a nerves in - heart

Br Heart J 1989;62:353-60

A study of nerves containing peptides in thepulmonary vasculature of healthy infants andchildren and of those with pulmonary hypertensionKIM M ALLEN, JOHN WHARTON,* JULIA M POLAK,* SHEILA G HAWORTH

From the Institute of Child Health and *Royal Postgraduate Medical School, London

SUMMARY Nerves containing peptides that supply the human intrapulmonary vasculature werestudied in 21 controls aged one month to 24 years and in 13 patients with pulmonary hypertensionaged 11 days to eight years. An indirect immunofluorescence technique was used to study thedistribution and relative density of nerve fibres containing the general neuronal marker, proteingene product 9 5; tyrosine hydroxylase; synaptophysin; neuropeptide tyrosine; vasoactiveintestinal polypeptide; substance P, somatostatin; and calcitonin gene related peptide. At all ages innormal and hypertensive lungs neuropeptide tyrosine was the predominant neuropeptideassociated with the pulmonary vascular nerves. In normal lungs the relative density ofnerve fibresincreased during childhood only in the arteries of the respiratory unit. Pulmonary hypertensionwas associated with the premature innervation of these arteries during the first year of life.

Innervation of small, abnormally thick-walled pre-capillary vessels by predominantly vasocon-strictor nerves may help to explain the susceptibility of infants to pulmonary hypertensive crises.

There is no reliable effective medical treatment forpulmonary hypertension. Children with potentiallyreversible pulmonary vascular disease can die after atechnically successful intracardiac repair because ofthe development of pulmonary hypertensive crisesthat become intractable to all forms of treatment.Pharmacolological management of pulmonaryhypertension might be improved by understandinghow the pulmonary vasculature is innervated in sucha lung.During normal development the wall structure of

the entire pulmonary arterial tree is remodelled.Connective tissue is deposited and medial smoothmuscle cells differentiate, acquiring a contractilerather than a synthetic phenotype.' The concentra-tion of contractile myofilaments increases and thetype of cytoskeletal proteins change.2 In childrenwith pulmonary hypertensive congenital heart dis-ease these changes are accelerated or abnormal.3 Therelation between remodelling of the pulmonaryarteries and pulmonary innervation in the humanlung is unclear. In the normal immature pig lungdifferentiation ofsmooth muscle cells is accompaniedRequests for reprints to Professor Sheila G Haworth, Institute ofChild Health, 30 Guilford Street, London WC1N 1EH.

Accepted for publication 27 July 1989

by an increase in the density of presumed sympath-etic nerves.4 Experimental studies showed trophicinteractions between sympathetic nerves and vas-cular smooth muscle cells.'The adult human pulmonary vasculature is

innervated by noradrenergic and cholinergicnerves67 but these nerves also contain other putativetransmitters, including several vasoactive peptides,which may have a regulatory role in the pulmonarycirculation. Examples of such peptides identified inmammalian pulmonary nerves include vasoactiveintestinal polypeptide8 (predominantly in parasym-pathetic nerve fibres), substance P and the neuro-kinins9 and calcitonin gene related peptide'0 (allpresent in sensory nerve fibres), and neuropeptidetyrosine" (in sympathetic fibres) as recently re-viewed by Lundberg et al."We used immunocytochemical techniques to

examine the innervation of the pulmonary vas-culature in the normal and pulmonary hypertensivehuman lung. The overall pattern of innervation wasstudied with an antiserum to general neuronal cyto-plasmic marker, protein gene product 9-5. Specificsubpopulations of nerve fibres were identified byantibodies to neuropeptides and to the enzymetyrosine hydroxylase which synthesises cate-cholamines.

