INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

DIFFERENTIAL RESPONSES of lumbar (LSNA), renal (RSNA), or adrenal sympathetic nerve activity (ASNA) to baroreceptor stimuli and other cardiovascular reflexes have been extensively studied in the rat. However, no direct and simultaneous comparisons of the full range of baroreflex reactivity for these sympathetic outputs have been described. In most studies, bolus injections of vasoactive drugs or a "slow ramp" of blood pressure changes (MAP 40-50 mmHg/30-60 s) is used. These methods do not allow for full expression of steady-state baroreflex sympathetic responses and may be the reason for disparate results from different laboratories (see Tables 1 and 2).

Baroreflex regulation of sympathetic nerve activity (SNA) consists of phasic and tonic components (24). The phasic component provides transient, "beat to beat" regulatory corrections for abrupt, short-lasting (up to a few seconds) disturbances of arterial pressure and is likely responsible for pulse pressure modulation of SNA. The tonic component provides long-lasting regulatory adjustments. For example, long-term baroreflex resetting in response to sustained elevation in MAP or electrical stimulation of the aortic nerve may be responsible for prolonged inhibition in RSNA lasting over 45 min after return in MAP to control or cessation of the stimulation (27, 41). Also, most of the neuromodulators of central baroreflex mechanisms, e.g., polypeptides, purines, or monoamines, exert long-lasting, tonic effects on SNA on microinjection into the nucleus of the solitary tract (28, 35-37). These responses last from several minutes to 1 h or longer. In this respect, evaluation of sympathetic responses evoked by baroreceptor stimulation and unloading with bolus injections of vasoactive drugs or slow ramp blood pressure changes may reflect mostly the phasic component of the regulation while missing the tonic component, which is important for the long-term adjustment of the sympathetic system to different physiological conditions. In contrast, steady-state baroreflex responses to gradual (step by step) increases and decreases in arterial pressure reflect both components of the baroreflex regulatory mechanisms (7). The latter method seems to be particularly suitable to compare differential control of regional sympathetic outputs by baroreflex mechanisms and long-lasting differential effects evoked by neuromodulators administered into specific brain stem baroreflex centers (35-37).

Whenever precise evaluation of differential baroreflex effects on regional sympathetic outputs is required, it is important to compare simultaneous responses from different sympathetic nerves recorded in one animal rather than the responses obtained in different animals. A direct comparison of the absolute level of neural activity recorded in whole nerve preparation is not possible because of nonphysiological factors (e.g., nerve-electrode contact, size of nerve bundle, amount and conductance of extracellular fluid between the electrodes, etc.). Usually, resting nerve activity is normalized to 100% and only relative changes of the activity may be compared. Therefore baroreflex stimulus-response functions obtained for SNA in different animals may start from relatively different resting levels determined by several physiological variables (e.g., surgical stress, temporal variation of the anesthesia, ventilation, blood volume, etc.) Consequently, comparison of baroreflex functions for two different nerves recorded in two different animals may contain a substantial and unknown error. This error is further enhanced if comparisons are made between neural recordings performed in different laboratories using different methods of measuring SNA (e.g., integrated electrical signal vs. burst frequency). Although transient baroreflex responses of separately recorded RSNA, LSNA, and ASNA may be found in numerous studies, no clear conclusions can be made from these comparisons. Therefore, we compared steady-state sigmoidal baroreflex stimulus-response curves for RSNA recorded simultaneously with LSNA or ASNA in urethan-chloralose-anesthetized male Sprague-Dawley rats. Our results indicate that baroreflex regulation of these sympathetic outputs shows wide regional variability in gain, range, and maximal inhibition.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

All protocols and surgical procedures employed in this study were reviewed and approved by the Institutional Animal Care and Use Committee and were performed in accordance with the Guiding Principles in the Care and Use of Animals endorsed by the American Physiological Society and published by the National Institutes of Health.

Design. Baroreflex reactivity of simultaneously recorded regional sympathetic outputs (RSNA and LSNA or RSNA and ASNA) was compared in 21 male Sprague-Dawley rats (350-400 g) (Charles River) under urethan-chloralose anesthesia. Full-range, steady-state baroreflex stimulus-response curves were assessed by increasing or decreasing MAP in steps lasting at least 1 min each via intravenous infusion of cumulative doses of phenylephrine (PE) or nitroprusside (SNP), respectively. Experimental data were approximated to logistic functions and expressed in terms of the classical mathematical model for evaluation of baroreflex reactivity (23).

