INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CONSIDERABLE INFORMATION ABOUT the development of baroreflex function has been described recently based on studies in chronically instrumented fetal and newborn lambs (11, 12, 16, 17). These data show that the baroreflex functions in utero are at a higher sensitivity than after birth and that there is a developmentally related decrease in the response that allows blood pressure to shift to higher levels as the lamb matures. However, studies of baroreflex development beyond the early perinatal period have received less attention. In this regard, the Sprague-Dawley rat has been used because baroreflex function continues to mature for several weeks after birth (18). During the third postnatal week there is a surge of sympathetic outflow to the heart coincident with development of central integration mechanisms that gradually subside during the fourth week. In the urethan-anesthetized preweaning rat, a postnatal depression in baroreflex function has been reported during the second and third week (day 20) of life that was gone by the fourth week (day 25) (6). Blood pressure rose from 34 to 75 mmHg during this period. Hence, considerable development of baroreflex function continues after birth during early postnatal life.

In the present investigation, we studied the baroreflex control of heart rate (HR) in young conscious rats just after weaning (age 22-25 days) when potential maternal influences are absent. We sought to characterize the response over a wide range of arterial pressures, to determine the relative contribution of sympathetic and parasympathetic activity to the response, and to determine the effect of an acute stress (restraint/light) on the gain of the response.

There is evidence that suggests that acute stress resets the baroreflex to higher pressures in adult rats (4). This is presumably the appropriate adaptation to stress because the shift to higher blood pressure would defend against hypoperfusion of the brain. Several studies using adult rats have reported that an action of ANG II within the brain contributes to the cardiovascular alterations during acute stress (8, 15). Because deranged control of blood pressure in the young may be a factor that predisposes to cardiovascular disease later in life (3, 21), we sought to characterize factors that may have a role in maintaining normal blood pressure in the postnatal period. Little is known about the development of central ANG II effects in early postnatal life, especially with regard to the interaction between ANG II and the baroreflex during stressful stimuli. We hypothesized that even at an early age, brain ANG II would play an important role in modulating baroreflex control of HR during acute stress.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

All experiments were performed using young male Sprague-Dawley (Harlan Sprague Dawley) rats. The rats were typically delivered from the vendor at age 21-22 days. Two days were allowed for acclimatization, and then the animals were anesthetized with a mixture of ketamine (140 mg/kg) and acepromazine (1.4 mg/kg ip) and prepared for baroreflex testing by surgically implanting a femoral venous and arterial catheter (PE-50 heat-fused to PE-10). The catheters were tied off and tunneled subcutaneously to exit at the nape between the scapulae. In rats that also received intracerebroventricular injections, a guide cannula (21 gauge, Plastics One) was stereotaxically implanted above the right lateral cerebroventricle and anchored in place using skull screws and cranioplastic cement.

Baroreflex testing. Baroreflex control of HR was tested 2 days postoperatively while the rats were conscious and freely moving. The arterial catheter was connected to the pressure transducer, and 2 h of stabilization were allowed before baroreflex testing. The baroreflex was assessed by delivering intravenous ramp infusions of phenylephrine (Sigma) or sodium nitroprusside (Sigma). The phenylephrine was administered initially at a rate of 0.83 µg/min and was gradually increased to 10 µg/min as needed to produce a gradual increase in mean arterial pressure (MAP) of 35-50 mmHg over a 90- to 120-s period. Nitroprusside, which was always delivered after the phenylephrine, was initially administered at a rate of 2.3 µg/min and gradually increased to 23 µg/min as needed to produce a gradual decrease in MAP of 35-50 mmHg over a 90- to 120-s period. HR was monitored throughout the experiment. The relationship between MAP and HR was determined every 15 s (4-7 time points) for each infusion (phenylephrine and nitroprusside), and linear regression (SigmaPlot, Jandel) was used to determine the slope of the line in each case. Thus the HR response to baroreceptor loading and unloading was assessed separately. Because we intended to test the effect of different interventions on baroreflex function in a paired paradigm (before and after the intervention), control rats had their baroreflex function tested in the conscious freely moving state on two separate occasions separated by at least 2 h.

