1838 Original article
Enhanced responses to ganglion blockade do not reflect sympathetic nervous system contribution to angiotensin II-induced hypertension John-Luis Morettia, Sandra L. Burkea, Roger G. Evansb, Gavin W. Lamberta and Geoffrey A. Heada,c Objective We examined whether a specific increase in sympathetic nervous system (SNS) activity accounts for the enhanced depressor response to ganglion blockade in angiotensin II (AngII)-induced hypertension in rabbits or whether it reflects a general increased sensitivity of arterial pressure to vasodilatation.
absent. In anaesthetized rabbits, methoxamine-induced decreases in hindlimb vascular conductance were greater in hypertensive than normotensive rabbits suggesting the presence of vascular hypertrophy of sufficient magnitude to explain increased responses to ganglion blockade and vasodilators.
Methods Rabbits were renal denervated or sham-operated and 2 weeks later AngII (50 ng/kg per min) infusion commenced. Mean arterial pressure (MAP) responses to ganglion blockade (pentolinium) and vasodilators nitroprusside and adenosine were measured 2–4 weeks later.
Conclusion Enhanced depressor responses to ganglion blockade in AngII hypertension do not reflect augmented SNS activity, but rather, augmented sympathetic vasoconstriction mediated by a vascular amplifier effect. J Hypertens 27:1838–1848 Q 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins.
Results Basal MAP was 74 W 2 mmHg and maximum hypotensive responses to pentolinium, nitroprusside and adenosine were S17 W 2, S17 W 1 and S21 W 2 mmHg. AngII increased MAP similarly in intact and renal denervated rabbits (R25 W 4 mmHg and R31 W 4 mmHg, respectively). In intact rabbits, depressor responses to pentolinium were augmented by 75% during AngII infusion but responses to vasodilators also increased by 73–106% suggesting general augmentation of vascular reactivity rather than a specific increase in SNS neural activity. Consistent with this notion, total noradrenaline spillover was similar in normal and AngII-treated rabbits. In renal denervated rabbits, AngII enhanced depressor responses to vasodilators but not pentolinium, suggesting that sympathetic activity may be reduced by AngII hypertension when renal nerves are
Journal of Hypertension 2009, 27:1838–1848
Introduction Elevated sympathetic activity has long been associated with the development of essential hypertension [1] but whether this is also the case in hypertension induced by chronic infusion of angiotensin II (AngII) has yet to be established. There are two lines of evidence supporting a greater role for the sympathetic nervous system (SNS) in AngII-induced hypertension. The first is that responses to ganglionic blockade and sympatholytic agents are enhanced in AngII-induced hypertension in rats [2,3] and rabbits [4,5]. The second is that prior renal denervation has been observed to delay or prevent the development of this form of hypertension, particularly in the case of experimental models in which AngII infusion is accompanied by high salt intake [6–8] or uninephrectomy 0263-6352 ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins
Keywords: AngII hypertension, blood pressure, ganglion blockade, heart rate, rabbits, renal denervation, sympathetic nervous system Abbreviations: AngII, angiotensin II; HR, heart rate; MAP, mean arterial pressure; MF, mid frequency power; NA, noradrenaline; PRA, plasma renin activity; SNS, sympathetic nervous system a Neuropharmacology Laboratory, Baker IDI Heart and Diabetes Institute, P.O. Box 6492 St Kilda Road Central, Melbourne, bDepartment of Physiology, Monash University and cDepartment of Pharmacology, Monash University, Clayton, Victoria, Australia
Correspondence to Professor Geoffrey A. Head, Baker IDI Heart and Diabetes Institute, Commercial Road Prahran, P.O. Box 6492, St Kilda Road Central, Melbourne, Victoria, 8008, Australia Tel: +61 3 8532 1332; fax: +61 3 8532 1100; e-mail:
[email protected] Received 1 August 2008 Revised 2 April 2009 Accepted 5 May 2009
[9,10]. These latter observations have been used to support the concept that activation of renal sympathetic nerve activity (RSNA) plays a particularly important role in the development of AngII-dependent hypertension. Clearly, there are a number of underlying assumptions in this argument. The observation of enhanced depressor responses to ganglion blockade in AngII-induced hypertension must be interpreted with care. This could be attributable to activation of sympathetic vasomotor drive but might also be due to increased responsiveness of the vasculature to sympathetic neurotransmitters or even an altered balance between the contributions to maintenance of mean arterial pressure (MAP) of cardiac output and total peripheral DOI:10.1097/HJH.0b013e32832dd0d8
