777

Vascular InjuryAugments Adrenergic Neurotransmission

Robert C. Candipan, MD, PhD; Paul T.C. Hsiun, BS;Richard Pratt, PhD; John P. Cooke, MD, PhD

Background We have observed persistent desensitizationto exogenous norepinephrine after balloon injury. We postu-lated that this desensitization may be due to a local increase inthe release of neuronal norepinephrine.Methods and Results New Zealand White rabbits under-

went left iliac artery angioplasty; 4 weeks later, both iliacarteries were harvested. Maximal response to exogenous nor-epinephrine was reduced in injured compared with noninjuredvessels (12.3+1.0 g versus 10.3±1.5 g; n=7, P=.056). Bycontrast, response to electrical stimulation (to induce neuronalnorepinephrine release) was significantly greater in injuredtissues (36±7% versus 14±3%; values expressed as percent ofmaximal contraction to exogenous norepinephrine; P=.025).Direct measurement of tissue norepinephrine revealed athreefold increase 4 weeks after injury (1236±410 versus466±97 pg/mg; injured versus noninjured). To determine ifdesensitization to exogenous norepinephrine was due to apersistent increase in neuronal norepinephrine release, the

W e and others have demonstrated that ballooninjury induces alterations in vascular reac-tivity that are secondary to changes in vas-

cular smooth muscle and endothelial function.1-4 Imme-diately after balloon injury, there is a marked reductionin the responsiveness to vasoconstrictor agents.2 Withinseveral days, the vascular smooth muscle recovers somecontractile function, and after 2 to 4 weeks, contractileresponses to potassium chloride have largely recovered.However, we have shown previously that an aberrationof adrenergic response persists up to 6 weeks afterballoon injury and is manifested by a reduction in thesensitivity to norepinephrine as well as a suppression ofthe maximal response.5A desensitization to exogenous catecholamines is

observed in tissues that are repeatedly stimulated.6-9We hypothesized that the desensitization after vascularinjury was due to a local and sustained increase inadrenergic neurotransmission in the area of injury. Wefurther postulated that this local facilitation of adren-ergic neurotransmission was due to activation of thetissue renin-angiotensin system. (It is known that afterballoon injury, there is increased expression of vascularangiotensinogen and angiotensin-converting enzyme[ACE]10-15; it is also known that angiotensin II [Ang II]facilitates the release of norepinephrine from adrener-gic nerve terminals.16-20) Accordingly, the present study

Received May 21, 1993; revision accepted November 6, 1993.From the Division of Cardiovascular Medicine, Stanford Uni-

versity School of Medicine, Stanford, Calif.Correspondence to John P. Cooke, MD, PhD, Falk Cardiovas-

cular Research Center, 300 Pasteur Dr, Stanford, CA 94305-5246.

experiments were repeated after chemical sympatholysis using6-hydroxydopamine (6-OHDA) (65 mg/kg). To determine ifactivation of vascular angiotensin II contributed to facilitationof adrenergic neurotransmission, other animals received rami-pril (RAM; 1 mg/kg per day). Both treatments were initiated7 days before angioplasty. In the 6-OHDA group there was noevidence of desensitization, judged by maximal response toexogenous norepinephrine (7.5_±0.6 versus 7.5+±0.8, nonin-jured versus injured). Similar results were obtained in RAManimals (9.9±0.8 versus 9.6±1.2, noninjured versus injured).

Conclusions This is the first study to demonstrate enhancedadrenergic neurotransmission after balloon injury. The facilita-tion of adrenergic neurotransmission may be due to increasedlocal concentrations of angiotensin II and is associated withdesensitization to exogenous norepinephrine. (Circulation.1994;89:777-784.)Key Words * restenosis * norepinephrine * angioplasty

*angiotensin

was designed to confirm our previous observations thatadrenergic response is downregulated after ballooninjury. Furthermore, we tested the hypothesis that thisabnormality was associated 'with an enhancement ofadrenergic neurotransmission. Finally, we sought todetermine if the persistent and local increase in cate-cholamine release may be due to activation of thevascular renin-angiotensin system.

MethodsAnimalsMale New Zealand White rabbits (weight, 2.5 to 3.0 kg)

were entered into the study after a 1-week period of acclima-tion in the housing facilities of the Stanford Department ofLaboratory Animal Medicine (DLAM), during which time theanimals were fed normal rabbit chow and received water adlibitum. All animals were inspected before the study by theDLAM veterinarian and monitored daily by DLAM techni-cians and the investigators. All experimental protocols wereapproved by the Administrative Panel on Laboratory AnimalCare of Stanford University and were performed in accor-dance with the recommendations of the American Associationfor the Accreditation of Laboratory Animal Care.

Before balloon angioplasty, animals were anesthetized usinga mixture of ketamine (5 mg/kg) and Rompun (35 mg/kg)given in an initial intravenous injection. The level of anesthesiawas continuously monitored during the procedure, and addi-tional anesthetic was given as needed. The left iliac artery wasused for balloon injury to allow the use of the contralateralartery as a noninjured control vessel.The left superficial femoral artery was exposed, isolated,

and the distal portion was ligated with a small surgical clip. A2F Fogarty arterial embolectomy catheter (Baxter HealthCare Co, Santa Ana, Calif) was inserted into the artery and

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

778 Circulation Vol 89, No 2 February 1994

advanced proximally into the iliac artery past the bifurcationinto the aorta. The balloon was inflated, then withdrawn whileinflated, then deflated. Three successive withdrawals with theinflated balloon were made. We have demonstrated previouslythat this degree of injury completely denudes the endotheliumand injures the underlying vascular smooth muscle.2' Theproximal portion of the superficial femoral artery was thenligated, and the site was checked for hemostasis. Animals wereallowed to recover from anesthesia before returning to theircages. Four weeks later, the animals were killed, and the iliacarteries were harvested for studies of vascular reactivity.