Contribution of sympathetic and parasympathetic activity. The contribution of the two arms of the autonomic nervous system to the baroreflex response in these young rats was assessed using intravenous injections of either atropine (Sigma, 0.5 mg/kg) or propranolol (Sigma, 1 mg/kg). Each drug was used in a separate group of rats (n = 5). The initial response was determined as outlined previously, and 2 h later the atropine or propranolol was given intravenously and then the baroreflex testing was repeated.

Effect of acute stress. Baroreflex function was assessed as outlined previously, and 2 h later the rats were placed in Plexiglas-restraining cages (5 × 3.5 × 10 cm) located 70 cm under two 150-W floodlights. The ambient temperature at the top of the restraining cages was 33°C. This combination of restraint, light, and heat has been shown to produce a reproducible effect on the hypothalamo-pituitary-adrenal axis (14). Thirty minutes after the beginning of the stress, baroreflex function was again assessed as above. This procedure typically took 20-30 min so the rats were restrained for ~1 h in total.

Effect of acute stress plus intracerebroventricular or intravenous losartan. The paired stress protocol outlined above was repeated except that the rats were given either intracerebroventricular losartan (AT₁ receptor antagonist, 10 µg in 5 µl) or intravenous losartan (10 mg/kg) just prior to being placed in the restraining cages (i.e., losartan was present during the stress but not during the initial baroreflex testing). At the end of the restraint period, the efficacy of the AT₁ receptor blockade was assessed by giving ANG II either intracerebroventricularly (10 ng) or intravenously (100 ng). In all cases, the effect of the ANG II on blood pressure was completely blocked. Control rats received two periods of baroreflex testing in the freely moving state (no stress during the second test), and intracerebroventricular losartan was given 30 min prior to the second assessment. A control group with intravenous losartan was not included.

Effect of acute stress plus intracerebroventricular -hCRH. The paired stress protocol was repeated in a separate group of rats (n = 7) that received an injection of intracerebroventricular -helical corticotropin-releasing hormone (-hCRH) receptor antagonist, 10 µg (Peninsula Laboratories); just prior to being placed in the restraining cage.

Data analysis. The slopes determined for baroreceptor loading and unloading were analyzed separately using either paired t-test (two means) or ANOVA for repeated measures (more than two means). Baseline MAP and HR were compared using similar analyses. Post hoc analysis was performed using Newman-Keuls test. For all analyses, a P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In these young rats, baroreceptor loading produced a greater slope in the relationship between MAP and HR than did baroreceptor unloading. This may have been due to the high-resting HR with limited room for further increase during decreases in MAP. Figure 1 depicts "average" responses for control rats obtained on two separate occasions separated by at least 2 h. Because the intracerebroventricular losartan-treated no-stress control rats (n = 4) had responses that were not statistically different from untreated no-stress control rats (n = 4), both groups were combined to create Fig. 1. Resting MAP and HR before beginning the first assessment of baroreflex function were 110 ± 3 mmHg and 525 ± 23 beats/min, and before beginning the second assessment they were 101 ± 4 mmHg and 547 ± 12 beats/min, respectively. There were no statistical differences between the first and second test baseline values. The average slope of the response to baroreceptor loading was not different between the two testing periods, but the slope of the response produced by nitroprusside was slightly (P < 0.05) decreased during the second test (Fig. 2A).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Average baroreflex response determined during 2 testing periods separated by at least 2 h. Lines were generated using the mean slopes and intercepts determined by linear regression. One-half the rats (n = 4) had no treatment during the first or second testing period. The other one-half (n = 4) were treated with intracerebroventricular losartan 30 min before the second testing period. The data were combined since there were no statistical differences between the two groups. r2, coefficient of determination; bpm, beats/min; HR, heart rate; MAP, mean arterial pressure. * P < 0.05.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effect of various treatments on the slopes of the baroreflex response. A: slopes produced during 2 unstressed testing periods (same slopes as shown in Fig. 1). B: effect of acute stress during the second testing period. C: effect of acute stress plus intracerebroventricular losartan during the second testing period. D: effect of acute stress plus intravenous losartan during the second testing period. * P < 0.05. The first vertical bar in each set represents the slope produced during the first control testing period. The second vertical bar represents the slope produced during the second testing period with the drug on board.