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Sympathetic drive in AngII hypertension Moretti et al. 1839
resistance. Furthermore, the development of vascular hypertrophy would also be expected to enhance responses to ganglion blockade through operation of the vascular amplifier [11]. Thus normal levels of sympathetic activity to arteries would cause greater degrees of vasoconstriction, which is effectively seen as increased ‘SNS vasomotor activity’. This phenomenon is not specific to the SNS as the hypertrophy or remodelling, or both of the resistance vasculature can amplify the effects of both vasoconstrictor and vasodilator agents on peripheral resistance, and attenuate their effects on vascular conductance. In the intact animal this translates to enhanced responses of MAP to vasoactive agents [11]. Such a phenomenon would include responses to ganglion blockade. To examine the potential role of the vascular amplifier mechanism in the enhanced response to ganglion blockade in AngII-induced hypertension, it is necessary to test whether the enhancement is specific for treatments that blunt sympathetic neural activity, as opposed to treatments that directly dilate the resistance vasculature. Augmentation of the depressor response to ganglion blockade has often been interpreted to indicate an enhanced role of the SNS in maintenance of arterial pressure in AngII-induced hypertension [3,4,12]. Indeed Li et al. [3] suggested that there are initial AngII-dependent actions to decrease excretion of salt and water but that there is subsequent activation of sympathetic pathways. Thus there has been a clear suggestion that increased SNS activity is the mechanism. However, in none of these studies were responses to ganglion blockade compared with responses to directly acting vasodilator agents. Recent observations have also called into question roles for activation of the SNS in general, or more specifically the renal sympathetic nerves, in AngII-induced hypertension, at least in rabbits with both kidneys intact and on a normal salt diet. With respect to the renal nerves, we and others have been unable to detect differences in RSNA between rabbits with 4–6 week established AngII-induced hypertension [12], 2K1C hypertension [13], and renal wrap hypertension [14] compared with normotensive control rabbits. However, there is evidence of an initial period of renal sympathoinhibition during the early stages of AngII-induced hypertension in both rabbits [13,15] and dogs [16]. Furthermore, as chronic renal denervation did lower arterial pressure in rabbits, this effect was certainly not greater in rabbits with AngII-induced hypertension than in normotensive controls [12]. Collectively, these data do not support a role of the renal sympathetic nerves in the maintenance of AngII-induced hypertension in animals with normal dietary salt intake and both kidneys intact. But they also do not exclude the possibility that the renal nerves contribute to the development of this form of hypertension, which can only really be tested by examining the
effects of renal denervation prior to the commencement of AngII treatment. Such studies have previously been performed in rabbits under conditions of high dietary salt intake [8], but not under conditions of normal dietary salt intake. Thus the current study had three major aims. The first was to determine whether prior renal denervation could alter the time course or magnitude of the development of AngII-induced hypertension. This experiment would determine whether there is a specific contribution of the renal nerves to the development of AngII-induced hypertension. The second aim was to measure whole body noradrenaline spillover to determine whether there is evidence for a general increase in sympathetic nerve activity in AngII-induced hypertension. The third aim was to determine whether depressor responses to the directly acting vasodilator agents sodium nitroprusside and adenosine, like those to the ganglion blocker pentolinium, are enhanced in AngII-induced hypertension. Such a finding would be consistent with the notion of a vascular amplifier phenomenon, rather than activation of sympathetic neural activity, accounting for the enhanced depressor response to ganglion blockade in AngII-induced hypertension. To confirm that the vascular amplifier phenomenon is present in rabbits with established AngII-induced hypertension, we also tested responses of MAP, hindlimb blood flow and hindlimb vascular resistance and conductance to systemic infusion of vasoactive agents, according to established techniques [11,17].
