Graduate

Social environment and the progression of atherosclerosis in the WHHL

Several years ago, we published the first demonstration that social environment can influence the progression of atherosclerosis in the WHHL model (McCabe et al., 2002). WHHL rabbits were assigned to one of three social/behavioral groups: an Unstable group, in which unfamiliar rabbits were paired daily, with the pairing switched each week, a Stable group, in which littermates were paired daily for the entire study, and an Individually-Caged group. The Stable group exhibited more affiliative social behavior and less agonistic behavior than the Unstable group, and significantly less aortic atherosclerosis than each of the other two groups. Although the Unstable and Individually-Caged groups had comparable aortic lesion areas, the severity of the disease progressed faster in the Unstable group, as indexed by a larger area of calcification and increased fibrous cap thickness in complex lesions (see Figure 1). Lesions in the Individually-Caged group consisted primarily of foams cells (i.e., grade 3). The Unstable group showed increased agonistic behavior and signs of chronic adrenocortical and gonadal activation, whereas the Individually-Caged group was relatively sedentary, had low glucocorticoid levels, and was hyperinsulinemic compared to the other groups. The study demonstrated that social environment can slow, as well as accelerate, the progression of atherosclerosis. It also emphasized the importance of behavioral factors in atherogenesis, even in a model of disease with strong genetic determinants.

Figure 1. Examples of atherosclerotic lesions in WHHL rabbits. Left photo, grade 3 lesion consisting primarily of foam cell infiltration into the vessel wall. Right photo, more complex grade 5a lesion containing foam cells, necrosis, and formation of a fibrous cap.

Glucocorticoids in the WHHL

The prior study suggested that social stress plays an important role in the progression of atherosclerosis in the WHHL model. One major effector in the stress response is the hypothalamic-pituitary-adrenocortical (HPA) system, which is likely to be involved in the behavioral modulation of disease. Because relatively little is known about the glucocorticoid responses in rabbits, we conducted a study (Szeto et al., 2004) to: 1) evaluate the rabbit glucocorticoid circadian rhythm, 2) compare plasma cortisol and corticosterone responses to social stress, and 3) examine strain differences (i.e., WHHL vs. New Zealand White, NZW) in rabbit glucocorticoid responses to assess whether WHHLs have an aberrant HPA system. It was found that male rabbits secrete both corticosterone and cortisol in a circadian rhythm that peaks in the afternoon and reaches a nadir at 0600h, i.e., approximately 12h out-of-phase with the human glucocorticoid rhythm. Both glucocorticoids responded similarly to social stress induced by repeated daily 4h pairings with another male rabbit; after 10 days of pairings, glucorticoid values were significantly correlated with the amount of defensive agonistic behavior exhibited. Finally, there were no significant strain differences in glucocorticoid circadian rhythms, baselines, or responses to social stress. These data suggest that glucocorticoid responses (e.g., circadian rhythms, responses to social stress) in the WHHL are similar to glucocorticoid responses in standard laboratory rabbits.

Insulin metabolic variables and behavioral interventions in the WHHL

In our first WHHL study, the fact that individually-caged WHHLs exhibited increased body weight, behavioral inactivity, hyperinsulinemia, and elevated heart rate suggested that insulin metabolic variables play a role in the progression of atherosclerosis in WHHL rabbits. A recently published study from our lab (Gonzales et al., 2005) sought to determine: a) if young, individually-caged WHHLs are insulin resistant relative to NZW rabbits and b) whether dietary or exercise interventions can improve insulin sensitivity and slow the development of atherosclerosis in these animals. Forty-two WHHLs were assigned to either a dietary, exercise, or control condition, and 12 NZWs were used as a comparison control group. The intervention ran from 3 to 7 months of age, and all animals received an intravenous glucose tolerance test at the beginning and end of the intervention. It was found that WHHLs were insulin resistant relative to NZWs at 3 months of age. Whereas the dietary intervention was effective in controlling insulin resistance, WHHLs in the exercise group without dietary restriction and the control group exhibited significant increases in insulin resistance. Despite the significant changes in insulin sensitivity, none of the interventions significantly influenced the progression of atherosclerosis. It was concluded that young WHHLs are insulin resistant during an early period when atherosclerosis is developing rapidly. Dietary restriction, but not exercise without weight control, is effective in controlling insulin metabolic variables in the WHHL model. Although dietary intervention can reduce cardiovascular risk factors such as insulin resistance, it is not effective in slowing the development of atherosclerosis in these genetically dyslipidemic animals. This may have been due, in part, to the fact that dietary restriction significantly increased plasma cholesterol relative to the other groups. Exercise training, without dietary control, also did not influence the progression of disease in WHHLs.

