Renal nerves modulate renin gene expression in the developing rat kidney with ureteral obstruction

Renal nerves modulate renin gene expression

in the developing rat kidney with ureteral

obstruction.

S S el-Dahr, … , R M Carey, R L Chevalier

J Clin Invest.

1991;

87(3)

:800-810.

https://doi.org/10.1172/JCI115083

.

Chronic unilateral ureteral obstruction (UUO) in newborn rats activates renin gene

expression in the obstructed kidney, and increases renin distribution along afferent

glomerular arterioles in both kidneys. To investigate the role of the renal nerves in this

response, 2-d-old Sprague-Dawley rats were subjected to UUO or sham operation.

Chemical sympathectomy was performed by injection of guanethidine, whereas, control

groups received saline vehicle. At 4-5 wk, renal renin distribution was determined by

immunocytochemistry, and renin mRNA levels were determined by Northern blot

hybridization. Compared to the saline-treated rats with UUO, renin remained localized to the

juxtaglomerular region in both kidneys of rats with UUO receiving guanethidine (P less than

0.05). Moreover, renin mRNA levels were eightfold lower in obstructed kidneys of rats

receiving guanethidine than in those receiving saline. Additional groups of rats with UUO

were subjected to unilateral mechanical renal denervation: renin gene expression in the

obstructed kidney was suppressed by ipsilateral but not by contralateral renal denervation.

These findings indicate that either chemical or mechanical denervation suppressed the

increase in renin gene expression of the neonatal kidney with ipsilateral UUO. We conclude

that the renal sympathetic nerves modulate renin gene expression in the developing kidney

with chronic UUO.

Research Article

Renal Nerves

Modulate Renin Gene Expression

in

the

Developing

Rat Kidney with Ureteral Obstruction

Samir Sayem El-Dahr,*R.ArielGomez,*Mark S.Gray,*Michael J.

Peach,*

RobertM.Carey,'Robert L.Chevalier*

With thetechnicalassistanceof MarjorieJ.Garmey* andBarbaraA.Thomhill*

Departmentsof*Pediatrics, tPharmacology, and lInternal Medicine, Universityof Virginia, Charlottesville, Virginia 22908

Abstract

Chronic unilateral ureteral obstruction (UUO) in newbornrats

activates renin geneexpression inthe obstructedkidney, and

increases renin distribution along afferent glomerular arteri-oles in both kidneys. To investigate the role of the renalnerves

in this response, 2-d-oldSprague-Dawleyratsweresubjected

toUUOorshamoperation.Chemical sympathectomywas

per-formed by injection of guanethidine, whereas, controlgroups

received saline vehicle. At 4-5 wk, renal renin distributionwas

determined by immunocytochemistry, and renin mRNA levels

weredetermined by Northern blot hybridization. Comparedto

the saline-treatedratswith UUO, renin remained localizedto

the juxtaglomerular region in both kidneys ofratswith UUO receiving guanethidine (P<0.05). Moreover, renin mRNA

lev-elswereeightfold lower in obstructed kidneys ofratsreceiving guanethidine than in those receiving saline. Additionalgroups

ofratswithUUOweresubjectedtounilateral mechanical renal denervation: renin gene expression in the obstructed kidney was suppressed by ipsilateral but not by contralateral renal

denervation. These findings indicate that either chemical or

mechanical denervation suppressed the increase in reningene

expression of the neonatal kidney with ipsilateral UUO. We concludethat the renal sympatheticnervesmodulatereningene

expression in the developing kidney with chronic UUO. (J. Clin. Invest. 1991. 87:800-810.) Key words: renalnerves*

gua-nethidine* sympathectomy-renin_*messengerRNA_*

immuno-cytochemistry * ureteral obstruction*newborn

Introduction

Thereareclose links between renal noradrenergicneuronsand

juxtaglomerular (JG)' cells which synthetize,store,andrelease

Portionsof thisstudywerepublishedin abstract form(1990. Kidney Int. 37:354A; 1990.Pediatr. Res.27:328A).

AddressreprintrequeststoRobert L.Chevalier, M.D.,Department of Pediatrics, MR4/2034, University ofVirginia Health Sciences

Center, Charlottesville, VA22908.