Instrumentation and measurements. Rats were anesthetized with a mixture of -chloralose (80 mg/kg) and urethan (500 mg/kg ip). A steady-state level of anesthesia was maintained by a continuous infusion (-chloralose, 8-16 mg · kg¹ · h¹, and urethan, 50-100 mg · kg¹ · h¹). The level of anesthesia was adjusted for each animal to completely inhibit corneal and "pedaling" reflexes. Rectal temperature was maintained between 37 and 38°C by a water heating pad (American Pharmceal, model K20). The rats were tracheotomized and allowed to spontaneously breathe room air. Catheters (PE-50) were inserted into the right femoral artery and both femoral veins to monitor arterial blood pressure, infuse vasoactive drugs, and continuously supplement anesthesia. Arterial pressure was measured with a Gould P23 ID pressure transducer, which was coupled to a Grass amplifier (model 7P122P). The arterial pressure and neural signals were digitized and recorded with Hemodynamic and Neural Data Analyzer (Biotech Products, Greenwood, IN). MAP and SNA were averaged over 1-s intervals. All data were stored on hard disk for subsequent analysis.

Neural recording. The lumbar and renal nerves were exposed through an abdominal incision in 12 animals. The left renal nerve was separated along a peripheral portion of the renal artery. The left lumbar sympathetic trunk was dissected at the level of L₃-L₅. In a second group of nine animals, the renal and adrenal nerves were exposed through a flank incision and a retroperitoneal approach. The same portion of the renal nerve as used for simultaneous recordings with lumbar nerve was separated from the renal artery. The adrenal nerve was exposed between the splanchnic ganglion and the adrenal gland. Usually the major bundle of the adrenal nerve directed to the caudal portion of adrenal gland was used. Approximately 5-8 mm of each nerve was dissected, cut peripherally to eliminate any afferent activity, and placed on stainless steel electrodes. Electrodes were constructed of multistranded stainless steel Teflon-coated wire (Medwire, Mount Vernon, NY) with the exposed ends wound into a single loop. The nerve bundles and electrodes were embedded in silicone gel (Wacker Sil-Gel 601A and 601B mixture) that was allowed to harden. The silicone gel provides electrical insulation of the nerve-electrode junction from surrounding body fluids, which results in a good signal-to-noise ratio and also mechanically protects the electrodes. When this technique is used, contact between the nerve and electrodes is stable over many hours. Finally, the wound was closed and the animal was allowed to stabilize for at least 45 min before starting the protocol.

Neural signals were initially amplified (2,000-10,000×) with a Grass P511 differential preamplifier and a high-impedance probe (HIP 511GA). The probe and animals were located inside a shielded Faraday cage. The amplification band was set at 100-1,000 Hz. Nerve activity was digitized, rectified, and averaged in 1-s intervals. Background noise was determined 30-60 min after the animal was euthanized. A direct comparison of the absolute level of nerve activity is not possible because of nonphysiological factors (e.g., nerve-electrode contact, size of nerve bundle). Thus resting nerve activity was normalized to 100%.

The ratio of preganglionic fiber activity to total nerve activity was evaluated at the end of each experiment as the percentage of neural activity remaining after ganglionic blockade (hexamethonium, 20 mg/kg iv). RSNA was almost completely postganglionic; only 2.8 ± 0.9% of the activity persisted after the ganglionic blockade. LSNA directed mostly to somatic vascular bed (hindquarter) was predominantly postganglionic; only 10.5 ± 2.0% of the activity remained after ganglionic blockade. The adrenal nerve consists of several separate bundles containing both pre- and postganglionic fibers with a very different ratio for each bundle. Therefore, according to the criteria of Carlsson and Skarphedinsson and co-workers (3, 4), ASNA was considered as predominantly preganglionic if the activity increased over 100% after administration of hexamethonium because of baroreflex responses caused by the decrease in MAP. Average ASNA after ganglionic blockade was 132.6 ± 10.5%.

Baroreflex response curves. To evaluate baroreflex control of SNA, arterial pressure was increased and decreased via intravenous infusion of PE (200 mg/ml) and SNP (200 mg/ml), respectively. Graded constant infusions were performed to produce stepwise (5-8 steps), sustained changes (1-2 min) in MAP to allow adequate time for full expression of steady-state sympathetic response. The doses of SNP and PE ranged from 2.5 to 80 mg · kg¹ · min¹. Total volume administered during gradual increases or decreases in MAP was 300 to 450 µl. Approximately 8-12 min were required to elevate blood pressure from resting values to ~160 mmHg or to decrease blood pressure to ~40 mmHg. Increases and decreases in MAP were separated by at least 15-min intervals.