Vascular ReactivityTwenty-eight days after balloon injury, animals were given

an overdose of intravenous pentobarbital. Both iliac vesselswere removed and immediately placed into oxygenated phys-iological saline solution (PSS) that was composed of thefollowing (mmol/L): NaCl, 118.3; KCl, 4.7; CaCl2, 2.5; MgSO4,1.2; KH2P04, 1.2; NaHCO3, 25.0; EDTA, 0.026; glucose, 11.1.In some experiments, the KCl concentration was increased byan equimolar replacement of NaCl with KCl. Adherent fat andconnective tissue was removed, and the iliac arteries were cutinto rings of 3- to 4-mm width. Two pairs of rings (left andright iliac arteries) from the proximal and distal iliac arterywere mounted horizontally on stainless steel stirrups throughthe lumen and connected to force transducers. The vascularrings were suspended in the organ chambers filled withoxygenated PSS at 37°C. Over a period of 60 minutes, ringswere progressively stretched to the optimal point of theirlength-tension relation (determined previously to be 2 g).2'21To determine the response to neuronally released endoge-

nous norepinephrine, the iliac rings were placed between twoparallel platinum plate electrodes to stimulate adrenergicnerve endings, as described previously.22 Frequency-responsecurves to electrical stimuli were obtained using electric im-pulses (square waves with amplitude of 9 V, duration of 2milliseconds, and frequency of 0 to 16 Hz) provided by a DCamplifier triggered by a stimulator (Grass SD9). After electri-cal stimulation studies, the vessels were washed repeatedlywith fresh oxygenated PSS and allowed to return to theirresting tension. To determine the response of the vascularrings to exogenous norepinephrine, dose-response curves wereobtained by adding to the bath increasing concentrations ofnorepinephrine (in half-log increments from 10-9 to 10-4mol/L). After the maximal response to exogenous norepineph-rine had been obtained, the vascular rings were washedrepeatedly with fresh PSS for 60 minutes, by which time theyhad regained their resting tension. To determine the ability ofthe vascular smooth muscle to contract to a non-receptor-mediated agonist, the rings were exposed to increasing con-centrations of KCl (40 to 80 mmol/L) in the presence ofpropranolol (10-6 mol/L) and phentolamine (10-6 mol/L) toblock the action of any neuronally released norepinephrine.All drug dilutions were prepared in distilled water made freshthe day of the experiment and were stored on ice. Norepineph-rine bitartrate and 6-OHDA were purchased from the SigmaChemical Co (St Louis, Mo). Ramipril was a gift of the UpjohnCo (Kalamazoo, Mich).

Total Norepinephrine AssayThe initial vascular reactivity studies revealed that vascular

injury was associated with a reduced response to exogenousnorepinephrine. To test the hypothesis that vascular injuryinduces a persistent increase in local neuronal norepinephrine,tissue norepinephrine was measured. Iliac artery segmentswere homogenized in a glass Potter-Elvenhjem homogenizerwith 1 mL of 0.1 mol/L HClO4 at 0°C to 4°C. The suspensionwas then centrifuged at 16 OOOg in a microfuge for 5 minutes.The resulting supernatant was then filtered through a 0.2-mmfilter before analysis.

Norepinephrine content was determined using high-pres-sure liquid chromatography and electrochemical detectionusing an ESA coulochem 5100A detector equipped with a3-mm C-18 reverse phase column. The mobile phase consistedof 98% 0.5 mol/L NaHPO4. H20, pH 3.6, 1.2 mmol/L 1-octanesulfonic acid, 0.2 mmol/L EDTA, 2% methanol, and wascycled at a flow rate of 1.5 mL/min. Graded concentrations ofnorepinephrine bitartrate were used as standards, and 3,4-dihydrobenzylamine (DHBA) served as the internal standard.All standard solutions were prepared fresh the day of theanalysis. Retention times were 5.2 minutes for norepinephrineand 14.8 minutes for DHBA.

Pharmacological Regimens. The initial in vitro studies ofvascular reactivity revealed that 4 weeks after vascular injury,there was a desensitization to exogenous norepinephrine. Thisabnormality was associated with an increased response tostimulation of adrenergic nerve endings. Furthermore, themeasurements of tissue norepinephrine revealed an increase 4weeks after balloon injury. We therefore postulated that thedesensitization to exogenous norepinephrine was due to per-sistently increased local release of neuronal norepinephrineafter the vascular injury. To further test this hypothesis, werepeated the vascular injury in animals that were chemicallydenervated. We reasoned that if the desensitization to norepi-nephrine was due to a persistent increase in local catechol-amine release induced by the vascular injury, this effect couldbe prevented by sympatholysis before the vascular injury.Accordingly, some animals received 6-hydroxydopamine (6-OHDA) given as an initial intravenous bolus of 5 mg/kgdissolved in 15 mL of sterile water 1 week before ballooninjury. Because of the sympathetic discharge that occursduring the infusion, the animals were sedated with ketamineduring administration of 6-OHDA. The initial infusion wasfollowed 24 to 48 hours later by a subsequent intravenous doseof 30 mg/kg 6-OHDA. This dose of 6-OHDA has previouslybeen demonstrated to produce a chemical denervation of thesympathetic nervous system which persists for 7 to 10 days.23'24To maintain the sympathetic denervation throughout thecourse of the study, an additional dose (30 mg/kg) was given 7days after the balloon injury.The above studies suggested that adrenergic desensitization

