- Email: [email protected]
Changes in regional vascular resistance in response to microinjection of l -glutamate into different antero-posterior coordinates of the RVLM in awake rats
Autonomic Neuroscience: Basic and Clinical 87 (2001) 301–309 www.elsevier.com / locate / autneu
Changes in regional vascular resistance in response to microinjection of L-glutamate into different antero-posterior coordinates of the RVLM in awake rats ´ Patrıcia M. de Paula, Benedito H. Machado* ˜ Preto, University of Sao ˜ Paulo, 14049 -900, Ribeirao ˜ Preto, SP, Brazil Department of Physiology, School of Medicine of Ribeirao Received 30 March 1999; accepted 7 March 2000
Abstract Changes in mean arterial pressure (MAP) and in regional vascular resistance (RVR, hindquarter, mesenteric and renal) induced by microinjection of L-glutamate into three different antero-posterior coordinates of the rostral ventrolateral medulla (RVLM) [1200–1600 mm (n510), 1601–2000 mm (n512) and 2001–2500 mm (n56) rostral to the obex] were investigated in unanesthetized rats. Guide cannulas directed towards the RVLM were implanted 4 days prior to the experiments. Doppler probes were implanted around the superior mesenteric, inferior abdominal aorta and left renal arteries and a catheter was inserted into the femoral artery and vein 1 day prior to the experiments. Insertion of the injector into the RVLM produced an increase in baseline MAP, which was back to control levels 2 min later, when L-glutamate was microinjected. Microinjection of L-glutamate (1 nmol / 30 nl) into the three antero-posterior coordinates of the RVLM produced an increase in MAP associated with a similar increase in hindquarter, mesenteric and renal vascular resistance, which were back to control 1 min later. Saline into the RVLM produced negligible effects on MAP and RVR. These findings suggest that the sympathetic vasomotor neurons involved in the regulation of the regional vascular resistance in rats are not topographically distributed in the antero-posterior coordinates of the RVLM. However, the experimental methods used to evaluate the topographic distribution of sympatho-vasomotor neurons in the RVLM and the measurement of the regional blood flow may not be precise enough to detect any possible differences. 2001 Elsevier Science B.V. All rights reserved. Keywords: Blood flow; Baroreflex; Sympathetic vasomotor neurons; Cardiovascular regulation, Vascular beds, Renal blood flow, Mesenteric blood flow, Hindquarter blood flow
1. Introduction The activation of the baroreceptors produces autonomic adjustments in order to maintain arterial blood pressure within a narrow range of variation. The sympathetic vasomotor tone is generated by neurons located in the rostral ventrolateral medulla (RVLM) and the activation of the baroreceptors produces the inhibitory modulation of these neurons through projections from the nucleus tractus solitarii (NTS) to the RVLM via the caudal ventrolateral medulla (Ross et al., 1984a,b; Willete et al., 1987; Urbanski and Sapru, 1988; Sun, 1995). In recent studies from our laboratory we verified that electrical stimulation of the aortic depressor nerve induced a greater vasodilatation in the hindquarter vascular bed than in other ter*Corresponding author. Tel.: 155-16-602-3015; fax: 155-16-6330017. E-mail address: [email protected] (B.H. Machado).
ritories (Machado et al., 1994; de Paula et al., 1999), supporting our previous finding that bilateral carotid occlusion (BCO) in intact rats produced major vasoconstriction in the mesenteric and renal vascular territories, while the BCO in rats with selective denervation of the aortic nerves produced major vasoconstriction in the hindquarter territory (Machado et al., 1992). Therefore, these data indicate that aortic baroreceptors play a selective and predominant role in the regulation of the hindquarter vascular resistance compared to other territories, suggesting that the activation of this specific subset of baroreceptor afferents produces selective inhibition of sympathetic neurons in the RVLM involved in the innervation of the vasculature of skeletical muscles. Chemical or electrical stimulation of RVLM neurons produced a major pressor response (Ross et al., 1984a,b; Dampney et al., 1985; Willete et al., 1987; Sun, 1995) due to vasoconstriction, which, however, was not generalized for all vascular territories because several lines of ex-
0165-1838 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S1566-0702( 00 )00283-6
302
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
perimental evidence obtained in cats indicate that the sympathetic vasomotor neurons in the RVLM are topographically distributed and that these different groups of neurons present a selective regulation of the vascular resistance of different vascular beds (Willete et al., 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994). Studies by Dean et al. (1992) also support this concept considering that neurons in the RVLM involved in kidney blood flow regulation are located more rostrally than those involved in skeletical muscle blood flow while studies by McAllen and Dampney (1990) have indicated that the sympathetic vasomotor neurons are topographically distributed in accordance with the type of vascular bed but not in relation to the body region. All the evidence presented above suggests that the regulation of regional blood flow is selectively regulated by specific baroreceptor afferent fibers as well as by different subpopulations of the sympathetic vasomotor neurons in the RVLM. However, most of these studies were performed on cats or on anesthetized animals and no studies explored the changes in regional vascular resistance in response to microinjection of L-glutamate into different subregions of the RVLM of conscious freely moving rats. Therefore, the aim of the present study was to determine whether the changes in mean arterial pressure (MAP) in response to L-glutamate microinjection into different antero-posterior coordinates of the RVLM are related to the increase in the resistance of some specific vascular bed. Part of these data have been published previously in Abstract form (de Paula and Machado, 1997).
2. Methods Male Wistar rats weighing 230–270 g were used in the present study. Four days before the experiments bilateral guide cannulas in the direction of the RVLM were implanted under tribromoethanol anesthesia (250 mg / kg, i.p.). One day before the experiments the rats were anesthetized with sodium pentobarbital (40 mg / kg, i.p.) for the implantation of miniaturized Doppler probes around the aorta and mesenteric and renal arteries as well as for catheterization of femoral artery and vein.
2.1. Implantation of the Doppler probes After a medial laparatomy a small 4-mm long segment of the inferior abdominal aorta, superior mesenteric or left renal artery was carefully isolated and probes with a specific internal diameter were placed around each vessel. The two extremities of the silastic probes embracing the vessels were fastened with ophthalmic 6-0 sutures. The wires of the probes were sutured to the internal abdominal wall in order to keep constant the position of the probes in relation to the vessels. After the abdominal suture, the wires were exteriorized subcutaneously on the back of the
rat where the connector was fixed. On the day of the experiments the wires were connected to the Doppler flowmeter (University of Iowa Bioengineering Facility) which in turn was connected to a four-channel recorder (Hewlett-Packard). The flowmeter was used to determine the velocity of blood flow through the changes in the frequency of the piezoelectric crystal (20 MHz, Doppler effect), which, according to Haywood et al. (1981) are directly related to absolute blood flow. In this case the resistance can be calculated as follows: Resistance5MAP (mmHg) / Doppler frequency (Hz), and consequently the changes in the vascular resistance of each vascular bed can be determined as percent change in relation to control. The changes in regional vascular resistance were calculated in each experiment considering the peak of changes in MAP as well as the peak of changes in regional blood flow. After the probe wires were connected to the recorder system, the efficacy of the probes was checked by recording pulsatile flow as well as by the intravenous injection of phenylephrine to produce a vasoconstriction and consequent reduction in blood flow. The tests for the efficacy of the Doppler probes in detecting changes in the blood flow of each vascular bed was performed at the end of the experimental protocols and one example of this test is shown in the tracings of Fig. 1. Each rat was implanted with 2 probes (aorta and mesenteric, aorta and renal, or mesenteric and renal) and pulsatile arterial pressure, mean arterial pressure and blood flow of two vascular beds were recorded.
2.2. Catheterization of the femoral artery and vein and arterial blood pressure recording Polyethylene PE-10 catheters connected to PE-50 (Clay Adams, Parsipanny, NJ, USA) with the length adjusted to the size of the rat were inserted into the femoral artery and vein, respectively, for measurement of arterial blood pressure and for phenylephrine injection. The arterial catheter was positioned in the abdominal aorta below the Doppler probe placed around the inferior abdominal aorta. Both arterial and venous catheters were exteriorized on the back of the rat together with the wires of the Doppler probes. Arterial pressure was recorded 24 h later under conscious freely moving conditions. The arterial catheter was connected to a Hewlett-Packard pressure transducer (267 BC, Hewlett-Packard, Watthan, MA, USA) and this to a four channel Hewlett-Packard polygraph. The Doppler flowmeter was also connected to the same polygraph for simultaneous recording of pulsatile arterial pressure (PAP), MAP and blood flow in two regional vascular beds.
