Skip Navigation

Cardiovascular Research 2003 59(1):169-180; doi:10.1016/S0008-6363(03)00335-3
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Norata, G.D.
Right arrow Articles by Eriksson, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Norata, G.D.
Right arrow Articles by Eriksson, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Gene expression and intracellular pathways involved in endothelial dysfunction induced by VLDL and oxidised VLDL

G.D. Norataa, A. Pirilloa, E. Callegaria, A. Hamstenb, A.L. Catapanoa,c and P. Erikssonb,*

aDepartment of Pharmacological Sciences, University of Milan, Milan, Italy
bKing Gustaf V Research Institute, Department of Medicine, Karolinska Hospital M1, Karolinska Institute, Stockholm 171 76, Sweden
cCentro per lo Studio e la Prevenzione delle Vasculopatie Periferiche, Ospedale Bassini, Cinisello Balsamo, Italy

per.eriksson{at}medks.ki.se

* Corresponding author. Tel.: +46-8-5177-3202; fax: +46-8-311-298.

Received 1 November 2002; accepted 24 February 2003


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objectives: The molecular mechanisms underlying the relationship between elevated plasma concentrations of triglyceride-rich lipoproteins and coronary artery disease remain uncertain. In the present work, we investigated the gene expression pattern and intracellular pathways in human endothelial cells incubated with very low density lipoproteins (VLDL). Moreover, as VLDL can enter the arterial wall and undergo oxidative modification, we compared the VLDL-induced expression pattern with the one of oxidised VLDL (Ox-VLDL). Methods: Total RNA from endothelial cells incubated with 75 µg/ml VLDL or Ox-VLDL and total RNA from endothelial cells under basal conditions were hybridised to identical microarrays containing 8411 genes. Seven clusters of expression profiles were identified. This pattern was validated by quantitative real-time PCR of selected genes. The intracellular pathway involved in VLDL or Ox-VLDL mediated endothelial responses were also investigated. Results and conclusion: VLDL predominantly activated the ERK1/2 pathway while P38 MAPK was the main target of Ox-VLDL. CREB and NF-KB were activated by both VLDL and Ox-VLDL. Real-time PCR demonstrated that VLDL induced matrix metalloproteinase-2 (5.47±1.74 fold), CD38 (2.38±0.23) and transforming growth factor-{alpha} (2.51±0.30) expression. Ox-VLDL was found to induce interleukin-15 (2.10±0.48) and macrophage migration inhibitory factor (3.19±0.07) expression. In addition, several genes implicated in endothelial cell activation and damage/proliferation were identified by the array analysis. Ox-VLDL was found to promote the generation of reactive oxygen species and exert a cytotoxic effect, while VLDL lacks these effects. These findings confirm the involvement of VLDL and Ox-VLDL in endothelial dysfunction and suggest new genes and molecular mechanisms involved in these actions.

KEYWORDS Endothelial cells; Gene expression; VLDL; Lipoprotein; Oxidised


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
The vascular endothelium is the primary site of dysfunction in many diseases, particularly cardiovascular disease (CVD). Triglyceride-rich lipoproteins, including VLDL, chylomicrons and their remnants, are recognised as CVD risk factors and have been shown to impair vasorelaxation [1,2]. Plasma concentrations of triglyceride-rich lipoproteins are more strongly related to thrombogenic conditions than is low density lipoprotein (LDL) [3] and the concentrations of soluble adhesion molecules are higher in individuals with hypertriglyceridaemia than in hypercholesterolaemic subjects [4]. Studies using HepG2 cells have investigated the intracellular signalling pathway induced by VLDL exposure [5]. VLDL seems to induce protein kinase C activity, resulting in activation of mitogen-activated protein kinase. Studies carried out in endothelial cells (ECs) indicate that VLDL can also activate nuclear factor-kB (NF-kB) [6], a transcription factor that plays an important role in the phenotypic modulation of ECs to a pro-inflammatory condition. Whether oxidation of VLDL augments or changes its effect on endothelial function is unknown. Exposure of β-VLDL to endothelial cells causes oxidation of this lipoprotein species [7,8] resulting in a two- to three-fold increased degradation by mouse peritoneal macrophages [7] and rabbit smooth muscle cells [8] compared with unoxidised β-VLDL. Isolated human VLDL is effectively oxidised in vitro on incubation with free radicals [9]. Ox-VLDL causes greater accumulation of cholesteryl ester in J774 macrophages than Ox-LDL [10] and induces the expression of monocyte chemoattractant protein-1 in rabbit peritoneal macrophages [11].

Candidate gene studies performed so far have been limited to a few genes. The recent development of gene array technology offers new opportunities to evaluate many genes at the same time. Evaluation of gene expression patterns could lead to a better understanding of the vascular endothelial dysfunction caused by triglyceride-rich lipoproteins [12]. In this study we evaluated the gene expression pattern of human endothelial cells incubated with VLDL and Ox-VLDL and investigated the intracellular pathways involved.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1 Lipoproteins
VLDL was isolated from fresh plasma from normolipidaemic subjects by preparative ultracentrifugation [13]. All subjects gave informed consent to participate in the study. The purity of the fractions separated was confirmed by gel electrophoresis; after staining with Oil Red O, the lane corresponding to the top fraction showed only one band in the pre-beta area (VLDL) while the bottom fraction showed one band in the beta area (LDL) and one band in the alpha area (HDL). The top fraction containing VLDL was dialysed in PBS containing 0.01% EDTA and the protein content was determined by the Lowry method [14]. VLDL (0.2 mg protein/ml) was oxidised with 20 µM CuSO4 for 24 h at 37°C. The oxidation was blocked by the addition of 40 µM butylated hydroxytoluene (BHT). The TBARS (thiobarbituric acid reactive substances) value of VLDL was 1±0.2 nmol/mg protein, whereas the TBARS value of Ox-VLDL was 45±3 nmol/mg protein. Native VLDL and Ox-VLDL were used within 8 h of preparation. Of note, addition of CuSO4 and BHT does not affect cell viability and gene expression [15].

