JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M702163200 on June 7, 2007

J. Biol. Chem., Vol. 282, Issue 32, 23698-23707, August 10, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/32/23698    most recent
M702163200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Watanabe, J.
Right arrow Articles by Reddy, S. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Watanabe, J.
Right arrow Articles by Reddy, S. T.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Differential Association of Hemoglobin with Proinflammatory High Density Lipoproteins in Atherogenic/Hyperlipidemic Mice

A NOVEL BIOMARKER OF ATHEROSCLEROSIS*Formula

Junji Watanabe{ddagger}§, Katherine J. Chou, James C. Liao, Yunan Miao||, Hsiang-Hui Meng||, Helen Ge||, Victor Grijalva{ddagger}, Susan Hama{ddagger}, Kathy Kozak**, Georgette Buga{ddagger}, Julian P. Whitelegge{ddagger}{ddagger}§§, Terry D. Lee||, Robin Farias-Eisner**, Mohamad Navab{ddagger}1, Alan M. Fogelman{ddagger}12, and Srinivasa T. Reddy{ddagger}§§§13

From the {ddagger}Atherosclerosis Research Unit, Department of Medicine/Cardiology, §Department of Molecular and Medical Pharmacology, Department of Chemical and Bimolecular Engineering, **Department of Obstetrics and Gynecology, {ddagger}{ddagger}The Semel Institute for Neuroscience and Human Behavior, and §§Molecular Biology Institute, University of California, Los Angeles, California 90095 and ||Division of Immunology, Beckman Research Institute, City of Hope, Duarte, California 91010

Received for publication, March 13, 2007 , and in revised form, May 15, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Studies in both mice and humans suggest that the anti- or proinflammatory nature of high density lipoprotein (HDL) may be a more sensitive predictor of risk for coronary heart disease events. In this study, we report the identification and characterization of two proteins (m/z 14,900 and 15,600) that are most dramatically associated with HDL in mouse models of atherosclerosis. Mass spectral analyses of proinflammatory HDL identified the two peaks to be hemoglobin (Hb) {alpha} and beta chains, respectively, with no apparent post-translational modification. Biochemical analysis confirmed the differential association of Hb with HDL from hyperlipidemic mice. We further show that HDL-associated Hb is predominantly in the oxyHb form with distinct physical and chemical properties. Furthermore oxyHb-containing proinflammatory HDL potently consumed nitric oxide and contracted arterial vessels ex vivo. Moreover Hb also was found differentially associated with HDL from coronary heart disease patients compared with healthy controls. Our data suggest that Hb contributes to the proinflammatory nature of HDL in mouse and human models of atherosclerosis and may serve as a novel biomarker for atherosclerosis.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Atherosclerosis is the leading cause of morbidity and mortality in Western society. The inverse relationship between HDL4 cholesterol and the risk of atherosclerosis is well established. Although HDL cholesterol is an epidemiological predictor of risk for coronary heart disease (CHD) (1), a significant number of CHD events occur in patients with normal LDL and HDL cholesterol levels (1, 2). Thus, there is a need for biomarkers with better predictive value (3-5).

HDL exerts anti-inflammatory functions by promoting reverse cholesterol transport and preventing the oxidation of LDL (6, 7). We have shown previously that the anti-inflammatory functions of HDL can be impaired in humans (3), rabbits (8), and mice (9) during inflammatory processes. This impaired HDL is proinflammatory in nature as characterized by (i) decreased levels and activity of anti-inflammatory, antioxidant factors including apolipoprotein A-I (apoA-I) and paraoxonase 1 (10), (ii) gain of proinflammatory enzymes such as serum amyloid A (8), (iii) increased lipid hydroperoxide content (11), (iv) reduced potential to efflux cholesterol (12), and (v) diminished ability to prevent LDL oxidation (13). The molecular changes and mechanisms that promote anti-inflammatory HDL conversion to proinflammatory HDL are currently unknown. The knowledge of molecular profiles that distinguish proinflammatory HDL from anti-inflammatory HDL will not only allow the development of novel biomarkers for the early detection of atherosclerosis but will also provide new strategies for therapeutic intervention of atherosclerosis.

The ProteinChip array technology coupled with surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) has been utilized to facilitate protein profiling of complex biological mixtures (14-17). In this study, we report the identification and characterization of two SELDI protein peaks, Hb {alpha} and beta chains, most dramatically associated with proinflammatory HDL in atherogenic/hyperlipidemic mice. We demonstrate that the association of Hb with HDL also plays an important role in the modulation of HDL function suggesting that Hb may serve as a novel biomarker as well as a therapeutic target for atherosclerosis.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Animal Experiments
Wild-type, low density lipoprotein receptor-deficient, and apoE null C57BL6/J wild-type female mice at an age of 8-12 weeks were used in experiments comparing the three mouse models. Mice were fed one of three diets for the periods described: chow diet (Ralston Purina Mouse Chow); atherogenic diet containing 15.8% fat, 1.25% cholesterol, and 0.5% cholate (Harlan Teklad, Madison, WI); or a western diet containing 42% fat and 0.15% cholesterol (Harlan Teklad). For short term studies mice were fed the described diet for 7 days (D7), and for long term studies mice were fed the described diet for 15 weeks (W15). Serum samples were isolated from overnight fasted mice, cryopreserved in 10% sucrose, and freshly frozen at -80 °C until use.

Human Samples
Human plasma samples were obtained from healthy donors and donors with stable CHD or CHD equivalents as defined by the National Cholesterol Education Program Adult Treatment Panel III criteria.

Lipoprotein Isolation
Pooled serum samples were fractionated by a gel permeation fast protein liquid chromatography (FPLC) system consisting of dual Superose 6 columns in series (Amersham Biosciences). Serum (0.5 ml) was eluted with sterile PBS at a flow rate of 0.5 ml/min, and fractionated every 1 ml. The first 10 1-ml fractions were discarded, and each subsequent 1-ml fraction was analyzed by cholesterol content (Thermo, Louisville, CO) and BCA protein assays (Piercer, Rockford, IL) according to the manufacturers' protocols. Very low density lipoprotein, LDL, HDL, and post-HDL fractions were pooled for all experiments. For some experiments, HDL from an individual serum sample was freshly isolated with LipiDirect HDL reagent (Polymedco, Cortland Manor, NY) according to the manufacturer's protocol. The supernatant containing HDL was assayed for cholesterol content and BCA protein assays. All lipoproteins were used within 48 h following isolation.

Assays to Determine the Inflammatory Properties of HDL
HDL properties were determined as described previously by measuring reactive oxygen species content with 2,7,7'dichlorofluorescein diacetate (H2DCFDA; Invitrogen) (18), paraoxonase activity (8), cellular cholesterol efflux (19), and monocyte chemotactic activity assay (3, 7).

