Augmented cell cycle protein expression and kinase activity in atherosclerotic rabbit vessels

Abstract

Cell proliferation within a primary atherosclerotic plaque is controversial. Identifying changes in cell cycle protein expression and the activities of their related kinases would provide valuable evidence of mitotic activity in the atherosclerotic lesion. Oxidized low-density lipoprotein has been shown to induce a significant increase in the total number of rabbit vascular smooth muscle cells in culture. In the present study, whole aortic cell extracts were harvested from rabbits fed a cholesterol-supplemented diet for eight weeks to induce modest plaque development, or 16 weeks to induce later, more severe plaque progression. Expression levels of cyclin A, cyclin-dependent kinase 4 (Cdk 4) and proliferating cell nuclear antigen were measured, as well as the activities of Cdk 4, Cdk 2 and Cdk 1. At both time points, the expression levels of cyclin A, Cdk 4 and proliferating cell nuclear antigen were significantly elevated. The activity of all three Cdks was also increased. There were no significant differences between moderate and more severe atherosclerosis. Surprisingly, tissues that neighboured the plaques, but did not show visible plaque formation on the vessel surface, also had significantly elevated cyclin A expression levels, but not as high as in the plaque areas. In conclusion, the primary atherosclerotic plaque exhibited elevated mitotic activity as shown by increased expression levels and activities of several cell cycle proteins. Expression levels were similar during moderate and severe atherosclerosis, and were even detected in nonatherosclerotic vascular tissue bordering the plaque.

Vascular cell proliferation is purported to be a critical event in the development and progression of atherosclerosis (1). Cell proliferation is a highly controlled process in any tissue and must be accompanied by increased expression and activation of cell cycle proteins (2). These proteins comprise the machinery, controlling movement through the cell cycle (and hence, proliferation) (3). Cyclins bind to cyclin-dependent kinases (Cdks) to form active complexes (4). Active Cdks phosphorylate key regulatory proteins, allowing passage from one phase of the cell cycle to the next (5). Movement from a quiescent G₀ state into the first phase of the cell cycle (G₁) requires cyclin D1/Cdk 4 complexes and Cdk 6 activation (3). Movement into the S phase (in which DNA synthesis occurs) requires cyclin E/Cdk 2 and cyclin A/Cdk 2 complexes, as well as the DNA polymerase cofactor – proliferating cell nuclear antigen (PCNA) (6). Finally, passage through the G₂ and M phases requires cyclin A/Cdk 1 and cyclin B1/Cdk 1 complexes (7). This induction of the cell cycle is tempered by the action of cell cycle inhibitory proteins such as p53, p21 and p27 (2).

Numerous studies (8–12) have demonstrated that these cell cycle proteins are induced in the balloon-injured vessels of animal restenosis models. However, the information regarding cell proliferation in restenosis conditions is not necessarily applicable to a primary atherosclerosis situation. To date, there have been few comparable studies focused on identifying cell proliferation in primary atherosclerosis situations; no studies focused on the response of cell cycle proteins in a primary atherosclerosis model. Several studies (13–15) documented increased thymidine labelling in plaques from atherosclerotic animals. This suggested that the plaque region was rich in mitotic activity. More recently, Orekhov et al (16) found a significant increase in PCNA-positive cells in lipid-rich atherosclerotic plaques and concluded that cell proliferation is stimulated in atherosclerosis. However, Marek et al (17) and Pickering et al (18) found little evidence of cell proliferative activity using PCNA as a marker of proliferation in the atheroma. Furthermore, other studies (19) have supported the hypothesis that decreased cell death through apoptosis, rather than increased cell proliferation, may be responsible for the growth of an atherosclerotic plaque. A more comprehensive examination of the expression of several different cell cycle proteins would support the evidence that cell proliferation is increased in a primary atherosclerotic plaque. An augmentation in cell cycle-dependent kinase-related activity would also be significant supportive evidence of mitotic activity in the plaque.

