Strain-Dependent Vascular Remodeling
The “Glagov Phenomenon” Is Genetically Determined
Background— Atherosclerosis of the carotid artery, called intima-media thickening (IMT), is a form of vascular remodeling that is an important predictor for cardiovascular events and has a strong genetic component.
Methods and Results— Recently, we established a mouse model of vascular remodeling based on partial ligation of the carotid, which is relevant to the “Glagov phenomenon.” We hypothesized that there would be genetically determined differences in outward remodeling and IMT induced by carotid flow alterations. We compared vascular remodeling among 5 inbred strains of mice. Despite similar changes in flow among the strains in the left carotid artery (LCA), we observed dramatic differences in remodeling of the partially ligated LCA relative to control. The smallest IMT volume (26±3 μm3) was found in C3H/HeJ mice, and the largest were in SJL/J (59±10 μm3) and FVB/NJ (81±6 μm3). Shear stress did not differ after ligation among strains. Lumen area decreased only when stenosis was ≥55%. IMT correlated significantly with outward remodeling among inbred strains (except C3H). There were significant strain-dependent differences in remodeling index (measured as vessel area/IMT), which suggest fundamental alterations in sensing or transducing hemodynamic signals among strains. Among hemodynamic factors, low shear stress and high heart rate were predictive for IMT. Specifically, heart rate (bpm: C3H, 592±6; SJL, 649±6; FVB, 683±7) but not systolic blood pressure (mm Hg: C3H, 116±2; SJL, 119±1; FVB, 136±1) was predictive.
Conclusions— The present study indicates that performing a genetic cross of these strains and total genome scan should identify genes that mediate vascular remodeling.
Received January 4, 2004; revision received March 9, 2004; accepted March 22, 2004.
The “Glagov phenomenon” refers to the observation that in the early stages of atherosclerosis, coronary arteries enlarge in relation to plaque area to preserve lumen until plaque area occupies ≥40% of vessel area.1 Recent clinical data demonstrate that regions of low shear stress develop progressive atherosclerosis as measured by increased intima-media thickening (IMT) and outward remodeling of coronary arteries.2,3 More importantly, IMT is associated with an increased risk of heart disease and stroke4 and has a strong genetic correlation.5
In a complete ligation and flow cessation mouse model,6 the roles of 15 genes in vascular remodeling have been studied in genetically altered mice (endothelial, inducible, and neuronal nitric oxide synthase; vimentin; P-selectin; osteopontin; dystrophin; steroid receptor coactivator-3; endothelin-B; p130; plasminogen; tissue and urokinase plasminogen activator; plasminogen activator inhibitor type 1; and matrix metalloproteinase 9 [see Data Supplement for references]). Lumen narrowing in this model of vascular remodeling is genetically controlled, as measured in inbred mouse strains.7 Unfortunately, the physiological relevance of this model to human cardiovascular disease is uncertain. We recently developed a partial ligation model (with maintenance of blood flow [BF]) of vascular remodeling with IMT.8 The purpose of this study was to provide the basis for a genetic analysis of vascular remodeling that has relevance to human diseases, such as carotid IMT and coronary atherosclerosis. We used 5 inbred mouse strains to evaluate mechanisms of outward remodeling and IMT.
Male and female C3H/HeJ (C3H), DBA/2J (DBA), C57Bl/6J (C57), FVB/NJ (FVB), and SJL/J (SJL) mice (8 weeks old, Jackson Laboratories, Bar Harbor, Me) were used in accordance with the guidelines of the National Institutes of Health and American Heart Association for the care and use of laboratory animals. All procedures were approved by the University of Rochester Animal Care Committee.
Mice were anesthetized with a mixture of ketamine and xylazine and maintained at 37°C as described previously.8 BF in the left common carotid artery (LCA) was reduced by partial ligation of the left external and internal carotid arterial branches. The branches of the left carotid artery were exposed but not ligated (sham operation). The animals were allowed to recover and housed individually under specific pathogen-free conditions with a 12/12-hour light/dark cycle.