353

on February 4, 2022 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.62.5.353 on 1 Novem

ber 1989. Dow

nloaded from

http://heart.bmj.com/

354

Patients and methods

The innervation of the intrapulmonary arteries andveins and the bronchial arteries was examined atnecropsy in 18 children and three adults withoutcardiopulmonary disease and in 13 children who hadpulmonary hypertension (table 1). In 11 children,pulmonary hypertension was caused by congenitalheart disease and two had primary idiopathic pul-monary hypertension of the newbom. Lung tissuewas obtained less than 24 hours after death. Blocks oftissue (1 cm2 x 0-5 cm) were dissected out from thehilum to the periphery to include airways and theirassociated arteries so that elastic, large and smallmuscular arteries were samples in addition to thealveolar region of the lung. Tissue blocks were fixedovernight (9-12 hours) at 4°C in a modified Bouin'ssolution.4 After the tissue was rinsed in severalchanges of phosphate buffered saline (0-1 mol/l,pH 7-2) containing 15% (w/v) sucrose and 0-1%sodium azide, cryostat blocks were prepared andstored in liquid nitrogen. Transverse cryostat sec-

tions (15 um thick) were collected on glass slidescoated with poly-L-lysine and allowed to dry for anhour at room temperature. Before immunostainingby the indirect immunofluorescence technique, sec-

tions were immersed in phosphate b.uffered salinecontaining 0.2% Triton X-100 for an hour at roomtemperature, rinsed in phosphate buffered saline,then counterstained with pontamine sky blue (BDHChemicals) to reduce background autofluorescence.The innervation was shown with primary anti-

sera"3 to general neuronal markers and specificneuropeptides (table 2). The general neuronal cyto-plasmic marker protein gene product 9-5 occurs in allperipheral nerves and in all regions of the neurone. 4

Synaptophysin is a membrane protein specificallyassociated with small secretory vesicles thought tocontain the classic neurotransmitters in nerve ter-minals." Antisera to both the sequences flanking theactive and carboxy terminals of the neuropeptidetyrosine precursor molecule were used and have beenshown to have an identical distribution pattem.'6

Table 1 Data on 21 controls and 13 infants and childrenwith pulmonary hypertension

Controls Pulmonary hypertension

Age No Age No

I mnth 3 i1 days 12mnths 2 2mnths 33 mnths 2 6-10 mnths 24 mnths 2 1-1-5 yrs 39-10 mniths 2 2-5-3-5 yrs 31-1-5yrs 3 8yrs 12-5-3-5 yr 4Adult 3

Allen, Wharton, Polak, Haworth

Tachykinin-like immunoreactivity was shown withantibodies raised to substance P but which cross reactwith the neurokinins A and B. Sections wereincubated with fluorescein isothiocyanate-con-jugated sheep anti-rabbit IgG (Wellcome Diagnos-tics) for an hour at room temperature. After a furtherrinse in phosphate buffered saline, sections weremounted in phosphate buffered saline/glycerol. Con-trols included the application of pre-immune serumas first layer and diluted primary antisera preabsor-bed with peptide (1-20 x 10-6 mol/l). Tissue sec-tions were examined by a fluorescence microscope(Olympus Vanox AH-2).We checked that the pulmonary vasculature in

those who died without cardiopulmonary disease wasindeed normal. Therefore blocks of tissue from areasof the lung adjacent to those used for the immuno-fluorescence study were prepared for light micro-scopical examination after staining with haematoxy-lin and eosin and with Miller's elastin stain counter-stained with van Geison's stain. The percentagearterial medial thickness (2 x medial thicknessdivided by external diameter) in arteries of differentsizes and the external diameter of arteries within therespiratory unit was determined and the findingswere compared with those of a previous extensivelight microscopical study of normal pulmonaryarterial development.'7

Results

FEATURES COMMON TO ALL THOSE EXAMINED ATALL AGESIntrapulmonary arteriesThe distribution of nerves accompanying the pul-monary arteries was assessed by the general neuronalmarker, protein gene product 9-5. Immunoreactivitywas seen in nerve trunks and single varicose fibres orfascicles (figs 1 and 2a and b). Nerve trunks contain-ing preterminal axons were found in the mid-outeradventitia ofarteries proximal to the respiratory unit.

