| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular Biology |
From the Instituto de Investigación Cardiovascular de Barcelona, CSIC-ICCC-Hospital de Sant Pau, Barcelona, Spain.
Correspondence to Prof Lina Badimon, Laboratori dInvestigació Cardiovascular, Hospital de la Santa Creu i Sant Pau, Avda. Sant Antoni Maria Claret 167, 08025 Barcelona, Spain. E-mail lbmucv{at}cid.csic.es
| Abstract |
|---|
|
|
|---|
Methods and Results By differential display analysis, we have studied the effect of high concentrations of native low density lipoprotein (LDL) on endothelial gene expression. mRNA levels of lysyl oxidase (LOX), an enzyme involved in collagen and elastin cross-linking, were downregulated by LDL treatment in endothelial cells in a dose- and time-dependent manner (80% of inhibition by 180 mg/dL LDL for 24 hours). This reduction of LOX expression was associated with a decrease in LOX activity (40% and 54% of inhibition after 24 and 48 hours of LDL treatment, respectively). LOX mRNA half-life was not modified by LDL, but transcriptional inhibition blocked the effect of LDL. Inhibition of LOX activity by either LDL or ß-aminopropionitrile, an inhibitor of LOX, increased endothelial permeability (192±0.19- and 3.37±0.74-fold, respectively). Interestingly, a reduction in LOX expression (3.5-fold) was observed in vivo in the vascular wall of hypercholesterolemic pigs.
Conclusions These findings suggest that LDL downregulation of LOX could contribute to the endothelial dysfunction caused by hypercholesterolemia, thus contributing to atherosclerotic plaque formation.
Key Words: endothelial cells low density lipoproteins lysyl oxidase vascular permeability hypercholesterolemia
| Introduction |
|---|
|
|
|---|
See page 1365
It has been reported that atherogenic concentrations of LDL alter the composition and permeability of the endothelial barrier by inducing changes in the basement membrane.5 Thus, the alteration in endothelial extracellular matrix (ECM) could play a key role in the endothelial dysfunction associated with hypercholesterolemia. One of the key enzymes involved in ECM maturation is lysyl oxidase (LOX). This enzyme, a copper-containing semicarbazide-sensitive amine oxidase,6 initiates the covalent cross-linking of collagen and elastin that is essential in maintaining ECM structure. LOX oxidatively deaminates peptidyl lysine residues of collagen and elastin, leading to the synthesis of peptidyl aldehydes, which condense spontaneously to form the mature and insoluble ECM.7 LOX has also been associated with tumor supression8,9 and chemotaxis,10 and in the last few years, different LOX isoforms, probably with different substrate specificity and function, have been identified.1115 Alterations in normal LOX expression have been linked to human diseases. Reduced LOX expression has been described in type IX Ehlers-Danlos syndrome and Menkes disease,16 whereas fibrotic diseases, such as hepatic, lung, or kidney fibrosis, have been associated with increases in LOX.17 The role of this enzyme in atherosclerosis remains to be elucidated because it has been suggested that increases18 and decreases1921 in LOX activity could be associated with this pathology.
The aim of the present study was to analyze the mechanisms involved in the initiation of atherosclerosis, targeting the endothelial functions regulated by LDL. We have identified LOX as a gene differentially regulated by LDL in endothelial cells. LOX expression and activity are downregulated by LDL in a time- and concentration-dependent manner. Moreover, LOX inhibition was associated with an increase in endothelial permeability in vitro. Finally, in pigs fed an atherogenic diet, a significant downregulation of vascular LOX expression was observed. Considering the multiple cell functions in which LOX could be involved and its potential role in the atherogenic process, the effect of downregulation exerted by LDL could be crucial in endothelial dysfunction.
