- Email: [email protected]
Baroreflex function in chronically stressed borderline hypertensive rats
Physiology& Behavior,Vol. 49, pp. 539-542. ©PergamonPress plc, 1991. Printedin the U.S.A.
0031-9384/91 $3.00 + .00
Baroreflex Function in Chronically Stressed Borderline Hypertensive Rats I J A M E S E. L A W L E R , B R I A N J. S A N D E R S , 2 R O N A L D H. C O X 3 A N D E D W A R D F. O ' C O N N O R
Physiology Program and Department of Psychology, University of Tennessee, Knoxville, TN 37996-0900 R e c e i v e d 11 October 1990
LAWLER, J. E., B. J. SANDERS, R. H. COX AND E. F. O'CONNOR. Baroreflexfunction in chronically stressed borderline hypertensive rats. PHYSIOL BEHAV 49(3) 539--542, 1991.--A number of previous studies have demonstrated that some aspect of baroreflex function is altered as hypertension develops. However, no studies have determined whether a chronic stressor can alter haroreflex function in the resting state. In the present study, male borderline hypertensive rats (BHR) were divided inW three groups: control, stressed daily for 5 weeks, and stressed daily for 11 weeks. At the appropriate time, 7 different dosages of angiotensin H (AH) were given intravenously as a bolus injection. Heart rate (interbeat interval, or IBI) and mean arterial pressure were tracked for 90 subsequent beats. For each group, intercorrelations between pressure and IBI were obtained. In addition, overall means of pressure and IBI for each dosage were computed for each group and plotted. Higher dosages of AII were required to produce a significant correlation between pressure and IBI in the 5-week group compared to control. In the 11-week group, only the highest dosage yielded a significant correlation between pressure and I]31. When data were expressed in terms of the highest pressure and its corresponding IBI for each group, the 5-week-stress group had a shift in set point compared to control. The 11-week group showed a shift in set point and a reduction in gain compared to control animals. The similarity between these findings and those for other models of hypertension is discussed, with special emphasis on the potential role of the central nervous system. Borderline hypertensive rat
Baroreflex function
Behavioral stress
Blood pressure
Central nervous system
activity showed a greater sensitivity in SHR than WKY per unit change in blood pressure, the authors concluded that SHR have impaired central neural integration of autonomic function. Using decerebmte animals, Gonzalez et al. (8) observed that SHR showed a smaller reduction in blood pressure to direct aortic nerve stimulation than WKY. These same investigators also observed a smaller reduction in sympathetic nerve activity to aortic nerve stimulation in SHR compared to WKY. The authors concluded that a central bulbospinal site might be important in the hypertension of the SHR. Prevention of hypertension with captopril in the SHR is accompanied by an increase in baroreceptor sensitivity in comparison to age-matched SHR not treated with captopril (5). Although the data are consistent with the hypothesis that the brain reninangiotensin system plays a role in the hypertension of the SHR, the authors note that the increased sensitivity of the baroreflex may also be due to the hypotensive effects of captopril per se. In summary, baroreflex function is compromised in a number of animal models of hypertension. In the SHR, data suggest that both peripheral and central factors are involved (10,23). Since the CNS may be at least partly implicated in baroreflex resetting, it is possible that environmental factors thought to play a role in the etiology of human hypertension could produce chronic changes in baroreflex function in animals. While salt intake has been investigated in several rat models, such as the Dab.1 and DOCA-NaC1 models (see above), there do not appear to be any studies which
ONE manner in which chronic increases in blood pressure can occur is through alterations in baroreflex function. Several animal models of hypertension show some degree of altered baroreflex sensitivity. For example, Matsuguchi and Schmid (16) observed an impaired baroreflex function in the deoxycorticosterone acetate-salt model of hypertension. Baroreflex function is reduced in hypertensive rats of the Lyon strain (22). Similarly, altered baroreflex sensitivity has been observed in a renal hypertensive model (15). Finally, Nosaka and Okamoto (19) have made similar observations in the spontaneously hypertensive rat (SHR). Alterations in baroreflex function in the SHR model have subsequently been studied by a number of investigators. Many of these studies have focused on afferent nerve activity, central resetting or altered sympathetic nervous system traffic. Sapru and Wang (20) found that treatment of SHR with antihypertensives led to a reversal of both the hypertension and the baroreceptor resetting present in untreated SHR. The fact that afferent nerve activity was decreased in hypertrophied aortas led the authors to conclude that such hypertrophy might be an important determinant of impaired baroreflex function. Using baroreceptor deafferentation and direct sympathetic nervous system recording, Judy and Farrell (9) concluded that central resetting rather than baroreceptor dysfunction was an important mechanism of the hypertension in the SHR. Luft et al. (14) found that the heart rate response of SHR was less sensitive to pressor and depressor substances than in WKY. Since splanchnic nerve
1This research was supported by HL-01395 and HL-19680, both to J.E.L. 2Present address: Department of Psychology, Drake University, Des Moines, IA 50311. 3Present address: Depamnent of Physical Education and Recreation, Miami University, Oxford, OH 45056. 539