4 weeks after vascular injury was due to persistent and localincreases in neurotransmitter release. A number of paracrinesubstances are known to facilitate adrenergic neurotransmis-sion. Ang II activates its corresponding receptor on theprejunctional nerve membrane to facilitate neuronal release ofnorepinephrine.16-20 It is also known that the vascular renin-angiotensin system is activated after balloon injury.10-15 Wereasoned that local elaboration of Ang II after balloon injurymight play a role in facilitating adrenergic neurotransmissionand thereby induce the desensitization to exogenous norepi-nephrine. Therefore, inhibition of converting enzyme activityshould interfere with Ang II facilitation of neurotransmissionand thereby restore the response to norepinephrine. Accord-ingly, ramipril was dissolved in the drinking water and given ina dose of 1 mg/kg per day. This dose was chosen because it didnot result in a significant drop in the blood pressure. Theanimals were allowed to drink water ad libitum. Water con-sumption was monitored daily, and the concentration oframipril was adjusted to administer the chosen dose. Ramipriltreatment was initiated 7 days before balloon injury and wascontinued for the duration of the experiment. Blood pressuremeasurements at baseline, before balloon injury, and at 4weeks after balloon injury were obtained. For these measure-ments, the central ear artery was cannulated, and the con-scious animal was allowed to rest quietly for at least 5 minutes;when the blood pressure had stabilized, the intra-arterialpressure was recorded.

ACE ActivityTo confirm that the oral administration of ramipril (1 mg/kg

per day) had a measurable effect, plasma ACE activity was

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

Candipan et al Adrenergic Neurotransmission After Balloon Injury 779

B

z

zLuI-

NUlIf EIIrtFMlN: (IVE) ELECTRICAL STIMULATION (Hz)FIG 1. A, Concentration-response curves to exogenous norepinephrine in rabbit iliac arteries exposed to balloon injury 4 weekspreviously in comparison with the contralateral noninjured iliac arteries. The injured tissues contract to norepinephrine to a lesserdegree, with a maximal response that is different from the uninjured tissues (12.3±1.0 vs 10.3±1.5; P=.056). B, Frequency-responsecurves to electrical stimulation of injured and uninjured iliac arteries. By contrast with their response to exogenous norepinephrine, theresponse to neuronal stimulation is qualitatively different, as the injured tissues appear to be more responsive.

determined. Blood samples were drawn directly from the heartafter the animal was killed into heparinized collecting tubes.The plasma was centrifuged and stored at -70°C until ana-lyzed for ACE activity.ACE activity was measured by the inhibitor binding tech-

nique as described by Cheung and Cushman.25 Fifteen millili-ters of plasma was aliquoted into 135 mL of distilled water.One hundred milliliters of the substrate hippuryl-L-histidyl-L-leucine (Hip-His-Leu) was added to each sample then incu-bated for 30 minutes at 37°C. The reaction was stopped with1.45 mL of 0.28N NaOH at room temperature. A graded seriesof concentrations of histidyl-L-leucine (H-L) were used togenerate a standard curve. To standards and all plasmasamples, 100 ,L O-phtaldialdehyde was added. The solutionwas then vortexed, incubated for 10 minutes at room temper-ature, and then covered with foil. The reaction was stoppedwith 200 ,L of 3N HCl and incubated for 30 minutes at roomtemperature, then covered with foil. All samples and standardswere centrifuged for 5 minutes at 2000g, and the supernatantwas analyzed by spectrofluorimetry (excitation wavelength of364 nm and emission wavelength at 486 nm). The ACE activityis expressed as nmol* min'* mL`.

Histological StudiesOn completion of the vascular reactivity studies, the iliac

rings were fixed in 10% buffered formalin, embedded inparaffin, sectioned, and stained with an elastic van Giesonstain for light microscopy and histomorphometric measure-ments. Measurements of intimal and medial cross-sectionalareas (millimeters squared) were made by a skilled observerblinded to the treatment groups. An eyepiece grid with 100counting points (line intersections), each 0.4 mm apart, wasused.2627 Total grid area (millimeters squared) was deter-mined by measurement of grid dimensions with a stage micro-meter at the same microscope magnification used for pointcounts. Total area of the intima (or media) was defined as thenumber of grid points overlying the intima (or media). Area ofeach was calculated by the formula 0.01 (P) xgrid area, whereP is the number of points counted over the intima (or media).At least five cross sections from each vascular segment wereanalyzed, and the values were averaged.One segment from a control animal (injured), which con-

tained thrombus that made assessment of intimal thickeningdifficult, was excluded from analysis.

Data AnalysisData are expressed as mean+SEM. Dose-response curves

to norepinephrine and frequency-response curves to electrical

stimuli are expressed as contractions in grams above theresting tension. The dose and frequency response curves arecharacterized by determining the maximal response to thedrug or stimulus. Statistical analysis to detect significantdifferences in vascular reactivity between injured and nonin-jured segments within the same experimental group was byStudent's t test for paired observations (two-tailed). Compar-isons between the three experimental groups were made byANOVA.

ResultsVascular Reactivity StudiesResponse to Exogenous and NeuronallyReleased Norepinephrine

Norepinephrine induced concentration-dependentincreases in tension in both the injured and uninjurediliac arteries. The maximal response to exogenous nor-epinephrine was reduced in the injured vessels(12.3± 1.0 versus 10.3+1 5; P=.056; Fig 1A).By contrast, the response to neuronal stimulation was

reversed. Electrical stimulation induced a frequency-dependent contraction caused by neuronally releasednorepinephrine, which tended to be greater in theinjured tissues (Fig 1B). When the response to neuron-ally released norepinephrine was expressed as a per-centage of the maximal response to exogenous norepi-nephrine, the differences between the injured andnoninjured tissues were even more apparent (Fig 2).Expressed in this way, the maximal response to electri-cal stimulation in the injured tissues was significantlygreater than that in the noninjured segments (36±7%versus 14±3%; n=7 in both groups; P=.025).

