During the last three decades, exercise has been increasingly used as a therapeutic intervention for patients with CAD (40), heart failure (3,29), and Type II diabetes (24), prescribed alone or in combination with medical therapy. Exercise is thought to be beneficial, in part, through direct effects on the vasculature (11), resulting in improvement of endothelial function in patients with risk factors for atherosclerosis or diagnosed CAD (12,13,45). Moreover, exercise has been shown to retard coronary plaque progression assessed by angiography (28). Improved endothelial function is the prevailing initiating mechanism invoked to explain the reduction in atherosclerosis with increased physical activity. Improved endothelial formation and release of nitric oxide (NO), endothelial-dependent hyperpolarizing factor, and reductions in vasoconstrictor prostanoids have been proposed to lead to reductions in adhesion molecule expression, e.g., vascular cell adhesion molecule (VCAM-1), and subsequent lesion formation (25,48). Indeed, inflammation not only is recognized as the major initiating factor of atherogenesis but is also associated with plaques in all stages of disease and is proposed to play a key role in acute coronary events (23,31). Increased physical activity is associated with reductions in cerebrovascular events, i.e., stroke, as well as coronary and peripheral artery disease. However, although improved endothelial-dependent dilation (EDD) with exercise is widely observed in peripheral and coronary arteries, the only reported effects of exercise on endothelial function in carotid arteries have been obtained in rabbits and are negative (7,50,51). Thus, evidence is lacking that exercise-induced reductions in cerebrovascular events can be linked to improved EDD and reduced adhesion molecule expression. Thus, we sought to determine whether chronic exercise training reduced carotid artery endothelial inflammation, i.e., VCAM-1 expression, and improved EDD using a swine model of atherosclerosis.
Assessment of endothelial function and inflammation in vivo could facilitate identification of active plaques at risk for rupture. Current methods of assessment of endothelial function in vivo, such as flow-mediated dilation and EDD, do not discriminate between the presence and absence of CAD in dyslipidemic patients (35) and do not correlate with the extent of coronary artery stenosis, carotid intima–media thickness (IMT), or intima–media (I/M) ratio (34). In addition, atherosclerotic plaque can be found without impairment of EDD (44).
Noninvasive, direct assessment of endothelial inflammation could provide a tool to monitor the response to treatment, because the reduction of inflammation could signify stabilization of the atherosclerotic plaques (33). Because inflammation of the endothelium is characterized by early expression of adhesion molecules such as VCAM-1, several human studies have focused on linking exercise to levels of soluble adhesion molecules, with inconsistent results (15,36). Therefore, we sought to examine the effects of exercise on inflammation using contrast-enhanced ultrasound (CEU) with VCAM-1–targeted microbubbles (MBs), a novel noninvasive approach to detect inflammation in vivo. We chose to examine these effects on carotid arteries, a vascular bed commonly used in humans to predict the extent of both cerebrovascular and CAD. We hypothesized that 1) exercise would reduce VCAM-1 expression in carotid arteries and that 2) VCAM-1–targeted MBs can detect carotid inflammation in vivo. To increase the translational potential of the study, we tested these hypotheses in the Rapacz familial hypercholesterolemic (FH) swine (32), an animal model that both develops spontaneous atherosclerosis and is of sufficient size to allow direct application of CEU as in humans.
MATERIALS AND METHODS
Fourteen male castrated Rapacz FH swine (32) were purchased from the University of Wisconsin Swine Research and Teaching Center. Pigs were individually housed in rooms maintained at 20°C–23°C with a 12:12-h light–dark cycle and had ad libitum access to water. Pigs were fed the University of Wisconsin gestation diet consisting of a corn- and soybean-based cholesterol-free, 3% fat diet for the entire duration of the study. At 10–11 months of age, pigs were familiarized with the treadmill and then randomly assigned into exercise-trained (Ex, n = 7) or cage-confined/sedentary (Sed, n = 7) groups.
Approximately 1 wk before the end of the 16- to 20-wk training program, pigs were sedated (telazol–xylazine) and underwent dual-energy x-ray absorptiometry (DEXA) scanning (QDR 4500A; Hologic, Inc., Bedford, MA) to assess body composition and to collect blood from the right external jugular vein. Animals were placed supine on the DEXA table in a fixed position and scanned once. Body composition (head, left and right thoracic and pelvic limbs, and trunk) was determined by an experienced technician using computer software (QDR Software for Windows XP, version 12.4, Hologic, Inc.). Percentage body fat was expressed as a ratio of fat mass (g) divided by the total mass (g) × 100.
At the end of the training program, swine were sedated with intramuscular telazol–xylazine, induced, and maintained under general anesthesia with isoflurane inhalation (approximately 1.5%–2%). Approximately 20 mL of blood was collected from the right external jugular vein shortly after induction. An ear vein catheter was placed followed by the administration of an initial loading dose of heparin (300 U·kg−1 intravenously) with maintenance doses of 100 U·kg−1 intravenously each hour. Heart rate, saturation of peripheral oxygen, rectal temperature, and electrocardiography were monitored during anesthesia. CEU with MBs (targeted and nontargeted) was performed first followed by in vivo measurement of EDD using acetylcholine (ACH) infusion. At the end of the experiments, the animal was euthanized and the common carotid artery (CCA), including the proximal carotid bifurcation (called the bifurcation later in the text) from the brachiocephalic trunk, was harvested and placed in cold physiologic saline solution. The left CCA and mid-portion of the right CCA were cleaned of fat and connective tissue and divided into several segments for immunohistochemistry/histology, Western blot, quantitative reverse transcription polymerase chain reaction (qRT-PCR), and in vitro endothelial function assessment with consistency between animals.
Lastly, forelimb skeletal muscles including the triceps brachii lateral head (both deep (D) and superficial (S) portions), triceps brachii accessory head (TAH), triceps brachii medial head, triceps brachii long head (both D and S portions), and deltoid were harvested and frozen for measurement of citrate synthase activity. Experimental protocols were approved by the University of Missouri Animal Care and Use Committee and in accordance with the “Principles for the Utilization and Care of Vertebrate Animals used in Testing, Research and Training.”