2.3. Implantion of the guide cannulas in the direction of the RVLM Four days prior to the experiments the rats were anesthetized with 2.5% tribromoethanol (Aldrich, Milwaukee, WI, USA,) and placed in a stereotaxic apparatus
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
303
Brazil) administered intramuscularly. The rats were then maintained in individual cages in a room with controlled light cycle and temperature for the next 4 days. Although the stereotaxic coordinates and the weight of the rats were standardized, the center of the microinjection sites in the RVLM varied, especially in relation to the antero-posterior coordinates. Therefore, in the histological analysis of the sites of microinjection we used the obex as an anatomical reference to locate the center of microinjection in the antero-posterior coordinate. For this reason we divided the rats into three groups according to the anteroposterior coordinate of the center of microinjection: 1200– 1600 mm rostral to the obex, 1601–2000 mm rostral to the obex and 2001 to 2500 mm rostral to the obex. The precise identification of the obex and of the distance of the center of microinjection from the obex was determined using serial 15 mm thick sections of the brainstem. In the histological analysis we also considered the caudal aspect of the facial nucleus, the olive complex and the compact formation of the nucleus ambiguus for complete identification of the rostro-caudal and dorso-ventral aspects of the RVLM.
2.4. Microinjections into the RVLM Fig. 1. Typical tracings of one rat representative of the group in which the microinjections were performed in the most rostral aspect of the RVLM (2001–2500 mm rostral to the obex) showing the pulsatile arterial pressure (PAP), mean arterial pressure (MAP) and pulsatile blood flow in hindlimb (HBF) and mesenteric (MBF) vascular beds (first column) and changes in pulsatile arterial pressure, mean arterial pressure, hindquarter and mesenteric blood flow in response to microinjection of L-glutamate [1 nmol / 30 nl, right side (second column)] or saline [30 nl, left side (third column)] into the RVLM (center of microinjection: 2160 mm rostral to the obex), and in response to intravenous (i.v.) injection of phenylephrine [0.5 mg / kg, i.v. (fourth column)]. Similar protocols were used in rats with the center of microinjections located in the other two antero-posterior coordinates studied (1200–1600 and 1601–2000 mm rostral to the obex). In some rats of this group as well as in some rats of the groups in which the microinjections were located along other antero-posterior coordinates we also measured renal blood flow (tracings not shown).
(David Kopf, Tujunga, CA, USA) for implantation of guide cannulas in the direction of the RVLM. Guide cannulas made with steel tubing (22 gauge) and 15 mm long were implanted in the direction of the RVLM according to the coordinates of Paxinos and Watson (1986) in relation to the lambda: antero-posterior: 23.2; lateral: 61.8; dorso-ventral: 26.5 mm. Through a small hole in the skull the guide cannula was introduced into the cerebellum and its inferior extremity was located 1 mm above the dorsal surface of the brainstem. The guide cannulas were fixed to the skull with small watch screws and acrylic cement (Simplex, DFL, Rio de Janeiro, Brazil) and the guide cannulas were obturated with a 15 mm long metal occluder until the time for the experiments. Additional tribromoethanol was injected in case of reaction to frequent toe pinching during stereotaxic surgery. After the stereotaxic surgery the rats received 0.2 ml (1.200.000 U) ˜ Paulo, of veterinary pentabiotic (Fontoura-Wyeth, Sao
Each rat was implanted with bilateral cannulas in the direction of the RVLM, but each side received only one microinjection of L-glutamate (1 nmol / 30 nl) or saline (30 nl) as a vehicle control. L-glutamate was dissolved in saline and the pH adjusted to values near 7.4 with sodium bicarbonate and the concentration used was always 1 nmol / 30 nl. L-glutamate or saline was microinjected with a Hamilton syringe (1 ml, Hamilton, Reno, NV, USA) connected to PE-10 tubing and to the needle injector (33 gauge). The needle injector was 3.5 mm longer than the guide cannula (15 mm), whose ventral tip was located 1 mm above the dorsal surface of the brainstem. All microinjections into the RVLM were made in a volume of 30 nl, which in a specific group of rats (n56) microinjected with Evans blue presented an antero-posterior spread in the range of 368649 mm. After the insertion of the injector into the RVLM we waited at least 2 min for MAP to return to baseline level and then L-glutamate or saline was microinjected over a period of less than 30 s and the injector removed. The cardiovascular changes produced by L-glutamate into the RVLM were immediate and the peak changes in MAP and in regional blood flow were recorded. After microinjection of L-glutamate into one side, for example, at least 30 min were allowed to elapse for the microinjection of saline into the controlateral side or vice-versa.
2.5. Histology At the end of the experiments the rats were anesthetized with ether and perfused with saline followed by 10% formalin. The brain was removed, stored in formalin for at
304
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
least 48 h, embedded in paraffin and cut into 15 mm thick coronal sections. The sections were stained by the Nissl method for histological analysis and only the rats with a positive site of microinjection into the RVLM were considered. Rats in which the microinjections of L-glutamate were located outside the RVLM were considered as a negative control of misplaced microinjections. In the histological analysis we identified the pathway of the injector in the direction of the RVLM and the center of the microinjection was identified in relation to its distance from the obex. We analyzed 28 rats that were responsive to L-glutamate into the RVLM and that presented positive histology; 14 of these 28 rats received microinjection of saline (vehicle) into the controlateral RVLM. Another 21 rats were considered as the misplaced microinjection group because the center of microinjection was located in areas dorsal and / or lateral to the RVLM.
2.6. Statistical analysis The data are presented as means6SEM and were analyzed by the paired or unpaired Student t-test according to the protocol and also by one-way ANOVA, and the differences among means were determined by the Student t-test with Bonferroni adjustments for multiple comparisons. In all analyses the level of significance was fixed at 5%.
3. Results
3.1. Cardiovascular responses to microinjection of Lglutamate into the RVLM Fig. 1 illustrates the tracings of one rat representative of the group in which the center of the microinjection was located in the RVLM 2001–2500 mm rostral to the obex, showing the increase in PAP, MAP and hindquarter and mesenteric blood flow in response to L-glutamate microinjection into the left RVLM. Microinjection of Lglutamate produced a large increase in MAP and a reduction in hindquarter and mesenteric blood flow, indicating an increase in vascular resistance. Microinjection of saline into the contralateral RVLM produced no changes in MAP or hindquarter and mesenteric blood flow. Fig. 1 also shows the reduction in hindquarter and mesenteric blood flow in response to i.v. injection of phenylephrine, indicating that the Doppler flow probes were effective in detecting the changes in blood flow. Fig. 1 is also representative of the experimental protocols used in the groups in which the center of the microinjections was located 1200–1600 mm or 1601–2000 mm rostral to the obex. The data related to the changes in MAP in this group, summarized in the upper part of Fig. 2, show that microinjection of L-glutamate produced a significant in-
crease in MAP when compared with the response to saline microinjection (4564 vs. 561 mmHg). The data related to the changes in regional vascular resistance, summarized in the upper part of Fig. 3, show that the increases in hindquarter (176632%), mesenteric (138612%) and renal (111623%) vascular resistance did not differ significantly from each other. The data related to the changes in MAP of the group of rats in which the center of the microinjection was located in the RVLM 1601 to 2000 mm rostral to the obex, is shown in the middle part of Fig. 2, show that microinjection of L-glutamate produced a significant increase in MAP when compared with the response to saline microinjection (4264 vs. 462 mmHg). The data related to the changes in regional vascular resistance, summarized in the middle part of Fig. 3, show that the increase in hindquarter (223636%), mesenteric (188643%) and renal (167661%) vascular resistance did not differ significantly from each other. The data related to the changes in MAP of the group of rats in which the center of the microinjection was located in the RVLM 1200–1600 mm rostral to the obex, summarized in the bottom part of Fig. 2, show that microinjection of L-glutamate produced a significant increase in MAP when compared with the response to saline microinjection (4063 vs. 261 mmHg). The data related to the changes in the regional vascular resistance, summarized in the bottom part of Fig. 3, show that the increase in hindquarter (147626%), mesenteric (103640%) and renal (116622%) vascular resistance did not differ significantly from each other. The data show that microinjection of L-glutamate into three different antero-posterior regions of the RVLM produced a large increase in MAP (Fig. 2) of similar magnitude in the three coordinates. The increase in MAP observed in these coordinates was also associated with an increase of similar magnitude in hindquarter, mesenteric and renal vascular resistance (Fig. 3). Statistical analysis indicated no difference in the magnitude of the increase in vascular resistance in the three vascular beds in response to injection of L-glutamate into each subregion of the RVLM. Comparison of the increase in vascular resistance for each vascular bed in the three antero-posterior coordinates also showed no significant differences. The increase in MAP produced by microinjection of L-glutamate into the RVLM of rats of the three groups studied (n528) was significantly greater than in the groups of rats in which L-glutamate was misplaced microinjected (n521) into areas adjacent to the RVLM (4262 vs. 661 mmHg).