| Methods |
|---|
|
|
|---|
In Vivo Animal Model
Female pigs (Landrace/Largewhite [Piensos Victoria SA, Barcelona, Spain], mean body weight at initiation 32±4 kg) were randomized into 2 groups: normolipemic animals (n=6) that were fed a normal chow and hyperlipemic animals (n=10) that were fed a cholesterol-rich diet (2% cholesterol, 1% cholic acid, and 20% beef tallow) for 100 days.23,24 At the end of the dietary period, the animals were euthanized with a thiopental overdose. Plasma cholesterol levels and hematologic parameters were measured at baseline and at euthanasia. Because in the porcine model the first atherosclerotic lesions develop in the abdominal aorta, rings of abdominal aorta were collected and frozen in liquid N2 to analyze gene expression. All procedures were in accordance with institutional guidelines and followed the American Physiological Society guidelines for animal research.
Plasma Biochemistry
Plasma total cholesterol was determined with an automatic analyzer (Kodak Ektachem DT System). Plasma lipoproteins (HDL cholesterol, LDL cholesterol, and VLDL cholesterol) were fractionated by using the validated methods of the Lipid Research Clinic Program25 and quantified spectrophotometrically (Kontron Instruments).
LDL Isolation
Porcine or human LDLs were obtained from fresh nonfrozen plasma by sequential ultracentrifugation (density 1.019 to 1.063 g/mL). LDLs used in the experiments were <72 hours old. The purity of the LDLs was assessed by agarose gel electrophoresis (Paragon System, Beckman). LDL samples had no detectable levels of endotoxin (Limulus Amebocyte Lysate test, BioWhittaker), and thiobarbituric acidreactive substance values were <1.5 nmol malonaldehyde/mg protein.
mRNA-DD Analysis
Cells grown in a 6-well plate, as indicated above, were incubated with LDL (180 mg/dL for 24 hours). Total RNA was isolated by using a QuickPrep total RNA kit (Pharmacia) or Ultraspec (Biotecx) according to the manufacturers instructions. mRNAdifferential display (DD) analysis was performed with a Delta RNA Fingerprinting kit (Clontech) by using 1 µg of total RNA as described previously.22 Reproducible DNAs from 3 independent experiments upregulated or downregulated by LDL were cloned and sequenced. Comparison of DNA homology with databases (GenBank) was performed by the use of BLAST, a National Center for Biotechnology Information server program.
Northern Blot Analysis
Total RNA was obtained as described above. RNA samples were fractionated in 1.2% agarose2% formaldehyde gels. RNA was transferred by capillarity to Nytran Supercharge (Schleicher & Schuell) membranes and UVcross-linked. Filters were hybridized by using the reamplified product from the DDpolymerase chain reaction (PCR) assay labeled with [
-32P]dATP as the LOX probe. A 28S ribosomal cDNA was used to normalize the results, as described previously.3 Filters were exposed to Agfa Curix RP2 x-ray films and quantified by densitometric analysis (Molecular Dynamics).
RT-PCR
Reverse transcription (RT)-PCR analysis was performed as described previously.22 The specific oligonucleotides selected were as follows: porcine LOX upper primer, 5'-tga aat agc tgc ttt ggt-3'; human LOX upper primer, 5'-caa gta aga ttt tcc agt gc; and porcine and human LOX lower primer, 5'-tct gta gat gcc agt ctg-3'. Amplification was carried out by 20 cycles of 94°C for 1 minute, 57°C for 1 minute, and 72°C for 30 seconds, followed by a final extension of 72°C for 7 minutes. Levels of GAPDH were used to normalize the results.3,22
LOX Activity
LOX activity was measured by a high-sensitivity fluorescence assay as previously described.26 Briefly, PAECs were plated (500 000 cells per dish) on 100-mm-diameter tissue culture dishes. Thirty-six hours after plating, cells were serum-depleted and fed with phenol redfree medium 199 for 18 hours. Cells were then reincubated with fresh medium containing 180 mg/dL LDL every 24 hours. LOX activity was measured in cell culture medium. The medium (200 µL) was incubated in the presence and absence of 500 µmol/L BAPN at 37°C for 30 minutes with 1 U/mL horseradish peroxidase, 10 µmol/L Amplex red (Molecular Probes), and 10 mmol/L 1,5-diaminopentane in 1.2 mol/L urea and 0.05 mol/L sodium borate (pH 8.2). The reaction was stopped on ice, and differences in fluorescence intensity (563-nm excitation wavelength and 587-nm emission wavelength) between samples with and without BAPN were determined. Results were normalized by cell protein content.
