Lipid Incorporation Inhibits Src-Dependent Assembly of Fibronectin and Type I Collagen by Vascular Smooth Muscle Cells
A vital role of vascular smooth muscle cells (SMCs) is to stabilize the artery wall by elaborating fibrils of type I collagen. This is especially important in atherosclerotic lesions. However, SMCs in these lesions can be laden with lipids and the impact of this modification on collagen fibril formation is unknown. To address this, we converted human vascular SMCs to a foam cell state by incubating them with either LDL or VLDL. Biochemical markers of a SMC phenotype were preserved. However, microscopic tracking revealed a profound perturbation in the ability of the cells to assemble collagen fibrils, reducing assembly by up to 79%. This dysfunction was mirrored by an inability of smooth muscle foam cells to assemble fibronectin. Lipid-loaded SMCs did not display a generalized defect in the actin cytoskeleton and the formation of vinculin-containing focal adhesion complexes was preserved. However, lipid-loaded SMCs were unable to assemble fibrillar adhesion complexes and clustering of tensin and α5β1 integrin was disordered. Moreover, phosphorylation of tensin, required for fibrillar adhesion complex formation, was suppressed by up to 57%, with a concomitant decrease in activation of Src and FAK and restriction of activated Src to the cell edges. Forced activation of Src-FAK signaling in lipid-engorged SMCs rescued both fibrillar adhesion formation and fibrillogenesis. We conclude that lipid accumulation by SMCs disables the machinery for collagen and fibronectin assembly. This previously unknown relationship between atherogenic lipids and integrin-based signaling could underlie plaque vulnerability.
Accumulation of lipids within the artery wall is a fundamental feature of atherosclerosis.1 Within vascular cells, lipids accumulate as cytoplasmic droplets of cholesterol esters and triglycerides; cells with an abundance of these droplets are termed foam cells. Macrophages are considered the primary source of foam cells in atherosclerosis. However, it has been established for many years that atherosclerotic lesions also contain abundant foam cells with a smooth muscle cell (SMC) identity.2–4 Consistent with this, SMCs in culture can accrue neutral lipids when challenged with hyperlipidemic serum, LDL, or VLDL.5,6
The fate of lipid-laden SMCs is uncertain and it may be incorrect to assume that it is identical to that of macrophage foam cells. Incorporation of lipids by macrophages will result in cell death once the macrophage’s capacity to esterify the otherwise toxic free cholesterol is exceeded.7 However, SMCs are more resistant than macrophages to the toxicity of free cholesterol.8 Therefore, lipid-loaded SMCs have the potential to reside in the artery wall for extended periods; the consequences of this are unknown.
A vital role played by healthy vascular SMCs is to stabilize the vessel wall through the elaboration of fibrils of type I collagen.9 Type I collagen is secreted from the cell as a soluble protein that must then polymerize into insoluble fibrils to confer mechanical properties. However, in contrast to collagen synthesis and secretion, collagen polymerization has not generally been studied as a process that can be regulated. This is because it has long been recognized that type I collagen can self-assemble into fibrils in the absence of cellular interactions.10 Recently, however, several groups have established that efficient polymerization of type I collagen is in fact highly dependent on cellular interactions and proceeds in an integrin- and cytoskeletal-dependent manner.11–13 Furthermore, type I collagen assembly has been found to be integrated with, and dependent on, fibronectin assembly,11–13 a polymerization process that is entirely dependent on cellular interactions. These findings raise the possibility that the polymerization of type I collagen may not be an irrevocable consequence of collagen secretion but a modifiable process that could be affected by alterations in cell structure or function.
One potential context for altered interplay between vascular SMCs and collagen fibrils is atherosclerosis. Many atherosclerotic lesions are deficient in collagen fibrils with plaque rupture and myocardial infarction as a consequence.9 Although healthy SMCs orchestrate the brisk assembly of collagen fibrils from soluble precursors,12 it is unknown whether SMCs laden with lipids, as found in atherosclerosis, are capable of performing this role. This is an important question given that lesions with high lipid content are precisely those requiring stabilization by fibrous collagen.
In this report, we describe a previously unknown molecular defect caused by lipid accumulation that renders SMCs incapable of efficiently assembling a fibrillar extracellular matrix.
