ORIGINAL ARTICLE

Vagal Stimulation, Through its Nicotinic Action, LimitsInfarct Size and the Inflammatory Response to Myocardial

Ischemia and Reperfusion

Laura Calvillo, PhD,* Emilio Vanoli, MD,†‡ Elisa Andreoli, PhD,‡ Alessandra Besana, PhD,*

Elisabetta Omodeo, PhD,§ Massimiliano Gnecchi, MD, PhD,†¶ Pietro Zerbi, MD,§

Gianluca Vago, MD,§ Giuseppe Busca,k and Peter J. Schwartz, MD†¶**††

Abstract: Vagal activity has protective effects in ischemic heart

disease. We tested whether vagal stimulation (VS) could modulate the

inflammatory reaction, a major determinant of cardiac injury after

ischemia/reperfusion. Four groups of male rats underwent myocardial

ischemia (30 minutes) and reperfusion (24 hours). One group un-

derwent VS (40 minutes), 1 VS plus atrial pacing (VS + Pacing), and

1 VS plus nicotinic inhibition by mecamylamine (VS + MEC). After

24 hours, the area at risk, infarct size, inflammation parameters, and

apoptosis were quantified. Infarct size was reduced in all VS-treated

rats (controls, 53 6 18%; VS, 6.5 6 3%; VS + Pacing, 23 6 6%;

VS + MEC, 33 6 9%; P , 0.005 vs. controls). The infarct size in the

VS + MEC group was larger than that in VS-treated animals, despite

similar heart rate, suggesting partial loss of protection. The number of

macrophages, neutrophils, and apoptotic cells in the area at risk and the

plasma cytokines levels were significantly reduced in all VS-treated

animals. In conclusion, VS decreases infarct size and inflammatory

markers during ischemia/reperfusion independent of the heart rate. The

anti-inflammatory and antiapoptotic properties of the nicotinic pathway

are the primary underlying mechanism. The vagally mediated modu-

lation of inflammatory responses may prove valuable in the clinical

management of acute coronary syndromes and of heart failure.

Key Words: vagal activity, ischemia/reperfusion, myocardial in-

farction, autonomic nervous system, infarct size, inflammatory

markers

(J Cardiovasc Pharmacol TM 2011;58:500–507)

INTRODUCTIONRecent clinical evidence suggests that chronic vagal

stimulation (VS) is feasible in man and that it might bebeneficial in the setting of chronic ischemic cardiomyopathyand heart failure.1,2 We are now advancing the hypothesis thatVS might have protective effects also at a much earlier stage,when the initial ischemic insult occurs.

Ischemia and reperfusion have complex and deleteriouseffects on the myocardium. Inflammatory mediators areessential in the healing process of injured tissue, but when,after reperfusion, their response is excessive, they may worsenmyocardial injury.3

Reperfusion is followed by rapid cellular infiltration ofneutrophils and monocytes and by the release of severalproinflammatory cytokines which orchestrate inflammationand tissue repair processes.3,4 Among them, lipopolysaccha-ride-inducible CXC chemokine (LIX) and monocyte chemo-attractant protein-1 (MCP-1) play a pivotal role.5,6 There isa fine balance between the inflammatory reaction, whichfollows reperfusion with the potential of extending myocardialdamage, and the mechanisms, which can positively affecttissue repair. A sound approach to an effective management ofthis important clinical problem has been hampered by what seemsto be a physiological catch 227: on one hand, inflammatorymediators are essential in the healing of injured tissue; on theother hand, an excessively intense inflammatory response afterreperfusion may actually worsen myocardial injury.

A rationale solution might be provided by identifyingmeans to physiologically modulate this inflammatory reaction.One of the areas, largely unexplored and of potential relevancein the modulation of the responses to ischemia andreperfusion, is the neural control of the heart. We considerthe possibility that vagal activity might represent 1 suchphysiological modulator. We focused on the vagus nervebecause there is evidence suggesting its significant role andbecause over the years, our experimental and clinical studieson the neural control of the heart8 have provided data relevantto the present study. We demonstrated the protective effect ofVS in the prevention of ischemia-induced ventricularfibrillation,9 the association between depressed vagal reflexes,and increased cardiac mortality after a myocardial infarc-tion,10,11 and more recently that chronic VS is not only feasibleand safe in man but that it can effectively contribute to themanagement of patients with advanced heart failure.1,2

Received for publication May 12, 2011; accepted June 28, 2011.From the *Laboratory of Cardiovascular Genetics, IRCCS Istituto Auxologico

Italiano, Milan, Italy; †Section of Cardiology, Department of Lung, Bloodand Heart, University of Pavia, Pavia, Italy; ‡Department of Cardiology,Policlinico di Monza, Monza, Italy; §Pathology Unit, Institute ofBiomedical Sciences, L. Sacco Hospital, The University of Milan, Italy;{Department of Cardiology, Fondazione IRCCS Policlinico S. Matteo,Pavia, Italy; kDepartment of Cardiovascular Medicine, The University ofMilan, Fondazione IRCCS Ca Granda, Ospedale Maggiore Policlinico,Milan, Italy; **Cardiovascular Genetics Laboratory, Hatter Institute forCardiovascular Research in Africa, Department of Medicine, University ofCape Town, South Africa; and ††Department of Family and CommunityMedicine, College of Medicine, King Saud University, Riyadh, SaudiArabia.