Table 2 Antisera used to testfor peptides in nerves

Antiserum to: Code Dilution

General neuronal markers:Protein gene product 9-5 1648 1:600Synaptophysin 1623 1:800

Specific neuropeptide subtypes:Neuropeptide tyrosine (NPY) 1Q86 1:400C-flanking peptide ofNPY 1411 1:600Vasoactive intestinal peptide 652 1:2000Somatostatin 1082 1:400Substance P 910 1:500Calcitonin gene related peptide 1208 1:200Atrial natriuretic peptide (1-28) 1607 1:200

Tyrosine hydroxylase 1731 1:100

(See " for further characterisation of these antisera.)



j.com/

Br H


ber 1989. Dow

nloaded from


Nerves in hypertensive pulmonary vasculature

Fig 1 Photomicrograph of transverse section of an elasticintrapulmonary artery showing nerve trunks in the adventitia(arrow) that are immunoreactive for protein gene product9.5. L, lumen; M, media. Scale bar = 50 m.

They ran mainly parallel to the vessels and decreasedin size and number from the hilum. Single, mainlyvaricose, nerve fibres and fascicles occurredpredominantly at the adventitial-medial border andin the inner part of the adventitia in all arteries. Nonerve fibres were seen within the media of smallmuscular arteries and only rarely penetrated into theoutermost lamellar unit of elastic arteries. Nervefibres were never seen accompanying arteries withinthe alveolar wall.Most of the nerve fibres immunostained for

protein gene product 9 5 seemed to be sympathetic inorigin, showing both tyrosine hydroxylase (fig 2a andb) and neuropeptide tyrosine immunoreactivity (fig3). Subpopulations of nerves showing immunoreac-tivity for other neuropeptides were far less common,fibres immunostained for vasoactive intestinal pep-tide being the most abundant ofthese. Immunostain-ing for vasoactive intestinal peptide was localised infine, sparsely distributed varicose fibres, most com-monly associated with small muscular arteriesimmediately proximal to the respiratory unit (fig 4).These nerves were occasionally seen with elastic andrespiratory unit arteries. Varicose fibres that wereimmunoreactive for calcitonin gene related peptidewere common in the nerve trunks but not else-where (fig 5). Somatostain-like and tachykinin-likeimmunoreactive fibres were rare in all types ofartery.

PA TB

mmuno reactivenerve fibre

RB

RespiratoryUnit

AD

Avolus

(b)Fig 2 (a) Tyrosine hydroxylase immunoreactive perivascular nerve fibres (arrow) at the adventitial-medial border of analveolar duct artery in a child aged two and a half. Scale bar = 50 pm. (b) Diagram of terminal bronchiolus (TB) andairways of respiratory unit accompanied by an innervated pulmonary artery (PA). RB, respiratory bronchiolus; AD,alveolar duct. Square indicates area shown in (a).

355



j.com/

Br H


ber 1989. Dow

nloaded from


4, _ l | YEcirculation were localised at the adventitial-medialborder and like the pulmonary arteries were

predominantly immunoreactive for neuropeptidetyrosine and tyrosine hydroxylase.

CHANGES IN NERVE SUBTYPE ANDDISTRIBUTION RELATED TO AGEIn controls, changes in nerve distribution with agewere confined to the arteries of the respiratory unit.In children aged 1-4 months, approximately onethird of arteries accompanying alveolar ducts werenot innervated but by the age of 2-5 almost all wereinnervated and in adulthood they were invariablyinnervated. While most nerves supplying arteriesaccompanying alveolar ducts were immunoreactiveto protein gene product 9-5 and neuropeptidetyrosine, rare fibres containing vasoactive intestinalpeptide were also found. Nerve fibres could not beshown accompanying veins lying in the alveolarregion in individuals aged 1-5 years or less. In thoseaged 2-5 years, however, a small proportion of theseveins were accompanied by protein gene product 9 5and fibres that were immunostained for neuropep-tide tyrosine.

Fig 3 Photomicrograph showing nerve fibres expressing FINDINGS IN PULMONARY HYPERTENSIONneuropeptide tyrosine immunoreactivity accompanying analveolar duct artery (arrow) and respiratory bronchiolar In the lungs of patients with pulmonary hypertensionartery (arrowhead) in a three month old child. Scale bar= Sg4um.