In contrast to the adrenergic response, there were nodifferences between the two groups in the responsive-ness to potassium chloride, a nonadrenergic agonist ofvascular smooth muscle contraction (Table 1).

Tissue NorepinephrineThe initial vascular reactivity studies suggested that

the desensitization to exogenous norepinephrine aftervascular injury was due to an increase in local release ofneuronal norepinephrine. To confirm this hypothesis,tissue norepinephrine was directly measured in injured

A

z1

0

zLuI-

1x10-9 1x104 1x10-7 1x104 1x10 5 1x10-4rIwOnlni=nunLnK fan%

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

780 Circulation Vol 89, No 2 February 1994

z

0

Ec-z0c)

z

X

0-+- INJURED (N=7)

-e- NON4NJURED (N=7)

5 10 15

ELECTRICAL STIMULATION (Hz)20

FIG 2. Frequency-response curves to electrical stimulation ex-pressed as a percentage of maximal response of each tissue toexogenous norepinephrine (NE). When the response of eachvascular segment to neuronal stimulation is expressed as apercentage of its capacity to respond to norepinephrine, thedifference between the two groups becomes more apparent.Because the injured tissues are desensitized to norepinephrine,the response to neuronal norepinephrine also should be re-duced. Paradoxically, however, the previously injured tissueshave a markedly increased responsiveness to neuronal stimula-tion (37±7% vs 14±3%; n=7 in each group; P=.025).

and noninjured iliac arteries 4 weeks after balloonangioplasty. In the uninjured iliac artery, norepineph-rine concentration was 466+97 pg/mg (n= 12). By con-

trast, there was a threefold elevation in tissue norepi-nephrine in the injured artery to 1236±410 pg/mg(n= 12, P<.05).

Effect of Chemical SympathectomyTo determine if the reduced responsiveness to exog-

enous norepinephrine in the injured vessels was due tolocal increases in neuronally released norepinephrine,the influence of neuronal norepinephrine was removedby chemical sympathectomy using 6-OHDA. In vascularsegments from the 6-OHDA-treated animals, the re-

sponse to electrical stimulation was essentially abol-ished in both the injured and noninjured vascularsegments. In the noninjured vascular segments, themaximal response to electrical stimulation was only0.16±0.08 g (2.1% of the maximal response to exoge-nous norepinephrine in these tissues). In the injuredsegments, the maximal response to electrical stimula-tion was only 0.04±0.04 g (0.5% of the maximal re-sponse to norepinephrine in these tissues; Table 2).These studies confirmed that the 6-OHDA treatmenthad induced a successful chemical denervation of thesympathetic nerve endings in the vessel wall.

In the 6-OHDA-treated animals, contractions toexogenous norepinephrine induced concentration-de-pendent increases in tension in both the injured andnoninjured tissues; the dose-response curves were iden-tical with maximum contractions that were not differentbetween the two groups (7.47±0.65 g versus 7.54±0.82g, noninjured versus injured; n=5 in each group; P=NS;Fig 3).

Role ofAng II in Adrenergic AlterationTo determine if an activation of the vascular renin-

angiotensin system in the vascular wall may be involvedin the alteration of adrenergic response after ballooninjury, one group of animals received ramipril beforeand after balloon injury until the vessels were harvestedfor study. The administration of ramipril reducedplasma ACE activity by 73% (from 76.6±5.1 to21.3±13.9 nmol* min-1 mL-1; vehicle [n=5] versusramipril [n=3], respectively). In vascular segments fromramipril-treated animals, exogenous norepinephrine in-duced concentration-dependent increases in tension inboth injured and noninjured segments that were notdifferent (maximal response, 9.99±0.86 g versus9.59±1.2 g, noninjured versus injured; n=7 in eachgroup; P=NS; Fig 4). Responses to electrical stimula-tion were not different between the injured and nonin-jured tissues (Table 2). Likewise, contractions to potas-sium chloride were not different between the twogroups (Table 1).

Histomorphometric AnalysisIn all experimental groups, the noninjured iliac arter-

ies had no intimal thickening with intimal/medial ratiosthat approached zero. In all three groups, the balloon-injured arteries showed varying degrees of intimal thick-ening (Table 3). The intimal/medial ratio of injuredsegments from animals treated with ramipril was signif-icantly reduced in comparison to 6-OHDA animals(0.97±0.15 versus 0.39+0.16, 6-OHDA treatment ver-sus ramipril treatment; P=.03). This difference waslargely due to a reduction in medial thickness in the6-OHDA-treated group in comparison with the rami-pril-treated animals (Table 3). There was no significantdifference in intimal/medial ratio between the control(0.64±0.12) and either of the experimental groups.

DiscussionThis is the first study to demonstrate a facilitation of

adrenergic neurotransmission after balloon injury. Thisenhancement of adrenergic neurotransmission is asso-ciated with a desensitization to exogenous norepineph-rine. Our data also suggest that local activation of thevascular renin-angiotensin system may play a role in the

TABLE 1. Contractions to KCI in Injured and Noninjured lliac Arteries

Control n=7 6-OHDA n=5 Ramipril n=5

KCI, mmol/L Noninjured Injured Noninjured Injured Noninjured Injured40 5.4±0.9 4.7±1.0 4.1+0.5 3.3+0.7 3.2+0.2 2.1+0.360 5.6±0.9 5.2+1.0 4.4±0.6 3.5+0.7 3.4±0.2 2.2±0.280 5.4±0.8 4.9±0.9 4.2±0.7 3.2+0.8 2.9±0.3 2.0±0.36-OHDA indicates 6-hydroxydopamine. Data are expressed as grams above resting tension and are shown as

mean+SEM.