The authors report no conflict of interest.Reprints: Peter J. Schwartz, MD, Department of Cardiology, Fondazione

IRCCS Policlinico S. Matteo, Viale Golgi 19, 27100 Pavia, Italy (e-mail:peter.schwartz@unipv.it).

Copyright � 2011 by Lippincott Williams & Wilkins

500 | www.jcvp.org J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011

A protective role of 24-hour VS in experimentalmyocardial ischemia was suggested because of its associationwith decreased infarct size, cytokine expression, and neutrophilinfiltration in reperfusion injury12,13 and represents an area ofactive research.14 To better understand the mechanisms un-derlying this apparent protection, we discriminated between thenicotinic and muscarinic actions of vagal activity and we alsotested whether a more clinically realistic 40-minute stimulationwould retain its efficacy.

MATERIALS AND METHODSOur investigation conforms with the Guide for the Care

and Use of Laboratory Animals published by the US NationalInstitutes of Health (NIH Publication No. 85-23, revised1996), it was approved by the Ethics Review Board of theItalian Ministry of Health, and all procedures were performedin accordance with the animal care guidelines of Federationof Laboratory Animal Science Associations.15 Ischemic injury,with or without VS, was made in male Sprague-Dawley rats(250–295 g).

Experimental ProtocolSprague-Dawley rats were randomized into 5 groups,

and results are reported for the 44 animals that completed theprotocol. All treatment animals received a bolus of 5 mg/kgcarprofen SC 30 minutes before surgery as analgesic. Theprotocol of each group is as follows:� Group 1 (n = 13): controls; ischemia (30 minutes) and

reperfusion (24 hours). The right vagus was exteriorized butnot stimulated.� Group 2 (n = 6): ischemia and reperfusion with VS (the right

cervical vagus was stimulated from 5 minutes beforeischemia to 5 minutes after reperfusion).� Group 3 (n = 9): same as group 2 but with the addition of left

atrial pacing to maintain heart rate at same level present ingroup 1.� Group 4 (n = 9): same as group 2 but with the addition of

mecamylamine (MEC) injection in the heart to inhibitnicotinic receptors.16

� Group 5 (n = 7): sham-operated rats, no ischemia andreperfusion.

Ischemia and ReperfusionAll rats were anesthetized with isoflurane and ventilated

through an endotracheal cannula. Anesthetized animals werecontinuous monitored for body temperature and electrocardio-gram (ECG).

The left coronary artery was ligated with 4-0 silk suture(Ethicon), and a plain knot was tied over 2 pieces of suture thatwere pulled after 30 minutes to reperfuse the heart.17 The chestwas closed under negative pressure. Ischemia was alwaysconfirmed by the appearance of discoloration of the cardiacsurface and ST elevation on ECG, whereas reperfusion wasverified by reddening of previously discolored area and by thepresence of arrhythmias on ECG. Sham-operated rats un-derwent the same surgical procedures, without ligation of theleft coronary artery.

Nicotinic InhibitionMEC, the inhibitor of nicotinic receptors, was injected in

the coronary arteries of the animals assigned to group 4. Afterclamping the ascending tract of the aortic arch, a bolus ofMEC (2.1 mg/kg) was injected in the left ventricle. The clampwas removed after the injection; overall, the procedure lastedless than 30 seconds.

VS and PacingElectrical stimulation was performed by a digital

stimulator Grass S88. The right vagus nerve was exteriorizedand isolated at the cervical level through a single midlineincision, and an electrode was placed around it. Stimulusparameters were 2.5 V, 0.5-ms pulse duration, and 8–10 Hzfrequency.9 Atrial pacing was performed by means of a29-gauge needle electrode placed in the left atrium, whereasthe reference electrode was placed on the chest. In the VS +Pacing group, heart rate was maintained at 330 beats perminute (bpm), (the average value observed in the control rats).