3.2. Histology The data related to the average distance of the center of microinjections into the RVLM in relation to the obex of the rats that received microinjection of L-glutamate in one
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
305
Fig. 2. Changes in mean arterial pressure (D MAP) in response to microinjection of L-glutamate ( L-glu) or saline (vehicle) into different antero-posterior coordinates of the RVLM (1200–1600, 1601–2000 and 2001–2500 mm rostral to the obex). * Different compared to the changes produced by microinjection of saline into the contralateral RVLM.
side and saline in the contralateral side are shown in the Table 1. The panels in Fig. 4 are photomicrographs of coronal sections of the brainstem showing the center of the microinjection in the RVLM of three rats representative of the three antero-posterior coordinates in relation to the obex of their respective groups. Fig. 5 is a line drawing of transverse sections of brain stem from the obex to 2500 mm rostral to the obex showing dark circles representative of the center of microinjections of L-glutamate into the RVLM of 28 rats (panel A) or the center of misplaced microinjections of L-glutamate into adjacent areas to RVLM of 21 rats (panel B).
4. Discussion Although previous studies have evaluated a possible differential control of regional vascular resistance by sympathetic vasomotor neurons in rats (Beluli and Weaver, 1991; Bonham and Jeske, 1989; Dean et al., 1992; Ermirio et al., 1993), there are no detailed studies related to the topographic distribution of sympathetic vasomotor neurons in the RVLM of this species. However, studies performed
in cats and rabbits have suggested a topographic distribution of the sympathetic vasomotor neurons in the RVLM (Dampney et al., 1985; Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; McAllen and May, 1994). However, these studies in different species were performed under anesthesia, which may be an additional factor influencing in the apparent distribution of these vasomotor neurons in the RVLM. In the present study we used awake animals in order to evaluate the cardiovascular changes produced by microinjection of Lglutamate into the RVLM in the absence of anesthetic effects, which may affect the processing of the circuits involved in the central neural control of the circulation, as we observed previously with the microinjection of Lglutamate into the NTS (Machado and Bonagamba, 1992). The first synapses of the arterial baroreceptors and chemoreceptor reflex inputs in the CNS occur at the level of the NTS and from this nucleus projections to the caudal and rostral medulla involved in the regulation of the sympathetic drive to the heart and vessels have been described (Guyenet, 1990; Spyer, 1990; Dampney, 1994). There is also evidence in the literature suggesting a topographic distribution of the sympathetic vasomotor
306
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
Fig. 3. Percent changes in vascular resistance (% DVR) of hindlimb (hind), mesenteric (mesent.) and renal vascular beds in response to microinjection of L-glutamate into different antero-posterior coordinates of the RVLM (1200–1600, 1601–2000 and 2001–2500 mm rostral to the obex).
neurons in the RVLM, which are selective for the innervation of each vascular bed in order to modulate the sympathetic drive to each of these vessels in a very precise manner (Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; McAllen and May, 1994). In addition, more recently Campos and McAllen (1997) also demonstrated a topographic distribution of the presympathetic neurons involved in the regulation of heart rate and in the contractility of the myocardium. In the present study we evaluated a possible anteroposterior topographic distribution of RVLM neurons using small (30 nl) microinjections of L-glutamate into the Table 1 Average distance of the centers of microinjections of L-glutamate into RVLM and saline into the contralateral RVLM from the obex (anteroposterior coordinates) of the three groups of rats studied a Antero-Posterior coordinates (mm)
L-glutamate
(one side)
Saline (contralateral side)
1200–1600 1601–2000 2001–2500
1444632 (10) 1773625 (13) 2237682 (6)
1473638 (5) 1799623(7) 2110635(2)
a The number of rats evaluated in each antero-posterior coordinate is given in parenthesis.
RVLM. Even this relatively small volume was shown to spread in the antero-posterior directions by 368649 mm when Evans blue was contained in the injectate. Accordingly a single microinjection was performed in each RVLM since multiple injections would be expected to excite different pools of neurons as the injector extended 3.5 mm beyond the guide-cannula and a diffusion pathway could be created at each insertion of the injector. The data show that the microinjection of L-glutamate into three different antero-posterior regions of the RVLM evoked a large increases in MAP at all three locations. These increases in MAP were also associated with an increase of similar magnitude in hindquarter, mesenteric and renal vascular resistance. Statistical analysis indicated no difference in the magnitude of the increase in vascular resistance in the three vascular beds in response to injection of L-glutamate into each subregion of the RVLM. Comparison of the increase in vascular resistance for each vascular bed in the three antero-posterior coordinates also showed no significant differences. A study by Ermirio et al. (1993) showed that electrical stimulation of sino-aortic and visceral nerves of rats produced a non-selective depolarization of neurons in the RVLM. A study by Beluli and Weaver (1991) indicated
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
307
Fig. 4. Photomicrographs of three coronal sections of the brainstem of rats representative of their respective group showing a typical site in the center of microinjection in the RVLM (1203) in the three antero-posterior coordinates studied. Panel A: center of microinjection 2250 mm rostral to the obex; Panel B: center of microinjection 1820 mm rostral to the obex; Panel C: center of microinjection 1520 mm rostral to the obex.
Fig. 5. Line drawing of a transverse section of brain stem from the obex to 2500 mm rostral to the obex [adapted from Paxinos and Watson (1986)] showing dark circles representative of the center of microinjections of L-glutamate into the RVLM of 28 rats (panel A) or the center of misplaced microinjections of L-glutamate into areas adjacent to the RVLM of 21 rats (panel B).
308
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
that neurons of the rat RVLM with a differential control of the sympathetic drive to the renal and splanchnic territories also showed no topographical organization within the RVLM. However, studies on cats have shown evidence supporting the concept that the sympathetic vasomotor neurons in the RVLM present a topographic distribution (Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994). Although our data for awake rats did not show a differential control between the three vascular beds studied and did not show a topographic distribution of sympathetic vasomotor neurons in the RVLM, at least in the antero-posterior coordinates, the data obtained by Beluli and Weaver (1991) supports the concept that regional vascular resistance is controlled by specific neurons in the RVLM. This is consistent with observations from this laboratory in the rat that the electrical stimulation of the aortic depressor nerve (Machado et al., 1994; de Paula et al., 1999) or activation of the chemoreflex (unpublished data) elicits a selective control of the different vascular beds studied. If the observations in the cat and rabbit are correct, why is there no topographical organization in the rat RVLM? The first possibility is that the RVLM of rats, which is significantly smaller than the RVLM of cats, impairs the characterization of a topographic distribution using microinjection with volumes of 30 nl. Whilst the dose of L-glutamate was the same in each region we have no knowledge of the variation of excitatory amino acid receptors present in each subregion of the RVLM (Bonham and Jeske, 1989) and with extensive spread of injectate an ‘average’ response was elicited from each site masking the regional differences. The alternative explanation is that sympathetic vasomotor neurons for the different vascular beds are randomly distributed in the antero-posterior, dorso-ventral and lateral aspects of the RVLM and present no topographic localization. The absence of evidence in favor of a topographic distribution in the present study might be related to the influence of the activation of the baroreceptor afferents in response to the large increase in arterial pressure produced by the microinjection of L-glutamate into the RVLM. This activation of the baroreceptors would be expected to excite sympathoinhibitory pathways involving CVLM neurons leading to an inhibition of RVLM neurons. This powerful feedback inhibition of sympathetic vasomotor neurons in the RVLM could mask any differential changes in regional vascular resistance induced by the microinjection of Lglutamate into specific subregions of the RVLM. However, it is important to note that several studies using chemical or electrical stimulation of RVLM observed a differential control of sympathetic activity by RVLM neurons, in spite of a major pressor response (Ross et al., 1984a,b; Dampney et al., 1985; Willete et al., 1983, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994; Sun, 1995).