Materials and Methods
An expanded Materials and Methods section is available in the online data supplement at http://circres.ahajournals.org.
Cell Culture and ECM Reagents
Experiments were performed using HITC6 SMCs, a nonimmortal human SMC line isolated from the internal thoracic artery,14,15 and primary cultures of SMCs harvested from internal thoracic artery fragments.12 Soluble type I collagen preparations were generated from rat tail tendons. Human plasma fibronectin was isolated by gelatin-sepharose chromatography.12,16
LDL (Svedberg flotation units, 0 to 12) and VLDL (Svedberg flotation units, 20 to 400) were isolated from hyperlipidemic subjects, recruited from the Lipid Clinic at the London Health Sciences Centre, University Campus (London, Ontario, Canada).
Generation of Smooth Muscle Foam Cells
SMCs were loaded with lipids by incubating for 24 hours with either LDL (150 μg cholesterol/mL media) or VLDL (50 μg cholesterol/mL media) in the presence of bovine milk lipoprotein lipase (0.1 U/mL media) in M199 with 3% FBS. Lipoprotein concentrations were chosen based on prior dose-ranging analyses to yield submaximal accumulation of intracellular lipids.6 Neutral lipid vesicles were visualized by staining cells with oil red O or Bodipy 493/503. Biochemical quantification of cellular lipids was determined as described previously.6 SMC apoptosis was assessed by fluorescence in situ end-labeling of DNA fragments.
Gene Transfer Into Human SMCs
Human SMCs overexpressing constitutively active Src Y527F were generated using a retroviral gene delivery system.17,18 Stable transductants were selected with G418 for 14 days.
Assessment of SMC-Mediated Collagen and Fibronectin Assembly
The assembly of type I collagen and fibronectin fibrils by SMCs was assessed using Texas red-labeled soluble precursors, as described.12,16
HITC6 SMCs were immunostained using monoclonal antibodies to detect vinculin, tensin or α5β1 integrin and a polyclonal antibody to detect phosphorylated Src.
Flow Cytometric Analysis of Surface Integrin Expression
Quantification α5β1 integrin and β1 integrin expression was carried out by indirect immunofluorescence staining and flow cytometry, as described previously.19
Immunoblotting and Immunoprecipitation
Expression of SMC contractile apparatus, adhesion, and signaling proteins was assessed by Western blot analysis with chemiluminescence detection. Phosphorylated tensin was evaluated by tensin immunoprecipitation and immunoblot analysis for phosphotyrosine.
Human Vascular SMCs Challenged With LDL or VLDL Convert to Foam Cells
To study the impact of lipid accumulation on SMC function, we generated stable populations of human SMCs laden with cytoplasmic lipid droplets. This was accomplished by incubating HITC6 SMCs with human LDL or human VLDL, in the presence of lipoprotein lipase. In both LDL and VLDL-rich environments, SMCs accumulated abundant neutral lipid droplets by 24 hours, as demonstrated by staining with Bodipy 495/503 or Oil Red O (Figure 1A). Droplets were distributed throughout the cytoplasm in all cells. There was no difference in cell numbers between control and lipoprotein-incubated cultures and no evidence for lipid-induced cytotoxicity, as assessed by trypan blue exclusion (percentage trypan blue-stained cells: untreated, 2.0±0.4%; LDL-exposed, 2.7±0.4%; VLDL-exposed 2.6±1.0%; P=NS) and lactate hydrogenase release (untreated, 25.8±1.5 U/L; LDL-exposed, 29.5±0.8 U/L; VLDL-exposed, 30.2±3.0 U/L; P=NS). Similarly, the percentage of apoptotic SMCs, as assessed by TUNEL, in lipoprotein-incubated cultures (1.5±0.1%, LDL; 1.3±0.1%, VLDL) was not different from that in control cultures (1.6±0.3, P=NS). Furthermore, engorgement of SMCs with lipid vesicles was reversible; 7 days after washout of lipoproteins from the medium, the size and number of intracellular lipid droplets declined to those of untreated SMCs (data not shown). Primary cultures of human SMCs also accumulated copious lipid droplets when exposed to lipoproteins (Figure I in the online data supplement).