Transendothelial Exchange
Endothelial permeability was determined by the exchange of FITC-dextran (Mr 40 000, Sigma Chemical Co) through the endothelial monolayer as described previously.27 In brief, PAECs (104) were seeded on Transwell-collagen porous membranes (3-µm pore size and 0.33-cm2 area, Costar) and cultured for 2 weeks to obtain a highly confluent culture. After 36 hours of incubation with LDL (140 mg/dL) or 16 hours with BAPN (100 µmol/L), FITC-dextran (10 µmol/L) was added to the upper compartment. After 6 hours, samples (50 µL) from the lower compartment were taken, and dextran transfer was determined in a fluorimeter (495-nm excitation wavelength and 530-nm emission wavelength).
Statistical Analysis
Data are expressed as mean±SD. Means were compared by ANOVA. For the in vivo study, statistical differences between groups were analyzed by the Mann-Whitney U test. Differences were considered significant at P<0.05.
| Results |
|---|
|
|
|---|
5-fold (P<0.001). Similar results were obtained by RT-PCR assays with specific oligonucleotides (data not shown).
|
Time- and Dose-Dependent Effect of LDL on LOX Expression in Endothelial Cells
Although DD experiments were performed in serum-deprived cells (2% FCS), the downregulation of LOX by atherogenic levels of LDL was also observed in the presence of 10% FCS, conditions that did not affect LOX basal expression (please see online Figure I, available at http://www.ahajournals.org). Thus, subsequent experiments were performed in serum-deprived cells. LOX mRNA levels decreased in a dose-dependent manner. As shown in Figure 2, LOX expression was significantly reduced by low LDL concentrations (74% inhibition by 50 mg/dL). To determine the time course of LOX mRNA expression, total RNA was isolated at different times from PAECs incubated with LDL (180 mg/dL). The LOX mRNA level, analyzed by RT-PCR, decreased after 12 hours of incubation with LDL, and expression levels remained low after 24 hours. Similar LDL-mediated downregulation effects on LOX expression were observed in human umbilical vein endothelial cells (please see online Figure II, available at http://www.ahajournals.org).
|
Effect of LDL on LOX Activity
LOX activity was determined in PAECs incubated with LDL (180 mg/dL). LDL significantly decreased LOX activity in the cell culture medium (40% and 54% of inhibition after 24 and 48 hours, respectively; Figure 3A).
|
Effect of LDL on Endothelial Barrier Function
Modulation of transendothelial exchange by LDL and BAPN, an inhibitor of LOX activity, was analyzed. As shown in Figure 3B, LDL and BAPN caused a significant increase in FITC-dextran transfer compared with control cultures (
2- and 3-fold, respectively), an effect that became apparent 6 hours after the addition of FITC-dextran. These results suggest that LOX inhibition disturbs endothelial barrier integrity.