Experimental treatments lasted for 16–20 wk during which Sed pigs were restricted to their pens and Ex pigs underwent an exercise-training regimen on treadmills as previously described (6). Briefly, pigs assigned to the Ex group completed a 16- to 20-wk endurance training program consisting of five exercise bouts per week. At the start of the training program, each exercise bout consisted of a 5-min warm-up, followed by 15 min of run at 5 mph and a 20- to 30-min run at 3 mph. The intensity and duration of the exercise bouts were increased weekly to maximize the training stimulus so that by week 10 of training, each exercise bout lasted 85 min and consisted of 5-min warm-up, a 15-min sprint run at 6–8 mph, a 60-min endurance run at 4–6 mph, and a 5-min cool down. Efficacy of the training program was determined from measurements of endurance time (from the treadmill exercise before and after test), forelimb skeletal muscle citrate synthase activity, and pig heart weight-to-body weight ratios.
MBs targeted to VCAM-1 antibody (VCAM-1 MBs) were prepared by Targeson Inc. (San Diego, CA) using a purified VCAM-1 antibody produced by 6G10 hybridoma (ATCC, Manassas, VA). Control MBs conjugated with rat immunoglobulin G antibody used to evaluate nonspecific attachment of MBs were also obtained from Targeson Inc. Pigs (Ex, n = 6; Sed, n = 6) were placed in dorsal recumbency. Proximal and distal portions of the left CCA were localized and the skin marked to ensure consistent transducer placement. Both proximal and distal segments of the left CCA were first imaged using 2-D brightness mode (HDI 5000; Philips, Andover, MA), followed by capture of background images using contrast mode (mechanical index of 0.06). Next, a bolus of 5 U (1.8–2.0 × 109 particles per milliliter) of VCAM-1–targeted or control MBs (2–3 μM in size, Targeson Inc.) was randomly administered intravenously. Approximately 3–4 and 6–7 min after administration, postcontrast images of both portions of the left CCA were captured. At 9 min, the mechanical index was increased to 0.6 to obtain postdestruction images. A delay of 20 min between injections was provided to allow dissipation of the first type of MBs administered. Images were reviewed offline using Image J software (National Institutes of Health, Bethesda, MD). Signal intensity from MBs present within the carotid arterial wall was measured using a region of interest (ROI) drawn around the near vessel wall including a minimal portion of the lumen adjacent to the wall. Post hoc review of the images demonstrated that the destruction process did not effectively destroy the MBs present in the field thus, we normalized the signal intensity measured in each ROI to the signal intensity measured on background images before each subsequent injection.
In vivo measurement of endothelial-dependent relaxation (EDD)
After CEU, pigs underwent assessment of carotid endothelial function using the endothelial agonist, ACH. To achieve local concentration of ACH of 10−6 and 10−5 M in the left CCA, total blood flow was determined as described previously by Tsutsui et al. (43) using the following equation: time-average velocity (cm·s−1) × π × (vessel diameter (cm)/2)2 × 60. The vessel diameter was first measured in a longitudinal plane with ultrasound calipers (HDI 5000, Philips) at R wave using 2-D brightness mode. With the transducer still in place, we switched to pulsed-wave Doppler ultrasonography and obtained the time-average velocity on the basis of three cardiac cycles. A JR SH guide catheter (Boston Scientific, Natick, MA) was introduced through a 7F sheath into the right femoral artery, and the distal tip of the guide catheter was positioned within the most proximal portion of the left CCA. The infusion protocol consisted of 2.5 min of saline, 5 min of ACH 10−6 M, 5 min of ACH 10−5 M, and 7.5 min of saline at a constant infusion rate of 1 mL·min−1 in 12 pigs (Ex, n = 6; Sed, n = 6). In two pigs, one of each group, only the first concentration of ACH (10−6 M) was infused. Before starting the infusion, gain settings and depth using 2-D brightness mode were optimized for both carotid portions and kept constant during the entire length of infusion. Immediately after initiation of the infusion, two ultrasound images of the proximal and distal segments of the left CCA were captured at R wave (baseline). The infusion protocol lasted 20 min during which two images per minute of each carotid portion were captured, for a total of 84 images per subject. Systolic and diastolic blood pressure was also recorded at baseline and every minute of the infusion protocol. Off-line analyses of proximal and distal CCA diameters were performed using a custom-designed edge detection and wall tracking software developed by MJ Davis (Vision Development Module; LabVIEW, National Instruments, Austin, TX). In brief, images were spatially calibrated using the original image file, and three regions of interest were placed along the longitudinal axis of the carotid. The internal edges of the artery were detected by thresholding and iterative regression procedures similar to a method previously described (9,38). Location of the ROIs was maintained for all images of the same pig. ROIs from each image and time point were averaged, and diameters were then normalized to mean arterial pressure and expressed as a percentage change in carotid diameter from saline infusion. Carotid arterial diameters after infusion of saline and ACH were normalized to mean arterial pressure to account for the systemic effect of ACH resulting in 3%–11% decrease in mean arterial pressure. Reduction in mean arterial pressure to up to 20% has been reported in rats using similar concentration of ACH (10). Heart rate was only marginally affected by the ACH infusion. To determine the maximal dilator response of the carotid artery, we attempted to infuse sodium nitroprusside (SNP), an endothelium-independent agonist, in the carotid artery. However, the dramatic decrease in blood pressure prevented us from repeating this protocol in other pigs.
In vitro assessment of EDD
Isometric tension was determined on arterial segments 3 to 4.5 mm in length from the mid-portion of the left and right CCA from five pigs in each group in a 20-mL bath filled with Krebs bicarbonate buffer maintained at 37°C with a gas mixture of 95% O2 and 5% CO2. Arterial segments were allowed 1 h to equilibrate to Krebs bicarbonate buffer. Before concentration–response curves, carotid rings were stretched to optimal length (Lo) using 30 mM KCl as described previously (50). Carotid rings were preconstricted with prostaglandin F2alpha (PGF2-α; 30 μM), and endothelium-dependent relaxation was assessed using ACH (10−10 to 10−4 M) and bradykinin (BK, 10−11 to 10−6 M). SNP (10−10 to 10−4 M) was used to measure endothelium-independent relaxation. A total of five carotid rings from each animal were studied in parallel. In three rings (control), vasorelaxation to agonists only was measured by adding cumulatively increasing doses of the selected drug to the organ bath while measuring changes in force. Two rings received NG-nitro-L-arginine methyl ester (L-NAME, 300 µM) to determine the role of NO in vasoactive responses by blocking NO synthase. Dose–response curves to ACH were studied first, followed by BK and SNP. At the end of each dose–response protocol, arterial rings were washed out several times by replacing bicarbonate buffer solution and arterial rings were allowed the necessary time to stabilize before initiation of the next protocol.