Quantitative droplet analysis revealed that LDL-incubated HITC6 SMCs contained 159±10 lipid vesicles per cell after 24 hours of exposure to lipoproteins. VLDL-incubated SMCs contained somewhat fewer droplets (125±10, P<0.05); however, the droplets were larger than those in LDL-exposed SMCs. As depicted in the frequency histograms of Figure 1B, 4.4% of droplets in LDL-incubated SMCs had a diameter above 0.8 μm and the maximum droplet diameter was 1.8 μm. In contrast, in VLDL-treated SMCs, 23.4% of droplets had a diameter greater than 0.8 μm (P<0.05), with a maximum diameter of 3.6 μm.
These morphological changes were associated with lipoprotein-specific differences in cellular lipid accumulation (Figure 1C). In LDL-incubated SMCs, cholesteryl ester content increased 3-fold (P<0.05), with no significant increase in triglyceride content. In VLDL-incubated SMCs, the increase in cholesteryl ester content was 1.8-fold, whereas triglyceride increased 5-fold (P<0.05 for both).
It has been previously reported that mouse SMCs challenged with cholesterol transdifferentiate into macrophages.20 Therefore, we next determined whether human SMCs incubated with lipoproteins retained a SMC phenotype. As depicted in Figure 1D, SMCs loaded for either 24 or 72 hours with LDL or VLDL displayed no changes in the abundance of heavy caldesmon, smooth muscle α-actin, or calponin-h1, compared to control SMCs not subjected to lipid loading.
Lipid-Loaded SMCs Fail to Assemble a Type I-Rich Collagen Fibril Matrix
The generation of lipid-laden cells that retained a SMC phenotype allowed us to determine whether lipid accumulation imparted defects in SMC-dependent functions. We were particularly interested in determining whether smooth muscle foam cells retained the ability to construct a collagen fibril matrix from soluble precursors, a polymerization process that is vital for vascular stability. To assess this, Texas red-labeled solubilized rat tail collagen was added to SMCs and fibril assembly was microscopically tracked. This approach had the advantage of functionally separating collagen fibril assembly from collagen synthesis and secretion.
Over 24 hours, SMCs not exposed to lipoproteins assembled an elaborate network of collagen fibrils on their surface (Figure 2, top row). In contrast, assembly of collagen on the surface of LDL-exposed smooth muscle foam cells was strikingly impaired (Figure 2, middle row). Quantitative analysis demonstrated that the fraction of apical cell surface area occupied by collagen fibrils was 30.1±6.9% in control cells but only 9.2±5.7% in LDL-incubated SMC foam cells (P<0.001). SMCs incubated with VLDL displayed an even greater defect in collagen fibril assembly (Figure 2, bottom row), with a near-complete abrogation of fibril formation (6.3±3.5%, P<0.001 versus control, P<0.05 versus LDL-incubated SMCs). Smooth muscle foam cells failed to efficiently assemble type I collagen fibrils regardless of whether or not lipoproteins were present in the medium at the time the fibril precursors were added, indicating that the defect was inherent to the SMC and not attributable to an interaction between lipoproteins and collagen precursors in the medium. Lipid accumulation did not impact collagen expression in SMCs, as assessed by quantitative real-time RT-PCR for proα1(I) collagen mRNA (supplemental Figure II).
Lipid-Loaded SMCs Fail to Assemble Fibronectin
We have previously shown that collagen assembly by SMCs requires the fibronectin-binding α5β1 integrin. Consistent with this, Sottile et al have established that collagen fibrils cannot assemble on fibronectin-null cells.13 Therefore, we next determined if the impairment in collagen assembly by lipid-loading was related to a defect in fibronectin assembly. As shown in Figure 3, SMCs not exposed to lipoproteins efficiently converted soluble fibronectin protomers into an elaborate network of fibronectin fibrils. In contrast, smooth muscle foam cells generated with LDL displayed a marked decrease in their ability to assemble fibronectin fibrils (38.3±7.6% versus 16.8±3.4% of the surface area, P<0.001). As with collagen assembly, SMCs incubated with VLDL displayed an even greater defect in assembling fibronectin (10.1±4.5%, P<0.001 versus control, P<0.05 versus LDL-incubated SMCs). Fibronectin mRNA expression was unaffected by lipid loading (supplemental Figure II). Importantly, inhibition of fibronectin assembly was not restricted to lipid-loaded HITC6 SMCs but was also seen with primary SMCs incubated with lipoproteins (supplemental Figure I).
