Alterations in Internal Elastic Lamina Permeability As a Function of Age and Anatomical Site Precede Lesion Development in Apolipoprotein Ee
http://www.100md.com
循环研究杂志 2005年第9期
The Department of Biomedical Engineering (K.L., G.M.S., M.S.P.), Case Western Reserve University, Cleveland
Departments of Cardiovascular Medicine and Cell Biology (K.L., F.F., M.S.P.), Cleveland Clinic Foundation, Ohio.
Abstract
Early atherosclerosis is characterized by the accumulation of plasma-borne macromolecules (eg, low-density lipoproteins) in the arterial intima, which is bordered by endothelial cells (EC) and the internal elastic lamina (IEL). This accumulation is believed to be secondary to increased EC permeability. We hypothesized that a decrease in IEL permeability may precede lesion development and contribute to macromolecular accumulation. To test this hypothesis, we quantified EC and IEL permeability in lesion-free areas of the thoracic and abdominal aortas of chow-fed C57BL/6 control and atherosclerotic-prone apolipoprotein E (apoE)-null mice at 3 and 5 months of age. Between 3 and 5 months of age, apoE-null mice begin to develop atherosclerotic lesions in the thoracic aorta. No significant differences in EC and IEL permeability were observed at either time in C57BL/6 control mice. In contrast, 78% and 19% decreases in IEL permeability of the thoracic aorta and abdominal aorta, respectively, were observed between 3 to 5 months of age in apoE-null mice (thoracic: 2.05±1.33 and 0.44±0.15 e/min, P<0.001; abdominal: 1.13±0.58 and 0.93±0.44 e/min, P<0.05). To further determine whether decreased IEL permeability is linked with atherosclerotic lesion development, we quantified IEL permeability in the greater and lesser curvature of the aortic arch. In apoE-null mice, the lesser curvature of the aortic arch develops lesions before the greater curvature. We found a significant and sustained decrease (59%) in IEL permeability in the lesser curvature of the aortic arch compared with the greater curvature. These data suggest that atherogenesis involves the pathological remodeling of the IEL, not the endothelium before lesion development. This remodeling may be attributable to local responses of the endothelium and smooth muscle cells to hyperlipidemia.
Key Words: atherogenesis extracellular matrix mathematical modeling endothelial dysfunction
Introduction
Early atherosclerotic lesions are characterized by an abnormal accumulation of lipoproteins within the intima in large arteries.1eC3 This abnormal accumulation increases the residence time of low-density lipoprotein (LDL) in the intima, ultimately leading to oxidation of the LDL particles and inflammation. Despite the importance of the initial increase in LDL in the arterial intima, the critical mechanisms involved in early-stage atherosclerosis, or atherogenesis, are not fully understood.
Two distinct tissue layers line the arterial intima, the site of early lipid accumulation: the endothelial cell (EC) layer on the luminal side and the internal elastic layer on its abluminal side. Much literature has focused on the EC layer as a barrier to macromolecular entry; however, recent studies have suggested that internal elastic lamina (IEL) permeability is not static and could contribute to pathological changes in the artery wall.4eC7
We have used horseradish peroxidase (HRP) (44 kDa) as a traceable macromolecule because it has intrinsic peroxidase activity, allowing its concentration in arterial tissue to be quantified,8eC12 and because its size is on the same order of magnitude with growth factors of interest. In prior studies, we have analyzed the arterial-wall transport of HRP in a model consisting of a spatially lumped intima (compartment) and a spatially distributed media.13 Using this approach, we have demonstrated dynamic changes in IEL permeability during endotoxemia that may contribute to vascular permeability during sepsis.10,13
Because abnormal macromolecular accumulation in the intima could be secondary to either increased EC permeability or decreased permeability of the IEL, these observations led us to hypothesize that IEL remodeling could precede atherogenesis. In this study, we quantified the EC and IEL permeability of the macromolecular tracer in thoracic and abdominal aorta separately as a function of age in C57BL/6 (wild-type [WT]) and apolipoprotein E (apoE)-null mice to determine whether IEL remodeling precedes the atherogenic process.
Materials and Methods
Animal Preparation
The Animal Research Committee approved all animal protocols, and all animals were housed in the Association for Accreditation of Laboratory Animal Care animal facility of Cleveland Clinic Foundation. The WT normal (3-month [n=10] and 5-month old [n=11]) and apoE-null mice (3-month [n=8] and 5-month old [n=10]) were anesthetized with 1 mL of ketamineeCHCl (100 mg/mL), 1.5 mL of saline, and 0.3 mL of xylazine (20 mg/mL) and injected for 15 minutes or 30 minutes with 0.1 mL HRP (10 mg/mL) via a jugular vein, cannulated with PE-10 tubing, to make a total 1 mg/mL HRP in the plasma. Numbers of tissue profiles (ns) of WT normal mice were ns=86 (3-month abdominal), ns=99 (3-month thoracic), ns=100 (5-month abdominal), and ns=101 (5-month thoracic aorta). Numbers of tissue profiles of apoE-null mice were ns=50 (3-month abdominal), ns=79 (3-month thoracic), ns=74 (5-month abdominal), ns=90 (5-month thoracic aorta), ns=49 (lesser curvature), and ns=41 (greater curvature of aortic arch).
Quantifying HRP Concentration in Plasma and Tissue
Representative tissue images of apoE-null mice are shown in 3-month abdominal (Figure 1A), 3-month thoracic (Figure 1B), 5-month abdominal (Figure 1C), and 5-month thoracic aorta (Figure 1D). HRP concentration profiles we obtained in WT and apoE-null mice (Jackson Laboratories, Bar Harbor, Me) 15 and 30 minutes after HRP injection using previously published methods.8,11 Briefly, animals were anesthetized and injected with HRP (50 mg/mL) via the jugular vein. Blood samples collected as a function of time after HRP injection were obtained by cardiac puncture. The HRP concentration in plasma was determined by measuring the colored reaction product of o-dianisidine and H2O2 as a substrate for peroxidase reaction.14
The HRP concentration profiles across tissue were obtained as described by Penn et al.9 Following HRP circulation, PBS was injected into left ventricle followed by perfusion of 10 mL of 2.5% ice-cold glutaraldehyde via a left ventricle. After 5 minutes, the aorta was removed and put into 2.5% ice-cold glutaraldehyde during 4 hours, followed by washing with PBS and water for 30 minutes in each solution. Tissue samples were reacted with 3,3'-diaminobenzidine and H2O2 in imidazole buffer (0.1 N, pH 7.6) for 30 minutes. The tissue was washed with PBS and serially dehydrated with graded ethanol solutions up to 95%, embedded with paraffin, and serially sectioned 5 e using rotary microtome.