Carotid arterial segments were cut longitudinally. TRIzol solution (TRI Reagent; Molecular Research Center, Inc., Cincinnati, OH) was deposited on the luminal side of the opened arterial segment in enough amounts to cover the entire surface. After 30 s, endothelial cells were scraped with a sterile scalpel and collected in 1 mL of TRIzol solution. Endothelial cell scraps were quick frozen in liquid nitrogen and stored at –80°C until processing. Total RNA was isolated according to the manufacturer’s published protocol. cDNA was transcribed using high-capacity cDNA reverse transcription kit (Applied Biosystems; Life Technologies Corp., Carlsbad, CA). A minus reverse transcriptase reaction was also performed to ensure no genomic DNA contamination. qRT-PCR was performed on a MyiQ Single-Color real-time polymerase chain reaction detection system (Bio-Rad, Herculese, CA) using reaction conditions optimized for each set of primers: VCAM-1 (sense, TCTGGAATTTACGTGTGCGAGGGA, antisense, AGCTGGAAAGCCATCACTAGAGCA), endothelial nitric oxide synthase (eNOS; sense, GCCTGAACAGCACAGGAGTT, antisense, GCTCTTCTAGCCGTGTGTCC), and β-actin (sense, GGACTTCGAGCAAGAGATGG, antisense, AGCACTGTGTTGTTGGCGTACAG), and gene expression was normalized to β-actin using the 2(−ΔΔC(t)) method (22). Linearity and efficiency of each polymerase chain reaction condition were verified by creating a standard curve plotting the critical threshold versus log of the dilution of cDNA.
Shortly after harvest, carotid arterial segments were opened with the luminal surface side-up, covered with approximately 80–100 µL of ice-cold homogenization buffer (50 mM Tris–HCl, pH 7.4, 0.1 mM ethylenediaminetetraacetic acid, 0.1 mM ethylene glycol tetraacetic acid, 0.1% mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 2 mM leupeptin, 1 mM pepstatin A, 10% vol/vol glycerol, 0.1% Triton X-100) and scraped with the dull surface of an X-Acto blade. Equal protein amounts (10 µg) of sample were electrophoresed on 4%–20% sodium dodecyl sulfate-polyacrylamide gels and transferred to a polyvinylidene difluoride membrane, blocked for 1 h and probed with anti-eNOS phosphorylated at serine residue 1177 (1:400, BD Transduction Labs #cat: 612392), anti-VCAM-1 antibody (1:100, from hybridoma 6G10), total eNOS (1:5000, BD Transduction Labs, Oxford, England), and β-actin (1:2000, Sigma, Washington, DC). Membranes were stripped between VCAM-1 and eNOS and between eNOS and β-actin using a Blot Restore Solution kit (Millipore, Billerica, MA). Goat anti-mouse (1:1000, Santa Cruz Biotechnologies, Santa Cruz, CA) and sheep anti-mouse (1:2500–10,000, GE Healthcare, Little Chalfont, UK) were used as secondary antibodies and were incubated for 1 h at room temperature. Immunoreactive bands were detected by chemiluminescence (ECL, GE Healthcare) and quantified using the National Institutes of Health Image J software.
Formalin-fixed tissue was embedded into paraffin, and consecutive sections were prepared and stained for eNOS phosphorylated at serine residue 1177 (1:800, BD Transduction Labs #cat: 612392), total eNOS (1:800, BD Transduction Labs), VCAM-1 (no dilution, from hybridoma 6G10), SRA-E5 (1:100, 1TransGenic Inc., Kumamoto, Japan), and nitrotyrosine (1:400, Chemicon, Temecula, CA) using an LSAB+ kit from Dako (Glostrup, Denmark). Photomicrographs of each stained sections from the bifurcation, as well as the proximal and distal CCA of all pigs, were captured at 10× magnification using Olympus MicroSuite Biological Suite Software connected to an Olympus BX61 motorized system microscope (Leeds Precision Instruments, Minneapolis, MN).
VCAM-1, scavenger receptor A (SRA), nitrotyrosine, phospho-eNOS, and eNOS staining were analyzed using Image-Pro Plus Software (version 6.2; Media Cybernetics, Inc., Bethesda, MD). Positive staining for each marker (except SRA) was expressed as the sum of positively stained endothelial areas divided by the luminal circumference of each arterial section. SRA staining was quantified by determining the percentage of intimal area positively stained in each arterial section.
Paraffin sections from the bifurcation, proximal, and distal CCA were stained for elastic fibers with Verhoeff–Van Gieson. Each arterial section was then captured on Olympus BX61 microscope as previously described for other markers. Maximal IMT, intimal areas, and medial areas were measured using Image-Pro Plus. IMT was defined as the distance between the external elastic lamina and the internal luminal border of the artery. Maximal IMT was measured three times at the widest part of the arterial section and averaged. Intimal area was defined as the area between the internal elastic lamina and the luminal border of the artery, whereas the area between the internal elastic lamina and external elastic lamina was called the medial area. Intima–media ratio (I/M) was calculated as the ratio of intimal over medial areas.
Fresh whole blood was submitted to the Veterinary Medical Diagnostic Laboratory of the University of Missouri the day of DEXA scanning for standard chemistry profiles including glucose. Frozen plasma collected the day of euthanasia was later submitted for quantification of triglycerides, total, LDL, and HDL cholesterol blood content. Results from FH swine were compared with values measured in non-FH domestic swine. Citrate synthase activity was determined using a method described by Srere (39).