Hyperlipidemia and oxidized LDL in the progression of atherosclerosis

Hyperlipidemia, and particularly LDL, has been shown to be a major risk factor for atherosclerosis. Given that WHHL rabbits are extremely dyslipidemic, we sought to examine the relationship of individual differences in plasma lipids and the extent of disease in these animals. Although native-LDL can stimulate atherogenesis, it has been reported that oxidative modification of LDL can trigger a variety of proatherogenic effects and lead to rapid disease progression. Therefore, we measured plasma ox-LDL, in addition to plasma cholesterol and triglycerides, from fasted samples at 3, 5, and 7 months of age in 73 WHHLs. It was found that total cholesterol measured at 3, 5 and 7 months is a significant predictor of the extent of atherosclerosis in the aortic arch measured at 7 months (r = 0.38, 0.45, 0.42, respectively; p < .01). Similarly, cholesterol measured at 5 and 7 months was significantly related to disease in the abdominal aorta (r = 0.29, 0.34, respectively; p< .25). ox-LDL measured at 5 and 7 months of age also significantly predicted atherosclerosis in the aortic arch (r = 0.39, 0.38, respectively; p<.01) and the abdominal aorta (r = 0.38, 0.33; p<.01). The relationship between ox-LDL and disease in the aortic arch was particularly strong in a subset of these WHHLs that were individually-caged (n = 23; 5 month ox-LDL, r = 0.72; 7 month ox-LDL, r = 0.69; p<.001). These findings are interesting in light of recent evidence that behavioral inactivity stimulates oxidative stress and atherosclerosis. As reported earlier, individually-caged WHHLs are sedentary relative to animals in other social conditions. These data emphasize the importance of hyperlipidemia in the mechanisms of disease in this model, and that these variables may account for a significant portion of the individual variability in disease. In addition, the study suggests that oxidative stress may play an important role in the progression of disease, especially in sedentary, individually-caged animals.

C-reactive protein and inflammation in the WHHL

Plasma CRP may be an indicator of coronary heart disease risk in humans, and recent evidence suggests that CRP itself is proatherogenic. Although the WHHL is an important animal model of human familial hypercholesterolemia and atherosclerosis, there are no reports in the literature of basal CRP detection or its relationship to atherosclerosis in this model. Therefore, in a recent study we sought to: a) describe the measurement of baseline plasma CRP in the rabbit (56 WHHL and 11 NZW) utilizing a newly available chicken-anti rabbit primary antibody for CRP, and b) examine the relationship of CRP sampled early in the disease process and the extent of atherosclerosis measured at seven months of age. It was found that plasma CRP was significantly greater in WHHLs than NZWs at 5 months of age, and that CRP levels sampled in WHHLs at 3 and 5 months of age were significantly correlated with the total area of atherosclerosis in the aortic arch measured at 7 months of age (r = 0.40, 0.35, respectively, <.01). This is important because the initial site of disease in the WHHL is in the aortic arch. CRP values measured at 5 months of age were also correlated with total area of aortic atherosclerosis measured at 7 months of age (r = 0.30, p<.05). This represents the first report of CRP measurement in the WHHL, and these results suggest that CRP, a sensitive circulating marker for inflammation, predicts the extent of atherosclerosis in this model.