Effect of DRB and Cycloheximide
To characterize the mechanisms involved in the downregulation of LOX expression by LDL, we analyzed the effect of 5,6-dichlorobenzimidazole (DRB), a transcriptional inhibitor, and cycloheximide, a protein synthesis inhibitor. PAECs were incubated with or without LDL for 24 hours, and then transcription was inhibited by DRB (50 µmol/L, time 0). LOX mRNA levels were analyzed by Northern analysis at different times (0, 6, 9, 12, and 24 hours). LDL-treated cells showed a half-life (
24 hours) similar to that of control cells; thus, the effect of LDL could not be attributable to a decrease in LOX mRNA stability (data not shown). DRB (Figure 4A) and cycloheximide (Figure 4B) blocked the effect of LDL on LOX mRNA levels, suggesting that transcriptional and posttranscriptional mechanisms are involved in the LDL-mediated effect.
|
In Vivo Effects of Hypercholesterolemia on LOX Expression in the Vessel Wall
LOX expression was analyzed by RT-PCR in porcine abdominal aortas from normolipemic and hypercholesterolemic pigs. Animals fed the hypercholesterolemic diet, compared with animals fed the normolipemic diet, showed higher plasma LDL cholesterol levels (333±120 versus 34±10.9 mg/dL, respectively; P<0.01). Vascular LOX mRNA levels were significantly reduced (
3-fold) by the hypercholesterolemic diet (P<0.01, Figure 5).
|
| Discussion |
|---|
|
|
|---|
We have analyzed the mechanisms involved in LOX regulation by LDL. Altogether, the experiments performed in the presence of cycloheximide and DRB suggest the involvement of a transcriptional mechanism, mainly because a decrease in LOX mRNA stability was not observed after LDL incubation. The long LOX mRNA half-life observed in our culture conditions (2% FCS) is in agreement with that reported in other cell types, such as smooth muscle cells maintained in low serum culture29 or quiescent IMR90 fibroblasts.30 However, in PAECs, serum deprivation did not alter LOX endothelial expression, and the reported superinduction of LOX mRNA by cycloheximide20 was not observed. Thus, differential regulation of LOX expression could be observed depending on the cell type. In this sense, it has been speculated that cell typespecific mechanisms could be involved in LOX regulation,31 probably mediated by different promoter elements. Supporting the data, we observed similar LOX downregulation by LDL in vascular cells from human and porcine origin. In addition, we observed that LOX downregulation by LDL is accompanied by a reduction in LOX activity. This inhibition was produced in the absence of any effect on bone morphogenetic protein 1 gene expression (authors unpublished data, 2002), the C-proteinase that processes pro-LOX to its active form with the highest efficiency.32 Thus, LDL specifically inhibited LOX expression/activity.
The increased endothelial permeability induced by BAPN, a specific inhibitor of LOX activity, and LDL suggests a role of LOX in the alteration of endothelial barrier function caused by LDL. Alterations in endothelial ECM composition by atherogenic concentrations of LDL have been reported5,33 and have been associated with an increase in endothelial permeability. However, there are some conflicting reports in the literature. Mildly oxidized LDL but not native LDL, at low concentrations and short incubation times (2 minutes), have been reported to increase endothelial permeability.34 Other authors testing nonatherosclerotic LDL concentrations concluded that LDL did not alter endothelial permeability.35 In the present study, we report that LOX downregulation is associated with an increased permeability. Moreover, not only does the inhibition of LOX maintain collagen in a soluble form, but this soluble form is also more susceptible to metalloproteinase degradation.36 On the other hand, because LOX activates the transcription of collagen III,19 LOX downregulation by LDL could trigger an additional decrease in some ECM components. Thus, LOX by different mechanisms could modulate ECM integrity and could play a key role in the modulation of endothelial function.
Interestingly, in vivo, in a model of early atherosclerosis, vascular LOX mRNA levels were downregulated by the hypercholesterolemic diet. No changes or even increases in vascular LOX activity have previously been reported in other animal models37,38; however, differences in LOX regulation between species and even differences in diet composition and types of lesions in these models could explain these discrepancies. Indeed, we have chosen the porcine animal model for the present study because it develops lesions resembling those in humans.39 After 100 days on a cholesterol-rich diet, pigs develop lesions that can be classified from arterial intimal thickening to type I and II according to the American Heart Association classification.40
Although the LDL-mediated LOX downregulation mechanism is largely unknown, preliminary data suggest that sterol regulatory elementbinding proteins are not involved because no sterol regulatory element has been identified in the LOX promoter and because ALLN, an inhibitor of sterol regulatory elementbinding protein catabolism, did not alter the LDL effect. On the other hand, LOX downregulation does not seem to be mediated by NO because endothelial NO synthase inhibition by NG-nitro-L-arginine methyl ester did not affect LOX expression (authors unpublished data, 2002).