Unless otherwise stated, outcome variables were compared using two-way ANOVA with repeated measures when appropriate, followed by post hoc testing for individual comparisons when justified. CEU signal intensity and animal characteristics were analyzed using two-tailed Student’s t-test or Mann–Whitney rank sum test where appropriate. Concentration–response curves for carotid arterial rings were analyzed with nonlinear regression curves to determine the half maximal effective concentration and then compared between groups using Student’s t-test. Maximal percentage relaxation was used to compare the effect of L-NAME between groups using two-way ANOVA on repeated measures. Statistical analyzes were performed with Sigma Stat Version 3.5 for Windows (Dundas Software, Erkrath, Germany) except linear regression for which GraphPad Prism 5.0d for Mac OS X (GraphPad Software, San Diego, CA) was used.
Clinical characteristics of exercise-trained (Ex) and Sed FH pigs are illustrated in Table 1. Efficacy of exercise training was demonstrated by a 33% increased endurance time between pre- and poststress tests in Ex as well as 34% and 36% increased muscle citrate synthase activity in the deltoid and TAH muscles in the Ex pigs, respectively. With exercise training, heart weight and heart weight-to-body weight ratio showed an increasing trend, which reached significance when heart weight was normalized to lean mass. Body weight and lean mass (total and individual anatomic regions) were not affected by training. Both Ex and Sed FH pigs were characterized by increased blood levels of total and LDL cholesterol and triglycerides, as shown previously (14); however, exercise training did not significantly alter overall or regional percentage body fat, triglycerides, LDL, HDL, or total cholesterol.
Effect of training on in vivo VCAM-1 expression.
Three minutes after injection of either type of MBs, the arterial wall was better visualized than on background images, and MBs were no longer seen freely circulating within the arterial lumen (Fig. 1). Signal intensity derived from specific attachment of VCAM-1–targeted MBs was determined as the ratio of signal intensity from VCAM-1-to-control MBs. Because no difference was seen between signal intensity measured at 3 and 6 min, we combined data from both time points. Irrespective of group, mean VCAM-1-to-control signal intensity was greater than 1.0, indicating elevated VCAM expression in the FH carotids. However, exercise training did not result in a significant decrease in VCAM-1-to-control signal intensity in either proximal (Ex, 1.46 ± 0.55; Sed, 1.94 ± 0.30; P = 0.24) or distal (Ex, 1.55 ± 0.65; Sed, 1.85 ± 0.46; P = 0.71) CCA (Fig. 2A, B). We validated the assessment of carotid inflammation in vivo by examining the correlation between the signal intensity derived from VCAM-1-to-control MBs ratio and VCAM-1 protein expression in the proximal and distal CCA. A linear relationship was found between VCAM-1-to-control MBs signal intensity and VCAM-1 protein expression in the proximal (r2 = 0.50, P < 0.01) and distal (r2 = 0.52, P < 0.01) CCA (Fig. 2E, F).
Effect of training on VCAM-1 and indices of disease
A twofold increase in VCAM-1 mRNA was found in Ex versus Sed in both proximal (P < 0.05, data not shown) and distal CCA (P < 0.05, data not shown), although this increase was not reflected at the protein level (Fig. 2C, D). Exercise training had no significant effect on any of the parameters studied, including endothelial VCAM-1 protein expression, intimal macrophage infiltration, endothelial oxidative stress, carotid maximal IMT, and carotid intima–media ratio. Conversely, a main effect of location of carotid segment was found for all immunohistochemical markers of disease severity. Indeed, endothelial VCAM-1 expression and intimal macrophage accumulation were approximately two- to fivefold greater in the bifurcation compared with the proximal and distal segments of the CCA (main effect of segment: eVCAM-1, P < 0.05; SRA, P < 0.001; Fig. 3).
Effect of training on expression of markers of endothelial cell health
Expression of eNOS mRNA and protein was not affected by exercise training in the proximal and distal CCA, although a trend for an increase in eNOS protein expression was found in Ex in the proximal CCA (2.66 ± 0.85 vs 1.0 ± 0.18 in Sed, P = 0.20; Fig. 4C). Similar findings were obtained for eNOS phosphorylation, an important regulator of eNOS activity (Ex, 2.23 ± 1.06, vs Sed, 1.0 ± 0.27, P = 0.62; Fig. 4A). The complex anatomy of the bifurcation precluded obtaining protein; therefore, we used immunohistochemistry to directly compare eNOS in the bifurcation, proximal, and distal CCA. Interestingly, the proportion of phosphorylated and total eNOS staining throughout the vessel circumference was greater in the distal CCA (two- to fourfold) compared with the bifurcation and proximal CCA, although the ratio of phosphorylated-to-total eNOS was similar throughout the carotid arteries (data not shown).
In vivo and in vitro assessment of endothelial-dependent vasoreactivity
Baseline carotid diameter, velocity, and blood flow as well as mean arterial pressure were similar between groups (Table 1). ACH infusion resulted in a significant increase in diameter of the proximal and distal CCA, relative to saline, in both groups (P < 0.01); however, this response was not affected by exercise training (Fig. 5A, B). A trend for a greater increase in carotid diameter with the highest dose of ACH (10−5 M) was seen in both portions of the vessel, although it did not reach statistical significance. When studied in vitro, internal diameter, length, resting tension, percentage stretch, and PGF2-α were not different between Ex and Sed in this study (Table 1). ACH and BK elicited dose-dependent relaxation in Ex and Sed (Fig. 5C, E). A significant interaction was found between training and concentration dose of either ACH (P < 0.05) or BK (P < 0.05); however, pairwise comparison procedures (Holm–Sidak method) failed to demonstrate a significant difference between Ex and Sed at any of the concentrations used. Moreover, the half maximal effective concentration for neither ACH nor BK was affected by training. Maximal EDD was observed at mid-range concentrations of ACH, similar to those used during local infusion. L-NAME blocked the majority of relaxation to both agonists, suggesting that NO is the major contributor to EDD in carotid arteries of FH pigs, an effect that was not modified by training (Fig. 5D, F). Carotid rings from both groups exhibited a dose-dependent relaxation to SNP that was not affected by training (data not shown).
The main goals of this study were to investigate the effects of training on carotid inflammation and to evaluate the use of VCAM-1–targeted MBs for detection of inflammation in vivo in a swine model of atherosclerosis. The major findings of this study are as follows: 1) contrary to our hypothesis, exercise training did not result in reduction of VCAM-1 protein expression along the endothelial surface of the carotid arteries; and 2) in vivo detection of inflammation using VCAM-1–targeted MBs correlated with the gold standard measure of in vitro VCAM-1 protein expression.