540
LAWLER, SANDERS, COX AND O'CONNOR
have investigated whether chronic stress can produce an alteration in baroreflex function. Part of the reason why chronic stress has not been investigated may be related to the failure of most stress studies to produce chronic alterations in blood pressure. In the last decade, we have developed a new animal model, the borderline hypertensive rat (BHR), which shows significant elevations in blood pressure in response to chronic stress (11) or dietary salt intake (12). Now that a model for chronic stress exists which shows large, permanent changes in blood pressure (13), studies can be performed which examine the role of altered baroreflex sensitivity in the observed blood pressure changes. In the present study, male BHR were subjected to either 5 or 11 weeks of chronic, daily, tail-shock stress. Intravenous bolus injections of angiotensin II (6) in the conscious, resting state were used to assess the sensitivity of the baroreflex in these groups compared to unstressed, control BHR. METHOD
Animals
Forty-eight mate BHR were used in this study. Inbred parent strains of female SHR and male WKY were purchased from Taconic Farms (Germantown, NY). Before breeding, indirect measurements of systolic blood pressure using tail plethysmography were obtained to determine that SHR and WKY were indeed hypertensive and normotensive, respectively. Male BHR offspring were weaned at 21 days, housed two to a cage, and maintained on laboratory rat chow and tap water ad lib throughout the remainder of study. Beginning at 8 weeks of age, all male BHR were gentled daily for 5 minutes. Weekly tail-cuff systolic blood pressures were also obtained in conscious, mildly restrained animals. The mean of 7 artifact-free determinations, taken once weekly in each animal, was defined as the systolic pressure for the week.
TABLE 1 PEARSON r CORRELATIONS FOR MEAN ARTERIAL BLOOD PRESSURE AND INTERBEAT INTERVAL FOR 10 SUCCESSIVE BEATS IMMEDIATELY PRECEDING THE MAXIMUM BLOOD PRESSURE RESPONSE TO VAR1OUS IV DOSAGES OF ANGIOTENSIN II IN UNSTRESSED ANIMALS (CONT), ANIMALS STRESSED FOR 5 WEEKS (EARLY) AND ANIMALS STRESSED FOR 11 WEEKS (LATEI
Cont Early Late
Sal
2 ng
4 ng
6 ng
20 ng
40 ng
60 ng
0.03 -0.57 -0.26
0.45 0.18 0.18
0.30 -0.12 0.52
0.87* -0.11 {).64
0.96* 0.68 0.62
0.80* 0.8t* 0.78* 0.85* (I.60 0.70*
*Depicts significantcorrelationsat p<0.025.
dosage. A Pearson r was then determined for the t0 mean pressures and their 10 corresponding interbeat intervals in each group and at each dosage. Since this resulted in 21 correlations (3 groups for 7 different dosages), a more stringent p-value (0,025) was chosen for significance. These data were also subjected to regression analysis for determination of slope and intercept. Additional analyses utilized plots of the average peak mean arterial pressure versus the average peak interbeat interval for each dosage of All in each of the three groups. For each dosage, the peak mean pressure was determined for each animal. The longest interbeat interval which occurred during the next three beats was taken as the IBI value for that dose of All. The peak mean arterial blood pressure and peak IBI were obtained for each animal and for each dose and averaged for the group. A line of best fit was then determined for each group. This line depicts the response of one group for the seven data points composed of saline and 2, 4, 6, 20, 40 and 60 ng/kg All. An a priori decision was made to compare the 5- and 1l-week groups to the control group. RESULTS
Procedure
At 10 weeks of age, animals were divided into 3 groups: an Early group (stressed for 5 weeks), a Late group (stressed for 11 weeks), and an unstressed control group which was age matched to the stress groups. Stressed groups received daily, two-hour sessions of pseudorandom, cutaneous tall shock (t mA, 1-s duration, averaging every 30 s). The presentation of all shocks was under microcomputer control. After the appropriate duration of stress (or at the appropriate age in control rats), femoral artery catheters were implanted under ether anesthesia. Following a 3-day recovery period, the blood pressure and heart rate responses to bolus injections of angiotensin II were determined. Dosages (0, 2, 4, 6, 20, 40, or 60 ng/kg AlI, IV) were randomly administered, with a minimum 5-minute recovery period between injections. All data were fed on-line to an IBM personal computer and 8-charmel analog to digital converter. Software stored the mean pressure and interbeat intervals for 90 consecutive beats for each injection. Raters blind to the group assignment and dosage then chose the highest pressure reached in the 90-beat interval. The mean pressure and interbeat intervals for that beat, and the 9 preceding beats, were determined for each animal at each dosage. Thus, each animal had 10 mean pressures and 10 corresponding interbeat intervals for each of 7 different All injections. Each peak pressure was averaged with all the other peak pressures for all BHR in a given group and a given AII dosage. In a similar manner, all pressures and interbeat intervals one beat before the highest pressure were averaged, etc., until 10 pressure and interbeat interval values were determined for a given group and a given AII