Data analysis. During PE and SNP infusions, SNA and MAP were averaged over 10 s of steady-state sustained changes in MAP. Resting values recorded before PE or SNP infusions were averaged for each curve. Sigmoidal logistic curves were approximated to experimental data points (10-18 points per curve) for each animal, according to the model described by Kent et al. (23), with the use of SYSTAT statistical software for Windows, version 5.02, and the formula

(1)

where P₁ is the upper plateau of the curve, P₂ is the lower plateau, P₃ is a coefficient describing the distribution of gain along the curve, and P₄ is MAP in the midpoint of the curve (BP₅₀). An example of this analysis is presented in Fig. 2. The distribution of gain as a function of MAP was calculated as the first derivative of equation 1. Maximum gain (G_max) was calculated according to the formula G_max = P₃ · (P₁P₂)/4. The range of baroreflex regulation was calculated as the difference between the upper and lower plateaus of the baroreflex curve (range P₁-P₂). Baroreflex response curves were constructed, and their parameters were calculated for each nerve in each animal and then averaged across animals.

A one-way ANOVA was used to evaluate the differences between all parameters characterizing the baroreflex response curves generated for each pair of nerves in each group of animals (RSNA vs. LSNA and RSNA vs. ASNA). A one-way ANOVA for independent measures was used to compare parameters characterizing the baroreflex curves between groups (LSNA vs. ASNA and RSNA vs. RSNA). The comparison of RSNA baroreflex curves recorded in both groups of animals allowed us to evaluate if experimental conditions and baroreflex reactivity between groups were similar. Consequently it allowed for reliable comparison between ASNA and LSNA recorded in separate groups of animals. Modified Bonferroni tests were used to compare individual means. In addition, significance of differences between averaged baroreflex functions calculated for each pair of nerves was evaluated across the MAP range in 10-mmHg intervals using paired t-test with Bonferroni adjustment for repeated measures (see Fig. 3). Also the MAP ranges for which the baroreflex gain was >30% of G_max were compared between all three sympathetic outputs (see Fig. 4). An -level of P < 0.05 was used to determine statistical significance.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

There were no differences between resting blood pressure levels (82.2 ± 2.0 vs. 85.7 ± 3.4 mmHg, P > 0.05) for the two groups of animals where RSNA vs. LSNA and RSNA vs. ASNA were compared, respectively. Examples of the neural responses to gradual increases and decreases in MAP are presented in Fig. 1 and the approximations of these responses to sigmoidal baroreflex function curves are shown in Fig. 2. Steady-state increases in MAP resulted in gradual inhibition of neural activity, whereas steady-state decreases in MAP resulted in gradual enhancement of neural activity in all three sympathetic outputs; however, these responses were nonuniform.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Sympathetic nerve activity (SNA) responses to gradual increases (top) and decreases (bottom) in mean arterial pressure (MAP). Maximal inhibition of lumbar SNA (LSNA) was smaller than the inhibition of renal SNA (RSNA) and adrenal SNA (ASNA), whereas maximal increase in ASNA was higher than the increases observed in RSNA and LSNA.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Examples of approximation of sigmoidal baroreflex curves to experimental data points obtained from the recordings presented in Fig. 1. In these 2 experiments baroreflex responses of RSNA vs. LSNA (top) and RSNA vs. ASNA (bottom) were compared.

Averaged baroreflex response curves for paired comparisons of SNA (RSNA vs. LSNA and RSNA vs. ASNA) are presented in Fig. 3. Characteristics of the baroreflex curves differed significantly between all three sympathetic outputs. In response to increases in MAP, LSNA was inhibited to a significantly lesser extent than RSNA and ASNA for MAP >100 mmHg. In response to decreases in MAP, the increase in ASNA was significantly higher than the increases in both RSNA (for MAP <70 mmHg) and LSNA (for MAP <60 mmHg). However, there were no differences between baroreflex responses of all three nerves in the range of MAP close to the operating point (70 mmHg < MAP < 100 mmHg).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Averaged baroreflex response curves for 2 pairs of comparisons of SNA. Vertical dashed lines indicate the operating points for each curve (control level of MAP). There were no differences between resting blood pressure levels for 2 groups of animals or between operating points for each comparison (P > 0.05).