Pretreatment of the rats with intravenous atropine before the second baroreflex test produced a significant increase in baseline HR (from 483 ± 18 to 552 ± 16 beats/min) without a significant effect on MAP (117 ± 7 mmHg before, 105 ± 14 mmHg after) and a marked decrease in the slope of the response produced by baroreceptor loading (Fig. 3, left). The response to baroreceptor unloading was not significantly affected by the atropine. It is estimated that 65% of the decrease in HR produced by increased blood pressure was due to increased parasympathetic tone. Propranolol treatment prior to the second baroreflex test significantly decreased resting HR (538 ± 12 to 457 ± 10 beats/min) without a significant effect on MAP (104 ± 6 mmHg before, 98 ± 12 mmHg after). The slope of the response to baroreceptor unloading was significantly decreased (~80%) by the propranolol (Fig. 3, right). The slope of the response to baroreceptor loading was decreased by 33%, although this effect was not statistically significant.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Effect of treatment with atropine (A) or propranolol (B) immediately before the second testing period on the slopes of the relationship between MAP and HR produced by baroreceptor loading (with phenylephrine) or unloading (with nitroprusside). See Fig. 2 legend for explanation of vertical bars.

Acute stress significantly affected the slope of the baroreflex response produced by both phenylephrine and nitroprusside (Fig. 2B). During the stress, the slope of the response during baroreceptor loading was decreased, whereas the slope of the response during baroreceptor unloading was increased. At the beginning of the baroreceptor test (30 min into the restraint), baseline MAP and HR were not different from the baseline values before the first testing (in the unrestrained state) period (Table 1). Pretreatment with intracerebroventricular losartan immediately before the beginning of the restraint stress prevented the expected stress-induced decrease in slope to phenylephrine infusions but had no effect on the stress-induced increase in slope to the nitroprusside (Fig. 2C). Baseline values for MAP and HR during the stress were not different from the prestress control period (Table 1). After pretreatment with intravenous losartan, the stress also produced a smaller decrease in the slope to baroreceptor loading and the response to nitroprusside was unaffected (Fig. 2D). Baseline MAP and HR were unaffected by the intravenous losartan (Table 1).

Pretreatment with the CRH receptor antagonist -hCRH before the beginning of the restraint had no effect on initial MAP or HR (105 ± 1 vs. 98 ± 7 mmHg and 494 ± 17 vs. 516 ± 15 beats/min, respectively). After intracerebroventricular -hCRH, the stress produced a significant decrease in the slope of the response to baroreceptor loading (Fig. 4). As with the other treatments, the stress-induced increase in the slope of the response to nitroprusside was unaffected by the CRH receptor antagonist.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of stress plus intracerebroventricular -helical corticotropin-releasing hormone (-hCRH), a CRH receptor antagonist, on the slopes of the baroreflex. See Fig. 3 legend for explanation of vertical bars.* P < 0.05.

In an effort to compare the magnitude of the stress effect with the different treatments during baroreceptor loading, one-way ANOVA was performed on the calculated percent of control slope (treatment slope/initial slope) for each treatment. A percentage normalization was used because the absolute slopes tended to vary from group to group, although there was no statistically significant difference (one-way ANOVA) in initial slopes. The stress produced a marked significant decrease in slope that was almost completely reversed by the intracerebroventricular losartan (Fig. 5, left 3 bars). The normalized slopes for the intravenous losartan and intracerebroventricular -hCRH also tended to be greater than the stress control group, but the difference was not significant (Fig. 5, right 2 bars).

View larger version (37K): [in this window] [in a new window]   Fig. 5.   Comparison of the magnitude of the effect of the stress (normalized to percent of initial slope) with the different treatments during baroreceptor loading with phenylephrine. Comparisons were made using one-way ANOVA followed by Newman-Keuls test. * P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The studies in the present investigation have provided a significant amount of information concerning baroreflex function in conscious 3-wk-old rats. These animals were just past weaning and no longer could be influenced by maternal interaction or substances in maternal milk.