SELDI ProteinChip Analysis
Sample Preparation—Lipoproteins or serum samples were processed on normal phase (NP20), strong anion exchange (Q10), and weak cation exchange (CM10) ProteinChip arrays according to the manufacturer's protocols (Ciphergen Biosystems, Fremont, CA). NP20, Q10, and CM10 arrays were equilibrated with binding buffer (PBS with 0.1% Triton X-100, pH 7.0) prior to use. Diluted samples with binding buffer (serum/HDL supernatant at 1:25, lipoprotein fractions at 1:2) were freshly prepared prior to use.

SELDI ProteinChip Data—The data were analyzed with ProteinChip data analysis software version 3.2 (Ciphergen Biosystems) as described previously (20). Sample statistics were performed on groups of profiles (anti-inflammatory HDL versus proinflammatory HDL and normal serum versus atherogenic serum). Protein differences (-fold changes of relative intensity) were calculated among the various groups. A protein was considered differentially associated between two groups if, when compared with the one group, statistically significant differences in its intensity were observed (p < 0.05).

Identification of Proteins Representing Specific m/z Peaks—SELDI protein peaks were identified as described previously (20, 21). Briefly HDL samples were separated by SDS-PAGE followed by silver staining. The bands with molecular weights corresponding to the peaks of interest were excised, and in-gel tryptic digestion was performed as described previously (22). Extracted peptides were subjected to microcapillary liquid chromatography/tandem mass spectrometry (LC/MS/MS) as described previously (22). The data were used to search mouse databases using Sonar MS/MSTM (Genomic Solutions) and TurboSEQUESTTM (Thermo Fisher Scientific, Waltham, MA).

Electrophoresis and Immunoblots
Isoelectric focusing (IEF), Tris/HCl gels, and all other reagents for electrophoresis were purchased from Bio-Rad. Serum (2 µl) or FPLC fractions (50 µl) were separated by SDS-PAGE, IEF (pH 3-10), Tris/HCl native (4-15%), or IEF Tris/HCl two-dimensional gels. For two-dimensional gels, each lane from IEF gels was cut and inserted into native gels. Samples were loaded on gels and transferred to nitrocellulose membrane (Amersham Biosciences). The membrane was immunoblotted against Hb at 1:1000 (MP Biomedicals, Irvine, CA) or apoA-I at 1:10,000 (Biodesign, Saco, ME). HRP-conjugated secondary antibody (Amersham Biosciences) was used at 1:10,000, and the bands were visualized with ECL detection reagent (Amersham Biosciences).

Analysis of Hb Association with ApoA-I-containing HDL
ELISA—Individual serum samples from mice fed a normal chow diet (n = 16) or an atherogenic diet (n = 16) for 7 days, apoE null female mice on a normal chow diet (n = 8), and human plasma samples from healthy donors and patients (n = 10 per group) were assayed for Hb in serum and apoA-I-containing HDL by direct and sandwich ELISA, respectively, according to the manufacturer's protocols (Abcam, Cambridge, MA).

Hb in Serum—Serum Hb was quantified by direct ELISA according to the manufacturer's protocol (Abcam). Primary antibodies against mouse Hb at 1:1000 (MP Biomedicals) or human Hb (Abcam) at 1:5000 followed by HRP-conjugated detection antibody at 1:5000 (Amersham Biosciences) were used. HRP was probed with 3,3',5,5'-tetramethylbenzidine solution (KPL, Inc., Gaithersburg, MD), and A450 was measured. HRP-conjugated detection antibody was used as a standard to convert the A450 of each sample to the concentration of detection antibody. The change of each protein was determined by comparing each concentration with the average of Hb in mice on chow diet, which was set to 100%.

Association of Hb with ApoA-I-containing HDL—The association of Hb with apoA-I-containing HDL was determined by sandwich ELISA according to the manufacturer's protocol (Abcam). Briefly 96-well polyvinyl chloride microfilter plates (BD Biosciences) were precoated with 1-5 µg/ml goat anti-mouse apoA-I (Biodesign) or rabbit anti-human apoA-I (Abcam) at 4 °C overnight. Rabbit anti-mouse Hb at 1:1000 (MP Biomedicals) and HRP-conjugated donkey anti-rabbit IgG detection antibody at 1:5000 (GE Healthcare) were used to detect mouse Hb. HRP-conjugated goat anti-human Hb (Abcam) was used for human Hb. HRP was probed with TMB solution (KPL, Inc.), and A450 was measured. HRP-conjugated detection antibody was used as a standard to convert the A450 of each sample to the concentration of the detection antibody. The change of each protein was determined by comparing each concentration with the average of Hb in mice on chow diet, which was set to 100%.

Spectrophotometric Determination of Hb—The concentrations of oxyHb and metHb were deconvoluted by fitting a set of pure species "basis spectra" to the measured spectra by means of linear regression as described previously (23). The deconvolution method was verified when the consumption of oxyHb to the generation of metHb exhibited a 1:1 ratio while the total hemoglobin was conserved in all samples.

Nitric Oxide (NO) Consumption Assay
NO consumption by HDL was tested with proliNONOate as described previously (24, 25). Briefly proliNONOate (Cayman Chemical), a pH-dependent NO timed-release donor, was added to a reaction chamber that contained 4 ml of PBS (pH 7.4) to initiate NO release. The generation of NO and the natural decay of NO in the presence of oxygen were monitored over time with a microchip sensor, ISO-NOPMC, connected to an NO meter, ISO-NO Mark II (World Precision Instrument, Sarasota, FL). After the NO level reached the maximum, HDL (10 µg/ml), Hb (10 µM), or PBS was added to the reaction chamber, and NO decay was observed. NO consumption by HDL or Hb was evaluated by the rate of NO decay and the half-life of NO. To determine whether NO was consumed by oxyHb in HDL all Hb species were converted to metHb, a less NO-reactive Hb species, with 20 µM potassium ferricyanide (K3Fe(CN)6; Sigma).

MS Analysis of Hb Derived from Red Blood Cells (RBCs) and HDL
HDL samples for MALDI analysis were spotted on Q10 ProteinChip arrays (Ciphergen Biosystems) as described above, and sinapinic acid was used as the matrix and placed on a custom adapter plate for analysis in a PE/Sciex PrOTOF 2000 MALDI time-of-flight mass spectrometer. Spectra acquired resulted from 250-300 laser shots.

HDL protein samples from mice (n = 4) fed an atherogenic diet for 7 days were pooled and separated on an Eldex MicroPro HPLC system using a Michrom PLRP-S reverse phase support with an aqueous acetonitrile (gradient 10-50%), 0.1% trifluoroacetic acid solvent system to yield purified Hb {alpha} and beta chains. Hb isolated from RBCs was similarly prepared. A few microliters of each collected HPLC fraction were analyzed by electrospray mass spectrometry using a Thermo-Finnigan LTQ-FT tandem linear ion trap-Fourier transform ion cyclotron resonance high resolution mass spectrometer (Thermo Fisher Scientific). Mass spectra shown were the average of 50-100 scans. After each reported mass value for the most abundant isotope peak, the mass difference (in units of 1.00235 Da) between the most abundant isotope peak and the monoisotopic peak is denoted in italics.

