Abstract
Aim: Postprandial hypertriglyceridemia (PHTG) is an independent risk factor for coronary heart diseases. PHTG exhibits accumulation of apoB-48 containing chylomicron remnants (CM-Rs) and apoB-100 containing VLDL remnants (VLDL-Rs), which are both known to be atherogenic. However, unlike VLDL-Rs, structural and functional characterization of CM-Rs remains to be elucidated due to challenges in separating CM-Rs from VLDL-Rs. Recently, we successfully isolated CM-Rs and VLDL-Rs utilizing anti-apoB-48 or apoB-100 specific antibodies. This study aimed to characterize the proteome of CM-Rs along with that of VLDL-Rs.
Methods: Eight healthy subjects were enrolled. Venous blood was drawn 3 hours after high-fat-containing meals. We isolated CM-Rs and VLDL-Rs from sera through combination of ultracentrifugation and immunoprecipitation using apoB-48 or apoB-100 specific antibodies, followed by shotgun proteomic analysis.
Results: We identified 42 CM-Rs or VLDL-Rs-associated proteins, including 11 potential newly identified proteins such as platelet basic protein (PPBP) and platelet factor 4, which are chemokines secreted from platelets. ApoA-I, apoA-IV, and clusterin, which are also known as HDL-associated proteins, were significantly more abundant in CM-Rs. Interestingly, apoC-I, which reduces the activity of lipoprotein lipase and eventually inhibits catabolism of remnant proteins, was also more abundant in CM-Rs. Moreover, we identified proteins involved in complement regulation such as complement C3 and vitronectin, and those involved in acute-phase response such as PPBP, serum amyloid A protein 2, and protein S100-A8, in both CM-Rs and VLDL-Rs.
Conclusions: We have firstly characterized the proteome of CM-Rs. These findings may provide an explanation for the atherogenic properties of CM-Rs.
Daisaku Masuda and Takeshi Okada contributed equally to this work.
Introduction
Atherosclerotic cardiovascular diseases (ASCVD), such as coronary heart disease (CHD) and stroke, are the leading cause of death all over the world. The risk factors for ASCVD include dyslipidemia, hypertension, diabetes mellitus, smoking, age, sex (male), body mass index and metabolic syndrome (MetS) based upon increased visceral adiposity. Regarding the treatment of dyslipidemia, intensive interventions against high low-density lipoprotein (LDL)-cholesterol (LDL-C) using statins, intestinal cholesterol transporter inhibitor (ezetimibe), proprotein convertase subtilisin kexin 9 (PCSK9) inhibitor, etc., have proved to reduce the primary and secondary ASCVD events.
Although LDL-C-lowering drugs have succeeded in reducing ASCVD risk, cardiovascular (CV) events have not been completely prevented. Thus, residual CV risk factors should be considered after the appropriate management of LDL-C levels. Among them, the importance of increased levels of triglycerides (TG) or reduced levels of high-density lipoprotein cholesterol (HDL-C) has been focused on and emphasized as a possible target for the prevention of ASCVD1, 2). Many epidemiological, as well as genetic studies, have demonstrated that fasting hypertriglyceridemia is associated with an accelerated risk for ASCVD. Furthermore, postprandial hyperlipidemia or postprandial hypertriglyceridemia (PHTG) has recently been established as one of the independent and residual risk factors for CHD2, 3). PHTG refers to a state of postprandial accumulation of intestine-derived chylomicrons (CMs) and CM remnants (CM-Rs)4, 5). CM-Rs have been assumed to have a variety of atherogenic properties6, 7). CMs and CM-Rs have one apolipoprotein B-48 (apoB-48) molecule per one lipoprotein particle; therefore, the serum level of apoB-48 indicates the number of apoB-48-containing lipoproteins.
It has been challenging to generate specific antibodies against apoB-48 due to its overlapping sequence of amino acid residues with apoB-100. However, we successfully developed monoclonal antibodies against human apoB-48 8) and established enzyme-linked immunoassay (ELISA)8) and chemiluminescent enzyme immunoassay (CLEIA)9) methods for measuring serum concentrations of apoB-48. We demonstrated that apoB-48 concentrations were significantly higher in patients with postprandial hyperlipidemia10, 11), MetS9, 12), CKD13), hypothyroidism14) and CHD15). Additionally, we found that serum apoB-48 levels were correlated with intima-media thickness (IMT) of carotid arteries in subjects with serum TG levels of 100-150 mg/dL15).
Notably, fasting serum concentrations of apoB-48 were only about 1/200 of those of liver-derived apoB-100 9). However, the content of apoB-48 in atherosclerotic plaque was proven to be twice as much as that of apoB-100 in carotid arteries of Watanabe Heritable Hyperlipidemic rabbits16). Considering their concentrations in serum, these data suggest a more pronounced contribution of CM-Rs to atherogenesis than apoB-100-containing lipoproteins. They also motivate us to characterize the structural and functional components of CM-Rs.
Proteomic analyses are promising tools for identifying and quantifying proteins associated with lipoproteins. The proteomes of HDL17, 18), LDL19), and VLDL19) have been extensively investigated, revealing associations with various functions in lipid metabolism, innate immunity, inflammation, protease regulation, and more. However, there have been no reports on proteomic analyses of CM-Rs thus far, likely due to the challenges in separating CM-Rs from apoB-100-containing lipoproteins.
Recently, we successfully isolated CM-Rs through a combination of ultracentrifugation and immunoprecipitation, utilizing anti-apoB-48 specific antibodies, as described above. In this study, we present, for the first time, the characterization of the protein composition of CM-Rs and compare it with that of VLDL-Rs.
Materials and Methods
1. Subjects
To isolate the CM-Rs, eight healthy volunteers were enrolled. All subjects were nonsmokers and had no apparent acute or chronic illnesses, especially primary or secondary dyslipidemia. The study protocol was reviewed and approved by the institutional review board of Osaka University. This study was performed in accordance with the ethical principles of the Declaration of Helsinki and the Ethical Guidelines for Clinical Research, enforced by the Ministry of Health, Labour and Welfare of Japan (2008). Each participant provided written informed consent for study participation. The basic characteristics of these volunteers are presented in Table 1.
