Locus Coeruleus Noradrenergic Modulation of Diurnal Corticosterone, Stress Reactivity and Cardiovascular Homeostasis in Male Rats

Abstract

Introduction:

Activation of the locus coeruleus-noradrenergic (LC-NA) system during awakening is associated with an increase in plasma corticosterone and cardiovascular tone. These studies evaluate the role of the LC in this corticosterone and cardiovascular response.

Methods:

Male rats, on day 0, were treated IP with either DSP4 (50 mg/ kg body weight) (DSP), a LC-NA specific neurotoxin, or normal saline (SAL). On day 10, animals were surgically prepared with jugular vein [Hypothalamic–pituitary–adrenal (HPA) axis] or carotid artery (hemodynamics) catheters and experiments performed on day 14. HPA axis activity, diurnally (circadian) and after stress [transient hemorrhage (14 mL/kg body weight) or airpuff-startle], and basal and post-hemorrhage hemodynamics were evaluated. On day 16, brain regions from a subset of rats were dissected for norepinephrine and corticotropin-releasing factor (CRF) assay.

Results:

In DSP rats compared to SAL rats: 1) regional brain norepinephrine was decreased but there was no change in median eminence or olfactory bulb CRF content; 2) during HPA axis acrophase, the plasma corticosterone response was blunted; 3) after hemorrhage and airpuff-startle, the plasma adrenocorticotropic hormone response was attenuated, whereas the corticosterone response was dependent on stressor category; 4) under basal conditions, hemodynamic measures exhibited altered blood flow dynamics and systemic vasodilation; and 5) after hemorrhage, hemodynamics exhibited asynchronous responses.

Conclusion:

LC-NA modulation of diurnal and stress-induced HPA axis reactivity occurs via distinct neurocircuits. The integrity of the LC-NA system is important to maintain blood flow dynamics. The importance of increases in plasma corticosterone at acrophase to maintain short- and long-term cardiovascular homeostasis is discussed.

Graphical Abstract

LC-NA modulation of HPA axis and hemodynamics DOI: #################

Introduction

The locus coeruleus (LC), the A6 noradrenergic (NA) cell group, is located in the dorsal rostral part of the brain stem on the lateral floor of the fourth ventricle. It projects the full length of the neuroaxis with extensive projections to cortical regions and the spinal cord [1–3]. The LC neurons are organized along the rostral-ventral extent of the nucleus. Neurons in the rostral region projecting towards forebrain and those in the ventral regions projecting towards spinal cord [2, 4] impart a functional dichotomy of the LC-NA system [5]. As a major component of the reticular activating system, the LC modulates alertness, attention, arousal, sleep-wake cycle, and diurnal/circadian activity patterns [1, 6–7]. The sleep-to-wake transition is also accompanied by tandem increases in cardiovascular tone and plasma glucocorticoids, corticosterone (CORT) in rat and cortisol in humans [1, 8–9]. However, despite considerable importance, the specific role of the LC in the modulation of many neuronal and physiological functions remains to be elucidated [1].

The increase in the plasma CORT during the sleep-to-wake transition coincides with the acrophase of hypothalamic–pituitary–adrenal (HPA) axis activity. This increase in CORT is necessary for glucocorticoid type II receptor function, gene expression and protein synthesis necessary to cope with any ensuing challenges to physiological and behavioral homeostasis. The CORT rhythm is known to regulate cognition, feedback, gluconeogenesis, immune, inflammatory and vascular sympathetic tone, many of the same functions that the LC is known to modulate [8, 10–14]. Whereas adrenocorticotropic hormone (ACTH) acts on the adrenal cortex to increase plasma CORT [15–16], at acrophase the increase is modulated by sympathoadrenal activity which primes adrenocortical sensitivity to ACTH and generates the surge in the plasma CORT [16–18]. This sympathetic innervation of the adrenals is predominantly from preganglionic fibers arising from spinal cord segments [19]. The paraventricular nucleus of the hypothalamus (PVN), a brain region with second order projections to the adrenals [20], is considered important for sympathoadrenal activation at acrophase [11, 18]. However, PVN lesions do not abolish the surge in plasma CORT at acrophase [21–22].

During the transition state, signals for arousal arising from pacemaker cells in the suprachiasmatic nucleus of the hypothalamus (SCN) communicate to the LC via the dorsomedial hypothalamic nucleus (DMH) [1, 3, 23]. Interestingly, DMH lesions suppress increases in plasma CORT at acrophase [24–25]. Because the DMH projection to the adrenals is third order and LC projection is second order [20], it was suggested that the LC may act as the relay between the DMH and the adrenals. However, the role of LC in the regulation of HPA axis activity and specifically, increases in plasma CORT at acrophase is not known.

LC-NA activity (as well as plasma ACTH and CORT) is also increased after somatic and visceral stressors including airpuff-startle and hemorrhage, and the LC is considered an integral part of the stress circuitry [3, 7, 26–30]. After hemorrhage, excitatory and inhibitory inputs to LC exhibit a spatial organization [29, 31] and such organization is considered important for functional heterogeneity within the LC [7, 10, 32]. It is also known that an increase in LC activity alters corticolimbic activity modifying PVN neuronal activity and HPA axis response [27]. However, little is known about the effect of corticolimbic circuitry in the modulation of HPA axis response after hemorrhage. Stress signals after hemorrhage are primarily conveyed to the PVN via medullary A2 noradrenergic neurons [28, 33]. However, an interaction, if any, between the medullary circuitry and the corticolimbic circuitry remains to be investigated. Corticolimbic circuitry is also known to modulate the HPA axis response to airpuff-startle [34]. However, the role of the LC in this regulation remains unknown.