Plasma cholesterol concentration levels were measured using infinity Cholesterol Reagent with Cholesterol standard solutions from Sigma.
Data Acquisition and Image Processing
To determine HRP equilibrium distribution coefficients in the tissue, 8 to 10 cylindrical rings sectioned from fresh aorta from 2 WT mice and 2 apoE-null mice were prepared. The tissue was incubated with HRP at defined concentrations containing albumin (4g/100 mL) in PBS in a slow-shaking plate for 24 hours before fixation with 2.5% glutaraldehyde. HRP reaction product was quantified by relative gray scale with the background image subtracted. This processing was accomplished using an imaging technique with a high spatial resolution and implemented by MATLAB (MathWorks, Inc). Experimental tissue HRP concentrations were calculated from standardized calibration between bulk HRP concentration and relative gray scale.
We quantified IEL permeability in the aorta with lesser and greater curvature in lesion-free areas of 5-month-old apoE-null mice. To recognize specific sites of the lesser and greater curvature, 11 o’clock position of aortic arches were cut during paraffin-embedding process. These cuts determine the 6 or 12 o’clock positions: the 12 o’clock position has greater curvature and 6 o’clock position as lesser curvature as shown in Figure 6. Separate site-specific examination was performed.
The thicknesses of IEL (IEL) and media were measured with microruler adjusted with the same magnification as the tissue sample. From the calibration curve, HRP concentration (milligrams per bulk fluid) in tissue space was calculated relative to free fluid space in the interstitium based on a standard volume fraction.
Mathematical Model and Parameter Estimation
A dynamic mass balance model of the HRP concentration distribution in the arterial wall was applied to analyze the experimental data.10eC12 In this model, HRP concentration is spatially lumped in the intima and spatially distributed in the media (Appendix). The input to this model is the plasma concentration that decreases exponentially. Figure 2 represents schematic diagram for this lumped-distributed model, which incorporates the following parameters: KE is the transport coefficient across the endothelium; PIEL, the permeability coefficient of the IEL; IEL, the thickness of the IEL; DM, the diffusion coefficient in the media; PA, the permeability coefficient between the adventitia and media; KDM, the degradation rate coefficient in the media. The unknown parameters estimated simultaneously were KE, PIEL, DM, and KDM. The parameter estimates were obtained by a 2-step process. First, all unknown parameters, KE, PIEL, DM, KDM, and PA, were estimated. Then, fixing the PA, we estimated KE, PIEL, DM, and KDM for better precision and accuracy. The thicknesses of IEL (IEL) and media were determined by direct measurement and are shown in the Table. These values were used for numerical analysis and parameter estimation. The parameters determining the plasma HRP concentration are the initial concentration Cp0 and time constant t0. The model equations were solved numerically as previously described.15,16 Optimal estimates of the parameters were obtained by minimizing the least-squares difference between experimental and corresponding model-predicted concentrations using an adaptive, nonlinear optimization algorithm, NL2SOL.17
Statistical Analysis
Statistically significant differences in parameter values between groups was determined by the 2-tail t test, assuming that all the measured concentration profiles were independent. The null hypothesis is that the means of the parameter values of any 2 groups are not statically different. All data in the text and figure are presented as mean±SD.
Results
Plasma Cholesterol and HRP Concentrations
Plasma cholesterol concentration in WT mice was significantly less than that observed in the apoE-null mice (92±27 mg/dL [WT] versus 529±111 mg/dL [apoE-null]; P<0.001). To determine whether the decay of HRP concentration after injection differs in WT and apoE-null mice, we quantified the decay by a single-exponential model. From a least-squares fit of the model output to the data, the half-life (t1/2) of HRP was estimated. Between WT and apoE-null mice, t1/2 was not statistically significant different (31 versus 29 minutes, respectively; P>0.3)
Tissue Concentration Profiles with Optimal Estimates of Model Parameter
The experimental HRP concentrations for both circulation times (15 or 30 minutes) compared with the model-predicted HRP concentrations with optimal parameter estimates are shown for each of the 8 groups (Figure 3, WT; Figure 4, apoE-null). On the vertical axes, the relative concentration is the HRP concentration in tissue normalized by its initial value Cp0. The first data point of horizontal axis in these figure represents the intimal HRP concentration.
For the 8 groups, the optimal estimated values of the endothelial transport coefficient (KE) and IEL permeability (PIEL) are shown in Figure 5A for WT and Figure 5B for apoE-null mice. The increased intimal concentration in the apoE-null mice could be explained by an increase in HRP entry (increase in KE), a decrease in HRP exit (decrease in PIEL), or a combination of both. In WT mice, KE for the control group did not differ significantly with age or anatomical location: 3-months (thoracic aorta: 0.12±0.05 minuteseC1; abdominal aorta: 0.09±0.02 minuteseC1); 5 months (thoracic aorta: 0.11±0.03 minuteseC1; abdominal aorta: 0.09±0.03 minuteseC1). KE was actually less in the apoE-null mice compared with WT (P<0.05). KE in the thoracic aorta in the apoE-null mice was less compared with abdominal aorta, especially in 3 and 5 months of thoracic aorta (P<0.05). This finding postulates that KE would be working as barriers for back flow, intima to lumen, of macromolecular transport as a function of age.
No site-specific or age-specific change is evident in IEL permeability in the WT mice. A 78% decrease in PIEL (2.05±1.33 to 0.44±0.15; P<0.001) occurred between 3 and 5 months in the apoE-null groups in the thoracic aorta. Also, a 19% decrease in PIEL (1.13±0.58 to 0.93±0.44; P<0.05) occurred between 3 and 5 months in the abdominal aorta.
Site-Specific Remodeling in Aortic Arch
An increased incidence of atherosclerosis occurred in the lesser curvature of aortic arch of apoE-null mice. Optimal parameter estimates KE and PIEL are shown for each site-specific aortic arch from lesion-free areas of 5-month-old apoE-null mice (Figure 6). The endothelial transport coefficient (KE) was not significantly different between the groups with greater or lesser curvature (0.09±0.03 and 0.08±0.02 minuteseC1, respectively; P>0.05). With the lesser curvature, the IEL permeability coefficient (PIEL) decreased by 59% (0.68±0.38 and 0.28±0.07 e/min for the greater and lesser curvature, respectively; P<0.001).