Unaltered carotid VCAM-1 expression with exercise training
To our knowledge, there are no human and very few animal studies (6,8,27) that have looked at the effect of regular exercise in patients with FH or animal models with FH. Exercise has been previously shown to exert effects that may vary depending on the intensity, type (aerobic vs resistance training), and duration of the training program (16). In our institution, efficacy of the training program has been traditionally assessed by increased forelimb muscle citrate synthase activity, longer endurance time on the treadmill, increased heart weight, and increased heart weight-to-body weight ratio, and our results for these variables are in line with those reported in previous studies using this training protocol and performed in FH pigs of similar age and in miniature swine (6,20). It is possible that 20 wk of exercise training is insufficient to induce a measurable change in VCAM-1 expression. In humans, it has been suggested that exercise alters arterial wall thickness over time frame of several months and years (41). Longitudinal studies using noninvasive assessment of VCAM-1 expression over longer exercise training program may be necessary to detect measurable attenuation in pro-inflammatory pathways, leading to changes in VCAM-1, such as maintenance of NO availability and repression of nuclear factor κB activation, an important transcription factor involved in VCAM-1 gene regulation (37,47).
The lack of exercise effect on VCAM-1 expression in the carotid arteries may also be associated with the unaltered blood lipid profile in FH pigs. These results are in line with those obtained in diet-induced hypercholesterolemia in miniature swine submitted to a similar training protocol and thus not unique to familial hypercholesterolemia (42). In a previous study by our group (unpublished data), carotid endothelial VCAM-1 was directly correlated with blood glucose and LDL cholesterol in Sed FH pigs. Perhaps the absence of effect of chronic exercise on the lipid profile or blood glucose may have contributed to similar levels of inflammation in this study.
Another factor that might have contributed to the lack of exercise effects on the expression of VCAM-1 is the early stage of atherosclerosis present in the carotid arteries in our subjects. Early in the development of atherosclerosis, the vessel wall may undergo remodeling changes such as outward remodeling to maintain normal shear rate and homeostatic functions of the endothelium. It is possible that in our pigs, activation of anti-inflammatory mechanisms may still be sufficient to counteract the hypercholesterolemic state in both Ex and Sed, thus limiting a generalized switch toward an endothelial inflammatory phenotype and maintaining relatively low levels of VCAM-1 expression in the carotid arteries. In this early state of compensating mechanisms, exercise may not provide additional benefits.
Moreover, exercise has been previously shown to exert effects that may differ between various vascular beds (16). In humans, endurance training may result in improvement of IMT in peripheral active regions (e.g., femoral arteries) although carotid arteries are unaffected (41). We limited our investigation to the CCA to increase the translational potential of our findings because this vessel is already commonly used in humans for measurements of IMT. The superficial and easily accessible location of the human carotid artery makes this arterial bed ideal for obtainment of high-resolution images noninvasively using percutaneous ultrasound in humans. An additional advantage of carotid arteries over other peripheral arteries such as the brachial is that the former can not only predict the extent of CAD but also assess local disease and therefore contribute to the stratification of risk for ischemic stroke (2). In our model, the effect of exercise on VCAM-1 expression in vascular beds other than the carotid arteries still needs to be investigated. In the miniature swine model with early CAD, 20 wk of exercise did not alter the expression of inflammatory markers including VCAM-1 in the coronary arteries, whereas antioxidant capacity was improved in the coronary arterioles (1), supporting regional differences for at least the antioxidant pathways in the coronary vasculature. VCAM-1 expression in the carotid arteries was not investigated in that study. It is plausible that exercise has limited effects on pathways directly involved in regulation of inflammation but rather contributes to activation of compensatory mechanisms such as up-regulation of antioxidant pathways.
Consistent with our finding that VCAM-1 is not reduced with exercise, we found no training effect on protein expression of other markers of atherosclerotic disease such as intimal macrophage accumulation, oxidative stress, IMT, and intima–media ratio in carotid arteries. The only significant effect of exercise found in this study was an increase in VCAM-1 mRNA, which did not translate to the protein level. The cause of discrepancy between mRNA and protein expression of VCAM-1 in CCA is unknown but could include transcriptional inhibition, posttranscriptional mechanisms, e.g., protein degradation, or inability to measure small differences in protein levels using Western blot or immunohistochemistry. Contrary to our expectations, KLF2, a transcription factor up-regulated by laminar flow and involved in atheroprotective mechanisms (46), was decreased with exercise. KLF2 is an important regulator of VCAM-1 gene expression in endothelial cells, through regulation of nuclear factor κB nuclear accumulation (21); thus, it is possible that the increase in VCAM-1 mRNA observed with exercise were mediated in part through KLF2.
Because controversy exists as to whether atherosclerosis plaque initiation has already begun before observation of altered EDD and because several studies have shown restoration of EDD with exercise, we concurrently measured EDD and VCAM-1 expression in carotid arteries. We found no evidence for impaired EDD in either group, despite the presence of early atherosclerosis, i.e., VCAM-1 expression and macrophage infiltration, providing evidence that endothelial inflammation precedes dysfunction. However, exercise training did not increase in vivo EDD as evidenced by the finding that carotid diameter after ACH/saline infusion was similar between Ex and Sed.
The lack of training effect on in vivo EDD is in accordance with the absence of a significant training effect on markers of endothelial NO production (phosphorylated and total eNOS) and on in vitro ED vasomotor function to ACH and BK performed on carotid artery rings. L-NAME almost abolished EDD in carotid arteries, indicating that in our study, relaxation is primarily dependent on NO release from an NO synthase–dependent pathway, which is in agreement with a previous study (44).