Because baroreflex curves for RSNA recorded in two different groups of animals were not different in terms of G_max, P₁, P₂, P₃, or P₄ (P > 0.05 for all comparisons), these data were combined for calculation of gain (Fig. 4) and other parameters characterizing the baroreflex stimulus-response functions (Table 3). Lack of differences between baroreflex curves for RSNA indicated that physiological conditions and baroreflex reactivity between the groups were similar. This allowed for reliable comparison of ASNA vs. LSNA although these parameters were recorded in separate groups of animals. ASNA exhibited the greatest range of baroreflex reactivity and highest maximum values. LSNA exhibited the smallest range of regulation and was inhibited to a significantly lesser extent than two other nerves. RSNA exhibited a significantly greater maximal gain than LSNA (P = 0.00014). G_max for ASNA tended to be greater than that for LSNA and smaller than that for RSNA (Table 3 and Fig. 4); however, the differences did not reach statistical significance (P = 0.090 and 0.080 vs. LSNA and RSNA, respectively). The gain for ASNA was distributed over a significantly wider range of MAP (P < 0.05) in comparison to both other sympathetic outputs. In other words, RSNA and LSNA are most sensitive to relatively small blood pressure changes around the operating point, whereas significant baroreflex sensitivity (>30% of G_max) in ASNA extends to both lower and higher blood pressure vs. that for RSNA or LSNA (Fig. 4). There were no significant differences between the midpoints of the curves (BP₅₀).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Baroreflex gain for RSNA, LSNA, and ASNA (top) and the range of MAP in which the gain was >30% of maximum gain (Gmax; bottom). * P < 0.05 vs. RSNA; # P < 0.05 vs. LSNA. Because baroreflex curves for RSNA recorded in 2 different groups of animals were not different, these data were combined.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This is the first study to compare the full range of steady-state baroreflex stimulus-response functions for RSNA recorded simultaneously with LSNA or ASNA. The major finding was that baroreflex regulation of these sympathetic outputs differs significantly among each other, showing regional variability in gain, range, and maximal inhibition. ASNA exhibited the greatest range of baroreflex regulation, the highest upper level of activity, and the widest distribution of the gain over the range of MAP. RSNA exhibited greater gain than LSNA. LSNA showed the smallest range and maximal inhibition in comparison to other sympathetic outputs. However, all three nerves responded similarly to baroreflex stimulation and unloading in the range of MAP close to the operating point.

Comparisons of baroreflex responses. Although baroreflex responses in RSNA, LSNA, and ASNA were analyzed in several studies, in all of them but one (45) the activity of each nerve was recorded in separate animals using different methods of evoking baroreflex responses and analyzing the data. Therefore conclusions from these comparisons are unclear and sometimes contradictory (see Tables 1 and 2). However, there are some general regularities across the studies performed in anesthetized rats that are similar to our results (Table 1). For example, despite the methods used in most studies, the gain for RSNA (1.67-6.24 %/mmHg) is greater than the gain for LSNA (0.95-2.4 %/mmHg) and the lower plateau for RSNA (4-15 %) is closer to zero activity than that for LSNA (14.5-29%). The "quasichronic" experiments show similar tendencies as those observed in experiments using anesthetized animals (see Table 2). Gain for RSNA tends to be greater than that for LSNA, and maximal inhibition in LSNA tends to be less effective in comparison to other outputs. In general, quasichronic experiments, in comparison to those performed with anesthetized animals, show slightly greater gains and markedly greater upper plateaus for all sympathetic outputs (Tables 1 and 2; i.e., compare Refs. 5 and 6).

Baroreflex responses of ASNA and RSNA were compared in only three studies (3, 38, 45). In two of them, only limited increases in MAP were used and the data were analyzed as a linear function (3, 45). In the other study, RSNA and ASNA were recorded in separate groups of quasiconscious rats (4-6 h after surgery) and under these conditions recent surgical stress may obscure any differences in reflex responses between groups (38). Although the authors did not find significant differences in baroreflex regulation of ASNA vs. RSNA under these circumstances, in all three studies maximal increases in ASNA tend to be greater than those for RSNA, consistent with our observations.

Our results are consistent with other studies indicating that the sympathetic output to the viscera has a greater range of baroreflex regulation than that to muscle or skin (12, 29).