INTRAPULMONARY VEINSThe nerves immunostained for protein gene product9-5 that accompanied the large proximal veins had adifferent distribution from that seen in proximalarteries; they formed a dense network throughout theadventitia (fig 6). Also these veins were more denselysupplied with varicose nerve terminals, and nervefibres could be found penetrating the muscle coat andextending to the subendothelium. There were moresynaptophysin immunoreactive fibres present thanin the arteries. In the wall of these veins and in thesurrovnding myocardial sheath most nerve fibres _.showed immunoreactivity for tyrosine hydroxylaseand neuropeptide tyrosine. In some of the large _proximal veins there was also a relatively densesubpopulation of somatostatin immunoreactivefibres, but fibres stained for other peptides were rare.Within the respiratory unit the veins were lessdensely innervated than the arteries. Atrialnatriuretic peptide was localised to myocardial cellsaround hilar veins but was not detected elsewhere inthe lung. Fig 4 Photomicrograph showing varicose nerve fibres

(arrows) that were immunoreactive for vasoactive intestinalBronchial arteries peptide accompanying small muscular arteries of a threeNerves accompanying the arteries of the systemic month old child. Scale bar = 50.um.

Allen, Wharton, Polak, Haworth356



j.com/

Br H


ber 1989. Dow

nloaded from


i aE 1 !||3|1in mature animals and nerve density in all arteriesincreased with age.' These differences may beassociated with the earlier extension of smoothmuscle to more peripheral arteries and the fasterdifferentiation of vascular smooth muscle cells in the

118pig.,In the human lung, nerves supplying the large

intrapulmonary arteries and veins differed in theirdistribution, because varicose nerve terminals wereoften seen penetrating the muscle coat in veins butnot arteries. This may reflect a difference in the wayin which the nerve stimulus is propagated in the two

E i _g j % S types of vessel. In the arteries, this may be due todiffusion ofneurotransmitter across the vessel wall orto electrocoupling of the medial smooth muscle cellsvia gap junctions."'92 In the veins, however, thegreater amount of connective tissue and fibroblastsmay hinder such processes and help explain the moredirect innervation of venous smooth muscle cells.

-~~~~~~ ~~Alternatively, the scarcity of nerves in the media ofarteries may be related to the higher intra-arterialpressure because pressure-induced degeneration ofsomatic nerves has been shown in the sheep carotid

20artery.Most perivascular nerves were presumed to be

Fig 5 Photomicrograph showing nervefibres (arrow) that sympathetic in origin in both normal and abnormalwere immunoreactive for calcitonin gene related peptide tissue at all ages. They expressed both tyrosineassociated with a muscular pre-acinar artery in the lung of atwo and a halfyear old child. Scale bar = 50 pm.

the density and distribution of the different sub-populatipns of nerve fibres within arteries and veinslying proximal to the respiratory unit were similar to _that in the-normal individuals at all ages. Within therespiratory unit, however, in all these patients E 2talveolar duct arteries (fig 2a) were accompanied byprotein gene product 9-5 (fig 7) and nerve fibres thatwere immunoreactive for neuropeptide tyrosine (fig8), even in those who were only 1 1 days old. The twopatients with idiopathic persistent pulmonary hyper-tension also showed extension of nerves into more

distal veins than was normal for their age. We foundno other differences in distribution or expression ofpeptide immunoreactivity.

Discussion

This study showed that the innervation ofthe humanpulmonary vasculature increases during childhood.Many arteries and veins of the respiratory unit didnot have any accompanying nerve fibres in childrenaged 1-4 months, but the proportion of innervatedvessels increased with age. This increase was morepronounced in the arterial than in the venous bed. Fig 6 Photomicrograph of a large proximal vein in a twoTheserfindinced mct trastwiththosein theve swhere month old child showing an extensive network offasciles andThese findings contrast with those in the pig where nerve fibres that were immunoreactive for neuropeptidethe distribution of nerves was as extensive at birth as tvrosine throughout the vessel wall. Scale bar = 50 um.