To assign the Hb {alpha} and beta chain variants, a portion of each of the collected HPLC fractions was digested with trypsin, and the resulting peptide mixture was analyzed by LC/MS/MS. Fragment ion spectra (MS/MS) were automatically collected for each eluting peptide and searched against the National Center for Biotechnology Information (NCBI) protein data base using SEQUEST to assign the sequence. A single amino acid substitution (Asn-68) in the sequence of {alpha}-1 Hb (NCBI accession number P01942 [GenBank] ) resulted in a match to the observed molecular weight. Similarly three amino acid substitutions (Gly-13, Ala-20, and Ala-139) in the sequence of beta-1 Hb (NCBI accession number P02088 [GenBank] ) provided a match to the observed molecular weight.

Vasoreactivity Studies
The effects of HDL on vasoreactivity were analyzed as described previously with some minor modifications (26-28). Briefly 2-mm-long thoracic arterial rings (150-250 µm in diameter) were prepared from 3-month-old C57BL6/J female mice. The rings were mounted between two 40-µm-diameter wires (one attached to an adjustable support and the other to a force transducer) in bath chambers on a small vessel wire myograph (Multi-Myograph 610M, Danish Myo Technology, Aarhus, Denmark) containing 5 ml of Krebs-bicarbonate solution (pH 7.4) at 37 °C gassed with 95% O2 + 5% CO2. Tension was continuously recorded on a computer connected to a PowerLab Model 4SP data acquisition system using Chart 5 software (ADInstruments Pty. Ltd, Bella Vista, New South Wales, Australia). All experiments were performed at 37 °C.

After the rings were equilibrated, submaximal precontraction with phenylephrine (0.1 µM) and relaxation with acetylcholine (ACh; 10 µM) were tested to determine the integrity of the endothelium. Following washing and equilibration, one cycle of contraction with phenylephrine and relaxation with ACh was performed to determine the maximum relaxation by ACh. To determine the effects of HDL on the arterial relaxation, the rings were contracted with phenylephrine, incubated with HDL (10 µg/ml) for 15 min, and relaxed with ACh. The direct arterial relaxation by HDL and effects of HDL on ACh-mediated relaxation were determined by changes in arterial contraction before and after relaxation by ACh. After the experiments were completed, endothelium-independent relaxation was evaluated with 3-isobutyl-1-methylxanthine and papervine (1 nM).

Statistical Analysis
All data were statistically analyzed by t test unless specified. Significance was determined as p < 0.05.


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 1.
Inflammatory properties of HDL from C57BL6/J mice after a short term feeding. HDL samples from C57BL6/J mice on chow (C) or atherogenic diet (A) for 7 days (D7) (n = 3 per group) were analyzed for cholesterol (A), lipid hydroperoxide content (B), paraoxonase (PON) activity (C), and cholesterol efflux (D) as described under "Experimental Procedures." Each bar represents average with 1 S.D. p values were calculated by t test. * shows p < 0.05. DCF, 2,7,7'dichlorofluorescein diacetate; HDL-C, HDL cholesterol.

 


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 2.
Elevated levels of m/z 14,900 and 15,600 are present in four mouse models of atherosclerosis. Serum (A) or HDL (B) samples from C57BL6/J mice on an atherogenic diet for 7 days (D7, n = 12), 15 weeks (W15, n = 12), or a western diet for 10 days (WD, n = 8); low density lipoprotein receptor null mice on a western diet for 8 weeks (LDLR-/-, n = 4); or 12-week-old apoE null mice on normal chow (apoE-/-, n = 12) were subjected to SELDI ProteinChip analysis with Q10 (pI < 7) ProteinChip arrays. The two SELDI peaks of interest (m/z 14,900 and 15,600) in serum or HDL from mice on an atherogenic diet were compared with those of serum or HDL from the corresponding control group of mice on chow diet or in the case of apoE null mice with age-matched C57BL6/J mice on chow diet, and the resulting intensities were statistically analyzed. Each bar represents average -fold increase with 1 S.D. in each peak of m/z 14,900 (white) and 15,600 (gray). The data presented are all statistically significant with a p value <0.05.

 


Figure 3
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 3.
Determination of pI values for the peaks representing m/z 14,900 and 15,600. A, individual serum samples from C57BL6/J mice (n = 8) fed either a normal chow (C) or atherogenic diet (A) for 7 days (D7) or 15 weeks (W15) were desalted and fractionated by anion exchange spin columns with buffers at different pH as shown. The eluted fractions were subjected to SELDI ProteinChip analysis with cation exchange (CM10: pI > 4) or anion exchange (Q10: pI < 4) ProteinChip arrays. Relative intensity at 14,900 and at 15,600 is shown. B, anion exchange column fractions representing pH 7.0 and 4.0 were separated by 15% SDS-PAGE and immunoblotted for Hb.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
C57BL6/J mice fed a normal chow have anti-inflammatory HDL, whereas C57BL6/J mice fed an atherogenic diet for both short term (D7) and long term (W15) have proinflammatory HDL (29, 30) as determined by decreased cholesterol, increased lipid hydroperoxide content, reduced paraoxonase 1 activity, and decreased cholesterol efflux (Fig. 1, A-D). SELDI ProteinChip analysis using strong anion exchange (Q10) ProteinChip arrays identified a set of 14 m/z peaks differentially associated with HDL from atherogenic diet-fed mice (supplemental Table 1).

SELDI Peaks at m/z 14,900 and 15,600 Are Associated with Atherogenic/Hyperlipidemic Serum and HDL—Two of the 14 peaks representing m/z 14,900 and 15,600 were elevated by severalfold compared with controls in serum (Fig. 2A) and HDL (Fig. 2B) obtained from four different mouse models of atherosclerosis/hyperlipidemia. All subsequent experiments for the identification and characterization of proteins representing SELDI peaks m/z 14,900 and 15,600 were performed on serum and HDL samples obtained from D7 and W15 groups of C57BL6/J mice on an atherogenic diet.

Identification of Candidate Proteins at m/z 14,900 and 15,600—To determine the identity of the proteins representing the two SELDI peaks, we first examined the pI range for the two peaks. Individual HDL samples obtained from D7 and W15 groups on chow or atherogenic diet were subjected to anion exchange (Q10) fractionation. The fractions were eluted with different pH buffers and further analyzed by SELDI-TOF-MS (Fig. 3A). Most of the intensity representing the two peaks at m/z 14,900 and 15,600 eluted with buffers at pH 7.5 or higher except HDL from the W15 group on atherogenic diet (not shown). Using an on-line TagIdent protein data base with the size determined from SELDI-TOF-MS analysis and the corresponding pI as determined by anion exchange fractionation, Hb {alpha} and beta chains were identified as potential candidate proteins for 14,900 and 15,600, respectively.


Figure 4
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 4.
Hb in mouse atherogenic serum and lipoproteins. Pooled serum samples from C57BL6/J mice (n = 8 per group) on either a normal chow (C) or atherogenic diet (A) for 7 days (D7) or 15 weeks (W15) were separated by 15% SDS-PAGE (A) or on IEF/native two-dimensional gels (B). RBC lysate from mice on chow was loaded on a second native gel as a standard for Hb. Gels were immunoblotted for Hb.