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

Candipan et al Adrenergic Neurotransmission After Balloon Injury 781

TABLE 2. Effect of Balloon Injury on Response to Electrical Stimulation in AnimalsTreated With 6-OHDA or Ramipril

Electrical 6-OHDA (n=5) Ramipril (n=7)Stimulation,Hz Noninjured Injured Noninjured Injured2 0.05±0.01 0.01±0.01 0.12±0.02 0.31 ±0.104 0.08±0.03 0.03±0.01 0.24±0.04 0.73±0.27

8 0.11±0.05 0.03±0.01 0.35±0.07 1.00±0.36

16 0.15±0.08 0.04±0.02 0.52±0.13 1.33±0.44

6-OHDA indicates 6-hydroxydopamine. Data are expressed as grams above resting tension and areshown as mean±SEM.

enhancement of adrenergic neurotransmission. Previ-ous studies have examined the role of endothelial andsmooth muscle cell dysfunction after balloon injury. Thecurrent investigation suggests a new avenue for explo-ration in understanding the pathophysiological mecha-nisms of vascular injury.We have observed previously that balloon injury

induces a persistent and progressive adrenergic dysfunc-tion in the rat aorta.5 This was manifested by reducedcontraction to exogenous norepinephrine of vascularrings harvested from injured rat aortas. This abnormal-ity persisted for up to 6 weeks after balloon injury. Thisabnormality was not due to a generalized depression ofvascular smooth muscle contractile function, becausethe sensitivity to potassium chloride was restored at atime when response to norepinephrine was depressed.5The major hypothesis of the current investigation wasthat the desensitization to exogenous norepinephrinewas secondary to local and sustained increases in neu-ronal release of norepinephrine. A desensitization toexogenous catecholamines can be seen in tissues thatare repeatedly stimulated.6 Desensitization of adrener-gic response may occur through several mechanismsincluding phosphorylation of the adrenergic receptor,sequestration and degradation of the receptor, and/or a

8-

z0

4

zLZ. | INJURED (N=5)

2- I /1 NON-INJURED (N=5)

1x1P 1xl108 1xlO-7 1x104 1x10-5 1x10 4NOREPINEPHRINE (M)

FIG 3. Dose-response curves to exogenous norepinephrine ininjured and noninjured iliac arteries from animals treated with6-hydroxydopamine (6-OHDA). This sympatholytic agent abol-ished neuronal release of norepinephrine in these vascularsegments, as determined by the response to electrical stimula-tion (Table 2). In the absence of neuronal influence, there is nodifference between the injured and noninjured vascular seg-ments. This is consistent with the hypothesis that the desensiti-zation to exogenous norepinephrine in injured tissues (Fig 1) isdue to chronic increases in the local release of catecholaminesin the injured segments.

reduction in its synthesis.6-9 We hypothesized that oneor more of these processes may be initiated afterballoon injury caused by a local and sustained increasein neurotransmission in the area of injury.

In the present study, we confirmed our previousobservations (using a different animal model) thatvascular injury results in a persistent desensitization toexogenous norepinephrine. Four weeks after ballooninjury, the response of the injured segment to norepi-nephrine was reduced in comparison to the uninjuredvascular segment (Fig 1A). By contrast, the response topotassium chloride was not altered 4 weeks after bal-loon injury. Potassium chloride induces a contraction ofthe vascular smooth muscle that is due to depolarizationand influx of extracellular calcium; this contraction isindependent of adrenoceptor function.To stimulate the neuronal release of norepinephrine,

we used well-established parameters of electrical stim-ulation.22 We were surprised by the marked enhance-ment of response to neuronal stimulation in the injuredtissues (Fig 1B). When the response to electrical stim-ulation was expressed as a percent of maximal responseof the tissues to exogenous norepinephrine (to take intoaccount the desensitization of the vascular smoothmuscle to norepinephrine), the maximal response inthe injured tissues was greater than twice that of theuninjured vessels (Fig 2). These data suggest that the

10

Z 6-0

Z 4I- N O N4INJURED(N=7)|

2- [ NIR E DN=J

1x109 1X1X8 1x10-7 1X10e lx105 1x104

NOREPINEPHRINE (M)FIG 4. Dose-response curves to exogenous norepinephrine ininjured and noninjured iliac arteries harvested from animalstreated with ramipril. The responsiveness to norepinephrine inthe injured tissues has been restored by the converting enzymeinhibitor ramipril. This is consistent with the hypothesis thatactivation of the renin-angiotensin system after vascular injuryplays a role in the increased neuronal release of endogenousnorepinephrine and subsequent desensitization to the neuro-transmitter.