Two groups of ligated or sham-operated animals of each strain were processed for morphological studies at 14 days after the surgery. Systolic blood pressure (SBP) and heart rate (HR) were measured by tail-cuff plethysmography (Visitech System). In separate experiments, the BF was measured on both carotids by use of an ultrasonic transit-time volume flowmeter (Transonic Systems, Inc) before and at termination.8
Animals were perfusion-fixed with 10% paraformaldehyde as described previously.9 The carotids were harvested and embedded in paraffin, and cross sections (4 μm) were made as described previously.8 Sections were stained with hematoxylin and eosin and analyzed by use of MCID image software (Imaging Research Inc). The lumen circumference was used to calculate lumen area. Vessel compartments volumes were calculated as described previously.8 Both sexes were used, because no sex differences were noted for any strain (Data Supplement Table 1s). Shear stress was calculated as described elsewhere.10
All results are reported as mean±SEM. Statistical tests were performed with StatView for MacIntosh, version 5.0.1, or MS Excel (regression statistics). Differences between groups were analyzed by means of a repeated-measures 1-way ANOVA followed by a Fisher’s post hoc test. A level of P<0.05 was regarded as significant.
Physiological and Hemodynamic Characteristics
Male mice were significantly heavier (≈3 g) than females among inbred strains. FVB mice were slightly heavier than other mouse strains at the age of 9 weeks (Data Supplement Table 2s). Ligated animals gained weight similar to their sham controls 2 weeks after ligation, except for C3H mice (≈5% difference, Data Supplement Table 2s). Initial levels of SBP differed among mouse strains (Figure 1A). C3H and SJL mice had relatively low SBP (≈120 mm Hg), whereas C57, DBA, and FVB had significantly higher SBP (≈130 mm Hg, Figure 1A). There were no significant changes in SBP among inbred mice after ligation compared with shams (data not shown). There were significant differences (≈100 bpm) in HR across inbred strains (Figure 1B), in the following order: C3H≈C57<DBA<SJL<FVB. The changes in HR among inbred mice after ligation did not differ compared with shams (data not shown).
The initial level of BF (0.5 mL/min) did not differ among inbred strains or between LCA and right carotid artery (RCA) (data not shown). Significant changes were observed in all strains 2 weeks after ligation compared with shams (Table). In response to partial ligation, BF decreased similarly for all strains in the LCA (≈80% decrease, Table). The level of BF decrease was similar to that reported by our laboratory in mice (baseline, 0.5 mL/min)8 and rats (baseline, 2.5 mL/min)11,12 in the same model. After ligation, BF increased significantly in the RCA (≈60% increase, Table). The BF changes were not statistically significant among inbred strains, although SJL trended to be less (P=0.07) compared with C57 in the RCA (Table). In the RCA, the increase in BF after ligation was ≈15% bigger than in rats.12
Strain-Dependent Flow-Stimulated Vascular Remodeling
Altered BF in the LCA and RCA resulted in dramatic vascular remodeling, with maximal responses at 2 weeks after ligation (Korshunov and Berk, unpublished data; and Reference 8). To determine genetic effects on vascular remodeling, we examined the responses to partial ligation in 5 inbred strains of mice.
There were small variations in lumen areas along carotid length because of the carotid dividing into branches (Data Supplement Figure 1s; compare areas at 200 versus 400 μm from bifurcation). Areas were used for calculation of the lumen volumes (Figure 2, A and B), as described elsewhere.8 In contrast to partial ligation in rats12 or complete ligation in mice,7 the 80% decrease in BF in the LCA of the inbred mice resulted in a variety of responses: lumen volume decreased significantly in C3H, had a trend to be smaller (P=0.08) in SJL, did not change in C57, and increased significantly in DBA and FVB mice compared with their shams (Figure 2A). There was a significant increase in the RCA lumen volume after ligation in C57 and FVB mice (Figure 2B). Other strains (C3H, DBA, and SJL) exhibited a tendency to increase RCA lumen compared with shams (Figure 2B). Because C3H shams had significantly bigger lumens compared with other strains, we calculated the ratio of RCA lumen volume to LCA lumen volume (Data Supplement Figure 2s). Among shams, there were no significant differences; the ratio was ≈1 (RCA=LCA=1.0, Data Supplement Figure 2s). After ligation, the RCA/LCA lumen ratios changed in the following order: FVB≈DBA(0.9)<C3H(1.3)<C57(1.5)<SJL(1.8). These data suggest that partial ligation caused the predicted lumen responses (an increase in lumen in high-flow RCA and decrease in low-flow LCA) only in C3H, C57, and SJL mice.7,12
Previous data from our laboratory in rats12 suggest that genetic factors significantly regulate shear stress and vascular remodeling, so we calculated changes in shear stress among inbred strains of mice (Table). There were no statistically significant differences in shear-stress reduction in the LCA among strains (≈80%). In the RCA, shear stress was not normalized in C3H compared with the other strains (Table).