CNS and plasma oxytocin, social environment and atherosclerosis in the WHHL

In our initial WHHL study we reported that WHHLs in a stable social environment, which is associated with increased affiliative behavior and less agonistic behavior, exhibited significantly less atherosclerosis relative to WHHLs in an unstable environment or individually-caged WHHLs. It has been suggested that affiliative social behavior is influenced by the release of OT in the hypothalamus and limbic system of the brain. Furthermore, it has been proposed that this nonapeptide regulates the release of brain CRH, thereby modulating the HPA and sympathetic axes of the body's stress response. Activation or inhibition of these response systems may be an important mechanism linking social behavior and the development of atherosclerosis. To address this hypothesis, we have used chronic microdialysis to measure the release of OT from the PVN of WHHLs in the three social conditions used in our initial study (Paredes et al., 2006). Under anesthesia and using aseptic surgical techniques, 34 WHHLs were chronically implanted with a guide cannula terminating in PVN. Following recovery from surgery, a microdialysis probe was inserted through the guide cannula into PVN and the region was perfused with artificial cerebrospinal fluid and dialysate samples were taken. Following the baseline measures, animals were placed in either an unstable, stable, or individually-caged social condition for one hour, and their behavior was videotaped and scored throughout the pairing. After the behavioral manipulation, the animals were separated and the microdialysis procedure and blood draw was repeated. Following this first day, the WHHLs were placed in the same social condition daily for three weeks and the microdialysis procedure was repeated at the end of the study (when the animals were 4.5 months old).

As we observed in our prior study, WHHLs in the Stable group exhibited significantly more affiliative, and less agonistic, behavior than the Unstable animals over the course of the study. Individually-caged rabbits spent significantly more time behaviorally inactive than the other groups. Importantly, the Stable group exhibited significantly less atherosclerosis in the aortic arch than the other groups (see Figure 2A). This finding replicates the results from our initial study, and it is interesting that this effect of social environment on disease is apparent at such an early time point (i.e., 4.5 months of age), and after only three weeks of social pairing. There was a general trend for plasma OT to be elevated in the Stable group relative to the other social groups (Figure 2B). It is also interesting that day 1 plasma OT for all animals was inversely correlated to the extent of abdominal aortic atherosclerosis measured 4 weeks later (r = -0.55, p<.002).

Figure 2. Panel A, extent of total atherosclerosis and aortic arch atherosclerosis as a function of social environment at 4.5 months of age. Panel B, plasma oxytocin at 4.5 months as a function of social environment.

There were no significant group differences in PVN OT responses to social pairing on day 1, however, by the end of the study (day 22) the Unstable group exhibited significant increases in PVN OT relative to the other groups (Figure 3A). Although this was not the hypothesized result, it does support the notion in the literature that CNS oxytocin is released as a response to stress, and may help to buffer further stress responses. In support of this idea, plasma catecholamine responses, which were greatest in the Unstable group, were significantly attenuated in this group by day 22 (Figure 3B). There were no group differences in plasma glucocorticoids. Day 1 PVN OT responses for all animals were positively correlated with disease in the aortic arch (r = 0.45, p<.05). The apparent inverse relationship of central and peripheral OT is not surprising, since this has been reported in the literature. Given the recent finding that OT receptors are located on aortic vascular cells (including endothelial cells) and the heart, we will examine the possibility that changes in plasma OT may directly affect vascular cells and influence the progression of atherosclerosis.

Figure 3. Panel A (top), PVN oxytocin responses as a function of social environment over 22 days. Panel B (bottom), plasma catecholamine responses as a function of social environment over 22 days.