Assembly of Fibrillar Adhesion Complexes Is Impaired in Smooth Muscle Foam Cells
Fundamental to fibronectin assembly is the formation of fibrillar adhesions, a specific protein assembly found on the apical surface of fibronectin-assembling cells. To evaluate fibrillar adhesion complexes in lipid-loaded SMCs, we immunostained for tensin, a primary component of fibrillar adhesions that has a critical role in their formation and the attendant assembly of fibronectin.21 As shown in Figure 4A, control SMCs had abundant tensin-containing fibrillar adhesion complexes, evident as linear accumulations that aligned parallel to the long axis of the cell. However, smooth muscle foam cells had a striking reduction of tensin accumulations, particularly on the apical surface. Residual tensin accumulation was observed along the cell periphery and also in isolated regions of the apical surface. Interestingly, double labeling for tensin and neutral lipid vesicles revealed that tensin-containing complexes were largely absent over areas of pronounced cytoplasmic lipid droplet accumulation but residual tensin complexes were present above lipid-free zones (Figure 4B). Western blot analysis revealed that total tensin expression was not different between control SMCs and LDL- or VLDL-exposed SMCs (Figure 4C), supporting a defect in tensin aggregation.
To determine whether these changes reflected a generalized defect in cytoskeletal architecture, smooth muscle foam cells were stained with TRITC-conjugated phalloidin. This revealed an actin microfilament bundle network similar to that of control SMCs, with the rare exception of subtle bending of actin filaments by large droplets in VLDL-loaded SMCs (supplemental Figure III).
In contrast to fibrillar adhesions, vinculin-containing focal adhesions were unaffected by lipid loading (Figure 4D). Therefore, lipid loading of SMCs leads to a selective inability of the cell to maintain the integrity of fibrillar adhesion complexes, with a redistribution of tensin out of those complexes closest to cytoplasmic lipid droplets.
Clustering of α5β1 Integrin Is Impaired in Smooth Muscle Foam Cells
The primary fibronectin-binding receptor for vascular SMCs is α5β1 integrin and movement of this integrin into fibrillar adhesion complexes is essential for fibronectin assembly.16,22 In unloaded SMCs, α5β1 integrin clustered primarily in streak-like aggregates on the apical surface of the cell, typical of fibrillar adhesion complexes (Figure 5A). However, in smooth muscle foam cells there were substantially fewer α5β1 integrin-containing fibrillar adhesion complexes and those observed were typically short. Double labeling for α5β1 integrin and neutral lipid vesicles revealed that α5β1 integrin-containing fibrillar complexes were least evident in regions of the plasma membrane overlying abundant lipid droplets, whereas interrupted accumulations were evident above droplet-free cytoplasm (Figure 5A, asterisk). Flow cytometry revealed that surface expression of α5β1 integrin was not different in smooth muscle foam cells generated by incubation with LDL (not shown) or VLDL than in control cells (Figure 5B). Likewise, there was no evidence for a decrease in surface expression of the total β1 integrin subunit pool. Thus, as with tensin, local accumulation of lipids within SMCs impairs the ability of nearby α5β1 integrin molecules to cluster into fibrillar adhesion complexes.
Accumulation of Lipid by SMCs Induces a Membrane-Based Signaling Defect
Fibrillar adhesion complexes form under the control of a signaling cascade that follows engagement of α5β1 integrin by fibronectin. To determine whether loading SMCs with lipid perturbs this cascade, we evaluated the steady-state activation status of key signaling components, 24 hours after addition of lipoproteins to the media. Phosphorylation of tensin is a distal activation step that is intimately linked to the egress of tensin out of focal adhesion complexes to form fibrillar adhesion complexes.23 As shown in Figure 6, the proportion of tensin that was phosphorylated was significantly lower in LDL-incubated SMCs than in unloaded SMCs (43.9±11.1% reduction, P<0.05). Similarly, in VLDL-incubated SMCs the fraction of phosphorylated tensin was significantly reduced (56.6±14.5% reduction, P<0.05).