A characteristic feature of this model is an increase in LCA adventitia that was observed for all strains except C3H (Figure 3A and Data Supplement Figure 3s). There was no increase in the RCA adventitia volume among inbred strains compared with shams (Figure 3B). However, adventitia volume in C3H shams was increased compared with the shams of the other strains (Figure 3, A and B).
Media areas (Data Supplement Figure 4s) were evaluated for the carotids and used to calculate volumes (Figure 3, C and D). The largest media response to ligation was in the LCA and was highly regulated by genetic factors (Figure 3C). The biggest response occurred in FVB (≈2.6-fold increase), with increases of 2-fold in C57, DBA, and SJL and no change in C3H compared with shams (Figure 3C). There was no increase in the RCA media volume among inbred strains compared with shams (Figure 3D). However, RCA media volume in DBA and SJL shams was smaller than in FVB (Figure 3D).
Robust intima changes occurred in ligated LCAs (Figure 3, E and F and Data Supplement Figure 5s). Of note, there was no detectable intima formation in the sham carotids or the RCA of ligated mice (data not shown). There were significant genetic determinants of intima formation: SJL and FVB developed much greater intima than other strains (Figures 3E and Data Supplement Figure 5s). In contrast to the complete ligation model in mice,6 intima formed without a longitudinal gradient relative to the ligation site (Data Supplement Figure 5s). Intima volume among inbred mouse strains differed by ≈30-fold: C3H≈DBA≈C57<FVB<SJL (Figure 3E). Determination of intima formation and the ratio of intima to media volume (Figure 3F) demonstrated that SJL remodeled largely by intima formation. FVB formed an intima to a smaller extent than SJL but significantly greater than other strains (Figure 3, E and F).
External Elastic Lamina
To evaluate outward versus inward vascular remodeling, we analyzed the carotid external elastic lamina (EEL) volumes (Figure 3, G and H). The RCA underwent outward remodeling in C57 and FVB, whereas the other strains (C3H, DBA, SJL) exhibited a trend to increase EEL (Figure 3H). The outward remodeling in the RCA correlated with lumen volume increase after ligation (Figure 2B). Partial ligation did not affect EEL volume in the LCA of C3H and C57 and resulted in significant positive (outward) remodeling in FVB, SJL, and DBA (Figure 3G). The maximal response was observed in FVB mice (180×10−6 μm3) versus their shams (85×10−6 μm3). Representative cross sections of the LCA (≈1000 μm from bifurcation) of strains with the biggest differences in intima and EEL volume are shown in Figure 4: C3H mice had very little intima formation and tended to inward remodeling (compare Figure 4A versus 4B). In contrast, FVB mice showed maximal outward remodeling with relatively large intima (Figure 4C versus 4D), and SJL had the biggest intima formation with outward remodeling (Figure 4E versus 4F).
Analysis of Vessel Remodeling Mechanisms
To evaluate possible mechanisms that regulate vessel remodeling, we plotted EEL area against intima+media area (Figure 5). Among multiple comparisons, significant correlations were found between outward remodeling (increased EEL area) and IMT (increased intima+media area) after ligation (Figure 5, B–E). In C3H mice, no correlation was observed (Figure 5A). Despite a significant correlation of EEL with IMT, the slopes differed significantly among C57, DBA, and FVB (these strains were “big responders,” with slopes >1.5, Figure 5, B–D) compared with SJL and C3H (strains that were “small responders,” with slopes <1.0, Figure 5, A and E). We summarized these data as a remodeling index (Figure 5F), which represents the slope of EEL versus IMT. To show that this was not a “threshold” effect, we also determined the maximal EEL areas (see legend of Figure 5). There was no correlation between the maximal EEL area and the remodeling index. For example, DBA mice had the greatest remodeling index (2.21) but one of the smallest maximal EEL areas (100×10−3 μm2) compared with C57 and FVB mice.
To evaluate whether there was a threshold for lumen remodeling (as observed by Glagov et al1), we plotted lumen area against stenosis as %stenosis=(intima+media area)/(EEL area×100) from all strains in the ligated LCAs (Figure 6). Correlation analysis for all data did not show a relationship between lumen area and %stenosis (not shown; R2=0.14, P>0.05). However, at %stenosis values <55%, lumen area remained constant, whereas at %stenosis >55%, there was a significant linear decline in lumen area (Figure 6).