Table 1 depicts the Pearson r for each All injection dosage in each of the 3 groups. An asterisk depicts significant correlations. As can be seen, control animals had significant correlations between pressure and interbeat interval for all dosages of 6 ng/kg or greater. Animals stressed for 5 weeks, however, had significant correlations only at the 40 and 60 ng/kg dosages, while animals stressed for 11 weeks had a significant correlation only at the highest dosage tested. Since the regression line is not meaningful in the absence of a significant correlation, the regressions could only be compared between groups at one All dosage (60 ng/kg), since this was the only dosage which yielded significant correlations for all 3 groups. Figure 1 depicts the regression lines for the 10 data points in control animals (triangles and dotted line), early stress animals (open circles and solid line), and late stress animals (closed circles and dashed line). The steepest slope (most sensitive lengthening of interbeat interval for a given rise in blood pressure) was found in the unstressed animals, while animals stressed for 11 weeks showed the shallowest slope. The regression equation for each of the groups was: IBI = 142.8 + 0.336 × BP (control); IBI = 142.7 + 0.278 x BP (early); and IBI = 149.5 + 0.263 × BP (late). Figure 2 compares the maximum pressor and IBI responses to each dose of AII in control and 5-week groups. These data, as do those in Fig. 1, suggest that there is no change in the sensitivity of the baroreflex when BHR stressed for 5 weeks are compared with controls. Figure 3, on the other hand, depicts the same data for control and 11-week groups. These data suggest a reduced sensitivity of the baroreflex in BHR stressed for 11 weeks com-
BAROREFLEX IN STRESSED RATS
541
BP-IBI RESPONSE TO ANG II
210
230 [
CONT 200 ~
.''''"
~
......it ~ .
|
£
LATE EARLY
190
200 : _m 180-
190 ~
180 170 120
I 130
I 140
I 150
I 160
I 170
I 180
170 160
AR]ERL,M. BP (mmHg)
120
FIG. 1. Regression lines relating pressure and interbeat interval in unstressed (CONT), 5-week-stressed(EARLY), and 11-week-stressed(LATE) animals in response to a 60 ng/kg IV bolus injection of angiotensin II. pared to control BHR. A striking resemblance is seen between the data depicted in Fig. 1 and those in Figs. 2 and 3, even though they represent two different ways of depicting the sensitivity of the baroreflex in this study. DISCUSSION Changes in baroreflex function have been noted in a number of animal models of hypertension (15, 16, 22). To the best of our knowledge, the present study is the first to examine whether chronic stress alters the sensitivity of the baroreflex. The data presented, although preliminary in nature, are nevertheless consistent in that BHR subjected to 5 weeks of stress have fairly minor changes in baroreflex function compared with BHR subjected to 11 weeks of stress. All three figures presented are consistent with the hypothesis that chronic stress in a genetically predisposed animal produces a gradual shift in the sensitivity of the baroreflex. In fact, the pattern observed in the present study has been found in other models of hypertension. For example, hypertenBP-IBI RESPONSE TO ANG II
2 o|
j
220*
.E
I OCON
|
160 L
120
/ .
! ~
.
130
.
.
14.0
.
150
160
I
I
170
180
190
MEAN BP (mm H9)
FIG. 2. Regression lines relating pressure and interbeat interval in unstressed (open circles) BHR and in BHR stressed for 5 weeks (f'dled circles). Each data point represents the mean ( __.SEM) for all the animals in a given group at one dosage of All.
I
t
I
I
I
!
130
140
150
160
170
180
190
MEAN BP (ram H9)
FIG. 3. Regression lines relating pressure and interbeat interval in unstressed (open circles) BHR and in BHR stressed for 11 weeks (filled circles). Each data point represents the mean ( ---SEM) for all the animals in a given group at one dosage of AII. sion of short duration results in a shift of the baroreflex curve to the right without a change in gain (21). This change in set point means that the reflex is not triggered until a higher pressure is reached. This is exactly what was seen in the BHR stressed for 5 weeks when compared to control animals. As hypertension persists, the sensitivity (gain) of the reflex is compromised (2). Thus, for a given change in blood pressure, a smaller change in heart rate is seen. This is exactly what was seen in the BHR stressed for 11 weeks when compared to control animals. Numerous investigators have debated the mechanisms responsible for the observed changes in baroreflex function. Some approaches emphasize vessel mechanics. In this view, alterations in distensibility of the vessel wall result in less mechanical distortion of the baroreceptors. Such changes have been noted in the SHR (1). Other approaches emphasize the role of the central nervous system. The nucleus and tractus solitarius (NTS), the point of termination of afferent fibers of the baroreflex, has been the primary focus of studies into central control of the baroreflex. Buchholz and Nathan (3) observed that lesions of the NTS produced chronic lability of blood pressure and eliminated the reflex bradycardia to phenylephrine. Local injection of an All antagonist into the NTS increased the sensitivity of the baroreflex (4). Finally, microinjection of vasopressin into the NTS reduced the sensitivity of the baroreflex (17). Other areas of the CNS also have an effect on the baroreflex. The most relevant to studies of chronic stress is the observation that direct stimulation of hypothalamic areas that lead to a defense reaction can override the baroreflex (7). If future work with stress in the BHR suggests an involvement of the CNS instead of peripheral adaptations, then it would seem that the hypothalamus would be an important structure on which to focus. Recent research from our laboratory (18) demonstrates that changes in hypothalamic levels of norepinephrine occur, but only after stress has been continued for many weeks. Whether or not there is a link between hypothalamic levels of norepinephrine and CNS overriding of the baroreflex awaits further research.