 


Figure 5
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 5.
Association of Hb with apoA-I in HDL from hyperlipidemic mice. A, very low density lipoprotein (VLDL), LDL, HDL, and post-HDL (pHDL) FPLC fractions from pooled serum samples of D7 and W15 groups were separated by 15% SDS-PAGE and immunoblotted for Hb. B, concentrations of Hb in HDL and post-HDL from pooled serum samples of D7 groups (n = 4 per group) were determined by spectrophotometer as described under "Experimental Procedures." Each bar represents average ± 1 S.D. Changes in Hb contents in individual serum (C) or in apoA-I-containing HDL (D) were determined by direct or sandwich ELISA, respectively, as described under "Experimental Procedures." Individual serum from wild-type mice on either normal chow or atherogenic diet for 7 days (D7 (C) or D7 (A), respectively) (n = 16 per group) and apoE null mice (n = 8) were tested. Circles and bars represent individual data and averages, respectively. A1/Hb indicates Hb detected in apoA-I-containing material. p values were calculated by t test. NC, no change.

 
Identification of the Two Biomarker Proteins in the Fractions at pH 7.0 and 4.0 as Hb {alpha} and beta Chains by Tandem Mass Spectrometry—Based on the literature the pI for free Hb ranges between 7.0 and 8.5. Interestingly the two peaks representing m/z 14,900 and 15,600 that were associated with HDL from atherogenic mice eluted at pH 7.0 and 4.0 (Fig. 3A). Western blot analysis revealed that HDL from only atherogenic mice, not mice on a normal chow, contained Hb in the fractions at pH 7.0 and 4.0 (Fig. 3B). To further confirm the identities of the two biomarkers, the peaks corresponding to m/z 14,900 and 15,600 in the anion exchange (Q10) fractions at pH 7.0 and 4.0 were separated by SDS-PAGE, extracted, digested with trypsin, and subjected to microcapillary LC/MS/MS analysis. The resulting fragments were searched against mouse protein data bases (Sonar and SEQUEST) and identified as Hb {alpha} and beta chains. These data suggested that Hb {alpha} and beta chains represent m/z 14,900 and 15,600 that are differentially associated with HDL from atherogenic mice.

Hb Associates Preferentially with HDL from Atherogenic/Hyperlipidemic Mice—To validate the presence of Hb in proinflammatory HDL from atherogenic mice by non-SELDI methods, we first tested serum samples from mice on the atherogenic diet by SDS-PAGE and immunoblotting for Hb. The total amount of Hb in serum samples was not significantly different between normal and atherogenic serum (Fig. 4A). This was further confirmed by SELDI ProteinChip analysis with normal phase (NP20) arrays (data not shown). However, IEF/native two-dimensional gels showed that unique Hb complexes at low pI were formed in atherogenic serum (Fig. 4B). Lipoprotein fractionation followed by SDS-PAGE analysis (Fig. 5A) and spectrophotometric analysis on Hb further confirmed the significant increases of Hb in HDL from atherogenic mice (Fig. 5B).

Accumulation of Hb-associated ApoA-I in HDL from Atherogenic/Hyperlipidemic Mice—To determine whether Hb was physically associated with HDL, ELISA analyses were performed on serum samples. Direct ELISA once again revealed no significant difference in serum Hb content among all groups (Fig. 5C). However, sandwich ELISA analyses using an apoA-I capture antibody combined with Hb detection antibody identified elevated Hb levels in HDL from atherogenic mice (Fig. 5D).

Accumulation of Hb-associated ApoA-I in HDL from Healthy Donor and Patient Samples—To determine whether Hb is differentially associated with HDL in humans, 10 samples each from healthy controls and patients with CHD were examined for apoA-I-associated Hb (Fig. 6). To determine whether cholesterol quantity affects the accumulation of Hb, each of the two groups was further divided into two groups based on cholesterol (Fig. 6A). HDL from healthy donors whether from a low cholesterol or high cholesterol group, as expected, was protective in a cell-based monocyte chemotactic assay (3) (Fig. 6B), whereas the two patient groups of patient samples were proinflammatory in the assay (Fig. 6B). More interestingly, the patient samples had significantly higher levels of apoA-I-associated Hb compared with healthy donor samples (Fig. 6C) that was independent of cholesterol levels, suggesting that Hb associates with proinflammatory HDL in mice and humans.

Characterization of Hb Associated with HDL Fractions—To characterize the distinct physical and chemical properties of Hb associated with HDL, serum samples and HDL fractions from D7 and W15 groups on chow or atherogenic diet were subjected to native gel electrophoresis (Fig. 7A) and IEF gel electrophoresis (Fig. 7B). Native gels showed the association of Hb with high molecular weight particles in atherogenic serum that was exclusively associated with HDL from the W15 group on atherogenic diet (Fig. 7A). IEF gels demonstrated that Hb in the atherogenic serum had a reduced pI value around 4, whereas normal Hb had a pI value around 8 (Fig. 7B). The transition of the normal/free Hb (pI 8) to the modified Hb associated with HDL in atherogenic serum (pI 4) was clearly seen in samples from the D7 group. These changes were not due to changes in RBCs because RBCs isolated from the same groups of mice showed Hb with a normal pI (Fig. 7B). These observations along with IEF/native two-dimensional gels (Fig. 4B) confirmed that Hb from atherogenic serum has multiple forms (based on pI) and associates primarily with HDL.


Figure 6
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 6.
Association of Hb with apoA-I in human HDL from patients. A, plasma samples from healthy donors (H) or patients (P) were grouped according to their LDL cholesterol concentration as lower (L) or higher (H) than 130 mg/dl. Each number represents average ± 1 S.D. of the concentrations of total, HDL, and very low density lipoprotein (VLDL)/LDL cholesterols in mg/dl. B, HDL from a pooled serum sample of each group was tested for monocyte chemotaxis assay as described "Experimental Procedures." Data are represented as average with 1 S.D. of the number of migrated monocytes in nine fields for each HDL. C, Hb contents in apoA-I-containing HDL in individual serum of each group were determined by sandwich ELISA. Circles and bars represent individual data and averages, respectively. p values were calculated by t test, and * represents p < 0.05. Buf, buffer. A1/Hb indicates Hb detected in apoA-I-containing material.

 


Figure 7
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 7.
Hb complexes in HDL from mouse atherogenic serum. Pooled serum samples from C57BL6/J mice (n = 8) fed either a normal chow (C) or atherogenic diet (A) for 7 days (D7) or 15 weeks (W15) were fractionated by FPLC. Pooled serum, HDL, and RBC lysates were loaded on native PAGE gels (4-15%) (A) or IEF gels (pH 3-10) (B). Gels were transferred and immunoblotted for Hb. Lysed RBCs were used as the standard for Hb. HMW represents Hb complexes at high molecular weight.

