Weight loss is an effective, nonpharmacologic method to reduce blood pressure and improve glucose tolerance in overweight hypertensive individuals (Katzel et al. 1995). Importantly, weight reduction also lowers blood pressure in nonobese hypertensive subjects (Imai et al. 1986). These observations suggest an important link between energy balance and blood pressure. In fact, a relationship between energy intake and blood pressure was recognized by Brozek et al. (1948) as a result of the Minnesota Starvation Study and observations of victims who endured prolonged semistarvation during World War II.

The mechanisms responsible for reductions in blood pressure produced by weight loss have not been adequately described. It has been hypothesized that augmented sympathetic nervous system activity contributes to the production of obesity-related hypertension (Landsberg 1986 and 1994). Furthermore, there are several lines of evidence from human studies that suggest that decreased energy intake reduces sympathetic nervous system activity. Weight reduction has been associated with reduced norepinephrine appearance rate in plasma (O'Dea et al. 1982) and reduced levels of plasma catecholamines (Sowers et al. 1982). Consumption of a reduced energy diet for 4 mo reduced body weight, blood pressure and directly measured muscle sympathetic nerve activity in obese women (Andersson et al. 1991). Finally, high renin hypertensive subjects in the Trial of Antihypertensive Interventions and Management were very responsive to beta-blockers (rather than diuretics) and weight loss (Blaufox et al. 1992). These observations support the hypothesis that the effects of weight loss on blood pressure could be mediated via reductions in sympathetic nervous system activity.

Body weight and energy balance also influence the cardiovascular system of spontaneously hypertensive rats (SHR).⁶ Young et al. (1978) demonstrated that 4 d of food deprivation or energy restriction (50% of energy consumed by free eating controls) produced significant reductions in systolic blood pressure of SHR. More prolonged food restriction (1-4 mo) has also generally been associated with lower blood pressure in SHR (Fernandes et al. 1986, Herlihy et al. 1991, Notargiacomo and Fries 1981, Wright et al. 1981). However, some studies have not observed a blood pressure-lowering effect of food restriction in SHR (Gradin and Persson 1990, Susic et al. 1990). Surprisingly, none of the full reports have reported directly measured blood pressure from conscious, unrestrained animals. This is important because SHR are hyperresponsive to the restraint stress associated with indirect assessment of blood pressure (Chiueh and Kopin 1979). Thus, food-restricted animals could exhibit less responsiveness to the restraint stress associated with indirect blood pressure assessment, but might not have significantly reduced blood pressure measured during basal conditions. Therefore, the first objective of this study was to determine if food restriction reduces blood pressure measured under basal conditions in unrestrained SHR.

The relationship between energy intake and sympathetic nervous system activity is complex. This complexity may exist because sympathetic neural outflow to organs and tissues is not likely to be homogeneously regulated. Bray (1991) proposed that there is a reciprocal relationship between sympathetic activity and energy intake. The concept is not consistent with the hypothesis that reduced sympathetic nervous system activity is directly responsible for the reductions in blood pressure that may be produced by prolonged energy restriction in SHR. Clearly, regional studies of norepinephrine turnover have suggested that food deprivation reduces sympathetic activity in heart, liver and pancreas in SHR (Einhorn et al. 1982, Young et al. 1979). Similar studies are not available for SHR exposed to chronic, mild food restriction. Reductions in norepinephrine turnover in specific organs do not necessarily provide an indication of the contribution of sympathetic activity in a region to the regulation of blood pressure. Therefore, we chose to quantify the reduction in blood pressure following complete pharmacologic ganglionic blockade as an index of sympathetic neural support of blood pressure. Thus, an additional objective of the study was to test directly the hypothesis that prolonged food restriction would reduce development of hypertension in SHR through sympathetic neural mechanisms.

MATERIALS AND METHODS

Food consumption for the ADLIB group was measured over 3- to 4-d periods and used to calculate food amounts for the FR group. Rats were maintained on this dietary regimen for approximately 8 wk. During the final week of the treatment period, rats were acclimated to the cages to be used for cardiovascular testing (2-3 sessions on different days, 4-5 h per session). No food or water was available while the rats were in the testing cages. Rats received their normal feeding the night before surgery.

On the day after surgery, the rats were again brought to the laboratory where measurements of base-line HR and BP were taken 3-4 h after they were placed in testing chambers. Following this base-line period, cardiac baroreflex function was assessed by bolus administration of graded doses of phenylephrine hydrochloride (0.5, 1.0, 2.0 and 4.0 µg/kg; Sigma). Doses were administered in volumes of <100 µL using Hamilton syringes. The order and dose of drug administration were randomized. Finally, blood pressures were obtained from some of the rats 7 d after cannulation.

