloned into a pGS expression vector as previously described.27
Transfection
Human embryonic kidney 293T/17 cells were grown in 10% FBS in
minimal essential medium with Earle's salts (MEM) at 37°C, 95%
humidity, and 5% CO2. Transfection was performed
with calcium phosphate precipitates as described
elsewhere32 by using 20 µg of plasmid DNA. Recombinant
enzyme was collected in MEM with either 1% or 10% FBS 96 hours after
removal of the DNA. Control medium was obtained by transfecting,
washing, and collecting medium as described above, except that the pGS
vector without the cDNA insert was used. In all cases, the control and
experimental media being compared were collected from an equal number
of simultaneously transfected cells, in the identical
preparation of culture medium, and the same duration of collection of
expressed enzyme. Expression medium was concentrated 10-fold with a
Centriprep 30 (Amicon) according to the manufacturer's
instructions.
Preparation of Cell Extracts
Adherent cells were trypsinized after collection of media,
suspended in PBS, and pelleted by centrifugation at
1000g for 10 minutes. The supernatant was removed and the
cells were washed by resuspension in PBS. The procedure was repeated 2
times, and the pelleted cells were resuspended in 1 mL of 50
mmol/L Tris-Cl, pH 8.0, containing 1 mmol/L
CaCl2 and 1% Triton X-100 (Sigma Chemical Co).
The suspension was then centrifuged at 12 000g and
the supernatant was removed for further use.
Activity Assays/Kinetics
Arylesterase activities were determined by using phenyl acetate
as a substrate according to Eckerson et al.33 In brief,
unless otherwise specified, all activities were measured at 270 nm in a
dual-beam spectrophotometer at 25°C in 50 mmol/L Tris-Cl, pH
8.0, and 1 mmol/L CaCl2, with the reference
cuvette being identical to the experimental one except for the addition
of expression medium to the experimental cuvette. Except for
determining kinetic constants, all assays were performed with 1
mmol/L phenyl acetate as a substrate. Background activities were
determined for all substrates at all concentrations by measuring the
activity in the control medium. This background activity was subtracted
to give the net activity of the expressed enzymes. Apparent Michaelis
constants (Km) and maximal velocities
(Vmax) were derived from Eadie-Hoffstee plots
through least-squares regression with the use of Statview software (SAS
Institute Inc).
Preparation of Lipid Vesicles and Proteoliposomes
Vesicles without apolipoprotein (PC/C vesicles) and
proteoliposome vesicles containing human apoA-I (AI/PC/C vesicles),
human apoA-II (AII/PC/C vesicles), or human apoE (E/PC/C vesicles) were
prepared by the cholate dialysis method essentially as described by
Chen and Albers.34 Purified, lyophilized preparations of
apo A-I, A-II, and E isolated from pooled, human plasma were a gift
from Dr Conrad Blum (Columbia University College of Physicians and
Surgeons, New York, NY).35 36 PC/C vesicles were prepared
with egg phosphatidylcholine/cholesterol in a molar ratio
of 250:12.5 in the presence of sodium cholate, whereas proteoliposomes
were prepared containing apolipoprotein/egg
phosphatidylcholine/cholesterol in a molar ratio of
0.8:250:12.5, also in the presence of sodium cholate as
described.34 Two separate vesicle batches containing
different concentrations of Tris-Cl and CaCl2
were made. Cholate was removed from the vesicles by extensive washing
in Centriflo CF25 ultrafiltration membrane cones (Amicon). For the
first batch, PC/C and AI/PC/C vesicles were prepared with 10
mmol/L Tris-HCl, pH 8.0, containing 0.2 mmol/L
CaCl2 as both the vesicle preparation and
dialysis buffer. For the second batch, we prepared PC/C, AI/PC/C,
AII/PC/C (molar ratio based on the monomeric form of apoA-II), or
E/PC/C vesicles; however, 50 mmol/L Tris-HCl, pH 8.0, containing
1.0 mmol/L CaCl2 was used for the dialysis
step. After dialysis, preparations were adjusted with dialysis buffer
to 36 µmol/L apoA-I (batch 1) or to 30 µmol/L
apolipoproteins (batch 2). Vesicles without apolipoproteins were
adjusted to an equal volume of dialysis buffer, and all preparations
were stored at 4°C until use. The apolipoprotein preparations alone
or when reconstituted as vesicles had no detectable arylesterase
activity.
Purification of Human PON1
Human serum PON1 (type Q) was purified as described by Gan et
al23 and later modified by Kuo and La
Du.22
Incubations/Size-Exclusion Chromatography
Media (500 µL, containing 1.9 mmol/L
CaCl2) from control (pGS vector without cDNA
insert), A20A21 mutant, and wild-type (type Q) PON1s were individually
incubated with 500 µL of 10 mmol/L Tris-HCl, pH 8.0/0.2
mmol/L CaCl2 buffer alone, with PC/C vesicles, or
with AI/PC/C vesicles overnight at 37°C. Aliquots of the control,
mutant, or wild-type media, incubated with or without vesicles, were
filtered (600 µL) through 0.45-µm Ultrafree-MC filters (Millipore
Corp). Aliquots (500 µL) were then separated by Superose 6HR
(exclusion limit 4x107
Mr) gel-filtration
chromatography (1x30-cm column) at a flow rate of 0.8
mL/min of 10 mmol/L Tris-HCl, pH 8.0, with the eluant monitored at
280 nm (ie, for turbidity and aromatic amino acids). Forty 0.8-mL
fractions were collected and analyzed individually or after
selective pooling. The experiment was repeated except that total
incubation volumes were increased to 600 µL (300 µL of cell medium
and 300 µL of buffer or vesicles), and the column was equilibrated
and eluted with 10 mmol/L Tris-HCl, pH 8.0, containing 0.2
mmol/L CaCl2.
Additional PC/C, AI/PC/C, AII/PC/C, and E/PC/C vesicles were prepared as described above. Concentrated (10-fold) control, A20A21, mutant, or wild-type PON1 media (400 µL) were each incubated with AI/PC/C vesicles (400 µL) in 25 mmol/L Tris-HCl, pH 8.0, containing 1.5 mmol/L CaCl2 overnight at 37°C. Additional incubations of 400 µL of wild-type PON1 media (not concentrated) plus 400 µL of AI/PC/C, AII/PC/C, E/PC/C, or PC/C vesicles in 25 mmol/L Tris-HCl, pH 8.0, containing a final concentration of 1.5 mmol/L CaCl2 were also performed overnight at 37°C. Aliquots were filtered through 0.45-µm Ultrafree-MC filters and separated by gel-filtration chromatography, and the fractions were collected as described above, except that the column was equilibrated with 50 mmol/L Tris-HCl, pH 8.0, containing 1 mmol/L CaCl2.
Polyacrylamide Gel Electrophoresis
(PAGE)/Immunostaining
PAGE was performed in the presence of SDS and
ß-mercaptoethanol except where noted for nondenaturing
electrophoresis. Gradient gels (4% to 20% polyacrylamide)
were purchased from Novex. Transfer to polyvinylidene
difluoride membranes (Bio-Rad) was performed as described
elsewhere.32 Mutant and wild-type PON1 proteins were
detected with chicken antiserum to purified human serum PON prepared by
HRP, Inc. Human apoA-I was detected with sheep antiserum to purified
human apoA-I (Boehringer Mannheim Co, catalog No. 726 478).
Mouse apoA-I was detected with a cross-reacting rabbit antiserum to
purified rat apoA-I. Rabbit IgG was visualized with alkaline
phosphatase–conjugated goat antiserum (Sigma, catalog No. A-3687).
Chicken and sheep IgG's were visualized with alkaline
phosphatase–conjugated rabbit antiserum (Sigma, catalog No. A-9171)
and donkey antiserum (Sigma, catalog No. A-7789),
respectively.