It is also necessary to consider the fact that our experiments were performed in awake rats while those experiments indicating a topographical distribution of sympathetic vasomotor neurons were all performed in anesthetized animals. The anesthetics may have a distorting effect on several neural mechanisms, particularly on neurotrasmission, producing in some cases opposite cardiovascular responses to the excitation of the paraventricular nucleus (Kannan et al., 1989) or NTS (Machado and Bonagamba, 1992) when compared with the data obtained in awake rats. Therefore, it is plausible to suggest that the absence of anesthesia in our experiments may explain, at least in part, our different findings in relation to other studies. A differential control of regional sympathetic outflow certainly exists in rats with intact arterial baroreceptors, as documented by Beluli and Weaver (1991), and a differential control of the regional vascular resistance by the baroreceptor (de Paula et al., 1999) and chemoreceptor afferents is also reported (unpublished data from our laboratory). However, the study by Beluli and Weaver (1991) suggest that the sympathetic vasomotor neurons in the RVLM of rats are not organized viscerotopically, which is compatible with our findings. However, it is important to consider that the quantitative change in sympathetic nerve activity recorded by Beluli and Weaver (1991) may not be necessarily converted to an equivalent change in blood flow. As such, we cannot not rule out the possibility that the measurement of regional blood flow may not be sensitive enough to detect discrete differences in relation to differential sympathetic control of the regional vascular resistance by RVLM neurons.
Acknowledgements The authors thank Leni G.H. Bonagamba and Jaci A. Castania for excellent technical assistance and Rubens F. de Melo for the histological preparations. This work was ˜ de Amparo a` Pesquisa do Estado supported by Fundac¸ao ˜ Paulo (FAPESP), Conselho Nacional de Desenvolde Sao ´ ´ vimento Cientıfico e Tecnologico (CNPQ) and Programa ˆ ´ de Apoio aos Nucleos de Excelencia (PRONEX).
References Beluli, D.J., Weaver. L.C., 1991. Differential control of renal and splenic nerves without medullary topography. Am J. Physiol., 260 (Heart Circ. Physiol.): H1072–H1079. Bonham, A.C., Jeske, I., 1989. Cardiorespiratory effects of DLhomocysteic acid in caudal ventrolateral medulla. Am. J. Physiol., 256 (Heart Circ. Physiol.): H688–H696. Campos, Jr, R.R., McAllen, R.M., 1997. Cardiac sympathetic premotor neurons. Am J. Physiol., 272 (Regulatory Integrative Comp. Physiol. 41): R615–R620. Dampney, R.A.L., 1994. Functional organization of central pathways regulating the cardiovascular system. Physiol. Rev. 74, 323–364.
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309 Dampney, R.A.L., McAllen, R.M., 1988. Differential control of sympathetic fibres supplying hindlimb skin and muscle by subretrofacial neurons in the cat. J. Physiol. 395, 41–56. Dampney, R.A.L., Goodchild, A.K., Tan, E., 1985. Vasopressor neurons in the rostral ventrolateral medulla of the rabbit. J. Auton. Nerv. System 14, 239–254. Dean, C., Seagard, J.L., Hopp, F.A., Kampine, J.P., 1992. Differential control of sympathetic activity to kidney and skeletal muscle by ventral medullary neurons. J. Auton. Nerv. System 37, 1–10. de Paula, P.M., Castania, J.A., Salgado, H.C., Machado, B.H., 1999. Hemodynamic responses to electrical stimulation of the aortic depressor nerve in awake rats. Am. J. Physiol. 277 (Regulatory Integrative Comp. Physiol. 46): R31–R38. de Paula, P.M., Machado, B.H., 1997. Microinjection of L-glutamate into different sites of rostral ventrolateral medulla of unanesthetized rats produced similar changes in regional vascular resistance. J. Auton. Nerv. Syst. 65, 82. Ermirio, R., Ruggeri, P., Molinari, C., Weaver, L.C., 1993. Somatic and visceral inputs to neurons of the rostral ventrolateral medulla. Am. J. Physiol. 265 (Regulatory Integrative Comp. Physiol.): R35–R40. Guyenet, P.G., 1990. Role of the ventral medulla oblongata in blood pressure regulation. In: Loewy A.C., Spyer, K.M. (Eds.), Central Regulation of Autonomic Functions. New York Oxford, Oxford University Press, pp. 145–167. Haywood, J.R., Shaffer, R.A., Fastenow, C., Fink, G.D., Brody, M.J., 1981. Regional blood flow measurement with pulsed Doppler flowmeter in conscious rat. Am J. Physiol., 241 (Heart Circ. Physiol.): H273–H278. Kannan, H., Hayashida, Y., Yamashita, H., 1989. Increase in sympathetic outflow by paraventricular nucleus stimulation in awake rats. Am. J. Physiol. 256 (Regulatory Integrative Comp. Physiol. 25): R1325– R1330. Lovick, T.A., 1987. Differential control of cardiac and vasomotor activity by neurones in nucleus paragigantocellularis lateralis in the cat. J. Physiol. 389, 23–35. Machado, B.H., Bonagamba, L.G.H., Castania, J.A., Menani, J.V., 1992. Changes in vascular resistance during carotid occlusion in normal and baroreceptor-denervated rats. Hypertension 19 (Suppl. II), II149– II153. Machado, B.H., Bonagamba, L.G.H., Castania, J.A., Menani, J.V., 1994.
309
Aortic baroreceptors play a predominant role in the regulation of hindlimb vascular resistance in rats. Am. J. Physiol., 267 (Regulatory Integrative Comp. Physiol. 36): R476–R480. Machado, B.H., Bonagamba, L.G.H., 1992. Microinjection of L-glutamate into the nucleus tractus solitarii increases arterial pressure in conscious rats. Brain Res. 576, 131–138. McAllen, R.M., May, C.N., 1994. Differential drives from rostral ventrolateral medullary neurons to three identified sympathetic outflows. Am. J. Physiol., 267 (Regulatory Integrative Comp. Physiol. 36): R935–R944. McAllen, R.M., Dampney, R.A.L., 1990. Vasomotor neurons in the rostral ventrolateral medulla are organized topographically with respect to type of vascular bed but not body region. Neurosci. Lett. 110, 91–96. Paxinos, G., Watson, C., 1986. The Brain in Stereotaxic Coordinates. Academic Press, New York. Ross, C.A., Ruggiero, D.A., Park, D.H., Joh, T.H., Sved, A.F., FernandezPardal, J., Saavedra, J.M., Reis, D.J., 1984a. Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area containing C1 adrenaline neurons on arterial pressure, heart rate, and plasma catecholamines and vasopressin. J. Neurosci. 4, 474–494. Ross, C.A., Ruggiero, D.A., Joh, T.H., Park, D.H., Reis, D.J., 1984b. Rostral ventrolateral medulla: selective projections to the thoracic autonomic cell column from the region containing C1 adrenaline neurons. J. Comp. Neurol. 228, 168–185. Spyer, K.M., 1990. The central nervous organization of reflex circulatory control. In: Loewy A.C., Spyer, K.M. (Eds.), Central Regulation of Autonomic Functions, New York, Oxford. Oxford University Press. Sun, M-K., 1995. Central neural organization and control of sympathetic nervous system in mammals. Prog. Neurob. 47, 157–233. Urbanski, R.W., Sapru, H.N., 1988. Putative neurotransmitters involved in medullary cardiovascular regulation. J. Auton. Nerv. Syst. 25, 181– 193. Willete, R.N., Barcas, P.P., Krieger, A.J., Sapru, H.N., 1983. Vasopressor and depressor areas in the rat medulla. Neuropharmacology 22 (9), 1071–1079. Willete, R.N., Punnen-Grandy, S., Krieger, A.J., Sapru, H.N., 1987. Differential regulation of regional vascular resistance by the rostral and caudal ventrolateral medulla in the rat. J. Auton. Nerv. Syst. 18, 143–151.