The purpose of the present study, therefore, was to determine whether alterations in the expression of cell cycle proteins and the activity of their related kinases occur in a model of primary atherosclerosis. In addition, it was hypothesized that cell cycle protein expression is more accelerated during moderate stages of plaque development and may be less dominant during the later, more severe, stages of plaque formation in an animal model of atherosclerosis.

METHODS

Animal model of atherosclerosis and plaque quantification

Male New Zealand white rabbits were maintained on either regular rabbit chow (control) or a 0.5% cholesterol diet (treated) for eight or 16 weeks. Aortas were harvested, cleaned, cut open lengthwise and into sections, and photographed using a Nikon Coolpix 995 camera (Nikon Canada). Photographs were digitized, and the amount of plaque was quantified using Imagespace software, version 3.2.1 (Molecular Dynamics, USA). The present study conforms with the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health (publication No. 85-23, revised 1996).

Preparation of tissue samples

Approximately 0.4 g wet weight of rabbit aortic tissue was finely chopped and added to a tube containing 1 mL of modified RIPA buffer (50 mM Tris-HCl [pH 7.4], 1% NP-40, 150 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 μM aprotinin, 1 μM leupeptin and 1 μM pepstatin). Homogenization was performed with a Polytron Homogenizer (Capitol Scientific, USA) for 1 min. The homogenate was ultracentrifuged for 30 min at 100,000 g and the supernatant was removed. The pellet was further subjected to nuclear protein extraction using a nuclear extraction reagent purchased from Thermo Fisher Scientific, USA; 250 μL of the nuclear extraction reagent was added to each tube. The tube was vortexed for 15 s to resuspend the pellet, then placed on ice for 10 min. The tube was vortexed again and returned to ice, and the process was repeated every 10 min for a total of 40 min. The tube was then centrifuged at 16,000 g in a microfuge for 10 min. The supernatant was removed and added to the original supernatant. The combined supernatants were mixed thoroughly and assayed for protein concentration using the modified Lowry assay (20). All samples were kept on ice throughout the experiments, and all centrifugations were performed at 4˚C.

Western blot analysis

For each sample, 50 μg of total protein was fractionated by sodium dodecyl sulphate polyacrylamide gradient gel electrophoresis for 4 h at 550 mV, 80 mA (constant current). Gels were calibrated using prestained molecular weight markers (Invitrogen Corporation, USA). Transfer onto nitrocellulose membrane was performed using a BioRad (Bio-Rad Laboratories, USA) apparatus for 75 min at 50 V (constant voltage). Following completion of the transfer, the membrane was placed in blocking buffer (a solution of wash buffer [10 mM Tris-HCl (pH 7.5), 100 mM NaCl, 0.1% Tween 20] plus 10% skim milk powder) for 1 h at room temperature. Antibody treatments for PCNA (Sigma-Aldrich Canada), cyclin A (Abcam, USA) and Cdk 4 (BD Transduction Laboratories, USA) were performed according to the manufacturer’s instructions. The membranes were washed five times in wash buffer, and primary antibody reactions were detected using horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescent detection reagents (Thermo Fisher Scientific, USA) according to the manufacturer’s instructions. Densitometry was performed on a Bio-Rad GS-670 Imaging Densitometer (Bio-Rad Laboratories, USA). Extensive preliminary analyses were conducted with many commercially available antibodies to a full compliment of cell cycle proteins. However, these were the only antibodies that reacted with rabbit tissue.

Kinase assay

Immunoprecipitation of Cdk 1, Cdk 2 and Cdk 4 was performed as described using antibodies from BD Transduction Laboratories (2). The immunoprecipitation reaction was performed overnight at 4°C. The next day, 20 μL of 50% protein G agarose beads (Calbiochem, EMD Chemicals Group, Germany) were added, and ultimately resuspended in kinase reaction buffer plus 0.2 μCi/μL [γ-³²P]ATP and the kinase substrate (0.2 μg/μL histone H1 [Gibco] for Cdc 2 and Cdk 2; 0.01 μg/μL glutathione S-transferase retinoblastoma protein pRb [Santa Cruz Biotechnology Inc, USA] for Cdk 4). Further kinase assay study details are published elsewhere (2). Samples were loaded onto a 10% gel and separated by sodium dodecyl sulphate polyacrylamide gel electrophoresis. The gel was stained with Coomassie blue to confirm equal amounts of kinase substrate in each sample, and then was destained and dried. Phosphorylated substrate was visualized by autoradiography and quantitated by densitometry.