Baroreceptor function is often assessed using a four-parameter logistic sigmoid model (7). However, in the present investigation, two separate linear models were used to analyze the relationship between MAP and HR, one for baroreceptor loading and one for baroreceptor unloading. There were a number of reasons why the linear model was used. First, whereas the data from some control rats could be fit into the sigmoid model (SigmaPlot, Jandel), this was not the case with others. To fit a sigmoid model there needs to be clear "plateaus" in HR at the upper and lower ends of MAP. This was not always the case; in fact, in some of these young rats there was a delayed secondary fall in HR shortly after peak MAP was reached during the phenylephrine infusions. The mechanism for this continued fall in HR in the face of a constant MAP is not presently known. A second reason for a lack of fit into the sigmoid model was the universal finding that the slope of the relationship between MAP and HR caused by baroreceptor loading was much steeper than the slope produced by baroreceptor unloading. Finally, the data obtained during the acute stress could never be fit into the sigmoid model. The average coefficient of determination (r²) for each line in Fig. 1 (range 0.88-0.97) shows that the linear model provided a good approximation of the relationship between MAP and HR. For all other regressions made, a similar range of r² values were obtained (0.86 the lowest and 0.97 the highest).

The relatively flat slope produced by baroreceptor unloading in the unstressed state could be because in young rats the resting blood pressure is near the upper end of the range of HR responsiveness. This fits with data reported for newborn lambs where the calculated "center point" (of HR response) for MAP was 85 mmHg, but the resting MAP was 69 mmHg, which was near the maximal HR (12). Interestingly, in adult sheep, the calculated center point (116 mmHg) and the resting MAP (120 mmHg) were almost identical (12). Thus baroreceptor function in these young animals may not be as capable of buffering against decreases in blood pressure as will eventually be the case as adults. It has been suggested that the surge in sympathetic outflow to the heart that occurs during the third and fourth postnatal week may provide the final adjustments of cardiac sensitivity to norepinephrine (18). The studies with propranolol showed that there is significant basal sympathetic tone in these young rats and that most of the rise in HR with blood pressure decreases was due to increased sympathetic outflow. The relatively flat HR response to decreases in blood pressure may be due to incomplete maturation of cardiac responsiveness to norepinephrine.

The studies with atropine showed that the young rats also had significant parasympathetic tone at rest. Furthermore, most of the fall in HR (~2/3) with blood pressure increases was due to increased parasympathetic outflow (with a minor contribution of sympathetic withdrawal). The greater slope produced during baroreceptor loading is consistent with the notion of earlier maturation of the parasympathetic arm of the baroreflex.

There have been a number of studies that have looked at the effect of various acute stressors on cardiovascular function in adult rats (4, 8, 15). However, the effect of stress in young immature rats has been largely ignored. Cardiovascular disease in adults could result from an accumulation of inappropriate responses to environmental stresses throughout life, including events at an early age when control mechanisms may still be developing. In the present investigation, the effect of acute stress on baroreflex function in these young 3-wk-old rats was determined. A combination of restraint, light, and heat was used to evoke a reproducible stress response. This type of paradigm has been used successfully by others to activate the stress response in adults (14). The acute stress affected the response to baroreceptor loading and baroreceptor unloading differently in these young rats. The slope of the response to sodium nitroprusside infusions was markedly increased during the stress, and the slope of the response to phenylephrine infusions was significantly reduced. This sort of acute adaptation would presumably favor increases in blood pressure and disfavor decreases, although baseline blood pressure was not affected by the acute stress. Although atropine and propranolol were not given during the acute stress paradigm, the results with these agents in the unstressed state suggest that the greater tachycardia during baroreceptor unloading was probably due to greater sympathetic outflow and that the decreased bradycardia during baroreceptor loading was probably due to less parasympathetic outflow.