REFERENCES 1. Andresen, M. C.; Kuraoka, S.; Brown, A. M. Baroreceptor function and changes in strain sensitivity in normotensive and spontaneously hypertensive rats. Circ. Res. 47:821-828; 1980.
2. Brown, A. M. Receptors under pressure: An update on baroreceptors. Circ. Res. 46:1-10; 1980. 3. Buchholz, R. A.; Nathan, M. A. Chronic lability of the arterial blood
542
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
LAWLER, SANDERS, COX AND O'CONNOR
pressure produced by electrolytic lesions of the nucleus tractus solitarii in the rat. Circ. Res. 54:227-238; 1984. Campagnole-Santos, M. J.; Diz, D. I.; Ferrario, C. M. Baroreceptor reflex modulation by angiotensin II at the nucleus tractus solitarii. Hypertension 1 l(Suppl. I): 167-171; 1988. Cheng, S. W. T.; Swords, B. H.; Kirk, K. A.; Berecek, K. H. Baroreflex function in lifetime-captopril-treated spontaneously hypertensive rats. Hypertension 13:63-69; 1989. Coleman, T. G. Arterial baroreflex control of heart rate in the conscious rat. Am. J. Physiol. 238:H515-H520; 1980. Gebber, G. L.; Snyder, D. W. Hypothalamic control of baroreceptor reflexes. Am. J. Physiol. 218:124-131; 1970. Gonzalez, E. R.; Krieger, A. J.; Sapru, H. N. Central resetting of baroreflex in the spontaneously hypertensive rat. Hypertension 5:346352; 1983. Judy, W. V.; Farrell, S. K. Arterial baroreceptor reflex control of sympathetic nerve activity in the spontaneously hypertensive rat. Hypertension 1:605-614; 1979. Krieger, E. M. Neurogenic mechanisms in hypertension: Resetting of the baroreceptors. Hypertension 8(Suppl. I):7-14; 1986. Lawler, J. E.; Barker, G. F.; Hubbard, J. W.; Allen, M. T. The effects of conflict on tonic levels of blood pressure in the genetically borderline hypertensive rat. Psychophysiology 17:363-370; 1980. Lawler, J. E.; Sanders, B. J.; Chen, Y.-F.; Nagahama, S.; Oparil, S. Hypertension produced by a high sodium diet in the borderline hypertensive rat. Clin. Exp. Hypertens. A9:1713-1731; 1987. Lawler, J. E.; Barker, G. F.; Hubbard, J. W.; Schaub, R. G. Effects of stress on blood pressure and cardiac pathology in rats with borderline hypertension. Hypertension 3:496-505; 1981. Luft, F. C.; Demmert, G.; Rohmeiss, P.; Unger, T. Baroreceptor
15. 16. 17. 18. 19. 20. 21.
22. 23.
reflex effect on sympathetic nerve activity in stroke-prone spontaneously hypertensive rats. J. Auton. Nerv. Syst. 17:199-209; 1986. McCubbin, J. W.; Green, J. H.; Page, I. H. Baroreceptor function in chronic renal hypertension. Circ. Res. 4:205-210; 1956. Matsuguchi, H.; Schmid, P. G. Pressor response to vasopressin and impaired baroreflex function in DOC-salt hypertension. Am. J. Physiol. 242:H44-H49; 1982. Michelini, L. S.; Bonagamba, L. G. H. Baroreceptor reflex modulation by vasopressin microinjected into the nucleus tractus solitarii of conscious rats. Hypertension ll(Suppl. I):75-79; 1988. Mitchell, V. P.; Lawler, J. E. Norepinephrine content of discrete brain nuclei in acutely and chronically stressed borderline hypertensive rats. Brain Res. Bull. 22:545-547; 1989. Nosaka, S.; Omamoto, K. Modified characteristics of the aortic baroreceptor activities in the spontaneously hypertensive rat. Jpn. Circ. J. 34:685-693; 1970. Sapru, H. N.; Wang, S. C. Modification of aortic baroreceptor resetting in the spontaneously hypertensive rat. Am. J. Physiol. 230:664674; 1976. Sleight, P.; Robinson, J. L.; Brooks, D. E.; Rees, P. M. Characteristics of single carotid sinus baroreceptor fibers and whole nerve activity in the normotensive and the renal hypertensive dog. Circ. Res. 41:750-758; 1977. Su, D. F.; Cerutti, C.; Barres, C.; Vincent, M.; Sassard, J. Blood pressure and baroreflex sensitivity in conscious hypertensive rats of Lyon strain. Am. J. Physiol. 251:H1111-HI 117; 1986. Thames, M. D. Arterial baroreflexes in hypertension. In: Guthrie, G. P.; Kotchen, T. A., eds. Hypertension and the brain. Mt. Kisco, NY: Futura; 1984.
0031-9384/91 $3.00 + .00
Baroreflex Function in Chronically Stressed Borderline Hypertensive Rats I J A M E S E. L A W L E R , B R I A N J. S A N D E R S , 2 R O N A L D H. C O X 3 A N D E D W A R D F. O ' C O N N O R
Physiology Program and Department of Psychology, University of Tennessee, Knoxville, TN 37996-0900 R e c e i v e d 11 October 1990
LAWLER, J. E., B. J. SANDERS, R. H. COX AND E. F. O'CONNOR. Baroreflexfunction in chronically stressed borderline hypertensive rats. PHYSIOL BEHAV 49(3) 539--542, 1991.--A number of previous studies have demonstrated that some aspect of baroreflex function is altered as hypertension develops. However, no studies have determined whether a chronic stressor can alter haroreflex function in the resting state. In the present study, male borderline hypertensive rats (BHR) were divided inW three groups: control, stressed daily for 5 weeks, and stressed daily for 11 weeks. At the appropriate time, 7 different dosages of angiotensin H (AH) were given intravenously as a bolus injection. Heart rate (interbeat interval, or IBI) and mean arterial pressure were tracked for 90 subsequent beats. For each group, intercorrelations between pressure and IBI were obtained. In addition, overall means of pressure and IBI for each dosage were computed for each group and plotted. Higher dosages of AII were required to produce a significant correlation between pressure and IBI in the 5-week group compared to control. In the 11-week group, only the highest dosage yielded a significant correlation between pressure and I]31. When data were expressed in terms of the highest pressure and its corresponding IBI for each group, the 5-week-stress group had a shift in set point compared to control. The 11-week group showed a shift in set point and a reduction in gain compared to control animals. The similarity between these findings and those for other models of hypertension is discussed, with special emphasis on the potential role of the central nervous system. Borderline hypertensive rat
Baroreflex function
Behavioral stress
Blood pressure
Central nervous system
activity showed a greater sensitivity in SHR than WKY per unit change in blood pressure, the authors concluded that SHR have impaired central neural integration of autonomic function. Using decerebmte animals, Gonzalez et al. (8) observed that SHR showed a smaller reduction in blood pressure to direct aortic nerve stimulation than WKY. These same investigators also observed a smaller reduction in sympathetic nerve activity to aortic nerve stimulation in SHR compared to WKY. The authors concluded that a central bulbospinal site might be important in the hypertension of the SHR. Prevention of hypertension with captopril in the SHR is accompanied by an increase in baroreceptor sensitivity in comparison to age-matched SHR not treated with captopril (5). Although the data are consistent with the hypothesis that the brain reninangiotensin system plays a role in the hypertension of the SHR, the authors note that the increased sensitivity of the baroreflex may also be due to the hypotensive effects of captopril per se. In summary, baroreflex function is compromised in a number of animal models of hypertension. In the SHR, data suggest that both peripheral and central factors are involved (10,23). Since the CNS may be at least partly implicated in baroreflex resetting, it is possible that environmental factors thought to play a role in the etiology of human hypertension could produce chronic changes in baroreflex function in animals. While salt intake has been investigated in several rat models, such as the Dab.1 and DOCA-NaC1 models (see above), there do not appear to be any studies which
ONE manner in which chronic increases in blood pressure can occur is through alterations in baroreflex function. Several animal models of hypertension show some degree of altered baroreflex sensitivity. For example, Matsuguchi and Schmid (16) observed an impaired baroreflex function in the deoxycorticosterone acetate-salt model of hypertension. Baroreflex function is reduced in hypertensive rats of the Lyon strain (22). Similarly, altered baroreflex sensitivity has been observed in a renal hypertensive model (15). Finally, Nosaka and Okamoto (19) have made similar observations in the spontaneously hypertensive rat (SHR). Alterations in baroreflex function in the SHR model have subsequently been studied by a number of investigators. Many of these studies have focused on afferent nerve activity, central resetting or altered sympathetic nervous system traffic. Sapru and Wang (20) found that treatment of SHR with antihypertensives led to a reversal of both the hypertension and the baroreceptor resetting present in untreated SHR. The fact that afferent nerve activity was decreased in hypertrophied aortas led the authors to conclude that such hypertrophy might be an important determinant of impaired baroreflex function. Using baroreceptor deafferentation and direct sympathetic nervous system recording, Judy and Farrell (9) concluded that central resetting rather than baroreceptor dysfunction was an important mechanism of the hypertension in the SHR. Luft et al. (14) found that the heart rate response of SHR was less sensitive to pressor and depressor substances than in WKY. Since splanchnic nerve
1This research was supported by HL-01395 and HL-19680, both to J.E.L. 2Present address: Department of Psychology, Drake University, Des Moines, IA 50311. 3Present address: Depamnent of Physical Education and Recreation, Miami University, Oxford, OH 45056. 539
540
LAWLER, SANDERS, COX AND O'CONNOR
have investigated whether chronic stress can produce an alteration in baroreflex function. Part of the reason why chronic stress has not been investigated may be related to the failure of most stress studies to produce chronic alterations in blood pressure. In the last decade, we have developed a new animal model, the borderline hypertensive rat (BHR), which shows significant elevations in blood pressure in response to chronic stress (11) or dietary salt intake (12). Now that a model for chronic stress exists which shows large, permanent changes in blood pressure (13), studies can be performed which examine the role of altered baroreflex sensitivity in the observed blood pressure changes. In the present study, male BHR were subjected to either 5 or 11 weeks of chronic, daily, tail-shock stress. Intravenous bolus injections of angiotensin II (6) in the conscious, resting state were used to assess the sensitivity of the baroreflex in these groups compared to unstressed, control BHR. METHOD
Animals
Forty-eight mate BHR were used in this study. Inbred parent strains of female SHR and male WKY were purchased from Taconic Farms (Germantown, NY). Before breeding, indirect measurements of systolic blood pressure using tail plethysmography were obtained to determine that SHR and WKY were indeed hypertensive and normotensive, respectively. Male BHR offspring were weaned at 21 days, housed two to a cage, and maintained on laboratory rat chow and tap water ad lib throughout the remainder of study. Beginning at 8 weeks of age, all male BHR were gentled daily for 5 minutes. Weekly tail-cuff systolic blood pressures were also obtained in conscious, mildly restrained animals. The mean of 7 artifact-free determinations, taken once weekly in each animal, was defined as the systolic pressure for the week.