Table 1.Characteristics and lipid profiles of subjects
| Age (year) |
|
42.1±8.9 |
| Sex (m/f) |
|
3/5 |
| Height (cm) |
|
172.5±9.9 |
| Weight (kg) |
|
79.1±18.9 |
| BMI |
|
26.3±3.8 |
|
Cholesterol (mg/dL) |
Triglyceride (mg/dL) |
| Total |
175.5±27.6 |
125.5±82.2 |
| CM (＞80nm) |
1.1±0.8 |
11.5±10.0 |
| VLDL (30-80nm) |
24.7±11.4 |
79.2±63.5 |
| LDL (16-30nm) |
96.2±20.9 |
23.3±7.4 |
| HDL (8-16nm) |
53.5±13.1 |
11.5±3.9 |
BMI, body mass index; CM, chylomicron; VLDL, very low density lipoprotein;
LDL, low density lipoprotein; HDL, high density lipoprotein Values are mean±SD
Subjects refrained from consuming caffeine or alcohol the day before the experiment and ingested a meal consisting of a hamburger and french fries (McDonald’s, Tokyo, Japan) and cola (Coca‑Cola(Japan) Company, Limited) following a 12-hour overnight fast. The composition of the meal is detailed in Supplementary Table 1.
Supplementary Table 1.The composition of the meal
|
Hamburger |
French fries |
Cola |
Total |
| Energy (kcal) |
530 |
424 |
140 |
1094 |
| Protein (g) |
27.1 |
5.3 |
0 |
32.4 |
| Fat (g) |
28.2 |
22 |
0 |
50.2 |
| Carbohydrate (g) |
41.9 |
51.4 |
35.1 |
128.4 |
| Cholesterol (mg) |
75 |
7 |
0 |
82 |
| NaCl (g) |
3.4 |
1.1 |
0 |
4.5 |
| Na (mg) |
1328 |
439 |
7 |
1774 |
| K (mg) |
404 |
950 |
0 |
1354 |
| Ca (mg) |
144 |
22 |
7 |
173 |
| P (mg) |
289 |
188 |
52 |
529 |
| Fe (mg) |
2.2 |
1.1 |
0 |
3.3 |
| Vitamin A (µg) |
75 |
0 |
0 |
75 |
| Vitamin B1 (mg) |
0.16 |
0.22 |
0 |
0.38 |
| Vitamin B2 (mg) |
0.25 |
0.03 |
0 |
0.28 |
| Niacin (mg) |
4.8 |
3.9 |
0 |
8.7 |
| Vitamin C (mg) |
2 |
36 |
0 |
38 |
| Fiber (g) |
2.9 |
5 |
0 |
7.9 |
Nine milliliters of venous blood were collected in a blood collection tube (VENOJECT II®) (TERUMO, Tokyo, Japan) three hours after the meal. During the test, subjects refrained from exercise and eating but had free access to water. Serum was prepared by centrifugation 30 minutes after collection at 2,000 ×g for 10 minutes at 4℃ and was used directly for lipoprotein isolation. Fractions with a density of d ＜0.94 g/mL (CMs fraction) and 0.94 ＜d＜1.006 g/mL (VLDL fraction) were isolated from human serum using density gradient ultracentrifugation following the method described by Gofman et al.20) with some modifications. Four mL of serum was pipetted into 16×76 mm polycarbonate ultracentrifuge tubes (Beckman Coulter Inc., CA, USA). A gradient was formed by carefully layering 2 mL of Dulbecco’s phosphate-buffered saline (D-PBS(-)) (d=1.006 g/mL) (Wako Pure Chemical Industries, Ltd., Osaka, Japan) on top of the serum. The samples were then centrifuged with XE-90 (Beckman Coulter Inc.) for 30 minutes at 4℃ at 20,000 rpm in a Type 70.1 Ti rotor (Beckman Coulter Inc.). Two milliliters of the upper fractions were subsequently removed from the ultracentrifuge tubes and collected as the large CM-containing fraction.
Following that, another gradient was formed by carefully layering 2 mL of D-PBS(-) on the lower fractions. The samples were centrifuged for 16 hours at 4℃ at 40,000 rpm. Two milliliters of the upper fractions were collected as a fraction containing VLDL and CM-Rs.
Lipoprotein immunoprecipitation was conducted on fractions containing CM-Rs and VLDL following the instructions provided by the manufacturer of Pureproteome® Protein G Magnetic Beads (MilliporeSigma, Darmstadt, Germany), utilizing 10 µg of mouse anti-apoB48 monoclonal antibodies (Fujirebio Inc., Tokyo, Japan) for the isolation of CM-Rs and 10 µg of mouse anti-apoB100 monoclonal antibodies (Fujirebio Inc.) for the isolation of VLDL-Rs8). Samples were incubated with beads, undergoing slow end-over-end mixing for 16 hours at 4℃. After incubation, the complex of the sample and magnetic beads was washed three times with D-PBS(-). Elution of lipoproteins contained in CM-Rs and VLDL-Rs was performed using a 3M NaSCN solution. Eluted samples were desalted with an Amicon Ultra-0.5 Centrifugal Filter Unit with Ultracel-3 membrane (MilliporeSigma). After desalting, protein concentration was measured using the BCA Protein Assay kit (Thermo Fisher Scientific K.K., MA, US).
4. Biochemical Analysis and Gel-Permeation High-Performance Liquid Chromatography of Lipoproteins
The serum samples for biochemical analysis were promptly frozen at -80℃ until analysis. The lipoprotein profile was analyzed with high sensitivity using gel-permeation high-performance liquid chromatography (GP-HPLC) (LipoSEARCH®; Skylight Biotech, Tokyo, Japan) with monitoring of cholesterol and TG21).