Sample Collection and Determination of Areaat Risk and Infarct Size

After 24 hours of reperfusion, the chest was reopenedunder anesthesia and ventilation. One milliliter of blood waswithdrawn in heparin and centrifuged for 14 minutes at 3500revolutions per minute at 4�C, and plasma was stored at220�C. Then the ligature around the coronary artery wastightened again and 2 mL of 5% Evans Blue in saline solutionwas injected into the right ventricular chamber to distinguishthe area at risk from the perfused area of the left ventricle. Theheart was then excised and cut transversally into 3 main slices.In the first slice, the septum and the area at risk were excisedand quickly frozen in liquid nitrogen. The second slice wasfixed in formalin for histology and immunohistochemistry.The third slice was frozen at 220�C in optimal cuttingtemperature compound for tissue inclusion and cut trans-versally in 1-mm serial sections with the McIlwain tissuechopper and incubated for 20 minutes at 37�C in triphenylte-trazolium chloride. Photographs of the sections were taken, andthe areas of perfused tissue (blue), viable ischemic tissue (red),and necrotic ischemic tissue (white) were measured with animage analyzer (Image Tool UTHSCSA for Windows) and wasused to quantify the perfused left ventricle, the area at risk, andthe necrotic area (Fig. 1). The necrotic area was expressed as thepercentage of the area at risk.

Determination of Macrophage and NeutrophilInfiltration, a-7 Subunit of the NicotinicAcetylcholine Receptor Expression on CardiacMacrophages, and Apoptosis

One coronal slice of the heart was formalin-fixed andparaffin-embedded; then, 4 serial sections were cut from eachblock and the following immunostaining reactions wereperformed.

MacrophagesAfter deparaffinization in xylene and endogenous

peroxidase blockade with 3% hydrogen peroxide for 20 minutes,antigen retrieval was carried out by microwave treatment in 0.01

� 2011 Lippincott Williams & Wilkins www.jcvp.org | 501

J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011 VS Protects From Ischemia/Reperfusion

M EDTA buffer. Slides were incubated for 2 hours at roomtemperature with a mouse antirat CD68 monoclonal antibody(1:100; AbD Serotec, Oxford, United Kingdom). After washingwith Tris buffer solution, slides were incubated with a ‘‘mouseon rat’’ HRP-polymer kit (Biocare Medical, Concord, CA) andSuper Sensitive IHC detection system (BioGenex, San Ramon,CA), respectively. Peroxidase activity was visualized withdiaminobenzidine. Slides were counterstained with hematoxylin.

Double Immunofluorescence Staining for Macro-phages and a-7nAChR

After incubation with the primary antibodies (mouseantirat CD68 at 1:50 and a rabbit antirat polyclonal antibodya-7nAChR 1:10 from Abcam, Cambridge, United Kingdom),slides were incubated with secondary antibodies (Alexa Fluor488 antirabbit and IgG and Alexa Fluor 594 antimouse IgG;Invitrogen, Carlsbad, CA) both at 1:500 dilutions. Nuclei werecounterstained with Hoechst (1:500).

Polymorphonuclear LeukocytesNaphthol AS-D chloroacetate esterase kit (Sigma-Aldrich,

Inc, St Louis, MO) was used to detect polymorphonuclearleukocytes (PMN).

ApoptosisIn situ detection of apoptosis was performed using

CardioTacs kit (Trevigen, Inc, Gaithersburg, MD) according tothe manufacturer’s instructions. Optimized conditions in ourlaboratory were permeabilization with proteinase K at 37�Cfor 15 minutes and labeling procedure with TdT enzyme for90 minutes.

Plasma Cytokines Expression

Cytokine Array

To detect LIX expression, we used a Rat Cytokine ArrayPanel A kit (R&D Systems) following the kit instructions.Briefly, the membranes were blocked with a blocking bufferand then plasma was incubated for 1 hour at room temperature.The blocking buffer was then removed from the membranes,and the sample–antibody mix was added for the overnightincubation at 4�C. After washing, the membranes wereincubated at room temperature for 30 minutes with 1.5 mL ofhorseradish peroxidase–conjugated streptavidin and then

thoroughly washed. The membranes were then developed byusing enhanced chemiluminescence technique (GE Health-care, formerly Amersham Biosciences), exposed to x-ray film,and processed by autoradiography.

To detect MCP-1 expression, we used a Rat CytokineAntibody Array kit (RayBiotech) following the kit instruc-tions. Briefly, the membranes were blocked with a blockingbuffer and then incubated with the plasma samples for 2 hoursat room temperature. The membranes were washed and thenincubated with 1 mL of primary biotin-conjugated antibody at4�C overnight. After washing, the membranes were incubatedat room temperature for 2 hours with 2 mL of horseradishperoxidase–conjugated streptavidin and then thoroughlywashed. The membranes were then developed by usingenhanced chemiluminescence technique, exposed to X-rayfilm and processed by autoradiography.18 For data analysis,autoradiographs of the arrays were scanned to determine thedensity of the protein array positions and pixel intensity wasmeasured with the image analyzer NIH ImageJ 1.41 softwarefor Windows. The values from scans were adjusted based onthe intensity of control spots on the filter corners.