Discussion
A great deal of literature has focused on the role that EC injury may play in the accumulation of plasm-borne macromolecules in the arterial intima.12,18eC21 However, recent studies indicate that remodeling the extacellular matrix of the arterial wall plays a role in atherogenesis.22eC25 Human atherosclerotic lesions have been shown to contain increased levels of the elastase enzymes, cathepsin S, K, and L,24 as well as a number of matrix metalloproteases.26 Coincident with the increase in elastase expression, there is a decrease in the expression of the endogenous inhibitors for elastases, cystatin C,24 and matrix metalloproteinases, tissue inhibitors of metalloproteases.27 Further emphasizing the role of extracellular remodeling in the atherosclerotic process through the upregulation of protease activity, the cathepsin S/LDL-receptor double-null mouse exhibits less atherosclerosis than the LDL-receptoreCnull mouse.28 In contrast, overexpression29 and underexpression30 of tissue inhibitor of metalloprotease-1 in apoE-null mouse results in decreased atherogenesis, suggesting that the effects of protease activation on atherogenesis may be distinct from lesion progression. Based on these observations, we hypothesize that remodeling of the endothelium or IEL could have a significant role in the accumulation of plasma-borne macromolecules in the intima before atherosclerotic lesion development.
To test whether endothelial or IEL remodeling preceded lesion development in apoE-null mice, we systematically quantified endothelial and IEL permeabilities in the aorta at key anatomical locations without evidence of lesion formation in WT and apoE-null mice at 3 and 5 months of age. An advantage of the apoE-null mouse model is that its predilection for atherosclerotic lesion development has been well defined at the anatomical sites of the aorta in this study.31,32 The aorta with lesser curvature exhibits more atherosclerosis than the aorta with greater curvature.33eC35
The rate and extent of lesion development is less with distance from the thoracic aorta to the abdominal aorta.31 Corresponding to the extent and timing of atherogenesis at each of these sites in apoE-null mice between 3 and 5 months of age, decreases in IEL permeability, as indicated by PIEL, were observed. No changes in endothelial (KE) or IEL (PIEL) permeability were observed in WT mice. Unexpectedly, KE in the 5-month thoracic aorta of the apoE-null mice was less than KE at 3 months in the apoE-null and any time point in the wild-type mice. This means that an increase in the concentration of plasma-borne macromolecules cannot be caused by transport across the EC layer.
In our previous studies, estimates of transport parameters that governed macromolecular movement in the artery wall were based on concentration profiles after a single circulation time. However, in this study, we obtained concentration profiles after 15- and 30-minute circulation times. Consequently, our mathematical model could distinguish degradation (or binding) processes for the intima and media. The model output fit all the concentration data when degradation occurred only in the media (see the online data supplement available at http://circres.ahajournals.org). That degradation in the intima is not consistent with the data suggests that the increased accumulation of HRP in the intima is not attributable to increased HRP degradation (or binding). Rather, the increased accumulation is a true reflection of a decrease in the ability of HRP in a free fluid space to exit the intima because of a decrease in the permeability of the IEL.
We did not find a change in the IEL thickness (IEL) at either ages or different anatomical regions (Table). However, a lack of significant change in the thickness of the IEL does not rule out IEL remodeling. Potential remodeling that could occur in the apoE-null mice includes increased collagen fibrils covering the fenestrae of the elastic lamina and/or remodeling of the elastic layers such that the total area of fenestrae is decreased.
To further elucidate the importance of remodeling of the IEL in atherogenesis, we compared endothelial and IEL permeability in the greater and lesser curvature of the aorta in apoE-null mice. The lesser curvature of the aorta exhibits greater atherosclerosis than the greater curvature,33,35 presumably because of the increased expression of the vascular cell adhesion molecule on the surface of the endothelium covering the lesser curvature.
This study provides evidence that the early increase in macromolecular accumulation in the arterial intima in response to hyperlipidemia may be associated with pathological remodeling of the IEL that precedes alteration in EC function. It is important to note that our conclusions are limited by the validity of the mathematical model to which we fit our data and from which we obtained our permeability coefficients for the endothelium and IEL. That said, our data would suggest that an intricate interplay may occur between the anatomical layers of the artery wall in the setting of hyperlipidemia. We hypothesize that (1) the smooth muscle cells or fibroblasts in the tunica media or the endothelium may initiate remodeling of the IEL because of modulation of protease expression in these cells by LDL,36 (2) the remodeling leads to a decrease in IEL permeability, (3) the decrease in IEL permeability leads to an increase in the residence of time of LDL in the intima,37 (4) the increase in the residence time of LDL in the intima increases the probability of LDL oxidation occurring,38,39 and (5) the presence of oxidized LDL in the intima causes vascular cell adhesion molecule expression in the overlying endothelium.40 Clearly this hypothesis, as well as the transport processes that regulate macromolecular movement into and across the artery wall, warrants further study.
Appendix
The macromolecular concentration in the intima changes with time based on transport processes across two boundaries: the EC layer and IEL:
where KE is the EC transport coefficient and PIEL is the IEL permeability coefficient. The input plasma concentration decreases exponentially:
In the media, partial differential governing equation was implemented, considering only diffusion effect but not convection effect. Usually the convection in the media region is negligible because pressure driven effects are blocked by internal elastic laminar barrier.11
where DM is the tracer diffusion coefficient and the degradation coefficient in the media is KDM.
The boundary conditions for media region were implemented with balancing macromolecule transport through IEL and adventitia.
where the permeability PIEL of the IEL to tracer is related to IEL thickness IEL as PIEL=KIELIEL. At the boundary between the adventitia and the media, the diffusion flux from the former must equal that of the latter so that
where PA is the effective permeability coefficient between adventitia and media. Initially (t=0), all concentrations in the intima and media are 0.
Acknowledgments
This work was supported by a grant from the Whitaker Foundation (to M.S.P.).
References
Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med. 1999; 340: 115eC126.
Glass CK, Witztum JL. Atherosclerosis. The road ahead. Cell. 2001; 104: 503eC516.
Osterud B, Bjorklid E. Role of monocytes in atherogenesis. Physiol Rev. 2003; 83: 1069eC1112.
Wong LC, Langille BL. Developmental remodeling of the internal elastic lamina of rabbit arteries: effect of blood flow. Circ Res. 1996; 78: 799eC805.
Tada S, Tarbell JM. The internal elastic lamina affects the distribution of macromolecules in the arterial wall: a computational study. Am J Physiol Heart Circ Physiol. 2004; 287: H905eCH913.
Briones AM, Gonzalez JM, Somoza B, Giraldo J, Daly CJ, Vila E, Gonzalez MC, McGrath JC, Arribas SM. Role of elastin in spontaneously hypertensive rat small mesenteric artery remodelling. J Physiol. 2003; 552: 185eC195.