TABLE 1 PEARSON r CORRELATIONS FOR MEAN ARTERIAL BLOOD PRESSURE AND INTERBEAT INTERVAL FOR 10 SUCCESSIVE BEATS IMMEDIATELY PRECEDING THE MAXIMUM BLOOD PRESSURE RESPONSE TO VAR1OUS IV DOSAGES OF ANGIOTENSIN II IN UNSTRESSED ANIMALS (CONT), ANIMALS STRESSED FOR 5 WEEKS (EARLY) AND ANIMALS STRESSED FOR 11 WEEKS (LATEI
Cont Early Late
Sal
2 ng
4 ng
6 ng
20 ng
40 ng
60 ng
0.03 -0.57 -0.26
0.45 0.18 0.18
0.30 -0.12 0.52
0.87* -0.11 {).64
0.96* 0.68 0.62
0.80* 0.8t* 0.78* 0.85* (I.60 0.70*
*Depicts significantcorrelationsat p<0.025.
dosage. A Pearson r was then determined for the t0 mean pressures and their 10 corresponding interbeat intervals in each group and at each dosage. Since this resulted in 21 correlations (3 groups for 7 different dosages), a more stringent p-value (0,025) was chosen for significance. These data were also subjected to regression analysis for determination of slope and intercept. Additional analyses utilized plots of the average peak mean arterial pressure versus the average peak interbeat interval for each dosage of All in each of the three groups. For each dosage, the peak mean pressure was determined for each animal. The longest interbeat interval which occurred during the next three beats was taken as the IBI value for that dose of All. The peak mean arterial blood pressure and peak IBI were obtained for each animal and for each dose and averaged for the group. A line of best fit was then determined for each group. This line depicts the response of one group for the seven data points composed of saline and 2, 4, 6, 20, 40 and 60 ng/kg All. An a priori decision was made to compare the 5- and 1l-week groups to the control group. RESULTS
Procedure
At 10 weeks of age, animals were divided into 3 groups: an Early group (stressed for 5 weeks), a Late group (stressed for 11 weeks), and an unstressed control group which was age matched to the stress groups. Stressed groups received daily, two-hour sessions of pseudorandom, cutaneous tall shock (t mA, 1-s duration, averaging every 30 s). The presentation of all shocks was under microcomputer control. After the appropriate duration of stress (or at the appropriate age in control rats), femoral artery catheters were implanted under ether anesthesia. Following a 3-day recovery period, the blood pressure and heart rate responses to bolus injections of angiotensin II were determined. Dosages (0, 2, 4, 6, 20, 40, or 60 ng/kg AlI, IV) were randomly administered, with a minimum 5-minute recovery period between injections. All data were fed on-line to an IBM personal computer and 8-charmel analog to digital converter. Software stored the mean pressure and interbeat intervals for 90 consecutive beats for each injection. Raters blind to the group assignment and dosage then chose the highest pressure reached in the 90-beat interval. The mean pressure and interbeat intervals for that beat, and the 9 preceding beats, were determined for each animal at each dosage. Thus, each animal had 10 mean pressures and 10 corresponding interbeat intervals for each of 7 different All injections. Each peak pressure was averaged with all the other peak pressures for all BHR in a given group and a given AII dosage. In a similar manner, all pressures and interbeat intervals one beat before the highest pressure were averaged, etc., until 10 pressure and interbeat interval values were determined for a given group and a given AII
Table 1 depicts the Pearson r for each All injection dosage in each of the 3 groups. An asterisk depicts significant correlations. As can be seen, control animals had significant correlations between pressure and interbeat interval for all dosages of 6 ng/kg or greater. Animals stressed for 5 weeks, however, had significant correlations only at the 40 and 60 ng/kg dosages, while animals stressed for 11 weeks had a significant correlation only at the highest dosage tested. Since the regression line is not meaningful in the absence of a significant correlation, the regressions could only be compared between groups at one All dosage (60 ng/kg), since this was the only dosage which yielded significant correlations for all 3 groups. Figure 1 depicts the regression lines for the 10 data points in control animals (triangles and dotted line), early stress animals (open circles and solid line), and late stress animals (closed circles and dashed line). The steepest slope (most sensitive lengthening of interbeat interval for a given rise in blood pressure) was found in the unstressed animals, while animals stressed for 11 weeks showed the shallowest slope. The regression equation for each of the groups was: IBI = 142.8 + 0.336 × BP (control); IBI = 142.7 + 0.278 x BP (early); and IBI = 149.5 + 0.263 × BP (late). Figure 2 compares the maximum pressor and IBI responses to each dose of AII in control and 5-week groups. These data, as do those in Fig. 1, suggest that there is no change in the sensitivity of the baroreflex when BHR stressed for 5 weeks are compared with controls. Figure 3, on the other hand, depicts the same data for control and 11-week groups. These data suggest a reduced sensitivity of the baroreflex in BHR stressed for 11 weeks com-
BAROREFLEX IN STRESSED RATS
541
BP-IBI RESPONSE TO ANG II
210
230 [
CONT 200 ~
.''''"