The LC is also known to regulate sympathetic nervous system (SNS) activity and cardiovascular homeostasis [3, 14, 30]. While a hierarchical role for the PVN in the regulation of LC activity has been proposed [3], lesions of the region do not affect basal or post-stressor hemodynamic responses [35–37]. Yet, the specific role(s) of the LC in the modulation of cardiovascular homeostasis is not known.

Accordingly, the present studies were designed to evaluate the role of the LC-NA system in the regulation of: 1) HPA axis diurnal and stressor-induced (hemorrhage and airpuff-startle) responses, and 2) the basal and post-hemorrhage hemodynamic response in rats treated systemically with DSP4 (N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine), a neurotoxin that denervates central LC-NA efferents [38–39].

DSP4 is a selective neurotoxin that denervates LC-NA efferents. Because it crosses the blood–brain barrier, DSP4 can be administered intraperitoneally [38–40] and does not require surgical manipulations. After administration, DSP4 cyclizes to form an aziridinium derivative, which is an irreversible inhibitor of the norepinephrine transporter and inhibits norepinephrine reuptake with long lasting degenerative effects on LC-NA efferents [38–40]. Thus, DSP4 eliminates the LC-NA spinal projection, but spares projections of the A5 and A7 noradrenergic cell groups [41]. Similarly, DSP4 spares A1 noradrenergic [42] and A2 noradrenergic [43] cell groups. Because of the specificity towards LC-NA efferents, the DSP4 treated rat model (DSP), was considered ideal for the present studies. Rats treated with normal saline (SAL), served as controls.

Materials and methods

Animals

Adult male Sprague–Dawley (SD) rats obtained from Harlan (<www.envigo.com>) were used. After arrival at the vivarium, they were housed, 3/cage, maintained under standard husbandry conditions with controlled humidity and temperature, on a 12:12 h light–dark cycle (lights ON at 0700 h), with food and water available ad libitum. The rats were used in the experimental studies after a one week acclimatization period (body weight: 230 – 290 g). Studies were conducted according to NIH Guidelines and protocols approved by the Institutional Animal Care and Use Committees.

DSP4 (DSP-4) solution

The LC-NA specific neurotoxin, DSP4 [38–39] was obtained commercially (Sigma, <www.sigmaaldrich.com>). In the present study, 100 mg DSP4 HCl (≈ 88 mg DSP4 base) was dissolved in 1760 μL deionized water (5 mg DSP4 base per 100 μL). The solution was made fresh before use, kept protected from direct light, and administered intraperitoneally (IP) at a dose of 100 μL/ 100 g body weight (50 mg/ kg, body weight) [38–39].

Experimental groups

At the end of the acclimatization period, body weights were recorded and the rats were distributed into two groups, SAL and DSP, matched closely for body weight. This was followed by IP, 100 μL/ 100 g body weight normal saline and DSP4 solution, to SAL and DSP rats, respectively. Thereafter, the rats were individually housed and maintained under standard husbandry conditions.

Experimental design

The day of SAL or DSP4 administration was designated as day 0. Body weight was recorded every other day until day 10. On day 10, DSP rats with greater than 5% decrease in body weight relative to their initial body weight were excluded (n=4) from the study. Remaining rats underwent surgery for catheter implantation under anesthesia [44]. After recovery of ambulation, the rats were individually housed in experimental cages [28, 45].

During stress studies data was collected for a period of 30 min. At the termination of the experiment, the rats were euthanatized with Euthasol® (Henry Schein, <https://www.henryschein.com>) solution in water (1:4), administered (100 to 125 μL/100 g body weight) through the indwelling catheters. Independent cohorts of rats were used for the different stress studies. Rats used for diurnal studies and rats that lost patency of catheters (resistance to blood flow), were euthanatized on day 16 and brains were collected. During this session, each rat received a lower dose (50 μL/ 100g body weight) of the diluted euthanasia solution and was rapidly decapitated with a rodent guillotine. The brains were harvested and regions of interest dissected for the determination of tissue norepinephrine or CRF content.

Catheterization surgery

The rats were prepared with a right carotid artery catheter to evaluate hemodynamic response after hemorrhage or a right jugular vein catheter for blood sampling to evaluate HPA axis characteristics, diurnally and after stressors. The anesthetic and surgical techniques for the implantation were performed according to published protocols [44, 46–47] with the following modifications for carotid artery catheterization. In the construction of carotid artery catheter, the PE-50 tubing was pushed the full length of the silastic tubing (3.0 to 4.0 cm) and the tubing was cut even [46]. During implantation, the catheter was advanced to about 4.0 cm into the artery from the incision site and an extra suture was tied behind the silastic tubing overlap. The length of the tubing exiting from the nape of the neck was cut to 3.5 to 5.0 cm and obturated with 1.0-cm long 22-gauge stainless steel rod / stylet.

Stress category

Rats were exposed to two ethologically relevant categories of stressors, either the physical stressor of transient hemorrhage, or the psychological stressor of airpuff-startle. Transient hemorrhage is both hypovolemic and hypotensive with the stress signals reaching the PVN via circuitry involving vagal afferents [28, 31]. In contrast, airpuff-startle is euvolemic and hypertensive [48] with the stressor signals reaching the PVN via non-vagal afferents [28].

Hemorrhage was performed by withdrawing 14 mL/ kg body weight (≈20% blood volume) in a 5.0 mL syringe containing 100 μL of heparin sodium solution (1000 USP units/ mL) at a constant rate over 3 min [28]. Body weight on day 10 was used to calculate hemorrhage volume. The shed blood was kept warm and at the termination of the experiment retransfused to the rat.

Airpuff-startle was performed by applying one episode of three successive 1 s long air blasts over a 5 s period beginning at time 0 [28].

Experimental protocol

Except for the collection of brain tissue for norepinephrine and CRF assay, all other experiments were carried out on day 14, four days after catheter implantation surgery. In addition, except for the diurnal studies, all other experiments were carried out between 0900 and 1300 h. In studies designed to evaluate the HPA axis response, a blood sample (300 μL) for plasma ACTH and CORT assay was collected and the sample volume was immediately replaced with heparinized saline (20 USP units/mL). Other details are described below.

Evaluation of regional brain norepinephrine and CRF:

Depletion of brain norepinephrine, in proportion to LC-NA afferents in a region, is a major neurotoxic effect of DSP4 in rats [38–39]. Accordingly, studies were designed to evaluate norepinephrine in selected brain regions in SAL and DSP rats. In addition, the effect of DSP4 on tissue CRF content in selected brain regions was also investigated. The dissection of brain regions of interest is described below

After the brain was isolated from the skull, it was collected into ice cold saline in a beaker on ice and swirled gently. It was then placed on a lightly moistened paper towel over ice and the dura and surface blood vessels were removed. The brain was rolled gently over the paper towel to remove adhering moisture and placed on an inverted petri dish cover on ice. With dorsal side up, both olfactory bulbs with stalk/peduncle were dissected. The tip of the brain in front of the stalk was cut with the scalpel and designated as the cortex. With ventral side up, the optic chiasma along with optic nerves were removed followed by dissection of a 3 mm cube of hypothalamus with the PVN. From a separate rat brain, the median eminence, a small swelling surrounding the infundibulum, was dissected with curved iridectomy scissors. Again, with dorsal side up, the two brain halves were separated and stretched on the petri dish. Thereafter, hippocampi from each side were removed and pooled. The brain regions were snap frozen on dry ice and stored at −80°C. The elapsed time starting from opening the skull till the end of dissection was ≤ 3 min.

Evaluation of diurnal HPA axis response:

SAL and DSP rats with indwelling right jugular vein catheters were used in this study. Blood samples for plasma ACTH and CORT assay were collected 1 hr after lights “ON” (nadir, AM/trough) and 30 min before lights “OFF” (acrophase, PM/peak/zenith). Extension tubing was connected at the time of blood sampling with minimal disturbance of the rats.

Evaluation of HPA axis response after stress:

SAL and DSP rats with indwelling right jugular vein catheters were used in this study. After the catheters were connected to extension tubing, the rats were allowed to stabilize for 90 min before initiation of the experimental paradigm which lasted for 30 min. Soon after collection of the 0 min blood sample, the rats were subjected to either hemorrhage or airpuff-startle while residing in their home cages [28].During hemorrhage and airpuff-startle, timed (0, 6, 9, 18 and 30 min) blood samples were collected and processed for plasma ACTH and CORT assay.

Evaluation of hemodynamic response after hemorrhage:

SAL and DSP rats with indwelling right carotid artery catheters were used in this study. On the day of the experiment the rats were transferred to test chambers designed to monitor hemodynamic responses. The carotid artery catheter was connected through a rotating swivel and a three-way stopcock assembly to a calibrated pressure transducer (Statham P23ID; Gould, Cleveland, OH, USA), positioned outside the chamber at the level of the heart. The length of the connection tubing (PE-50) was the same for all rats. Thereafter, blood pressure (force of blood against artery walls) and heart rate (heartbeats per minute) responses were monitored continuously and the rats were allowed to stabilize for 90 min. The unit of blood pressure values is millimeters of mercury (mmHg) and that of heart rate is beats (pulse) per minute (bpm).

At the end of the stabilization period, basal/ control blood pressure and heart rate values were recorded. Thereafter, the carotid artery catheter was disconnected from the transducer and the rats were subjected to hemorrhage by withdrawing 14 mL/ kg body weight blood. The catheter was reconnected to the transducer, recalibrated and hemodynamic responses were monitored until the end of 30 min experimental paradigm.

Hemodynamic responses from four rats were collected simultaneously. For data collection, the phasic pressure outputs were fed to an amplifier-recorder (Model 8802; Gould) and to a computerized data acquisition system equipped with an analog-to-digital converter board that recorded hemodynamic indices. This included heart rate, diastolic blood pressure (diastolic = pressure against artery walls when the heart is at rest (the basal state) between beats and heart muscle relaxes allowing blood to fill the chambers), systolic blood pressure (systolic = pressure against artery walls at the cusp of the beat, when the heart muscle contracts and blood is ejected), and mean arterial pressure (MAP = average pressure in the artery throughout one cardiac cycle, systole to diastole). Other variables were calculated offline. Pulse pressure (mmHg), the net increase in the arterial pressure during one cardiac cycle, was calculated as the difference between systolic and diastolic.

The Liljestrand and Zander equation [49] was adapted to calculate stroke volume (volume of blood ejected during each beat) from systolic and diastolic values. Other hemodynamic indices, cardiac output and systemic vascular resistance (SVR: resistance to blood flow through vasculature) were calculated from heart rate, MAP and stroke volume using standard equations (Table 2).

Table 2.

Basal hemodynamic parameters in SAL and DSP rats.

Indices	Group		SAL vs. DSP
	SAL	DSP	t-value #	p-value
Heart rate (bpm)	409±15.7	456±16.7	2.052	0.074
Pressure (mmHg):
Systolic	137±5.1	122±6.5	1.809	0.108
Diastolic	100±2.5	79±3.4	4.939	0.001
Mean arterial (MAP)	112±3.5	94±3.6	3.488	0.008
Pulse	37±2.7	43±5.7	0.957	0.367
Volume:
Stroke volume (mL)	0.16±0.01	0.21±0.02	2.410	0.043
Cardiac output (mL/min)	64±3.3	97±11.9	2.699	0.027
cardiac indexa	20±1.1	36±4.0	3.758	0.006
SVR (unit)	1.76±0.06	1.03±0.14	4.742	0.002
SVR indexb	0.56±0.02	0.39±0.06	2.837	0.022

Values: mean ± SE, n = 5 rats/ group. Stroke volume = pulse pressure divided by (Systolic + Diastolic); cardiac output = stroke volume x Heart rate; SVR (systemic vascular resistance) = MAP divided by Cardiac output.

^acardiac index = Cardiac output/ 100 g body weight.

^bSVR index = SVR/ 100 g body weight. unit = mmHg/mL/min;

^#= unpaired t-test

Blood samples

After collection, the blood samples were transferred into 1.5 ml microcentrifuge tubes containing 10 μL EDTA (100 mg/mL EDTA; tetra sodium salt; Sigma) on ice. At the end of the experiment, the blood samples were centrifuged at 12,000 g for 7 min at 4°C and the plasma samples were stored at −20°C until assay [28] for ACTH (Allegro HS-ACTH, Nichols Institute, San Juan Capistrano, CA) and CORT (ImmuChemE Double Antibody, ICN Bio- medicals, Costa Mesa, CA). The unit for ACTH and CORT values are pg/mL and ng/mL respectively.

Tissue analysis

Norepinephrine content was analyzed in cortex, hippocampus, hypothalamus and olfactory bulb. On the day of assay, a 5 mL polystyrene sonication tube containing 1.0 mL of 1.0 M perchloric acid was weighed (initial) and frozen tissue sample was added directly into the perchloric acid in the tube and reweighed (final). Thereafter the tube was placed in ice and homogenized by sonication for 25–30 sec. The difference in the final and initial weights was recorded in milligram as the tissue weight. The homogenate was centrifuged at 12,000 g for 10 min at 4°C. After centrifugation, the supernatant was filtered through 0.22 micron syringe filter and norepinephrine in the filtrate was determined by HPLC-EC [50]. The norepinephrine values are ng/g tissue.

CRF content was analyzed in the median eminence and olfactory bulbs as described previously [51] and expressed as pg/mg protein.

Data Analysis

Plasma ACTH and CORT, and hemodynamic responses between SAL and DSP rats were evaluated by repeated measures analysis of variance (ANOVA) across time [52]. A value of 5 (reliable lower assay limit) was added to all ACTH and CORT numbers to correct for values below the assay detection limit. To evaluate reproducibility of stress response across studies, a priori, ANOVA was performed for each group independently along with post-hoc comparisons. Within a treatment group, when the overall response (main effect) was significant, post-hoc multi-comparisons were performed using Newman-Keuls’ procedure [52]. Between treatment groups, when the overall treatment effect and/or interaction was significant, the significance of differences between specific time points were tested by computing “F-values” for the simple main effects. The sum of changes over the basal values were treated as net response. Other statistical analyses, paired and unpaired (independent) samples t-tests, and bivariate (Pearson) correlation were performed using SPSS version 26.0 (IBM Corp., 2016) or StatView™II (Abacus Concepts, Inc.) statistical packages available for Macintosh computers (Apple Inc.). Unless otherwise indicated values are mean ± SE and values of p ≤ 0.05, 2-tailed, were treated as significant. Units of values are provided in the table and figures.

Results

Effects of LC-NA denervation on metabolic and neurochemical indices

After DSP4 administration, the rats lost body weight (Fig. 1) that was evident by day 2, whereas all SAL rats exhibited sustained growth (Fig.1). However, by day 6 the rats showed signs of recovery. Although body weight loss and recovery varied between rats, by day 10, DSP rats established a growth rate that was similar to that of SAL rats. Thus, body weight between day 0 and day 10 in SAL (r₂₃ = 0.799, p < 0.001) and DSP (r₂₃ = 0.708, p < 0.001) rats were positively correlated (Pearson). However, between SAL (309±2.7) and DSP (287±5.1) rats, body weight on day 10 were significantly different (Fig. 1, Table 1).

Fig. 1:

Body weight of rats treated on day 0 with either saline (SAL: control group) or the LC-NA denervating neurotoxin DSP4 (DSP: experimental group). Analysis of body weight between SAL and DSP rats by repeated measures ANOVA revealed a treatment effect (F_1,48 = 1.917, p = 0.173) that was not significant, however the interaction (F_5,240 = 44.09, p < 0.001) was significant. Values, mean ± SE, n = 25 rats/ group. ^§ and ^§§ = p<0.05 and p<0.01 respectively, SAL vs. DSP rats.

Table 1.

Basal metabolic and neurochemical indices in SAL and DSP rats.

Indices	n	Group		SAL vs. DSP
		SAL	DSP	t-value #	p-value
Body weight (g):	25
Day 0		257±3.8	265±3.0	1.604	0.115
Day 10		309±2.7	287±5.1	3.833	0.0004
Tissue norepinephrine (ng/g):	3
Cortex		209±10.3	13.3±1.5	18.899	0.0001
Olfactory bulb		217±14	32.3±5.3	12.220	0.0003
Hippocampus		281±36	5.7±0.7	7.744	0.0015
Hypothalamus		3125±201	2030±138	4.486	0.0109
HPA axis:
Plasma:	19
ACTH (pg/ml)		30.3±3.3	27.7±2.9	0.569	0.573
CORT (ng/ ml)		34.4±5.3	25.0±2.1	1.642	0.109
CRF (pg/mg protein):	3
Median eminence		4282±655	3884±843	0.373	0.7281
Olfactory bulb		465±37	441±34	0.484	0.6537

Values, mean ± SE, based on n rats/ group. SAL and DSP, rats treated with saline and DSP4 respectively. Plasma ACTH and CORT values are average of 0 min values from hemorrhage and startle experiments, and nadir values from diurnal experiments.

^#= independent sample/unpaired t-test.

In DSP rats compared to SAL rats, norepinephrine content of cortex was 6%, hippocampus was 2% and that of olfactory bulb was 15%, and the differences were significant (Table 1). However, in DSP rats, norepinephrine content of hypothalamus, a brain region that receives only moderate LC-NA efferents [2] was 65% of that in SAL rats. Nevertheless, the differences were significant (t₄ = 4.486, p < 0.011). Comparing SAL and DSP rats, CRF content in neither the median eminence nor the olfactory bulb were significantly different (Table 1). Furthermore, neither basal plasma ACTH nor CORT concentrations were different (Table 1).

Effect of LC-NA denervation on diurnal pattern of HPA axis response

Overall, for plasma ACTH (Fig. 2a), between SAL and DSP rats, by repeated measures ANOVA, treatment effect (F_1,16 = 2.231, p = 0.155) and interaction (F_1,16 = 0.122, p = 0.731) were not significant. Thus, the differences between SAL and DSP rats at nadir (t₈ = 0.973, p = 0.359) or at acrophase (t₈ = 1.139, p = 0.288) were not significant. As to be expected, for plasma ACTH, the differences between nadir (30.5±3.8) and acrophase (59±6.7) in DSP rats (t₄ = 7.838, p < 0.002) and between nadir (38.3±7.1) and acrophase (72±8.8) in SAL rats (t₄ = 15.06, p < 0.001) were significant.

Fig. 2:

Diurnal plasma ACTH (1a) and plasma CORT (1b) responses at nadir and acrophase of HPA axis activity after denervation of LC-NA efferents in DSP rats compared to intact SAL rats. The rats were maintained under 12:12 h light–dark cycle; lights ON at 0700 h. Blood samples were collected 60min after lights “ON,” (nadir) and again 30min before lights “OFF,” (acrophase). Values, mean ± SE, n = 5 rats/ group. ** = p<0.01 AM vs. PM, ^§§ = p<0.01 SAL vs. DSP rats.

Comparing plasma CORT (Fig. 2b), between SAL and DSP rats, by repeated measures ANOVA resulted in a treatment effect (F_1,16 = 11.54, p < 0.004) and interaction (F_1,16 = 15.21, p < 0.002) that were significant. Thus, the differences between SAL and DSP rats in the plasma CORT at nadir were not significant (t₈ = 0.978, p = 0.357), but they were at acrophase (t₈ = 3.775, p < 0.006). This suggested the diurnal rise in plasma CORT was significantly blunted in the DSP4 treated rats. In addition, for plasma CORT (Fig. 2b), the differences between nadir (24.2±3.3) and acrophase (56.9±11.2) in the DSP rats only approached significance (t₄ = 2.657, p = 0.057), whereas that between nadir (20.9±0.9) and acrophase (105±5.9) in the SAL rats (t₄ = 13.78, p < 0.001) were significant.

Effect of LC-NA denervation on plasma ACTH and CORT after hemorrhage

Plasma ACTH and CORT increased after hemorrhage and remained elevated even at 30 min when the experiment was terminated (Fig. 3). For plasma ACTH (Fig. 3a), the increases in SAL (F_4,24 = 16.06, p < 0.001) and DSP (F_4,24 = 9.332, p < 0.001) rats were significant (independent group ANOVA). Between groups, the treatment effect (F_1,12 = 4.849, p = 0.048) was significant with DSP rats exhibiting an attenuated response. Plasma CORT (Fig. 3b) increases in response to hemorrhage were significant in both SAL (F_4,24 = 52.35, p < 0.001) and DSP (F_4,24 = 34.66, p < 0.001) rats (independent group ANOVA). Between groups, the interaction was significant with an exaggerated CORT response observed in DSP rats at 30 min.

Fig. 3:

Hemorrhage (HEM) induced plasma ACTH (2a) and plasma CORT (2b) responses after denervation of LC-NA efferents in DSP rats compared to intact SAL rats. Soon after min 0 blood sample, 20% calculated blood volume was withdrawn via indwelling jugular vein catheter in three-minute duration. Values, mean ± SE, n = 7 rats/group. * and ** = p<0.05 and p<0.01 respectively, vs. min 0 value (independent group ANOVA). ^§ and ^§§ = p<0.05 and p<0.01 respectively, SAL vs. DSP rats.

Effect of LC-NA denervation on plasma ACTH and CORT after airpuff-startle

Plasma ACTH and CORT increased after airpuff-startle and returned towards 0 min values by the end of 30 min (Fig. 4). Plasma ACTH (Fig. 4a), responses were significant in both SAL (F_4,24 = 4.361, p < 0.009) and DSP (F_4,24 = 4.322, p = 0.009) rats (independent group ANOVA). Between SAL and DSP rats the interaction (F_4,48 = 2.639, p = 0.045) was significant, and as with the hemorrhage stressor, there was an attenuation of the plasma ACTH response in DSP rats compared to SAL rats. Plasma CORT (Fig. 4b) increases in response to airpuff-startle were significant in both SAL (F_4,24 = 10.73, p < 0.001) and DSP (F_4,24 = 28.80, p < 0.001) rats (independent group ANOVA). The treatment effect between SAL and DSP groups was significant with an attenuated CORT response observed in DSP rats.

Fig. 4:

Airpuff-startle (APS) induced plasma ACTH (3a) and plasma CORT (3b) responses after denervation of LC-NA efferents in DSP rats compared to intact SAL rats. Soon after min 0 blood sample, the rats were subjected to a short episode airpuff-startle of 5s duration. Values, mean ± SE, n = 7 rats/ group. * and ** = p<0.05 and p<0.01 respectively, vs. min 0 value (independent group ANOVA). ^§ and ^§§ = p<0.05 and p<0.01 respectively, SAL vs. DSP.

Effect of LC-NA denervation on hemodynamic plasticity

This was evaluated by monitoring hemodynamic indices after 20% hemorrhage in SAL and DSP rats (Fig. 5). Because cardiovascular homeostasis is regulated by long-(tonic) and short-term (reflex) mechanisms [53–54], the 0 min hemodynamic values recorded before the rats were subjected to hemorrhage were treated as the basal indices and were analyzed independently (Table 2).

Fig. 5:

Hemorrhage (HEM) induced hemodynamic responses after denervation of LC-NA efferents in DSP rats compared to intact SAL rats. Soon after min 0 blood sample, 20% calculated blood volume was withdrawn via indwelling carotid artery catheter in three-minute duration. Values, mean ± SE, n = 5 rats/ group. * and ** = p<0.05 and p<0.01 respectively, vs. min 0 (independent group ANOVA).

Hemodynamics basal:

Analysis of basal hemodynamic indices in SAL and DSP rats are presented in Table 2. In the DSP rats compared to SAL rats, heart rate exhibited a nonsignificant elevation (p = 0.074). Differences in systolic and pulse pressure were not significant, however differences in all other indices were significant (Table 2). In DSP rats the decreases in diastolic and MAP were accompanied by an increase in cardiac output and a decrease in SVR. This effect was not simply due to differences in body weight because between group differences in the cardiac index and SVR index were also significant (Table 2).

Hemodynamics after hemorrhage:

The effects of hemorrhage on the response of heart rate and blood pressure indices are shown in Figure 5. Immediately after hemorrhage (3 min), in both SAL and DSP rats, all hemodynamic indices decreased (Fig. 5). Although the basal values of blood pressure indices were low, in DSP rats the values at 3 min did not decrease below that of SAL rats. Accordingly, the decreases were smaller in DSP rats and larger in SAL rats. For heart rate, the basal values in DSP rats compared to SAL rats were moderately elevated (Fig. 5). However, the response after hemorrhage paralleled that of SAL rats and the net changes between groups were similar.

In SAL rats, the overall responses of heart rate (F_5,20 = 3.111, p = 0.031), systolic (F_5,20 = 9.834, p<0.001), diastolic (F_5,20 = 5.262, p = 0.003) and MAP (F_5,20 = 6.444, p = 0.001) were significantly different following hemorrhage (independent group ANOVA). In contrast, heart rate (F_5,20 = 0.601, p = 0.700), systolic (F_5,20 = 1.608, p = 0.204), diastolic (F_5,20 = 1.690, p = 0.183) and MAP (F_5,20 = 0.877, p = 0.514) responses to hemorrhage were not significantly different in DSP rats (Fig. 5) (independent group ANOVA). Comparing SAL and DSP rat responses, the interaction for systolic (F_5,40 = 2.952, p = 0.023), diastolic (F_5,40 = 3.103, p < 0.019) and MAP (F_5,40 = 2.646, p = 0.037) were significantly different (Fig. 5). Pearson correlation (r) between heart rate and MAP was significant in SAL group (r₂₈ = 0. 0.660, p < 0.001) but not in DSP group (r₂₈ = 0. 046, p = 0.811). Accordingly, after hemorrhage the hemodynamic responses in SAL rats were synchronous, while those in DSP rats were asynchronous.

Discussion

DSP4 rat model

The results on body weight and tissue norepinephrine in DSP rats compared to SAL rats are consistent with those reported by other investigators [38–39]. Furthermore, in DSP rats, norepinephrine content of brain regions innervated densely by LC-NA efferents [1, 3] decreased to marginal levels (Table 1). Accordingly, DSP4 treatment significantly and reliably denervates LC-NA efferents. Our data demonstrate that DSP4 treatment does not meaningfully alter CRF in the median eminence [51].

LC-NA modulation of diurnal HPA axis response

Our study is the first to identify potential role of LC-NA neurons as a major neuroanatomical orchestrator of regulation of plasma CORT responses at acrophase. In DSP rats at acrophase, the plasma ACTH response was normal, whereas plasma CORT response was suppressed (Fig. 2). Accordingly, a potential role of LC-NA neurons in the control of splanchnic sympathetic nerve activity and adrenal cortical secretory function at acrophase is suggested. This is in sharp contrast to the response of plasma ACTH between SAL and DSP rats which did not differ at nadir or at acrophase.

Electrical and chemical stimulation of LC activates adrenal medullary secretory and SNS responses independent of forebrain projections [55–56]. LC-NA neuronal projections to the adrenals are second order [20] and the LC may transduce signals to the adrenals via cholinergic preganglionic sympathetic fibers arising from the spinal cord that are critical for adrenal cortical secretory responses at acrophase [16–18]. DSP4 treatment denervates LC-NA projections to the spinal cord [41] and suggests that the lack of central drive to the adrenals suppresses the increases in plasma CORT normally observed at acrophase.

Our results also suggest that the neurocircuitry modulating increases in plasma ACTH and CORT at acrophase is different from that after stress. In studies discussed above, plasma ACTH at acrophase was not different between SAL and DSP rats, but following either hemorrhage or airpuff-startle stressor (Fig. 3, 4) plasma ACTH response was decreased in DSP rats. Based on central components of arousal system [1, 23] and our data, a circuitry regulating of diurnal adrenocortical function is proposed (Fig. 6).

Fig. 6:

Schematic representation of potential neurocircuitry in the central control of HPA axis activity at diurnal acrophase. Broken arrow indicates the specific effect of the projection needs reevaluation or not clearly established; see text for details. ACTH: adrenocorticotropic hormone, Adr: adrenal gland, CORT: corticosterone, DMH: dorsomedial hypothalamus, LC: locus coeruleus, ME: median eminence, Pit: pituitary, PVN: paraventricular nucleus of hypothalamus, SC: spinal cord, SCN: Suprachiasmatic nucleus, ZT: zeitgebers.

The SCN-DMH-LC constitutes the primary arousal circuitry [1, 23]. The extant literature indicates that DMH lesions, but not PVN lesions, suppress plasma CORT responses [21–22, 24–25]. After PVN lesion, CRF content in the median eminence is decreased, whereas pituitary ACTH content is increased [57]. Chemical stimulation of the DMH increases plasma ACTH in conscious rats, whereas similar stimulation of PVN does not [58]. Electrophysiological studies have identified DMH neurons projecting either to LC or median eminence. A third group of neurons that send projections to both LC and median eminence have also been described [59]. Interestingly, at acrophase plasma ACTH response is uncoupled from CRF heteronuclear RNA expression in the PVN neurons [60]. In addition, the sparse LC-NA efferents to PVN [2] do not influence plasma CORT response at acrophase [61]. These observations along with the results of present study suggest a potential role for DMH in the regulation of plasma CORT at acrophase via LC-NA system, but, perhaps surprisingly, the role of the DMH in plasma ACTH responses at acrophase remains to be clarified.

LC-NA modulation of HPA axis response after stress

As indicated, after hemorrhage, medullary A2-noradrenergic neurons relay hemodynamic signals to the PVN in the activation of HPA axis [2, 28, 33, 62–63]. In the present study, after hemorrhage the DSP rats responded with increases in plasma ACTH and CORT, suggesting that DSP4 treatment spared the A2-noradrenergic neurons. However, in DSP rats, the ACTH response was attenuated and the net response in DSP rats was 73% of that in SAL rats. The results are comparable to the plasma ACTH response following restraint stress in rats with selective denervation of LC-NA efferents to the medial prefrontal cortex [27]. We observed that attenuation of the ACTH response was more pronounced following airpuff-startle and the net response in DSP rats was 48% of that in SAL rats. After airpuff-startle, corticolimbic circuitry participates in the modulation of plasma ACTH response [34]. Accordingly, the LC-NA system modulates plasma ACTH regardless of stressor category (physical/hemodynamic or psychological).

Radley et al. (2008), have suggested that an increase in LC-NA activity restrains gamma-aminobutyric acid (GABA) inhibitory tone on PVN-CRF neurons via a corticolimbic circuit [27]. However, direct application of norepinephrine locally into the PVN elicits increases in plasma ACTH [64]. It is likely that after stress the LC-NA-corticolimbic circuitry synergistically modulates CRF neuronal activity in the PVN. Moreover, it is well-established that the LC-NA system enhances the signal-to-noise ratio of incoming stimuli [1, 3].

In both SAL and DSP rats, plasma CORT increased after hemorrhage and airpuff-startle (Fig. 3b, 4b). This was in sharp contrast to the response of the DSP group at acrophase when plasma CORT responses were blunted (Fig. 2b). However, the net percentage increase in plasma CORT in DSP rats compared to SAL rats after hemorrhage or airpuff-startle was, respectively, 158% versus 62%. Accordingly, in DSP rats the pattern of response was dependent upon the stressor category. However, duration of the stressor and magnitude of increases in plasma ACTH also need to be taken into consideration.

Stressors increases the activity of baroreceptors that communicate with the medullary cardiovascular centers for appropriate vasomotor function, and to maintain cardiovascular homeostasis and sympathoadrenal activity [53–54, 65–67]. Increases in sympathoadrenal activity facilitate the release of vasodilatory factors that increase adrenal blood flow, ACTH presentation rate and plasma CORT [15, 68–69]. Both hemorrhage [70] and plasma ACTH [15] have been shown to increase adrenal blood flow. After hemorrhage, the increase in plasma ACTH was larger than that after airpuff-startle exposure (Fig. 3, 4). In addition, SVR was reduced in DSP rats (Table 2). Thus, in the DSP group, these factors might interact and contribute towards the large increase in plasma CORT after hemorrhage. At this time, a specific role for individual factors still needs to be systematically investigated. Regardless of the differences in the pattern of plasma CORT responses, our results suggest that adrenocortical activity after stress is modulated by neurocircuitry that is different from that at acrophase (Fig. 6). The rostral ventrolateral medulla, including the C1 adrenergic cell group is a major component of the baroreflex and after stress may serve as a potential neuroanatomical substrate in the modulation of sympathoadrenal activity [54, 65–67].

LC-NA modulation of hemodynamics

To our knowledge, this is the first study to calculate stroke volume for rats using the Liljestrand and Zander equation primarily derived for human subjects [49]. In human studies, the calculated values are multiplied by a constant [49]. In the present study, the calculated cardiac output values (stroke volume x heart rate) for SAL rats were comparable to that reported by other investigators using thermodilution technique [71], therefore a correction factor was not applied. However, when the values for cardiac output reported in the literature are different [70], the use of a correction factor of two, corrected the discrepancy. The application of this equation greatly helped in the calculation of the volume components of hemodynamic indices in freely moving rats.