The major findings of this study are that flow-induced vascular remodeling of the mouse carotid is highly related to the Glagov phenomenon observed in humans and has significant genetic determinants. Specifically, we observed that partial ligation of the LCA to establish a large flow reduction (−80%) stimulated growth of the intima, media, and adventitia with outward vascular remodeling. Most relevant to the Glagov phenomenon, we observed that IMT correlated significantly with outward remodeling (EEL area) among inbred strains (except C3H mice), similar to data we reported previously.8 A novel finding was the strong genetic influence on the Glagov phenomenon. Specifically, there were large variations in vascular remodeling in the ligated LCAs among strains: the smallest response in vessel wall compartment volumes was found in C3H mice; moderate changes were noted for C57 and DBA mice, whereas the greatest responses occurred in SJL and FVB mice (Figure 3). Interestingly, despite a significant correlation of EEL with IMT, there were also significant strain-dependent differences in remodeling index (measured as a correlation slope). The differences in remodeling index among these 5 mouse strains prove that vascular remodeling has strong genetic influences and suggest that fundamental alterations in sensing or transducing hemodynamic signals contribute to these genetic differences. Because shear stress changes did not differ among strains, a primary role for cells in the vessel wall (as opposed to endothelial cells) is suggested for the early stages of vessel remodeling. Among other hemodynamic variables that we measured, HR but not SBP influenced IMT.
The results reported here are remarkably similar to recent studies in human coronaries. Specifically, remodeling in native and stented human arteries, measured by intracoronary ultrasound, biplane coronary angiography, and coronary BF, demonstrate that regions of low flow and low shear stress exhibit increased IMT and outward remodeling.2,3 To evaluate the magnitude of lumen preservation in our mouse model, an important clinical measure in human atherosclerosis, we plotted lumen area against stenosis. As shown in Figure 6, we found that the vessel compensated for IMT growth until the %stenosis was >55%. This transition point at 55% is similar to Glagov’s observation, although the transition point for lumen decrease in mice was higher than in humans (≈40%).1 The higher value in mice may be because of differences in species, artery types (carotid versus coronary), and other factors.
It is important to note that the present model also results in remodeling of the RCA and therefore provides insight into remodeling associated with high flow and high (initial) shear stress. As expected, the RCA underwent outward remodeling in response to increased shear stress (Table) with increased lumen volume (Figure 2).13 There was adequate RCA outward remodeling to normalize shear stress in all strains except C3H. Of clinical relevance, high shear stress was not associated with IMT, consistent with the concept that high shear stress is atheroprotective, as shown for human coronary arteries.2 Genetic factors appeared to play a small role in the response to high shear stress, because there were no significant differences in the RCA among inbred mouse strains.
We believe that our model represents a significant advance over previous animal models used to study vascular remodeling. First, significant IMT occurred in 4 of 5 mouse strains in response to low flow, whereas there was no IMT formation in partially ligated low-flow carotids of rats or rabbits.7,12 Second, the present model has important advantages compared with the complete carotid ligation model in mice.7 The absence of thrombosis and maintenance of low flow are the major differences of the present model compared with the complete ligation model.6 Thrombus formation in the complete ligation model is a much more complex process, with many factors (thrombin, tissue factor, platelets, and white blood cells) present that may affect neointima formation. In addition, the presence of low flow in our model is more representative of the hemodynamic environment in human coronaries associated with IMT and remodeling. The partial ligation model also mimics the Glagov phenomenon much better, with both IMT and outward remodeling. Third, in the present model, there is also vascular remodeling in the high-flow RCA. We believe that the genes and cellular processes responsible for remodeling in response to high flow differ from those of low flow,12 making this model particularly versatile.
We identified 3 factors that influence vascular remodeling in the present study: genes (strain), shear stress, and HR. It has been shown previously in rats12 and mice7 that the genetic background is a critical determinant of remodeling, with unique responses among inbred strains. In the present study, there was a strong correlation between outward remodeling and IMT formation in the carotid: strains with very large outward remodeling (SJL and FVB) had the largest IMT. A similar correlation was reported previously in the complete carotid ligation model in hypercholesterolemic apolipoprotein E-knockout mice.14 However, even among the large remodelers, SJL mice remodeled primarily by increased intima formation, whereas FVB mice remodeled primarily by increased media. We also observed large differences between these 2 strains in remodeling index, consistent with a limitation of outward remodeling in SJL mice compared with FVB. Shear stress was an important determinant of remodeling in the RCA, although there was no genetic component, on the basis of similar responses among the 5 strains. HR but not SBP correlated significantly with remodeling. HR was shown previously to correlate with IMT in experimental15,16 and clinical17 studies. A strong relationship between HR-corrected QT interval and IMT was found in humans with early atherosclerotic disease.17 Of interest, reducing HR (by sinoatrial node ablation) in cynomolgus monkeys decreased plaque burden in coronary15 and carotid16 arteries. BP, especially pulse pressure, is also known to affect carotid IMT.18 The basal level of SBP was not associated with the magnitude of IMT in inbred mice in this study. Because BP is more difficult to measure than HR, the lack of association with BP could be because of technical problems. However, we believe that our BP measurements (tail-cuff) are reliable, on the basis of several reports that used direct hemodynamic measurements in C3H,19 C57, and FVB,20 in which HR and BP values were similar to those reported here.