5. Isolation and Cultivation of Mouse Resident Peritoneal Macrophages
Mouse resident peritoneal macrophages were obtained from 12-week-old male C57BL/6J mice fed a normal chow diet. Mice were anesthetized with isoflurane (Wako Pure Chemical Industries, Ltd.) following the guidelines of the Ethics Review Committee for Animal Experimentation of Osaka University School of Medicine. Sterile ice-cold D-PBS(-) (Wako Pure Chemical Industries, Ltd.) was injected into the cavity of each mouse by peritoneal lavage. This fluid was carefully collected and centrifuged at 1,500 rpm for 5 min. After centrifugation, the supernatant was withdrawn, and the cell pellet was resuspended in D-MEM medium (Wako Pure Chemical Industries, Ltd.), containing 100 IU/ml of penicillin, 100 µg/ml of streptomycin, and plated in 4-well culture slides (Corning, NY, US) coated with type I-C collagen (Nitta Gelatin Inc., Osaka, Japan). Cells were incubated in a humidified CO2 (5%) incubator at 37 ℃ for 3 hours to allow adherence. D-MEM with all additions was replaced daily, and macrophages were used within 3 days from harvesting.
6. Evaluation of Foam Cell Formation by Purified CM-Rs
Ten micrograms of isolated CM-Rs protein or D-PBS(-) (as a negative control) were added to the medium in each well of culture slides where mouse peritoneal macrophages were incubated. After a 16-hour incubation, Oil Red O staining was performed to investigate whether CM-Rs could induce the transformation of macrophages into foam cells.
7. Shotgun Proteomics Analysis by LC-MS/MS
Ten micrograms of protein samples were solubilized in 50mM Tris-HCl (pH 9.0) containing 6M urea and 5% sodium deoxycholate, reduced with 10mM dithiothreitol for 60 minutes at 37℃, and alkylated with 55mM iodoacetamide for 30 minutes in the dark at 25℃. The reduced and alkylated samples were diluted 10-fold with 50mM Tris-HCl (pH 9.0) and digested with trypsin at 37℃ for 16 hours (trypsin-to-protein ratio of 1:20 (w/w)). An equal volume of ethyl acetate was added to each sample solution, and the mixtures were acidified with a final concentration of 0.5% trifluoroacetic acid. The mixtures were shaken for 1 minute and centrifuged at 15,700 x g for 2 minutes. Then, the aqueous phase was collected. Digested samples were desalted with C18-StageTips.
LC-MS/MS analysis was conducted using UltiMate 3000 Nano LC systems (Thermo Fisher Scientific) coupled to a Q-Exactive hybrid quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific) with a nano-electrospray ionization source. The digested sample was injected by an autosampler and enriched on a C18 reverse-phase trap column (100 µm I.D. x 5 mm length, Thermo Fisher Scientific) at a flow rate of 4 µL/min. Subsequently, the sample was separated by a C18 reverse-phase column (75 µm I.D. x 150 mm length, Nikkyo Technos Co. Ltd., Tokyo, Japan) at a flow rate of 300 nL/min with a linear gradient from 2% to 35% mobile phase B. Mobile phase B consisted of 95% acetonitrile with 0.1% formic acid, whereas mobile phase A consisted of 2% acetonitrile with 0.1% formic acid. The peptides were ionized using nano-electrospray ionization in positive ion mode22-24).
8. Data Processing and Relative Quantification Based on Spectral Counts
The raw data files were analyzed by Mascot Distiller v2.3 (Matrix Science, London, UK) to create peak lists on the basis of the recorded fragmentation spectra. Peptides and proteins were identified by Mascot v2.3 (Matrix Science) against UniProt database with a precursor mass tolerance of 10 ppm, a fragment ion mass tolerance of 0.01Da and strict trypsin specificity allowing for up to 1 missed cleavage. The carbamidomethylation of cysteine and the oxidation of methionine were allowed as variable modification.
Normalized spectral counts were calculated using Scaffold™ version 5_3_2 (Proteome Software Inc., OR, US). Database search files generated by Mascot were imported into Scaffold. Peptide and protein identifications were accepted if established at ＞95.0% probability, as specified by the Peptide Prophet25) and Protein Prophet algorithms26), respectively. Each identified protein required at least two unique peptides to be part of the dataset. To account for MS/MS sampling differences between individuals, Scaffold outputs were expressed based on the unweighted spectral count assigned to each identified protein. To enhance identification stringency, proteins had to be present in at least 5 of the 8 groups analyzed.
To assess differences in relative protein abundance between CM-Rs and VLDL-Rs, the Spectrum index (SpI) was calculated. The SpI27) is defined as:
SpI
＝
(
S
¯
48
S
¯
48
＋
S
¯
100
×
N
48
D
N
48
T
)
−
(
S
¯
100
S
¯
48
+
S
¯
100
×
N
100
D
N
100
T
)
where
S
¯
48
is the mean spectrum count for a given protein among apoB-48 containing lipoproteins;
S
¯
100
is the mean spectrum count for a given protein among apoB-100 containing lipoproteins;
N
48
D
and
N
100
D
are the numbers of apoB-48 containing lipoproteins and apoB-100 containing lipoproteins in which the proteins of interest are detected;
N
48
T
and
N
100
T
are the total number of samples. Values of the SpI can range from -1 to +1. A value close to 0 indicates that the relative peptide abundance is about equal in CM-Rs and VLDL-Rs. A positive value suggests enrichment of peptides derived from the protein in CM-Rs.
9. Statistical Analysis
All data are presented as mean±SD. Statistical analyses were conducted using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which serves as a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria), or GraphPad Prism Ver.10 software program (GraphPad Software, CA). Two-group comparisons were executed through either Student’s t-test or Mann-Whitney U test, as deemed appropriate. Statistical significance was set at a P value of 0.05.