Changes in regional vascular resistance in response to microinjection of L-glutamate into different antero-posterior coordinates of the RVLM in awake rats ´ Patrıcia M. de Paula, Benedito H. Machado* ˜ Preto, University of Sao ˜ Paulo, 14049 -900, Ribeirao ˜ Preto, SP, Brazil Department of Physiology, School of Medicine of Ribeirao Received 30 March 1999; accepted 7 March 2000
Abstract Changes in mean arterial pressure (MAP) and in regional vascular resistance (RVR, hindquarter, mesenteric and renal) induced by microinjection of L-glutamate into three different antero-posterior coordinates of the rostral ventrolateral medulla (RVLM) [1200–1600 mm (n510), 1601–2000 mm (n512) and 2001–2500 mm (n56) rostral to the obex] were investigated in unanesthetized rats. Guide cannulas directed towards the RVLM were implanted 4 days prior to the experiments. Doppler probes were implanted around the superior mesenteric, inferior abdominal aorta and left renal arteries and a catheter was inserted into the femoral artery and vein 1 day prior to the experiments. Insertion of the injector into the RVLM produced an increase in baseline MAP, which was back to control levels 2 min later, when L-glutamate was microinjected. Microinjection of L-glutamate (1 nmol / 30 nl) into the three antero-posterior coordinates of the RVLM produced an increase in MAP associated with a similar increase in hindquarter, mesenteric and renal vascular resistance, which were back to control 1 min later. Saline into the RVLM produced negligible effects on MAP and RVR. These findings suggest that the sympathetic vasomotor neurons involved in the regulation of the regional vascular resistance in rats are not topographically distributed in the antero-posterior coordinates of the RVLM. However, the experimental methods used to evaluate the topographic distribution of sympatho-vasomotor neurons in the RVLM and the measurement of the regional blood flow may not be precise enough to detect any possible differences. 2001 Elsevier Science B.V. All rights reserved. Keywords: Blood flow; Baroreflex; Sympathetic vasomotor neurons; Cardiovascular regulation, Vascular beds, Renal blood flow, Mesenteric blood flow, Hindquarter blood flow
1. Introduction The activation of the baroreceptors produces autonomic adjustments in order to maintain arterial blood pressure within a narrow range of variation. The sympathetic vasomotor tone is generated by neurons located in the rostral ventrolateral medulla (RVLM) and the activation of the baroreceptors produces the inhibitory modulation of these neurons through projections from the nucleus tractus solitarii (NTS) to the RVLM via the caudal ventrolateral medulla (Ross et al., 1984a,b; Willete et al., 1987; Urbanski and Sapru, 1988; Sun, 1995). In recent studies from our laboratory we verified that electrical stimulation of the aortic depressor nerve induced a greater vasodilatation in the hindquarter vascular bed than in other ter*Corresponding author. Tel.: 155-16-602-3015; fax: 155-16-6330017. E-mail address: [email protected] (B.H. Machado).
ritories (Machado et al., 1994; de Paula et al., 1999), supporting our previous finding that bilateral carotid occlusion (BCO) in intact rats produced major vasoconstriction in the mesenteric and renal vascular territories, while the BCO in rats with selective denervation of the aortic nerves produced major vasoconstriction in the hindquarter territory (Machado et al., 1992). Therefore, these data indicate that aortic baroreceptors play a selective and predominant role in the regulation of the hindquarter vascular resistance compared to other territories, suggesting that the activation of this specific subset of baroreceptor afferents produces selective inhibition of sympathetic neurons in the RVLM involved in the innervation of the vasculature of skeletical muscles. Chemical or electrical stimulation of RVLM neurons produced a major pressor response (Ross et al., 1984a,b; Dampney et al., 1985; Willete et al., 1987; Sun, 1995) due to vasoconstriction, which, however, was not generalized for all vascular territories because several lines of ex-
0165-1838 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S1566-0702( 00 )00283-6
302
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
perimental evidence obtained in cats indicate that the sympathetic vasomotor neurons in the RVLM are topographically distributed and that these different groups of neurons present a selective regulation of the vascular resistance of different vascular beds (Willete et al., 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994). Studies by Dean et al. (1992) also support this concept considering that neurons in the RVLM involved in kidney blood flow regulation are located more rostrally than those involved in skeletical muscle blood flow while studies by McAllen and Dampney (1990) have indicated that the sympathetic vasomotor neurons are topographically distributed in accordance with the type of vascular bed but not in relation to the body region. All the evidence presented above suggests that the regulation of regional blood flow is selectively regulated by specific baroreceptor afferent fibers as well as by different subpopulations of the sympathetic vasomotor neurons in the RVLM. However, most of these studies were performed on cats or on anesthetized animals and no studies explored the changes in regional vascular resistance in response to microinjection of L-glutamate into different subregions of the RVLM of conscious freely moving rats. Therefore, the aim of the present study was to determine whether the changes in mean arterial pressure (MAP) in response to L-glutamate microinjection into different antero-posterior coordinates of the RVLM are related to the increase in the resistance of some specific vascular bed. Part of these data have been published previously in Abstract form (de Paula and Machado, 1997).
2. Methods Male Wistar rats weighing 230–270 g were used in the present study. Four days before the experiments bilateral guide cannulas in the direction of the RVLM were implanted under tribromoethanol anesthesia (250 mg / kg, i.p.). One day before the experiments the rats were anesthetized with sodium pentobarbital (40 mg / kg, i.p.) for the implantation of miniaturized Doppler probes around the aorta and mesenteric and renal arteries as well as for catheterization of femoral artery and vein.
2.1. Implantation of the Doppler probes After a medial laparatomy a small 4-mm long segment of the inferior abdominal aorta, superior mesenteric or left renal artery was carefully isolated and probes with a specific internal diameter were placed around each vessel. The two extremities of the silastic probes embracing the vessels were fastened with ophthalmic 6-0 sutures. The wires of the probes were sutured to the internal abdominal wall in order to keep constant the position of the probes in relation to the vessels. After the abdominal suture, the wires were exteriorized subcutaneously on the back of the
rat where the connector was fixed. On the day of the experiments the wires were connected to the Doppler flowmeter (University of Iowa Bioengineering Facility) which in turn was connected to a four-channel recorder (Hewlett-Packard). The flowmeter was used to determine the velocity of blood flow through the changes in the frequency of the piezoelectric crystal (20 MHz, Doppler effect), which, according to Haywood et al. (1981) are directly related to absolute blood flow. In this case the resistance can be calculated as follows: Resistance5MAP (mmHg) / Doppler frequency (Hz), and consequently the changes in the vascular resistance of each vascular bed can be determined as percent change in relation to control. The changes in regional vascular resistance were calculated in each experiment considering the peak of changes in MAP as well as the peak of changes in regional blood flow. After the probe wires were connected to the recorder system, the efficacy of the probes was checked by recording pulsatile flow as well as by the intravenous injection of phenylephrine to produce a vasoconstriction and consequent reduction in blood flow. The tests for the efficacy of the Doppler probes in detecting changes in the blood flow of each vascular bed was performed at the end of the experimental protocols and one example of this test is shown in the tracings of Fig. 1. Each rat was implanted with 2 probes (aorta and mesenteric, aorta and renal, or mesenteric and renal) and pulsatile arterial pressure, mean arterial pressure and blood flow of two vascular beds were recorded.
2.2. Catheterization of the femoral artery and vein and arterial blood pressure recording Polyethylene PE-10 catheters connected to PE-50 (Clay Adams, Parsipanny, NJ, USA) with the length adjusted to the size of the rat were inserted into the femoral artery and vein, respectively, for measurement of arterial blood pressure and for phenylephrine injection. The arterial catheter was positioned in the abdominal aorta below the Doppler probe placed around the inferior abdominal aorta. Both arterial and venous catheters were exteriorized on the back of the rat together with the wires of the Doppler probes. Arterial pressure was recorded 24 h later under conscious freely moving conditions. The arterial catheter was connected to a Hewlett-Packard pressure transducer (267 BC, Hewlett-Packard, Watthan, MA, USA) and this to a four channel Hewlett-Packard polygraph. The Doppler flowmeter was also connected to the same polygraph for simultaneous recording of pulsatile arterial pressure (PAP), MAP and blood flow in two regional vascular beds.