Physiological significance of differential responses. Maximal gain for RSNA was significantly higher than that for LSNA and tended to be higher than the gain for ASNA. This may be related to a greater responsiveness of RSNA than other sympathetic outputs to afferent baroreceptor C-fiber stimulation (1, 34). Because baroreflex responses to C-fiber stimulation develop slowly and cause very powerful sympathoinhibition (1, 11, 27), it is likely that the steady-state method of stimulation of baroreceptors applied in our study allows C-fiber baroreceptor responses to fully develop; these responses contributed to the high gain for RSNA. Therefore the method used in our study was able to detect differences in gain that are not so evident when transient stimulation is used, especially when different nerve activity is recorded in different animals (Tables 1 and 2). Interestingly, greater gain for RSNA is consistent with the powerful cardiac modulation of this sympathetic output by the pulse pressure wave at resting MAP. Smaller modulation is observed in LSNA, and weak or no modulation is observed in preganglionic ASNA (3).

It should be stressed that the gain for ASNA is distributed over a significantly greater range of MAP than that for two other nerves (Fig. 4). In other words, ASNA still responds proportionally to changes in MAP when the other sympathetic nerves have already reached their plateaus. It is likely that RSNA and LSNA participate mostly in phasic baroreflex regulation on a beat-to-beat basis, closer to the operating point, whereas ASNA may be involved mainly in long-term regulatory responses over a wider range of MAP. This "inert" feature of ASNA is consistent with the greatest range of baroreflex regulation, the highest maximum increase in ASNA vs. RSNA and LSNA, weak cardiac modulation of ASNA, and the humoral nature of this regulatory mechanism operating via release of catecholamines into the blood stream. In fact, long-lasting decreases in MAP during hemorrhagic shock are accompanied by transient sympathoexcitation reversed into sympathoinhibition in RSNA and postganglionic ASNA; however, preganglionic ASNA directed to the adrenal medulla increases gradually and remains markedly elevated under these circumstances (3, 45). Many other experimental factors show dissociation between responses of ASNA and RSNA. For example, hypoglycemia leads to a greater increase in ASNA than RSNA (4); stimulation of cardiac chemoreceptors or paraventricular nucleus of the hypothalamus or intravenous injections of morphine increase ASNA while inhibiting RSNA (9, 15, 22). Interestingly, this dissociation is related only to preganglionic fibers directed to the adrenal medulla, whereas postganglionic fibers directed to the adrenal cortex and vascular system of the adrenal gland react similarly to RSNA (3, 4). We made a similar observation in three additional experiments where RSNA was simultaneously recorded with predominantly postganglionic ASNA (only 37.7 ± 8.0% of activity remained after ganglionic blockade). There were no differences between these two outputs in BP₅₀, range, and upper and lower plateaus of baroreflex responses; however, G_max for RSNA still tended to be higher than that for ASNA (4.9 ± 1.0 vs. 3.8 ± 0.8 %/mmHg, respectively; P = 0.016).

We did not observe differences between maximal inhibition of ASNA and RSNA; however, LSNA was inhibited to a lesser extent than two other sympathetic outputs. The smaller inhibition of LSNA suggests that relatively few nerve fibers of the lumbar sympathetic trunk may be influenced by the baroreceptor reflex. Alternatively, activation of the baroreceptor reflex may increase the activity of some lumbar sympathetic fibers such that the overall suppression of LSNA appears to be less than in other sympathetic nerves. This is consistent with the nonhomogeneous content of sympathetic fibers in the lumbar sympathetic trunk, which contains fibers directed to the hindlimb and the tail, a major thermoregulatory organ in the rat. For example, in normothermic rats, tail skin blood flow is very low and cutaneous vasodilation in response to increases in MAP is limited, suggesting that powerful, tonic sympathetic vasoconstrictor activity directed to cutaneous vessels cannot be fully inhibited by stimulation of baroreceptors (32). Similarly, direct neural recordings showed that in response to the same increase in MAP, sympathetic fibers directed to the skin are inhibited to a lesser extent than those directed to the muscles or to the kidney in normothermic rats and cats (12, 29). In addition, there is recent evidence that lumbar sympathetic nerves mediate active sympathetic vasodilation in rat hindlimb via release of nitric oxide from sympathetic terminals (8). These features of LSNA should be taken into consideration if LSNA is used as an index of overall sympathetic activity.

Perspectives