Tada S, Tarbell JM. Fenestral pore size in the internal elastic lamina affects transmural flow distribution in the artery wall. Ann Biomed Eng. 2001; 29: 456eC466.
Penn MS, Chisolm GM. Relation between lipopolysaccharide-induced endothelial cell injury and entry of macromolecules into the rat aorta in vivo. Circ Res. 1991; 68: 1259eC1269.
Penn MS, Koelle MR, Schwartz SM, Chisolm GM. Visualization and quantification of transmural concentration profiles of macromolecules across the arterial wall. Circ Res. 1990; 67: 11eC22.
Penn MS, Saidel GM, Chisolm GM. Vascular injury by endotoxin: changes in macromolecular transport parameters in rat aortas in vivo. Am J Physiol. 1992; 262: H1563eCH1571.
Penn MS, Saidel GM, Chisolm GM. Relative significance of endothelium and internal elastic lamina in regulating the entry of macromolecules into arteries in vivo. Circ Res. 1994; 74: 74eC82.
Rangaswamy S, Penn MS, Saidel GM, Chisolm GM. Exogenous oxidized low-density lipoprotein injures and alters the barrier function of endothelium in rats in vivo. Circ Res. 1997; 80: 37eC44.
Penn MS, Rangaswamy S, Saidel GM, Chisolm GM. Macromolecular transport in the arterial intima: comparison of chronic and acute injuries. Am J Physiol. 1997; 272: H1560eC1570.
Avrameas S, Guilbert B. Enzyme-immunoassay for the measurement of antigens using peroxidase conjugates. Biochimie. 1972; 54: 837eC842.
Schiesser WE, Sikbi CA, eds. Computational Transport Phenomena. New York: Cambridge University Press; 1997.
Hindmarsh AC. ODEPACK, a systemized collection of ode solvers. In: Stepleman R, ed. Scientific Computing. Amsterdam: North-Holland Pub; 1983: 509eC516.
Dennis JE, Gay DM, Welsch RE. Algorithm 573 NL2SOL: an adaptive nonlinear least-square algorithm [E4]. ACM Trans Math Softw. 1981; 7: 369eC383.
Hessler JR, Morel DW, Lewis LJ, Chisolm GM. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis. 1983; 3: 215eC222.
Lin SJ, Jan KM, Schuessler G, Weinbaum S, Chien S. Enhanced macromolecular permeability of aortic endothelial cells in association with mitosis. Atherosclerosis. 1988; 73: 223eC232.
Lin SJ, Jan KM, Chien S. Role of dying endothelial cells in transendothelial macromolecular transport. Arteriosclerosis. 1990; 10: 703eC709.
Ross R. Mechanisms of atherosclerosis—a review. Adv Nephrol Necker Hosp. 1990; 19: 79eC86.
Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251eC262.
Torzewski M, Suriyaphol P, Paprotka K, Spath L, Ochsenhirt V, Schmitt A, Han SR, Husmann M, Gerl VB, Bhakdi S, Lackner KJ. Enzymatic modification of low-density lipoprotein in the arterial wall. A new role for plasmin and matrix metalloproteinases in atherogenesis. Arterioscler Thromb Vasc Biol. 2004; 24: 2130eC2136.
Liu J, Sukhova GK, Sun JS, Xu WH, Libby P, Shi GP. Lysosomal cysteine proteases in atherosclerosis. Arterioscler Thromb Vasc Biol. 2004; 24: 1359eC1366.
Wilson D, Massaeli H, Russell JC, Pierce GN, Zahradka P. Low matrix metalloproteinase levels precede vascular lesion formation in the JCR: LA-cp rat. Mol Cell Biochem. 2003; 249: 151eC155.
Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest. 1994; 94: 2493eC2503.
George SJ, Johnson JL, Angelini GD, Newby AC, Baker AH. Adenovirus-mediated gene transfer of the human TIMP-1 gene inhibits smooth muscle cell migration and neointimal formation in human saphenous vein. Hum Gene Ther. 1998; 9: 867eC877.
Sukhova GK, Zhang Y, Pan JH, Wada Y, Yamamoto T, Naito M, Kodama T, Tsimikas S, Witztum JL, Lu ML, Sakara Y, Chin MT, Libby P, Shi GP. Deficiency of cathepsin S reduces atherosclerosis in LDL receptor-deficient mice. J Clin Invest. 2003; 111: 897eC906.
Rouis M, Adamy C, Duverger N, Lesnik P, Horellou P, Moreau M, Emmanuel F, Caillaud JM, Laplaud PM, Dachet C, Chapman MJ. Adenovirus-mediated overexpression of tissue inhibitor of metalloproteinase-1 reduces atherosclerotic lesions in apolipoprotein E-deficient mice. Circulation. 1999; 100: 533eC540.
Silence J, Collen D, Lijnen HR. Reduced atherosclerotic plaque but enhanced aneurysm formation in mice with inactivation of the tissue inhibitor of metalloproteinase-1 (TIMP-1) gene. Circ Res. 2002; 90: 897eC903.
Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. ApoE-deficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb. 1994; 14: 133eC140.
Reddick RL, Zhang SH, Maeda N. Atherosclerosis in mice lacking apo E. Evaluation of lesional development and progression. Arterioscler Thromb. 1994; 14: 141eC147.
Iiyama K, Hajra L, Iiyama M, Li H, DiChiara M, Medoff BD, Cybulsky MI. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ Res. 1999; 85: 199eC207.
Nakashima Y, Raines EW, Plump AS, Breslow JL, Ross R. Upregulation of VCAM-1 and ICAM-1 at atherosclerosis-prone sites on the endothelium in the ApoE-deficient mouse. Arterioscler Thromb Vasc Biol. 1998; 18: 842eC851.
Gerrity RG. The role of the monocyte in atherogenesis. I. Transition of blood-borne monocytes into foam cells in fatty lesions. Am J Pathol. 1981; 103: 181eC190.
Wilson D, Massaeli H, Pierce GN, Zahradka P. Native and minimally oxidized low density lipoprotein depress smooth muscle matrix metalloproteinase levels. Mol Cell Biochem. 2003; 249: 141eC149.
Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits. II. Selective retention of LDL vs. selective increases in LDL permeability in susceptible sites of arteries. Arteriosclerosis. 1989; 9: 908eC918.
Parthasarathy S, Steinberg D, Witztum JL. The role of oxidized low-density lipoproteins in the pathogenesis of atherosclerosis. Annu Rev Med. 1992; 43: 219eC225.
Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989; 320: 915eC924.