Density Centrifugation
For density centrifugation, concentrated mutant
or wild-type medium (75 µL) was adjusted to 1.21 g/mL with KBr in
50 mmol/L Tris-HCl, pH 8.0, with 1 mmol/L
CaCl2. Samples were centrifuged at
100 000g for 24 hours in a fixed-angle rotor in a TL100
ultracentrifuge (Beckman Instruments). To remove KBr from the
upper 0.5-mL (d<1.21 g/mL) and the lower 2.0-mL infranatant
(d>1.21 g/mL) fractions, the samples were exchanged into
50 mmol/L Tris-HCl, pH 8.0, with 1 mmol/L
CaCl2 by using Centricon 30 filters (Amicon). The
d<1.21 g/mL and d>1.21 g/mL fractions were
assayed for arylesterase activity, separated by SDS-PAGE, and assessed
for PON1 immunoreactivity by Western blotting.
Pooled sera (60 µL) from 6 C57BL/6J wild-type or 6 apoA-I–deficient
mice (3 females and 3 males in each pool)37 obtained from
Jackson Laboratories (Bar Harbor, Me) were incubated with 60 µL of
either PC/C vesicles or 50 mmol/L Tris-HCl with 1 mmol/L
CaCl2, pH 8.0, for 16 hours at 37°C. After
incubation, the samples were separated by isopyknic density gradient
ultracentrifugation in an SW 40.1 rotor (Beckman
Instruments). In brief, 100 µL of the incubation mixture was adjust
to a volume of 2.2 mL and d=1.15 g/mL (with KBr, containing
50 mmol/L Tris-HCl with 1 mmol/L CaCl2)
and layered above 2.2-mL density layers of d=1.25 and
d=1.20 g/mL and below 2.2-mL density layers of
d=1.10 and d=1.05 g/mL.
Centrifugation was performed at 40 000g for
24 hours, and 11 fractions 
1.1 mL each were carefully aspirated with
an elongated capillary pipet. Fraction densities were determined
gravimetrically. Fractions were assessed for arylesterase activity.
Fractions from the wild-type mice were desalted,
delipidated,38 separated by SDS-PAGE, and assessed
for mouse apoA-I immunoreactivity by Western blotting.
Incubation of Purified, Detergent-Free, Human Serum PON1 and
Purified, Delipidated, Human Serum ApoA-I
Detergent was removed from purified human serum PON1 by using an
Extracti-Gel D column and following the manufacturer's recommendations
(Pierce Chemical Co). PON1 (18 µg) and apoA-I (345 µg) were then
incubated either individually or in combination with a 30-fold molar
excess of PON1 in 50 mmol/L Tris-HCl, 1 mmol/L
CaCl2, pH 8.0, at 37°C for 16 hours in a volume
of 700 µL containing 0.35 mL of the PC/C vesicles described above.
After incubation, the mixture was adjusted to a density of 1.21 g/mL
with KBr and a final volume of 3.5 mL in 50 mmol/L Tris-HCl,
1 mmol/L CaCl2, pH 8.0. Binding of PON1 to
phospholipid was determined after ultracentrifugation.
The top 0.5-mL, middle 1.5-mL, and bottom 1.5-mL fractions of PON1 were
then assayed for arylesterase activity for PON1 localization.
To determine whether PON1 and apoA-I could associate in the absence of phospholipids, the 2 proteins were incubated individually and together with a 30-fold molar excess of PON1 in 50 mmol/L Tris-HCl, 1 mmol/L CaCl2, pH 8.0, at 37°C for 16 hours in a volume of 3.5 mL. From this, 30-µL aliquots were taken for analysis. The presence or absence of binding between PON1 and apoA-I was determined by assessment of altered electrophoretic migration after nondenaturing PAGE and immunoblotting with anti–apoA-I antiserum alone or in combination with anti-PON1 antiserum.
LDL Oxidation
Human serum LDL was obtained from PerImmune, Inc. Before LDL
oxidation, the lipoprotein was dialyzed against EDTA-free PBS, pH 7.4,
under N2 at 4°C. LDL oxidation and peroxidation
were analyzed as described previously.6 In brief,
the lipoproteins (100 µg of protein per mL) were incubated with
10 µmol/L CuSO4 in air in the absence or
presence of the indicated concentrations of recombinant PON1 enzymes
for up to 4 hours at 37°C. LDL oxidation was measured directly in the
medium by the thiobarbituric acid–reactive substances (TBARS) assay at
532 nm with the use of malondialdehyde to generate a standard curve.
Lipoprotein oxidation was also determined by the lipid peroxides test,
which analyzes their capacity to convert iodide to
I2, as measured photometrically at 365 nm.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The A20A21 Mutant PON1 Has a Cleaved N-Terminal Signal Peptide and Reduced Arylesterase Activity
Kidney 293T/17 cells were transiently transfected with pGS expression vectors containing wild-type or mutant PON1 cDNA. After extensive washing of the cells, cellular proteins were assessed by SDS-PAGE and immunostaining for PON1 (Figure 2
). Figure 2 shows that the
recombinant A20A21 mutant enzyme is truncated compared with the
wild-type recombinant enzyme and that cleavage occurs intracellularly.
The estimated sizes of the wild-type and mutant recombinant proteins
were 44.2 and 41.6 kDa, respectively. The calculated size difference of
2.6 kDa is consistent with that expected if the enzyme were
cleaved as predicted at the 22nd or 23rd residue.30 In
addition, by using phenyl acetate as the substrate (ie, arylesterase
activity), the N-terminal mutant enzyme shows an 4-fold
increase in apparent Km (Table 1). The mutant enzyme preparation also
had a substantial decrease in activity, showing an 17-fold lower
apparent Vmax than the wild-type preparation.
|
|
The PON1 N-Terminal Hydrophobic Signal Sequence Is
Required for Binding to Phospholipids, Proteoliposomes, and Serum
Lipoproteins
Bovine lipoproteins present in the expression medium were used
to model PON1's association with HDL with and without its
N-terminal peptide. Samples of concentrated, recombinant,
wild-type and A20A21-mutant PON1 expression media were subjected to
density ultracentrifugation at d=1.21 g/mL
to determine whether the expressed enzymes would colocalize with
lipoproteins in the top (ie, d<1.21 g/mL) or bottom (ie,
d>1.21 g/mL) free protein fraction. Figure 3 demonstrates that the wild-type PON1
was present with lipoproteins in the top fractions at
d<1.21 g/mL. It is also apparent that the wild type was
present in the bottom fraction. This could be due to incomplete
removal of the top fraction before the remaining sample was
concentrated for analysis, or it could be that human
recombinant PON1 is displaced from bovine lipoproteins under high salt
conditions during ultracentrifugation. Displacement of
apolipoproteins during ultracentrifugation is known to
occur.39 40 The presence of bovine apoA-I in all
fractions shows that apoA-I–containing lipoproteins were present
in FBS and were detected by cross-reacting antibodies contaminating the
PON1 chicken polyclonal antiserum. This cross-reactivity was observed
in concentrated samples only, and the identity of the 28-kDa band as
bovine apoA-I in the medium was also confirmed in other experiments
with specific antiserum to apoA-I (data not shown). It is notable that
apoA-I was found predominantly in the d<1.21 g/mL (top)
fraction in both the wild-type and mutant PON1 preparations. Mutant
PON1, on the other hand, was present exclusively in the
d>1.21 g/mL fraction, as expected for free proteins. Thus,
it appears that the N-terminus was required for association
with lipoproteins. However, the inability of mutant PON1 to associate
with bovine lipoproteins could have been due to an inability to bind
apoA-I, a reduced affinity for bovine apoA-I, an inability to bind to
phospholipids, or denaturation of the mutant enzyme.
|
To better understand the specific requirements for PON1 binding to
lipoproteins or their associated phospholipids, we utilized synthetic
vesicles and proteoliposomes with defined constituents. We incubated
wild-type and N-terminal–deleted PON1 expression media with
AI/PC/C and PC/C vesicles overnight at 37°C and followed the elution
profile of PON1 and apoA-I by gel-filtration
chromatography on Superose 6HR. The large size of the
vesicles allowed fractionation by size-exclusion
chromatography to assess binding of PON1 enzyme to the
vesicles. The large exclusion limit of the column resin also allowed us
to distinguish between binding to the vesicles versus extraction of
phospholipids by PON1. The presence of vesicles containing the expected
apoA-I proteins and lipid was indicated by
immunoblotting for apoA-I and through
spectrophotometric absorbance at 280 nm, which detected protein and
light scattering by the vesicles (Figure 4). Adjacent fractions were collected for
selective pooling (ie, pools A through D) and assayed for the presence
of PON1 and apoA-I by Western blotting with antisera against human PON1
and human apoA-I (Figure 4, insets). Wild-type PON1 eluted early
with the large, AI/PC/C vesicles in pools A and B, with lesser amounts
trailing into pool C. In contrast, the N-terminal–deleted
enzyme did not associate with the vesicles and eluted predominantly in
pool C. ApoA-I–deficient PC/C vesicles were also present
predominantly in pools A and B. When incubated with PC/C vesicles, the
wild-type enzyme bound to them and eluted in pools A, B, and C. As with
the AI/PC/C vesicles, the mutant protein did not associate with the
PC/C vesicles and was mostly present in pool C.