Previously reported effects of exercise on endothelial function in carotid arteries are limited to rabbits and have failed to demonstrate an exercise response in either normal or hypercholesterolemic animals (7,50,51). We are unaware of other studies examining chronic exercise on carotid EDD in large mammals such as swine. The invasive nature of the procedure has limited evaluation of carotid endothelial function in humans. In our study, we did not evaluate EDD in other arterial beds and thus are unaware if the lack of exercise effects in FH is unique to the carotid arteries or uniformly distributed throughout the arterial tree. In a previous study from our institution and performed on a different subset of FH pigs of similar age, exercise did not improve EDD in coronary arteries but reduced maximal contractile response to endothelin-1 (6). More studies would be needed to determine whether exercise has beneficial effects on carotid arteries that were not assessed in this study.
Another factor to consider for the lack of exercise effects on carotid EDD is that carotid endothelial function may have been unimpaired in our FH pigs. In a diet-induced swine model of early vascular disease, hypercholesterolemia was shown to impair ED vasomotor function in carotid arteries (44). Although we did not directly compare our FH swine to non-FH, the relative relaxation to ACH and BK obtained in our study was greater to that reported for non-FH hypercholesterolemic swine and comparable with that of pigs fed a chow diet (44), leading us to conclude that EDD was not impaired in carotid arteries in our FH swine. The effect of exercise training on vasomotor function of healthy individuals is equivocal (30). Indeed, in healthy Ex pigs, 20 wk of exercise training does not result in further enhancement of peripheral conduit arteries (brachial and femoral arteries) compared with their Sed but otherwise healthy counterparts. Evidence indicates that EDD is transiently enhanced early after the onset of training in response to increased shear rate and then lost with continuation of the training and structural remodeling (16,30). Because the goal of our study was not to directly assess arterial remodeling over time, we did not measure carotid luminal diameter and wall thickness before or during training. However, Ex animals neither revealed a greater carotid diameter nor decreased wall thickness compared with Sed pigs at the end of the training program.
The relatively small number of animals per group and the variability of the response to either agonist limit the statistical power of these analyses; thus, there should be caution in interpreting negative findings. Although expensive, long-term studies with a larger number of animals in which EDD could be assessed over time would be needed to definitely conclude that exercise training has no effect on carotid arterial EDD or eNOS protein (19,30).
Validation of in vivo detection of inflammation
Our data from CEU with VCAM-1–targeted MBs were positively correlated protein expression in both portions of the CCA, validating our method and minimizing the effect of the transducer location on the interpretation of the in vivo results. This is consistent with studies using rodent models of atherosclerosis in which preferential attachment of VCAM-1–targeted agents to regions of early plaque formation was demonstrated (17,18,26). Interestingly, despite the early stages of atherosclerosis in the CCA as indicated by less than 10% of the intima infiltrated by macrophages and intima/media ratios of only 1.5% to 3.5%, the greater signal intensity obtained with the VCAM-1–targeted MBs relative to the control MBs suggests that CEU is a very sensitive approach for detection of initiation of disease. If these results are proven to be similar in humans, this could dramatically affect our ability to detect early signs of carotid atherosclerosis in patients at risks, resulting in earlier and more adapted treatment, and may ultimately help reduce the incidence of cerebrovascular events.
An important limitation of our study was the inability to distinguish MBs attached to the endothelium apart from those potentially adhered to endothelial cells of the vasa vasorum. In pigs, the depth of the carotid arteries requires increased penetration of the ultrasound beam at the expense of the axial resolution. In addition, the early stage of atherosclerosis in our pigs limited any substantial intimal thickening, limiting our ability to easily identify the intima, media, and adventitia individually. Because it was not possible to accurately isolate endothelial-specific signal, we measured the signal intensity in the carotid arterial wall, including a minimal portion of the lumen. We therefore may have potentially included signal from VCAM-1–targeted MBs adhered to activated endothelial cells within vessels belonging to the most inner portion of the adventitia. We did not attempt to measure VCAM-1 staining in the adventitia, although we believe that adhered VCAM-1–targeted MBs in the adventitia would have contributed only a small portion of the signal detected if any because VCAM-1 staining in the adventitia of proximal and distal carotid arterial sections was either absent or minimal in all pigs except one with moderate adventitial staining. Nonetheless, none of pigs had more VCAM-1 staining in the adventitia compared with the endothelium.
In retrospect, it would have been interesting to obtain CEU data from the bifurcation of the carotid arteries because this region had the greatest expression of endothelial VCAM-1 and would have provided a wider range of VCAM-1 expression for our regression analysis. However, ultrasound imaging of the bifurcation requires removal of the first rib and therefore not suited for the development of a noninvasive approach designed to be used in longitudinal studies.
Nonuniformity of markers of atherosclerosis and endothelial cell health within the CCA
A consistent finding in our study is the increased levels of markers of atherosclerotic disease in the bifurcation compared with the proximal and distal segments of the CCA. Indeed, increased levels of carotid VCAM-1 and intimal infiltrated macrophages were associated with increased IMT, a commonly used index of disease severity previously shown to predict cardiovascular event (5). These findings are consistent with the preferential development of atherosclerotic lesions at bifurcations and curvatures of the arterial tree, regions associated with low shear stress, and oscillatory or turbulent flow. Similarly, eNOS was significantly increased in the distal CCA compared with the bifurcation and proximal CCA, both in its phosphorylated and nonphosphorylated forms. On the other hand, oxidative stress was unexpectedly greater in the distal CCA, as shown by a twofold increase in nitrotyrosine staining compared with the proximal CCA. It is generally accepted that oxidative stress is associated with atherosclerosis and increased levels are usually found at regions of disturbed flow (4). Given the close proximity between the proximal CCA and the flow divider, we would predict more disturbed flow at the proximal than the distal CCA and consequently lower levels of oxidative stress in the distal versus proximal CCA. Although the reason for the discrepant finding is unknown, nitrotyrosine is a marker of peroxynitrite, which is formed by the reaction of superoxide with NO. Therefore, paradoxically, the greater nitrotyrosine may result from more abundant NO in the distal compared with the proximal CCA. Overall, these results support an inverse relationship between markers of disease and endothelial NO production and are consistent with the preferential development of atherosclerotic lesion at regions of disturbed flow such as bifurcations.
In conclusion, this study suggests that exercise training has no effect on endothelial inflammation or atherosclerosis and does not contribute to improve EDD in carotid arteries of FH swine. These conclusions cannot be extrapolated to the entire arterial tree because exercise effects are known to be complex, vascular bed specific and may involve broader end points than those we examined. Of importance, this study validated the use of CEU with VCAM-1–targeted MBs to detect carotid inflammation in vivo and provides the foundation for the use of VCAM-1–targeted MBs to noninvasively determine the inflammatory response to treatment over time.