Results
1. Lipoprotein Profile of Serum and Isolated CM-Rs
Table 1 and Fig.1 display the fundamental characteristics of 8 healthy volunteers and their postprandial lipid profiles assessed using LipoSEARCH®. Postprandial total cholesterol and triglyceride (TG) levels were 175.5±27.6 mg/dL and 125.5±82.2 mg/dL, respectively. Individual lipoprotein profiles of cholesterol and TG monitoring are depicted in Fig.1A and 1B. The VLDL fraction exhibited abundant TG levels, indicating the presence of significant remnant lipoproteins in these postprandial sera. CM-Rs were isolated from this VLDL fraction using a combination of ultracentrifugation and immunoprecipitation with anti-apoB-48 specific antibodies. The lipoprotein profiles of the isolated CM-Rs are presented in Fig.1C and 1D. In comparison to the lipoprotein profile of postprandial serum, the particle size of isolated lipoproteins varied, ranging from small CMs to large LDL, suggesting a wide distribution of CM-Rs.
2. Evaluation of the Atherogenicity of Isolated CM-Rs
The uptake of isolated CM-Rs by mouse resident peritoneal macrophages is illustrated in Fig.2. Mouse resident peritoneal macrophages incubated with CM-Rs (Fig.2A) displayed larger areas of Oil Red O staining-positive lipid droplets compared to the control (Fig.2B). These findings suggest that CM-Rs are taken up by macrophages, promoting foam cell formation, which is consistent with a previous report28).
3. Enumeration of CM-Rs Associated Proteins
We applied stringent criteria for the identification of CM-Rs or VLDL-Rs-associated proteins to ensure specificity. Initially, we considered 7 keratin (KRT) proteins (KRT1, KRT2, KRT5, KRT6B, KRT9, KRT10, KRT77) as potential contaminations18). Subsequently, we identified 42 CM-Rs or VLDL-Rs-associated proteins in the current study (Table 2). Among these, 11 proteins such as platelet basic protein (PPBP), platelet factor 4 variant (PF4V1), hornerin (HRNR), platelet factor 4 (PF4), calmodulin-like protein 5 (CALML5), suprabasin (SBSN), fatty acid-binding protein 5 (FABP5), filaggrin-2 (FLG2), angiogenin (ANG), prothrombin (F2), and shugoshin-like 1 (SGOL1) might be newly identified CM-Rs or VLDL-Rs-associated proteins. This is inferred from their absence in previous VLDL proteomic studies29-35). Notably, these proteins, excluding SGOL1, are also listed as HDL-associated proteins in the HDL Proteome Watch18), suggesting that the identified proteins are indeed part of the cargo of lipoproteins.
Table 2.CM-R and VLDL-R associated proteins identified in this study
| Protein name |
Protein short name |
CM-R (n = 8)
|
VLDL-R (n = 8)
|
p value
|
| Apolipoprotein B |
APOB |
145.4 |
(8/8) |
252.3 |
(8/8) |
0.038
|
| Apolipoprotein E |
APOE |
134.6 |
(8/8) |
105.4 |
(8/8) |
0.105 |
| Apolipoprotein C-III |
APOC3 |
35.0 |
(8/8) |
35.0 |
(8/8) |
0.959 |
| Apolipoprotein C-II |
APOC2 |
25.2 |
(8/8) |
28.7 |
(8/8) |
0.574 |
| Apolipoprotein D |
APOD |
19.2 |
(8/8) |
18.7 |
(8/8) |
0.959 |
| Apolipoprotein A-I |
APOA1 |
21.2 |
(8/8) |
12.5 |
(5/8) |
0.049
|
| Apolipoprotein A-IV |
APOA4 |
24.5 |
(8/8) |
11.5 |
(8/8) |
＜0.001
|
| Platelet basic protein |
PPBP |
12.6 |
(8/8) |
11.1 |
(8/8) |
0.645 |
| albumin |
ALB |
9.3 |
(8/8) |
7.6 |
(8/8) |
0.442 |
| Fibrinogen alpha chain |
FGA |
11.8 |
(8/8) |
3.4 |
(4/8) |
0.010
|
| Apolipoprotein C-IV |
APOC4 |
11.3 |
(8/8) |
6.5 |
(8/8) |
0.105 |
| Clusterin |
CLU |
9.3 |
(8/8) |
3.2 |
(7/8) |
＜0.001
|
| Platelet factor 4 variant |
PF4V1 |
7.6 |
(8/8) |
5.7 |
(8/8) |
0.279 |
| Apolipoprotein A-II |
APOA2 |
7.0 |
(8/8) |
3.8 |
(5/8) |
0.064 |
| Prenylcysteine oxidase 1 |
PCYOX1 |
5.5 |
(7/8) |
6.2 |
(6/8) |
0.721 |
| Hornerin |
HRNR |
5.9 |
(8/8) |
3.9 |
(4/8) |
0.362 |
| Histidine-rich glycoprotein |
HRG |
5.1 |
(6/8) |
1.8 |
(2/8) |
0.102 |
| Transthyretin |
TTR |
6.2 |
(5/8) |
0.2 |
(1/8) |
0.026
|
| Apolipoprotein L1 |
APOL1 |
4.7 |
(8/8) |
3.6 |
(7/8) |
0.442 |
| Apolipoprotein C-I |
APOC1 |
4.8 |
(8/8) |
1.4 |
(6/8) |
＜0.001
|
| Platelet factor 4 |
PF4 |
11.1 |
(8/8) |
6.8 |
(6/8) |
0.125 |
| Apolipoprotein M |
APOM |
4.6 |
(8/8) |
1.1 |
(5/8) |
0.005
|
| Calmodulin-like protein 5 |
CALML5 |
3.3 |
(7/8) |
2.1 |
(5/8) |
0.203 |
| Apolipoprotein A-V |
APOA5 |
4.2 |
(6/8) |
0.1 |
(1/8) |
0.013
|
| suprabasin |
SBSN |
3.0 |
(8/8) |
1.2 |
(4/8) |
0.013
|
| paraoxonase/arylesterase 1 |
PON1 |
3.8 |
(7/8) |
0.3 |
(2/8) |
0.002
|
| Vitronectin |
VTN |
2.5 |
(7/8) |
1.2 |
(2/8) |
0.041
|
| Complement C3 (Fragment) |
C3 |
2.4 |
(8/8) |
1.6 |
(7/8) |
0.130 |
| Fatty acid-binding protein 5 |
FABP5 |
2.0 |
(8/8) |
1.6 |
(6/8) |
0.697 |