2.3. Implantion of the guide cannulas in the direction of the RVLM Four days prior to the experiments the rats were anesthetized with 2.5% tribromoethanol (Aldrich, Milwaukee, WI, USA,) and placed in a stereotaxic apparatus
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
303
Brazil) administered intramuscularly. The rats were then maintained in individual cages in a room with controlled light cycle and temperature for the next 4 days. Although the stereotaxic coordinates and the weight of the rats were standardized, the center of the microinjection sites in the RVLM varied, especially in relation to the antero-posterior coordinates. Therefore, in the histological analysis of the sites of microinjection we used the obex as an anatomical reference to locate the center of microinjection in the antero-posterior coordinate. For this reason we divided the rats into three groups according to the anteroposterior coordinate of the center of microinjection: 1200– 1600 mm rostral to the obex, 1601–2000 mm rostral to the obex and 2001 to 2500 mm rostral to the obex. The precise identification of the obex and of the distance of the center of microinjection from the obex was determined using serial 15 mm thick sections of the brainstem. In the histological analysis we also considered the caudal aspect of the facial nucleus, the olive complex and the compact formation of the nucleus ambiguus for complete identification of the rostro-caudal and dorso-ventral aspects of the RVLM.
2.4. Microinjections into the RVLM Fig. 1. Typical tracings of one rat representative of the group in which the microinjections were performed in the most rostral aspect of the RVLM (2001–2500 mm rostral to the obex) showing the pulsatile arterial pressure (PAP), mean arterial pressure (MAP) and pulsatile blood flow in hindlimb (HBF) and mesenteric (MBF) vascular beds (first column) and changes in pulsatile arterial pressure, mean arterial pressure, hindquarter and mesenteric blood flow in response to microinjection of L-glutamate [1 nmol / 30 nl, right side (second column)] or saline [30 nl, left side (third column)] into the RVLM (center of microinjection: 2160 mm rostral to the obex), and in response to intravenous (i.v.) injection of phenylephrine [0.5 mg / kg, i.v. (fourth column)]. Similar protocols were used in rats with the center of microinjections located in the other two antero-posterior coordinates studied (1200–1600 and 1601–2000 mm rostral to the obex). In some rats of this group as well as in some rats of the groups in which the microinjections were located along other antero-posterior coordinates we also measured renal blood flow (tracings not shown).
(David Kopf, Tujunga, CA, USA) for implantation of guide cannulas in the direction of the RVLM. Guide cannulas made with steel tubing (22 gauge) and 15 mm long were implanted in the direction of the RVLM according to the coordinates of Paxinos and Watson (1986) in relation to the lambda: antero-posterior: 23.2; lateral: 61.8; dorso-ventral: 26.5 mm. Through a small hole in the skull the guide cannula was introduced into the cerebellum and its inferior extremity was located 1 mm above the dorsal surface of the brainstem. The guide cannulas were fixed to the skull with small watch screws and acrylic cement (Simplex, DFL, Rio de Janeiro, Brazil) and the guide cannulas were obturated with a 15 mm long metal occluder until the time for the experiments. Additional tribromoethanol was injected in case of reaction to frequent toe pinching during stereotaxic surgery. After the stereotaxic surgery the rats received 0.2 ml (1.200.000 U) ˜ Paulo, of veterinary pentabiotic (Fontoura-Wyeth, Sao
Each rat was implanted with bilateral cannulas in the direction of the RVLM, but each side received only one microinjection of L-glutamate (1 nmol / 30 nl) or saline (30 nl) as a vehicle control. L-glutamate was dissolved in saline and the pH adjusted to values near 7.4 with sodium bicarbonate and the concentration used was always 1 nmol / 30 nl. L-glutamate or saline was microinjected with a Hamilton syringe (1 ml, Hamilton, Reno, NV, USA) connected to PE-10 tubing and to the needle injector (33 gauge). The needle injector was 3.5 mm longer than the guide cannula (15 mm), whose ventral tip was located 1 mm above the dorsal surface of the brainstem. All microinjections into the RVLM were made in a volume of 30 nl, which in a specific group of rats (n56) microinjected with Evans blue presented an antero-posterior spread in the range of 368649 mm. After the insertion of the injector into the RVLM we waited at least 2 min for MAP to return to baseline level and then L-glutamate or saline was microinjected over a period of less than 30 s and the injector removed. The cardiovascular changes produced by L-glutamate into the RVLM were immediate and the peak changes in MAP and in regional blood flow were recorded. After microinjection of L-glutamate into one side, for example, at least 30 min were allowed to elapse for the microinjection of saline into the controlateral side or vice-versa.
2.5. Histology At the end of the experiments the rats were anesthetized with ether and perfused with saline followed by 10% formalin. The brain was removed, stored in formalin for at
304
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
least 48 h, embedded in paraffin and cut into 15 mm thick coronal sections. The sections were stained by the Nissl method for histological analysis and only the rats with a positive site of microinjection into the RVLM were considered. Rats in which the microinjections of L-glutamate were located outside the RVLM were considered as a negative control of misplaced microinjections. In the histological analysis we identified the pathway of the injector in the direction of the RVLM and the center of the microinjection was identified in relation to its distance from the obex. We analyzed 28 rats that were responsive to L-glutamate into the RVLM and that presented positive histology; 14 of these 28 rats received microinjection of saline (vehicle) into the controlateral RVLM. Another 21 rats were considered as the misplaced microinjection group because the center of microinjection was located in areas dorsal and / or lateral to the RVLM.
2.6. Statistical analysis The data are presented as means6SEM and were analyzed by the paired or unpaired Student t-test according to the protocol and also by one-way ANOVA, and the differences among means were determined by the Student t-test with Bonferroni adjustments for multiple comparisons. In all analyses the level of significance was fixed at 5%.
3. Results
3.1. Cardiovascular responses to microinjection of Lglutamate into the RVLM Fig. 1 illustrates the tracings of one rat representative of the group in which the center of the microinjection was located in the RVLM 2001–2500 mm rostral to the obex, showing the increase in PAP, MAP and hindquarter and mesenteric blood flow in response to L-glutamate microinjection into the left RVLM. Microinjection of Lglutamate produced a large increase in MAP and a reduction in hindquarter and mesenteric blood flow, indicating an increase in vascular resistance. Microinjection of saline into the contralateral RVLM produced no changes in MAP or hindquarter and mesenteric blood flow. Fig. 1 also shows the reduction in hindquarter and mesenteric blood flow in response to i.v. injection of phenylephrine, indicating that the Doppler flow probes were effective in detecting the changes in blood flow. Fig. 1 is also representative of the experimental protocols used in the groups in which the center of the microinjections was located 1200–1600 mm or 1601–2000 mm rostral to the obex. The data related to the changes in MAP in this group, summarized in the upper part of Fig. 2, show that microinjection of L-glutamate produced a significant in-
crease in MAP when compared with the response to saline microinjection (4564 vs. 561 mmHg). The data related to the changes in regional vascular resistance, summarized in the upper part of Fig. 3, show that the increases in hindquarter (176632%), mesenteric (138612%) and renal (111623%) vascular resistance did not differ significantly from each other. The data related to the changes in MAP of the group of rats in which the center of the microinjection was located in the RVLM 1601 to 2000 mm rostral to the obex, is shown in the middle part of Fig. 2, show that microinjection of L-glutamate produced a significant increase in MAP when compared with the response to saline microinjection (4264 vs. 462 mmHg). The data related to the changes in regional vascular resistance, summarized in the middle part of Fig. 3, show that the increase in hindquarter (223636%), mesenteric (188643%) and renal (167661%) vascular resistance did not differ significantly from each other. The data related to the changes in MAP of the group of rats in which the center of the microinjection was located in the RVLM 1200–1600 mm rostral to the obex, summarized in the bottom part of Fig. 2, show that microinjection of L-glutamate produced a significant increase in MAP when compared with the response to saline microinjection (4063 vs. 261 mmHg). The data related to the changes in the regional vascular resistance, summarized in the bottom part of Fig. 3, show that the increase in hindquarter (147626%), mesenteric (103640%) and renal (116622%) vascular resistance did not differ significantly from each other. The data show that microinjection of L-glutamate into three different antero-posterior regions of the RVLM produced a large increase in MAP (Fig. 2) of similar magnitude in the three coordinates. The increase in MAP observed in these coordinates was also associated with an increase of similar magnitude in hindquarter, mesenteric and renal vascular resistance (Fig. 3). Statistical analysis indicated no difference in the magnitude of the increase in vascular resistance in the three vascular beds in response to injection of L-glutamate into each subregion of the RVLM. Comparison of the increase in vascular resistance for each vascular bed in the three antero-posterior coordinates also showed no significant differences. The increase in MAP produced by microinjection of L-glutamate into the RVLM of rats of the three groups studied (n528) was significantly greater than in the groups of rats in which L-glutamate was misplaced microinjected (n521) into areas adjacent to the RVLM (4262 vs. 661 mmHg).