We next evaluated the phosphorylation FAK, a key kinase upstream of tensin phosphorylation involved in remodeling of adhesion complexes.24,25 As illustrated (Figure 6), steady-state levels of FAK in SMCs were unaltered by lipid loading. However, there was a striking decrease in the abundance of phosphorylated FAK (tyrosine 397) in both LDL- and VLDL-exposed SMCs (P<0.05). We also evaluated the activation of Src, an activator (and reciprocal target) of FAK that is necessary for tensin phosphorylation and fibronectin assembly.26 Compared to unloaded SMCs, the proportion of Src that was phosphorylated on tyrosine 418 in LDL- or VLDL-incubated SMCs was significantly reduced (P<0.05). Suppression of FAK and Src activation was also observed in lipid-engorged primary SMCs (supplemental Figure IV).
We also examined the activation of ERK1/2 in lipid-loaded SMCs. ERK1/2 can be activated by the ECM but, unlike Src and FAK, they do not have direct targets in adhesion complexes at the plasma membrane. Interestingly, there was no change in activation of ERK1/2 on conversion of SMCs to foam cells, based on the proportion of pERK.
Finally, we evaluated the in situ activation of Src in lipid-loaded SMCs by immunostaining for phosphorylated Src. Under control conditions, active Src localized in linear aggregates at the cell periphery and the apical cell plasma membrane. In contrast, in lipid-loaded SMCs, phosphorylated Src localized almost exclusively at the cell edges, where discrete accumulations decorated the cell boundary. These accumulations generally stained more intensely than those of control SMCs with a globular morphology (Figure 6B), indicating abnormally localized Src activity in addition to a net decrease in overall activity.
Defective ECM Assembly in Foam Cells Is Mediated by Src Inactivation
To ascertain whether this lipid-induced disruption of Src-based signaling caused the impaired fibrillar adhesion formation, we infected SMCs with retrovirus containing cDNA encoding constitutively active Src mutant (Y527F).27 Human SMCs expressing mutant Src displayed Src activation, as evidenced by phosphorylation of Tyr418,28 and also FAK activation, indicated by phosphorylation of Tyr397 (Figure 7A). Furthermore, these residues remained activated in SMCs following incorporation of lipids from LDL or VLDL, confirming the forced activation of Src-FAK signaling in these cells by Src Y527F. Interestingly, well-formed tensin-containing fibrillar adhesion complexes were present in lipid-engorged SrcY527F-SMCs, even in regions overlying abundant cytoplasmic lipid droplets (Figure 7B, arrows). Furthermore, the impaired assembly of fibronectin fibrils in lipid-loaded SMCs was rescued by the forced activation of FAK-Src signaling (Figure 7C). Thus, lipid accumulation disrupts the machinery for forming/maintaining fibrillar adhesion complexes and ECM fibrils by deactivating a Src-FAK-tensin axis.
We have established that accumulation of lipids by vascular SMCs disrupts the ability of SMCs to assemble a matrix of fibronectin and type I collagen fibrils. To our knowledge, this represents the first identification of an acquired defect in the assembly of fibronectin or collagen. It is further noteworthy that this defect in fibril assembly arises from a widespread cardiovascular risk factor. We have also elucidated the mechanistic underpinnings of this defect. In particular, on lipid accumulation, α5β1 integrin fails to translocate within the plasma membrane as required to polymerize fibronectin and, secondarily, type I collagen. This abnormality in α5β1 integrin clustering was not part of a generalized defect in cytoskeletal architecture but, rather, a feature of the inability of the cell to assemble fibrillar adhesion complexes. The inability to form fibrillar adhesion complexes, in turn, was linked to the deactivation and regionally abnormal activation of Src and to disruption of the Src-FAK-tensin signaling axis required for remodeling fibrillar adhesion complexes. These findings thus uncover a novel, pathological relationship between intracellular lipids and the machinery for building a fibrillar ECM.