Statistical AnalysisContinuous variables are presented as mean 6 SD. Their

comparisons between groups were performed by 1-way analysisof variance. All variables, with the exception of heart rate and areaat risk, for which the analysis of variance showed significantdifferences were further analyzed by post hoc comparisons usingthe Bonferroni method. Whenever the assumption of homoge-neity of variance, evaluated by the Levene test, was questionable,the Games–Howell test for multiple comparisons was used. P ,0.05 was considered significant.

RESULTS

Effect of VS on Infarct Size and Heart RateAt baseline, heart rate was similar in all groups (Table 1).

Similarly, the area at risk, calculated as percentage of the leftventricle, was not significantly different (61 6 16% in the controlgroup, 52 6 8% in the VS group, 60 6 10% in the VS + Pacinggroup, and 48 6 8% in the VS + MEC group; P = NS). VSmarkedly decreased infarct size, compared with control, from53 6 18% to 6.5 6 3% (P , 0.001). Keeping heart ratecontrolled by atrial pacing blunted, but did not eliminate, the

FIGURE 1. Determination of area atrisk and infarcted zone. Evans Blue insaline solution was injected into theright ventricle to distinguish the areaat risk from the perfused area of theleft ventricle. Photographs of thesections were taken, and the areas ofperfused tissue (blue), viable ische-mic tissue (red), and necrotic tissue(white) were measured with animage analyzer and used to quantifythe perfused left ventricle, the areaat risk, and the necrotic area. Twenty-four hours after surgery, the infarcted area was significantly reduced in VS and in VS + Pacing(VS + PAC)–treated animals with respect to controls.

502 | www.jcvp.org � 2011 Lippincott Williams Wilkins

Calvillo et al J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011

protective effect of VS, as infarct size was still significantlyreduced to 23 6 6% in the VS + Pacing group (P , 0.01; Fig. 1).Inhibition of cardiac nicotinic receptors by MEC reduced theprotective effect of VS because the infarct size in group 4 (VS +MEC) was significantly larger (33 6 9% vs. 6.5 6 3%, P ,0.001) despite a similar reduction in heart rate during ischemia(234 6 12 bpm in the VS group, 251 6 17 bpm in the VS +MEC group; P = NS).

VS and Inflammatory Response

Effects on Infiltrating Cells

Compared to control animals, VS significantly reducedthe number of infiltrated macrophages in all treatment groupswith the exception of VS + MEC (Figs. 2A, B and Table 1,

P , 0.001 for all comparisons). Pacing did not affect the resultof VS, as the number of infiltrating macrophages in the area atrisk remained essentially the same in these 2 groups (491 6240 and 600 6 215 in the VS and the VS + Pacing groups,respectively). Cardiac nicotinic receptor inhibition associatedwith VS produced a nonsignificantly greater number ofmacrophages compared with VS (962 6 381 vs. 491 6 240,NS). In the sham-operated animals, the number of infiltratedmacrophages in the left ventricle free wall was extremely low(33 6 28) and significantly lower compared with all the othergroups (Table 1). The macrophages infiltrating the infarctedand reperfused myocardial tissue expressed immunoreactivityfor the a-7nAChR (Fig. 3).

VS had a similar effect in reducing the number of PMNin myocardial tissue 24 hours after ischemia and reperfusion.

TABLE 1. Effect of VS and Nicotinic Receptor Inhibition on Heart Rate, Area at Risk, Infarct Size, Apoptosis, Macrophages, andPMN After Ischemia and Reperfusion

GroupsBasal

HR (bpm)

HR AfterIschemia orSham (bpm)

AAR/LV(%)

IS/AAR(%)

No.Macrophages

in AARNo. PMNin AAR

No. ApoptoticCell inAAR

Control 332 6 43 334 6 39 61 6 16 53 6 18 1406 6 343 174 6 59 190 6 88

VS 303 6 38 234 6 15† 52 6 8 6.5 6 3† 491 6 240†§ 48 6 39** 49 6 27*

VS + Pacing 301 6 41 330 6 0{ 60 6 10 23 6 6**f 600 6 215†¤ 60 6 41** 59 6 10*

VS + MEC 327 6 25 251 6 17† 48 6 8 33 6 9*{ 962 6 381¤ 142 6 85§ 153 6 67§§Sham 344 6 54 303 6 49 NA NA 33 6 28† 42 6 27† 45 6 9*

*P , 0.05, **P , 0.01, †P , 0.001 vs. control.§P , 0.05, fP , 0.01, {P , 0.001 VS + Pacing or VS + MEC vs. VS.§P , 0.05 vs. sham; ¤P , 0.01 vs. sham.AAR, area at risk; HR, heart rate; IS, infarct size; LV, left ventricle; NA, not applicable.