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

782 Circulation Vol 89, No 2 February 1994

TABLE 3. Histomorphometric Analysis

Vehicle Group (n=7) Ramipril Group (n=8) 6-OHDA Group (n=5)

Measurement Injured Noninjured Injured Noninjured Injured NoninjuredIntimal area, mm2 0.13±0.03 0.0 0.08±0.03 0.0 0.15±0.02 0

Medial area, mm2 0.19±0.02 0.22±0.07 0.23±0.02 0.24±0.03 0.16+0.01* 0.21±0.06I/M 0.64±0.12 0.0 0.30±0.16 0.0 0.97±0.15* 0

6-OHDA indicates 6-hydroxydopamine; l/M, intimal/medial ratio.*Significantly different from value in ramipril group by ANOVA (P<.05).

neuronal release of norepinephrine is facilitated afterballoon injury. Our central hypothesis was furtherstrengthened by direct measurement of tissue norepi-nephrine levels. Four weeks after vascular injury, weobserved a threefold increase in the concentration oftissue norepinephrine. The increase in local catechol-amine levels could be explained by a number of mech-anisms including an increase in the number of presyn-aptic nerve terminals, an enhanced synthesis andrelease of neurotransmitter from individual nerve ter-minals, a reduction in the uptake or metabolism oflocally released neurotransmitter, or from increaseduptake of circulating norepinephrine.

If enhanced local release of norepinephrine is re-sponsible for the desensitization to exogenous norepi-nephrine, then removal of this neuronal influenceshould restore the responsiveness to exogenous norepi-nephrine. To test this hypothesis, we induced a chemicalsympatholysis by administering 6-OHDA to some ani-mals. The efficacy of the denervation was established bystudying the response of the vascular rings to electricalstimulation, which was essentially abolished. In theabsence of neuronally released norepinephrine, wefound that the response to exogenous norepinephrine ofthe injured and noninjured vessels was not different (Fig3). Generally, 6-OHDA treatment induces a supersen-sitivity to exogenous norepinephrine. We did not notean increased sensitivity to norepinephrine in the iliacarteries of chemically denervated rabbits. This probablyis due to a countervailing effect on vascular smoothmuscle reactivity, because the contractions to potassiumchloride were also depressed in this group (Table 1).The direct effect of 6-OHDA treatment to reduce iliacartery contractility may be in part due to an effect onvascular smooth muscle growth, because the medialthickness was reduced in the injured vessels of theseanimals (Table 3). Our findings are similar to those ofFronek,'28 who observed that chemical denervation with6-OHDA was associated with increased sensitivity toexogenous norepinephrine in most vascular beds of therabbit but not the auricular or iliac arteries. Neverthe-less, in our study, adrenergic denervation was associ-ated with an equalization of the response to exogenousnorepinephrine in the injured and uninjured vessels.This is consistent with our central hypothesis thatvascular injury induces a local increase in tissue cate-cholamines that is manifested by a desensitization toexogenous norepinephrine.The mechanism by which adrenergic neurotransmis-

sion is facilitated after balloon injury may involve thevascular renin-angiotensin system. Ang II is known tofacilitate the release of norepinephrine from adrenergicnerve endings. In the isolated portal vein and pulmo-

nary artery of the rabbit, Ang II augments the release oftritiated norepinephrine evoked by electrical stimula-tion.16'7 Ang II enhances the appearance of adrenergicneurotransmitters in the venous effluent from canineskin and kidney during stimulation of the sympatheticnerves.18'19 This probably is due to direct interaction ofAng II with its corresponding receptor on the prejunc-tional adrenergic nerve terminal, the existence of whichhas been previously demonstrated by autoradiographicstudies.20 Finally, it is known that after balloon injury,there is increased expression of vascular angiotensino-gen and ACE.10-" Evidence suggests that this activationof vascular renin-angiotensin contributes to the alter-ations in vascular structure after balloon injury.13-'5

If the desensitization to exogenous norepinephrine isdue to Ang II-facilitated release of neuronal norepi-nephrine, inhibitors of Ang II synthesis should restorethe response to exogenous norepinephrine. Indeed, inthe ramipril-treated animals, the response to exogenousnorepinephrine is not different in the injured segments(Fig 4). This finding is consistent with the hypothesisthat injury-induced activation of the vascular renin-angiotensin system leads to higher tissue levels of AngII, which facilitates local norepinephrine release. How-ever, we do not exclude other effects of ramipril (ie, onlocal kinin concentrations) that could be responsible forthe observed effect. To summarize, the present investi-gation confirms our previous observation that ballooninjury induces a persistent desensitization to exogenousnorepinephrine. This aberration of adrenergic responseappears to be due to a local facilitation of neuronallyreleased neurotransmitter. Injury-induced activation ofthe vascular renin-angiotensin system may play a role inthe facilitated adrenergic neurotransmission.

In light of these observations, the importance ofneurotransmitters in the pathophysiology of restenosisafter balloon angioplasty must be considered. Catechol-amines have long been known to stimulate smoothmuscle cell growth. In vitro studies have revealed thatcatecholamines promote growth of cultured smoothmuscle cells in association with increased expression ofthe proto-oncogene c-myc.29-31 In vivo studies suggest arole for local catecholamines in vascular growth. Be-van3233 surgically denervated the central ear artery ofthe rabbit and observed a reduction in medial growthmanifested by reduced wall thickness and protein con-tent.3233 Hart et a134 found that the characteristic in-crease in vessel wall to lumen ratio that is observed inhypertensive arteries does not occur after sympathec-tomy. An infusion of Ang II increases DNA synthesis invascular smooth muscle of the rat thoracic aorta; thiseffect is antagonized by the a,-adrenergic antagonistprazosin.35 The mitogenic effects of catecholamines on