Data analysis

Data are reported as mean ± SEM. Results were analyzed by one-way ANOVA (control versus treatments) followed by a Dunnett’s post hoc test. The statistics were analyzed using the SigmaStat (Systat Software Inc, USA) program, with P<0.05 considered to be statistically significant.

RESULTS

A high-cholesterol diet was used for two different durations to induce a moderate and a more severe atherosclerotic plaque formation in the aortas of rabbits. Atherosclerotic plaque formation was assessed through analysis of digital photos of the aortas (Figure 1). A total of 59 aortas were examined. Animals fed a 0.5% cholesterol diet for eight or 16 weeks developed significant amounts of plaque in their aortas compared with animals fed a regular chow diet. Eight weeks of cholesterol feeding resulted in plaque coverage of 36.2% of the aorta, while 16 weeks of feeding resulted in 78.8% plaque coverage (Figure 1). None of the animals in the control group developed plaque in their aortas. These data and the intrasubject variation are consistent with observations from previous studies (21,22). To evaluate the status of cell proliferation during the development of plaque, total protein lysates of aortic tissue from each treatment group were analyzed by Western blot for the expression of cell cycle proteins. Levels of PCNA were significantly increased (by 43.9% after eight weeks and by 52.0% after 16 weeks of dietary intervention) in the aortas of rabbits fed the cholesterol diet versus controls (Figure 2). There was no significant difference between PCNA levels in the eight-week aortas compared with the 16-week aortas. Similarly, expression of Cdk 4 and cyclin A were also significantly increased in the aortas of cholesterol-fed rabbits. Cdk 4 levels were increased by 18.9% in eight-week aortas and 26.0% in 16-week aortas, while cyclin A expression was increased by 57.4% at eight weeks and by 76.7% at 16 weeks (Figure 2). There were no significant differences between expression levels at eight weeks and 16 weeks for either Cdk 4 or cyclin A.

Figure 1).

Sections of rabbit aortas showing atherosclerotic plaque build-up due to cholesterol feeding. From top to bottom: control, eight weeks and 16 weeks. Bar graph shows comparison of the amount of plaque in the aortas (n = at least 12 aortas per group), expressed as a percentage of the total area ± SEM. *P<0.05 versus control; ^†P<0.05 versus eight weeks

Figure 2).

Representative autoradiograms from Western blots showing expression of proliferating cell nuclear antigen (PCNA), cyclin-dependent kinase 4 (Cdk 4) and cyclin A in rabbit aortas from control animals, and eight- and 16-week cholesterol-fed animals, respectively. Protein expression was analyzed by densitometry and reported as a percentage of control ± SEM. *P<0.05 versus control (n=6)

Increased expression of cell cycle proteins suggests cell proliferation. However, to confirm this, the activities of Cdks were examined. The activities of Cdk 4, Cdk 2 and Cdk 1 were assessed in protein extracts from aortas of control and cholesterol-fed rabbits (Figure 3). Cdk 4 activity was significantly increased in both eight- and 16-week aortas (26.8% and 24.4%, respectively) relative to controls. The activity of Cdk 2 was elevated by 40.2% in eight-week aortas, and 16.4% in 16-week aortas. Finally, Cdk 1 activity increased by 51.7% at eight weeks and 58.7% at 16 weeks of cholesterol feeding compared with aortas from control animals. Levels of activity did not differ significantly between eight and 16 weeks in the aortas of cholesterol-fed rabbits for any of the Cdks studied.

Figure 3).