FIGURE 2. Macrophage and PMNinfiltration in the area at risk. Immu-nohistochemical staining shows mac-rophage and PMN infiltration incontrol animals (A and C, respec-tively) and in animals treated with VS(B and D) (magnification, 320).

� 2011 Lippincott Williams & Wilkins www.jcvp.org | 503

J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011 VS Protects From Ischemia/Reperfusion

Compared with control animals, the number of PMN wasreduced significantly (P , 0.01) in both VS and VS + Pacinggroups (Figs. 2C, D and Table 1). In contrast, the number ofPMN in the VS + MEC group was not significantly differentfrom the control group and from the VS and VS + Pacinggroups.

In the sham-operated animals, the number of PMN was42 6 27, markedly lower versus controls (P , 0.001), lowerversus VS + MEC group (P , 0.05), but similar to that of VSand VS + Pacing groups.

Effects on CytokinesA rat cytokine-array approach, to detect the presence of

proteins in the plasma, was used to identify specific cytokinesreleased after myocardial ischemia and reperfusion injury.Data from a representative array are shown in Figure 4. Thesignal intensity for the 2 cytokines involved in the recruitmentof neutrophils (LIX) and macrophages (MCP-1) was 2-foldhigher in the heart of control rats when compared with that insham-operated rats. VS, even in presence of controlled HR,decreased the signal intensity by almost 2-fold, thus reversing

FIGURE 3. Immunostaining for thea-7nAChR and CD68. A, Staining fora-7nAChR in green. B, CD68 stain-ing in red; the microphotograph in(C) represents the merge of (A) and(B). D, Another representative dou-ble staining for a-7nAChR andCD64, clearly demonstrating thatthe infiltrating macrophages coex-press a-7nAChR receptor.

FIGURE 4. Rat protein array analysis was used todetermine the differences in the level of rat plasmacytokines during ischemia and reperfusion injury(control). Cytokines levels were decreased by VS +Pacing (VS + PAC) treatment. Autoradiographs of thearrays were scanned to determine the density of theprotein array positions. The values from scans wereadjusted based on the intensity of control spots onthe filter corners, and the level decreases for specificcytokines are shown.

504 | www.jcvp.org � 2011 Lippincott Williams Wilkins

Calvillo et al J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011

the consequences of ischemia and reperfusion on cytokineplasma levels. Local inhibition of nicotinic receptors in VS +MEC–treated rats did not affect the plasma levels of LIX andMCP-1 with respect to VS-treated animals.

VS and ApoptosisCompared with control myocardial tissue, which con-

tained 190 6 88 apoptotic cells per area at risk, the myocardiumof rats undergoing VS showed (Fig. 5) a marked and significantreduction in their number (49 6 27), which was not modified bypacing (59 6 10) (P , 0.05 for both comparisons). Cardiacnicotinic receptor inhibition resulted in an increased number ofapoptotic cells with respect to VS group (153 6 67 in VS +MEC group vs. 49 6 27 in VS-treated rats, P # 0.05). In thesham-operated animals, this number was 45 6 9, almostidentical to that observed with VS and similar to that of VS +Pacing (NS); In contrast, it was significantly (P , 0.05) lowercompared with controls and also with VS + MEC, thus showingthat nicotinic receptor inhibition had largely interfered with theprotective effect of VS.

DISCUSSIONThe main finding of the present study is that a brief

period of VS, independent of its muscarinic action, drasticallylimits infarct size and favorably attenuates the physiologicalresponse to acute myocardial ischemia and reperfusion by itsnicotinic pathway. This protective effect is independent fromheart rate changes, is demonstrated by a significant reductionin the number of macrophages, PMN, and apoptotic cells, andis paralleled by decreased levels of circulating proinflamma-tory cytokines.