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

Candipan et al Adrenergic Neurotransmission After Balloon Injury 783

the vessel wall may be due to regulation of growth-related gene expression by the a1-adrenergic receptor.36To summarize, the evidence is strong that endoge-

nous catecholamines modulate the growth of medialsmooth muscle cells. It remains to be establishedwhether or not endogenous catecholamines promotethe growth of neointimal cells after vascular injury.However, Fingerle and colleagues37 have studied theeffects of two adrenergic antagonists, prazosin andurapidil, on myointimal hyperplasia after balloon injuryin rat carotid arteries. They observed a reduction in thesize of the neointima in urapidil-treated animals ascompared with control animals and reduced neointimaformation in both prazosin-treated and urapidil-treatedanimals when intimal DNA content was measured.Their data suggest that antagonism of the adrenergicsystem can inhibit myointimal hyperplasia. In support oftheir findings, we found that inhibition of convertingenzyme appeared to restore normal adrenergic neuro-transmission and that this effect was associated with atrend toward reduced myointimal hyperplasia, an obser-vation previously reported by others.10,38 By contrast, wedid not observe a decrease in the degree of myointimalhyperplasia in the injured vessels of our chemicallydenervated animals. This result could be explained by adirect effect of 6-OHDA on vascular smooth muscle cellgrowth.We and others have shown that the endothelium

regenerating after balloon injury has altered function,manifested by reduced nitric oxide-dependent vasodi-lation.1,2,21 Exogenous nitric oxide can inhibit the neu-ronal release of norepinephrine39; therefore, it is possi-ble that local catecholamine release may also befacilitated in the setting of balloon injury, where thesynthesis and/or degradation of endogenous nitric oxidemay be altered.

SummaryThe present investigation provides the first evidence

for a facilitation of adrenergic neurotransmission afterballoon injury. This effect is associated with a desensi-tization of the vascular smooth muscle to norepineph-rine. An injury-induced activation of the vascular renin-angiotensin system may be responsible for thefacilitated release of adrenergic neurotransmitter.

AcknowledgmentsThis work was supported in part by grants from the National

Heart, Lung, and Blood Institute (1PO1-HLA8638-01), the In-stitute of Biological and Clinical Investigation of Stanford Uni-versity, the University of California Tobacco-Related DiseaseProgram (1RT215), and Upjohn Pharmaceuticals. Dr Candipanis the recipient of a National Research Service Award(1F32HL08779). Mr Hsiun is supported by a Medical ScholarsGrant from Stanford University. Dr Cooke is the recipient of aVascular Academic Award from the NHLBI (1K07HCO2660-2).The authors thank Joan Rosel and Cecily Denson for their expertsecretarial and administrative assistance.

References1. Shimokawa H, Flavahan NA, Vanhoutte PM. Natural course of the

impairment of endothelium-dependent relaxations after balloonendothelium removal in porcine coronary arteries: possible dys-function of a pertussis toxin-sensitive G protein. Circ Res. 1989;65:740-753.

2. Weidinger FF, McLenachan JM, Cybulsky MI, Gordon JB,Rennke HG, Hollenberg NK, Fallon JT, Ganz P, Cooke JP. Per-

sistent dysfunction of regenerated endothelium after balloonangioplasty of rabbit iliac artery. Circulation. 1990;81:1667-1679.

3. Shimokawa H, Tomoike H, Nabeyama S, Yamamoto H, Araki H,Nakamura M, Ishii Y, Tanaka K. Coronary artery spasm inducedin atherosclerotic miniature swine. Science. 1983;221:560-562.

4. Egashira K, Tomoike H, Yamamoto Y, Yamada A, Hayashi Y,Nakamura M. Histamine-induced coronary spasm in regions ofintimal thickening in miniature pigs: roles of serum cholesterol andspontaneous or induced intimal thickening. Circulation. 1986;74:826-837.

5. Cooke JP, Dasari P, Krieger JE, Rakugi H, Dzau VJ, Pratt RE.Persistent endothelial and adrenergic dysfunction after angio-plasty. JAm Coll Cardiol. 1991;17:59A. Abstract.

6. Caron MG, Lefkowitz RJ. Catecholamine receptors: structure,function, and regulation. In: Recent Progress in Hormone Research.Vol 48. New York: Academic Press; 1993:277-290.

7. Bouvier M, Collins S, O'Dowd BF, Campbell PT, de Blasi A,Kobilka BK, MacGregor C, Irons GP, Caron MG, Lefkowitz RJ.Two distinct pathways for cAMP-mediated down-regulation of theA-adrenergic receptor. J Biol Chem. 1989;264:16786-16792.

8. Benovic JL, Bouvier M, Caron MG, Lefkowitz RJ. Regulation ofadenylyl cyclase-coupled 3-adrenergic receptors. Annu Rev CellBiol. 1988;4:405-428.

9. Oron Y, Straub RE, Traktman P, Gershengorn MD. DecreasedTRH receptor mRNA activity precedes homologous down-regulation: assay in oocytes. Science. 1987;238:1406-1408.

10. Powell JS, Clozel JP, Muller RKM, Kuhn H, Hefti F, Hosang M,Baumgartner HR. Inhibitors of angiotensin-converting enzymeprevent proliferation injury after vascular injury. Science. 1989;245:186-188.

11. Rakugi H, Krieger J, Wang DS, Dzau VJ, Pratt RE. Induction ofangiotensin converting enzyme in neointima after balloon injury.Circulation. 1991;84(suppl II):II-113. Abstract.

12. Rakugi H, Jacob HJ, Krieger JE, Ingelfinger JR, Pratt RE.Vascular injury induces angiotensinogen gene expression in themedia and neointima. Circulation. 1993;87:283-290.

13. Osterrieder W, Muller RKM, Powell JS, Clozel JP, Hefti F, Baum-gartner HR. Role of angiotensin II in injury-induced neointimaformation in rats. Hypertension. 1991;18(suppl II):II-60-II-64.

14. Daemen MJAP, Lombardi DM, Bosman FT, Schwartz SM. Angio-tensin II induces smooth muscle cell proliferation in the normaland injured rat arterial wall. Circ Res. 1991;68:450-456.