Data were analyzed via ANOVA, linear regression and Student's t test procedures (GB-Stat, Version 1.0, Dynamic Microsystems, Silver Spring, MD). ANOVA (2 × 3) was used to evaluate blood pressure, heart rate and body weight data. Significant differences between individual means were determined using Tukey's post-hoc analysis (Bruning and Kintz 1977). Student's t test was used to analyze decrements in BP produced by ganglionic blockade. ANOVA (2 × 4) was used to evaluate pressor responses to phenylephrine. Linear regression analysis was used to evaluate cardiac baroreflex function. Statistical significance was set at P < 0.05. 

RESULTS

Table 1. Blood pressure and heart rate values of conscious, unrestrained, spontaneously hypertensive rats (SHR) given free access to food (ADLIB) or restricted access to food (FR) for 8 wk1,2 [View Table]

Because of catheter failures, cardiovascular data were obtained from eight rats in the ADLIB group and nine rats in the FR group. Of these, four ADLIB and five FR rats were studied 7 d after surgery. Rats in the FR group had significantly lower mean (20 mmHg), systolic (27 mmHg) and diastolic (15 mmHg) blood pressures recorded several hours after catheterization (Table 1). On the day of surgery, HR were not significantly different but were reasonably low, suggesting that the rats had recovered partially from surgical stress.

Mean, systolic and diastolic pressures were significantly lower in the FR group on both the day after surgery and 7 d after surgery (Table 1). The magnitude of the differences in BP between the FR and ADLIB groups was relatively consistent on the three measurement days. There was no significant effect of chronic food restriction on resting HR in SHR on any measurement day.

The BP response to administration of graded bolus doses of the -receptor agonist phenylephrine was not different in the ADLIB and FR groups (Fig. 1). Linear regression analysis of the reflex reduction in HR in response to elevations in pressure produced by phenylephrine is shown in Figure 2. Regression lines for food-restricted rats had a significantly greater slope relating the change in HR to the change in MAP than those for rats given free access to food.

The maximum depressor response to ganglionic blockade produced by intravenous administration of hexamethonium chloride (30 mg/kg) and atropine methyl nitrate (0.1 mg/kg) was significantly greater in the ADLIB group (Fig. 3). As a result, there was no effect of chronic food restriction on MAP after ganglionic blockade (ADLIB = 93 ± 6 mmHg; FR = 95 ± 4 mmHg).

DISCUSSION

The results are generally consistent with early reports demonstrating that food deprivation and short-term food restriction lower blood pressure in SHR (Einhorn et al. 1982, Young et al. 1978). Four days of starvation has been shown to be associated with reduced cardiac, hepatic and pancreatic norepinephrine turnover (Einhorn et al. 1982, Young and Landsberg 1979), suggesting that reduced sympathetic activity could produce the hypotensive effect of food deprivation. Subsequent studies have examined the influence of more prolonged and less severe food restriction on BP in SHR. The amount of food provided in these studies was 50-67% of the amount consumed by controls, which is generally consistent with the level of restriction that extends life span in laboratory rats (for review see Masoro 1985). A similar degree of restriction (to 60%) was chosen for the current study. For the most part, previous studies suggest that systolic BP measured indirectly in restrained rats (tail plethysmography) is also reduced by chronic food restriction (Fernandes et al. 1986, Notargiacomo and Fries 1981, Wright et al. 1981). However, there are exceptions to the observation that food restriction lowers BP in SHR (Gradin and Persson 1990, Susic et al. 1990). To date, these studies all used indirect methods to assess BP in restrained rats. Our findings from conscious, unrestrained rats are consistent with a preliminary report suggesting that food restriction lowers BP measured in conscious, unrestrained SHR (Herlihy et al. 1991). We obtained 30-min averages of BP acquired over three different experimental sessions. Thus, the findings provide strong support for the hypothesis that food restriction lowers resting BP in SHR.

Prolonged food restriction produces a wide array of actions on the cardiovascular system (for review see Herlihy and Thomas 1992). It is important to note that the hypotensive effects of food restriction are not limited to SHR. Swoap et al. (1995) demonstrated marked food restriction-induced reductions in blood pressure in rats made hypertensive either by combined nephrectomy-deoxycorticosterone acetate treatment or by abdominal aortic coarctation. Interestingly, similar degrees of food restriction seem to produce minimal effects on blood pressure in normotensive rats. This suggests that food restriction may produce greater effects on sympathetic support of blood pressure when it is elevated. This is the case in SHR (Leenen et al. 1994).

The effects of chronic food restriction on resting HR in rats is variable. Herlihy et al. (1992) reported lower resting HR in Fischer 344 rats that had consumed an energy-restricted diet for about a year. In contrast, the present results suggest that food restriction does not lower HR in SHR. Similarly, Swoap et al. (1995) did not observe an influence of food restriction on HR.