These results provide new insights on how the neuralcontrol of the heart may modulate cardiac responses to life-threatening events and limit, or worsen, the damage producedby ischemia and reperfusion. Also, they may offer a clue tounderstand, at least in part, some recent results obtained bychronic VS in patients with heart failure, which are not entirelyexplained by heart rate changes.1,2

Cardiac Responses DuringIschemia–Reperfusion and the Role of VS

At the time of reperfusion, blood containing leukocytesenters an area rich in chemotactic factors and inflammatorymediators.19,20 VS inhibits not only 2 of the key players in this

process, LIX (IL-8 analogue in the rat) and MCP-1, but alsothe infiltration by PMN and macrophages, which is essential inthe entire innate immune process after myocardial reperfusioninjury.21

LIX is a murine chemokine with almost the same functionas IL-8, responsible for 80% of PMN infiltration into an ischemicmyocardium.5 MCP-1 (CCL2) is a small cytokine with profibroticproperties that recruits monocytes, memory T cells, and dendriticcells to sites of tissue injury and seems to regulate fibrous tissuedeposition in the injured heart, critically regulating also mono-nuclear cell recruitment and activation in healing myocardialinfarcts.6,22 All these cytokines have proinflammatory propertiesthat seem to be clinically relevant. LIX attracts PMN in the site ofinjury, thus allowing the intense reaction triggered by myeloper-oxidase (MPO). MPO is a peroxidase present in PMN thatdamages cardiomyocytes during reperfusion. MPO serum levelsin patients with acute coronary syndrome, monitored during a 6-month follow-up, correlate significantly with increased risk forsubsequent cardiovascular events.23

MCP-1, responsible for the recruitment of monocytes tosites of inflammation, seems to play a critical role inatherosclerosis and in remodeling after myocardial infarction.In a large cohort of patients with acute coronary syndromes, anelevated baseline level of MCP-1 was associated both withtraditional risk factors for atherosclerosis and with anincreased risk for death or myocardial infarction.24 Moreover,in patients with congestive heart failure, MCP-1 levels weresignificantly inversely correlated with left ventricular ejectionfraction and were particularly high in those with heart failureof ischemic origin.25

VS inhibits also the apoptotic process in the ischemicand reperfused myocardium, a possible explanation beinga PI3K/Akt/HIF-1alpha pathway activation inside the targetcells. This would be supported by the evidence thatacetylcholine had protective effects on rat cardiomyocytessubjected to normoxia/hypoxia by increasing Akt phosphor-ylation and preventing hypoxia-induced apoptosis andmitochondrial membrane potential collapse.26 Activation ofthe same pathway by VS could explain our results.

When taken together, these data strongly suggest that theimmune reactions occurring during acute myocardial ischemiacan worsen prognosis and that the molecules analyzed in thepresent study and modified by VS are likely to play a majorrole in the harmful side effects of innate immune responses.

FIGURE 5. Apoptotic cells in the leftventricle. Myocardial tissue fromcontrol animals (A) had significantlyhigher number of apoptotic cells/slice compared with vagal stimu-lated rats (B) (magnification, 320).

� 2011 Lippincott Williams & Wilkins www.jcvp.org | 505

J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011 VS Protects From Ischemia/Reperfusion

VS and Nicotinic InhibitionThe specific functional role of the nicotinic pathway in

our results was assessed by inhibiting nicotinic receptors withMEC during VS. The myocardial nicotinic receptors wereinhibited by a local injection in the heart without affecting thesystemic cholinergic preganglionic neurons. Despite a similarheart rate–lowering effect, cardiac nicotinic receptor inhibitionby MEC largely prevented the reduction of both the infarct sizeand the number of macrophages, PMN and apoptotic cellsproduced by VS alone. Furthermore, local inhibition of thenicotinic receptors did not affect the plasma content of LIXand MCP-1 with respect to VS-treated animals, suggestingthat the nicotinic pathway importantly contributed to thevagally mediated cardioprotection.

It is true that 77% of the infarct size reduction producedby VS was lost after inhibition of the nicotinic receptors, butthis also means that 23% is contributed by the muscariniccomponent. We have elected not to block muscarinic receptorsbecause the ensuing large increase in heart rate would haveintroduced a significant confounder.

The Cholinergic Anti-inflammatory PathwayWe explored whether it might be reasonable to suggest

an involvement of the cholinergic anti-inflammatory pathwayin myocardial ischemia and reperfusion injury. This pathwayis a bidirectional communication between the brain andthe immune system that seems to play a critical role in thecontrol of inflammation as cholinergic neurons inhibit acuteinflammatory response and, conversely, inflammation inperipheral tissue alters neural signaling in hypothalamus.27,28

It seems that one of the main cell types involved in thisprocess is the macrophage, through the a-7nAChR on itssurface.28–30

We tested whether the macrophages infiltrated in theheart would express a-7nAChR on their surface after ischemiaand reperfusion, which would suggest a possible involvementof the cholinergic anti-inflammatory pathway in myocardialreperfusion injury. By performing a double staining withimmunofluorescence on macrophages infiltrated in the myo-cardial injured tissue, we did indeed observe that they exhibitedimmunoreactivity for the subunit a7 on their surface. Themacrophages expressing a-7nAChR were detected in severalzones of the area at risk.