Khan BV, Parthasarathy SS, Alexander RW, Medford RM. Modified low density lipoprotein and its constituents augment cytokine-activated vascular cell adhesion molecule-1 gene expression in human vascular endothelial cells. J Clin Invest. 1995; 95: 1262eC1270.(Kwangdeok Lee, Farhad For)
Departments of Cardiovascular Medicine and Cell Biology (K.L., F.F., M.S.P.), Cleveland Clinic Foundation, Ohio.
Abstract
Early atherosclerosis is characterized by the accumulation of plasma-borne macromolecules (eg, low-density lipoproteins) in the arterial intima, which is bordered by endothelial cells (EC) and the internal elastic lamina (IEL). This accumulation is believed to be secondary to increased EC permeability. We hypothesized that a decrease in IEL permeability may precede lesion development and contribute to macromolecular accumulation. To test this hypothesis, we quantified EC and IEL permeability in lesion-free areas of the thoracic and abdominal aortas of chow-fed C57BL/6 control and atherosclerotic-prone apolipoprotein E (apoE)-null mice at 3 and 5 months of age. Between 3 and 5 months of age, apoE-null mice begin to develop atherosclerotic lesions in the thoracic aorta. No significant differences in EC and IEL permeability were observed at either time in C57BL/6 control mice. In contrast, 78% and 19% decreases in IEL permeability of the thoracic aorta and abdominal aorta, respectively, were observed between 3 to 5 months of age in apoE-null mice (thoracic: 2.05±1.33 and 0.44±0.15 e/min, P<0.001; abdominal: 1.13±0.58 and 0.93±0.44 e/min, P<0.05). To further determine whether decreased IEL permeability is linked with atherosclerotic lesion development, we quantified IEL permeability in the greater and lesser curvature of the aortic arch. In apoE-null mice, the lesser curvature of the aortic arch develops lesions before the greater curvature. We found a significant and sustained decrease (59%) in IEL permeability in the lesser curvature of the aortic arch compared with the greater curvature. These data suggest that atherogenesis involves the pathological remodeling of the IEL, not the endothelium before lesion development. This remodeling may be attributable to local responses of the endothelium and smooth muscle cells to hyperlipidemia.
Key Words: atherogenesis extracellular matrix mathematical modeling endothelial dysfunction
Introduction
Early atherosclerotic lesions are characterized by an abnormal accumulation of lipoproteins within the intima in large arteries.1eC3 This abnormal accumulation increases the residence time of low-density lipoprotein (LDL) in the intima, ultimately leading to oxidation of the LDL particles and inflammation. Despite the importance of the initial increase in LDL in the arterial intima, the critical mechanisms involved in early-stage atherosclerosis, or atherogenesis, are not fully understood.
Two distinct tissue layers line the arterial intima, the site of early lipid accumulation: the endothelial cell (EC) layer on the luminal side and the internal elastic layer on its abluminal side. Much literature has focused on the EC layer as a barrier to macromolecular entry; however, recent studies have suggested that internal elastic lamina (IEL) permeability is not static and could contribute to pathological changes in the artery wall.4eC7
We have used horseradish peroxidase (HRP) (44 kDa) as a traceable macromolecule because it has intrinsic peroxidase activity, allowing its concentration in arterial tissue to be quantified,8eC12 and because its size is on the same order of magnitude with growth factors of interest. In prior studies, we have analyzed the arterial-wall transport of HRP in a model consisting of a spatially lumped intima (compartment) and a spatially distributed media.13 Using this approach, we have demonstrated dynamic changes in IEL permeability during endotoxemia that may contribute to vascular permeability during sepsis.10,13
Because abnormal macromolecular accumulation in the intima could be secondary to either increased EC permeability or decreased permeability of the IEL, these observations led us to hypothesize that IEL remodeling could precede atherogenesis. In this study, we quantified the EC and IEL permeability of the macromolecular tracer in thoracic and abdominal aorta separately as a function of age in C57BL/6 (wild-type [WT]) and apolipoprotein E (apoE)-null mice to determine whether IEL remodeling precedes the atherogenic process.
Materials and Methods
Animal Preparation
The Animal Research Committee approved all animal protocols, and all animals were housed in the Association for Accreditation of Laboratory Animal Care animal facility of Cleveland Clinic Foundation. The WT normal (3-month [n=10] and 5-month old [n=11]) and apoE-null mice (3-month [n=8] and 5-month old [n=10]) were anesthetized with 1 mL of ketamineeCHCl (100 mg/mL), 1.5 mL of saline, and 0.3 mL of xylazine (20 mg/mL) and injected for 15 minutes or 30 minutes with 0.1 mL HRP (10 mg/mL) via a jugular vein, cannulated with PE-10 tubing, to make a total 1 mg/mL HRP in the plasma. Numbers of tissue profiles (ns) of WT normal mice were ns=86 (3-month abdominal), ns=99 (3-month thoracic), ns=100 (5-month abdominal), and ns=101 (5-month thoracic aorta). Numbers of tissue profiles of apoE-null mice were ns=50 (3-month abdominal), ns=79 (3-month thoracic), ns=74 (5-month abdominal), ns=90 (5-month thoracic aorta), ns=49 (lesser curvature), and ns=41 (greater curvature of aortic arch).
Quantifying HRP Concentration in Plasma and Tissue
Representative tissue images of apoE-null mice are shown in 3-month abdominal (Figure 1A), 3-month thoracic (Figure 1B), 5-month abdominal (Figure 1C), and 5-month thoracic aorta (Figure 1D). HRP concentration profiles we obtained in WT and apoE-null mice (Jackson Laboratories, Bar Harbor, Me) 15 and 30 minutes after HRP injection using previously published methods.8,11 Briefly, animals were anesthetized and injected with HRP (50 mg/mL) via the jugular vein. Blood samples collected as a function of time after HRP injection were obtained by cardiac puncture. The HRP concentration in plasma was determined by measuring the colored reaction product of o-dianisidine and H2O2 as a substrate for peroxidase reaction.14
The HRP concentration profiles across tissue were obtained as described by Penn et al.9 Following HRP circulation, PBS was injected into left ventricle followed by perfusion of 10 mL of 2.5% ice-cold glutaraldehyde via a left ventricle. After 5 minutes, the aorta was removed and put into 2.5% ice-cold glutaraldehyde during 4 hours, followed by washing with PBS and water for 30 minutes in each solution. Tissue samples were reacted with 3,3'-diaminobenzidine and H2O2 in imidazole buffer (0.1 N, pH 7.6) for 30 minutes. The tissue was washed with PBS and serially dehydrated with graded ethanol solutions up to 95%, embedded with paraffin, and serially sectioned 5 e using rotary microtome.
Plasma cholesterol concentration levels were measured using infinity Cholesterol Reagent with Cholesterol standard solutions from Sigma.