3.2. Histology The data related to the average distance of the center of microinjections into the RVLM in relation to the obex of the rats that received microinjection of L-glutamate in one
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
305
Fig. 2. Changes in mean arterial pressure (D MAP) in response to microinjection of L-glutamate ( L-glu) or saline (vehicle) into different antero-posterior coordinates of the RVLM (1200–1600, 1601–2000 and 2001–2500 mm rostral to the obex). * Different compared to the changes produced by microinjection of saline into the contralateral RVLM.
side and saline in the contralateral side are shown in the Table 1. The panels in Fig. 4 are photomicrographs of coronal sections of the brainstem showing the center of the microinjection in the RVLM of three rats representative of the three antero-posterior coordinates in relation to the obex of their respective groups. Fig. 5 is a line drawing of transverse sections of brain stem from the obex to 2500 mm rostral to the obex showing dark circles representative of the center of microinjections of L-glutamate into the RVLM of 28 rats (panel A) or the center of misplaced microinjections of L-glutamate into adjacent areas to RVLM of 21 rats (panel B).
4. Discussion Although previous studies have evaluated a possible differential control of regional vascular resistance by sympathetic vasomotor neurons in rats (Beluli and Weaver, 1991; Bonham and Jeske, 1989; Dean et al., 1992; Ermirio et al., 1993), there are no detailed studies related to the topographic distribution of sympathetic vasomotor neurons in the RVLM of this species. However, studies performed
in cats and rabbits have suggested a topographic distribution of the sympathetic vasomotor neurons in the RVLM (Dampney et al., 1985; Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; McAllen and May, 1994). However, these studies in different species were performed under anesthesia, which may be an additional factor influencing in the apparent distribution of these vasomotor neurons in the RVLM. In the present study we used awake animals in order to evaluate the cardiovascular changes produced by microinjection of Lglutamate into the RVLM in the absence of anesthetic effects, which may affect the processing of the circuits involved in the central neural control of the circulation, as we observed previously with the microinjection of Lglutamate into the NTS (Machado and Bonagamba, 1992). The first synapses of the arterial baroreceptors and chemoreceptor reflex inputs in the CNS occur at the level of the NTS and from this nucleus projections to the caudal and rostral medulla involved in the regulation of the sympathetic drive to the heart and vessels have been described (Guyenet, 1990; Spyer, 1990; Dampney, 1994). There is also evidence in the literature suggesting a topographic distribution of the sympathetic vasomotor
306
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
Fig. 3. Percent changes in vascular resistance (% DVR) of hindlimb (hind), mesenteric (mesent.) and renal vascular beds in response to microinjection of L-glutamate into different antero-posterior coordinates of the RVLM (1200–1600, 1601–2000 and 2001–2500 mm rostral to the obex).
neurons in the RVLM, which are selective for the innervation of each vascular bed in order to modulate the sympathetic drive to each of these vessels in a very precise manner (Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; McAllen and May, 1994). In addition, more recently Campos and McAllen (1997) also demonstrated a topographic distribution of the presympathetic neurons involved in the regulation of heart rate and in the contractility of the myocardium. In the present study we evaluated a possible anteroposterior topographic distribution of RVLM neurons using small (30 nl) microinjections of L-glutamate into the Table 1 Average distance of the centers of microinjections of L-glutamate into RVLM and saline into the contralateral RVLM from the obex (anteroposterior coordinates) of the three groups of rats studied a Antero-Posterior coordinates (mm)
L-glutamate
(one side)
Saline (contralateral side)
1200–1600 1601–2000 2001–2500
1444632 (10) 1773625 (13) 2237682 (6)
1473638 (5) 1799623(7) 2110635(2)
a The number of rats evaluated in each antero-posterior coordinate is given in parenthesis.
RVLM. Even this relatively small volume was shown to spread in the antero-posterior directions by 368649 mm when Evans blue was contained in the injectate. Accordingly a single microinjection was performed in each RVLM since multiple injections would be expected to excite different pools of neurons as the injector extended 3.5 mm beyond the guide-cannula and a diffusion pathway could be created at each insertion of the injector. The data show that the microinjection of L-glutamate into three different antero-posterior regions of the RVLM evoked a large increases in MAP at all three locations. These increases in MAP were also associated with an increase of similar magnitude in hindquarter, mesenteric and renal vascular resistance. Statistical analysis indicated no difference in the magnitude of the increase in vascular resistance in the three vascular beds in response to injection of L-glutamate into each subregion of the RVLM. Comparison of the increase in vascular resistance for each vascular bed in the three antero-posterior coordinates also showed no significant differences. A study by Ermirio et al. (1993) showed that electrical stimulation of sino-aortic and visceral nerves of rats produced a non-selective depolarization of neurons in the RVLM. A study by Beluli and Weaver (1991) indicated
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
307
Fig. 4. Photomicrographs of three coronal sections of the brainstem of rats representative of their respective group showing a typical site in the center of microinjection in the RVLM (1203) in the three antero-posterior coordinates studied. Panel A: center of microinjection 2250 mm rostral to the obex; Panel B: center of microinjection 1820 mm rostral to the obex; Panel C: center of microinjection 1520 mm rostral to the obex.
Fig. 5. Line drawing of a transverse section of brain stem from the obex to 2500 mm rostral to the obex [adapted from Paxinos and Watson (1986)] showing dark circles representative of the center of microinjections of L-glutamate into the RVLM of 28 rats (panel A) or the center of misplaced microinjections of L-glutamate into areas adjacent to the RVLM of 21 rats (panel B).
308
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309
that neurons of the rat RVLM with a differential control of the sympathetic drive to the renal and splanchnic territories also showed no topographical organization within the RVLM. However, studies on cats have shown evidence supporting the concept that the sympathetic vasomotor neurons in the RVLM present a topographic distribution (Lovick, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994). Although our data for awake rats did not show a differential control between the three vascular beds studied and did not show a topographic distribution of sympathetic vasomotor neurons in the RVLM, at least in the antero-posterior coordinates, the data obtained by Beluli and Weaver (1991) supports the concept that regional vascular resistance is controlled by specific neurons in the RVLM. This is consistent with observations from this laboratory in the rat that the electrical stimulation of the aortic depressor nerve (Machado et al., 1994; de Paula et al., 1999) or activation of the chemoreflex (unpublished data) elicits a selective control of the different vascular beds studied. If the observations in the cat and rabbit are correct, why is there no topographical organization in the rat RVLM? The first possibility is that the RVLM of rats, which is significantly smaller than the RVLM of cats, impairs the characterization of a topographic distribution using microinjection with volumes of 30 nl. Whilst the dose of L-glutamate was the same in each region we have no knowledge of the variation of excitatory amino acid receptors present in each subregion of the RVLM (Bonham and Jeske, 1989) and with extensive spread of injectate an ‘average’ response was elicited from each site masking the regional differences. The alternative explanation is that sympathetic vasomotor neurons for the different vascular beds are randomly distributed in the antero-posterior, dorso-ventral and lateral aspects of the RVLM and present no topographic localization. The absence of evidence in favor of a topographic distribution in the present study might be related to the influence of the activation of the baroreceptor afferents in response to the large increase in arterial pressure produced by the microinjection of L-glutamate into the RVLM. This activation of the baroreceptors would be expected to excite sympathoinhibitory pathways involving CVLM neurons leading to an inhibition of RVLM neurons. This powerful feedback inhibition of sympathetic vasomotor neurons in the RVLM could mask any differential changes in regional vascular resistance induced by the microinjection of Lglutamate into specific subregions of the RVLM. However, it is important to note that several studies using chemical or electrical stimulation of RVLM observed a differential control of sympathetic activity by RVLM neurons, in spite of a major pressor response (Ross et al., 1984a,b; Dampney et al., 1985; Willete et al., 1983, 1987; Dampney and McAllen, 1988; McAllen and Dampney, 1990; Dean et al., 1992; Dampney, 1994; McAllen and May, 1994; Sun, 1995).