This work was supported by the Canadian Institutes of Health Research Fellowship and National Heart, Lung, and Blood Institute Grant HL52490.
We thank Jan Ivey and Dr. Darla Tharp for their help during in vivo procedures, Dave Harah for tissue collection, Dr. Arturo Arce for work with ring studies, Pam Thorne for work with ring studies and measurement of citrate synthase activity, Miles Tanner for help with mRNA studies, Jennifer Casati and Alexa Bermudez for work with immunohistochemistry, Melissa Morehead for histologic measurements, and Jack Rychak from Targeson Inc. for production of MBs.
The authors have no conflicts of interest pertaining to this work.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Arce-Esquivel AA, Kreutzer KV, Rush JW, Turk JR, Laughlin MH. Exercise does not attenuate early CAD progression in a pig model. Med Sci Sports Exerc
. 2012; 44 (1): 27–38.
2. Barnett HJ, Gunton RW, Eliasziw M, et al.. Causes and severity of ischemic stroke in patients with internal carotid artery stenosis. JAMA
. 2000; 283 (11): 1429–36.
3. Belardinelli R, Georgiou D, Cianci G, Purcaro A. Randomized, controlled trial of long-term moderate exercise training in chronic heart failure: effects on functional capacity, quality of life, and clinical outcome. Circulation
. 1999; 99 (9): 1173–82.
4. Bonomini F, Tengattini S, Fabiano A, Bianchi R, Rezzani R. Atherosclerosis and oxidative stress. Histol Histopathol
. 2008; 23 (3): 381–90.
5. Bots ML, Baldassarre D, Simon A, et al.. Carotid intima–media thickness and coronary atherosclerosis: weak or strong relations? Eur Heart J
. 2007; 28 (4): 398–406.
6. Bunker AK, Laughlin MH. Influence of exercise and perivascular adipose tissue on coronary artery vasomotor function in a familial hypercholesterolemic porcine atherosclerosis model. J Appl Physiol
. 2010; 108 (3): 490–7.
7. Chen HI, Li HT. Physical conditioning can modulate endothelium-dependent vasorelaxation in rabbits. Arterioscler Thromb
. 1993; 13 (6): 852–6.
8. Company JM, Booth FW, Laughlin MH, et al.. Epicardial fat gene expression after aerobic exercise training in pigs with coronary atherosclerosis: relationship to visceral and subcutaneous fat. J Appl Physiol
. 2010; 109 (6): 1904–12.
9. Davis MJ. An improved, computer-based method to automatically track internal and external diameter of isolated microvessels. Microcirculation
. 2005; 12 (4): 361–72.
10. Granstam SO, Granstam E, Fellstrom B, Lind L. Effects of acetylcholine and nitroprusside on systemic and regional hemodynamics in hypertensive rats. Clin Exp Hypertens
. 1998; 20 (2): 223–43.
11. Green DJ. Exercise training as vascular medicine: direct impacts on the vasculature in humans. Exerc Sport Sci Rev
. 2009; 37 (4): 196–202.
12. Hambrecht R, Adams V, Erbs S, et al.. Regular physical activity improves endothelial function in patients with coronary artery disease by increasing phosphorylation of endothelial nitric oxide synthase. Circulation
. 2003; 107 (25): 3152–8.
13. Hambrecht R, Wolf A, Gielen S, et al.. Effect of exercise on coronary endothelial function in patients with coronary artery disease. N Engl J Med
. 2000; 342 (7): 454–60.
14. Hasler-Rapacz J, Prescott MF, Von Linden-Reed J, Rapacz JM Jr, Hu Z, Rapacz J. Elevated concentrations of plasma lipids and apolipoproteins B, C-III, and E are associated with the progression of coronary artery disease in familial hypercholesterolemic swine. Arterioscler Thromb Vasc Biol
. 1995; 15 (5): 583–92.
15. Hatunic M, Finucane F, Burns N, Gasparro D, Nolan JJ. Vascular inflammatory markers in early-onset obese and type 2 diabetes subjects before and after three months’ aerobic exercise training. Diab Vasc Dis Res
. 2007; 4 (3): 231–4.
16. Jasperse JL, Laughlin MH. Endothelial function and exercise training: evidence from studies using animal models. Med Sci Sports Exerc
. 2006; 38 (3): 445–54.
17. Kaufmann BA, Carr CL, Belcik JT, et al.. Molecular imaging of the initial inflammatory response in atherosclerosis: implications for early detection of disease. Arterioscler Thromb Vasc Biol
. 2010; 30 (1): 54–9.
18. Kaufmann BA, Sanders JM, Davis C, et al.. Molecular imaging of inflammation in atherosclerosis with targeted ultrasound detection of vascular cell adhesion molecule-1. Circulation
. 2007; 116 (3): 276–84.
19. Laughlin MH, Pollock JS, Amann JF, Hollis ML, Woodman CR, Price EM. Training induces nonuniform increases in eNOS content along the coronary arterial tree. J Appl Physiol
. 2001; 90 (2): 501–10.
20. Laughlin MH, Welshons WV, Sturek M, et al.. Gender, exercise training, and eNOS expression in porcine skeletal muscle arteries. J Appl Physiol
. 2003; 95 (1): 250–64.
21. Lin Z, Natesan V, Shi H, et al.. A novel role of CCN3 in regulating endothelial inflammation. J Cell Commun Signal
. 2010; 4 (3): 141–53.
22. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods
. 2001; 25 (4): 402–8.
23. Maresi E, Midulla R, Cospite V, et al.. Transmural coronary inflammation triggers simultaneous multivessel rupture of unstable plaques. Ital Heart J
. 2003; 4 (7): 488–91.
24. Marwick TH, Hordern MD, Miller T, et al.. Exercise training for type 2 diabetes mellitus: impact on cardiovascular risk: a scientific statement from the American Heart Association. Circulation
. 2009; 119 (25): 3244–62.