It should be noted that neither baseline MAP nor HR was significantly increased by the stress in any group (Table 1). This was surprising because both parameters are usually reported to be increased by acute stress in adult rats (4, 15). The reason for the lack of increase in MAP or HR is not known. It is possible that the vasodilation in the tail circulation due to the rise in body temperature during the stress may have countered stress-induced vasoconstriction in other vascular beds and resulted in no change in total peripheral resistance. The lack of HR increase during stress could be due to incomplete maturation of cardiac response to sympathetic activation as mentioned previously.

A central action of ANG II has been implicated in the control of baroreflex function in both young and adult animals (13, 16, 17). In newborn sheep, the acute injection of losartan into the lateral cerebroventricles lowered baseline arterial pressure and shifted the baroreflex response to a lower midpoint pressure without affecting the gain (17). Interestingly, fourth cerebroventricle injection of losartan also shifted the curve toward lower pressure but only in 8-wk-old lambs and not in newborns. This suggests that development of central ANG II circuits continues during early postnatal life. In the present study, intracerebroventricular injection of losartan had no effect on baseline blood pressure or on the baroreflex response in the unstressed animals. This is most likely due to species differences between sheep and rats since we have never seen an affect of intracerebroventricular losartan on baseline blood pressure in adult or young rats (19, 20). Hence, central ANG II does not appear to contribute tonically to control of baroreflex function in young rats, at least in brain areas that are accessible by intracerebroventricular losartan.

On the other hand, in the presence of acute stress, a central role for ANG II in control of baroreflex function was established. The attenuated slope of the baroreflex in response to baroreceptor loading produced by acute stress was virtually abolished by intracerebroventricular losartan (Figs. 2 and 5). There is abundant evidence that central ANG II decreases vagal tone to the heart (9, 10, 13). The data from the present study are consistent with the idea that stress-induced increases in central ANG II decreased baroreflex responsiveness (to baroreceptor loading) presumably by inhibiting parasympathetic outflow to the heart. A role of peripheral ANG II in attenuating the baroreflex cannot be ruled out. Intravenous losartan also may have abolished partially the stress-induced changes (Fig. 5). Most studies suggest that central cardiovascular effects of circulating ANG II are mediated at the area postrema (13). The acute stress presumably increased renin secretion with subsequent increased generation of circulating ANG II. Further studies will be needed to determine the relative contribution of brain ANG II and peripheral ANG II to the observed effect.

The enhancement of the tachycardic response to baroreceptor unloading produced by the acute stress was not affected by losartan delivered via either route. This suggests central ANG II does not contribute to the stress-induced enhancement of the sympathetic arm of the baroreflex. This was somewhat surprising because it was reported recently that intracerebroventricular losartan blocked stress-induced increases in circulating catecholamines in adult rats (5). It is not known whether the lack of effect of losartan in the present study was because the rats were immature or whether stress affects mechanisms for sympathetic outflow to baroreceptor unloading differently than it affects basal circulating catecholamines.

A primary neurohumoral response to acute stress is increased release of CRH into the hypophyseal portal blood from neurons of the parvocellular paraventricular nucleus. Fisher (2) reported that intracerebroventricular administration of CRH in adult rats had effects on baroreceptor function that were similar to those produced by acute stress, that is, the CRH produced decreases in reflex gain with a shift toward higher midrange pressures. Interestingly, it was the parasympathetic arm of the response that was affected by the CRH. In the present investigation, intracerebroventricular -hCRH, a CRH receptor antagonist, may have partially attenuated the stress-induced decrease in responsiveness to baroreceptor loading. It is not known whether the CRH and ANG II affected separate pathways or whether they were involved in the same pathway. CRH-containing neurons in the paraventricular nucleus of the hypothalamus also express the ANG II AT₁ receptor, and it could be speculated that the ANG II effect is proximal to the CRH effect (1). Stress-induced increases in central ANG II (or peripheral ANG II via circumventricular organs) could produce increased expression and release of CRH from paraventricular nucleus neurons, which then affect baroreflex function via projections to the medulla. This fits with recent data in adult rats showing that intracerebroventricular losartan did not prevent stress-induced increases in circulating ACTH or corticosterone, but it did decrease the expression of paraventricular CRH mRNA in the hypothalamus (5).

Perspectives