Representative autoradiograms from kinase assays showing activity of cyclin-dependent kinase 4 (Cdk 4), Cdk 2 and Cdk 1 in rabbit aortas from control animals, and eight- and 16-week cholesterol-fed animals, respectively. Activity was analyzed by densitometry of autoradiograms (n=4 for each kinase). Activity is reported as a percentage of control ± SEM. *P<0.05 versus control

The amount of plaque in the aortas of rabbits fed cholesterol for eight and 16 weeks differed significantly (Figure 1). However, the entire aorta was used for expression analysis. Thus, the samples from the animals fed the diet for eight weeks contained less plaque than the vessels obtained from the animals fed the diet for 16 weeks. The results, therefore, could be viewed as biased and an unfair comparison of expression within the plaque at the two stages of atherosclerosis. It may be a more appropriate comparison to examine expression levels of cell cycle proteins specifically from plaque regions in both groups to obtain a better, unbiased analysis of cell proliferation in moderate and more severe plaque development, without the confounding influence of the vessel that did not contain any visible plaque. Plaque tissue was carefully dissected away from areas without plaque. This was particularly time consuming and difficult, and generated relatively little tissue in the eight-week aortic samples, in which approximately 65% of the vessel wall did not contain visible evidence of plaque formation. Because the 16-week samples contained even less plaque-containing tissue (approximately 20%), expression analysis could not be completed on nonplaque tissue in this group. The plaque and nonplaque areas of these aortas were homogenized separately and total protein from each was used for Western blots. Expression levels of PCNA, Cdk 4 and cyclin A were evaluated in these samples. Levels of cyclin A were increased by 71.5% in the plaque portions of eight-week aortas versus control aortas (Figure 4). As expected, cyclin A expression was not stimulated as much in the nonplaque regions. However, interestingly, the levels of cyclin A in nonplaque areas of eight-week aortas were significantly increased by 36.4% versus control aortas. By comparison, cyclin A expression in whole eight-week aortas was 57.4% greater than the control aortas. Similar qualitative effects were observed for the other cell cycle proteins. For example, in the plaque areas, PCNA expression was elevated by 40.7% versus controls (P<0.05) and, in the nonplaque areas, PCNA expression was elevated by 25.9% (P>0.05). Levels of Cdk 4 were also significantly increased by 56.5% versus controls in plaques found within eight-week aortas (P<0.05). Nonplaque areas had Cdk 4 levels that were 23.1% greater than in controls (P>0.05).

Figure 4).

Analysis of Western blot autoradiograms by densitometry showing cyclin A, proliferating cell nuclear antigen (PCNA) and cyclin-dependent kinase 4 (Cdk 4) expression in rabbit aortas from control animals, and eight-week plaque and eight-week nonplaque areas. Expression is shown as a percentage of control ± SEM. *P<0.05 versus control (n=4)

DISCUSSION

Proliferation of cells within the vascular wall has long been assumed to be an important component of the atherosclerotic process (23,24). Cell proliferation necessarily involves an upregulation of the components of the cell cycle machinery. The present investigation demonstrated, for the first time, that both the expression of cell cycle proteins (Figure 2) and the activity of their related kinases (Figure 3) are elevated in the aortas of cholesterol-fed rabbits. This would strongly support the hypothesis that cell proliferation is stimulated in the vessel wall during atherosclerosis. In turn, this conclusion supports the research of Orekhov et al (16) but is in opposition to the work of Marek et al (17) and Pickering et al (18). Previous studies have used PCNA as the sole marker of cell proliferation. Using PCNA as the only marker of cell proliferation has limitations that may lead to false-positive results (17). We used several different cell cycle proteins that are induced at different points within the cell cycle as markers of cell cycle progression. This avoids the concern that false-positive results may be obtained from cells entering and exiting the cell cycle at different times. It remains uncertain how the cell cycle inhibitory proteins reacted to the atherogenic environment. Cell cycle progression is a combined result of opposing actions from the cell cycle-inducing and inhibitory proteins (2). Without availability of antibodies to any of the inhibitory cell cycle proteins that will be functional in the rabbit preparations, we could not assess this possibility. However, the measurement of kinase activity provides further evidence in support of cell cycle stimulation and avoids any fixation or antibody artifacts associated with PCNA (17) that may have produced inaccurate results. The activities of all of these cell cycle kinases were elevated within the atherosclerotic vessels. This strongly supports the contention that cell proliferation is stimulated within the atherosclerotic vessel wall.