Chronic food restriction also increases the sensitivity of the cardiac baroreflex in both normotensive (Herlihy et al. 1991) and hypertensive rats (Herlihy et al. 1992). We chose to focus on the reflex bradycardic response to increases in pressure produced by bolus administration of phenylephrine. Because of the transient nature of the increase in MAP produced by bolus phenylephrine injections, the reflex bradycardia is mediated almost exclusively by increased vagal outflow (Coleman 1980). This would suggest that food restriction may influence parasympathetic function as well as sympathetic outflow. Normotensive rats exposed to 1 y of food restriction exhibited augmented increases in HR after atropine administration, suggesting augmented vagal outflow at rest (Herlihy and Thomas 1992). Furthermore, recent studies in humans using analysis of HR variability to assess autonomic tone indicated that 10% weight loss produces decreases in resting HR and sympathetic control of HR, as well as increased vagal tone regulating HR (Arronne et al. 1995). Thus, energy restriction appears to have substantial impact on the autonomic nervous system, leading to increases in vagal tone and decreases in sympathetic tone at rest. The outcomes of these adaptations appear to include augmented cardiac baroreflex sensitivity and reduced BP.

What are the signals activated by energy restriction that produce modulation of autonomic tone and reductions in BP? At present, we have no definitive experimental evidence to answer this question. Landsberg (1994) has recently reviewed the evidence linking hyperinsulinemia, sympathetic nervous system function and blood pressure. Thus, one possibility is that insulin levels in the food-restricted rats are reduced and may participate in lowering sympathetic tone. It should be noted that we did not assess plasma insulin concentrations in this study. Spontaneously hypertensive rats have elevated plasma insulin levels (Verma et al. 1994). Insulin infusion elevates blood pressure in SHR (Brand et al. 1994) and directly measured sympathetic nerve activity in humans (Anderson et al. 1991). Central administration of insulin also elevates lumbar sympathetic nerve activity in anesthetized Wistar rats (Muntzel et al. 1994), and treatment of SHR with hypoglycemic agents lowers plasma insulin levels and BP (Verma et al. 1994). Therefore, it seems possible that the reductions in blood pressure and sympathetic nervous system activity that result from hypocaloric diets are mediated, at least in part, by a reduction in plasma insulin.

There are several limitations to our experimental design that should be carefully considered. We did not control for differences in mineral consumption between the FR and ADLIB groups. Could lower sodium consumption in the FR group account for the reductions in BP in SHR? The SHR in the FR group consumed the equivalent of a 0.58 g/100 g NaCl diet (60% of a 0.96 g/100 g NaCl diet). We believe this is unlikely to have substantially influenced the BP response to food restriction because consumption of very low sodium diets (0.01-0.05 g/100 g) for several weeks does not reduce BP in SHR (Winternitz and Oparil 1982). However, we cannot completely discount this possibility. Furthermore, we cannot eliminate the possibility that reduced consumption of other minerals could have contributed to the antihypertensive actions of food restriction.

Another important limitation of our experimental design is that the food restriction was imposed during the development of hypertension in SHR. Thus food restriction blunted the development of hypertension in this group. Will food restriction lower established hypertension in SHR? Leenen et al. (1994) provide data that suggest that sympathetic tone is a very important component of BP in SHR that are 3-6 mo old. Thus if food restriction lowers BP via sympathetic mechanisms, SHR with established hypertension might be responsive to food restriction. This idea is supported by findings from studies of Wright et al. (1981) who fed 4-mo-old SHR 65% of control food consumption and noted a significant reduction in systolic blood pressure. Thus, we believe that food restriction can both attenuate the development of hypertension and lower established BP in SHR.

Finally, we have relied on an indirect measure of sympathetic nervous system activity, i.e., the decrement in BP in response to ganglionic blockade, to test the hypothesis that food restriction lowers BP via sympathetic neural mechanisms. Ganglionic blockade eliminates the tonic effects of both the parasympathetic and sympathetic branches of the autonomic nervous system. Because muscarinic receptor blockade with atropine alone increases HR but has minimal effects on the blood pressure of SHR (Chen et al. 1995), it is reasonable to conclude that the hypotensive actions of hexamethonium are primarily the result of removal of tonic sympathetic nervous system activity. Vasopressin and angiotensin II are released in response to the hypotension produced by ganglionic blockade and may produce a partial compensatory increase in BP. However, combined angiotensin II (AT-1) receptor and vasopressin (V-1) receptor blockade does not affect the magnitude of the peak depressor response to ganglionic blockade (Santajuliana et al. 1996). This finding provides additional support for this approach as a method to quantify sympathetic support of blood pressure. However, it is clear that additional indices of sympathetic nervous system function such as plasma norepinephrine levels, plasma norepinephrine spillover or direct measures of sympathetic nerve activity are required to examine more completely the relationship between energy intake and autonomic nervous system function in hypertensive animals.