~
......it ~ .
|
£
LATE EARLY
190
200 : _m 180-
190 ~
180 170 120
I 130
I 140
I 150
I 160
I 170
I 180
170 160
AR]ERL,M. BP (mmHg)
120
FIG. 1. Regression lines relating pressure and interbeat interval in unstressed (CONT), 5-week-stressed(EARLY), and 11-week-stressed(LATE) animals in response to a 60 ng/kg IV bolus injection of angiotensin II. pared to control BHR. A striking resemblance is seen between the data depicted in Fig. 1 and those in Figs. 2 and 3, even though they represent two different ways of depicting the sensitivity of the baroreflex in this study. DISCUSSION Changes in baroreflex function have been noted in a number of animal models of hypertension (15, 16, 22). To the best of our knowledge, the present study is the first to examine whether chronic stress alters the sensitivity of the baroreflex. The data presented, although preliminary in nature, are nevertheless consistent in that BHR subjected to 5 weeks of stress have fairly minor changes in baroreflex function compared with BHR subjected to 11 weeks of stress. All three figures presented are consistent with the hypothesis that chronic stress in a genetically predisposed animal produces a gradual shift in the sensitivity of the baroreflex. In fact, the pattern observed in the present study has been found in other models of hypertension. For example, hypertenBP-IBI RESPONSE TO ANG II
2 o|
j
220*
.E
I OCON
|
160 L
120
/ .
! ~
.
130
.
.
14.0
.
150
160
I
I
170
180
190
MEAN BP (mm H9)
FIG. 2. Regression lines relating pressure and interbeat interval in unstressed (open circles) BHR and in BHR stressed for 5 weeks (f'dled circles). Each data point represents the mean ( __.SEM) for all the animals in a given group at one dosage of All.
I
t
I
I
I
!
130
140
150
160
170
180
190
MEAN BP (ram H9)
FIG. 3. Regression lines relating pressure and interbeat interval in unstressed (open circles) BHR and in BHR stressed for 11 weeks (filled circles). Each data point represents the mean ( ---SEM) for all the animals in a given group at one dosage of AII. sion of short duration results in a shift of the baroreflex curve to the right without a change in gain (21). This change in set point means that the reflex is not triggered until a higher pressure is reached. This is exactly what was seen in the BHR stressed for 5 weeks when compared to control animals. As hypertension persists, the sensitivity (gain) of the reflex is compromised (2). Thus, for a given change in blood pressure, a smaller change in heart rate is seen. This is exactly what was seen in the BHR stressed for 11 weeks when compared to control animals. Numerous investigators have debated the mechanisms responsible for the observed changes in baroreflex function. Some approaches emphasize vessel mechanics. In this view, alterations in distensibility of the vessel wall result in less mechanical distortion of the baroreceptors. Such changes have been noted in the SHR (1). Other approaches emphasize the role of the central nervous system. The nucleus and tractus solitarius (NTS), the point of termination of afferent fibers of the baroreflex, has been the primary focus of studies into central control of the baroreflex. Buchholz and Nathan (3) observed that lesions of the NTS produced chronic lability of blood pressure and eliminated the reflex bradycardia to phenylephrine. Local injection of an All antagonist into the NTS increased the sensitivity of the baroreflex (4). Finally, microinjection of vasopressin into the NTS reduced the sensitivity of the baroreflex (17). Other areas of the CNS also have an effect on the baroreflex. The most relevant to studies of chronic stress is the observation that direct stimulation of hypothalamic areas that lead to a defense reaction can override the baroreflex (7). If future work with stress in the BHR suggests an involvement of the CNS instead of peripheral adaptations, then it would seem that the hypothalamus would be an important structure on which to focus. Recent research from our laboratory (18) demonstrates that changes in hypothalamic levels of norepinephrine occur, but only after stress has been continued for many weeks. Whether or not there is a link between hypothalamic levels of norepinephrine and CNS overriding of the baroreflex awaits further research.