Methods Thirty-four New Zealand White rabbits (2.3–3.2 kg) were used in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. Rabbits were fed a normal pellet diet (0.21% sodium) supplemented with daily vegetables. This level of sodium is considered normal for laboratory rabbits [18]. Experimental design
Rabbits were randomized to one of four groups: renal denervation and AngII treatment (n ¼ 6); renal denervation and vehicle treatment (n ¼ 5); sham denervation and AngII treatment (n ¼ 7); sham denervation and vehicle treatment (n ¼ 3). Cardiovascular function was studied in each conscious rabbit on four occasions; before and 2 weeks after surgery for renal or sham denervation [12], and 2 and 4 weeks after subsequently commencing an infusion of AngII (50 ng/kg per min s.c.) or isotonic saline (0.9% NaCl) vehicle via an osmotic minipump [12]. MAP and heart rate (HR) were also determined 5 weeks after commencing infusions. For assessing cardiovascular function, blood (1 ml) was first collected from an ear artery for plasma renin activity
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1840 Journal of Hypertension
2009, Vol 27 No 9
(PRA) determination [19]. Baseline measurements of MAP and HR were made over 60 min via an ear artery [12] before the vasodilators sodium nitroprusside (20, 63 and 200 mg/kg per min; Fluka AG, Buchs, Switzerland) and adenosine (100, 300 and 1000 mg/kg per min; Sigma, St Louis, Missouri, USA) were infused via an ear vein for 3 min at each dose. The agents were administered in random order, 60 min apart. After a further 60 min, the ganglion blocker pentolinium (6 mg/kg i.v., Sigma) was administered. Another experiment was conducted in conscious rabbits to examine the effects of acutely increasing MAP in rabbits with normal vasculature (vehicle treated rabbits, n ¼ 3) and of acutely lowering MAP in rabbits with chronic AngII-induced hypertension (n ¼ 4). In vehicletreated rabbits, AngII (50 ng/kg per min i.v.) was infused for 30–60 min before treatment with pentolinium and vasodilators as described earlier. Four weeks after commencing AngII treatment, candesartan (10 mg/kg and 10 mg/kg per h i.v.) was infused for 30 min before treatment with vasodilators and pentolinium as described above. Thus, normotensive rabbits were acutely made hypertensive and hypertensive rabbits were acutely made normotensive. A final experiment was conducted 4–5 weeks after commencing infusion of AngII (n ¼ 4) and vehicle (n ¼ 4) in anaesthetized (sodium pentobarbitone, 90– 150 mg/kg plus 30–50 mg/h) rabbits with intact renal nerves. Hindlimb blood flow was measured via a transit-time ultrasound flow probe (2SB; Transonic Systems Inc, Ithaca, New York, USA) placed around the femoral artery. MAP was measured via an ear artery and drugs were administered via an ear vein. Rabbits were pretreated with candesartan (10 mg/kg and 10 mg/kg per h i.v.), pentolinium (6 mg/kg i.v.) and an infusion of sodium nitroprusside (2 1 mg/kg per min) titrated to reduce MAP to approximately 40 mmHg. After MAP had been stable for 5 min, ascending doses of methoxamine (1, 3, 10, 30 and 60 mg/kg per min, Sigma) were administered, each over 6 min.
described [21]. Whole body (total) noradrenaline spillover was calculated as: Total noradrenaline spillover ¼ noradrenaline clearance noradrenaline arterial concentration where noradrenaline clearance ¼ [3H] noradrenaline infusion rate/arterial [3H] noradrenaline concentration. Data analysis
MAP and HR derived from the pressure pulse were digitized and averaged over 2 s. The control period before administration of vasodilators or ganglion blockade in conscious rabbits was measured over 15 min. Responses to vasodilators and ganglion blockade were measured over 1 min during the maximum fall in MAP. For the normalization of the response to ganglion blockade and for the assessment of the effects of acutely altering arterial pressure, the depressor responses to the two vasodilators were averaged. Diedrich et al. [22] have developed a methodology, which uses ganglionic blockade to determine the contribution of the SNS to blood pressure. This technique involves combining the fall in blood pressure (on the Y axis) with the change in low frequency power (on the X axis) before and after several doses of ganglionic blockade. The slope of the resulting line reflects the contribution of activity of the SNS to blood pressure and was shown to be superior to ganglion blockade or power spectral analysis alone [22]. We have adapted this technique for the rabbit using the single dose of pentolinium and the corresponding ‘autonomic’ frequency band for the rabbit which is the mid frequency (MF, 0.2–0.4 Hz) as described previously [23]. Responses to methoxamine in anesthetized rabbits were averaged over the final minute of each infusion. Hindlimb vascular resistance was calculated as MAP/hindlimb blood flow, and hindlimb vascular conductance as its reciprocal. Statistical analysis
At the end of the final experiment, the rabbit was euthanized and kidneys and left ventricle were removed and weighed. Kidneys were snap frozen in liquid nitrogen and stored at 808C for later determination of noradrenaline content by HPLC [20]. In a separate group of animals that received either no treatment (n ¼ 5) or AngII (n ¼ 4) infusion, total noradrenaline spillover was determined from a 2 ml blood sample. The method involved infusing intravenously ring labelled [3H]-noradrenaline (Perkin Elmer) at 90 nCi/kg per min. Catecholamines were extracted from plasma using alumina adsorption and quantified using HPLC with colorimetric detection as previously
Values are expressed as mean standard error of the mean (SEM) or mean difference standard error of the difference (SED). Repeated measures analysis of variance provided within-animal (effects of denervation, of AngII treatment and of methoxamine) and betweenanimal (group) contrasts. One-way analysis of variance was used for data collected at a single time-point. Type 1 error was controlled using Bonferroni and GreenhouseGeisser corrections [24].