 
Hb in HDL Has Distinct Properties but Is Not Modified—To determine whether the distinct properties of Hb associated with HDL from atherogenic mice are due to protein modification(s), MALDI analysis was performed for Hb extracted from RBCs and Q10 ProteinChip arrays spotted with HDL from the D7 group. MALDI spectra showed prominent peaks (supplemental Fig. 2) in the mass range consistent with Hb {alpha} and beta chains from RBCs (Table 1). These peaks were absent in the spectra of samples derived from HDL from normal sera (supplement Fig. 1). Alternatively a more accurate analysis of Hb was performed by other MS techniques. Following Hb isolation by HPLC from RBCs and HDL from atherogenic sera, electrospray MS analysis on a high resolution ion cyclotron resonance mass spectrometer yielded most abundant isotope mass values of 14,980.79 and 15,616.18 for the two proteins (Table 2). Electron capture dissociation spectra of the ions corresponding to Hb {alpha} and beta chains (data not shown) were consistent with the assigned sequences and measured mass values. These were further confirmed by LC/MS/MS analysis of the peptide mixture after digestion with trypsin (data not shown). These MS analyses yielded mass values for both Hb {alpha} and beta chains (Table 2) that are the same within experimental error as the values for the Hb obtained from RBCs. Thus, Hb present on HDL obtained from atherogenic mice was not different from Hb in RBCs. This suggests that the distinct properties/characteristics of Hb in atherogenic serum are most likely due to its association with HDL.


View this table:
[in this window]
[in a new window]

  		TABLE 1
Hb mass values from MALDI analysis of proinflammatory HDL

The observed values are in good agreement with calculated values for the normal Hb obtained from RBCs from normal mice of the same strain. The variant assignment is based on LC/MS/MS analysis of the peptides resulting from the tryptic digest of the Hb isolated from RBCs. D7A = HDL from mice fed an atherogenic diet for 7 days.

 


View this table:
[in this window]
[in a new window]

  		TABLE 2
Hb mass values from electrospray analysis of the HPLC-separated chains

Shown is the comparison of the theoretical mass values for mouse Hb {alpha} and beta chains and those observed from the electrospray mass spectra of individual chains isolated by reverse phase liquid chromatography of the HDL fraction from atherogenic mice and red blood cells from normal control mice.

 
Hb Associated with HDL Is Predominantly in an "Oxy" Form and Consumes NO—Because Hb in HDL is not modified, we next examined the state of Hb in HDL samples. Hb is known to exist in three forms, deoxyHb (Fe2+), oxyHb (Fe2+), and metHb (Fe3+). Spectrophotometric analyses of HDL from normal and atherogenic mice revealed that the major Hb species associated with HDL from atherogenic mice is the oxyHb form (Fig. 8) with small amounts of metHb; however, we did not detect any deoxyHb. Moreover we further quantified the composition of HDL-bound Hb and compared it with Hb prepared from RBCs and found similar proportions of oxyHb (87.3 ± 3.1% for RBCs and 75.9 ± 5.2% for the atherogenic HDL) and metHb (12.7% for RBCs and 24.1% for the atherogenic HDL). Interestingly in sharp contrast to normal HDL (Fig. 9, C and F), HDL from atherogenic mice consumed NO readily (Fig. 9, D and F); this was prevented when the oxyHb HDL was converted to the inert metHb using potassium ferricyanide (Fig. 9E). Furthermore spectrophotometric analyses of HDL revealed that HDL from atherogenic mice contained stable NO products (Fig. 9G). These data suggest that HDL from atherogenic mice could participate in the sequestration of NO through the Hb complexes associated with it.


Figure 8
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 8.
Spectrophotometric determination of Hb in proinflammatory HDL. Amounts and forms of Hb were determined from pooled FPLC fractions containing HDL using a Beckman DU 640 spectrophotometer. A, the spectra of all samples and pure species were scanned from 380 to 700 nm. B, spermine NONOate, a slow timed-release NO donor, was added to HDL from mice fed an atherogenic diet for 7 days to observe the conversion of oxyHb to metHb. The concentrations of oxyHb and metHb were deconvoluted by fitting a set of pure species basis spectra to the measured spectra by means of linear regression. The concentrations of Hb in pooled serum (C), HDL (D), and post-HDL (pHDL) from pooled serum (E) (n = 4 per group) of mice (n = 12 per group) on a normal chow (C) or atherogenic diet (A) for 7 days (D7) were determined by spectrophotometer as described under "Experimental Procedures." Each bar represents average -fold with 1 S.D. p values were calculated by t test, and * shows p < 0.05.

 


Figure 9
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 9.
NO consumption assay and NO content in HDL from atherogenic mice. Hb or HDL was incubated with PROLI NONOate, a pH-dependent NO timed-release donor, to determine the NO consumption. PBS at pH 7.4 was used to start NO release. A, in the absence of sample, the generation of NO from the NO donor and the natural decay of NO in the presence of oxygen was monitored in electrical current (pA) over time with a microchip sensor connected to an NO meter. After the NO level reached the maximum, Hb purified from RBCs (B) or HDL (10 µg/ml) from mice fed a normal chow (C) or atherogenic diet (D) for 7 days was added to the reaction chamber, and NO decay was observed. E, HDL from atherogenic mice was pretreated with potassium ferricyanide (20 µM), and Hb was converted to metHb before addition to the reaction chamber. F, effect of purified Hb from RBC lysate or HDL isolated from either chow-fed mice (HDL (C)) or atherogenic diet (HDL (A)) on half-life of NO using the NO consumption assay (n = 4 per group). Each bar represents average with 1 S.D. G, the concentrations of ONOO- in HDL (n = 4 per group) from pooled serum of mice (n = 12 per group) on a normal chow (HDL (C)) or atherogenic diet (HDL (A)) for 7 days was determined by the absorbance at 302 nm. Open circles and bars represent individual or average concentrations of ONOO-, respectively. p values were calculated by t test.

 
HDL Containing OxyHb Causes Impaired Arterial Vasoconstriction—NO consumption by proinflammatory HDL suggests that Hb content, particularly oxyHb, in proinflammatory HDL may play a role in the vasomotor function. To test this hypothesis, we performed vasoreactivity experiments and analyzed vasoconstriction and vasodilation in normal mouse arteries in the presence of HDL from either mice fed a chow diet or an atherogenic diet (Fig. 10A). HDL from atherogenic mice caused further vasoconstriction of arteries precontracted with 0.1 µM phenylephrine (11%) (Fig. 10B) and impaired acetylcholine-mediated relaxation (~50%) (Fig. 10C) when compared with normal HDL, which weakly relaxed arteries (3%) and had no influence on acetylcholine-mediated vasodilation. Thus, HDL from atherogenic mice may promote vasomotor dysfunction when compared with normal/anti-inflammatory HDL presumably through sequestration of NO. These studies suggest that Hb·HDL complexes have the potential to impair vasomotor function in vivo.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Protein profiling is an efficient way to determine differentially expressed and/or associated proteins in complex biological mixtures, such as serum and lipoproteins, and allows rapid evaluation of biologically important functions. We have utilized SELDI-TOF-MS system to facilitate protein profiling of complex biological mixtures (14-17) and reported previously the identification of biomarkers for the early detection of ovarian cancer (20, 21). We have now utilized the discovery capabilities of the SELDI technique and identified protein peaks that differentiate proinflammatory HDL from normal/anti-inflammatory HDL.

Biochemical analysis of the protein profiles identified Hb {alpha} chain (14,900) and beta chain (15,600) as two of the potential biomarkers associated with proinflammatory HDL in mouse models of atherosclerosis/hyperlipidemia. These identifications were confirmed by the different MS techniques described above. We were initially concerned with this finding because mechanical handling and/or sample preparation could potentially result in RBC lysis and Hb release. However, subsequent careful repetitions of these experiments determined that Hb was not an artifact but rather a specific and significant marker of proinflammatory HDL present in atherogenic serum. First, we showed that total Hb mass was not significantly different between serum samples obtained from chow- and atherogenic diet-fed mice (Figs. 4A and 5C). Second, during the conversion of normal HDL into proinflammatory HDL (chow versus atherogenic diet), we did not find any modification of Hb (Fig. 7B and Table 2) in RBC lysates obtained from the same mice, once again suggesting that Hb association with proinflammatory HDL is a specific phenomenon under proinflammatory conditions. Third, the association of Hb with proinflammatory HDL was dependent on the extent of proinflammatory conditions as evident by comparing Hb in D7 and W15 lipoprotein samples (Figs. 5 and 7). Finally in four different models of atherosclerosis/hyperlipidemia including apoE null mice, which have proinflammatory HDL on a chow diet, Hb was associated with HDL. Taken together, these results suggest that HDL-associated Hb is a biomarker of proinflammatory HDL.


Figure 10
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 10.
Effect of Hb-containing HDL on vasoreactivity. A, thoracic aorta was precontracted with phenylephrine (PE; 0.1 µM) followed by incubation with HDL (10 µg/ml) and relaxed with ACh (10 µM). HDL was isolated by FPLC from pooled serum obtained from either mice on a normal chow (HDL (C)) (shown by a gray line) or atherogenic diet for 7 days (HDL (A)) (shown by a black line). Upon addition of the atherogenic HDL from mice fed an atherogenic diet, both basal NO release and acetylcholine-stimulated release were impaired compared with the addition of normal HDL from mice fed normal chow that caused a weak relaxation to basal NO release and a strong relaxation to acetylcholine. The artery relaxation assay was repeated with HDL (10 µg/ml) from pooled samples (n = 4 per group) from mice on a normal chow or atherogenic diet. The vascular response to HDL (B) and the effects of HDL on ACh-mediated relaxation (C) are represented as percent changes. Each bar represents the average of changes on vascular response of each group with 1 S.D. p values were calculated by t test.

 
Physiological pI for Hb has been reported to be between pI 7.0 and 8.5. We found that Hb associated with proinflammatory HDL has at least two Hb species with distinct pI values (Figs. 3 and 7B). After 15 weeks on the atherogenic diet all of the Hb associated with HDL demonstrated this abnormal pI. The changes in the pI of Hb after feeding the atherogenic diet may likely relate to its association with HDL, not Hb, induced by the diet because Hb associated with HDL showed no modification, and its sequence was identical to the Hb from RBCs (Table 2). Although refolding of the Hb polypeptide chains in the modified HDL microenvironment may account for some of the changes it is unlikely that the drastic alterations observed in the properties are a result of subtle misfolding. To our knowledge, the different forms of Hb (oxy, met, and deoxy) are not vastly different from one another in pI. However, it should be noted that HDL is a huge complex with an overall pI around 5. Therefore, one likely explanation is that the pI of Hb is not altered, but rather the association, as a complex with HDL, is what makes the Hb travel at a different pI on IEF gels. Furthermore as will be discussed below, Hb might be sequestered to HDL through a physiological scavenger pathway that involves haptoglobin.

We saw modest differences in the composition of Hb species between RBCs and atherogenic HDL. It is surprising that an extracellular Hb species is kept in its ferrous form in the absence of metHb reductase, because in the presence of oxygen Hb undergoes rapid autoxidation to metHb. Thus it appears that the heme moiety in the HDL-bound Hb may be protected from autoxidation by HDL components and/or other scavenger proteins such as hemopexin discussed below. Hb associated with proinflammatory HDL was the pro-oxidant form, oxyHb, which consumed NO (Fig. 9) and contracted arterial vessels (Fig. 10). Recent studies demonstrated that Hb could oxidize LDL and alter the antiatherogenic properties of HDL, including its ability to efflux cholesterol (13). Taken together, Hb associated with proinflammatory HDL might not be only a biomarker of proinflammatory HDL but also might be involved in the conversion of HDL from anti-inflammatory to proinflammatory.

Haptoglobin (Hp) and hemopexin (Hx) are plasma proteins with the highest binding affinity for Hb (Kd {approx} 1p M) and heme (Kd < 1p M), respectively. They are expressed mainly in the liver and belong to the family of acute phase proteins whose synthesis is induced during inflammatory processes (31, 32). It is well established that Hb (the most abundant and functionally important protein in erythrocytes), once released from RBCs, becomes highly toxic because of the oxidative properties of heme, which participates in the Fenton reaction to produce reactive oxygen species causing cell injury (33). The toxicity of heme is increased by heme hydrophobicity, which enables it to intercalate into lipid membranes and other lipophilic compartments when not associated with proteins (34). Usually low amounts of extravascular hemolysis occur during enucleation of erythroblasts and destruction of senescent erythrocytes, thus causing Hb release into plasma. Under intravascular hemolysis-linked pathologic conditions, such as hemorrhage, hemoglobinopathies, ischemia reperfusion, or malaria, large amounts of free Hb are released (35). Once in the plasma, free Hb rapidly dissociates in dimers that are bound by Hp. Metabolism of plasma Hb is considered a main function of tissue macrophages, which can take up Hb·Hp complexes through the macrophage scavenger receptor CD163 (36, 37) and internalize them for degradation (38). Interestingly a very recent study identified low density lipoprotein receptor-related protein/CD91 (39) as the receptor responsible for scavenging hemopexin·heme complexes. Lipoprotein receptor-related protein/CD91 is expressed in several cell types including macrophages and hepatocytes, which can internalize the heme·Hx complex through receptor-mediated endocytosis (40). In the experiments reported here, we did not see a release of Hb from RBCs but rather a conversion of existing Hb into an alternative form (not reported previously) that associates with HDL fractions. It is possible that under oxidative stress conditions, Hb·Hp·Hx complexes are formed and associate with HDL for rapid clearance from the circulation. Indeed Hp has been reported to associate with apoA-I, the major protein component of HDL (41-44). Hp association of apoA-I alters HDL function (45, 46).

Hb is a known marker for injuries and diseases associated with glycemia, oxidative stress, hypertension, insulin resistance, obesity, and diabetes (47-49). Hb is also considered to be toxic because free Hb is also a potential oxidant due to its heme (iron) and heme-bound reactive radicals (50), which have also been shown to oxidize LDL in vivo (51-53). Our findings show for the first time that in an oxidative stress environment Hb associates with HDL fractions in mice. Proinflammatory HDL in atherogenic serum contains lipid hydroperoxides, lacks paraoxonase activity, activates monocyte, fails to prevent the oxidation of LDL, and exhibits less cholesterol efflux (54). Here we report that Hb specifically associates with proinflammatory HDL in atherogenic mice. It is possible that Hb association with HDL might be involved in the conversion of HDL from anti-inflammatory to proinflammatory. However, further direct studies are needed before we can conclude this relationship between Hb and HDL function.

In conclusion, Hb with distinct physical and chemical properties associates with proinflammatory HDL in animal models of atherosclerosis and in human samples in preliminary experiments. If these data can be extended to large human population studies, Hb-associated HDL may serve as a novel marker of proinflammatory HDL.


   FOOTNOTES
 
* This work was supported by NHLBI, National Institutes of Health Grants 1RO1HL71776 (to S. T. R.) and HL-30568 (to A. M. F., M. N., and S. T. R.) and the Laubisch, Castera, and M. K. Grey Funds at UCLA. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains supplemental Table 1 and Figs. 1-5. Back

1 Principals in Bruin Pharma. Back

2 An officer in Bruin Pharma. Back

3 To whom correspondence should be addressed: Dept. of Medicine/Cardiology, Dept. of Molecular and Medical Pharmacology, University of California Los Angeles, 650 Charles E. Young Dr. S., A8-131, CHS, Los Angeles, CA 90095. Tel.: 310-206-3915; Fax: 310-206-3605; E-mail: sreddy{at}mednet.ucla.edu.

4 The abbreviations used are: HDL, high density lipoprotein; apo, apolipoprotein; CHD, coronary heart disease; ELISA, enzyme-linked immunosorbent assay; Hb, hemoglobin; LDL, low density lipoprotein; NO, nitric oxide; RBC, red blood cell; SELDI, surface-enhanced laser desorption/ionization; TOF, time-of-flight; MS, mass spectrometry; MS/MS, tandem mass spectrometry; LC, liquid chromatography; HPLC, high pressure liquid chromatography; FPLC, fast protein liquid chromatography; PBS, phospate-buffered saline; IEF, isoelectric focusing; HRP, horseradish peroxidase; ACh, acetylcholine; MALDI, matrix-assisted laser desorption ionization; Hp, haptoglobin; Hx, hemopexin; proliNONOate, 1-(hydroxy-NNO-azoxy)-L-proline. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Gordon, T., Castelli, W. P., Hjortland, M. C., Kannel, W. B., and Dawber, T. R. (1977) Am. J. Med. 62, 707-714[CrossRef][Medline] [Order article via Infotrieve]
  2. Castelli, W. P., Garrison, R. J., Wilson, P. W., Abbott, R. D., Kalousdian, S., and Kannel, W. B. (1986) J. Am. Med. Assoc. 256, 2835-2838[Abstract]
  3. Ansell, B. J., Navab, M., Hama, S., Kamranpour, N., Fonarow, G., Hough, G., Rahmani, S., Mottahedeh, R., Dave, R., Reddy, S. T., and Fogelman, A. M. (2003) Circulation 108, 2751-2756[Abstract/Free Full Text]
  4. Navab, M., Ananthramaiah, G. M., Reddy, S. T., Van Lenten, B. J., Ansell, B. J., Fonarow, G. C., Vahabzadeh, K., Hama, S., Hough, G., Kamranpour, N., Berliner, J. A., Lusis, A. J., and Fogelman, A. M. (2004) J. Lipid Res. 45, 993-1007[Abstract/Free Full Text]
  5. Ridker, P. M. (2002) Circulation 105, 2-4[Free Full Text]
  6. Navab, M., Hama, S. Y., Anantharamaiah, G. M., Hassan, K., Hough, G. P., Watson, A. D., Reddy, S. T., Sevanian, A., Fonarow, G. C., and Fogelman, A. M. (2000) J. Lipid Res. 41, 1495-1508[Abstract/Free Full Text]
  7. Navab, M., Hama, S. Y., Cooke, C. J., Anantharamaiah, G. M., Chaddha, M., Jin, L., Subbanagounder, G., Faull, K. F., Reddy, S. T., Miller, N. E., and Fogelman, A. M. (2000) J. Lipid Res. 41, 1481-1494[Abstract/Free Full Text]
  8. Van Lenten, B. J., Hama, S. Y., de Beer, F. C., Stafforini, D. M., McIntyre, T. M., Prescott, S. M., La Du, B. N., Fogelman, A. M., and Navab, M. (1995) J. Clin. Investig. 96, 2758-2767[Medline] [Order article via Infotrieve]
  9. Van Lenten, B. J., Wagner, A. C., Nayak, D. P., Hama, S., Navab, M., and Fogelman, A. M. (2001) Circulation 103, 2283-2288[Abstract/Free Full Text]
  10. Mackness, M. I., Durrington, P. N., and Mackness, B. (2004) Am. J. Cardiovasc. Drugs 4, 211-217[CrossRef][Medline] [Order article via Infotrieve]
  11. Navab, M., Berliner, J. A., Subbanagounder, G., Hama, S., Lusis, A. J., Castellani, L. W., Reddy, S., Shih, D., Shi, W., Watson, A. D., Van Lenten, B. J., Vora, D., and Fogelman, A. M. (2001) Arterioscler. Thromb. Vasc. Biol. 21, 481-488[Abstract/Free Full Text]
  12. Hayek, T., Oiknine, J., Brook, J. G., and Aviram, M. (1994) Biochem. Biophys. Res. Commun. 205, 1072-1078[CrossRef][Medline] [Order article via Infotrieve]
  13. Navab, M., Ananthramaiah, G. M., Reddy, S. T., Van Lenten, B. J., Ansell, B. J., Hama, S., Hough, G., Bachini, E., Grijalva, V. R., Wagner, A. C., Shaposhnik, Z., and Fogelman, A. M. (2005) Ann. Med. 37, 173-178[CrossRef][Medline] [Order article via Infotrieve]
  14. Fung, E. T., Thulasiraman, V., Weinberger, S. R., and Dalmasso, E. A. (2001) Curr. Opin. Biotechnol. 12, 65-69[CrossRef][Medline] [Order article via Infotrieve]
  15. Issaq, H. J., Veenstra, T. D., Conrads, T. P., and Felschow, D. (2002) Biochem. Biophys. Res. Commun. 292, 587-592[CrossRef][Medline] [Order article via Infotrieve]
  16. Rubin, R. B., and Merchant, M. (2000) Am. Clin. Lab. 19, 28-29[Medline] [Order article via Infotrieve]
  17. Weinberger, S. R., Dalmasso, E. A., and Fung, E. T. (2002) Curr. Opin. Chem. Biol. 6, 86-91[CrossRef][Medline] [Order article via Infotrieve]
  18. Navab, M., Hama, S. Y., Hough, G. P., Subbanagounder, G., Reddy, S. T., and Fogelman, A. M. (2001) J. Lipid Res. 42, 1308-1317[Abstract/Free Full Text]
  19. Navab, M., Anantharamaiah, G. M., Reddy, S. T., Hama, S., Hough, G., Grijalva, V. R., Wagner, A. C., Frank, J. S., Datta, G., Garber, D., and Fogelman, A. M. (2004) Circulation 109, 3215-3220[Abstract/Free Full Text]
  20. Kozak, K. R., Amneus, M. W., Pusey, S. M., Su, F., Luong, M. N., Luong, S. A., Reddy, S. T., and Farias-Eisner, R. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 12343-12348[Abstract/Free Full Text]
  21. Kozak, K. R., Su, F., Whitelegge, J. P., Faull, K., Reddy, S., and Farias-Eisner, R. (2005) Proteomics 5, 4589-4596[CrossRef][Medline] [Order article via Infotrieve]
  22. Gomez, S. M., Bil, K. Y., Aguilera, R., Nishio, J. N., Faull, K. F., and Whitelegge, J. P. (2003) Mol. Cell. Proteomics. 2, 1068-1085[Abstract/Free Full Text]
  23. Vaughn, M. W., Huang, K. T., Kuo, L., and Liao, J. C. (2000) J. Biol. Chem. 275, 2342-2348[Abstract/Free Full Text]
  24. Liao, J. C., Hein, T. W., Vaughn, M. W., Huang, K. T., and Kuo, L. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 8757-8761[Abstract/Free Full Text]
  25. Vaughn, M. W., Huang, K. T., Kuo, L., and Liao, J. C. (2001) Nitric Oxide 5, 18-31[CrossRef][Medline] [Order article via Infotrieve]
  26. Garban, H. J., Buga, G. M., and Ignarro, L. J. (2004) J. Cardiovasc. Pharmacol. 43, 638-644[CrossRef][Medline] [Order article via Infotrieve]
  27. Nofer, J. R., van der Giet, M., Tolle, M., Wolinska, I., von Wnuck Lipinski, K., Baba, H. A., Tietge, U. J., Godecke, A., Ishii, I., Kleuser, B., Schafers, M., Fobker, M., Zidek, W., Assmann, G., Chun, J., and Levkau, B. (2004) J. Clin. Investig. 113, 569-581[CrossRef][Medline] [Order article via Infotrieve]
  28. Yuhanna, I. S., Zhu, Y., Cox, B. E., Hahner, L. D., Osborne-Lawrence, S., Lu, P., Marcel, Y. L., Anderson, R. G., Mendelsohn, M. E., Hobbs, H. H., and Shaul, P. W. (2001) Nat. Med. 7, 853-857[CrossRef][Medline] [Order article via Infotrieve]
  29. Shih, D. M., Xia, Y. R., Wang, X. P., Miller, E., Castellani, L. W., Subbanagounder, G., Cheroutre, H., Faull, K. F., Berliner, J. A., Witztum, J. L., and Lusis, A. J. (2000) J. Biol. Chem. 275, 17527-17535[Abstract/Free Full Text]
  30. Hedrick, C. C., Hassan, K., Hough, G. P., Yoo, J. H., Simzar, S., Quinto, C. R., Kim, S. M., Dooley, A., Langi, S., Hama, S. Y., Navab, M., Witztum, J. L., and Fogelman, A. M. (2000) Arterioscler. Thromb. Vasc. Biol. 20, 1946-1952[Abstract/Free Full Text]
  31. Altruda, F., Poli, V., Restagno, G., Argos, P., Cortese, R., and Silengo, L. (1985) Nucleic Acids Res. 13, 3841-3859[Abstract/Free Full Text]
  32. Bowman, B. H., and Kurosky, A. (1982) Adv. Hum. Genet. 12, 189-261, 453-454[Medline] [Order article via Infotrieve]
  33. Puppo, A., and Halliwell, B. (1988) Biochem. J. 249, 185-190[Medline] [Order article via Infotrieve]
  34. Balla, J., Jacob, H. S., Balla, G., Nath, K., Eaton, J. W., and Vercellotti, G. M. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 9285-9289[Abstract/Free Full Text]
  35. Wagener, F. A., Abraham, N. G., van Kooyk, Y., de Witte, T., and Figdor, C. G. (2001) Trends Pharmacol. Sci. 22, 52-54[CrossRef][Medline] [Order article via Infotrieve]
  36. Fabriek, B. O., Dijkstra, C. D., and van den Berg, T. K. (2005) Immunobiology 210, 153-160[CrossRef][Medline] [Order article via Infotrieve]
  37. Schaer, D. J., Schaer, C. A., Buehler, P. W., Boykins, R. A., Schoedon, G., Alayash, A. I., and Schaffner, A. (2006) Blood 107, 373-380[Abstract/Free Full Text]
  38. Kristiansen, M., Graversen, J. H., Jacobsen, C., Sonne, O., Hoffman, H. J., Law, S. K., and Moestrup, S. K. (2001) Nature 409, 198-201[CrossRef][Medline] [Order article via Infotrieve]
  39. Hvidberg, V., Maniecki, M. B., Jacobsen, C., Hojrup, P., Moller, H. J., and Moestrup, S. K. (2005) Blood 106, 2572-2579[Abstract/Free Full Text]
  40. Hunt, R. C., Hunt, D. M., Gaur, N., and Smith, A. (1996) J. Cell. Physiol. 168, 71-80[CrossRef][Medline] [Order article via Infotrieve]
  41. Kunitake, S. T., Carilli, C. T., Lau, K., Protter, A. A., Naya-Vigne, J., and Kane, J. P. (1994) Biochemistry 33, 1988-1993[CrossRef][Medline] [Order article via Infotrieve]
  42. Porta, A., Cassano, E., Balestrieri, M., Bianco, M., Picone, R., De Stefano, C., and Abrescia, P. (1999) Zygote 7, 67-77[CrossRef][Medline] [Order article via Infotrieve]
  43. Rademacher, B. E., and Steele, W. J. (1987) Anal. Biochem. 160, 119-126[CrossRef][Medline] [Order article via Infotrieve]
  44. Spagnuolo, M. S., Cigliano, L., D'Andrea, L. D., Pedone, C., and Abrescia, P. (2005) J. Biol. Chem. 280, 1193-1198[Abstract/Free Full Text]
  45. Balestrieri, M., Cigliano, L., Simone, M. L., Dale, B., and Abrescia, P. (2001) Mol. Reprod. Dev. 59, 186-191[CrossRef][Medline] [Order article via Infotrieve]
  46. Cigliano, L., Spagnuolo, M. S., Dale, B., Balestrieri, M., and Abrescia, P. (2001) Steroids 66, 889-896[CrossRef][Medline] [Order article via Infotrieve]