25. Mathur N, Pedersen BK. Exercise as a mean to control low-grade systemic inflammation. Mediators Inflamm
. 2008; 2008: 109502.
26. Nahrendorf M, Jaffer FA, Kelly KA, et al.. Noninvasive vascular cell adhesion molecule-1 imaging identifies inflammatory activation of cells in atherosclerosis. Circulation
. 2006; 114 (14): 1504–11.
27. Napoli C, Williams-Ignarro S, De Nigris F, et al.. Long-term combined beneficial effects of physical training and metabolic treatment on atherosclerosis in hypercholesterolemic mice. Proc Natl Acad Sci U S A
. 2004; 101 (23): 8797–802.
28. Niebauer J, Hambrecht R, Velich T, et al.. Attenuated progression of coronary artery disease after 6 years of multifactorial risk intervention: role of physical exercise. Circulation
. 1997; 96 (8): 2534–41.
29. O’Connor CM, Whellan DJ, Lee KL, et al.. Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA
. 2009; 301 (14): 1439–50.
30. Padilla J, Newcomer SC, Simmons GH, Kreutzer KV, Laughlin MH. Long-term exercise training does not alter brachial and femoral artery vasomotor function and endothelial phenotype in healthy pigs. Am J Physiol Heart Circ Physiol
. 2010; 299 (2): H379–85.
31. Piek JJ, van der Wal AC, Meuwissen M, et al.. Plaque inflammation in restenotic coronary lesions of patients with stable or unstable angina. J Am Coll Cardiol
. 2000; 35 (4): 963–7.
32. Prescott MF, McBride CH, Hasler-Rapacz J, Von Linden J, Rapacz J. Development of complex atherosclerotic lesions in pigs with inherited hyper-LDL cholesterolemia bearing mutant alleles for apolipoprotein B. Am J Pathol
. 1991; 139 (1): 139–47.
33. Reilly SD, Litovsky SH, Steinkampf MP, Caulfield JB. Statins improve human coronary atherosclerotic plaque morphology. Tex Heart Inst J
. 2008; 35 (2): 99–103.
34. Rohani M, Jogestrand T, Kallner G, Jussila R, Agewall S. Morphological changes rather than flow-mediated dilatation in the brachial artery are better indicators of the extent and severity of coronary artery disease. J Hypertens
. 2005; 23 (7): 1397–402.
35. Rueda-Clausen CF, Lopez-Jaramillo P, Luengas C, del Pilar Oubina M, Cachofeiro V, Lahera V. Inflammation but not endothelial dysfunction is associated with the severity of coronary artery disease in dyslipidemic subjects. Mediators Inflamm
. 2009; 2009: 469169.
36. Saxton JM, Zwierska I, Hopkinson K, et al.. Effect of upper- and lower-limb exercise training on circulating soluble adhesion molecules, hs-CRP and stress proteins in patients with intermittent claudication. Eur J Vasc Endovasc Surg
. 2008; 35 (5): 607–13.
37. Shu HB, Agranoff AB, Nabel EG, et al.. Differential regulation of vascular cell adhesion molecule 1 gene expression by specific NF-kappa B subunits in endothelial and epithelial cells. Mol Cell Biol
. 1993; 13 (10): 6283–9.
38. Simmons GH, Padilla J, Young CN, et al.. Increased brachial artery retrograde shear rate at exercise onset is abolished during prolonged cycling: role of thermoregulatory vasodilation. J Appl Physiol
. 2011; 110 (2): 389–97.
39. Srere PA. Citrate synthase. Methods Enzymol
. 1969; 13: 3–11.
40. Taylor RS, Brown A, Ebrahim S, et al.. Exercise-based rehabilitation for patients with coronary heart disease: systematic review and meta-analysis of randomized controlled trials. Am J Med
. 2004; 116 (10): 682–92.
41. Thijssen DH, Cable NT, Green DJ. Impact of exercise training on arterial wall thickness in humans. Clin Sci (Lond)
. 2012; 122 (7): 311–22.
42. Thomas TR, Pellechia J, Rector RS, Sun GY, Sturek MS, Laughlin MH. Exercise training does not reduce hyperlipidemia in pigs fed a high-fat diet. Metabolism
. 2002; 51 (12): 1587–95.
43. Tsutsui JM, Xie F, Cano M, et al.. Detection of retained microbubbles in carotid arteries with real-time low mechanical index imaging in the setting of endothelial dysfunction. J Am Coll Cardiol
. 2004; 44 (5): 1036–46.
44. Turk JR, Henderson KK, Vanvickle GD, Watkins J, Laughlin MH. Arterial endothelial function in a porcine model of early stage atherosclerotic vascular disease. Int J Exp Pathol
. 2005; 86 (5): 335–45.
45. Walsh JH, Bilsborough W, Maiorana A, et al.. Exercise training improves conduit vessel function in patients with coronary artery disease. J Appl Physiol
. 2003; 95 (1): 20–5.
46. Wang N, Miao H, Li YS, et al.. Shear stress regulation of Kruppel-like factor 2 expression is flow pattern-specific. Biochem Biophys Res Commun
. 2006; 341 (4): 1244–51.
47. Wilson SH, Caplice NM, Simari RD, Holmes DR Jr, Carlson PJ, Lerman A. Activated nuclear factor-kappaB is present in the coronary vasculature in experimental hypercholesterolemia. Atherosclerosis
. 2000; 148 (1): 23–30.
48. Wilund KR. Is the anti-inflammatory effect of regular exercise responsible for reduced cardiovascular disease? Clin Sci (Lond)
. 2007; 112 (11): 543–55.
49. Woodman CR, Turk JR, Williams DP, Laughlin MH. Exercise training preserves endothelium-dependent relaxation in brachial arteries from hyperlipidemic pigs. J Appl Physiol. 2003; 94 (5): 2017–26.
50. Yang AL, Chen HI. Chronic exercise reduces adhesion molecules/iNOS expression and partially reverses vascular responsiveness in hypercholesterolemic rabbit aortae. Atherosclerosis
. 2003; 169 (1): 11–7.
51. Yang AL, Jen CJ, Chen HI. Effects of high-cholesterol diet and parallel exercise training on the vascular function of rabbit aortas: a time course study. J Appl Physiol
. 2003; 95 (3): 1194–200.