Results Effects of renal denervation
Prior to renal denervation or sham surgery, MAP and HR averaged 74 1 mmHg and 187 6 beats/min,
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Sympathetic drive in AngII hypertension Moretti et al. 1841
Baseline mean arterial pressure and changes in responses to ganglion blockade with pentolinium and vasodilatation with adenosine and sodium nitroprusside before and after renal denervation or sham surgery
Table 1
Basal MAP (mmHg) Pentolinium DMAP (mmHg) Adenosine DMAP (mmHg) Nitroprusside DMAP (mmHg)
73.6 18.1 21.5 17.9
Renal denervated 66.2 19.5 20.1 22.0
SEM
P
Basal
Sham operated
SEM
P
1.7 2.3 1.8 1.8
MM
74.0 15.6 20.1 17.0
70.9 17.7 21.8 17.7
1.4 1.9 2.0 1.4
NS NS NS NS
NS NS NS
Values are mean SEM and mean difference SED using within animal variance. MM P < 0.01, NS P > 0.05 for comparisons between basal responses and those measured after renal denervation or sham surgery. MAP, mean arterial pressure.
respectively, across all rabbits. MAP had reduced by 7 2 mmHg (P < 0.01) 2 weeks after renal denervation compared with 3 2 mmHg (P > 0.05) 2 weeks after sham denervation (Table 1). By contrast, basal HR was not significantly altered by sham surgery or renal denervation (data not shown). Renal denervation also reduced PRA by 67% compared with levels in the sham-operated rabbits, (4.9 0.4 ng Ang I/ml per h, P < 0.001). Renal denervation did not alter the size of either the left ventricle or the kidneys. Noradrenaline levels in denervated kidneys were 7% those of innervated kidneys (88 9 ng/g tissue weight vs. 6 1 ng/g, P < 0.001) but were not altered by AngII infusion. Effects of chronic AngII infusion
Treatment with AngII or vehicle began 2 weeks after renal denervation or sham surgery, and continued for 5 weeks. In sham-operated rabbits, MAP averaged over weeks 2–5 of AngII treatment was 25 4 mmHg mmHg higher than the pretreatment level (P < 0.001, Fig. 1). Treatment with AngII also markedly reduced PRA (84%, P < 0.001). AngII treatment also increased MAP in renal denervated rabbits (by 31 4 mmHg; P < 0.001), an effect that was indistinguishable from that in rabbits with intact renal nerves (P ¼ 0.3 for between group comparison, Fig. 1). Vehicle treatment did not significantly alter MAP in either denervated or sham denervated rabbits (Fig. 1). Left ventricular weight from AngII treated rabbits was 21% greater than that of vehicle treated rabbits (1.4 0.1 g/kg and 1.1 0.0 g/kg, respectively, P < 0.01) and kidney weight was 18% greater in those animals (2.6 0.1 g/kg vs. 2.2 0.1 g/kg, respectively, P < 0.01).
7 1 mmHg, 23 2 mmHg and 22 2 mmHg at doses of 20, 63 and 200 mg/kg per min, respectively. Increasing doses of adenosine evoked falls in MAP of 1 1 mmHg, 5 1 mmHg and 21 2 mmHg. To match the magnitude of depressor responses to the vasodilator agents with those to pentolinium, we averaged responses to sodium nitroprusside over all three doses but only used the response to the greatest dose of adenosine (1000 mg/kg per min). The average depressor response to three increasing doses of nitroprusside was 17 2 mmHg, whereas the average response to 1000 mg/kg per min of adenosine was 21 2 mmHg. The magnitude of the responses to sodium nitroprusside and adenosine did not differ significantly from the response to pentolinium (P ¼ 0.10).