Data Acquisition and Image Processing
To determine HRP equilibrium distribution coefficients in the tissue, 8 to 10 cylindrical rings sectioned from fresh aorta from 2 WT mice and 2 apoE-null mice were prepared. The tissue was incubated with HRP at defined concentrations containing albumin (4g/100 mL) in PBS in a slow-shaking plate for 24 hours before fixation with 2.5% glutaraldehyde. HRP reaction product was quantified by relative gray scale with the background image subtracted. This processing was accomplished using an imaging technique with a high spatial resolution and implemented by MATLAB (MathWorks, Inc). Experimental tissue HRP concentrations were calculated from standardized calibration between bulk HRP concentration and relative gray scale.
We quantified IEL permeability in the aorta with lesser and greater curvature in lesion-free areas of 5-month-old apoE-null mice. To recognize specific sites of the lesser and greater curvature, 11 o’clock position of aortic arches were cut during paraffin-embedding process. These cuts determine the 6 or 12 o’clock positions: the 12 o’clock position has greater curvature and 6 o’clock position as lesser curvature as shown in Figure 6. Separate site-specific examination was performed.
The thicknesses of IEL (IEL) and media were measured with microruler adjusted with the same magnification as the tissue sample. From the calibration curve, HRP concentration (milligrams per bulk fluid) in tissue space was calculated relative to free fluid space in the interstitium based on a standard volume fraction.
Mathematical Model and Parameter Estimation
A dynamic mass balance model of the HRP concentration distribution in the arterial wall was applied to analyze the experimental data.10eC12 In this model, HRP concentration is spatially lumped in the intima and spatially distributed in the media (Appendix). The input to this model is the plasma concentration that decreases exponentially. Figure 2 represents schematic diagram for this lumped-distributed model, which incorporates the following parameters: KE is the transport coefficient across the endothelium; PIEL, the permeability coefficient of the IEL; IEL, the thickness of the IEL; DM, the diffusion coefficient in the media; PA, the permeability coefficient between the adventitia and media; KDM, the degradation rate coefficient in the media. The unknown parameters estimated simultaneously were KE, PIEL, DM, and KDM. The parameter estimates were obtained by a 2-step process. First, all unknown parameters, KE, PIEL, DM, KDM, and PA, were estimated. Then, fixing the PA, we estimated KE, PIEL, DM, and KDM for better precision and accuracy. The thicknesses of IEL (IEL) and media were determined by direct measurement and are shown in the Table. These values were used for numerical analysis and parameter estimation. The parameters determining the plasma HRP concentration are the initial concentration Cp0 and time constant t0. The model equations were solved numerically as previously described.15,16 Optimal estimates of the parameters were obtained by minimizing the least-squares difference between experimental and corresponding model-predicted concentrations using an adaptive, nonlinear optimization algorithm, NL2SOL.17
Statistical Analysis
Statistically significant differences in parameter values between groups was determined by the 2-tail t test, assuming that all the measured concentration profiles were independent. The null hypothesis is that the means of the parameter values of any 2 groups are not statically different. All data in the text and figure are presented as mean±SD.
Results
Plasma Cholesterol and HRP Concentrations
Plasma cholesterol concentration in WT mice was significantly less than that observed in the apoE-null mice (92±27 mg/dL [WT] versus 529±111 mg/dL [apoE-null]; P<0.001). To determine whether the decay of HRP concentration after injection differs in WT and apoE-null mice, we quantified the decay by a single-exponential model. From a least-squares fit of the model output to the data, the half-life (t1/2) of HRP was estimated. Between WT and apoE-null mice, t1/2 was not statistically significant different (31 versus 29 minutes, respectively; P>0.3)
Tissue Concentration Profiles with Optimal Estimates of Model Parameter
The experimental HRP concentrations for both circulation times (15 or 30 minutes) compared with the model-predicted HRP concentrations with optimal parameter estimates are shown for each of the 8 groups (Figure 3, WT; Figure 4, apoE-null). On the vertical axes, the relative concentration is the HRP concentration in tissue normalized by its initial value Cp0. The first data point of horizontal axis in these figure represents the intimal HRP concentration.
For the 8 groups, the optimal estimated values of the endothelial transport coefficient (KE) and IEL permeability (PIEL) are shown in Figure 5A for WT and Figure 5B for apoE-null mice. The increased intimal concentration in the apoE-null mice could be explained by an increase in HRP entry (increase in KE), a decrease in HRP exit (decrease in PIEL), or a combination of both. In WT mice, KE for the control group did not differ significantly with age or anatomical location: 3-months (thoracic aorta: 0.12±0.05 minuteseC1; abdominal aorta: 0.09±0.02 minuteseC1); 5 months (thoracic aorta: 0.11±0.03 minuteseC1; abdominal aorta: 0.09±0.03 minuteseC1). KE was actually less in the apoE-null mice compared with WT (P<0.05). KE in the thoracic aorta in the apoE-null mice was less compared with abdominal aorta, especially in 3 and 5 months of thoracic aorta (P<0.05). This finding postulates that KE would be working as barriers for back flow, intima to lumen, of macromolecular transport as a function of age.
No site-specific or age-specific change is evident in IEL permeability in the WT mice. A 78% decrease in PIEL (2.05±1.33 to 0.44±0.15; P<0.001) occurred between 3 and 5 months in the apoE-null groups in the thoracic aorta. Also, a 19% decrease in PIEL (1.13±0.58 to 0.93±0.44; P<0.05) occurred between 3 and 5 months in the abdominal aorta.
Site-Specific Remodeling in Aortic Arch
An increased incidence of atherosclerosis occurred in the lesser curvature of aortic arch of apoE-null mice. Optimal parameter estimates KE and PIEL are shown for each site-specific aortic arch from lesion-free areas of 5-month-old apoE-null mice (Figure 6). The endothelial transport coefficient (KE) was not significantly different between the groups with greater or lesser curvature (0.09±0.03 and 0.08±0.02 minuteseC1, respectively; P>0.05). With the lesser curvature, the IEL permeability coefficient (PIEL) decreased by 59% (0.68±0.38 and 0.28±0.07 e/min for the greater and lesser curvature, respectively; P<0.001).
Discussion
A great deal of literature has focused on the role that EC injury may play in the accumulation of plasm-borne macromolecules in the arterial intima.12,18eC21 However, recent studies indicate that remodeling the extacellular matrix of the arterial wall plays a role in atherogenesis.22eC25 Human atherosclerotic lesions have been shown to contain increased levels of the elastase enzymes, cathepsin S, K, and L,24 as well as a number of matrix metalloproteases.26 Coincident with the increase in elastase expression, there is a decrease in the expression of the endogenous inhibitors for elastases, cystatin C,24 and matrix metalloproteinases, tissue inhibitors of metalloproteases.27 Further emphasizing the role of extracellular remodeling in the atherosclerotic process through the upregulation of protease activity, the cathepsin S/LDL-receptor double-null mouse exhibits less atherosclerosis than the LDL-receptoreCnull mouse.28 In contrast, overexpression29 and underexpression30 of tissue inhibitor of metalloprotease-1 in apoE-null mouse results in decreased atherogenesis, suggesting that the effects of protease activation on atherogenesis may be distinct from lesion progression. Based on these observations, we hypothesize that remodeling of the endothelium or IEL could have a significant role in the accumulation of plasma-borne macromolecules in the intima before atherosclerotic lesion development.
To test whether endothelial or IEL remodeling preceded lesion development in apoE-null mice, we systematically quantified endothelial and IEL permeabilities in the aorta at key anatomical locations without evidence of lesion formation in WT and apoE-null mice at 3 and 5 months of age. An advantage of the apoE-null mouse model is that its predilection for atherosclerotic lesion development has been well defined at the anatomical sites of the aorta in this study.31,32 The aorta with lesser curvature exhibits more atherosclerosis than the aorta with greater curvature.33eC35
The rate and extent of lesion development is less with distance from the thoracic aorta to the abdominal aorta.31 Corresponding to the extent and timing of atherogenesis at each of these sites in apoE-null mice between 3 and 5 months of age, decreases in IEL permeability, as indicated by PIEL, were observed. No changes in endothelial (KE) or IEL (PIEL) permeability were observed in WT mice. Unexpectedly, KE in the 5-month thoracic aorta of the apoE-null mice was less than KE at 3 months in the apoE-null and any time point in the wild-type mice. This means that an increase in the concentration of plasma-borne macromolecules cannot be caused by transport across the EC layer.
In our previous studies, estimates of transport parameters that governed macromolecular movement in the artery wall were based on concentration profiles after a single circulation time. However, in this study, we obtained concentration profiles after 15- and 30-minute circulation times. Consequently, our mathematical model could distinguish degradation (or binding) processes for the intima and media. The model output fit all the concentration data when degradation occurred only in the media (see the online data supplement available at http://circres.ahajournals.org). That degradation in the intima is not consistent with the data suggests that the increased accumulation of HRP in the intima is not attributable to increased HRP degradation (or binding). Rather, the increased accumulation is a true reflection of a decrease in the ability of HRP in a free fluid space to exit the intima because of a decrease in the permeability of the IEL.
We did not find a change in the IEL thickness (IEL) at either ages or different anatomical regions (Table). However, a lack of significant change in the thickness of the IEL does not rule out IEL remodeling. Potential remodeling that could occur in the apoE-null mice includes increased collagen fibrils covering the fenestrae of the elastic lamina and/or remodeling of the elastic layers such that the total area of fenestrae is decreased.
To further elucidate the importance of remodeling of the IEL in atherogenesis, we compared endothelial and IEL permeability in the greater and lesser curvature of the aorta in apoE-null mice. The lesser curvature of the aorta exhibits greater atherosclerosis than the greater curvature,33,35 presumably because of the increased expression of the vascular cell adhesion molecule on the surface of the endothelium covering the lesser curvature.
This study provides evidence that the early increase in macromolecular accumulation in the arterial intima in response to hyperlipidemia may be associated with pathological remodeling of the IEL that precedes alteration in EC function. It is important to note that our conclusions are limited by the validity of the mathematical model to which we fit our data and from which we obtained our permeability coefficients for the endothelium and IEL. That said, our data would suggest that an intricate interplay may occur between the anatomical layers of the artery wall in the setting of hyperlipidemia. We hypothesize that (1) the smooth muscle cells or fibroblasts in the tunica media or the endothelium may initiate remodeling of the IEL because of modulation of protease expression in these cells by LDL,36 (2) the remodeling leads to a decrease in IEL permeability, (3) the decrease in IEL permeability leads to an increase in the residence of time of LDL in the intima,37 (4) the increase in the residence time of LDL in the intima increases the probability of LDL oxidation occurring,38,39 and (5) the presence of oxidized LDL in the intima causes vascular cell adhesion molecule expression in the overlying endothelium.40 Clearly this hypothesis, as well as the transport processes that regulate macromolecular movement into and across the artery wall, warrants further study.
Appendix
The macromolecular concentration in the intima changes with time based on transport processes across two boundaries: the EC layer and IEL:
where KE is the EC transport coefficient and PIEL is the IEL permeability coefficient. The input plasma concentration decreases exponentially:
In the media, partial differential governing equation was implemented, considering only diffusion effect but not convection effect. Usually the convection in the media region is negligible because pressure driven effects are blocked by internal elastic laminar barrier.11
where DM is the tracer diffusion coefficient and the degradation coefficient in the media is KDM.
The boundary conditions for media region were implemented with balancing macromolecule transport through IEL and adventitia.
where the permeability PIEL of the IEL to tracer is related to IEL thickness IEL as PIEL=KIELIEL. At the boundary between the adventitia and the media, the diffusion flux from the former must equal that of the latter so that
where PA is the effective permeability coefficient between adventitia and media. Initially (t=0), all concentrations in the intima and media are 0.
Acknowledgments
This work was supported by a grant from the Whitaker Foundation (to M.S.P.).
References
Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med. 1999; 340: 115eC126.
Glass CK, Witztum JL. Atherosclerosis. The road ahead. Cell. 2001; 104: 503eC516.
Osterud B, Bjorklid E. Role of monocytes in atherogenesis. Physiol Rev. 2003; 83: 1069eC1112.
Wong LC, Langille BL. Developmental remodeling of the internal elastic lamina of rabbit arteries: effect of blood flow. Circ Res. 1996; 78: 799eC805.
Tada S, Tarbell JM. The internal elastic lamina affects the distribution of macromolecules in the arterial wall: a computational study. Am J Physiol Heart Circ Physiol. 2004; 287: H905eCH913.
Briones AM, Gonzalez JM, Somoza B, Giraldo J, Daly CJ, Vila E, Gonzalez MC, McGrath JC, Arribas SM. Role of elastin in spontaneously hypertensive rat small mesenteric artery remodelling. J Physiol. 2003; 552: 185eC195.
Tada S, Tarbell JM. Fenestral pore size in the internal elastic lamina affects transmural flow distribution in the artery wall. Ann Biomed Eng. 2001; 29: 456eC466.