In addition to hypercholesterolemia, other atherosclerotic risk factors, such as hyperhomocysteinemia, could be associated with decreases in LOX activity, inasmuch as homocysteine thiolactone and other analogues are irreversible inhibitors of LOX.21 Thus, LOX downregulation could be involved in the pathogenesis of atherosclerosis. In the present study, we have observed that hypercholesterolemia downregulates vascular LOX expression, an effect that could be related to an increase in endothelial permeability in the early steps of the atherosclerotic process. Further experiments focusing on the regulation of LOX activity in the course of LDL-induced vascular changes are necessary to clarify the role of LOX in this process.
| Acknowledgments |
|---|
Received May 30, 2002; accepted July 16, 2002.
| References |
|---|
|
|
|---|
in rat aortic smooth muscle cells. Arterioscler Thromb Vasc Biol. 2000; 20: 982988.This article has been cited by other articles:
![]() |
C. Rodriguez, J. Martinez-Gonzalez, B. Raposo, J. F. Alcudia, A. Guadall, and L. Badimon Regulation of lysyl oxidase in vascular cells: lysyl oxidase as a new player in cardiovascular diseases Cardiovasc Res, July 1, 2008; 79(1): 7 - 13. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Aydin, S. Signorelli, T. Lechleitner, M. Joannidis, C. Pleban, P. Perco, W. Pfaller, and P. Jennings Influence of microvascular endothelial cells on transcriptional regulation of proximal tubular epithelial cells Am J Physiol Cell Physiol, February 1, 2008; 294(2): C543 - C554. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bouez, C. Reynaud, E. Noblesse, A. Thepot, C. Gleyzal, J. Kanitakis, E. Perrier, O. Damour, and P. Sommer The Lysyl Oxidase LOX Is Absent in Basal and Squamous Cell Carcinomas and Its Knockdown Induces an Invading Phenotype in a Skin Equivalent Model Clin. Cancer Res., March 1, 2006; 12(5): 1463 - 1469. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Sullivan, T. H. Teal, I. P. Luttrell, K. B. Tran, M. A. Peters, and H. Wessells Microarray analysis reveals novel gene expression changes associated with erectile dysfunction in diabetic rats Physiol Genomics, October 17, 2005; 23(2): 192 - 205. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Schieffer, T. Selle, A. Hilfiker, D. Hilfiker-Kleiner, K. Grote, U. J.F. Tietge, C. Trautwein, M. Luchtefeld, C. Schmittkamp, S. Heeneman, et al. Impact of Interleukin-6 on Plaque Development and Morphology in Experimental Atherosclerosis Circulation, November 30, 2004; 110(22): 3493 - 3500. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Krettek, G. K. Sukhova, and P. Libby Elastogenesis in Human Arterial Disease: A Role for Macrophages in Disordered Elastin Synthesis Arterioscler. Thromb. Vasc. Biol., April 1, 2003; 23(4): 582 - 587. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Rodriguez, B. Raposo, J. Martinez-Gonzalez, V. Llorente-Cortes, G. Vilahur, and L. Badimon Modulation of ERG25 expression by LDL in vascular cells Cardiovasc Res, April 1, 2003; 58(1): 178 - 185. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. A. Bank and V. W.M. van Hinsbergh Lysyl Oxidase: New Looks on LOX Arterioscler. Thromb. Vasc. Biol., September 1, 2002; 22(9): 1365 - 1366. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2002 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |