B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Nitric oxide inhibition in paraventricular nucleus oncardiovascular and autonomic modulation after exercisetraining in unanesthetized rats

Rosiane Batista Mastelaria, Hugo Celso Dutra de Souzab, Adriane Lenharda,Fernando Morgan de Aguiar Corrêac, Marli Cardoso Martins-Pingea,⁎aDepartment of Physiological Sciences, Center of Biological Sciences, State University of Londrina, Londrina, Paraná, BrazilbDepartment of Biomechanics and Rehabilitation, School of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, São Paulo, BrazilcDepartment of Pharmacology, School of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, São Paulo, Brazil

A R T I C L E I N F O

⁎ Corresponding author. Departamento de CCEP: 86051-990, Paraná, Brazil. Fax: +55 11 33

E-mail address: martinspinge@uel.br (M.C.

0006-8993/$ – see front matter © 2010 Elsevidoi:10.1016/j.brainres.2010.12.049

A B S T R A C T

Article history:Accepted 13 December 2010Available online 21 December 2010

It is well known that regular physical exercise alter cardiac function and autonomicmodulation of heart rate variability (HRV). The paraventricular nucleus of hypothalamus(PVN) is an important site of integration for autonomic and cardiovascular responses, wherenitric oxide (NO) plays an important role. The aim of our study was to evaluate thecardiovascular parameters and autonomic modulation by means of spectral analysis afternitric oxide synthase (NOS) inhibition in the PVN in conscious sedentary (S) or swimmingtrained (ST) rats. After swimming training protocol, adult male Wistar rats, instrumentedwith guide cannulas to PVN and femoral artery and vein catheters were submitted to meanarterial pressure (MAP) and heart rate (HR) recording. At baseline, the physical traininginduced a resting bradycardia (S: 374±5, ST: 346±1 bpm) and promoted adaptations in HRVcharacterized by an increase in high-frequency oscillations (HF; 26.43±6.91 to 88.96±2.44)and a decrease in low-frequency oscillations (LF; 73.57±6.91 to 11.04±2.44) in normalizedunits. The microinjection of Nω-nitro-L-arginine methyl ester (L-NAME) in the PVN ofsedentary and trained rats promoted increase in MAP and HR. L-NAME in the PVN did notsignificantly alter the spectral parameters of HRV of sedentary animals, however in thetrained rats increased LF oscillations (11.04±2.44 to 27.62±6.97) and decreased HFoscillations (88.96±2.44 to 72.38±6.97) in normalized units compared with baseline. Ourresults suggest that NO in the PVN may collaborate to cardiac autonomic modulation afterexercise training.

© 2010 Elsevier B.V. All rights reserved.

Keywords:Arterial pressureHeart rate variabilityBrainSympatheticL-NAMESwimming

1. Introduction

Physical exercise is characterized by a situation that changesbody homeostasis, since it implies in instantaneous increase

iências Fisiológicas, Cen71 4467.Martins-Pinge).

er B.V. All rights reserved

in energy demand of the muscles exercised, and therefore thebody as a whole. Thus, to meet the new metabolic demand,several physiological adaptations are necessary and, amongthem those related to cardiovascular function. These

tro de Ciências Biológicas, Universidade Estadual de Londrina,

.

69B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

adjustments include the cardiovascular remodeling of theheart and skeletal muscle circulation, improvement of auto-nomic control of heart, and the appearance of bradycardia atrest (Souza et al., 2007a, 2007b; Medeiros et al., 2004; Negrão etal., 1993).

It has been proposed that areas of the central nervoussystem, known for its involvement in cardiovascular control,appear to show altered activity in response to an exercisesession in trained rats (Ichiyama et al., 2002). Among the areasinvolved in the exercise most studied in cardiovascularadaptations are the rostral ventrolateral medulla (RVLM)(Martins-Pinge et al., 2005a; Becker et al., 2005; Mueller,2007), the nucleus of solitary tract (NTS) (De Souza et al.,2001; Mueller and Hasser, 2006; Michelini and Stern, 2009) andparaventricular nucleus (PVN) (Jackson et al., 2005; de Abreuet al., 2009; Michelini and Stern, 2009).

The paraventricular nucleus (PVN) of the hypothalamus isknown to be a place of integration for autonomic andneuroendocrine responses (Coote, 1995; Herman et al., 2000;Michelini and Stern, 2009). Morphological and electrophysio-logical studies showed that the PVN is reciprocally connectedto other areas of the brain that are involved in cardiovascularfunction (Kleiber et al., 2008). These regions include the NTS,RVLM and through lateral spinal thoracolumbar cell column,location of sympathetic preganglionic neurons (Dampney,1994; Kleiber et al., 2008). Through these reciprocal intercon-nections with autonomic centers in the brain stem and spinalcord, as well as projections to the median eminence andposterior pituitary (neurohypophysis), the PVN is capable ofgenerating complex patterns of neuroendocrine and auto-nomic, thus controlling homeostasis.

There is increasing evidence that nitric oxide (NO) acts as agaseous neurotransmitter to modulate synaptic function inthe central nervous system. NO synthase (NOS) positiveneurons have been identified in several regions of the brainand in the paraventricular nucleus (PVN) of the hypothalamusthe presence of NO synthase suggest that NO may play a rolein endocrine and autonomic regulation of cardiovascularresponses (Stern et al., 2003). In vitro studies had providedevidences for a modulation of neuronal activity by NO (Hornet al., 1994). Also, in vivo studies demonstrated that nitricoxide originating fromNMDA activation seems to increase therelease of GABA from presynaptic terminals, suggesting aninhibitory role for NO in the PVN (Bains and Ferguson, 1997;Zhang et al., 1997; Zhang and Patel, 1998). Studies in intact andanesthetized rats have shown that administration of an NOdonor into the PVN produces decreases in renal sympatheticnerve discharge, arterial pressure (AP) and heart rate (HR)(Zhang et al., 1997; Zhang and Patel, 1998). Conversely,microinjection of NOS blockers into the PVN producesexcitatory responses (Zhang and Patel, 1998; Martins-Pingeet al., 2004, 2005b). However, recent data has pointed to anexcitatory role for NO in the PVN of conscious rats (Busnardoet al., 2010). Anyway in all these data NO has a tonic effect inPVN cardiovascular responses.

Regular physical exercise, which is known to promote afavorable cardiovascular state, improved endothelial functionvia several mechanisms. Indeed, it augments blood flow andlaminar shear stress, resulting in increased NO productionand bioavailability (Di Francescomarino et al., 2009). Regular

exercise is also able to modify the cardiac autonomic balanceby increasing the parasympathetic and decreasing sympa-thetic activity (de Abreu et al., 2009). The exercise trainingpromotes an increase in the capacity of the autonomicnervous system to meet the demands of the cardiovascularsystem, increasing the heart rate variability (HRV) (Souza et al.,2007a, 2007b; Tezini et al., 2008). A recent study from ourlaboratory (de Abreu et al., 2009) suggests a contribution ofneurons in the PVN to cardiac autonomic modulation afterphysical training. Also, no data in the literature has addressedthe study of NO effects in central areas involved in cardiovas-cular regulation and it modulatory effects after exercisetraining.

To investigate the autonomic modulation on HRV and thesystolic arterial pressure (SAP) in the frequency domainweusedthe spectral analysis bymeans autoregressivemethod. Spectralanalysis is a tool that decomposes cardiovascular signals intotheir frequency components and quantifies the power of eachcomponent (Malliani et al., 1991). Three main spectral compo-nents are distinguished for the assessment of a spectrum in ashort record period of 2–5 min (Sayers, 1973; Hirsh and Bishop,1981; Askelrod et al., 1981; Pagani et al., 1986; Malliani et al.,1991): very low frequency (VLF), low frequency (LF) and highfrequency (HF). The distribution of power and frequency of LFand HF are not fixed but can vary in relation to changes inautonomic modulations on these periods in the heart (Paganiet al., 1986; Malliani et al., 1991; Furlan et al., 1990). Thephysiological explanation of the VLF component is not welldefined, and the existenceof a physiological process assigned tochanges of this period are still questioned (Malik, 1996).

Based in those evidences, our present study investigatedthe effects of NO synthesis inhibition into the PVN oncardiovascular autonomic adaptations induced by previousphysical training in conscious rats. In order to do so, we haveused the evaluation of heart rate variability and systolicarterial pressure by means of spectral analysis.

2. Results

2.1. Histological analysis

The histological sections were examined microscopically andcompared to the rat brain atlas (Paxinos and Watson, 1998).Only the rats in which the sites of microinjection were locatedin the area of PVN (inside PVN) were used for data analysis(Fig. 1).

2.2. Characterization of baseline parameters of sedentaryand trained rats

The baseline values for mean arterial pressure (MAP) andheart rate (HR) for each group were considered as a meanvalue of the period of basal recording (±30 min) of eachanimal. Physical training did not alter baseline MAP (seden-tary, n=7, 110±2 mm Hg, and Trained, n=8, 109±1 mm Hg).However, the exercise training promoted a decrease inbaseline heart rate (HR): sedentary 374±5 bpm; trained 346±1 bpm (P<0.05). The resting bradycardia was observed in ourothers studies with the same exercise training protocol and

Fig. 1 – Representative photomicrograph (A) and schematic representation (B) of injection sites in serial sections from the rostral(1.4 mm) to the caudal (2.12 mm) extent of the region of the paraventricular nucleus (PVN). Distance posterior to bregma isshown for each section according to the atlas of Paxinos and Watson (1998): ♦, microinjections inside PVN; ○, microinjectionsoutside PVN. Abbreviations: 3V, thirt ventricle; AH, anterior hypothalamic nucleus.

70 B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

71B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

has been used as amarker of effectiveness of exercise trainingfor other groups too (Evangelista et al., 2005; Zanesco andAntunes, 2007; Sanches et al., 2009).

The modulation of autonomic nervous system by spectralanalysis showed that physical training promoted an increasein the variability of pulse interval, an increase of HF inabsolute and normalized values and a decrease of LF inabsolute and normalized units (Fig. 3). The analysis of systolicblood pressure showed no changes in baseline parametersafter physical training (Fig. 4).

2.3. Cardiovascular responses to L-NAME or salinemicroinjections into the PVN of sedentary or trained rats

The microinjection of L-NAME in the sedentary group (n=7)produced an increase in mean arterial pressure (19±2 mmHg)and heart rate (83±5 bpm). After physical training by swim-ming (n=8), the cardiovascular effects of L-NAME were notdifferent formean arterial pressure (19±2 mmHg) and also didnot change for heart rate (84±5 bpm) (Fig. 2).

2.4. Spectral analysis of pulse interval after L-NAMEmicroinjection into the PVN in sedentary and trained rats

The microinjection of L-NAME in the PVN did no significantlyalter the spectral parameters of the pulse interval of sedentaryanimals, however, in the trained animals, the microinjectionof L-NAME promoted an increase in LF component in absolute

Fig. 2 – Changes inmean arterial pressure (MAP) (A) and heartrate (HR) (B) post L-NAME microinjection in sedentary (blackbars) or exercise trained (gray bars) group into the PVN (left)or outside the PVN (right). The number of animals per groupis described in the results section. *P<0.05.

and normalized units and a decrease of HF in normalizedvalues (Fig. 3).

2.5. Spectral analysis of systolic arterial pressure afterL-NAME microinjection into the PVN in sedentary and trainedrats

After administration of L-NAME in the PVN, the spectralanalysis of systolic blood pressure showed an increase invariability only in the sedentary animals with no significantchange in other parameters (Fig. 4).

3. Discussion

It is well established in the literature that the holding ofregular exercise alter cardiovascular function and autonomicbalance. The purpose of our study was to test the hypothesisthat these changes would be associated with a differentialmodulatory effect of NO in PVN neurons between sedentaryand trained animals. Our data revealed that at baselinephysical training produces as well as bradycardia at rest, achange in autonomic modulation with increased heart ratevariability, decreased LF component and increased HF com-ponent. However, no changes were observed after physicaltraining on autonomic modulation of systolic blood pressurein conscious rats.

Changes in HRV suggest that exercise training causedadaptations in cardiac autonomic balance characterized by apredominance of parasympathetic autonomic componentover the sympathetic component. This statement is basedon studies that showed that the HF oscillations are due,mostly, by autonomic parasympathetic modulation, whereasthe LF oscillations were derived at least in part, by the twocomponents of autonomic, sympathetic and parasympathetic(Japundzic et al., 1990; Cerutti et al., 1991; Daffonchio et al.,1995). The genesis of these mechanisms of autonomiccardiovascular control is central, but there is a discussion onthe modulation of these mechanisms for interactions ofafferents from peripheral reflex arc, as the baroreflex (Mallianiet al., 2006; Eckberg, 2009).

In fact, our results regarding the effects of exercise trainingon autonomic modulation are consistent with the literature,and confirm that physical training promotes greater auto-nomic nervous system's ability to meet the demands of thecardiovascular system (Tezini et al., 2008; Souza et al., 2009).The cause of these changes is still uncertain, however a recentstudy from our laboratory (de Abreu et al., 2009), evaluated theautonomic modulation after microinjections of bicuculline ormuscimol in the PVN of sedentary and exercise trained rats.Our data showed that heart rate variability in sedentaryanimals was increased compared with trained ones. In thatwork, after the inhibition of the PVN with muscimol, thedifference in variability between sedentary and trained ratsdisappeared, suggesting a contribution of neurons in the PVNto autonomic modulation during physical training.

The PVN is a complex structure composed of functionallydifferent subsets of neurons, including magnocellular neuro-endocrine, parvocellular neuroendocrine and parvocellularpreautonomic neurons (Swanson et al., 1980; Swanson and

Pulse Interval

Var

ianc

e, m

s2

0

10

20

30

Sedentary Vehicle

Sedentary L-NAME

Trained Vehicle

Trained L-NAME

LF, m

s2

0

10

20

30

HF

, ms2

0

10

20

30

LF, n

u

0

40

80

120

HF

, nu

0

40

80

120

Control Trained

Control Trained Control Trained

Control Trained Control Trained

*

*

*

*

*

#

#

#

+

+

+

Fig. 3 – Spectral parameters of pulse interval (PI) calculated from time series using autoregressive spectral analysis, after PVNmicroinjections of L-NAME in control and trained rats. Values are MEAN ± SEM. * P<0.05 compared to Vehicle–Sedentary;+P<0.05 compared to L-NAME–Sedentary; #P<0.05 compared to Vehicle–Trained.

72 B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

Sawchenko, 1983). Through reciprocal interconnections withautonomic centers in the brainstem and spinal cord, the PVNoccupies a pivotal position within the central neuronalcircuitry involved in the integration of autonomic andendocrine responses (Swanson et al., 1980; Swanson andSawchenko, 1983).

NO-producing neurons are found in brain areas involved inautonomic activity including the PVN (Miyagawa et al., 1994),the NTS and the RVLM (Dun et al., 1994, 1995). Several reportsindicate that the PVN is an important site mediating NOactions, where either endogenously produced or exogenouslyapplied, NO elicits a reduction in arterial pressure and renalsympathetic nerve activity (Hashiguchi et al., 1997; Zhang

et al., 1997; Li et al., 2001). The literature also demonstratesthat NO tonically inhibits spontaneous firing activity ofmedulla oblongata PVN projecting neurons (Li et al., 2002).However, there is general agreement that the magnocellularneurosecretory neurons of the PVN constitute a major cellularsource of nitric oxide (Stern, 2004).

In our study, the administration of L-NAME in the PVN ofsedentary and trained animals promoted an increase in meanarterial pressure and heart rate. Those effects has beenobserved by others (Zhang and Patel, 1998; Zhang et al.,2001), even utilizing anesthetized animals. In our study nodifferences in these parameters between sedentary andtrained groups were observed. However, the autonomic

Systolic Arterial Pressure

Var

ianc

e, m

mH

g2

0

10

20

30

Sedentary Vehicle

Sedentary L-NAME

Trained Vehicle

Trained L-NAME

LF, m

mH

g2

0

6

12

18

HF

, mm

Hg2

0

6

12

18

Control Trained

Control Trained Control Trained

*

Fig. 4 – Spectral parameters of systolic arterial pressure (SAP) calculated from time series using autoregressive spectralanalysis, after PVN microinjections of L-NAME in control and trained rats. Values are MEAN ± SEM. *P<0.05 compared toVehicle–Sedentary.

73B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

modulation after administration of L-NAME in the PVN wasdifferent between sedentary and trained groups. In thesedentary group, the administration of L-NAME did notmodifythe spectral parameters of HRV compared to baseline.However, in the trained group the administration of L-NAMEinduced an increase in LF component in absolute andnormalized units and a decrease in HF in normalized units.As physical training changed the autonomic modulation tothe pulse interval, observing the baseline, and it is known thatphysical training also promotes changes in NO production(Jankord et al., 2009; Kuru et al., 2009), we believe that thechanges observed in the autonomic modulation by theadministration of L-NAME in the PVN of trained rats may bedue to a lack of NO on neurotransmission at the level of thePVN, where it modulates the autonomic function.

With regard to SAP variability, it was observed in ourstudy that blockade of NO synthesis into the PVN using L-NAME caused a marked increase in total variance (Fig. 4),which were attenuated in animals submitted to previousphysical training. The cause of this increase is uncertain andneeds further investigation. However, some studies haveshown that chronic inhibition of NO synthase would promoteincreases in total variance and LF oscillations in SAP (Nafz etal., 1997; Souza et al., 2001). The latter was not observed inour results. Those studies indicate that NO was accomplish

an important role on the modulation of the variability of SAPthrough peripheral mechanisms. In this case, NO has abuffering action on LF oscillations, opposing the vascularsympathetic modulation. This buffering action would betriggered particularly by the increased shear stress inendothelial cells, consequently promoting production andrelease of NO. On the other hand, a recent study showedthat after L-NAME treatment in conscious rats, the spectralanalysis of HRV and SAP did not show up different comparedwith control animals (Dos Santos et al., 2010). Actually, theauthors demonstrated that hypertension induced by L-NAMEtreatment did not correlate with sympathetic overactivity.Although we did not used L-NAME as a treatment, but asingle microinjection in the PVN, this study is in line with ourfindings as we did not observed differences in the autonomicmodulation after L-NAME in sedentary rats.

The literature found that NO-mediated inhibition of RSNAwithin the PVN was blunted in rats with heart failure (HF)(Zhang et al., 2001). It suggests that an altered NO mechanismmay contribute to the increased sympathetic nerve activity inthe HF state. Also, other study indicated that exercise trainingimproves the altered NO mechanism within the PVN andrestores NO-mediated changes in renal sympathetic nerveactivity in rats with HF (Zheng et al., 2005). However, no datafrom the literature studied the physiological effects of exercise

74 B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

training on NO neurotransmission in the PVN over autonomicand cardiovascular aspects.

In summary our study observed that nitric oxide blockadeinto the PVN showed tonic cardiovascular effects in sedentaryas trained rats with no differences between them. Wealso observed that after L-NAME microinjection in the PVN,in the sedentary group, the heart rate and arterial pressurevariability was not modified, suggesting that sedentarysubjects have not significant NO tonic influence over auto-nomicmodulation in the PVN. However in the trained animalsthe autonomic aspects were changed after L-NAME, suggest-ing a differential effect for NO in PVN neurotransmission onheart rate variability in exercise trained subjects. We believethat future studies are necessary to evaluate the neurochem-istry aspects of NO in the PVN under exercise trainingsituations.

4. Experimental procedures

4.1. Animals

The ethics committee for animal experiments of the StateUniversity of Londrina, Brazil, approved all experimentalprocedures involved in this study. Fifteen male Wistar rats(280–300 g) were divided into two groups, control and trainedrats, and kept in individual cages with controlled temperature(21 °C) under a dark/light cycle of 12 h. The animals wereallowed to obtain food and water ad libitum before theexperiment.

4.2. Swimming training protocol

The swimming training (ST) protocol was performed in a glasstank with vertical sides filled with lukewarm water (31±1 °C)with 4000 cm2 surface area and 60 cmdeep and occurred at thesame time every day (between 11:00 a.m. and 1:00 p.m.). TheST consisted of 20 swimming sessions (5 days/week, 4 weeks)of 60 min, according to Martins-Pinge et al., 2005a.

4.3. Surgical proceedings

One day after the ST been complete, the rats were anesthe-tized with sodium pentobarbital (40 mg/kg, IP) and placedprone in a stereotaxic apparatus (David Kopf Instruments)with the incisor bar 5 mm below the interaural line. Theanimals were implanted with two guide cannulas directedto the PVN (1.8 mm posterior to bregma, 0.5 mm lateralto midline and 7.6 mm below the surface). The guide cannulaimplantation technique was an adaptation based on aprevious work (Martins-Pinge et al., 1997). After this the ratswere allowed 3–5 days for recuperation from the surgery.

Twenty-four hours before the experiments, under sodiumpentobarbital (40 mg/kg, IP) anesthesia, a polyethylene cath-eter was inserted into the abdominal aorta through thefemoral artery. Another catheter was inserted into the femoralvein and both catheters were externalized dorsally to recordarterial pressure (AP) and heart rate (HR) and for drugadministration, respectively.

4.4. Measurement of cardiovascular parameters

On the day of the experiment, the animals were kept in theircages and basal recordings were obtained for at least 30minbefore start the protocol. The mean arterial blood pressure(MAP) and heart rate (HR) were recorded by a MLT0380 bloodpressure transducer connected to a Powerlab system 4/20T(ADInstruments).

4.5. Microinjections of drugs into the PVN

A standard 30–33 G stainless steel injector cannula wasconnected to a Hamilton syringe (7101) and positioned intothe guide cannula. Bilateral L-NAME (1 M/100 nl) or saline(0.9%) were microinjected (100 nl) into the PVN. This protocolwas used in control and trained rats. The choice of L-NAMEdose was based in a previous study (Zhang and Patel, 1998).

The values of mean arterial pressure (MAP) and HR inresponse to microinjections were considered at the peak ofresponses. Microinjections into the PVN were performed withthe rats showing no external signals of discomfort. Aftermicroinjectionof L-NAME into thePVN, nobehavioral responseswere observed.

Just one administration was performed in each PVN. At theend of the experiment, Evans blue dye (2%) was microinjected(100 nl) into each experimental site of the brain for histologicalverification.

4.6. Histology

The rats were sacrificed with an overdose of ether and thebrain was removed and fixed in 10% formaldehyde. Forhistological identification of the injection sites, the brainstem was cut coronally into 40-μm-thick sections and stainedwith 1% neutral red. The histological sections were examinedmicroscopically and compared to the rat brain atlas (Paxinosand Watson, 1998).

Just before infusion, all drugs were dissolved in physiolog-ical saline. L-NAME and sodium pentobarbital were obtainedfrom Sigma Chemical Co.

4.7. Heart rate and systolic arterial pressure variability

The baseline arterial pressure (AP) and HR recorded during a30-min period were processed by a customized computersoftware which applies an algorithm to detect cycle-to-cycleinflection points in the pulsatile AP signal, thus determiningbeat-by-beat values of systolic and diastolic pressures. Beat-by-beat pulse interval (PI) series from pulsatile AP signal werealso generated by measuring the length of time betweenadjacent systolic waves. From the baseline 40-min recordingperiod, the time series of PI and systolic AP (SAP) were dividedinto contiguous segments of 300 beats, overlapped by half.After calculating mean value and variance of each segment,they were submitted to amodel-based autoregressive spectralanalysis as described elsewhere (Malliani et al., 1991; Rubini etal., 1993; Task Force of the European Society Of Cardiology andthe North American Society of Pacing and Electrophysiology,1996). Briefly, a modeling of the oscillatory componentspresented in stationary segments of beat-by-beat time series

75B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

of SAP and PI was calculated based on Levinson–Durbinrecursion, with the model order chosen according to Akaike'scriterion (Malliani et al., 1991). This procedure allows anautomatic quantification of the center frequency and power ofeach relevant oscillatory component present in the time series.The oscillatory components were labeled as having very low(VLF: 0.01–0.20 Hz), low (LF: 0.20–0.75 Hz) or high frequency (HF:0.75–2.50 Hz). The power of LF and HF components of heart ratevariability was also expressed in normalized units, obtained bycalculating the percentage of the LF and HF variability withrespect to the total power after subtracting the power of the VLFcomponent (frequencies < 0.20 Hz). The normalization proce-dure tends to minimize the effect of total variance changes onthe absolute values of LF andHF variability (Malliani et al., 1991;Rubini et al., 1993; Task Force of the European Society ofCardiology and the North American Society of Pacing andElectrophysiology, 1996). The ratio between LF and HF wascalculated inorder to obtain the index of autonomicmodulationbalance for determining the cardiac variability.

4.8. Statistical analysis

All data are reported as mean ± SEM. Changes in maximalresponses induced by microinjection of drugs into the PVNwere analyzed by paired Student's t test. Differences betweengroups were assessed by two-way ANOVA followed by theTukey–Kramer test. The criterion for statistical significancewas P<0.05.

R E F E R E N C E S

Askelrod, S., Gordon, D., Ubel, F.A., Shannon, D.C., Barger, A.C.,Cohen, R.J., 1981. Power spectrum analysis of heart ratefluctuation: a quantitative probe of beat to beat cardiovascularcontrol. Science 213, 220–222.

Bains, J.S., Ferguson, A.V., 1997. Nitric oxide regulatesNMDA-driven GABAergic inputs to type I neurones of therat paraventricular nucleus. J. Physiol. 499, 733–746.

Becker, L.K., Santos, R.A., Campagnole-Santos, M.J., 2005.Cardiovascular effects of angiotensin II and angiotensin-(1-7)at the RVLM of trained normotensive rats. Brain Res. 1040 (1–2),121–128.

Busnardo, C., Crestani, C.C., Tavares, R.F., Resstel, L.B., Correa, F.M.,2010. Cardiovascular responses to L-glutamate microinjectioninto thehypothalamicparaventricularnucleusaremediatedbyalocal nitric oxide-guanylate cyclasemechanism. Brain Res. 1344,87–95.

Cerutti, C., Gustin, M.P., Paultre, C.Z., Lo, M., Julien, C., Vincent, M.,Sassard, J., 1991. Autonomic nervous system andcardiovascular variability in rats: a spectral analysis approach.Am. J. Physiol. Heart Circ. Physiol. 261, H1292–H1299.

Coote, J.H., 1995. Cardiovascular function of the paraventricularnucleus of the hypothalamus. Biol. Signals 4, 142–149.

Daffonchio, A., Franzelli, C., Radaelli, A., Castiglioni, P., Di Rienzo,M., Mancia, G., Ferrari, A.U., 1995. Sympathectomy andcardiovascular spectral components in consciousnormotensive rats. Hypertension 25 (6), 1287–1293.

Dampney, R.A., 1994. Functional organization of central pathwaysregulating the cardiovascular system. Physiol. Rev. 74,323–364.

DeAbreu, S.B., Lenhard,A.,Mehanna,A., deSouza,H.C., Correa, F.M.,Hasser, E.M., Martins-Pinge, M.C., 2009. Role of paraventricular

nucleus in exercise training-induced autonomic modulation inconscious rats. Auton. Neurosci. 148 (1–2), 28–35.

De Souza, C.G., Michelini, L.C., Fior-Chadi, D.R., 2001. Receptorchanges in the nucleus tractus solitarii of the rat after exercisetraining. Med. Sci. Sports Exerc. 33 (9), 1471–1476.

Di Francescomarino, S., Sciartilli, A., Di Valerio, V., Di Baldassarre,A., Gallina, S., 2009. The effect of physical exercise onendothelial function. Sports Med. 39 (10), 797–812.

Dos Santos, F.M., Martins Dias, D.P., da Silva, C.A., Fazan Jr., R.,Salgado, H.C., 2010. Sympathetic activity is not increased inL-NAME hypertensive rats. Am. J. Physiol. Regul. Integr. Comp.Physiol. 298 (1), R89–R95.

Dun, N.J., Dun, S.L., Forstermann, U., 1994. Nitric oxide synthaseimmunoreactivity in rat pontine medullary neuronsNeuroscience 59, 429–445.

Dun, N.J., Dun, S.L., Hwang, L.L., Forstermann, U., 1995. Infrequentco-existence of nitric oxide synthase and parvalbumin,calbindin and calretinin immunoreactivity in rat pontineneurons. Neurosci. Lett. 191, 165–168.

Eckberg, D.L., 2009. Point:counterpoint: respiratory sinus arrhythmia isdue to a centralmechanism vs. respiratory sinus arrhythmia is dueto the baroreflex mechanism. J. Appl. Physiol. 106 (5), 1740–1742.

Evangelista, F.S.,Martuchi, S.E., Negrão, C.E., Brum, P.C., 2005. Lossofresting bradycardia with detraining is associated with intrinsicheart rate changes. Braz. J. Med. Biol. Res. 38 (7), 1141–1146.

Furlan, R., Guzetti, S., Crivellaro, W., et al., 1990. Continuous24-hours assessment of the neural regulation os systemicarterial pressure and RR variabilities in ambulant subjects.Circulation 81, 537–547.

Hashiguchi, H., Ye, S.H., Ross-Cisneros, F., Alexander, N., 1997.Central nitric oxide donors attenuate cardiovascular and centralnorepinephrine responses to stress. Am. J. Physiol. 272,R1447–R1453.

Herman, J.P., Eyigor, O., Ziegler, D.R., Jennes, L., 2000. Expression ofionotropic glutamate receptor subunit mRNAs in thehypothalamic paraventricular nucleus of the rat. J. Comp.Neurol. 422, 352–362.

Hirsh, J.A., Bishop, B., 1981. Respiratory sinus arrhythmia inhumans: how breathing pattern modulates heart rate.American Journal of Physiology 241, H620–H629.

Horn, T., Smith, P.M., McLaughlin, B.E., Bauce, L., Marks, G.S.,Pittman, Q.J., Ferguson, A.V., 1994. Nitric oxide actions inparaventriuclar nucleus: cardiovascular and neurochemicalimplications. Am. J. Physiol. 266, 306–313.

Ichiyama, R.M., Gilber, A.B., Waldrop, T.G., Iwamoto, G.A., 2002.Changes in exercise activation of diencephalic and brainstemcardiorespiratory areas after training. Brain Res. 947, 225–233.

Jackson, K., Silva, H.M., Zhang, W., Michelini, L.C., Stern, J.E., 2005.Exercise training differentially affects intrinsic excitability ofautonomic and neuroendocrine neurons in the hypothalamicparaventricular nucleus. J. Neurophysiol. 94, 3211–3220.

Jankord, R., McAllister, R.M., Ganjam, V.K., Laughlin, M.H., 2009.Chronic inhibition of nitric oxides synthase augments theACTH response to exercise. Am. J. Physiol. Regul. Integr. Comp.Physiol. 296, R728–R734.

Japundzic, N., Grichois, M.L., Zitoun, P., Laude, D., Elghozi, J.L.,1990. Spectral analysis of blood pressure and heart rate inconscious rats: effects of autonomic blockers. J. Auton. Nerv.Syst. 30 (2), 91–100.

Kleiber, A.C., Zheng, H., Schultz, H.D., Peuler, J.D., Patel, K.P., 2008.Exercise training normalizes enhanced glutamate-mediatedsympathetic activation from the PVN in heart failure. Am. J.Physiol. Regul. Integr. Comp. Physiol. 2, 1–43.

Kuru, O., Senturk, U.K., Koçer, G., Ozdem, S., Baskurt, O.K., Çetin,A., Vesilkaya, A., Gunduz, F., 2009. Effect of exercise training onresistance arteries in rats with chronic NOS inhibition. J. Appl.Physiol. 107, 896–902.

Li, Y.F., Mayhan, W.G., Patel, K.P., 2001. NMDA-mediated increasein renal sympathetic nerve discharge within the PVN: role of

76 B R A I N R E S E A R C H 1 3 7 5 ( 2 0 1 1 ) 6 8 – 7 6

nitric oxide. Am. J. Physiol. Heart Circ. Physiol. 281,H2328–H2336.

Li,Y.F.,Roy, S.K.,Channon,K.M., Zucker, I.H., Patel,K.P., 2002. Effectofin vivo gene transfer of nNOS in the PVNon renal nerve dischargein rats. Am. J. Physiol. Heart Circ. Physiol. 282, H594–H601.

Malik, 1996. Heart rate variability. European Heart Journal 17,354–381.

Malliani, A., Pagani, M., Lombardi, F., Cerutti, S., 1991.Cardiovascular neuro regulation explored in thefrequency domain. Circulation 84, 1482–1494.

Malliani, A., Julien, C., Billman, G.E., Cerutti, S., Piepoli, M.F.,Bernardi, L., Sleight, P., Cohen,M.A., Tan, C.O., Laude, D., Elstad,M., Toska, K., Evans, J.M., Eckberg, D.L., 2006. Cardiovascularvariability is/is not an index of autonomic control ofcirculation. J. Appl. Physiol. 101, 684–688.

Martins-Pinge, M.C., Baraldi-Passy, I., Lopes, O.U., 1997. Excitatoryeffects of nitric oxide within the rostral ventrolateral medullaof freely moving rats. Hypertension 30, 704–707.

Martins-Pinge, M.C., Mueller, P.J., Foley, C.M., Friskey, S., Heesch,C., Hasser, E.M., 2004. Cardiovascular effects of nitric oxide andGABA in the paraventricular nucleus of the hypothalamus inconscious rats. Faseb J. 18 (1), A668.

Martins-Pinge, M.C., Becker, L.K., Garcia, M.R.L., Zoccal, D.B., Neto,R.V., Basso, L.S., Souza, H.C., Lopes, O.U., 2005a. Attenuatedpressor responses to amino acids in the rostral ventrolateralmedulla after swimming training in conscious rats. Auton.Neurosci. 122, 21–28.

Martins-Pinge, M.C., Mueller, P.J., Foley, C.M., Heesch, C.M., Hasser,E.M., 2005b. Tonic cardiovascular effects in the PVN ofconscious rats: GABA, Glutamate and nitric oxide interactions.Faseb J. 19 (4), A598.

Medeiros, A., Oliveira, E.M., Gianolla, R., Casarini, D.E., Negrão, C.E.,Brum, P.C., 2004. Swimming training increases cardiac vagalactivity and induces cardiac hypertrophy in rats. Braz. J. Med.Biol. Res. 37, 1909–1917.

Michelini, L.C., Stern, J.E., 2009. Exercise-induced neuronalplasticity in central autonomic networks: role incardiovascular control. Exp. Physiol. 94 (9), 947–960.

Miyagawa, A., Okamura, H., Ibata, Y., 1994. Coexistence of oxytocinand NADPH-diaphorase in magnocellular neurons of theparaventricular and the supraoptic nuclei of the rathypothalamus. Neurosci. Lett. 171, 13–16.

Mueller, P.J., Hasser, E.M., 2006. Putative role of the NTS inalterations in neural control of the circulation exercise trainingin rats. Am. J. Physiol. Regul. Integr. Comp. Physiol. 290 (2),R383–R392.

Mueller, P.J., 2007. Exercise training attenuates increases in lumbarsympathetic nerve activity produced by stimulation of therostral ventrolateral medulla. J. Appl. Physiol. 102, 803–813.

Nafz, B., Wagner, C.D., Persson, P.B., 1997. Endogenous nitric oxidebuffers blood pressure variability between 0.2 and 0.6 Hz in theconscious rat. Am. J. Physiol. 272, H632–H637.

Negrão, C.E., Irigoyen, M.C., Moreira, E.D., Brum, P.C., Freire, P.M.,Krieger, E.M., 1993. Effect of exercise training on RSNA,baroreflex control, and blood pressure responsiveness. Am. J.Physiol. 265 (2 Pt 2), R365–R370.

Pagani, M., Lombardi, F., Guzzetti, S., Rimoldi, O., Furlan, R.,Pizzinelli, P., Sandrone, G., Malfatto, G., Dell'Orto, S., Piccaluga,E., et al., 1986. Power spectral anlysis of heart rate and arterialpressure variability as a marker of sympatho-vagal interactionin man and conscious dog. Circulation Research 59, 178–193.

Paxinos, G., Watson, C., 1998. The Rat Brain in StereotaxicCoordinates, 4a ed. Academic Press, New York.

Rubini, R., Porta, A., Baselli, G., Cerutti, S., Paro, M., 1993. Powerspectrum analysis of cardiovascular variability monitored bytelemetry in conscious unrestrained rats. J. Auton. Nerv. Syst.45, 181–190.

Sanches, I.C., Sartori, M., Jorge, L., Irigoyen, M.C., De Angelis, K.,2009. Tonic and reflex cardiovascular autonomic controlin trained-female rats. Braz. J. Med. Biol. Res. 42 (10),942–948.

Sayers, B.M., 1973. Analysis of heart rate variability. Ergonomics16, 17–32.

Souza, H.C., Ballejo, G., Salgado, M.C., Da Silva, V.J., Salgado, H.C.,2001. Cardiac sympathetic overactivity and decreasedbaroreflex sensitivity in L-NAME hypertensive rats. Am. J.Physiol. Heart Circ. Physiol. 280 (2), H844–H850.

Souza, H.C., Penteado, D.M., Martin-Pinge, M.C., Barbosa Neto, O.,Teixeira, V.P., Blanco, J.H., Silva, V.J., 2007a. Nitric oxidesynthesis blockade increases hypertrophy and cardiac fibrosisin rats submitted to aerobic training. Arq. Bras. Cardiol. 89 (2),88–93.

Souza, H.C., De Araújo, J.E., Martins-Pinge, M.C., Cozza, I.C.,Martins-Dias, D.P., 2009. Nitric oxide synthesis blockadereduced the baroreflex sensitivity in trained rats. Auton.Neurosci. 150 (1–2), 38–44.

Souza, S.B., Flues, K., Paulini, J., Mostarda, C., Rodrigues, B., Souza,L.E., Irigoyen, M.C., De Angelis, K., 2007b. Role of exercisetraining in cardiovascular autonomic dysfunction andmortality in diabetic ovariectomized rats. Hypertension 50,786–791.

Stern, J.E., Li, Y., Zhang, W., 2003. Nitric oxide: a local signallingmolecule controlling the activity of pre-autonomic neurones inthe paraventricular nucleus of the hypothalamus. Acta Physiol.Scand. 177 (1), 37–42.

Stern, J.E., 2004. Nitric oxide and homeostatic control: anintracellular signalling molecule contributing to autonomicand neuroendocrine integration? Prog. Biophys. Mol. Biol. 84,197–215.

Swanson, L.W., Sawchenko, P.E., 1983. Hypothalamic integration:organization of the paraventricular and supraoptic nuclei.Annu. Rev. Neurosci. 6, 269–324.

Swanson, L.W., Sawchenko, P.E., Wiegand, S.J., Price, J.L., 1980.Separate neurons in the paraventricular nucleus project to themedian eminence and to the medulla or spinal cord. Brain Res.198, 190–195.

Task Force Of The European Society Of Cardiology And TheNorth American Society Of Pacing And Electrophysiology, 1996.Heart rate variability: standards of measurement,physiological interpretation and clinical use. Circulation 93,1043–1065.

Tezini, G.C., Silveira, L.C., Maida, K.D., Blanco, J.H., Souza, H.C.,2008. The effect of ovariectomy on cardiac autonomic controlin rats submitted to aerobic physical training. Auton. Neurosci.143 (1–2), 5–11.

Zanesco, A., Antunes, E., 2007. Effects of exercise training on thecardiovascular system: pharmacological approachesPharmacol. Ther. 114 (3), 307–317.

Zhang, K., Mayhan, W.G., Patel, K.P., 1997. Nitric oxide within theparaventricular nucleus mediates changes in renalsympathetic nerve activity. Am. J. Physiol. 273, 864–872.

Zhang, K., Patel, K.P., 1998. Effect of nitric oxide within theparaventricular nucleus on renal sympathetic nerve discharge:role of GABA. Am. J. Physiol. Regul. Integr. Comp. Physiol. 275,728–734.

Zhang, K., Li, Y.F., Patel, K.P., 2001. Blunted nitric oxide-mediatedinhibition of renal nerve discharge within PVN of rats withheart failure. Am. J. Physiol. Heart Circ. Physiol. 281,995–1004.

Zheng, H., Li, Y.F., Cornish, K.G., Zucker, I.H., Patel, K.P., 2005.Exercise training improves endogenous nitric oxidemechanisms within the paraventricular nucleus in ratswith heart failure. Am. J. Physiol. Heart Circ. Physiol. 288,2332–2341.