Endogenous HDL is retained by the column and elutes in
pools B and C. It is apparent that when the expression medium is
subjected to size-exclusion chromatography without the
addition of synthetic vesicles, both the mutant and wild-type proteins
are present in pools B and C. However, wild-type PON1 distributes
in both pools B and C, whereas the mutant PON1 predominates in pool C,
with proteins unassociated with lipoproteins.
|
Compared with no-vesicle incubations (ie, medium alone), the majority
of wild-type PON1 protein shifts to follow the same fractionation
pattern as the AI/PC/C vesicles, whereas the
N-terminal–mutant fractionation pattern shows no pronounced
shift of PON1 to the AI/PC/C vesicles. The shifting of the wild-type
enzyme to larger size fractions relative to the medium alone on
incubation with AI/PC/C and PC/C vesicles suggested either that the
wild type could bind to both apoA-I and phospholipid or that the
binding in each case was due to association with phospholipid. The
reduced mutant enzymatic activity in unconcentrated preparations made
it impossible to reliably follow the A20A21 activity across the
fractions to determine whether the lack of binding was due to the loss
of the N-terminus or simply to denaturation. To address this
problem, wild-type and mutant PON1 expression media were concentrated
to allow the loading of more activity on the column after incubation
with AI/PC/C vesicles in the presence of increased
CaCl2 (to enhance stability of the mutant
enzyme). Analysis of gel-filtration fractions for activity and
of PON1 immunoreactivity (Figure 5)
demonstrated that both the mutant and wild-type proteins were
correlated with their respective activity profiles. Thus, although the
mutant protein was active, it could bind to neither PC/C nor AI/PC/C
vesicles. These findings strongly suggested that binding of the
wild-type enzyme to lipoproteins, AI/PC/C vesicles, and PC/C vesicles
was mediated through the retained N-terminal hydrophobic
signal sequence. The observation that the presence or absence of apoA-I
had no impact on A20A21 PON1 binding to vesicles also indicated that
there were not two distinct sites on PON1 for binding of phospholipid
and apoA-I, respectively. Furthermore, the exclusion limit of the
column used was far larger than that anticipated if PON1 were simply
extracting phospholipid from the vesicles. Clearly, PON1 was
associating directly with the vesicles. This result suggested that PON1
had transferred between the phospholipid surfaces of the lipoproteins
and vesicles. It seemed that the vesicle binding was due to direct
phospholipid binding to the N-terminal peptide; however, the
presence of bovine lipoproteins in all expression media could have
facilitated the transfer of PON1 to the vesicles through the cotransfer
of apoA-I.
|
Wild-Type PON1 Does Not Require ApoA-I to Bind to
Phospholipid
To clarify the nature of PON1 binding to the synthetic vesicles,
we tested whether purified PON1 from human sera could bind directly to
PC/C vesicles in the presence and absence of apoA-I. The methods used
to purify human PON1 remove virtually all contaminating apoA-I. Thus,
if apoA-I were required for PON1 to associate with phospholipid, only a
very low percentage of purified PON1 would be able to bind as a result
of apoA-I contamination. We incubated purified PON1 with PC/C vesicles
in the presence and absence of a 30-fold molar excess of apoA-I and
separated the vesicles by density ultracentrifugation
at d=1.21 g/mL. Figure 6A
shows that PON1 activity predominantly bound to vesicles in the
d<1.21 g/mL fraction. The addition of apoA-I to the
vesicles did not increase vesicle PON1 binding. Thus, we concluded that
PON1 is capable of binding to phospholipids directly, without the
addition of apolipoproteins.
|
Wild-Type PON1 Does Not Bind Directly to ApoA-I in the Absence
of Phospholipid
The above results do not preclude the possibility of a direct
association between PON1 and apoA-I under some circumstances.
Therefore, we determined whether PON1 could bind directly to apoA-I in
the absence of lipid. Purified, delipidated apoA-I and detergent-free
PON1 were incubated under identical conditions as presented in
Figure 6A
but without the PC/C vesicles. Nondenaturing PAGE and
immunoblotting were performed to assess whether PON1
and apoA-I had associated, as indicated by a change in the migration of
these proteins in combination, compared with the proteins alone. No
alteration in the migration patterns of the 2 proteins was observed
when incubated together (Figure 6B). Therefore, we were unable
to obtain any evidence for a direct association between the 2
proteins.
Phospholipid Competes With Nonionic Detergent for Binding to the
N-Terminus to Stimulate PON1 Activity
The addition of nonionic detergents directly to serum or during
PON1 purification decreases PON1 arylesterase and paraoxonase
activities by 50%.22 23 41 Phospholipid addition
restores these activities.22 We speculated that detergent
induced the loss of activity resulting from phospholipid removal from
PON1's N-terminus. To test this hypothesis, A20A21 mutant
and wild-type PON1s were treated with Triton X-100 alone, Triton X-100
in combination with PC/C vesicles, and with PC/C vesicles alone. Mutant
PON1 was unaffected by all 3 treatments (Figure 7). In contrast, the wild type lost
nearly half of its activity after treatment with detergent, whereas
treatment with detergent plus PC/C vesicles protected this activity.
Wild-type PON1 activity was also stimulated by PC/C vesicles alone
(17% increase). The degree of stimulation was less than that reported
by Kuo and La Du,22 but this was expected, because
the wild-type enzyme was already partially stimulated by the
phospholipids present in FBS. These results suggest that PON1's
N-terminal binding to phospholipid stimulates the enzyme's
activity.
|
ApoA-I Stabilizes PON1 Activity In Vitro
We observed that the presence of phospholipid enhanced wild-type
PON1 activity during size-exclusion chromatography
compared with medium alone (Figure 8). We
also observed that apoA-I further augmented this effect (Figure 8).
To determine whether this effect was unique to apoA-I or a
generalized feature of apolipoproteins, wild-type recombinant PON1 was
incubated with AI/PC/C, AII/PC/C, E/PC/C, or PC/C control vesicles.
After size-exclusion chromatography, PON1 activity was
determined in pooled fractions (Figure 9). Minor differences in the shapes of
the activity profiles for the individual vesicle preparations is
likely, secondary to variations in the size distribution within each
vesicle type. However, when the total activity between treatments was
compared through summation of activities across all fractions, it was
observed that the PC/C, AII/PC/C, and E/PC/C vesicles had only 77%,
74%, and 82% of AI/PC/C's activity, respectively. These data were
obtained at increased CaCl2 concentration
compared with that shown in Figure 8, in which activities were
54% and 13% of the apoA-I–treated enzyme in the presence of PC/C
vesicles and medium alone. Thus, the stabilizing effects of
phospholipid, apoA-I, and CaCl2 appear to be
additive. Nonetheless, the presence of apoA-I in vesicle preparations
consistently resulted in higher activity than that seen with
other treatments.
|
|
PON1 Is Expressed at Reduced Levels in HDL in the Absence of
ApoA-I, and It Is Less Stable
Although apoA-I did not appear to be essential for binding of PON1
to phospholipid, the close in vivo association between the two proteins
combined with the in vitro evidence for PON1's stabilization by apoA-I
suggested that apoA-I may play a similar role in the in vivo expression
of PON1. To determine whether apoA-I was required for PON1 expression
in vivo, we measured the arylesterase activity in apoA-I–deficient
mice.37 Sera obtained from age-matched, male and female,
control and apoA-I–deficient C57BL/6 mice were assayed for
arylesterase activity. Sex-dependent dimorphism in arylesterase
activity was observed in both control and apoA-I–knockout mice, with
females always expressing higher levels than males (Table 2). In control animals, the degree of
difference between the sexes diminished as the mice aged, mainly due to
a decrease of enzyme activity in the females. In apoA-I–deficient
mice, arylesterase levels, on the other hand, remained stable.