357Nerves in hypertensive pulmonary vasculature

--l.-.- -.-. --.10.---- -.-- --- -- r--,--

I



j.com/

Br H


ber 1989. Dow

nloaded from


occur in parallel with an increase in the amount ofsmooth muscle in peripheral arteries and with anincrease in the differentiation of that muscle. Sym-pathetic nerves may be exerting a long term trophiceffect on the vascular muscle cells. Certainly inculture, vascular smooth muscle cells undergo anaccelerated maturation in the presence of sympath-etic nerves and also show increased incorporation ofprotein in the presence of contractile agonists.5"

In all the controls and children with pulmonaryhypertension vasoactive intestinal peptide andsomatostatin (presumed parasympathetic) and

B tachykinin and calcitonin gene related peptide (sen-sory) immunoreactive nerve fibres were more spar-sely distributed than the relatively dense, presumedsympathetic innervation of the pulmonary vas-culature. Previous studies showed both nerves andspecific binding sites in the small muscular pulmon-ary arteries of the adult lung that were immunoreac-tive to vasoactive intestinal peptide.2526 In the presentstudy. at all ages vasoactive intestinal peptide fibresseemed most abundant in the small muscular arteriesjust proximal to the respiratory unit-arteries thatcan be considered as the resistance vessels ofthe lung.Extensive innervation of these vessels suggests thatFig 7 Photomicrograph showing nerves that were**

immunoreactive for protein gene product 9-5 localised to the the flow of blood into the respiratory unit andadventitial-medial border of abnormality thick walled intra-acinar arteries in a two month old child with pulmonary _hypertension. Scale bar = SO pm.

hydroxylase and neuropeptide tyrosine immuno-reactivity. Neuropeptide tyrosine has a direct effectmediating vasoconstriction and an indirect effectmodulating noradrenergic vasoconstriction via bothpre and post synaptic mechanisms.""2 In the mus-cular pulmonary arteries of the rabbit, neuropeptide _tyrosine was found to potentiate noradrenalineevoked vasoconstriction,2' but little is known aboutthe distribution and density of specific neuropeptidetyrosine receptors in the pulmonary vasculature.

In children with pulmonary hypertension whowere less than one year old, the arteries of therespiratory unit seemed to be prematurely innervatedby sympathetic-like nerve fibres. Associated changesin the smooth muscle cells ofthese peripheral arteriesalso occur over a similar period of time. Cells whichwould normally be pericytes, intermediate cells, orsmooth muscle cells with a synthetic phenotypebecome hypertrophied and have an increased con-centration of myofilaments and dense bodies andthus show ultrastructural features typical ofdifferen-tiated smooth muscle cells with a contractile pheno- Fig 8 Photomicrograph showingfibres (arrow) that weretype.' Many of these changes are an acceleration of immunoreactive to neuropeptide tyrosine running around thethe normal pattern of development. In both the adventitial border of an alveolar duct artery in a two monthnormal and pulmonary hypertensive lung, the old patient with pulmonary hypertension. Scale bardevelopment ofthe sympathetic innervation seems to 50 plm.

358 Allen, Wharton, Polak, Haworth



j.com/

Br H


ber 1989. Dow

nloaded from


Nerves in hypertensive pulmonary vasculature 359

capillary bed is partly under neurogenic control.Vasoactive intestinal peptide,2527 substance P,2

and calcitonin gene related peptide' are all potentvasodilators and all have been shown to relax pul-monary arteries in man or other species. As withvasoactive intestinal peptide, specific binding sitesfor substance P'3 and calcitonin gene related peptide"have been localised to medial smooth muscle cells ofhuman intrapulmonary arteries by in vitroautoradiography. Whether or not the action of thevasoactive peptides is endothelium-dependent isparticularly important in the context of pulmonaryhypertension because endothelial damage and de-nudation can be seen in the small muscular arteries ofmany pulmonary hypertensive children after-the ageofsix months.3 Substance P seems to require an intactendothelium'2 to be effective whereas vasoactiveintestinal peptide and calcitonin gene-related peptideseem to be less endothelium dependent.27"' If in thechildren with pulmonary hypertension endothelialdamage reduces the effectiveness of endothelialdependent vasodilator neuropeptides, this mightexplain why pulmonary hypertensive crises developin some children but not others. All the patients withpulmonary hypertension had abnormally thickwalled, prematurely innervated respiratory unitarteries. Numerous other vasoactive substancesacting on and produced by the endothelium may bejust as important. Studies on receptor localisation,density, and kinetics and on the interaction betweenvasoactive peptides, endothelium, and vascularsmooth muscle in the immature lung are indicated.The present study indicates that the vasoconstric-