Vascular NAD(P)H oxidase, angiotensin receptors, and inflammation as a function of social environment

It has been demonstrated that vascular oxidative stress plays an important role in the progression of atherosclerosis in the WHHL model. Increased vascular NAD(P)H oxidase-mediated superoxide production appears to be stimulated by activation of AT1 receptors on vascular cells by local or circulating angiotensin. Antagonism of angiotensin converting enzyme, which blocks angiotensin II production, reduces atherosclerosis in WHHLs by up to 40%. Similarly, administration of the powerful antioxidant, probucol, significantly reduces disease in this model. Therefore, we recently undertook a study to examine the influence of social environment on vascular NAD(P)H oxidase activity and AT1 receptor binding in the WHHL (Nation et al., submitted) In addition, we also used the recently-developed rabbit CRP assay to measure plasma levels of this inflammatory marker in these animals. WHHLs were assigned to Unstable, Stable or Individually-caged groups and blood was drawn every 4 weeks for measurement of lipids and CRP. At 5 months of age, the animals were sacrificed, the aortas removed and sectioned, and the fresh tissue was homogenized for biochemical analysis. It was found that vascular NADPH oxidase activity in the aortic arch was greatest for the Individually-caged group (Figure 4A). In contrast, CRP values were lowest for the Individually-caged animals, but were elevated in the Unstable group (Figure 4B). Taken together, these preliminary data suggest that early in the disease process, atherosclerosis may be progressing through different (albeit overlapping) mechanisms. Specifically, oxidative stress may be particularly important in the development of disease in the Individually-caged group, whereas inflammation, as indexed by CRP, remains low at this early time point. These animals are behaviorally sedentary, and recently it has been shown that behavioral inactivity increases oxidative stress, endothelial dysfunction and atherosclerosis. The elevated levels of CRP in the Unstable group suggest that inflammatory processes may be exacerbated in these animals, perhaps via chronic sympathetic nervous system activation. Stable animals exhibit moderate inflammation and oxidative stress relative to the other groups.

Figure 4. Panel A (top), NADPH oxidase activity in the aortic arch at 5 months of age as a function of social environment. Panel B (bottom), plasma CRP at 5 months of age as a function of social environment.

The influence of oxytocin on NAD(P)H oxidase and CAMs in cultured endothelial cells

Given our findings that plasma OT is elevated in Stable WHHLs and the plasma OT inversely predicts future disease in all animals, we have begun to examine the influence of OT on vascular cells in vitro. In preliminary studies (Szeto et al., submitted) using cultured human aortic endothelial cells, we measured endothelial NAD(P)H oxidase activity and cell adhesion molecule (CAM) expression following incubation with OT. Initially, we confirmed that these cells express oxytocin receptors with a western blot (Figure 5A), and then we established that OT administration significantly reduced the TNF-alpha-stimulated vCAM and iCAM expression in these cells (Figure 6A and B, and Figure 5B). Next, we established that NAD(P)H oxidase activity could be increased with administration of TNF-alpha and PMA (Figure 7A), but following 16 hours of incubation with OT, NADPH oxidase activity could be significantly reduced (Figure 7B). These data suggest that OT has antioxidant (i.e., NAD(P)H oxidase inhibition) and antiflammatory (i.e., suppression of vCAM and iCAM expression) actions on vascular endothelial cells. For the proposed work, we will use these cultured human endothelial cells, in addition to cultured WHHL endothelial cells, smooth muscle cells, and macrophages to examine the influence of OT and other substances (e.g., catecholamines, LDL, insulin, cytokines, CRP) on oxidative stress and inflammatory mechanisms in vitro.

Figure 5. (A) Western blot analysis of oxytocin receptor (OTR) expression in cultured human endothelial cells, and (B) the effect of oxytocin ICAM-1 in TNF-α-stimulated endothelial cells.

Figure 6. Expression (Western Blot) of VCAM-1 (A) and ICAM-1 (B) in cells. Cells were incubated with oxytocin (1 µM) for 16 hours, then treated with TNF-α (1 ng/ml) for 1 hour.

Figure 7. Effect of TNF-α and PMA (A) and oxytocin (B) on NADPH oxidase activity in cultured human endothelial cells as determined with lucigninen-enhanced chemilumine