In the present study, we used cholesterol-fed New Zealand white rabbits as models for primary atherosclerosis. When maintained on a 0.5% cholesterol diet, these rabbits consistently developed plaques in their aortas. The duration of the dietary intervention was used to manipulate the severity of the atheroma. Eight weeks on the 0.5% cholesterol-supplemented diet resulted in moderate plaque development, and a 16-week dietary intervention induced more severe plaque formation. Our initial results using the entire vessel to obtain cell extracts demonstrated that the expression of cell cycle proteins was similar during the moderate and more severe stages of atherogenesis – results that were surprising. We had expected to detect augmented cell proliferation during the moderate stages of atherosclerosis. However, it was possible that the results in the eight-week samples were biased by a greater ‘contamination’ of the vessels with tissue that did not contain any plaque. If expression was normal in the area of the vessel that was not infiltrated by the plaque, this would artificially produce lower expression values. Therefore, areas of visible plaque in eight-week aortas were separated from nonplaque areas and analyzed independently. Expression of the cell cycle proteins was consistently greater in the plaque than in the neighbouring sections of the vessel that did not contain plaque. It was, however, similar in the plaque-containing samples from the eight-week vessels to that found in the later-stage atherosclerotic plaques in aortas from rabbits that underwent the 16-week dietary intervention. This suggests that cell cycle protein expression remains relatively constant from moderate to more severe stages of atherosclerosis. Interestingly, nonplaque areas of moderately atherosclerotic aortas also expressed higher levels of cell cycle protein than control samples. It appears that these ‘preatherosclerotic’ areas may be initiating cell proliferation as a first step in plaque generation.

The mechanism responsible for the augmentation of cell proliferation during atherosclerosis is likely complex and multifactorial. Vascular cell proliferation has been shown to be induced by basic fibroblast growth factor (25), acidic fibroblast growth factors (26), insulin-like growth factor (27), vascular endothelial growth factor (28), native low-density lipoprotein (2), oxidized low-density lipoprotein (2,29), adrenomedullin (30), nitric oxide (25), ATP and ADP (31), phospholipase D (32), lysophosphatidylcholine (29), mitogen-activated protein kinase activation (33,34) and phosphatidylinositol 3-kinase activation (35), to identify just a few. Therefore, the mechanisms responsible for inducing cell cycle protein expression and the related kinase activity in the present study may involve any one or more of these factors. For example, oxidized low-density lipoprotein can increase the expression and activation of cell cycle proteins in vascular cells (2). The status of the cell within the cell cycle may also influence the response of cell cycle proteins to mitogens such as oxidized low-density lipoprotein (36). The precise mechanism that induces cell cycle expression in vivo in atherosclerosis may prove to be exceedingly difficult to isolate and clearly identify. However, this should not detract from its significance. Strategies that target specific cell cycle proteins have proven valuable in restenosis (37) and may be worthy of study in primary atherosclerosis now that we are certain that they are altered in the atherosclerotic lesion.

The present investigation provided the first evidence that the expression and activity of several cell cycle proteins are elevated in atherosclerotic lesions of cholesterol-fed rabbits. These data strengthen the contention that cell proliferation is augmented during plaque initiation and development. The present investigation also demonstrated that the augmentation of cell proliferation was independent of the stage of the atherosclerotic plaque. Interestingly, even the area of the atherosclerotic vessel that did not contain visible plaque formation appeared to be in the process of initiating cell proliferation. Our results provide important new insights into the involvement of cell cycle proteins and cell proliferation in the process of primary atherogenesis.