Neuronal and humoral factors regulate hemodynamics and maintain cardiovascular homeostasis [53–54, 65–67]. Because, the major focus of this experiment was deciphering the role of LC-NA system in this process, the effect of denervation of LC-NA efferents on hemodynamic plasticity during long-term (basal) and short-term (hemorrhage), cardiovascular homeostasis is of interest. }. In the present study, the increase in plasma CORT after hemorrhage did not normalize cardiovascular indices comparable to that of SAL rats (Fig. 5). Accordingly, the effect of CORT in maintaining normal cardiovascular tone is unlikely to be mediated by non-genomic mechanisms [12–13].

Hemodynamic plasticity, basal (long-term):

Based on blood pressure and flow dynamics (Table 2) our data suggest that the DSP group exhibited systemic vasodilation. Thus, the increased cardiac output that accompanied low MAP (Table 2) was considered important to maintain tissue perfusion. However, this increased venous return activates the sinoatrial node (stretch receptors). This is known to inhibit vagal outflow and enhance sympathetic action on the sinoatrial node to increase heart rate [72–73] (Table 2). The scenario is similar to the Bainbridge reflex [72–73], therefore it is possible that in the DSP group this reflex function maintained cardiovascular homeostasis due to loss of modulatory LC-NA efferents.

The effect of DSP4 treatment (IP) on peripheral SNS activity is transient and by two weeks normal or increased catecholamine turnover is observed [74–75]. Accordingly, the vasodilatory condition may be due to a loss of vascular alpha1 adrenergic receptor function [76]. Because a normal glucocorticoid status is important in maintaining vascular smooth muscle tone [77], a role for CORT itself in the vasodilatory effect in the DSP group must be considered. Similar to our current observations, it has been reported that rats exhibit low MAP and increased heart rate after adrenalectomy [78–80].

Glucocorticoid type I and type II receptors and the two isoforms of the glucocorticoid regulatory enzyme, 11 beta-hydroxysteroid dehydrogenases (11beta-HSD), are present in endothelial and vascular smooth muscle cells [77, 81]. Activation of GR, upregulates alpha1 adrenergic receptor function and facilitates receptor-G protein coupling [78–79], whereas the glucocorticoid type II receptor antagonist, RU486, decreased blood pressure [82]. Additionally, local increases in glucocorticoid after inhibition of 11beta-HSD activity facilitate the vasoconstrictor actions of norepinephrine and angiotensin II [77, 81]. During diurnal rhythm, the concentration of plasma CORT required to bind and activate glucocorticoid type II receptors and regulation of gene transcription occurs only at acrophase. In the present study, such an increase in plasma CORT was suppressed in the DSP rats (Fig. 2). With special reference to cardiovascular homeostasis, perhaps the lack of a significant increase in plasma CORT affected glucocorticoid type II receptor dependent gene transcription, especially that of alpha1 adrenergic receptor number, with a resultant decrease in blood pressure [78, 80, 82]. Moreover, the reduced CORT concentration might allow the endothelial cells to increase production of the vasorelaxants prostacyclin and nitric oxide [77, 81]. Therefore, the LC may modulate long-term cardiovascular homeostasis by regulating diurnal acrophase CORT status.

Hemodynamic plasticity, after hemorrhage (short-term):

Immediately after hemorrhage the pattern of heart rate and blood pressure response in SAL and DSP groups were similar (Fig. 5, Table 2). It is known that hemorrhage activates baroreceptor and chemoreceptor reflexes and increases vasoconstrictor factors (norepinephrine, epinephrine, vasopressin and angiotensin) to maintain blood pressure in awake and anesthetized animal models [28–29, 36]. The increase in plasma vasopressin is most important for the vasoconstrictor function [82–83]. In patients with autonomic failure, an increase in plasma vasopressin increases blood pressure response, whereas a similar increase in normal subjects is without any effect [84–85]. It is possible that a similar mechanism maintained blood pressure in the DSP rats. Therefore, our data suggest that the underlying vasodilation affected the interaction between the multitude of factors controlling hemodynamics thereby affecting the stability of the hemodynamic measures in DSP rats. Regardless of the mechanisms, it seems unlikely that the LC-NA system modulates short-term cardiovascular homeostasis.

Conclusion

Regarding HPA axis reactivity, our results from the DSP group suggests that the LC-NA neuronal system is an integral part of neuroanatomical substrate modulating activity of PVN CRF neurons and sympathoadrenal activity via independent, non-overlapping neurocircuits. After stress, the LC-NA system modulated only the plasma ACTH response, whereas it did not modulate plasma CORT response. In addition, the modulatory effect of the LC-NA system on PVN CRF neurons appeared to be synergistic. In sharp contrast to that after stress, during diurnal acrophase the plasma ACTH response was maintained, whereas the plasma CORT response was blunted. Thus, at acrophase, the LC-NA system modulated only plasma CORT response and not plasma ACTH response. Thus, LC-NA neuronal modulation of HPA axis reactivity during the normal diurnal activity period and after stress appears to be dissociated. A potential neurocircuitry in the LC-NA activation sympathoadrenal function during acrophase is highlighted.

Regarding cardiovascular homeostasis, our results from DSP rats suggested that the LC-NA effect on hemodynamics is indirect. This is based upon our observation that in DSP rats the blunted plasma CORT response at acrophase was accompanied by a reduced basal SVR. Because a normal CORT status is important to maintain vascular sympathetic tone, the blunted plasma CORT response altered hemodynamics and blood flow dynamics that affected long-term and short-term maintenance of cardiovascular homeostasis.