Despite longstanding recognition that smooth muscle-derived foam cells are present in human atherosclerotic lesions,3,4 there has been little data to ascertain whether these cells have functional consequences for atherosclerosis. This may be partly attributable to a paucity of models whereby SMCs become engorged with lipid vesicles, as found in atherosclerosis. The lipid-loaded SMCs generated in this study offer several advantages in this regard. First, the SMCs used are human cells that closely mimic the behavior of vascular SMCs in vivo.14,15 Second, the cells were loaded with lipid by exposure to human lipoproteins, a process that was nontoxic and reversible. Third, the cells displayed the classic morphology of foam cells, with copious cytoplasmic droplets of neutral lipids. Interestingly, droplet size varied between LDL-and VLDL-exposed SMCs, with larger droplets in VLDL-incubated SMCs. This difference is consistent with the increased lipid burden from the combined load of cholesterol esters and triglyceride within VLDL.6 The result was a more striking defect in fibril assembly but qualitatively the effects were similar to those of LDL-incubated SMCs.
Recently, mouse SMC-derived foam cells were found to shift their phenotype to a macrophage-like state.20 This differs from the foam cells generated in the present study that retained a SMC phenotype, indicated by persistently abundant smooth muscle α-actin, calponin h1, and h-caldesmon. Direct comparison between the two models of SMC-derived foam cells cannot be made because of differences in SMC species and the nature of lipid delivery. Rong et al20 used a cyclodextrin-based approach to deliver cholesterol whereas we used lipoproteins, so that lipids would enter the cell in a receptor-mediated pathway akin to that which occurs during foam cell formation in vivo. Regardless, the phenotypic stability of the lipid-engorged SMCs generated in the present study provided a valuable opportunity to determine whether SMC functions were altered by lipids, unconfounded by global shifts in cell identity.
Despite the importance of collagen polymerization, it has received little attention as a potentially modifiable process or a point of potential dysfunction in disease. This may be because type I collagen has the capacity to assemble into fibrils spontaneously, an entropy-driven reaction that has been extensively studied in cell-free systems.10 However, the extent to which collagen self-assembly accounts for fibril formation in tissues is uncertain and evidence has emerged that cells play a major role in orchestrating collagen fibril assembly.11–13 Cell-based control of fibrillogenesis likely underlies the diverse collagen architectures in different tissues and is probably essential for rapid deposition of appropriately aligned fibrils,29 including during acute tissue repair. Our finding that collagen polymerization was a vulnerable process that can be destabilized by intracellular lipids thus suggests a novel paradigm for abnormal tissue repair under conditions of lipid excess.
It has recently been identified that the assembly of type I collagen depends on the assembly of fibronectin.12,13 Our finding that defective collagen assembly by smooth muscle foam cells is linked to impaired fibronectin assembly is consistent with this relationship. Fibronectin assembly requires actin-dependent movement of fibronectin-bound α5β1 integrin and tensin out of focal adhesions and along the cell surface to form fibrillar adhesions.22 The stretching force imposed on soluble fibronectin unfolds the globular protein to expose cryptic fibronectin-binding sites, enabling the polymerization process.30 We found that expression of α5β1 integrin and tensin was normal in SMCs loaded with lipid; however, neither protein appropriately assembled into fibrillar adhesion complexes. This was a selective defect in protein assembly as focal adhesion complexes were unaffected. This selectivity is consistent with studies wherein a dominant-interfering tensin protein impaired fibronectin assembly but did not disrupt focal adhesions21 and may reflect the particularly dynamic nature of fibrillar adhesion complexes.
The movement of tensin and α5β1 integrin out of focal adhesion complexes into fibrillar adhesion complexes requires the phosphorylation of tensin.23 This process is linked to Src signaling, a necessity indicated by the phenotype of Src-deficient mouse embryonic fibroblasts which contain hypophosphorylated tensin and fail to form fibrillar adhesions.23 Embryonic fibroblasts from FAK-null mice also fail to form tensin-containing fibrillar adhesion complexes25 and FAK activation is tightly and reciprocally linked to Src activity. Thus, it is noteworthy that the abundance of phosphorylated tensin was reduced in smooth muscle foam cells and that both Src and FAK were less active. Furthermore, the impaired assembly of fibrillar adhesions and fibronectin in lipid-loaded SMCs was rescued by the forced activation of FAK-Src signaling. Thus, rather than directly affecting the integrity of cytoskeletal structures, accumulated lipids within SMCs adversely impact the signaling cascade necessary for assembling proteins into fibrillar adhesions.