It is also necessary to consider the fact that our experiments were performed in awake rats while those experiments indicating a topographical distribution of sympathetic vasomotor neurons were all performed in anesthetized animals. The anesthetics may have a distorting effect on several neural mechanisms, particularly on neurotrasmission, producing in some cases opposite cardiovascular responses to the excitation of the paraventricular nucleus (Kannan et al., 1989) or NTS (Machado and Bonagamba, 1992) when compared with the data obtained in awake rats. Therefore, it is plausible to suggest that the absence of anesthesia in our experiments may explain, at least in part, our different findings in relation to other studies. A differential control of regional sympathetic outflow certainly exists in rats with intact arterial baroreceptors, as documented by Beluli and Weaver (1991), and a differential control of the regional vascular resistance by the baroreceptor (de Paula et al., 1999) and chemoreceptor afferents is also reported (unpublished data from our laboratory). However, the study by Beluli and Weaver (1991) suggest that the sympathetic vasomotor neurons in the RVLM of rats are not organized viscerotopically, which is compatible with our findings. However, it is important to consider that the quantitative change in sympathetic nerve activity recorded by Beluli and Weaver (1991) may not be necessarily converted to an equivalent change in blood flow. As such, we cannot not rule out the possibility that the measurement of regional blood flow may not be sensitive enough to detect discrete differences in relation to differential sympathetic control of the regional vascular resistance by RVLM neurons.
Acknowledgements The authors thank Leni G.H. Bonagamba and Jaci A. Castania for excellent technical assistance and Rubens F. de Melo for the histological preparations. This work was ˜ de Amparo a` Pesquisa do Estado supported by Fundac¸ao ˜ Paulo (FAPESP), Conselho Nacional de Desenvolde Sao ´ ´ vimento Cientıfico e Tecnologico (CNPQ) and Programa ˆ ´ de Apoio aos Nucleos de Excelencia (PRONEX).
References Beluli, D.J., Weaver. L.C., 1991. Differential control of renal and splenic nerves without medullary topography. Am J. Physiol., 260 (Heart Circ. Physiol.): H1072–H1079. Bonham, A.C., Jeske, I., 1989. Cardiorespiratory effects of DLhomocysteic acid in caudal ventrolateral medulla. Am. J. Physiol., 256 (Heart Circ. Physiol.): H688–H696. Campos, Jr, R.R., McAllen, R.M., 1997. Cardiac sympathetic premotor neurons. Am J. Physiol., 272 (Regulatory Integrative Comp. Physiol. 41): R615–R620. Dampney, R.A.L., 1994. Functional organization of central pathways regulating the cardiovascular system. Physiol. Rev. 74, 323–364.
P.M. de Paula, B.H. Machado / Autonomic Neuroscience: Basic and Clinical 87 (2001) 301 – 309 Dampney, R.A.L., McAllen, R.M., 1988. Differential control of sympathetic fibres supplying hindlimb skin and muscle by subretrofacial neurons in the cat. J. Physiol. 395, 41–56. Dampney, R.A.L., Goodchild, A.K., Tan, E., 1985. Vasopressor neurons in the rostral ventrolateral medulla of the rabbit. J. Auton. Nerv. System 14, 239–254. Dean, C., Seagard, J.L., Hopp, F.A., Kampine, J.P., 1992. Differential control of sympathetic activity to kidney and skeletal muscle by ventral medullary neurons. J. Auton. Nerv. System 37, 1–10. de Paula, P.M., Castania, J.A., Salgado, H.C., Machado, B.H., 1999. Hemodynamic responses to electrical stimulation of the aortic depressor nerve in awake rats. Am. J. Physiol. 277 (Regulatory Integrative Comp. Physiol. 46): R31–R38. de Paula, P.M., Machado, B.H., 1997. Microinjection of L-glutamate into different sites of rostral ventrolateral medulla of unanesthetized rats produced similar changes in regional vascular resistance. J. Auton. Nerv. Syst. 65, 82. Ermirio, R., Ruggeri, P., Molinari, C., Weaver, L.C., 1993. Somatic and visceral inputs to neurons of the rostral ventrolateral medulla. Am. J. Physiol. 265 (Regulatory Integrative Comp. Physiol.): R35–R40. Guyenet, P.G., 1990. Role of the ventral medulla oblongata in blood pressure regulation. In: Loewy A.C., Spyer, K.M. (Eds.), Central Regulation of Autonomic Functions. New York Oxford, Oxford University Press, pp. 145–167. Haywood, J.R., Shaffer, R.A., Fastenow, C., Fink, G.D., Brody, M.J., 1981. Regional blood flow measurement with pulsed Doppler flowmeter in conscious rat. Am J. Physiol., 241 (Heart Circ. Physiol.): H273–H278. Kannan, H., Hayashida, Y., Yamashita, H., 1989. Increase in sympathetic outflow by paraventricular nucleus stimulation in awake rats. Am. J. Physiol. 256 (Regulatory Integrative Comp. Physiol. 25): R1325– R1330. Lovick, T.A., 1987. Differential control of cardiac and vasomotor activity by neurones in nucleus paragigantocellularis lateralis in the cat. J. Physiol. 389, 23–35. Machado, B.H., Bonagamba, L.G.H., Castania, J.A., Menani, J.V., 1992. Changes in vascular resistance during carotid occlusion in normal and baroreceptor-denervated rats. Hypertension 19 (Suppl. II), II149– II153. Machado, B.H., Bonagamba, L.G.H., Castania, J.A., Menani, J.V., 1994.
309
Aortic baroreceptors play a predominant role in the regulation of hindlimb vascular resistance in rats. Am. J. Physiol., 267 (Regulatory Integrative Comp. Physiol. 36): R476–R480. Machado, B.H., Bonagamba, L.G.H., 1992. Microinjection of L-glutamate into the nucleus tractus solitarii increases arterial pressure in conscious rats. Brain Res. 576, 131–138. McAllen, R.M., May, C.N., 1994. Differential drives from rostral ventrolateral medullary neurons to three identified sympathetic outflows. Am. J. Physiol., 267 (Regulatory Integrative Comp. Physiol. 36): R935–R944. McAllen, R.M., Dampney, R.A.L., 1990. Vasomotor neurons in the rostral ventrolateral medulla are organized topographically with respect to type of vascular bed but not body region. Neurosci. Lett. 110, 91–96. Paxinos, G., Watson, C., 1986. The Brain in Stereotaxic Coordinates. Academic Press, New York. Ross, C.A., Ruggiero, D.A., Park, D.H., Joh, T.H., Sved, A.F., FernandezPardal, J., Saavedra, J.M., Reis, D.J., 1984a. Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area containing C1 adrenaline neurons on arterial pressure, heart rate, and plasma catecholamines and vasopressin. J. Neurosci. 4, 474–494. Ross, C.A., Ruggiero, D.A., Joh, T.H., Park, D.H., Reis, D.J., 1984b. Rostral ventrolateral medulla: selective projections to the thoracic autonomic cell column from the region containing C1 adrenaline neurons. J. Comp. Neurol. 228, 168–185. Spyer, K.M., 1990. The central nervous organization of reflex circulatory control. In: Loewy A.C., Spyer, K.M. (Eds.), Central Regulation of Autonomic Functions, New York, Oxford. Oxford University Press. Sun, M-K., 1995. Central neural organization and control of sympathetic nervous system in mammals. Prog. Neurob. 47, 157–233. Urbanski, R.W., Sapru, H.N., 1988. Putative neurotransmitters involved in medullary cardiovascular regulation. J. Auton. Nerv. Syst. 25, 181– 193. Willete, R.N., Barcas, P.P., Krieger, A.J., Sapru, H.N., 1983. Vasopressor and depressor areas in the rat medulla. Neuropharmacology 22 (9), 1071–1079. Willete, R.N., Punnen-Grandy, S., Krieger, A.J., Sapru, H.N., 1987. Differential regulation of regional vascular resistance by the rostral and caudal ventrolateral medulla in the rat. J. Auton. Nerv. Syst. 18, 143–151.