15. Bell L, Madri JA. Influence of the angiotensin system on endo-thelial and smooth muscle cell migration. Am J Pathol. 1990;137:7-12.

16. Endo T, Starke K, Bangerte A, Taube HD. Presynaptic receptorsystems on the noradrenergic neurones of the rabbit pulmonaryartery. Naunyn Schmiedebergs Arch Pharmacol. 1977;296:229-247.

17. Hughes J, Roth RH. Evidence that angiotensin enhances transmitterrelease during sympathetic nerve stimulation. Br J Pharmacol. 1971;41:239-255.

18. Zimmerman BG, Gisslen J. Pattern of renal vasoconstriction andtransmitter release during sympathetic stimulation in presence ofangiotensin and cocaine. JPharmacol Exp Ther. 1968;163:320-329.

19. Zimmerman BG, Whitmore L. Effect of angiotensin and phenoxy-benzamine on release of norepinephrine in vessels during sympa-thetic nerve stimulation. Int J Neuropharmacol. 1967;6:27-38.

20. Urata H, Healy B, Stewart RW, Bumpus FM, Husain A. Angio-tensin II receptors in normal and failing human hearts. J ClinEndocrinol Metab. 1989;69:54-66.

21. Weidinger FF, McLenachan JM, Cybulsky MI, Fallon JT, HKol-lenberg NK, Cooke JP, Ganz P. Hypercholesterolemia enhancesmacrophage recruitment and dysfunction of regenerated endo-thelium after balloon injury of the rabbit iliac artery. Circulation.1991;84:755-767.

22. Cooke JP, Shepherd JT, Vanhoutte PM. The effect of warming onadrenergic neurotransmission in canine cutaneous vein. Circ Res.1984;54:547-553.

23. Aprigliano 0, Hermsmeyer K. In vitro denervation of the portalvein and caudal artery of the rat. JPharmacol Exp Ther. 1976;198:568-577.

24. Thoenen H, Tranzer JP. The pharmacology of 6-hydroxydopa-mine. Annu Rev Pharmacol. 1973;13:169-180.

25. Cheung HS, Cushman DW. Inhibition of homogeneous angiotensin-converting enzyme of rabbit lung by synthetic venom peptides ofBothrop's jararaca. Biochim Biophys Acta. 1973;293:451-463.

26. Aherne MA, Dunnill MS. Morphometry. Sevenoaks, Kent,England: Edward Arnold; 1992.

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

784 Circulation Vol 89, No 2 February 1994

27. Manwaring L, O'Connell DL, Bhagwandeen BS, Zardawi IM,Dobson AJ. Morphometric analysis of coronary artery stenosis: an

accuracy and reliability study. J Pathol. 1988;156:111-117.28. Fronek K. Evaluation of long-term chemical sympathectomy in

adult rabbits. Am J Physiol. 1980;238(Heart Circ Physiol7):H527-H532.

29. Blaes N, Boissel JP. Growth-stimulating effect of catecholamineson rat aortic smooth muscle cells in culture. J Cell Physiol. 1983;116:167-172.

30. Chamley JH, Campbell GR. Trophic influences of sympatheticnerves and cyclic AMP on differentiation and proliferation ofisolated smooth muscle cells in culture. Cell Tissue Res. 1975;161:497-510.

31. Starksen NF, Simpson PC, Bishopric N, Coughlin SR, Lee WMF,Escobedo JA, Williams LT. Cardiac myocyte hypertrophy is asso-

ciated with c-myc proto-oncogene expression. Proc Natl Acad SciUSA. 1986;83:8348-8350.

32. Bevan RD. Effect of sympathetic denervation on smooth muscle cellproliferation in growing rabbit ear artery. Circ Res. 1975;37:14-19.

33. Bevan RD. Trophic effects of peripheral adrenergic nerves on

vascular structure. Hypertension. 1984;6(suppl III):III-19-III-26.

34. Hart MN, Heistad DD, Brody MJ. Effect of chronic hypertensionand sympathetic denervation on wall/lumen ratio of cerebralvessels. Hypertension. 1980;2:419-423.

35. van Kleef EM, Smits JFM, De May JGR, Cleutjens JPM,Lombardi DM, Schwartz SM, Daemen MJAP. a1-Adrenoceptorblockade reduces the angiotensin II-induced vascular smoothmuscle cell DNA synthesis in the rat thoracic aorta and carotidartery. Circ Res. 1992;70:1122-1127.

36. Majesky MW, Daemen MJ, Schwartz SM. Alpha 1-adrenergicstimulation of platelet-derived growth factor A-chain geneexpression in rat aorta. J Biol Chem. 1980;265:1082-1088.

37. Fingerle J, Sanders KH, Fotev Z. Alpha1-receptor antagonistsurapidil and prazosin inhibit neo-intima formation in rat carotidartery induced by balloon catheter injury. Basic Res Cardiol. 1991;86(suppl I):75-81.

38. Capron L, Heudes D, Chajara A, Bruneval P. Effect of ramipril, aninhibitor of angiotensin converting enzyme, on the response of ratthoracic aorta to injury with a balloon catheter. J CardiovascPharmacol. 1991;18:207-211.

39. Tesfamarian B, Cohen RA. Inhibition of adrenergic vasocon-striction by endothelial cell shear stress. Circ Res. 1988;63:720-725.

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from

R C Candipan, P T Hsiun, R Pratt and J P CookeVascular injury augments adrenergic neurotransmission.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1994 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.89.2.777

1994;89:777-784Circulation. 

http://circ.ahajournals.org/content/89/2/777World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 25, 2017http://circ.ahajournals.org/

Dow

nloaded from