Age-related fluctuations in mouse serum PON1 was also noted by Li et
al,28 who observed a 3-fold variation in serum
organophosphatase activities between 1-day-old animals and adults, with
a peak activity occurring at 20 days. The mice used in our study were
obtained when they were 60 days old, and future studies will need to
focus on whether the decrease in arylesterase activity in control
females is age related or due to environmental changes.
|
To determine whether reduced arylesterase levels in the
apoA-I–knockout mice resulted from an inability of PON to bind to
apoA-I–deficient HDL, sera from control and apoA-I–deficient animals
were pooled and subjected to isopyknic density
ultracentrifugation, and an analysis of
arylesterase activity and PON1 distribution in the density fractions
was performed. Figure 10A shows
arylesterase activity in the isopyknic density gradient fractions of
pooled sera from the apoA-I–deficient (3 females and 3 males) and
control (3 females and 3 males) mice. Total arylesterase activity was
reduced and was present in a single, very high density peak
(d=1.18 to 1.22 g/mL) in the apoA-I–deficient mice. In
control mice, 2 peaks of arylesterase activity were observed, one at
d=1.07 to 1.15 g/mL and the other at d=1.18 to
1.22 g/mL. This distribution is consistent with observations in
humans, in whom PON1 associates with both HDL2
and HDL3 subtypes.42 The
apoA-I–deficient serum, which lacked the larger peak but retained the
smaller, denser peak, suggests that most PON1 can still associate with
HDL, independent of apoA-I. Mice lacking apoA-I show an increase in
expression of apoA-II and apoE.43 44 It is likely that the
smaller peak of activity associated with the VHDL fractions in the
apoA-I–deficient mouse represents an HDL species containing 1
or more of these apolipoproteins. These data suggest that PON1 is
secreted less efficiently or is less stable in the absence of
apoA-I.
|
PON1's presence in apoA-I–deficient HDL and its ability to bind to
phospholipid directly suggests that PON1 binds to HDL through
phospholipids, like those present in cell membranes, and could be
competitively removed from HDL by phospholipids. To test this
hypothesis, we incubated pooled sera from control mice and from
apoA-I–deficient mice with PC/C vesicles, followed by fractionation
through isopyknic ultracentrifugation. As shown in
Figure 10B, virtually all arylesterase activity in both control
and apoA-I–knockout mice was transferred to the d<1.06
g/mL fraction containing the vesicles (ie, top fraction). Again, we
questioned whether this was due to a transfer of PON1 and
apolipoproteins in tandem. The apoA-I antiserum used for these studies
had no immunoreactivity in the deficient animals, as verified by
Western blotting of delipidated serum (Figure 10C). If the
transfer of PON1 to lipid were mediated through apolipoproteins, it
would likely be a generalized feature and not unique to apoA-I. To
directly assess whether apoA-I cotransferred with PON1 to phospholipid
vesicles, desalted and delipidated aliquots from fractionated control
serum incubated without PC/C vesicles (Figure 10A, solid
circles) and with PC/C vesicles (Figure 10B, solid circles)
were immunostained for apoA-I (Figure 10D). Figure 10D (bottom panel) shows that fractions 1 and 2, which
contained virtually all PON1 activity added (Figure 10B, solid
circles), lacked apoA-I. These data demonstrate that PON1 transferred
independently to the phospholipid vesicles. The observed shifting of
apoA-I to less-dense HDL fractions on incubation with PC/C vesicles is
likely due to phospholipid enrichment of endogenous HDL by
serum enzymes, such as phospholipid transfer protein. Thus, we
demonstrated that PON1's association with HDL was mediated through
phospholipid binding to the retained N-terminal hydrophobic
peptide and that PON1 can transfer from HDL to other phospholipid
surfaces through this retained peptide.
We observed that arylesterase activity in apoA-I–deficient sera
decreased more rapidly than that in sera from control animals after
storage at 4°C. This finding suggested that PON1 may be less stable
in apoA-I–deficient HDL. To directly test this observation, diluted,
pooled sera from control and apoA-I–deficient animals were incubated
separately at 37°C overnight. Control serum retained nearly 90% of
its original activity while the activity of the knockout mice serum had
fallen to 65% of pretreatment values (Figure 11). These data suggest that PON1 is
less stable in the absence of apoA-I and provide further evidence that
apoA-I enhances PON1's activity and stability in vivo through a
preferred presentation on HDL phospholipid.
|
The PON1 N-Terminal Hydrophobic Signal Sequence Is
Not Essential for Protection Against LDL Oxidation
A recent report has suggested that phospholipids themselves may
protect LDL from oxidation.45 We questioned whether the
observed protection of LDL by PON1 was due to the phospholipids bound
to the N-terminus of PON1. Figure 12 presents data from the copper
ion–induced oxidation of purified LDL in the presence of control
expression medium, PON1 (type Q) recombinant medium, and medium
containing the recombinant N-terminal mutant. It is evident
that equal amounts of mutant and wild-type activity result in equal
protection against LDL oxidation, as measured by peroxide
production and TBARS formation. In this assay, wild-type and
N-terminal–deleted PON1 enzymes were added by activity
rather than mass. This required that the medium for the wild-type PON1
be diluted to 0.02 arylesterase units per mL with an
appropriate–mutant PON1 reduced copper ion–induced LDL oxidation by
35% in both the peroxides and TBARS assays (Figure 12).
Nonetheless, the decreased Vmax and increased
Km of the N-terminal mutant
(Table 1) with phenyl acetate as the substrate suggested that
its affinity for other lipophilic substrates and rate of hydrolysis
might also be reduced. This concept is consistent with the fact
that more mutant protein was present in the LDL oxidation assays
for the preparation to have equal activity. To determine whether this
activity represented the same amount of an equally
protective mutant protein (diluted by additional inactive, unstable
mutant protein) compared with wild-type protein or more of an active
mutant protein with reduced affinity will require the development of
techniques to separate or distinguish active PON1 protein from inactive
enzyme in the same preparation. Even so, the fact that the
N-terminal mutant inhibited the oxidative damage of LDL
despite its inability to bind phospholipids demonstrates that PON1's
protective actions are not simply through presentation of
phospholipids and provides further evidence for a catalytic
mechanism.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
PON1 is distinctive in that it interacts with serum lipoproteins. This association, in combination with studies showing that PON1 can protect LDL from oxidative damage in artificial systems, suggests that PON1 has a protective role in vivo and inhibits the development of vascular and coronary diseases. A positive correlation between apoA-I and PON1 levels in vivo, combined with its absent or low activities observed in apoA-I deficiencies such as Tangier disease and fish-eye disease,46 47 suggested that apoA-I was required for the expression of PON1 as well as its binding to HDL. Nonetheless, before this current investigation, no data were available about the role of apoA-I in PON1 expression, the means of PON1's association with HDL, or the structural requirements for this association.