The changes in the low-flow LCA resemble human carotid artery IMT, which is associated with an increased risk of heart disease and stroke.4 It is well known that genetic factors are important for cardiovascular diseases.21 Approximately 40% of the variability in the carotid IMT was shown recently to be dependent on family history.5 To date, association studies of polymorphisms and IMT have not yielded strong candidate genes.21,22 Therefore, we propose that future experiments using QTL analysis of a cross involving strains in the present study (ligated C3H×SJL or C3H×FVB) will be useful to elucidate genetic mechanisms of outward remodeling and IMT in response to flow reduction. Because of the strong association of carotid IMT with cardiovascular events, these genetic mechanisms may provide important insights into atherogenesis and the pathogenesis of coronary artery disease.
This study was supported by National Institutes of Health grant HL-62826 to Dr Berk. We thank Mary Georger and David Nagel for performing histological measurements.
Supplementary materials are available in the online-only Data Supplement at http://www.circulationaha.org.
Stone PH, Coskun AU, Kinlay S, et al. Effect of endothelial shear stress on the progression of coronary artery disease, vascular remodeling, and in-stent restenosis in humans: in vivo 6-month follow-up study. Circulation. 2003 108: 438–444.
Wentzel JJ, Janssen E, Vos J, et al. Extension of increased atherosclerotic wall thickness into high shear stress regions is associated with loss of compensatory remodeling. Circulation. 2003; 108: 17–23.
Fox CS, Polak JF, Chazaro I, et al. Genetic and environmental contributions to atherosclerosis phenotypes in men and women: heritability of carotid intima-media thickness in the Framingham heart study. Stroke. 2003; 34: 397–401.
Kumar A, Lindner V. Remodeling with neointima formation in the mouse carotid artery after cessation of blood flow. Arterioscler Thromb Vasc Biol. 1997; 17: 2238–2244.
Korshunov VA, Berk BC. Flow-induced vascular remodeling in the mouse: a model for carotid intima-media thickening. Arterioscler Thromb Vasc Biol. 2003; 23: 2185–2191.
Geary RL, Kohler TR, Vergel S, et al. Time course of flow-induced smooth muscle cell proliferation and intimal thickening in endothelialized baboon vascular grafts. Circ Res. 1994; 74: 14–23.
Miyashiro JK, Poppa V, Berk BC. Flow-induced vascular remodeling in the rat carotid diminishes with age. Circ Res. 1997; 81: 311–319.
Ibrahim J, Miyashiro JK, Berk BC. Shear stress is differentially regulated among inbred rat strains. Circ Res. 2003; 92: 1001–1009.
Langille BL, Reidy MA, Kline RL. Injury and repair of endothelium at sites of flow disturbances near abdominal aortic coarctations in rabbits. Arteriosclerosis. 1986; 6: 146–154.
Ivan E, Khatri JJ, Johnson C, et al. Expansive arterial remodeling is associated with increased neointimal macrophage foam cell content: the murine model of macrophage-rich carotid artery lesions. Circulation. 2002; 105: 2686–2691.
Beere PA, Glagov S, Zarins CK. Retarding effect of lowered heart rate on coronary atherosclerosis. Science. 1984; 226: 180–182.
Beere PA, Glagov S, Zarins CK. Experimental atherosclerosis at the carotid bifurcation of the cynomolgus monkey: localization, compensatory enlargement, and the sparing effect of lowered heart rate. Arterioscler Thromb. 1992; 12: 1245–1253.
Festa A, D’Agostino R Jr, Rautaharju P, et al. Is QT interval a marker of subclinical atherosclerosis in nondiabetic subjects? The Insulin Resistance Atherosclerosis Study (IRAS). Stroke. 1999; 30: 1566–1571.
Liang YL, Shiel LM, Teede H, et al. Effects of blood pressure, smoking, and their interaction on carotid artery structure and function. Hypertension. 2001; 37: 6–11.
Shusterman V, Usiene I, Harrigal C, et al. Strain-specific patterns of autonomic nervous system activity and heart failure susceptibility in mice. Am J Physiol. 2002; 282: H2076–H2083.