2.2 Cell culture
The EAhy926 line [16], a hybridoma cell line with the characteristics of human endothelial cells, which is widely used as a model of macrovascular endothelium, was cultured under standard conditions in MEM+10% FCS and 1% HAT (5x10–3 M hypoxanthine, 2x10–5 M aminopterin, 8x10–4 M thymidine). Human umbilical vein endothelial cells (HUVEC) were isolated according to established procedures [17], cultured under standard conditions in medium M-199 containing 20% FCS (fetal calf serum), heparin (15 U/ml) and ECGF (endothelial cell growth factor, 20 µg/ml) (Roche, Italy) and used within the 4th passage as a sparse culture.

2.3 Microarray analysis
Total RNA extraction was performed using RNAwiz (Ambion, TX, USA) according to the instructions given by the manufacturer. The samples were treated with DNase I according to the DNA free protocol (Ambion, TX, USA). Ten to 20 µg of Total RNA from EAhy926 were used in each RT reaction. The cDNA synthesis was primed using oligo-dT primers. [{alpha}-33P]dATP-labelled cDNA was prepared using the Strip-EZ RT kit (Ambion, TX, USA) and hybridised to the Atlas Plastic Human 8 k microarray (Clontech Laboratories Inc. Palo Alto, CA, USA) following the instructions of the manufacturer. The microarrays were exposed on a phosphor plate (Fuji BAS 2040, Fujifilm, Tokyo, Japan) for 6–16 h and quantified on a BAS 2500 Bio-Imaging Analyzer (Fujifilm).

2.4 Data analysis
The images were imported into the Array Gauge version 1.2 program (Fujifilm). Intensities were calculated by subtracting the local background of each spot and normalised using a global normalisation procedure when comparing the different arrays. Only genes with average normalised intensity of 0.1 (10 times higher than the threshold of detection, i.e. 0.01) or above were studied. The expression ratios reported are the average from four separate experiments (genes that showed a standard deviation higher that 35% of the average were discarded). Using these criteria 1620 genes out of 2213 genes detected on the array remained for further analysis.

2.5 Real-time polymerase chain reaction
Twenty nanograms of mRNA from each sample were reverse transcribed using superscript II according to the manufacturer's manual (Invitrogen, USA). Two microlitres of cDNA were amplified by real-time PCR with 1x TaqMan universal PCR mastermix (Applied Biosystem, USA), 200 µM of each primer and 1.25 pM of probe. The primers (Interactiva, and Operon-Qiagen, Germany) and probes (Applied Biosystem) used are specified in Table 1. β-actin was used as a housekeeping gene to normalise for RNA loading. Each sample was analysed in duplicates using ABI prism 7000 (Applied Biosystem). The PCR amplification was related to a standard curve.


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

  		Table 1 Primers and probes sequences

 
2.6 Antibodies and immunoblotting
The rabbit polyclonal phospho-p38 MAPK antibody specific for dual-phosphorylated 180Thr and 182Tyr of p38 MAPK, the mouse monoclonal phospho-p44/42 MAPK antibody specific for dual phosphorylated 202Thr and 204Tyr of p44 and p42 MAP kinases, the rabbit polyclonal phospho-IkB-alpha antibody specific for phosphorylated 32Ser of Ik-B alpha and the rabbit polyclonal phospho-CREB antibody specific for phosphorylated 133Ser of CREB were from New England Biolabs, (Germany). The monoclonal antibody against β-actin was from Sigma. Peroxidase-conjugated anti-mouse IgG was from Sigma and peroxidase-conjugated anti rabbit IgG was from BioRad (Italy). All antibodies were diluted 1:1000 except β-actin (1:10 000). The immunoblotting was performed as described [15].

2.7 Reactive oxygen species (ROS) generation
ROS was measured as described [18]. ECs were pre-incubated in culture medium containing DCFH-DA for 1 h to establish a stable intracellular level of the probe. The same concentration of DFCH-DA was present throughout the incubation with the lipoproteins. DFCH-DA penetrating the cells is initially converted into DCFH by cellular esterase, then DFCH is in turn oxidised to DCF in the presence of ROS. The DCF fluorescence intensity of the cells is an index of the intracellular levels of ROS and can be determined by fluorescence spectrophotometry with excitation and emission wavelengths at 475 and 525 nm, respectively.

2.8 Cell viability and morphology
Cell viability was evaluated after exposure to VLDL or Ox-VLDL using the MTT test [19] and the LDH release assay [20]. VLDL and Ox-VLDL were added in serum-free medium at increasing concentrations (10 to 75 µg protein/ml) for 6 and 18 h and at 75 µg/ml for increasing periods of time (2 to 18 h). For the assays cells in 12-well plates (6x104 cells/well) were exposed to VLDL and Ox-VLDL. Cell viability was determined as the ratio of lipoprotein-exposed cells versus control cells. The MTT and the LDH assays were performed as described [15]. Nuclear morphology was analysed with the fluorescent dye bisbenzimide (Hoechst 33258) as described [15].

2.9 Statistical analysis
Data are shown as mean±S.D. Differences in gene expression were evaluated by a Student's t-test (Statsoft Statistica Package). A value of P≤0.05 was considered significant.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1 Experimental set-up and overview of regulation events
The experimental set-up was designed to analyse endothelial cell gene expression in the presence of native VLDL or Ox-VLDL in relation to basal conditions. The EAhy926 cells were synchronised in a serum-free medium for 18 h; then VLDL (75 µg/ml) or Ox-VLDL (75 µg/ml) were added for 6 h. This protein content corresponds to approximately 350 µg/ml of triglycerides, which is in the physiological range [21]. Control cells were incubated for 6 h with the experimental medium containing the same percentage of PBS that was added with the stimuli. A human gene expression array (Clontech) that contains around 8400 genes was used. Radioactively labelled cDNA probes generated from total RNA from control cells, cells incubated with VLDL and cells incubated with Ox-VLDL, were hybridised in parallel to identical cDNA arrays. mRNA from 1620 of the 8411 arrayed genes (19.2%) were expressed in endothelial cells under basal conditions. Genes with a ratio of 2.0 or above were considered positively regulated whereas those that had a ratio of 0.5 or below were considered negatively regulated. The expression ratios reported are the average from four separate experiments. Genes that showed a standard deviation higher that 35% of the average were discarded. Using these criteria, 115 genes were up-regulated with VLDL and 206 were up-regulated with Ox-VLDL, 33 of which were up-regulated by both VLDL and Ox-VLDL. Eighty-one were down-regulated with VLDL and 106 were down-regulated with Ox-VLDL. Twenty-nine of these genes were down-regulated by both (Fig. 1A). To investigate whether the gene expression in VLDL-treated cells or in Ox-VLDL-treated cells correlate with the expression levels of the individual genes in control cells, the average intensities from four separate experiments in VLDL-treated cells or in Ox-VLDL-treated cells were plotted in a regression analysis against the average intensities of the individual genes from four separate experiments in control cells (Fig. 1A). The overall gene expressions were highly correlated (R2=0.85 for VLDL and R2=0.87 for Ox-VLDL) between stimulated and control cells at an interval of approximately three orders of magnitude, however some genes deviated from the main trend.


Figure 1
View larger version (46K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 cDNA microarray analysis of gene expression in response to VLDL and Ox-VLDL. (A) Regression analysis and scatter diagram of genes up-regulated, unmodified, and down-regulated by VLDL and Ox-VLDL in human endothelial cells. Genes with a ratio of 2.0 or above were considered positively regulated (green) whereas those that had a ratio of 0.5 or below were considered negatively regulated (red). The blue spots represent the unmodified genes. The x-axis shows the signal intensity of control cells while the y-axis shows the signal intensity of VLDL-treated or Ox-VLDL treated cells. The formulae defines the regression line. The data represent the average of four separate experiments. (B) Analysis of housekeeping gene expression. Radioactively labelled cDNA probes generated from total RNA from control cells (B), cells incubated with VLDL for 6 h (C) and cells incubated with Ox-VLDL for 6 h (D), were hybridised in parallel to identical cDNA arrays. Hybridisation patterns were assessed by autoradiography for 8 to 24 h and analysed by scanning densitometry. The results illustrated are representative of four different experiments. All housekeeping genes were assigned a number (E) and the ratios reflecting their expression (mean±S.D. of four different experiments) are presented in the Table.

 
Several genes that are supposed to be unresponsive under several conditions (i.e. housekeeping genes) are spotted on the array (Fig. 1B); the expression of these genes was investigated to validate the experimental set-up. Under our conditions all the housekeeping genes showed a ratio between 0.7 and 1.3 apart from GAPDH which showed an expression of 2.3 times in cells incubated with Ox-VLDL (Fig. 1B). This result is in agreement with recent observations suggesting that mammalian GAPDH cannot be regarded as a housekeeping gene [22].

3.2 Gene cluster analysis
An agglomerative cluster analysis was used to identify the typical response patterns and establish the relationships between the different gene expression profiles. Seven clusters resulted from the analysis. As expected, a majority of genes showed no changes in gene expression with either VLDL or Ox-VLDL (cluster 4). Genes contained in cluster 1 showed an increased expression with both VLDL and Ox-VLDL treatment (Table 2) (genes with a ratio higher than 3 are shown); genes contained in clusters 2 and 3 showed an increased expression with VLDL or Ox-VLDL, respectively (Table 2). Genes in cluster 4 were unaffected by both VLDL and Ox-VLDL. Genes belonging to clusters 5 and 6 showed a decreased expression with Ox-VLDL or VLDL, respectively (Table 3) (genes with a ratio lower than 3 are shown), while genes in cluster 7 showed a decreased expression with both VLDL and Ox-VLDL (Table 3).


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

  		Table 2 Genes up-regulated by VLDL and Ox-VLDL (cluster 1)

 

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

  		Table 3 Genes down-regulated by VLDL (cluster 5)

 
3.3 Quantitative real-time PCR
To validate the pattern and threshold expression levels in the analysis, we analysed the expression of matrix metalloprotease 2 (MMP-2), interleukin 15 (IL-15) and matrix metalloprotease 7 (MMP-7) by quantitative real-time PCR in both EAhy926 and HUVECs. Data from HUVECs are presented in Fig. 2. The expression of these genes was similar in HUVECs and Eahy926 cells (data not shown). MMP-2 was up-regulated by VLDL and Ox-VLDL on the array while IL-15 was up-regulated only by Ox-VLDL; MMP-7 neither changed with VLDL nor with Ox-VLDL. MMP-2 showed a medium intensity level on the array, while IL-15 and MMP-7 had a low intensity level. The expression of MMP-2 in the gene array showed ratios of 3.90±1.21 and 2.31±0.21 in the VLDL- and Ox-VLDL-treated cells, respectively. The corresponding quantitative real-time PCR showed 5.47±1.74- and 3.98±2.33-fold increases (Fig. 2). The ratio of the IL-15 gene expression in the gene array was 1.49±0.35 for VLDL-treated cells and 2.10±0.48 for ox-VLDL-treated cells. The corresponding quantitative real-time PCR showed 1.34±0.26- and 4.32±0.75-fold increases (Fig. 2). The expression of MMP-7 did not change (1.26±0.92- and 1.24±0.85-fold increases with VLDL and Ox-VLDL, respectively) which is in agreement with the array data (1.01±0.12 and 0.83±0.56 with VLDL and Ox-VLDL, respectively). To investigate whether the effect of VLDL or Ox-VLDL is concentration-dependent, we analysed the expression of transforming growth factor alpha (TGF{alpha}) (2.52±1.09-fold activation by VLDL in array analysis), CD38 (3.49±2.13-fold activation by VLDL in array analysis) and macrophage migration inhibitory factor (MIF) (2.4±0.61-fold activation by Ox-VLDL in array analysis) in HUVECs incubated with increasing concentrations (5 to 75 µg/ml) of VLDL or Ox-VLDL (Fig. 3). VLDL induced CD 38 and TGF{alpha} mRNA expression in a concentration-dependent manner, while Ox-VLDL induced MIF expression in a concentration-dependent manner (Fig. 3).