We have now demonstrated that PON1's association with HDL requires the retention of its hydrophobic N-terminal signal peptide. Through site-directed mutagenesis, a cleavable signal peptide was created. Using egg PC/C vesicles and separation techniques based on density or size, we have demonstrated that PON1 associates with lipoproteins and lipid-containing particles by binding phospholipids directly to its hydrophobic signal peptide. Furthermore, the significance of the coexistence of PON1 and apoA-I in HDL and the apparent protein-protein interaction between the two proteins were also investigated by a variety of methods, including characterization of PON1 activity and distribution in the sera of apoA-I–deficient mice and physical studies of purified human PON1 and human apoA-I. In both wild-type and apoA-I–deficient mouse sera, PON1 activity was associated with HDL. When PC/C vesicles were incubated with either wild-type or apoA-I–deficient mouse serum, essentially all PON1 activity shifted to the vesicles independently of apoA-I, suggesting that PON1 binding to HDL is primarily due to the binding of PON1's N-terminus to phospholipid, rather than to apoA-I. Consistent with these findings was our inability to induce a physical association between PON1 and apoA-I in a lipid-free, in vitro system. Hence, it appears that lipid is required for the association of apoA-I and PON1. Although apoA-I was not required for PON1 expression, phospholipid binding, or HDL association, the presence of apoA-I conferred increased PON1 heat stability in mouse serum and consistently resulted in higher arylesterase activity in in vitro studies, a feature not seen with other apolipoproteins tested. These data suggest that apoA-I provides an optimal presentation of phospholipid for PON1 binding, stabilization, and stimulation. Hence, distinct HDL subtypes containing apoA-I may compete for PON1 by presenting a preferred configuration of phospholipids for binding to the hydrophobic N-terminus rather than by binding to apolipoproteins. Serum PON1 levels are altered by diet,29 acute-phase reactants (as is apoA-I),48 pregnancy,49 and disorders affecting apoA-I metabolism.46 47 These findings, our observations of fluctuations in arylesterase activities unique to female mice expressing apoA-I, and the observed stabilizing actions of apoA-I on PONI activity suggest that hormonal and other circulating factors may alter PON1's biological activities, in part through their effects on the expression and metabolism of apoA-I.
Recombinant and serum-purified paraoxonase preparations likely contain both catalytically active and inactive PON1 protein. At present, we are unable to distinguish active from inactive PON1 protein in these preparations. Consequently, it is not possible to rigorously explain the observed lower apparent Vmax (activity per mg protein) of the N-terminal–mutant preparations. This preparation may contain significant amounts of inactive but still immunoreactive protein that confounds our efforts to quantify the amount of active enzyme present. When purified PON1, recombinant PON1, or mouse serum was added to phospholipid vesicles, activity was protected under several experimental conditions. This finding provides strong evidence that PON1 is stabilized or stimulated in the presence of phospholipid. Presumably, the decrease in apparent Vmax of the mutant enzyme is due to its inability to bind phospholipid. Thus, the mutant enzyme may be unable to access substrates efficiently, which normally partition into phospholipids. The observed increase in Km for phenyl acetate hydrolysis by the mutant enzyme indicates that interaction of the wild-type enzyme with its substrates involves the retained hydrophobic N-terminus. All known substrates of PON1 are hydrophobic and partition into a lipid phase, and therefore the protein's association with HDL or phospholipid may be important for its enzymatic function. The lipophilicity of the N-terminus or of its bound lipid may facilitate the binding of substrate and increase the concentration in the immediate environment of the enzyme. Similarly, these qualities may influence the transfer of substrate to the active site or its release.
Surprisingly, both mutant and native recombinant PON1s were effective in blocking copper ion–induced oxidation of LDL. These results were obtained on the basis of the addition of equal units of activity in the assay, which therefore required more mutant protein to be present to achieve activity levels equal to that of the wild-type PON1. As discussed above, it is not possible to determine how much of the mutant protein was active. The two recombinant enzymes appear to be equally effective in protecting LDL against oxidation, thus suggesting that the kinetic parameters for phenyl acetate hydrolysis are not representative of those for the products of LDL oxidation. In contrast, it may be argued that more mutant protein was required for equal protection to overcome the reduced affinity for substrates, as predicted by the increased Km for phenyl acetate. Nonetheless, it is clear that phospholipid binding through the N-terminus is not essential for protection of LDL from oxidation. It has recently been shown that the active site and cofactor requirements for LDL protection by PON1 differ from those for arylesterase activity,6 and it is possible that PON1 activity with the products of LDL oxidation is not significantly altered by loss of the N-terminus. Alternatively, PON1 may also act stoichiometrically, perhaps through reactions involving its free sulfhydryl group (C283). This site may allow PON1 to act as a "biological buffer" to absorb damage induced during oxidation. It may also be in or near the catalytic site that provides enzymatic protection of LDL from oxidation.6 It is important to note that there is no evidence for the direct binding of PON1 to LDL. PON1's function in reducing LDL oxidation may require lipid exchange or absorption of oxidative damage products rather than direct binding of PON1 to LDL.
There is little evidence for the presence of PON1's circulating in plasma as a free protein in vivo. The current study demonstrates that PON1 binds phospholipids with high affinity. The finding that phospholipids competitively remove PON1 from HDL suggests that PON1 may exchange with other phospholipid depots, such as cellular membranes. Although we did not directly measure nor explore this further in the current study, on the basis of our findings, we postulate that PON1 may migrate between HDL and cell membranes. Apolipoproteins show preferential binding for specific phospholipid types,50 as do other HDL proteins such as lecithin-cholesterol acyltransferase.51 PON1 not only binds to phospholipids but is also stimulated by them; however, phospholipid subtypes vary in their degree of stimulation.22 Because of an optimal presentation of commonly preferred phospholipids, apoA-I may serve as a shuttle for PON1 distribution from the liver to distal sites of lipid damage. Therefore, the association of PON1 with HDL may be required for the appropriate physiological distribution of the enzyme for it to have a significant biological impact.
The retention of an N-terminal hydrophobic signal peptide in
a secreted protein is an extremely unusual feature. We are aware of
only 1 other such protein with this rare characteristic, the human
haptoglobin-related protein. Interestingly, the haptoglobin-related
protein associates with PON1 and apoA-I in an anti-trypanosomal HDL
subset that exhibits peroxidase activity and is referred to as
trypanosomal lytic factor.52 Trypanosomal lytic
factor protects humans from infection with Trypanosoma
brucei, possibly through oxidative damage resulting from its
peroxidase activity. Furthermore, HDL is also known to
bind53 and to provide in vivo protection against
endotoxin.54 55 These observations, combined with the
strong evidence for PON1's antiatherogenic activities and its
localization to sites of oxidative damage, including the interstitium,
vascular smooth muscle, and atherosclerotic
plaques,2 56 57 suggest that retention of the
N-terminal hydrophobic leader sequence may have evolved to
facilitate the delivery and transfer of lipid-binding proteins to
inactivate toxic lipid products resulting from
oxidative damage, infection, and inflammation. Such a model is
presented in Figure 13. We
suggest that PON1 leaves the hepatocyte plasma membrane by
binding to apoA-I–associated phospholipids under nonequilibrium
conditions to circulate with HDL. Under conditions more closely
approximating a steady state, the ability of phospholipids to
competitively remove PON1 from HDL would allow the enzyme to diffuse
from the relatively small, phospholipid-binding area of HDL to the
extensive phospholipid depots present at sites of lipid damage and
LDL accumulation, such as endothelial cells, the
interstitium, and smooth muscle cell membranes.
|
|
Acknowledgments |
|---|
The authors are grateful to Dr Nobuyo Maeda and associates (University of North Carolina, Chapel Hill) for providing apoA-I–deficient mice for distribution to the scientific community. The authors are also grateful to Dr David Baltimore (Rockefeller University, New York, NY) for providing human embryonic kidney 293T/17 cells and to Dr Conrad Blum (Columbia University, New York, NY) for providing human apolipoproteins. The authors also thank Dr Trudy Forte (University of California, Berkeley) for helpful comments regarding this manuscript. We also thank Arnold D. Essenburg and Jeffrey Kreick, both of Parke-Davis Pharmaceuticals (Ann Arbor, Mich), for their technical assistance with portions of these studies.
Received December 10, 1998; accepted February 18, 1999.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. La Du BN. Human serum paraoxonase/arylesterase. In: Kalow W, ed. Pharmacogenetics of Drug Metabolism. New York, NY: Pergamon Press;1992:51–91.
2. Watson AD, Berliner JA, Hama SY, La Du BN, Faull KF, Fogelman AM, Navab M. Protective effect of high density lipoprotein associated paraoxonase: inhibition of the biological activity of minimally oxidized low density lipoprotein. J Clin Invest. 1995;96:2882–2891.