tor peptide, neuropeptide tyrosine, is the predomin-ant neuropeptide in the pulmonary circulation at allages, in both the normal and pulmonary hypertensivelung. In infants with pulmonary hypertension, thepremature innervation of peripheral, abnormallythick walled, pulmonary arteries may be important inthe aetiology of pulmonary hypertensive crises-thevasoconstrictor episodes which characterise thepostoperative course of many young children afterintracardiac repair.

This work was supported by grants from the BritishHeart Foundation and the National Fund forResearch into Crippling Diseases. SGH is supportedby the British Heart Foundation.

References

1 Allen K, Haworth SG. Human postnatal pulmonaryarterial remodelling: ultrastructural studies ofsmoothmuscle cell and connective tissue maturation. LabInvest 1988;58:702-9.

2 Allen K, Haworth S. Cytoskeletal features of immaturepulmonary vascular smooth muscle cells and theinfluence of pulmonary hypertension on normalhuman development. J Pathol 1989;158:311-7.

3 Haworth S, Hall S, Burford B, Allen K, Mills A. Theeffect of pulmonary hypertensive congenital heartdisease on the post-natal remodelling of the pulmon-ary vasculature [Abstract]. JAm Coll Cardiol 1988;11(suppl A):135A.

4 Wharton J, Haworth SG, Polak J. Posmatal develop-ment ofthe innervation and paraganglia in the porcinepulmonary arterial bed. J Pathol 1988;154:19-27.

5 Chamley J, Campbell GR, Burnstock G. Dedifferentia-tion, redifferentiation and bundle formation ofsmooth muscle cells in tissue culture: influence of cellnumber and nerve fibres. J Embryol Exp Morphol1974;32:297-323.

6 Amenta F, Cavalloti C, Ferrante F, Tonelli F. Cholin-ergic innervation of the human pulmonary circula-tion. Acta Anat (Basel) 1983;117:58-64.

7 Kay K. Pulmonary vasculature and nerves. Com-parative morphological features of the pulmonaryvasculature in mammals. Am Rev Respir Dis1983;128:S53-7.

8 Dey R, Shannon W, Said S. Localisation of VIP-immunoreactive nerves in airways and pulmonaryvessels of dogs, cats and human subjects. Cell TissueRes 1981;220:231-8.

9 Hua X-Y, Theodorsson-Northeim E, Brodin E, Lund-berg J, Hokfelt T. Multiple tachykinins (neurokininA, neuropeptide K and substance P) in capsaicin-sensitive sensory neurones in the guinea pig and rat.Regul Pept 1985;13:1-19.

10 Wharton J, Gulbenkian S, Mulderry P, et al. Capsaicininduces a depletion of calcitonin gene-related peptide(CGRP)-immunoreactive nerves in the cardio-vascular system of the guinea pig and rat. J AutonNerv Syst 1986;16:289-309.

11 Sheppard MN, Polak JM, Allen JM, Bloom SR.Neuropeptide tyrosine (NPY): a newly discoveredpeptide is present in the mammalian respiratory tract.Thorax 1984;39:326-30.

12 Lundberg J, Martling C-R, H6kfelt T. Airways, oralcavity and salivary glands: classical transmitters andpeptides in sensory and autonomic motor neurones.In: Bjorkland A, Hofelt T, Owman C, eds. Handbookof chemical neuroanatomy, vol 6. Amsterdam:Elsevier, 1988:391-444.

13 Wharton J, Gulbenkian S, Merighi A, et al. Immuno-histochemical and ultrastructural localization of pep-tide-containing nerves and myocardial cells in thehuman atrial appendage. Cell Tissue Res1988;254:155-66.

14 Gulbenkian S, Wharton J, Polak J. The visualisation ofcardiovascular innervation in the guinea pig using anantiserum to protein gene product 9-5 (PGP 9-5). JAuton Nerv Syst 1987;18:235-47.