Penn MS, Chisolm GM. Relation between lipopolysaccharide-induced endothelial cell injury and entry of macromolecules into the rat aorta in vivo. Circ Res. 1991; 68: 1259eC1269.
Penn MS, Koelle MR, Schwartz SM, Chisolm GM. Visualization and quantification of transmural concentration profiles of macromolecules across the arterial wall. Circ Res. 1990; 67: 11eC22.
Penn MS, Saidel GM, Chisolm GM. Vascular injury by endotoxin: changes in macromolecular transport parameters in rat aortas in vivo. Am J Physiol. 1992; 262: H1563eCH1571.
Penn MS, Saidel GM, Chisolm GM. Relative significance of endothelium and internal elastic lamina in regulating the entry of macromolecules into arteries in vivo. Circ Res. 1994; 74: 74eC82.
Rangaswamy S, Penn MS, Saidel GM, Chisolm GM. Exogenous oxidized low-density lipoprotein injures and alters the barrier function of endothelium in rats in vivo. Circ Res. 1997; 80: 37eC44.
Penn MS, Rangaswamy S, Saidel GM, Chisolm GM. Macromolecular transport in the arterial intima: comparison of chronic and acute injuries. Am J Physiol. 1997; 272: H1560eC1570.
Avrameas S, Guilbert B. Enzyme-immunoassay for the measurement of antigens using peroxidase conjugates. Biochimie. 1972; 54: 837eC842.
Schiesser WE, Sikbi CA, eds. Computational Transport Phenomena. New York: Cambridge University Press; 1997.
Hindmarsh AC. ODEPACK, a systemized collection of ode solvers. In: Stepleman R, ed. Scientific Computing. Amsterdam: North-Holland Pub; 1983: 509eC516.
Dennis JE, Gay DM, Welsch RE. Algorithm 573 NL2SOL: an adaptive nonlinear least-square algorithm [E4]. ACM Trans Math Softw. 1981; 7: 369eC383.
Hessler JR, Morel DW, Lewis LJ, Chisolm GM. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis. 1983; 3: 215eC222.
Lin SJ, Jan KM, Schuessler G, Weinbaum S, Chien S. Enhanced macromolecular permeability of aortic endothelial cells in association with mitosis. Atherosclerosis. 1988; 73: 223eC232.
Lin SJ, Jan KM, Chien S. Role of dying endothelial cells in transendothelial macromolecular transport. Arteriosclerosis. 1990; 10: 703eC709.
Ross R. Mechanisms of atherosclerosis—a review. Adv Nephrol Necker Hosp. 1990; 19: 79eC86.
Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251eC262.
Torzewski M, Suriyaphol P, Paprotka K, Spath L, Ochsenhirt V, Schmitt A, Han SR, Husmann M, Gerl VB, Bhakdi S, Lackner KJ. Enzymatic modification of low-density lipoprotein in the arterial wall. A new role for plasmin and matrix metalloproteinases in atherogenesis. Arterioscler Thromb Vasc Biol. 2004; 24: 2130eC2136.
Liu J, Sukhova GK, Sun JS, Xu WH, Libby P, Shi GP. Lysosomal cysteine proteases in atherosclerosis. Arterioscler Thromb Vasc Biol. 2004; 24: 1359eC1366.
Wilson D, Massaeli H, Russell JC, Pierce GN, Zahradka P. Low matrix metalloproteinase levels precede vascular lesion formation in the JCR: LA-cp rat. Mol Cell Biochem. 2003; 249: 151eC155.
Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest. 1994; 94: 2493eC2503.
George SJ, Johnson JL, Angelini GD, Newby AC, Baker AH. Adenovirus-mediated gene transfer of the human TIMP-1 gene inhibits smooth muscle cell migration and neointimal formation in human saphenous vein. Hum Gene Ther. 1998; 9: 867eC877.
Sukhova GK, Zhang Y, Pan JH, Wada Y, Yamamoto T, Naito M, Kodama T, Tsimikas S, Witztum JL, Lu ML, Sakara Y, Chin MT, Libby P, Shi GP. Deficiency of cathepsin S reduces atherosclerosis in LDL receptor-deficient mice. J Clin Invest. 2003; 111: 897eC906.
Rouis M, Adamy C, Duverger N, Lesnik P, Horellou P, Moreau M, Emmanuel F, Caillaud JM, Laplaud PM, Dachet C, Chapman MJ. Adenovirus-mediated overexpression of tissue inhibitor of metalloproteinase-1 reduces atherosclerotic lesions in apolipoprotein E-deficient mice. Circulation. 1999; 100: 533eC540.
Silence J, Collen D, Lijnen HR. Reduced atherosclerotic plaque but enhanced aneurysm formation in mice with inactivation of the tissue inhibitor of metalloproteinase-1 (TIMP-1) gene. Circ Res. 2002; 90: 897eC903.
Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. ApoE-deficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb. 1994; 14: 133eC140.
Reddick RL, Zhang SH, Maeda N. Atherosclerosis in mice lacking apo E. Evaluation of lesional development and progression. Arterioscler Thromb. 1994; 14: 141eC147.
Iiyama K, Hajra L, Iiyama M, Li H, DiChiara M, Medoff BD, Cybulsky MI. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ Res. 1999; 85: 199eC207.
Nakashima Y, Raines EW, Plump AS, Breslow JL, Ross R. Upregulation of VCAM-1 and ICAM-1 at atherosclerosis-prone sites on the endothelium in the ApoE-deficient mouse. Arterioscler Thromb Vasc Biol. 1998; 18: 842eC851.
Gerrity RG. The role of the monocyte in atherogenesis. I. Transition of blood-borne monocytes into foam cells in fatty lesions. Am J Pathol. 1981; 103: 181eC190.
Wilson D, Massaeli H, Pierce GN, Zahradka P. Native and minimally oxidized low density lipoprotein depress smooth muscle matrix metalloproteinase levels. Mol Cell Biochem. 2003; 249: 141eC149.
Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits. II. Selective retention of LDL vs. selective increases in LDL permeability in susceptible sites of arteries. Arteriosclerosis. 1989; 9: 908eC918.
Parthasarathy S, Steinberg D, Witztum JL. The role of oxidized low-density lipoproteins in the pathogenesis of atherosclerosis. Annu Rev Med. 1992; 43: 219eC225.
Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989; 320: 915eC924.
Khan BV, Parthasarathy SS, Alexander RW, Medford RM. Modified low density lipoprotein and its constituents augment cytokine-activated vascular cell adhesion molecule-1 gene expression in human vascular endothelial cells. J Clin Invest. 1995; 95: 1262eC1270.(Kwangdeok Lee, Farhad For)