| Prolactin-inducible protein |
PIP |
2.1 |
(8/8) |
1.8 |
(7/8) |
0.721 |
| Dermcidin |
DCD |
2.4 |
(8/8) |
0.9 |
(4/8) |
0.074 |
| Beta-2-glycoprotein 1 |
APOH |
1.9 |
(5/8) |
1.1 |
(2/8) |
0.236 |
| Filaggrin-2 |
FLG2 |
1.4 |
(7/8) |
1.2 |
(5/8) |
0.203 |
| Angiogenin |
ANG |
1.6 |
(7/8) |
0.6 |
(2/8) |
0.041
|
| Desmoplakin |
DSP |
0.7 |
(4/8) |
1.9 |
(5/8) |
0.143 |
| Selenoprotein P |
SEPP1 |
1.6 |
(6/8) |
0.4 |
(2/8) |
0.049
|
| Prothrombin |
F2 |
1.5 |
(7/8) |
0.2 |
(1/8) |
0.005
|
| Cathelicidin antimicrobial peptide precursor |
CAMP |
1.2 |
(6/8) |
0.7 |
(4/8) |
0.217 |
| Serum amyloid A protein |
SAA2 |
1.3 |
(5/8) |
1.2 |
(4/8) |
0.557 |
| Protein S100-A8 |
S100A8 |
0.5 |
(5/8) |
0.5 |
(3/8) |
0.957 |
| Shugoshin-like 1 |
SGOL1 |
0.7 |
(5/8) |
0.3 |
(2/8) |
0.236 |
| Lysozyme C |
LYZ |
0.7 |
(6/8) |
0.1 |
(1/8) |
0.035
|
The CM-R and VLDL-R proteome was analyzed by LC-MS/MS, and data were analyzed by searching the Uniprot database with Mascot 2.3 (MatrixScience). Values represent the mean spectral counts of peptides. The number of subjects with the protein identified is noted in parentheses. Statistical significance was calculated with the Mann-Whitney U test.
We applied an equal amount of CM-Rs or VLDL-Rs proteins (10 µg for each sample) to LC-MS/MS to assess the relative protein abundance between the two groups, employing spectral counting (Table 2). From these data, we generated a volcano plot (Fig.3A) and a spectral index for each protein (Fig.3B), revealing a substantial difference in the proteome of CM-Rs compared to that of VLDL-Rs. When comparing equal weights of proteins, the spectral counting of apoB was significantly lower in CM-Rs, consistent with the fact that one CM-R particle contains a lighter apoB-48 molecule (264 kDa), while one VLDL-R particle carries a heavier apoB-100 molecule (540 kDa)19). On the other hand, HDL-associated proteins such as apoA-I, ApoA-IV, apoA-V, apoM, clusterin (CLU), and paraoxonase/arylesterase 1 (PON1) were significantly more abundant in CM-Rs, possibly reflecting their origin from small intestine-derived HDL as well as their attachment during chylomicron formation process.
We also identified proteins involved in the complement pathway and regulation, such as complement C3 and vitronectin (VTN), as well as those associated with acute-phase response, including PPBP, serum amyloid A protein 2 (SAA2), and protein S100-A8 (S100A8), in both CM-Rs and VLDL-Rs. These proteins may contribute to the atherogenicity of these remnant lipoproteins. Additionally, apoC-I, implicated in the metabolism of triglyceride-rich lipoproteins (TGLs) by inhibiting their binding to VLDL receptor (VLDL-R), low-density lipoprotein receptor (LDL-R), and LDL receptor-related protein (LRP)36), was found to be more abundant in CM-Rs compared to VLDL-Rs. Furthermore, F2, associated with acute-phase response and hemostasis17), was also more abundant in CM-Rs. These findings may collectively contribute to the atherogenicity of CM-Rs in addition to that of VLDL-Rs.
Discussion
It has been exceptionally challenging to isolate CM-Rs from other lipoproteins, such as apoB-100-containing lipoproteins (VLDL, IDL, and LDL), to individually evaluate their characteristics and atherogenicity. There are likely two main reasons for this difficulty. The first reason may stem from the specific gravity of remnant lipoproteins. Since the method developed by Gofman et al.37), lipoprotein separation has been achieved through ultracentrifugation. In this approach, fractionation is based on the differences in specific gravity among lipoproteins, which are subject to fluctuations due to the action of various enzymes and lipid transfer proteins such as lipoprotein lipase (LPL), hepatic lipase (HL), cholesteryl ester transfer protein (CETP), and others. This fluctuation is particularly pronounced in triglyceride-rich lipoproteins like CMs and VLDLs. ApoB-100-containing lipoproteins are distributed across a relatively limited range of particle sizes, including VLDL, IDL, LDL, and small dense LDL, with a significantly larger number of particles, which makes their analysis relatively straightforward. On the other hand, although CMs are originally larger than VLDL, during lipolysis, CM-Rs progressively shrink, sometimes reaching the size of LDL or even large HDL particles38), indicating the broad distribution of apoB-48-containing lipoproteins in each lipoprotein fraction38). Consequently, isolating apoB-48-containing CM-Rs and apoB-100-containing VLDL-Rs separately from serum using only the ultracentrifugation method proves to be challenging.