Figure 2
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Quantitative real-time RT–PCR and cDNA microarrays for IL-15, MMP-2 and MMP-7. The evaluation of the expression levels of IL-15, MMP-2 and MMP-7 is shown for control, VLDL and Ox-VLDL treated cells. (A) Representative real-time PCR experiment for MMP-2 and beta-actin. (B) Results of the real-time PCR (mean±S.D. of mRNA from four different experiments); the y-axis indicates the fold induction of the genes (corrected for the beta-actin content). (C) The results from the cDNA microarrays (mean±S.D. of mRNA from four different experiments). (D) Images of the IL-15, MMP-2 and MMP-7 genes from a representative experiment with the microarrays.

 

Figure 3
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Concentration-dependent effects of VLDL and Ox-VLDL on CD 38, MIF and TGF{alpha} mRNA expression. HUVEC were incubated for 6 h with VLDL or Ox-VLDL (5 to 75 µg/ml). The mRNA expression levels for CD38, MIF and TGF{alpha} are shown. Data are the average of four separate experiments±S.D. *P<0.01 vs. corresponding VLDL or Ox-VLDL values.

 
3.4 Intracellular kinase signalling
The microarray analysis showed that VLDL and Ox-VLDL modulate the expression of several kinases and transcription factors. It has previously been shown that VLDL can modulate MAP kinase activity in smooth muscle cells [23]. Thus we addressed the question of whether VLDL and Ox-VLDL can activate MAPK pathways in HUVECs. In our experiments cells were pre-incubated with serum-free medium for at least 4 h and then VLDL or Ox-VLDL were added for 5 up to 40 min. VLDL, but not Ox-VLDL, activated ERK1/2 and the peak of phosphorylation was reached after 5 to 10 min of stimulation. p38 MAPK was activated by VLDL with a peak of phosphorylation after 5 min but to a lesser extent compared with the activation induced by Ox-VLDL which induced a sustained phosphorylation for up to 40 min (Fig. 4). Several transcription factors are activated through MAPK-dependent pathways. Both VLDL and Ox-VLDL activated CREB with a peak of activity around 10 min, in agreement with the observation that both ERK1/2, via p90RSK, and p38 MAPK, via MSK-1, could activate CREB. The activation observed with VLDL was stronger than the one observed with Ox-VLDL. Ik-B alpha phosphorylation results in the release and nuclear translocation of active NF-kB. The basal level of phosphorylation of Ik-B alpha was high, suggesting that the pathway is activated also under basal conditions. A slight effect on phosphorylation was observed after 5 and 10 min with both VLDL and Ox-VLDL (Fig. 4).


Figure 4
View larger version (60K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Effects of VLDL and Ox-VLDL on intracellular kinase pathways. HUVEC were incubated for 5 to 40 min with VLDL or Ox-VLDL (75 µg/ml). Cell proteins were subjected to SDS–PAGE and immunoblotting with mouse monoclonal phospho-p44/42 MAPK antibody, rabbit polyclonal phospho-p38 MAPK antibody, rabbit polyclonal phospho-CREB antibody, rabbit polyclonal phospho-IkB-alpha antibody and mouse monoclonal anti-beta actin antibody. For experimental details see Methods. The panels illustrate a representative result of four experiments.

 
3.5 Oxidative stress, cell viability and cell death
Several genes involved in oxidative stress, cell proliferation and cell death are modulated during stimulation with either VLDL or Ox-VLDL. The effect of lipoproteins on intracellular ROS generation was investigated using DCFH-DA. A total of 23.31% of cells incubated with Ox-VLDL showed an increase in ROS generation as compared to 7.26% of control cells and 1.31% of VLDL-treated cells (Fig. 5). The ROS generation induced by Ox-VLDL was confirmed with microscope analysis of cell fluorescence (Fig. 5). The observation that the percentage of cells showing increase in ROS generation is higher in control cells than in VLDL-treated cells is in agreement with the fact that cells were pre-incubated and treated in serum-free medium, i.e. control cells can suffer oxidative stress in vitro due to lack of serum while addition of VLDL provides at least in part a survival condition. To investigate this hypothesis MTT transformation and LDH release were measured. After incubation with VLDL or Ox-VLDL (5 to 75 µg/ml) for 6 and 18 h only cells incubated with Ox-VLDL showed a dose-dependent decrease in viability compared with control cells (58.42% after 6 h and 18.90% after 18 h at 75 µm/ml). Cells incubated with VLDL showed no change or a small increase in cell viability (122.15% after 6 h and 108.33% after 18 h at 75 µm/ml) (Fig. 6). The LDH release assay confirmed these findings (Fig. 6). The analyses of cell and nuclear morphology demonstrated the presence of shrinkage and blebs in cells incubated with Ox-VLDL (Fig. 7). The effect is specific for Ox-VLDL and is not due to the presence of copper or BHT in the medium as control cells incubated with 20 µM CuSO4 and 40 µM BHT did not show any changes in cell morphology or viability as previously reported [15] (data not shown).


Figure 5
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Effects of VLDL and Ox-VLDL on ROS generation in ECs. HUVEC were incubated for 4 h with VLDL or Ox-VLDL (75 µg/ml). The DCF fluorescence intensity of the cells was used as an index of intracellular levels of ROS. The upper part of the figure shows the cytofluorimetric analysis of cells incubated with VLDL (red) or Ox-VLDL (green) compared with the basal condition (blue). The graph shows the percentage of DCF-positive cells in all experimental conditions.

 

Figure 6
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 MTT test and LDH release in VLDL- and Ox-VLDL-treated ECs. Endothelial cells were incubated for 6 and 18 h with increasing concentrations of VLDL and Ox-VLDL (5 to 75 µg protein/ml). At the end of the incubation the MTT transformation (A and B) and LDH release (C and D) were evaluated (for experimental details see Methods). Data are the average of four separate experiments ±S.D. *P<0.001 **P<0.01 vs. VLDL values.

 

Figure 7
View larger version (72K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Cell and nuclear morphology of ECs incubated with VLDL or Ox-VLDL. Panels A through F: Endothelial cells were incubated for 18 h in control medium (A and D), in medium containing VLDL (75 µg protein/ml) (B and E), or in medium containing Ox-VLDL (75 µg protein/ml) (C and F). At the end of the experiment, the morphology of the cells (A–C) and of the nuclei (D–F), evaluated by Hoechst 33258 staining, was studied. The results shown are representative of four different experiments. Magnifications: x400 and x1000.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Hypertriglyceridaemia is considered a risk factor for CVD [24]. Although the molecular mechanisms for the relationship between elevated triglyceride-rich lipoproteins and CVD remain uncertain, evidence is accumulating to suggest that endothelial dysfunction is involved [12,25]. Triglyceride-rich lipoproteins can cross the endothelial barrier and enter the arterial wall, placing them in a position where they may promote endothelial damage [26]. Studies aimed at characterising the endothelial response to triglyceride-rich lipoproteins are required to distinguish how these lipoproteins contribute to the pathophysiology of atherosclerosis. In the present study we investigated the gene expression pattern in human endothelial cells incubated with VLDL. Moreover, as VLDL can enter the arterial wall [26] and undergo oxidative modification [7–9], we compared the VLDL induced expression pattern with the one of Ox-VLDL. The list of genes that were up-regulated by both VLDL and Ox-VLDL comprises genes that to the best of our knowledge, apart from MMP-2, have not been described in the context of atherosclerosis. Most of them are involved in intracellular signalling and transcriptional regulation, suggesting massive endothelial cell activation by triglyceride-rich lipoproteins. The following discussion focuses mainly on genes involved in inflammation and cytotoxicity that could account for the endothelial dysfunction observed with triglyceride-rich lipoproteins [12,25,26].

4.1 Cell signalling
VLDL induced the expression of the NF-kB subunit p65 (2.76±0.76-fold), thus sustaining the pro-inflammatory activation of ECs. In contrast, there was no effect by Ox-LDL on p65 expression (0.93±0.18), which is in agreement with previous findings that VLDL but not Ox-VLDL induces nuclear staining for the activated RelA (p65) subunit of NF-kappaB in aortic endothelial cells [6]. On the other hand, Ox-VLDL increased the expression of transcription factor AP-2 gamma (3.84±1.02), TFIIA-alpha/beta-like factor (11.37±5.67), and CRE-Bpa (2.16±0.12), all transcription factors implicated in vascular wall cell functions [27]. The observation that several intracellular kinase network members and transcription factors are specifically induced by either VLDL or Ox-VLDL suggests that different pathways can be activated by native as opposed to oxidised lipoproteins. This hypothesis is supported by the observation that VLDL mainly activates the ERK 1/2 pathway as previously shown in smooth muscle cells [23] while Ox-VLDL activates the p38 MAPK pathway, sustaining activation for up to 40 min. The ERK 1/2 pathway is involved in cell proliferation while the p38 MAPK pathway is involved in inflammation and apoptosis [28]. Accordingly, VLDL activating the ERK1/2 pathway induced the expression of genes involved in cell proliferation and survival while prolonged activation of the p38 MAPK by Ox-VLDL could lead to inflammation and cell death. CREB can be activated both through the ERK1/2 and the p38 MAPK pathways. Thus, the activation of CREB by VLDL can be the result of ERK1/2 and p38 MAPK activation while Ox-VLDL affects CREB phosphorylation through activation of p38 MAPK. The NF-kB pathway appeared to be activated also under basal conditions as Ik-B alpha was phosphorylated in control cells. Moreover, both VLDL and Ox-VLDL induced a slight phosphorylation in agreement with previous observation [6]. As several other transcription factors can be activated trough MAPK-dependent pathways, the selective activation of p38 MAPK or ERK1/2 can result in the activation of other transcription factors involved in the gene responses observed with the array analysis.

4.2 Endothelial cell activation
The activation of the endothelium is a crucial step in the recruitment of immune cells to the vessel wall [29]. Adhesion molecules play a key role in the early stages of atherosclerosis [30] and the plasma concentrations of soluble adhesion molecules are elevated in individuals with hypertriglyceridaemia [4]. ECs incubated with VLDL showed an increased expression of major histocompatibility complex class II DR beta 5 (2.57±1.02), integrin alpha M (2.24±0.76), a disintegrin and metalloproteinase domain 15 (ADAM 15) (2.11±1.13) and CD38 (3.49±2.13), all potentially connected with cell–cell adhesion. Of note, toll-like receptor 4, recently identified in atherosclerotic lesions [31], was increased by VLDL treatment (2.25±1.44) suggesting another mechanism involved in the endothelial cell activation by VLDL. Several growth factors, cytokines, and chemokines and their receptors have been shown to be induced by VLDL and Ox-VLDL. Somatomedin C (5.64±2.89), TGF{alpha} (2.52±1.09) and cardiotrophin-like cytokine (2.47±0.43) were specifically induced by VLDL.

Ox-VLDL induced the expression of the small inducible cytokine subfamily A member 21 (2.83±0.71), a potent chemoattractant for lymphocytes [32], the interleukin 12 receptor beta 1 (2.32±0.32) and interleukin 15 (IL-15) (2.10±0.48), a cytokine recently discovered in mouse and human atherosclerotic lesions [33]. Also, Ox-VLDL increased the expression of macrophage migration inhibitory factor (MIF), an inflammatory cytokine that shows increased expression during the progression of atherosclerosis [34]. The expression of IL-15 and MIF were investigated in detail using real-time PCR. These experiments confirmed the expression pattern detected with the arrays and suggest a mechanism that may contribute to T-cell recruitment and macrophage retention in the vascular wall.

It has previously been shown that VLDL increases PAI-1 expression in ECs [35]. Unfortunately, there was no functional probe for PAI-1 on the microarray. However, real-time PCR together with PAI-1 protein measurements confirmed an effect of VLDL on PAI-1 expression (data not shown).

4.3 Oxidative stress, cell proliferation and death
Oxidative stress has been proposed to be a link between the triglyceride-rich lipoproteins and endothelial damage [36]. Ox-VLDL induced the expression of several stress response proteins such as peroxiredoxin 1 (3.80±1.72), glutathione-S-transferase like (2.07±0.84) and the heat shock 27 kD protein 1 (2.13±1.19), previously shown to be increased in atherosclerotic lesion of Apo E-deficient mice [33]. In addition, Ox-VLDL induced tumour protein p53-binding protein (14.29±5.43) and BCL2-antagonist/killer 1 (2.61±1.37) that modulate apoptotic cell death, as well as up-regulation of several DNA damage signalling/repair proteins such as three prime repair exonuclease 2 (2.85±0.66), excision repair cross-complementing rodent repair (2.72±0.66) and RAD23 homolog B (2.31±0.41). Ox-VLDL also decreased the expression of interleukin enhancer binding factor 3 (0.31±0.05). Overall, this suggests a cytotoxic role of Ox-VLDL on endothelial cells at a concentration lower than that reported previously for other species of modified lipoproteins such as oxidised LDL [37]. Native VLDL instead induced the expression of small GTP binding protein Rac1 (2.36±0.85), MAPKKK-1 (12.16±3.95), cyclin-dependent kinase 2 (2.37±1.27), myeloid cell nuclear differentiation antigen (2.31±0.59) and serum-inducible kinase (2.47±1.08), all involved in cell proliferation. This suggests a pro-proliferating role of VLDL in endothelial cells, which is in agreement with recent findings in smooth muscle cells [23]. Furthermore, only VLDL induced CD38 expression in a concentration-dependent manner, and it has recently been shown that CD38+ cells divide faster than CD38– cells [38]. These findings are also supported by the observation that Ox-VLDL induced a massive ROS generation in ECs while in VLDL-treated cells, the ROS generation was even lower than that observed in control cells. Moreover, the analysis of LDH release and MTT transformation showed that Ox-VLDL induces cell death while VLDL showed an opposite effect. Also, the analysis of cell and nucleus morphology only showed the presence of blebbing and shirking in cells incubated with Ox-VLDL.

Several other proteins that are still functionally unclassified were induced by VLDL or Ox-VLDL, e.g. the tumour necrosis factor alpha-induced protein 1 (3.92±2.08 and 3.71±2.34 fold up-regulation for VLDL and Ox-VLDL, respectively). The elucidation of their functions will be important to suggest new pathways linking VLDL and Ox-VLDL to endothelial function.

In conclusion, we studied the modulation of gene expression in human endothelial cells by VLDL or Ox-VLDL. The expression of several genes implicated in EC activation and damage/proliferation was found to be perturbed, confirming ECs as the site of the dysfunction observed in vivo during hypertriglyceridaemia. Different intracellular kinases as well as transcription factors may play a key role in the modulation of the downstream events induced by VLDL, or Ox-VLDL, suggesting that these lipoproteins have different targets for the regulation of gene expression. In vivo, the endothelium faces both VLDL and Ox-VLDL. Thus, the overall effect could depend on the pro-oxidant micro-environment that promotes conversion of VLDL to Ox-VLDL. Further studies are necessary to evaluate the significance of these findings.

Time for primary review 34 days.


   Acknowledgements
 
This work was supported by grants from the Swedish Medical Research Council (8691 and 12660), the Swedish Heart–Lung Foundation, the King Gustaf V and Queen Victoria Foundation, the Marianne and Marcus Wallenberg Foundation, the Torsten and Ragnar Söderberg Foundation, the Foundation for Old Servants and the Professor Nanna Svartz Foundation, the European Community (BMH4-CT98-3191 to ALC) and MIUR (Ministero Istruzione Università e Ricerca to GDN). We thank Alessandro Margutti for the software assistance, Carl Whatling and Cristina Banfi for helpful discussions.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Lewis T.V., Dart A.M., Chin-Dustin J.P. Endothelium-dependent relaxation by acetylcholine is impaired in hypertriglyceridemic humans with normal levels of plasma LDL cholesterol. J Am Coll Cardiol (1999) 33:805–812.[Abstract/Free Full Text]
  2. Doi H., Kugiyama K., Ohgushi M., et al. Remnants of chylomicron and very low density lipoprotein impair endothelium-dependent vasorelaxation. Atherosclerosis (1998) 137:341–349.[CrossRef][Web of Science][Medline]
  3. Simpson H.C.R., Meade T.W., Stirling Y., et al. Hypertriglyceridaemia and hypercoagulability. Lancet (1983) 1:786–790.[Web of Science][Medline]
  4. Hackman A., Abe Y., Insull W., et al. Levels of soluble cell adhesion molecules in patients with dyslipidemia. Circulation (1996) 93:1334–1338.[Abstract/Free Full Text]
  5. Banfi C., Mussoni L., Ris P., et al. Very low density lipoprotein-mediated signal transduction and plasminogen activator inhibitor type 1 in cultured HepG2 cells. Circ Res (1999) 85:208–217.[Abstract/Free Full Text]
  6. Dichtl W., Nilsson L., Goncalves I., et al. Very low-density lipoprotein activates nuclear factor-kappaB in endothelial cells. Circ Res (1999) 84:1085–1094.[Abstract/Free Full Text]
  7. Parthasarathy S., Quinn M.T., Schewenke D.C., Carew T., Steinberg D. Oxidative modification of beta-very low density lipoprotein: potential role in monocyte recruitment and foam cell formation. Arteriosclerosis (1989) 9:389–404.
  8. Horrigan S., Campbell J.H., Campbell G.R. Oxidation of β-very low density lipoprotein by endothelial cells enhances its metabolism by smooth muscle cells in culture. Arterioscler Thromb Vasc Biol (1991) 11:279–289.[Abstract/Free Full Text]
  9. Mohr D., Stoker R. Radical mediated oxidation of isolated human very low density lipoprotein. Arterioscler Thromb Vasc Biol (1994) 14:1186–1192.[Abstract/Free Full Text]
  10. Whitman S.C., Sawyes C.G., Miller D.B., Wolfe B.M., Huff M.W. Oxidized type IV hypertriglyceridemic VLDL-remnants cause greater macrophage cholesteryl ester accumulation than oxidized LDL. J Lipid Res (1998) 39:1008–1020.[Abstract/Free Full Text]
  11. Wang G.P., Deng Z.D., Ni J., Qu Z.L. Oxidized low density lipoprotein and very low density lipoprotein enhance expression of monocyte chemoattractant protein-1 in rabbit peritoneal exudate macrophages. Atherosclerosis (1997) 133:31–36.[CrossRef][Web of Science][Medline]
  12. Sattar N., Petrie J.R., Jaap A.J. The atherogenic lipoprotein phenotype and vascular endothelial dysfunction. Atherosclerosis (1998) 138:229–235.[CrossRef][Web of Science][Medline]
  13. Havel R.J., Eder H.A., Bragdon J.H. The distribution and chemical composition of ultracentrifugally separated lipoprotein in human serum. J Clin Invest (1955) 34:1345–1353.[Web of Science][Medline]
  14. Lowry O.H., Rosebrough H.J., Farr A.R., Randall R.J. Protein measurement with the Folin-phenol reagent. J Biol Chem (1951) 193:265–275.[Free Full Text]
  15. Pirillo A., Norata G.D., Zanelli T., Catapano A.L. Overexpression of HSP70i in COS-1 cells fails to protect from the cytotoxicity of Ox-LDL. Arterioscler Thromb Vasc Biol (2001) 21:348–354.[Abstract/Free Full Text]
  16. Edgell C.J.S., McDonald C.C., Graham J.B. Permanent cell line expressing human factor VIII-related antigen established by hybridisation. Proc Natl Acad Sci USA (1983) 80:3734–3737.[Abstract/Free Full Text]
  17. Jaffe E.A., Nachman R.L., Becker C.J., Minick C.R. Culture of human endothelial cells derived from umbilical veins: identification by morphological and immunological criteria. J Clin Invest (1973) 52:2745–2749.[Web of Science][Medline]
  18. Hsieh H.J., Cheng C.C., Wu A.T., et al. Increase of reactive oxygen species (ROS) in endothelial cells by shear flow and involvement of ROS in shear-induced c-fos expression. J Cell Physiol (1998) 175:156–162.[CrossRef][Web of Science][Medline]
  19. Denizot F., Lang R. Rapid colorimetric assay for cell growth and survival. Modifications to tetrazolium dye procedure giving improved sensitivity and reliability. J Immunol Methods (1986) 89:271–277.[CrossRef][Web of Science][Medline]
  20. Windholz M. Lactate dehydrogenase. In: The Merck Index—Windholz M., ed. (1983) Rahway, NJ, USA: Merck & Co.
  21. Betteridge D.J., Illingworth D.R., Shepherd J., eds. Lipoproteins in health and disease. (1999) Arnold.
  22. Sirover M.A. New insights into an old protein: the functional diversity of mammalian glyceraldehydes-3-phosphate dehydrogenase. Biochim Biophys Acta (1999) 1432:159–184.[CrossRef][Medline]
  23. Zhao D., Letterman J., Schreiber B.M. Beta-migrating very low density lipoprotein (beta VLDL) activates smooth muscle cell mitogen-activated protein (MAP) kinase via G protein-coupled receptor-mediated transactivation of the epidermal growth factor (EGF) receptor: effect of MAP kinase activation on beta VLDL plus EGF-induced cell proliferation. J Biol Chem (2001) 276:30579–30588.[Abstract/Free Full Text]
  24. Gotto A.M. Triglyceride. The forgotten risk factor. Circulation (1998) 97:1027–1028.[Free Full Text]
  25. Lundman P., Eriksson M., Schenck-Gustafsson K., Karpe F., Tornvall P. Transient tryglyceridemia decreases vascular reactivity in young, healthy men without risk factors for coronary heart disease. Circulation (1997) 96:3266–3268.[Abstract/Free Full Text]
  26. Nordestagaard B.G., Wooton R., Lewis B. Selective retention of VLDL, IDL and LDL in the arterial intima of genetically hyperlipidaemic rabbits in vivo. Arterioscler Thromb Vasc Biol (1995) 15:534–542.[Abstract/Free Full Text]
  27. Napoli C., Quehenberger O., De Nigris F., et al. Mildly oxidized low density lipoprotein activates multiple apoptotic signaling pathways in human coronary cells. FASEB J (2000) 14:1996–2007.[Abstract/Free Full Text]
  28. Seger R., Krebs E.G. The MAPK signaling cascade. FASEB J (1995) 9:726–735.[Abstract]
  29. Hansson G.K. Immune mechanisms in atherosclerosis. Arterioscler Thromb Vasc Biol (2001) 21:1876–1890.[Abstract/Free Full Text]
  30. Chia M.C. The role of adhesion molecules in atherosclerosis. Crit Rev Clin Lab Sci (1998) 35:573–602.[CrossRef][Web of Science][Medline]
  31. Edfeldt K., Swedenborg J., Hansson G.K., Yan Z.Q. Expression of toll-like receptors in human atherosclerotic lesions: a possible pathway for plaque activation. Circulation (2002) 105:1158–1161.[Abstract/Free Full Text]
  32. Nagira M., Imai T., Hieshima K., et al. Molecular cloning of a novel human CC chemokine secondary lymphoid-tissue chemokine that is a potent chemoattractant for lymphocytes and mapped to chromosome 9p13. J Biol Chem (1997) 272:19518–19524.[Abstract/Free Full Text]
  33. Wuttge D.M., Sirsjö A., Eriksson P., Stemme S. Gene expression in atherosclerotic lesion of apoE deficient mice. Mol Med (2001) 7:383–392.[Web of Science][Medline]
  34. Burger-Kentischer A., Goebel H., Seiler R., et al. Expression of macrophage migration inhibitory factor in different stages of human atherosclerosis. Circulation (2002) 105:1561–1566.[Abstract/Free Full Text]
  35. Stiko-Rahm A., Wiman B., Hamsten A., Nilsson J. Secretion of plasminogen activator inhibitor-1 from cultured human umbilical vein endothelial cells is induced by very low density lipoprotein. Arteriosclerosis (1990) 10:1067–1073.[Abstract/Free Full Text]
  36. Plotnick G.D., Corretti M.C., Vogel R.A. Effect of antioxidant vitamins on the transient impairment of endothelium-dependent brachial artery vasoactivity following a single high-fat meal. J. Am. Med. Assoc. (1997) 278:1682–1686.[Abstract/Free Full Text]
  37. Pirillo A., Zhu W., Norata G.D., et al. Oxidized lipoproteins and endothelium. Clin Chem Lab Med (2000) 38:155–160.[CrossRef][Web of Science][Medline]
  38. Tangye S.G., Avery D.T., Deenick E.K., Hodgkin P.D. Intrinsic differences in the proliferation of naive and memory human B cells as a mechanism for enhanced secondary immune responses. J Immunol (2003) 170:686–694.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
G. D. Norata, P. Marchesi, A. Pirillo, P. Uboldi, G. Chiesa, V. Maina, C. Garlanda, A. Mantovani, and A. L. Catapano
Long Pentraxin 3, a Key Component of Innate Immunity, Is Modulated by High-Density Lipoproteins in Endothelial Cells
Arterioscler Thromb Vasc Biol, May 1, 2008; 28(5): 925 - 931.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
G. D. Norata and A. L. Catapano
Triglyceride-Rich Lipoproteins From Normotrygliceridemic Subjects and Hyperlipidemic Patients Differently Affect Endothelial Cell Activation and Gene Expression Patterns
Circ. Res., April 13, 2007; 100(7): e81 - e81.
[Full Text] [PDF]

Home page
 J Antimicrob ChemotherHome page
J. J. R. Blanco, I. S. Garcia, J. G. Cerezo, J. M. P. S. de Rivera, P. M. Anaya, P. G. Raya, J. G. Garcia, J. R. A. Lopez, F. J. B. Hernandez, and J. J. V. Rodriguez
Endothelial function in HIV-infected patients with low or mild cardiovascular risk
J. Antimicrob. Chemother., July 1, 2006; 58(1): 133 - 139.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
G. D. Norata, G. Tibolla, P. M. Seccomandi, A. Poletti, and A. L. Catapano
Dihydrotestosterone Decreases Tumor Necrosis Factor-{alpha} and Lipopolysaccharide-Induced Inflammatory Response in Human Endothelial Cells
J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 546 - 554.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
G. D. Norata, E. Callegari, M. Marchesi, G. Chiesa, P. Eriksson, and A. L. Catapano
High-Density Lipoproteins Induce Transforming Growth Factor-{beta}2 Expression in Endothelial Cells
Circulation, May 31, 2005; 111(21): 2805 - 2811.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
C. Thiemermann
Inhibition of the activation of nuclear factor kappa B to reduce myocardial reperfusion injury and infarct size
Cardiovasc Res, July 1, 2004; 63(1): 8 - 10.
[Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
G.D. Norata, E. Callegari, H. Inoue, and A.L. Catapano
HDL3 Induces Cyclooxygenase-2 Expression and Prostacyclin Release in Human Endothelial Cells Via a p38 MAPK/CRE-Dependent Pathway: Effects on COX-2/PGI-Synthase Coupling
Arterioscler Thromb Vasc Biol, May 1, 2004; 24(5): 871 - 877.
[Abstract] [Full Text]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
J. Rius, J. Martinez-Gonzalez, J. Crespo, and L. Badimon
Involvement of Neuron-Derived Orphan Receptor-1 (NOR-1) in LDL-Induced Mitogenic Stimulus in Vascular Smooth Muscle Cells: Role of CREB
Arterioscler Thromb Vasc Biol, April 1, 2004; 24(4): 697 - 702.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
J. Herrmann
New kits on the blot--can we microarray the future of atherosclerosis?
Cardiovasc Res, November 1, 2003; 60(2): 220 - 222.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Norata, G.D.
Right arrow Articles by Eriksson, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Norata, G.D.
Right arrow Articles by Eriksson, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?