Furthermore, our finding of the restriction of activated Src to the cell edges implicates the induction of regionally compromised signaling on lipid loading. The dichotomous relationship between the location of lipid vesicles and sites of α5β1 integrin and tensin assembly further highlights a spatially heterogeneous defect. Recent studies have established that intracellular lipid droplets are not inert accumulations but dynamic structures that recruit a variety of proteins to their coat.31,32 This recruitment can entail trafficking of signaling and scaffolding proteins out of the plasma membrane.32 Altered regional fluidity of the plasma membrane of lipid-loaded SMCs might also underlie the disturbed Src-FAK-tensin axis. Thus, cellular processes that depend on lateral protein mobility and interactions in the plasma membrane may be especially sensitive to high lipid environments.
A relative lack of collagen fibrils in atherosclerotic lesions is a major reason for plaque rupture and myocardial infarction.9 To date, this paucity of collagen fibrils has been linked primarily to the elaboration of collagen-degrading enzymes.33,34 The present study raises the possibility that an additional basis for vulnerable atherosclerotic plaque may be aberrant polymerization of collagen by lipid-loaded SMCs. The potential for SMCs to have their repair capacity disabled by lipids should be recognized when considering strategies for stabilizing diseased arteries.
We thank Dr Robert A. Hegele (Robarts Research Institute) for providing the patient blood samples.
Sources of Funding
This work was supported by grants from the Canadian Institutes of Health Research (FRN-11715) and Heart and Stroke Foundation of Canada (T5675, PRG4854). J.G.P. holds a Career Investigator Award from the Heart and Stroke Foundation of Ontario.
Original received September 12, 2008; revision received February 4, 2009; accepted February 10, 2009.
Faggiotto A, Ross R, Harker L. Studies of hypercholesterolemia in the nonhuman primate. I. Changes that lead to fatty streak formation. Arteriosclerosis. 1984; 4: 323–340.
Argmann CA, Sawyez CG, Li S, Nong Z, Hegele RA, Pickering JG, Huff MW. Human smooth muscle cell subpopulations differentially accumulate cholesteryl ester when exposed to native and oxidized lipoproteins. Arterioscler Thromb Vasc Biol. 2004; 24: 1290–1296.
Rong JX, Kusunoki J, Oelkers P, Sturley SL, Fisher EA. Acyl-coenzymeA (CoA): cholesterol acyltransferase inhibition in rat and human aortic smooth muscle cells is nontoxic and retards foam cell formation. Arterioscler Thromb Vasc Biol. 2005; 25: 122–127.
Lee RT, Libby P. The unstable atheroma. Arterioscler Thromb Vasc Biol. 1997; 17: 1859–1867.
Williams BR, Gelman RA, Poppke DC, Piez KA. Collagen fibril formation. Optimal in vitro conditions and preliminary kinetic results. J Biol Chem. 1978; 253: 6578–6585.
Velling T, Risteli J, Wennerberg K, Mosher DF, Johansson S. Polymerization of type I and III collagens is dependent on fibronectin and enhanced by integrins alpha 11beta 1 and alpha 2beta 1. J Biol Chem. 2002; 277: 37377–37381.
Sottile J, Hocking DC. Fibronectin polymerization regulates the composition and stability of extracellular matrix fibrils and cell-matrix adhesions. Mol Biol Cell. 2002; 13: 3546–3559.
Li S, Fan YS, Chow LH, Van Den Diepstraten C, van Der Veer E, Sims SM, Pickering JG. Innate diversity of adult human arterial smooth muscle cells: cloning of distinct subtypes from the internal thoracic artery. Circ Res. 2001; 89: 517–525.
Li S, Sims S, Jiao Y, Chow LH, Pickering JG. Evidence from a novel human cell clone that adult vascular smooth muscle cells can convert reversibly between noncontractile and contractile phenotypes. Circ Res. 1999; 85: 338–348.
Rocnik EF, Van Der Veer E, Cao H, Hegele RA, Pickering JG. Functional linkage between the endoplasmic reticulum protein Hsp47 and procollagen expression in human vascular smooth muscle cells. J Biol Chem. 2002; 277: 38571–28578.
Small TW, Bolender Z, Bueno C, O'Neil C, Nong Z, Rushlow W, Rajakumar N, Kandel C, Strong J, Madrenas J, Pickering JG. Wilms’ tumor 1-associating protein regulates the proliferation of vascular smooth muscle cells. Circ Res. 2006; 99: 1338–1346.
Pickering JG, Uniyal S, Ford CM, Chau T, Laurin MA, Chow LH, Ellis CG, Fish J, Chan BM. Fibroblast growth factor-2 potentiates vascular smooth muscle cell migration to platelet-derived growth factor: upregulation of alpha2beta1 integrin and disassembly of actin filaments. Circ Res. 1997; 80: 627–637.
Rong JX, Shapiro M, Trogan E, Fisher EA. Transdifferentiation of mouse aortic smooth muscle cells to a macrophage-like state after cholesterol loading. Proc Natl Acad Sci USA. 2003; 100: 13531–13536.
Pankov R, Cukierman E, Katz BZ, Matsumoto K, Lin DC, Lin S, Hahn C, Yamada KM. Integrin dynamics and matrix assembly: tensin-dependent translocation of alpha(5)beta(1) integrins promotes early fibronectin fibrillogenesis. J Cell Biol. 2000; 148: 1075–1090.
Ren XD, Kiosses WB, Sieg DJ, Otey CA, Schlaepfer DD, Schwartz MA. Focal adhesion kinase suppresses Rho activity to promote focal adhesion turnover. J Cell Sci. 2000; 113: 3673–3678.
Ilic D, Kovacic B, Johkura K, Schlaepfer DD, Tomasevic N, Han Q, Kim JB, Howerton K, Baumbusch C, Ogiwara N, Streblow DN, Nelson JA, Dazin P, Shino Y, Sasaki K, Damsky CH. FAK promotes organization of fibronectin matrix and fibrillar adhesions. J Cell Sci. 2004; 117: 177–187.
Wierzbicka-Patynowski I, Schwarzbauer JE. Regulatory role for Src and phosphatidylinositol 3-kinase in initiation of fibronectin matrix assembly. J Biol Chem. 2002; 277: 19703–19708.
Cooper JA, Gould KL, Cartwright CA, Hunter T. Tyr527 is phosphorylated in pp60c-src: implications for regulation. Science. 1986; 231: 1431–1434.
Canty EG, Starborg T, Lu Y, Humphries SM, Holmes DF, Meadows RS, Huffman A, O'Toole ET, Kadler KE. Actin filaments are required for fibripositor-mediated collagen fibril alignment in tendon. J Biol Chem. 2006; 281: 38592–38598.
Baneyx G, Baugh L, Vogel V. Fibronectin extension and unfolding within cell matrix fibrils controlled by cytoskeletal tension. Proc Natl Acad Sci USA. 2002; 99: 5139–5143.
Brasaemle DL, Dolios G, Shapiro L, Wang R. Proteomic analysis of proteins associated with lipid droplets of basal and lipolytically stimulated 3T3-L1 adipocytes. J Biol Chem. 2004; 279: 46835–46842.
Pol A, Martin S, Fernandez MA, Ingelmo-Torres M, Ferguson C, Enrich C, Parton RG. Cholesterol and fatty acids regulate dynamic caveolin trafficking through the Golgi complex and between the cell surface and lipid bodies. Mol Biol Cell. 2005; 16: 2091–2105.
Deguchi JO, Aikawa E, Libby P, Vachon JR, Inada M, Krane SM, Whittaker P, Aikawa M. Matrix metalloproteinase-13/collagenase-3 deletion promotes collagen accumulation and organization in mouse atherosclerotic plaques. Circulation. 2005; 112: 2708–2715.
Dollery CM, Libby P. Atherosclerosis and proteinase activation. Cardiovasc Res. 2006; 69: 625–635.