The lack of a commercial antibody able to neutralizebioactivity in vivo hampered our possibility of inhibitinga-7nAChR receptors on infiltrated macrophages or in othercells in the rat. As a consequence, we regard the a-7nAChRstaining on macrophages surface just as a further stepsuggesting the involvement of the cholinergic anti-inflamma-tory pathway in myocardial reperfusion injury. However,despite the evidence of an anti-inflammatory and cardiopro-tective property of the nicotinic component of VS, we couldnot discriminate between the different subunits of the nicotinicreceptor. It is reasonable to posit that nicotinic protection couldactually be mediated by the a-7 receptor subunit, but theevidence for the real contribution of the cholinergic anti-inflammatory pathway to the myocardial response to reperfu-sion injury must await further proof.

Relationship With Heart Failure of IschemicOrigin and Clinical Implications

The present data are clinically relevant, given that thestrong evidence, experimental and clinical,13–21 linking de-pressed vagal activity to poor outcome in patients with ischemicheart disease and heart failure, has already prompted a first-in-man assessment of the feasibility and safety of chronic VS withencouraging results.1,2 Our data show that even a brief VS,lasting only 40 minutes during ischemia and reperfusion canprotect the myocardium from inflammatory and ischemic injury.Uemura et al13 observed cardioprotection after 24 to 72 hours ofcontinuous VS after ischemia and reperfusion. A shorter (,1hour) VS would be clinically implementable and might beconsidered in conjunction with angioplasty or other clinicalconditions with a potential for reperfusion injury, such as bypassgrafting or cardiac transplant.

VS shows a remarkable ability to modulate innateimmune responses, cellular networks, and organ homeostasis.A major difference between a simple pharmacologicalinhibition of cytokines or leukocytes and VS is that the latterrepresents an already existing physiological anti-inflammatorymechanism, which just needs reinforcement. The possibility offine-tuning, physiologically a complex process involvingcytokines, PMN, and neurohormones, is especially attractivein light of the current failure of pharmacological interventionsaimed to modulate the inflammatory responses to ischemicheart disease and heart failure.31,32 The fact that all theseresponses can be modulated by VS points to unexpected crosstalk between pathways present in the myocardium and raisesthe possibility of novel therapeutic interactions.33

ACKNOWLEDGMENTThe authors are grateful to Dr Lidia Cova and Dr Renato

Pataccini for their very helpful technical advices and for theirprofessional support, to Dr. Carla Spazzolini for expertstatistical assistance, and to Pinuccia De Tomasi for experteditorial support.

REFERENCES1. Schwartz PJ, De Ferrari GM, Sanzo A, et al. Long term vagal stimulation

in patients with advanced heart failure. First experience in man. Eur JHeart Fail. 2008;10:884–891.

2. De Ferrari GM, Crijns HJ, Borggrefe M, et al; for the CardioFitMulticenter Trial Investigators. Chronic vagus nerve stimulation: a newand promising therapeutic approach for chronic heart failure. Eur Heart J.2011;32:847–855.

3. Entman ML, Smith CW. Postreperfusion inflammation: a model forreaction to injury in cardiovascular disease. Cardiovasc Res. 1994;28:1301–1311.

4. Hawkins HK, Entman ML, Zhu JY, et al. Acute inflammatory reactionafter myocardial ischemic injury and reperfusion. Development and use ofa neutrophil-specific antibody. Am J Pathol. 1996;148:1957–1969.

5. Chandrasekar B, Smith JB, Freeman GL. Ischemia-reperfusion of ratmyocardium activates nuclear factor-kb and induces neutrophil infiltrationvia lipopolysaccharide-induced CXC chemokine. Circulation. 2001;103:2296–2302.

6. Frangogiannis NG, Dewald O, Xia Y, et al. Critical role of monocytechemoattractant protein-1/CC chemokine ligand 2 in the pathogenesis ofischemic cardiomyopathy. Circulation. 2004;115:584–592.

7. Heller J. Catch 22. Great Britain, United Kingdom: Jonathan Cape Ltd;1962.

506 | www.jcvp.org � 2011 Lippincott Williams Wilkins

Calvillo et al J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011

8. Levy MN, Schwartz PJ. Vagal Control of the Heart: ExperimentalBasis and Clinical Implications. Armonk, NY: Futura Publishing Co;1994:644.

9. Vanoli E, De Ferrari GM, Stramba-Badiale M, et al. Vagal stimulation andprevention of sudden death in conscious dogs with a healed myocardialinfarction. Circ Res. 1991;68:1471–1481.

10. Schwartz PJ, Vanoli E, Stramba-Badiale M, et al. Autonomic mechanismsand sudden death. New insights from analysis of baroreceptor reflexes inconscious dogs with and without a myocardial infarction. Circulation.1988;78:969–979.