Fig. 1
Noradrenaline spillover
Arterial levels of noradrenaline, clearance of noradrenaline and total noradrenaline spillover were similar in control and AngII hypertensive rabbits (Table 2). Responses to pentolinium, sodium nitroprusside and adenosine
As we have found previously [12], the maximum fall in MAP was observed between 3.5–4.5 min after pentolinium administration. Prior to renal denervation or sham surgery, this averaged 17 2 mmHg across all 21 rabbits. The depressor responses to nitroprusside were
Mean arterial pressure (MAP, mmHg) and heart rate (HR, beats/min) measured before (basal) and after sham surgery (grey line) or renal denervation (black line) and after 2, 4 and 5 weeks (W) of treatment with angiotensin II (solid lines) or vehicle (dashed lines) in conscious rabbits. Error bars are SEM indicating between animal variance.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1842 Journal of Hypertension
2009, Vol 27 No 9
Arterial levels of noradrenaline, total noradrenaline clearance and total noradrenaline spillover in control rabbits and rabbits with hypertension induced by chronic infusion of angiotensin II
Table 2
Arterial NA (pg/ml) NA Clearance (ml/min) Total NA spillover (ng/min) N
Control
ANGII
Control SEM
Control SEM
47.4 6.4 385.0 46.6 18.9 4.5 5
67.4 18.9 291.3 40.0 21.5 7.9 4
Values are mean SEM, n, is the number of rabbits per group. AngII, angiotensin II; NA, noradrenaline.
Responses to pentolinium, adenosine and nitroprusside were not significantly altered by either renal denervation or sham surgery. Furthermore, when tested 2 weeks after surgery these responses were indistinguishable in sham compared with renal-denervated rabbits (Table 1).
Relationship between decreases in spectral power and blood pressure during ganglion blockade
We determined the slopes of the relationship between the reduction in MF power on the X-axis and the reduction in MAP on the Y-axis produced by pentolinium. This analysis was performed in individual rabbits before and after sham surgery or renal denervation and also 2 and 4 weeks after commencing AngII infusion. This slope has been suggested in human studies to be a better index of the contribution of sympathetic activity to maintenance of arterial blood pressure than ganglionic blockade alone [22]. The slope of this relationship
Fig. 2
Responses to ganglion blockade and vasodilators were measured 2 and 4 weeks after beginning treatment with AngII or vehicle. In sham-operated rabbits, the depressor response to pentolinium increased by 75% during AngII infusion (averaged over weeks 2 and 4) from 14 2 to 25 3 mmHg (P < 0.05, Fig. 2). By contrast, in renal denervated rabbits, the response to pentolinium did not significantly change over the same period (Fig. 2). The depressor responses to both nitroprusside and adenosine in sham-operated rabbits also increased markedly during AngII treatment, by 106% (P < 0.001) and 73% (P < 0.01), respectively (Fig. 2). This increased response was seen at all three doses of nitroprusside (Fig. 2). Unlike the response to pentolinium, in the renal denervated rabbits the depressor responses to nitroprusside and adenosine were also increased during AngII infusion (þ107%, P < 0.001 and þ88%, P < 0.01 respectively, Fig. 2). This increased response occurred at the two highest doses of nitroprusside (Fig. 2). Vehicle treatment did not significantly alter depressor responses to ganglion blockade or vasodilators in either renal denervated or sham-operated rabbits (data not shown). Normalization of responses to ganglion blockade
To estimate the contribution of the vasculature to the magnitude of the depressor response to ganglion blockade, we normalized the response to ganglion blockade to that of the vasodilators by dividing the hypotensive response to pentolinium before and 4 weeks after AngII treatment by the equivalent averaged responses to nitroprusside and adenosine. In sham-operated rabbits, the normalized response to pentolinium was similar before (1.14 0.17) and 4 weeks after (1.20 0.14) commencing AngII treatment (P > 0.05, Fig. 3). In renal denervated rabbits, the normalized response to pentolinium was 46% less during AngII treatment than before it commenced (1.30 0.24 vs. 0.70 0.10, P < 0.05, Fig. 3).
Effects of sham surgery (left panels, open symbols) or renal denervation (right panels, closed symbols) on depressor responses to ganglion blockade with pentolinium (upper) and the directly acting vasodilators adenosine (middle) and sodium nitroprusside (lower). Responses were measured 2 weeks after sham or renal denervation (denerv) (pretreatment) and 2 and 4 weeks (W) after commencement of angiotensin II infusion. A single dose of pentolinium (6 mg/kg) and three doses of adenosine of 100 (circles), 300 (squares) and 1000 (triangles) mg/kg per min and three doses of nitroprusside of 20 (circles), 63 (squares) and 200 (triangles) mg/kg per min were given. Error bars are SEM indicating between animal variance. P < 0.05, P < 0.01, P < 0.001 and NS P > 0.05 for comparison of responses at 2 and 4 weeks after treatment with responses measured before treatment. MAP, mean arterial pressure.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Sympathetic drive in AngII hypertension Moretti et al. 1843
Fig. 3
Hindlimb vascular conductance and vascular calibre in anaesthetized rabbits
MAP after treatment with candesartan and pentolinium (before methoxamine infusion) was similar in both groups, but hindlimb blood flow and hence vascular conductance in the AngII treated group was approximately half that of the vehicle treated group (Fig. 6, P < 0.05). Thus vascular resistance was greater in the AngII-treated group (P < 0.05, Fig. 6).
Effects of renal denervation and AngII-induced hypertension on depressor responses to ganglion blockade and directly acting vasodilators. Left panels: average changes in mean arterial pressure (DMAP) in response to ganglion blockade with pentolinium and the vasodilators sodium nitroprusside and adenosine (responses averaged) in sham denervated (upper) and denervated (lower) rabbits before (Pretreat) and 4 weeks (W4) after treatment with angiotensin II (grey bars). Right panels show the ratio of the ganglion blockade response to that of the vasodilators in the same animals. Error bars are SEM indicating between animal variance. P < 0.05 for comparison of responses measured after 4 weeks of angiotensin II treatment with responses measured before treatment commenced.
under basal conditions was 19 6 and 20 4 mmHg/ mmHg2 in the groups that subsequently underwent sham denervation and renal denervation respectively (Fig. 4). Sham denervated animals showed an increased slope during 2 and 4 weeks of AngII infusion (26 4 mmHg/mmHg2, P ¼ 0.02). By contrast, renal denervation and subsequent AngII infusion did not significantly change the slope of the relationship (Fig. 4). Effects of acutely altering mean arterial pressure
Acute infusion of AngII in vehicle treated rabbits increased MAP by 27 2 mmHg from 76 3 mmHg (P < 0.001). In those rabbits, the depressor response to ganglion blockade after acute AngII was indistinguishable from that measured before (Fig. 5, upper left panel). By contrast, the average response to the vasodilators was greater during acute AngII treatment (38 6 mmHg vs. 21 4 mmHg, P < 0.05, Fig. 5, upper right panel). In rabbits with chronic AngII-induced hypertension, acute treatment with candesartan lowered MAP by 13 1 mmHg from 91 4 mmHg (P < 0.001). The depressor response to ganglion blockade (30 3 mmHg) was greater than that measured in the same rabbits before candesartan treatment (17 3 mmHg, P < 0.05) but the average response to the vasodilators was not significantly altered (Fig. 5, lower panel).
Methoxamine produced increases in MAP and hindlimb vascular resistance and falls in hindlimb blood flow and vascular conductance (Fig. 6). Over all doses, MAP was not significantly different in the two groups. However, hindlimb blood flow and vascular conductance were less, over all doses, in the AngII-treated group than in the vehicle group and vascular resistance was greater in the AngII-treated group (all P < 0.001, Fig. 6). Furthermore, the response ranges, from baseline to maximum, of hindlimb vascular conductance were 64% less in the AngII-treated group than the vehicle group, respectively (P < 0.05, Fig. 6) but the response range of vascular resistance was not significantly different between the groups. On the basis of the relationship between vasoconstriction and MAP in the upper panel of Fig. 4, we have calculated the predicted depressor responses to ganglion blockade in AngII-treated rabbits based on their respective basal MAP and the observed 14 mmHg depressor response in the normotensive rabbits (insert, Fig. 6). This was done by assuming the same level of vasodilatation (i.e. same shift along the x axis for both groups). The predicted value of 23 mmHg is closely similar to the observed depressor response of 25 mmHg produced by ganglion blockade in rabbits with AngII-induced hypertension (upper panel Fig. 2). Relationship between basal mean arterial pressure and the depressor response to sodium nitroprusside
An overall regression of the level of basal MAP at the beginning of each experiment with the subsequent decrease in MAP caused by nitroprusside from all rabbits, all experiments, indicated a very strong underlying linear relationship (r ¼ 0.72, df ¼ 104, t ¼ 10.2, P < 0.0001, Fig. 7). Indeed basal MAP explained over 50% of the total variance in the responses. When only the depressor responses were used from those animals with blood pressure acutely altered by candesartan (to normalize blood pressure) or acute Ang II treatment (to acutely raise blood pressure), the correlation was lost (r ¼ 0.26, df ¼ 6, P ¼ 0.6).
Discussion Our current study provides new findings that have important implications for our understanding of the pathogenesis of hypertension during chronic infusion
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1844 Journal of Hypertension
2009, Vol 27 No 9
Fig. 4
Left Panels: mean arterial pressure (MAP) plotted against midfrequency MAP power (MF, 0.2–0.4 Hz) before and after pentolinium in conscious rabbits under basal conditions, and after sham denervation (top left), renal denervation (lower left) and 2 and 4 weeks (W) after commencing angiotensin II infusion. Right Panels: average slope of the relationship between fall in MAP and reduction in MF power calculated for individual animals during basal recordings (B), sham (S) or renal denervation (D) and after 2 and 4 weeks of angiotensin II treatment. Error bars are SEM indicating between animal variance. P < 0.05 for comparison of slopes measured at 2 and 4 weeks after treatment with responses measured before treatment.
of AngII. We found that prior renal denervation did not blunt or delay the development of AngII-induced hypertension, which suggests that renal sympathetic nerves are not obligatory for the development of AngII-induced hypertension. Furthermore, we found that total noradrenaline spillover was similar in control and AngII hypertensive rabbits suggesting that there is no change to overall sympathetic nerve activity. This conclusion is further supported by observations of reduced [15] or normal [12] RSNA, normal whole body noradrenaline spillover and relatively normal renal sympathetic neuroeffector function [25,26] in AngII-induced hypertension in rabbits. Furthermore, in rats on normal salt intake, whole body noradrenaline spillover changes little during the first 2 weeks of AngII-induced hypertension, although it does appear to increase when AngII-induced hypertension is accompanied by high-salt intake [27].
Taken together with our previous studies showing no change in RSNA in AngII hypertensive rabbits [12], it appears that neither renal nor nonrenal sympathetic activity are likely changed in this form of hypertension. A major aspect of our study was to show that the depressor responses to the directly acting vasodilator agents adenosine and sodium nitroprusside were enhanced as much as those to the ganglion blocker pentolinium. The magnitude of the response to ganglion blockade or sympatholytic agents has often been utilized as an indirect measure of sympathetic vasomotor nerve activity [28,29]. Thus the common observation of an enhanced response to ganglion blockade in various models of hypertension has been considered as evidence of increased sympathetic vasomotor nerve activity and therefore used to suggest a ‘neurogenic’ contribution to the development and/or
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Sympathetic drive in AngII hypertension Moretti et al. 1845
Fig. 5
Effects of acutely altering mean arterial pressure (MAP) on depressor responses to ganglion blockade (left panels) and to directly acting vasodilators (right panels). Depressor responses were measured before (open bars) and after (grey bars) intravenous treatment with angiotensin II (AngII, 50 ng/kg per min) to acutely increase MAP in vehicle treated rabbits (upper panels) or with candesartan (10 mg/kg and 10 mg/kg per h) to acutely lower MAP in chronic AngII treated rabbits (lower panels). Note that in the lower panel, the open bars showing responses before acute treatment are on the right of each pair. The numbers inside the bars are the resting MAP preceding the administration of pentolinium or vasodilators. Error bars are SEM indicating between animal variance. P < 0.05 for control response vs. response after treatment. HT, hypertensive; NT, normotensive.
maintenance of hypertension, including AngII-induced hypertension [3,4]. Our current findings show that such conclusions are not always valid, and that depressor responses to vasodilatation per se, regardless of whether they are mediated through withdrawal of sympathetic vasomotor nerve activity or direct actions on vascular smooth muscle, appear to be enhanced in AngII-induced hypertension. Thus enhanced responses to ganglion blockade likely do not reflect an increase in sympathetic nerve activity, but rather a ‘nonspecific enhancement’ of the vasoconstrictor affects the nerve activity. This could be considered as greater ‘sympathetic vasomotor tone’ in AngII hypertension but our findings suggest that the underlying mechanism involves changed vascular structure or possibly greater levels of vasoconstriction. In support of the former, we observed that fully dilated hindlimb vasculature of anaesthetized hypertensive rab-
bits had lower basal levels of conductance and markedly diminished vasoconstrictor-induced reductions in hindlimb conductance compared with normotensive controls at all levels of vascular tone studied. This confirms in our study with AngII-induced hypertension, the presence of vascular hypertrophy, or remodelling or both, which was similar in magnitude to that reported by Wright and Angus [11] using renal cellophane wrap hypertensive rabbits. Importantly, the relatively subtle differences in the relationship between MAP and vasoconstriction, between the normotensive and AngII-treated groups, can completely account for the enhanced response to ganglion blockade in the hypertensive group, as shown by the inset in Fig. 6. These observations support the hypothesis that AngII-induced hypertension is associated with hypertrophy, or remodelling, or both of the arterial resistance vasculature, which could account for the enhanced depressor responses to both ganglion blockade and direct vasodilatation. Collectively, our findings provide strong evidence that the SNS plays little direct role in the development of AngII-induced hypertension in rabbits. It might be suggested that any increase in vasoconstriction, including that caused by increased sympathetic pressor activity, might result in increased depressor responses to vasodilators. Our study showed that increasing vasoconstriction acutely with AngII results in greater depressor responses to vasodilators in normal rabbits with normal vasculature. However, we did not observe any attenuation of depressor responses to dilators in the chronic AngII treated animals when the blood pressure was reduced to normal levels (Fig. 5). We also found a strong linear relationship between the basal MAP and the subsequent depressor response to nitroprusside. However, this relationship was lost when acute changes to MAP were induced, suggesting that vasoconstriction was not the cause. Together, these findings suggest the presence of the vascular amplification rather than vasoconstriction is most likely responsible for the greater depressor responses to dilators in chronically AngII treated rabbits. In the present study we also included a relatively new method of combining the changes in blood pressure with changes to specific frequencies of blood pressure variability that have been shown to be predominantly driven by the SNS. The combination of these two measures has been suggested by Diedrich et al. [22] to be a superior method for determining the contribution of the SNS. This index was increased by 58% in the intact AngII infused rabbits, which was similar to the 75% increase in the response to ganglionic blockade. This observation is hardly surprising, given that both of these indices reflect the integrated vasomotor response to sympathetic drive. Therefore, they should be affected both by changes in SNS activity per se and
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1846 Journal of Hypertension
2009, Vol 27 No 9
Fig. 6
Haemodynamic responses to graded vasoconstriction in anaesthetized rabbits. Baseline values (dose 0) and responses to increasing doses of methoxamine were determined under conditions of ganglion blockade and AT1 receptor blockade. Sodium nitroprusside was infused intravenously to reduce mean arterial pressure (MAP) to approximately 40 mmHg before infusions of methoxamine commenced. Error bars are SEM indicating between animal variance. P < 0.001 for comparison of angiotensin II-treated (closed circles) with vehicle-treated (open circles) rabbits. Inset in top panel shows the predicted falls in MAP produced by ganglion blockade, assuming equal degrees of dilatation in vehicle (open bar) and angiotensin II-treated (solid bar) rabbits, based on the respective relationships between levels of vasoconstriction and MAP in the two groups of rabbits.
Fig. 7
vascular neuroeffector function, including structurally based vascular amplification. Thus, it appears that as this new index of sympathetic nerve activity might be an improvement on the standard approach of quantifying the depressor response to ganglion blockade, it may still be influenced by nonspecific factors such as the vascular amplifier effect. A further consideration in attempting to compare the depressor responses to dilators with those of ganglion blockade is the contribution effect of baroreflexes. However, we have previously shown that there is no change to the gain of the cardiac or sympathetic baroreflex at this time during chronic AngII treatment [12]. Another potentially confounding factor might be differences in haemodynamic responses to vasodilators compared with ganglion blockade. For example, responses to ganglion blockade, but not dilators may involve reduced cardiac output. However, this is also not likely as Wright and Angus [11] showed that the depressor response to ganglion block in renal wrap hypertensive rabbits was due to a reduction in total peripheral resistance with no change to cardiac output.
Linear regression between resting mean arterial pressure (MAP, mmHg) and the magnitude of the depressor response to sodium nitroprusside (SNP) from all rabbits (n ¼ 21), from all times (weeks 0– 4) in the study showing a highly significant, P ¼