REFERENCES 1. Andresen, M. C.; Kuraoka, S.; Brown, A. M. Baroreceptor function and changes in strain sensitivity in normotensive and spontaneously hypertensive rats. Circ. Res. 47:821-828; 1980.
2. Brown, A. M. Receptors under pressure: An update on baroreceptors. Circ. Res. 46:1-10; 1980. 3. Buchholz, R. A.; Nathan, M. A. Chronic lability of the arterial blood
542
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
LAWLER, SANDERS, COX AND O'CONNOR
pressure produced by electrolytic lesions of the nucleus tractus solitarii in the rat. Circ. Res. 54:227-238; 1984. Campagnole-Santos, M. J.; Diz, D. I.; Ferrario, C. M. Baroreceptor reflex modulation by angiotensin II at the nucleus tractus solitarii. Hypertension 1 l(Suppl. I): 167-171; 1988. Cheng, S. W. T.; Swords, B. H.; Kirk, K. A.; Berecek, K. H. Baroreflex function in lifetime-captopril-treated spontaneously hypertensive rats. Hypertension 13:63-69; 1989. Coleman, T. G. Arterial baroreflex control of heart rate in the conscious rat. Am. J. Physiol. 238:H515-H520; 1980. Gebber, G. L.; Snyder, D. W. Hypothalamic control of baroreceptor reflexes. Am. J. Physiol. 218:124-131; 1970. Gonzalez, E. R.; Krieger, A. J.; Sapru, H. N. Central resetting of baroreflex in the spontaneously hypertensive rat. Hypertension 5:346352; 1983. Judy, W. V.; Farrell, S. K. Arterial baroreceptor reflex control of sympathetic nerve activity in the spontaneously hypertensive rat. Hypertension 1:605-614; 1979. Krieger, E. M. Neurogenic mechanisms in hypertension: Resetting of the baroreceptors. Hypertension 8(Suppl. I):7-14; 1986. Lawler, J. E.; Barker, G. F.; Hubbard, J. W.; Allen, M. T. The effects of conflict on tonic levels of blood pressure in the genetically borderline hypertensive rat. Psychophysiology 17:363-370; 1980. Lawler, J. E.; Sanders, B. J.; Chen, Y.-F.; Nagahama, S.; Oparil, S. Hypertension produced by a high sodium diet in the borderline hypertensive rat. Clin. Exp. Hypertens. A9:1713-1731; 1987. Lawler, J. E.; Barker, G. F.; Hubbard, J. W.; Schaub, R. G. Effects of stress on blood pressure and cardiac pathology in rats with borderline hypertension. Hypertension 3:496-505; 1981. Luft, F. C.; Demmert, G.; Rohmeiss, P.; Unger, T. Baroreceptor
15. 16. 17. 18. 19. 20. 21.
22. 23.
reflex effect on sympathetic nerve activity in stroke-prone spontaneously hypertensive rats. J. Auton. Nerv. Syst. 17:199-209; 1986. McCubbin, J. W.; Green, J. H.; Page, I. H. Baroreceptor function in chronic renal hypertension. Circ. Res. 4:205-210; 1956. Matsuguchi, H.; Schmid, P. G. Pressor response to vasopressin and impaired baroreflex function in DOC-salt hypertension. Am. J. Physiol. 242:H44-H49; 1982. Michelini, L. S.; Bonagamba, L. G. H. Baroreceptor reflex modulation by vasopressin microinjected into the nucleus tractus solitarii of conscious rats. Hypertension ll(Suppl. I):75-79; 1988. Mitchell, V. P.; Lawler, J. E. Norepinephrine content of discrete brain nuclei in acutely and chronically stressed borderline hypertensive rats. Brain Res. Bull. 22:545-547; 1989. Nosaka, S.; Omamoto, K. Modified characteristics of the aortic baroreceptor activities in the spontaneously hypertensive rat. Jpn. Circ. J. 34:685-693; 1970. Sapru, H. N.; Wang, S. C. Modification of aortic baroreceptor resetting in the spontaneously hypertensive rat. Am. J. Physiol. 230:664674; 1976. Sleight, P.; Robinson, J. L.; Brooks, D. E.; Rees, P. M. Characteristics of single carotid sinus baroreceptor fibers and whole nerve activity in the normotensive and the renal hypertensive dog. Circ. Res. 41:750-758; 1977. Su, D. F.; Cerutti, C.; Barres, C.; Vincent, M.; Sassard, J. Blood pressure and baroreflex sensitivity in conscious hypertensive rats of Lyon strain. Am. J. Physiol. 251:H1111-HI 117; 1986. Thames, M. D. Arterial baroreflexes in hypertension. In: Guthrie, G. P.; Kotchen, T. A., eds. Hypertension and the brain. Mt. Kisco, NY: Futura; 1984.