Furthermore, the second reason for the difficulty in separating CM-Rs from VLDL-Rs may be the challenge in discriminating apoB isoforms. Serum apoB circulates in two distinct isoforms: apoB-100 and apoB48. Human apoB-48 is exclusively synthesized in the small intestines, while apoB-100 is produced in the liver. These two forms of apoB are encoded by a single APOB gene from a single mRNA and share a common N-terminal sequence39). Due to the identical amino acid sequence of apoB-48 and apoB-100, developing apoB-48 specific antibodies that do not recognize apoB-100 posed an extremely challenging task. As mentioned earlier, we successfully established specific monoclonal antibodies against human apoB-48 8). In the present study, we employed a combination of immunoprecipitation and ultracentrifugation, utilizing monoclonal apoB-48 antibodies affixed to magnetic beads. These monoclonal antibodies selectively recognize only the C-terminus of human apoB-48, exhibiting no reactivity to human apoB-100. Since a single copy of apoB-48 per particle is present on the CM-Rs40), and apoB-100 is exclusive to VLDL-Rs in humans, we were able, for the first time, to successfully isolate pure CM-Rs and VLDL-Rs from human serum and analyze each proteome.
Remnant lipoproteins have been reported to exhibit a variety of proatherogenic properties and functions7). These lipoproteins can penetrate the intima of the arterial wall, inducing inflammation by upregulating the expression of interleukin (IL)-1b, monocyte chemoattractant protein (MCP)-1, CD40, early growth response factor (Egr)-1, and other factors. Additionally, remnant lipoproteins have the capacity to transform macrophages into foam cells, instigate endothelial dysfunction by reducing nitric oxide (NO) production, and induce apoptosis in endothelial cells. They also accelerate the cell proliferation of smooth muscle cells and enhance the production of plasminogen activator inhibitor (PAI)-1 from endothelial cells. However, the extent of the contribution of CM-Rs and VLDL-Rs to atherogenesis has been a subject of controversy.
In the current study, we analyzed the proteome of CM-Rs as apoB-48-containing lipoproteins and VLDL-Rs as apoB-100-containing lipoproteins. We identified 42 CM-Rs or VLDL-Rs-associated proteins, including 11 newly identified proteins. Among these, HRNR, CALML5, FLG2, ANG, and SGOL1 may be potential contaminations as they are predominantly located in the cytoplasm or extracellular matrix. Conversely, PPBP, PF4V1, PF4, SBSN, FABP5, and F2 are primarily found in circulation. These 6 proteins are also listed as HDL-associated proteins in the HDL Proteome Watch18), suggesting their presence in the cargo of CM-Rs or VLDL-Rs. Notably, PPBP and PF4 are chemokines secreted from platelets and have been reported to be present in atherosclerotic lesions41). Although direct evidence for the proatherogenic role of PPBP is limited, a causal role of PF4 in atherogenesis was suggested by reduced lesion sizes in PF4 and Apoe double-deficient mice42). PPBP and PF4 may contribute to the atherogenic properties of CM-Rs or VLDL-Rs.
Since we applied the same weight of CM-Rs and VLDL-Rs proteins to LC-MS/MS, the spectral counting of apoB was significantly lower in CM-Rs, which has one apoB-48 molecule per particle, compared to VLDL-Rs. Conversely, CM-Rs contain more HDL-associated proteins such as apoA-I, apoA-IV, apoA-V, and PON1, likely originating from small intestine-derived HDL. Concerning other proteins, apoC-I has been reported to be predominantly associated with HDL and VLDL in humans so far36). In this study, we observed that apoC-I was more abundant in CM-Rs compared to VLDL-Rs. While the impact of apoC-I on atherosclerosis remains a subject of debate, it is known that apoC-I reduces the activity of lipoprotein lipase (LPL) and stimulates VLDL production, resulting in increased plasma triglyceride levels. Additionally, apoC-I directly binds lipopolysaccharide (LPS) and facilitates LPS presentation to innate immune receptors on the surface of macrophages, thereby enhancing the inflammatory response43). Thus, our findings may provide an explanation for the atherogenic properties of CM-Rs.
There are several limitations to this study. First, due to the different molecular weights of apoB-48 and apoB-100, we were unable to directly compare the amounts of CM-R or VLDL-R associated proteins per particle. Nevertheless, we successfully characterized the proteome of CM-Rs for the first time. Second, due to the limited quantity of isolated lipoproteins, we could not perform western blotting to validate the quantification. However, the mean spectral counts of apoB in CM-Rs and VLDL-Rs were considered relevant given their molecular weights, suggesting that the quantitativeness in this study appears reasonably valid. Third, this study involved blood sampling from healthy volunteers 3 hours after a high-fat-containing meal. We did not perform a time course analysis due to column availability constraints. However, there is a possibility that the proteome differs between fasting, early postprandial, and late postprandial stages. Additionally, a previous study showed that apoC-III is recognized as an exchangeable component among lipoproteins. Specifically, in patients with hypertriglyceridemia, apoC-III is bound to triglyceride-rich lipoproteins, whereas it is bound to HDL in normolipidemic subjects44). Thus, the CM-Rs proteome of patients with dyslipidemia or metabolic syndrome may differ from that of healthy subjects, warranting investigation in future studies.
Conflict of Interest
FUJIREBIO INC. shared the cost for the apoB-48 measurements in the context of a joint research collaboration.
Acknowledgement
We express our gratitude to Takuya Kobayashi, Masumi Asaji, Ayami Saga, and Kyoko Ozawa for their outstanding administrative and technical support. This study received support from the Center for Medical Research and Education, Graduate School of Medicine, Osaka University.