15 Wiedenmann B, Franke W. Identification and localiza-tion of synaptophysin, an integral membrane glyco-protein of Mr 38,000 characteristic of presynapticvesicles. Cell 1985;41:1017-28.

16 Gulbenkian S, Wharton J, Hacker G, Varndell I, BloomS, Polak J. Co-localisation of neuropeptide tyrosine



j.com/

Br H


ber 1989. Dow

nloaded from


360 Allen, Wharton, Polak, Haworth(NPY) and its C-terminal flanking peptide (CPON).Peptides 1985;6:1237-43.

17 Haworth SG, Hislop AA. Pulmonary vascular develop-ment: normal values of peripheral vascular structure.Am J Cardiol 1983;52:578-83.

18 Haworth SG, Hall SM, Chew M, Allen K. Thinning offetal pulmonary arterial wall and postnatal re-modelling: ultrastructural studies on the respiratoryunit arteries. Virchows Arch [A] 1987;411:161-71.

19 Burnstock G. Neuropeptides as trophic factors. In:Polak JM, Lindenlaub, E, eds. Systemic role ofregulatory peptides. Stuttgart: FK Schattauer Verlag,1982:423-40.

20 Keatinge W, Torrie C. Action of sympathetic nerves oninner and outer muscle of sheep carotid artery, andeffect of pressure on nerve distribution. J Physiol(Lond) 1976;257:699-712.

21 Pernow J, Saria A, Lundberg J. Mechanisms under-lying pre- and postjunctional effects of neuropeptideY in sympathetic vascular control. Acta Physiol Scand1986;126:239-49.

22 Ekblad E, Edvinsson L, Wahlestedt C, Uddman R,Hakanson R, Sundler F. Neuropeptide Y co-existsand co-operates with noradrenaline in perivascularnerve fibres. Reg Pept 1984;8:225-35.

23 Wahlestedt C, Edvinsson L, Ekblad E, Hakanson R.Neuropeptide Y potentiated noradrenaline-evokedvasoconstriction: mode of action. J Pharmacol ExpTher 1985;234:735-41.

24 Geisterfer A, Peach M, Owens G. Angiotensin IIinduces hypertrophy, not hyperplasia, of culturedrat aortic smooth muscle cells. Circ Res 1988;62:749-56.

25 Barnes PJ, Cadieux A, Carstairs JR, Greenberg B,Polak JM, Rhoden K. Vasoactive intestinal peptide inbovine pulmonary artery: localisation, function andreceptor autoradiography. Br J Pharmacol 1986;89:157-62.

26 Carstairs J, Barnes PJ. Visualisation of vasoactiveintestinal peptide receptors in human and guinea piglung. J Pharmacol Exp Ther 1986;239:249-55.

27 Greenberg B, Rhoden K, Barnes P. Relaxant effects ofvasoactive intestinal peptide and histidine isoleucinein human and bovine pulmonary arteries. BloodVessels 1987;24:45-50.

28 Tanaka D, Grunstein M. Vasoactive effects of sub-stance P on isolated rabbit pulmonary artery. J ApplPhysiol 1985;58:1291-7.

29 McCormack D, Mak J, Barnes P. Calcitonin gene-related peptide (CGRP) causes non-endothelialdependent relaxation of human pulmonary arteriesand veins [Abstract]. Am Rev Respir Dis 1988;137:372.

30 Carstairs J, Barnes P. Autoradiographic mapping ofsubstance P receptors in lung. Eur J Pharmacol1986;127:295-6.

31 Mak J, Barnes P. Autoradiographic localization ofcalcitonin gene-related peptide (CGRP) binding sitesin human and guinea pig lung. Peptides 1988;9:957-63.

32 D'Orleans-Juste P, Dion S, Drapeau G, Regoli D.Different receptors are involved in the endothelium-mediated relaxation and smooth muscle contractionof the rabbit pulmonary artery in response to sub-stance P and related neurokinins. Eur J Pharmacol1985;125:37-44.



j.com/

Br H


ber 1989. Dow

nloaded from


a nerves in - heart

Documents