11. La Rovere MT, Bigger JT Jr, Marcus FI, et al; for the ATRAMI (AutonomicTone and Reflexes After Myocardial Infarction) Investigators. Baroreflexsensitivity and heart-rate variability in prediction of total cardiac mortalityafter myocardial infarction. Lancet. 1998;351:478–484.

12. Katare RG, Ando M, Kakinuma Y, et al. Differential regulation of TNFreceptors by vagal nerve stimulation protects heart against acute ischemicinjury. J Mol Cell Cardiol. 2010;49:234–244.

13. Uemura K, Zheng C, Li M, et al. Early short-term vagal nerve stimulationattenuates cardiac remodeling after reperfused myocardial infarction.J Card Fail. 2010;16:689–699.

14. Li W, Olshanky B. Inflammatory cytokines and nitric oxide in heart failureand potential modulation by vagus nerve stimulation. Heart Fail Rev.2011;16:137–145.

15. Nicklas W. FELASA guidelines for the accreditation of health monitoringprogrammes and for testing laboratories involved in health monitoring.Lab Anim. 2010;44:69.

16. Li XW, Wang H. Non-neural nicotinic alpha 7 receptor, a new endothelialtarget for revascularization. Life Sci. 2006;78:1863–1870.

17. Michael LH, Entman ML, Hartley CJ, et al. Myocardial ischemia andreperfusion: a murine model. Am J Physiol. 1995;269:H2147–H2154.

18. Xu Y, Kulkosky J, Acheampong E, et al. HIV-1-mediated apoptosis ofneuronal cells: proximal molecular mechanisms of HIV-1-inducedencephalopathy. Proc Natl Acad Sci U S A. 2004;101:7070–7075.

19. Pinckard RN, Olson MS, Kelley RE, et al. Antibody-independent activationof human C1 after interaction with heart subcellular membranes. J Immunol.1973;110:1376–1382.

20. Ley K. Histamine can induce leukocyte rolling in rat mesenteric venules.Am J Physiol. 1994;267:H1017–H1023.

21. Frangogiannis NG, Youker KA, Rossen RD, et al. Cytokines and themicrocirculation in ischemia and reperfusion. J Mol Cell Cardiol. 1998;30:2567–2576.

22. Frangogiannis NG. Chemokines in the ischemic myocardium: frominflammation to fibrosis. Inflamm Res. 2004;53:585–595.

23. Baldus S, Heeschen C, Meinertz T, et al; CAPTURE Investigators.Myeloperoxidase serum levels predict risk in patients with acute coronarysyndromes. Circulation. 2003;108:1440–1445.

24. de Lemos JA, Morrow DA, Sabatine MS, et al. Association betweenplasma levels of monocyte chemoattractant protein-1 and long-termclinical outcomes in patients with acute coronary syndromes. Circulation.2003;107:690–695.

25. Aukrust P, Ueland T, Muller F, et al. Elevated circulating levels of C-Cchemokines in patients with congestive heart failure. Circulation. 1998;97:1136–1143.

26. Kakinuma Y, Ando M, Kuwabara M, et al. Acetylcholine from vagalstimulation protects cardiomyocytes against ischemia and hypoxiainvolving additive non-hypoxic induction of HIF-1a. FEBS Lett. 2005;579:2111–2118.

27. Tracey KJ. The inflammatory reflex. Nature. 2002;420:853–859.28. Pavlov VA, Tracey KJ. The cholinergic anti-inflammatory pathway. Brain

Behav Immun. 2005;19:493–499.29. Wang H, Yu M, Ochani M, et al. Nicotinic acetylcholine receptor a7

subunit is an essential regulator of inflammation. Nature. 2003;421:384–388.

30. Pavlov VA, Wang H, Czura CJ, et al. The cholinergic anti-inflammatorypathway: a missing link in neuroimmunomodulation. Mol Med. 2003;9:125–134.

31. Mann DL, McMurray JJ, Packer M, et al. Targeted anticytokine therapy inpatients with chronic heart failure: results of the Randomized EtanerceptWorldwide Evaluation (RENEWAL). Circulation. 2004;109:1594–1602.

32. Heymans S, Hirsch E, Anker SD, et al. Inflammation as a therapeutictarget in heart failure? A scientific statement from the TranslationalResearch Committee of the Heart Failure Association of the EuropeanSociety of Cardiology. Eur J Heart Fail. 2009;11:119–129.

33. De Ferrari GM, Schwartz PJ. Vagus nerve stimulation: from pre-clinical toclinical application: challenges and future directions. Heart Fail Rev.2011;16:195–203.

� 2011 Lippincott Williams & Wilkins www.jcvp.org | 507

J Cardiovasc Pharmacol � � Volume 58, Number 5, November 2011 VS Protects From Ischemia/Reperfusion