References
- 1) Goldberg IJ, Eckel RH, McPherson R: Triglycerides and heart disease: still a hypothesis? Arterioscler Thromb Vasc Biol, 2011; 31: 1716-1725
- 2) Hirata A, Okamura T, Sugiyama D, Kuwabara K, Kadota A, Fujiyoshi A, Miura K, Okuda N, Ohkubo T, Okayama A, Ueshima H, Group NDR: The Relationship between Very High Levels of Serum High-Density Lipoprotein Cholesterol and Cause-Specific Mortality in a 20-Year Follow-Up Study of Japanese General Population. J Atheroscler Thromb, 2016; 23: 800-809
- 3) Murad MH, Hazem A, Coto-Yglesias F, Dzyubak S, Gupta S, Bancos I, Lane MA, Erwin PJ, Berglund L, Elraiyah T, Montori VM: The association of hypertriglyceridemia with cardiovascular events and pancreatitis: a systematic review and meta-analysis. BMC Endocr Disord, 2012; 12: 2
- 4) Cohn JS, Marcoux C, Davignon J: Detection, quantification, and characterization of potentially atherogenic triglyceride-rich remnant lipoproteins. Arterioscler Thromb Vasc Biol, 1999; 19: 2474-2486
- 5) Masuda D, Nishida M, Arai T, Hanada H, Yoshida H, Yamauchi-Takihara K, Moriyama T, Tada N, Yamashita S: Reference interval for the apolipoprotein B-48 concentration in healthy Japanese individuals. J Atheroscler Thromb, 2014; 21: 618-627
- 6) Masuda D, Yamashita S: Postprandial Hyperlipidemia and Remnant Lipoproteins. Journal of Atherosclerosis and Thrombosis, 2017; 24: 95-109
- 7) Fujioka Y, Ishikawa Y: Remnant lipoproteins as strong key particles to atherogenesis. J Atheroscler Thromb, 2009; 16: 145-154
- 8) Uchida Y, Kurano Y, Ito S: Establishment of monoclonal antibody against human Apo B-48 and measurement of Apo B-48 in serum by ELISA method. Journal of Clinical Laboratory Analysis, 1998; 12: 289-292
- 9) Hanada H, Mugii S, Okubo M, Maeda I, Kuwayama K, Hidaka Y, Kitazume-Taneike R, Yamashita T, Kawase R, Nakaoka H, Inagaki M, Yuasa-Kawase M, Nakatani K, Tsubakio-Yamamoto K, Masuda D, Ohama T, Matsuyama A, Ishigami M, Nishida M, Komuro I, Yamashita S: Establishment of chemiluminescence enzyme immunoassay for apolipoprotein B-48 and its clinical applications for evaluation of impaired chylomicron remnant metabolism. Clinica Chimica Acta, 2012; 413: 160-165
- 10) Sato I, Ishikawa Y, Ishimoto A, Katsura S, Toyokawa A, Hayashi F, Kawano S, Fujioka Y, Yamashita S, Kumagai S: Significance of measuring serum concentrations of remnant lipoproteins and apolipoprotein B-48 in fasting period. J Atheroscler Thromb, 2009; 16: 12-20
- 11) Masuda D, Sakai N, Sugimoto T, Kitazume-Taneike R, Yamashita T, Kawase R, Nakaoka H, Inagaki M, Nakatani K, Yuasa-Kawase M, Tsubakio-Yamamoto K, Ohama T, Nakagawa-Toyama Y, Nishida M, Ishigami M, Masuda Y, Matsuyama A, Komuro I, Yamashita S: Fasting serum apolipoprotein B-48 can be a marker of postprandial hyperlipidemia. J Atheroscler Thromb, 2011; 18: 1062-1070
- 12) Kinoshita M, Ohnishi H, Maeda T, Yoshimura N, Takeoka Y, Yasuda D, Kusano J, Mashimo Y, Saito S, Shimamoto K, Teramoto T: Increased serum apolipoprotein B48 concentration in patients with metabolic syndrome. J Atheroscler Thromb, 2009; 16: 517-522
- 13) Okubo M, Hanada H, Matsui M, Hidaka Y, Masuda D, Sakata Y, Yamashita S: Serum apolipoprotein B-48 concentration is associated with a reduced estimated glomerular filtration rate and increased proteinuria. Journal of atherosclerosis and thrombosis, 2014; 21: 974-982
- 14) Mugii S, Hanada H, Takeoka K, Hidaka Y, Masuda D, Ohama T, Toyama Y, Yamashita S: [Clinical significance of apolipoprotein B-48 (apoB-48) in patients with thyroid disease]. Rinsho Byori, 2009; 57: 1058-1063
- 15) Nakatani K, Sugimoto T, Masuda D, Okano R, Oya T, Monden Y, Yamashita T, Kawase R, Nakaoka H, Inagaki M, Yuasa-Kawase M, Tsubakio-Yamamoto K, Ohama T, Nishida M, Ishigami M, Komuro I, Yamashita S: Serum apolipoprotein B-48 levels are correlated with carotid intima-media thickness in subjects with normal serum triglyceride levels. Atherosclerosis, 2011; 218: 226-232
- 16) Proctor SD, Mamo JC: Intimal retention of cholesterol derived from apolipoprotein B100- and apolipoprotein B48-containing lipoproteins in carotid arteries of Watanabe heritable hyperlipidemic rabbits. Arterioscler Thromb Vasc Biol, 2003; 23: 1595-1600
- 17) Shah AS, Tan L, Long JL, Davidson WS: Proteomic diversity of high density lipoproteins: our emerging understanding of its importance in lipid transport and beyond. Journal of lipid research, 2013; 54: 2575-2585
- 18) Davidson WS, Shah AS, Sexmith H, Gordon SM: The HDL Proteome Watch: Compilation of studies leads to new insights on HDL function. Biochim Biophys Acta Mol Cell Biol Lipids, 2022; 1867: 159072
- 19) Gianazza E, Zoanni B, Mallia A, Brioschi M, Colombo GI, Banfi C: Proteomic studies on apoB-containing lipoprotein in cardiovascular research: A comprehensive review. Mass Spectrom Rev, 2023; 42: 1397-1423
- 20) Gofman JW, Delalla O, Glazier F, Freeman NK, Lindgren FT, Nichols AV, Strisower B, Tamplin AR: The serum lipoprotein transport system in health, metabolic disorders, atherosclerosis and coronary heart disease. J Clin Lipidol, 2007; 1: 104-141
- 21) Toshima G, Iwama Y, Kimura F, Matsumoto Y, Miura M: LipoSEARCH®; Analytical GP-HPLC method for lipoprotein profiling and its applications. J Biol Macromal, 2013; 13: 21-32
- 22) Okada T, Ohama T, Takafuji K, Kanno K, Matsuda H, Sairyo M, Zhu Y, Saga A, Kobayashi T, Masuda D, Koseki M, Nishida M, Sakata Y, Yamashita S: Shotgun proteomic analysis reveals proteome alterations in HDL of patients with cholesteryl ester transfer protein deficiency. Journal of clinical lipidology, 2019; 13: 317-325
- 23) Rappsilber J, Mann M, Ishihama Y: Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips. Nature protocols, 2007; 2: 1896-1906
- 24) Masuda T, Tomita M, Ishihama Y: Phase transfer surfactant-aided trypsin digestion for membrane proteome analysis. Journal of Proteome Research, 2008; 7: 731-740
- 25) Keller A, Nesvizhskii AI, Kolker E, Aebersold R: Empirical Statistical Model To Estimate the Accuracy of Peptide Identifications Made by MS / MS and Database Search. Analytical chemistry, 2002; 74: 5383-5392
- 26) Nesvizhskii AI, Keller A, Kolker E, Aebersold R: A statistical model for identifying proteins by tandem mass spectrometry. Analytical chemistry, 2003; 75: 4646-4658
- 27) Fu X, Gharib Sa, Green PS, Aitken ML, Frazer Da, Park DR, Vaisar T, Heinecke JW: Spectral index for assessment of differential protein expression in shotgun proteomics. Journal of proteome research, 2008; 7: 845-854
- 28) Yoshimura N, Kinoshita M, Teramoto T: Isolation and characterization of apolipoprotein B48-containing lipoproteins with a monoclonal antibody against apolipoprotein B48. J Atheroscler Thromb, 2009; 16: 740-747
- 29) Mancone C, Amicone L, Fimia GM, Bravo E, Piacentini M, Tripodi M, Alonzi T: Proteomic analysis of human very low-density lipoprotein by two-dimensional gel electrophoresis and MALDI-TOF/TOF. Proteomics, 2007; 7: 143-154
- 30) Banfi C, Brioschi M, Barcella S, Wait R, Begum S, Galli S, Rizzi A, Tremoli E: Proteomic analysis of human low-density lipoprotein reveals the presence of prenylcysteine lyase, a hydrogen peroxide-generating enzyme. Proteomics, 2009; 9: 1344-1352
- 31) Hoofnagle AN, Heinecke JW: Lipoproteomics: using mass spectrometry-based proteomics to explore the assembly, structure, and function of lipoproteins. Journal of lipid research, 2009; 50: 1967-1975
- 32) Sun HY, Chen SF, Lai MD, Chang TT, Chen TL, Li PY, Shieh DB, Young KC: Comparative proteomic profiling of plasma very-low-density and low-density lipoproteins. Clin Chim Acta, 2010; 411: 336-344
- 33) Lepedda AJ, Nieddu G, Zinellu E, De Muro P, Piredda F, Guarino A, Spirito R, Carta F, Turrini F, Formato M: Proteomic analysis of plasma-purified VLDL, LDL, and HDL fractions from atherosclerotic patients undergoing carotid endarterectomy: identification of serum amyloid A as a potential marker. Oxid Med Cell Longev, 2013; 2013: 385214
- 34) Savinova OV, Fillaus K, Jing L, Harris WS, Shearer GC: Reduced apolipoprotein glycosylation in patients with the metabolic syndrome. PLoS One, 2014; 9: e104833
- 35) Dashty M, Motazacker MM, Levels J, de Vries M, Mahmoudi M, Peppelenbosch MP, Rezaee F: Proteome of human plasma very low-density lipoprotein and low-density lipoprotein exhibits a link with coagulation and lipid metabolism. Thromb Haemost, 2014; 111: 518-530
- 36) Rouland A, Masson D, Lagrost L, Verges B, Gautier T, Bouillet B: Role of apolipoprotein C1 in lipoprotein metabolism, atherosclerosis and diabetes: a systematic review. Cardiovasc Diabetol, 2022; 21: 272
- 37) Gofman JW, Lindgren FT, Elliott H: Ultracentrifugal Studies of Lipoproteins of Human Serum. Journal of Biological Chemistry, 1949; 179: 973-979
- 38) Masuda D, Nakagawa-Toyama Y, Nakatani K, Inagaki M, Tsubakio-Yamamoto K, Sandoval JC, Ohama T, Nishida M, Ishigami M, Yamashita S: Ezetimibe improves postprandial hyperlipidaemia in patients with type IIb hyperlipidaemia. Eur J Clin Invest, 2009; 39: 689-698
- 39) Greeve J: Inhibition of the synthesis of apolipoprotein B-containing lipoproteins. Handb Exp Pharmacol, 2005: 483-517
- 40) Phillips ML, Pullinger C, Kroes I, Kroes J, Hardman DA, Chen G, Curtiss LK, Gutierrez MM, Kane JP, Schumaker VN: A single copy of apolipoprotein B-48 is present on the human chylomicron remnant. Journal of Lipid Research, 1997; 38: 1170-1177
- 41) Gleissner CA, von Hundelshausen P, Ley K: Platelet chemokines in vascular disease. Arterioscler Thromb Vasc Biol, 2008; 28: 1920-1927
- 42) Sachais B, Turrentine T, Dawicki McKenna J, Rux A, Rader D, Kowalska A: Elimination of platelet factor 4 (PF4) from platelets reduces atherosclerosis in C57Bl/6 and apoE-/- mice. Thrombosis and Haemostasis, 2007; 98: 1108-1113
- 43) Berbee JF, Coomans CP, Westerterp M, Romijn JA, Havekes LM, Rensen PC: Apolipoprotein CI enhances the biological response to LPS via the CD14/TLR4 pathway by LPS-binding elements in both its N- and C-terminal helix. J Lipid Res, 2010; 51: 1943-1952
- 44) Fredenrich A, Giroux LM, Tremblay M, Krimbou L, Davignon J, Cohn JS: Plasma lipoprotein distribution of apoC-III in normolipidemic and hypertriglyceridemic subjects: comparison of the apoC-III to apoE ratio in different lipoprotein fractions. J Lipid Research, 1997; 38: 1421-1432