3. Mackness MI, Arrol S, Abbott C Durrington PN. Protection of low-density lipoprotein against oxidative modification by high-density lipoprotein associated paraoxonase. Atherosclerosis. 1993;104:129–135.[Medline] [Order article via Infotrieve]
4. Mackness MI, Arrol S, Durrington PN. Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein. FEBS Lett. 1991;286:152–154.[Medline] [Order article via Infotrieve]
5. Aviram M, Rosenblat M, Bisgaier CL, Newton RS, Primo-Parmo SL, La Du BN. Paraoxonase inhibits high-density lipoprotein oxidation and preserves its functions: a possible peroxidative role for paraoxonase. J Clin Invest. 1998;101:1581–1590.[Medline] [Order article via Infotrieve]
6.
Aviram M, Billecke S, Sorenson R, Bisgaier C, Newton
R, Rosenblat M, Erogul J, Hsu C, Dunlop C, La Du B. Paraoxonase active
site required for protection against LDL oxidation involves its free
sulfhydryl group and is different from that required for its
arylesterase/paraoxonase activities: selective action of human
paraoxonase allozymes Q and R. Arterioscler Thromb Vasc
Biol. 1998;18:1617–1624.
7. Mackness MI, Mackness B, Durrington PN, Connelly PW, Hegele RA. Paraoxonase: biochemistry, genetics and relationship to plasma lipoproteins. Curr Opin Lipidol. 1996;7:69–76.[Medline] [Order article via Infotrieve]
8.
McElveen J, Mackness MI, Colley CM, Peard T, Warner S,
Walker CH. Distribution of paraoxon hydrolytic activity in the serum of
patients after myocardial infarction. Clin Chem. 1986;32:671–673.
9.
Hegele RA, Brunt JH, Connelly PW. A polymorphism
of the paraoxonase gene associated with variation in plasma
lipoproteins in a genetic isolate. Arterioscler Thromb Vasc
Biol. 1995;15:89–95.
10. Ruiz J. Diabetes mellitus and the late complications: influence of the genetic factors. Diabetes Metab. 1997;23(suppl 2):57–63.
11.
Sanghera DK, Saha N, Aston CE, Kamboh MI. Genetic
polymorphism of paraoxonase and the risk of coronary heart
disease. Arterioscler Thromb Vasc Biol. 1997;17:1067–1073.
12. Serrato M, Marian AJ. A variant of human paraoxonase/arylesterase (HUMPONA) gene is a risk factor for coronary artery disease. J Clin Invest. 1995;96:3005–3008.
13. Sanghera DK, Aston CE, Saha N, Kamboh MI. DNA polymorphisms in two paraoxonase genes (PON1 and PON2) are associated with the risk of coronary heart disease. Am J Hum Genet. 1998;62:36–44.[Medline] [Order article via Infotrieve]
14. Heinecke JW, Lusis AJ. Paraoxonase-gene polymorphisms associated with coronary heart disease: support for the oxidative damage hypothesis? Am J Hum Genet. 1998;62:20–24.[Medline] [Order article via Infotrieve]
15.
Zama T, Murata M, Matsubara Y, Kawano K, Aoki N,
Yoshino H, Watanabe G, Ishikawa K, Ikeda Y. A 192Arg variant of the
human paraoxonase (HUMPONA) gene polymorphism is associated with an
increased risk for coronary artery disease in the Japanese.
Arterioscler Thromb Vasc Biol. 1997;17:3565–3569.
16. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density lipoprotein as a protective factor against coronary heart disease: the Framingham Study. Am J Med. 1977;62:707–714.[Medline] [Order article via Infotrieve]
17.
Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD,
Castelli WP, Knoke JD, Jacobs DR Jr, Bangdiwala S, Tyroler HA.
High-density lipoprotein cholesterol and
cardiovascular disease: four prospective American
studies. Circulation. 1989;79:8–15.
18. Shih DM, Gu L, Xia YR, Navab M, Li WF, Hama S, Castellani LW, Furlong CE, Costa LG, Fogelman AM, Lusis AJ. Mice lacking serum paraoxonase are susceptible to organophosphate toxicity and atherosclerosis. Nature. 1998;394:284–287.[Medline] [Order article via Infotrieve]
19. Kelso GJ, Stuart WD, Richter RJ, Furlong CE, Jordan-Starck TC, Harmony JA. Apolipoprotein J is associated with paraoxonase in human plasma. Biochemistry. 1994;33:832–839.[Medline] [Order article via Infotrieve]
20. Blatter MC, James RW, Messmer S, Barja F, Pometta D. Identification of a distinct human high-density lipoprotein subspecies defined by a lipoprotein-associated protein, K-45: identity of K-45 with paraoxonase. Eur J Biochem. 1993;211:871–879.[Medline] [Order article via Infotrieve]
21. Tanaka-Kawai H, Yomoda S. Molecular weight and substrate characteristics of human serum arylesterase following purification by immuno-affinity chromatography. Clin Chim Acta. 1993;215:127–138.[Medline] [Order article via Infotrieve]
22. Kuo CL, La Du BN. Comparison of purified human and rabbit serum paraoxonases. Drug Metab Dispos. 1995;23:935–944.[Abstract]
23. Gan KN, Smolen A, Eckerson HW, La Du BN. Purification of human serum paraoxonase/arylesterase: evidence for one esterase catalyzing both activities. Drug Metab Dispos. 1991;19:100–106.[Abstract]
24. Adkins S, Gan KN, Mody M, La Du BN. Molecular basis for the polymorphic forms of human serum paraoxonase/arylesterase: glutamine or arginine at position 191, for the respective A or B allozymes. Am J Hum Genet. 1993;52:598–608.[Medline] [Order article via Infotrieve]
25. Furlong CE, Richter RJ, Chapline C, Crabb JW. Purification of rabbit and human serum paraoxonase. Biochemistry. 1991;30:10133–10140.[Medline] [Order article via Infotrieve]
26. Hassett C, Richter RJ, Humbert R, Chapline C, Crabb JW, Omiecinski CJ, Furlong CE. Characterization of cDNA clones encoding rabbit and human serum paraoxonase: the mature protein retains its signal sequence. Biochemistry. 1991;30:10141–10149.[Medline] [Order article via Infotrieve]
27.
Sorenson RC, Primo-Parmo SL, Kuo CL, Adkins S,
Lockridge O, La Du BN. Reconsideration of the catalytic center and
mechanism of mammalian paraoxonase/arylesterase. Proc Natl Acad
Sci U S A. 1995;92:7187–7191.
28. Li WF, Matthews C, Disteche CM, Costa LG, Furlong CE. Paraoxonase (PON1) gene in mice: sequencing, chromosomal localization and developmental expression. Pharmacogenetics. 1997;7:137–144.[Medline] [Order article via Infotrieve]
29. Shih DM, Gu L, Hama S, Xia YR, Navab M, Fogelman AM, Lusis AJ. Genetic-dietary regulation of serum paraoxonase expression and its role in atherogenesis in a mouse model. J Clin Invest. 1996;97:1630–1639.[Medline] [Order article via Infotrieve]
30.
von Heijne G. A new method for predicting signal
sequence cleavage sites. Nucleic Acids Res. 1986;14:4683–4690.
31. von Heijne G. Signals for protein targeting into and across membranes. Subcell Biochem. 1994;22:1–19.[Medline] [Order article via Infotrieve]
32. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K. Introduction of DNA Into Mammalian Cells. New York, NY: John Wiley & Sons; 1993. Ch 9.
33. Eckerson HW, Romson J, Wyte C, La Du BN. The human serum paraoxonase polymorphism: identification of phenotypes by their response to salts. Am J Hum Genet. 1983;35:214–227.[Medline] [Order article via Infotrieve]
34. Chen CH, Albers JJ. Characterization of proteoliposomes containing apoprotein A-I: a new substrate for the measurement of lecithin:cholesterol acyltransferase activity. J Lipid Res. 1982;23:680–691.[Abstract]
35. Blum CB, Dell RB, Palmer RH, Ramakrishnan R, Seplowitz AH, Goodman DS. Relationship of the parameters of body cholesterol metabolism with plasma levels of HDL cholesterol and the major HDL apoproteins. J Lipid Res. 1985;26:1079–1088.[Abstract]
36. Blum CB, Aron L, Sciacca R. Radioimmunoassay studies of human apolipoprotein E. J Clin Invest. 1980;66:1240–1250.
37.
Williamson R, Lee D, Hagaman J, Maeda N. Marked
reduction of high density lipoprotein cholesterol in mice
genetically modified to lack apolipoprotein A-I. Proc Natl Acad
Sci U S A. 1992;89:7134–7138.
38. Scanu AM, Edelstein C. Solubility in aqueous solutions of ethanol of the small molecular weight peptides of the serum very low density and high density lipoproteins: relevance to the recovery problem during delipidation of serum lipoproteins. Anal Biochem. 1971;44:576–588.[Medline] [Order article via Infotrieve]
39. Gibson JC, Rubinstein A, Bukberg PR, Brown WV. Apolipoprotein E-enriched lipoprotein subclasses in normolipidemic subjects. J Lipid Res. 1983;24:886–898.[Abstract]
40. Bisgaier CL, Sachdev OP, Megna L, Glickman RM. Distribution of apolipoprotein A-IV in human plasma. J Lipid Res. 1985;26:11–25.[Abstract]
41. Kitchen BJ, Masters CJ, Winzor DJ. Effects of lipid removal on the molecular size and kinetic properties of bovine plasma arylesterase. Biochem J. 1973;135:93–99.[Medline] [Order article via Infotrieve]
42. Zech R, Rockseisen M, Kluge K, Dewald K, Armstrong VW, Chemnitius JM. Lipoproteins and hydrolysis of organophosphorus compounds. Chem Biol Interact. 1993;87:85–94.[Medline] [Order article via Infotrieve]
43.
Li H, Reddick RL, Maeda N. Lack of apoA-I is not
associated with increased susceptibility to
atherosclerosis in mice. Arterioscler
Thromb. 1993;13:1814–1821.
44. Plump AS, Azrolan N, Odaka H, Wu L, Jiang X, Tall A, Eisenberg S, Breslow JL. ApoA-I knockout mice: characterization of HDL metabolism in homozygotes and identification of a post-RNA mechanism of apoA-I up-regulation in heterozygotes. J Lipid Res. 1997;38:1033–1047.[Abstract]
45. Graham A, Hassall DG, Rafique S, Owen JS. Evidence for a paraoxonase-independent inhibition of low-density lipoprotein oxidation by high-density lipoprotein. Atherosclerosis. 1997;135:193–204.[Medline] [Order article via Infotrieve]
46.
Mackness MI, Walker CH, Carlson LA. Low A-esterase
activity in serum of patients with fish-eye disease. Clin
Chem. 1987;33:587–588.
47. Mackness MI, Peuchant E, Dumon MF, Walker CH, Clerc M. Absence of `A'-esterase activity in the serum of a patient with Tangier disease. Clin Biochem. 1989;22:475–478.[Medline] [Order article via Infotrieve]
48. Van Lenten BJ, Hama SY, de Beer FC, Stafforini DM, McIntyre TM, Prescott SM, La Du BN, Fogelman AM, Navab M. Anti-inflammatory HDL becomes pro-inflammatory during the acute phase response: loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures. J Clin Invest. 1995;96:2758–2767.
49. Weitman SD, Vodicnik MJ, Lech JJ. Influence of pregnancy on parathion toxicity and disposition. Toxicol Appl Pharmacol. 1983;71:215–224.[Medline] [Order article via Infotrieve]
50. Forte TM, Bielicki JK, Goth-Goldstein R, Selmek J, McCall MR. Recruitment of cell phospholipids and cholesterol by apolipoproteins A-II and A-I: formation of nascent apolipoprotein-specific HDL that differ in size, phospholipid composition, and reactivity with LCAT. J Lipid Res. 1995;36:148–157.[Abstract]
51. Bonomo EA, Swaney JB. Effect of phosphatidylethanolamine on the properties of phospholipid-apolipoprotein complexes. Biochemistry. 1990;29:5094–5103.[Medline] [Order article via Infotrieve]
52.
Smith AB, Esko JD, Hajduk SL. Killing of trypanosomes
by the human haptoglobin-related protein. Science. 1995;268:284–286.
53. Massamiri T, Tobias PS, Curtiss LK. Structural determinants for the interaction of lipopolysaccharide binding protein with purified high density lipoproteins: role of apolipoprotein A-I. J Lipid Res. 1997;38:516–525.[Abstract]
54. La Du BN, Aviram M, Billecke S, Navab M, Primo-Parmo SL, Sorenson RC, Standiford TJ. On the biological role(s) of paraoxonases. Chem Biol Interact. In press.
55.
Levine DM, Parker TS, Donnelly TM, Walsh A, Rubin AL.
In vivo protection against endotoxin by plasma high density
lipoprotein. Proc Natl Acad Sci U S A. 1993;90:12040–12044.
56.
Mackness B, Hunt R, Durrington PN, Mackness MI.
Increased immunolocalization of paraoxonase, clusterin, and
apolipoprotein A-I in the human artery wall with the progression of
atherosclerosis. Arterioscler Thromb Vasc
Biol. 1997;17:1233–1238.
57. Mackness MI, Mackness B, Arrol S, Wood G, Bhatnagar D, Durrington PN. Presence of paraoxonase in human interstitial fluid. FEBS Lett. 1997;416:377–380.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  J. R. Batuca, P. R. J. Ames, M. Amaral, C. Favas, D. A. Isenberg, and J. Delgado Alves Anti-atherogenic and anti-inflammatory properties of high-density lipoprotein are affected by specific antibodies in systemic lupus erythematosus Rheumatology, January 1, 2009; 48(1): 26 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Christoffersen, J. Ahnstrom, O. Axler, E. I. Christensen, B. Dahlback, and L. B. Nielsen The Signal Peptide Anchors Apolipoprotein M in Plasma Lipoproteins and Prevents Rapid Clearance of Apolipoprotein M from Plasma J. Biol. Chem., July 4, 2008; 283(27): 18765 - 18772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Moren, S. Deakin, M.-L. Liu, M.-R. Taskinen, and R. W. James HDL subfraction distribution of paraoxonase-1 and its relevance to enzyme activity and resistance to oxidative stress J. Lipid Res., June 1, 2008; 49(6): 1246 - 1253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Deakin, X. Moren, and R. W. James HDL Oxidation Compromises its Influence on Paraoxonase-1 Secretion and its Capacity to Modulate Enzyme Activity Arterioscler Thromb Vasc Biol, May 1, 2007; 27(5): 1146 - 1152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Thomas-Moya, M. Gianotti, A. M. Proenza, and I. Llado The age-related paraoxonase 1 response is altered by long-term caloric restriction in male and female rats J. Lipid Res., September 1, 2006; 47(9): 2042 - 2048. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Christoffersen, L. B. Nielsen, O. Axler, A. Andersson, A. H. Johnsen, and B. Dahlback Isolation and characterization of human apolipoprotein M-containing lipoproteins J. Lipid Res., August 1, 2006; 47(8): 1833 - 1843. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rosenblat, L. Gaidukov, O. Khersonsky, J. Vaya, R. Oren, D. S. Tawfik, and M. Aviram The Catalytic Histidine Dyad of High Density Lipoprotein-associated Serum Paraoxonase-1 (PON1) Is Essential for PON1-mediated Inhibition of Low Density Lipoprotein Oxidation and Stimulation of Macrophage Cholesterol Efflux J. Biol. Chem., March 17, 2006; 281(11): 7657 - 7665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. Blatter Garin, X. Moren, and R. W. James Paraoxonase-1 and serum concentrations of HDL-cholesterol and apoA-I J. Lipid Res., March 1, 2006; 47(3): 515 - 520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Ferretti, T Bacchetti, F Principi, F Di Ludovico, B Viti, V A Angeleri, M Danni, and L Provinciali Increased levels of lipid hydroperoxides in plasma of patients with multiple sclerosis: a relationship with paraoxonase activity Multiple Sclerosis, December 1, 2005; 11(6): 677 - 682. [Abstract] [PDF]
|
|
|
|
|
|
D. I. Draganov, J. F. Teiber, A. Speelman, Y. Osawa, R. Sunahara, and B. N. La Du Human paraoxonases (PON1, PON2, and PON3) are lactonases with overlapping and distinct substrate specificities J. Lipid Res., June 1, 2005; 46(6): 1239 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Ferretti, T. Bacchetti, C. Moroni, S. Savino, A. Liuzzi, F. Balzola, and V. Bicchiega Paraoxonase Activity in High-Density Lipoproteins: A Comparison between Healthy and Obese Females J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1728 - 1733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Faber, O. Axler, B. Dahlback, and L. B. Nielsen Characterization of apoM in normal and genetically modified mice J. Lipid Res., July 1, 2004; 45(7): 1272 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Ferretti, T. Bacchetti, D. Busni, R. A. Rabini, and G. Curatola Protective Effect of Paraoxonase Activity in High-Density Lipoproteins against Erythrocyte Membranes Peroxidation: A Comparison between Healthy Subjects and Type 1 Diabetic Patients J. Clin. Endocrinol. Metab., June 1, 2004; 89(6): 2957 - 2962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. W. Connelly, G. F. Maguire, and D. I. Draganov Separation and quantitative recovery of mouse serum arylesterase and carboxylesterase activity J. Lipid Res., March 1, 2004; 45(3): 561 - 566. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Delgado Alves, S. Kumar, and D. A. Isenberg Cross-reactivity between anti-cardiolipin, anti-high-density lipoprotein and anti-apolipoprotein A-I IgG antibodies in patients with systemic lupus erythematosus and primary antiphospholipid syndrome Rheumatology, July 1, 2003; 42(7): 893 - 899. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. G. Cabana, C. A. Reardon, N. Feng, S. Neath, J. Lukens, and G. S. Getz Serum paraoxonase: effect of the apolipoprotein composition of HDL and the acute phase response J. Lipid Res., April 1, 2003; 44(4): 780 - 792. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rosenblat, D. Draganov, C. E. Watson, C. L. Bisgaier, B. N. La Du, and M. Aviram Mouse Macrophage Paraoxonase 2 Activity Is Increased Whereas Cellular Paraoxonase 3 Activity Is Decreased Under Oxidative Stress Arterioscler Thromb Vasc Biol, March 1, 2003; 23(3): 468 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Rozenberg, D. M. Shih, and M. Aviram Human Serum Paraoxonase 1 Decreases Macrophage Cholesterol Biosynthesis: Possible Role for Its Phospholipase-A2-Like Activity and Lysophosphatidylcholine Formation Arterioscler Thromb Vasc Biol, March 1, 2003; 23(3): 461 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Parolini, G. Chiesa, Y. Zhu, T. Forte, S. Caligari, E. Gianazza, M. G. Sacco, C. R. Sirtori, and E. M. Rubin Targeted Replacement of Mouse Apolipoprotein A-I with Human ApoA-I or the Mutant ApoA-IMilano. EVIDENCE OF APOA-IM IMPAIRED HEPATIC SECRETION J. Biol. Chem., February 7, 2003; 278(7): 4740 - 4746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Josse, C. Ebel, D. Stroebel, A. Fontaine, F. Borges, A. Echalier, D. Baud, F. Renault, M. le Maire, E. Chabrieres, et al. Oligomeric States of the Detergent-solubilized Human Serum Paraoxonase (PON1) J. Biol. Chem., August 30, 2002; 277(36): 33386 - 33397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Desideri, M. C. Marinucci, G. Tomassoni, P. G. Masci, A. Santucci, and C. Ferri Vitamin E Supplementation Reduces Plasma Vascular Cell Adhesion Molecule-1 and von Willebrand Factor Levels and Increases Nitric Oxide Concentrations in Hypercholesterolemic Patients J. Clin. Endocrinol. Metab., June 1, 2002; 87(6): 2940 - 2945. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Ng, G. F. Maguire, J. Wylie, A. Ravandi, W. Xuan, Z. Ahmed, M. Eskandarian, A. Kuksis, and P. W. Connelly Oxidative Stress Is Markedly Elevated in Lecithin:Cholesterol Acyltransferase-deficient Mice and Is Paradoxically Reversed in the Apolipoprotein E Knockout Background in Association with a Reduction in Atherosclerosis J. Biol. Chem., March 29, 2002; 277(14): 11715 - 11720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Forte, G. Subbanagounder, J. A. Berliner, P. J. Blanche, A. O. Clermont, Z. Jia, M. N. Oda, R. M. Krauss, and J. K. Bielicki Altered activities of anti-atherogenic enzymes LCAT, paraoxonase, and platelet-activating factor acetylhydrolase in atherosclerosis-susceptible mice J. Lipid Res., March 1, 2002; 43(3): 477 - 485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Deakin, I. Leviev, M. Gomaraschi, L. Calabresi, G. Franceschini, and R. W. James Enzymatically Active Paraoxonase-1 Is Located at the External Membrane of Producing Cells and Released by a High Affinity, Saturable, Desorption Mechanism J. Biol. Chem., February 1, 2002; 277(6): 4301 - 4308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. W. Castellani and A. J. Lusis ApoA-II Versus ApoA-I: Two for One Is Not Always a Good Deal Arterioscler Thromb Vasc Biol, December 1, 2001; 21(12): 1870 - 1872. [Full Text] [PDF]
|
|
|
|
|
|
P. Holvoet, K. Peeters, S. Lund-Katz, A. Mertens, P. Verhamme, R. Quarck, D. Stengel, M. Lox, E. Deridder, H. Bernar, et al. Arg123-Tyr166 Domain of Human ApoA-I Is Critical for HDL-Mediated Inhibition of Macrophage Homing and Early Atherosclerosis in Mice Arterioscler Thromb Vasc Biol, December 1, 2001; 21(12): 1977 - 1983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Brushia, T. M. Forte, M. N. Oda, B. N. La Du, and J. K. Bielicki Baculovirus-mediated expression and purification of human serum paraoxonase 1A J. Lipid Res., June 1, 2001; 42(6): 951 - 958. [Abstract] [Full Text]
|
|
|
|
|
|
M. A. Deeg, E. L. Bierman, and M. C. Cheung GPI-specific phospholipase D associates with an apoA-I- and apoA-IV-containing complex J. Lipid Res., March 1, 2001; 42(3): 442 - 451. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Billecke, D. Draganov, R. Counsell, P. Stetson, C. Watson, C. Hsu, and B. N. L. Du Human Serum Paraoxonase (pon1) Isozymes Q and R Hydrolyze Lactones and Cyclic Carbonate Esters Drug Metab. Dispos., November 1, 2000; 28(11): 1335 - 1342. [Abstract] [Full Text]
|
|
|
|
|
|
Bart De Geest, D. Stengel, M. Landeloos, M. Lox, L. Le Gat, D. Collen, P. Holvoet, and E. Ninio Effect of Overexpression of Human Apo A-I in C57BL/6 and C57BL/6 Apo E-Deficient Mice on 2 Lipoprotein-Associated Enzymes, Platelet-Activating Factor Acetylhydrolase and Paraoxonase : Comparison of Adenovirus-Mediated Human Apo A-I Gene Transfer and Human Apo A-I Transgenesis Arterioscler Thromb Vasc Biol, October 1, 2000; 20 (10): e68 - e75. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Kudchodkar, A. G. Lacko, L. Dory, and T. V. Fungwe Dietary Fat Modulates Serum Paraoxonase 1 Activity in Rats J. Nutr., October 1, 2000; 130(10): 2427 - 2433. [Abstract] [Full Text]
|
|
|
|
|
|
C. C. Hedrick, K. Hassan, G. P. Hough, J. Yoo, S. Simzar, C. R. Quinto, S.-M. Kim, A. Dooley, S. Langi, S. Y. Hama, et al. Short-Term Feeding of Atherogenic Diet to Mice Results in Reduction of HDL and Paraoxonase That May Be Mediated by an Immune Mechanism Arterioscler Thromb Vasc Biol, August 1, 2000; 20(8): 1946 - 1952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. I. Draganov, P. L. Stetson, C. E. Watson, S. S. Billecke, and B. N. La Du Rabbit Serum Paraoxonase 3 (PON3) Is a High Density Lipoprotein-associated Lactonase and Protects Low Density Lipoprotein against Oxidation J. Biol. Chem., October 20, 2000; 275(43): 33435 - 33442. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||