Receivedfor publication5February1990 and in revisedform 11

October1990.

1.Abbreviations usedinthispaper:AA,afferent arteriolarsegment;JG,

juxtaglomerular; MAP,meanarterialpressure;RAS,

renin-angioten-sin system;UUO,unilateral ureteral obstruction.

renin. Anatomically, the sympathetic nerve terminals are found in closeassociationwiththe cellsof theafferentarteriole and the granular JG cell (1, 2). On a functional basis, renal

sympathetic nerve activity modulates renin release

indepen-dent of thebaroreceptor and macula densa mechanisms,

pre-sumably byanaction mediatedby

B,

adrenergicreceptors

lo-catedonJGcells(3). Whatisnot known,however,isthe roleof thesympathetic nerves in the modulation of renin gene expres-sion. Ofparticular interest is the contribution ofthe renal nerves in themodulation of renin gene expression in

physio-logicorpathophysiologic statescharacterizedby enhanced in-trarenal reninproduction.Of equal importance is the interac-tion between the renal nervesand theintrarenal renin-angio-tension system (RAS) at thelevel ofgene expression during earlydevelopment,atime in whichintrarenalreninexpression

andsynthesisaremarkedly enhanced(4,5), whereas functional innervation of the kidneys is incomplete(6). Physiologic

stud-ies indicate that the intrarenal RAS is activated in animals subjectedtochronicunilateral ureteralobstruction (UUO)

(7-12). In contrast to the adult rat kidney, inwhich renin gene

expression isnotactivated by chronic UUO (13), 1-mo-oldrats

subjected to UUO since day 2 of life manifest a significant increaseinrenincontent and in thepercent of JG apparatuses

containing reninand its mRNA in the obstructed kidney as compared withsham-operatedcontrols(14).Interestingly,the

distributionofimmunoreactive renin alsowasfoundtobe

in-creasedintheintact opposite kidney, suggesting communica-tion betweenthe twokidneys(14). Thefactorsthat modulate

reningeneexpressionin response to UUO and themechanisms

underlying

theinteractionsbetween the twokidneys havenot

been

investigated

previously.

The presentstudywas

designed

to

investigate

the roleofthe

sympathetic

renalnervesin

modulating

reningene

expression

inresponse tochronic UUOin thenewbornrat.Wereasoned

thatgiven the close anatomical and functional relationships betweenthe renalnervesand

renin-containing

JG

cells,

activa-tionof reningene

expression

in

chronically

obstructed

kidneys

could be modulatedbytherenalnerves.

To testthis

hypothesis,

newbornrats

subjected

toUUOor

sham-operationweretreated

chronically

with

guanethidine

or saline

injections.

At 1 moofage, the intrarenal distribution of

immunoreactive reninand reninmRNAlevelswereassessed and the effects of

sympathetic

denervation

(guanethidine)

or

vehicle(saline) administrationwerecomparedamong the

dif-ferentgroups.In

_addition,

to assesswhether the effects of

gua-nethidine on renin gene

expression

were the result of renal

denervationandnotduetoother

systemic

orlocal effects ofthe

drug,

kidney

renin mRNAlevels wereexamined in

similarly

obstructedrats

subjected

tomechanical orsham denervation

of eithertheobstructedorintact

opposite kidneys.

800 El-Dahretal.

J.Clin.Invest.

©TheAmericanSocietyfor ClinicalInvestigation, Inc. 0021-9738/91/03/0800/11 $2.00

Methods

ProtocolNo. 1:systemicsympathectomywithguanethidine. These

ex-perimentswereperformedon59 2-d-oldSprague-Dawleyratswhich

were divided into fourgroups. Twogroups weresubjectedto sham operation and receivedsalineorguanethidine (Sham+S,n= 13;and

Sham+G,n= 12; respectively).Twoadditionalgroupsweresubjected

toleft complete ureteral obstruction and received salineor guanethi-dine(UUO+S,n= 12;andUUO+G,n= 22;respectively).Under

halothane andoxygenanesthesia,asmallincisionwasmadein theleft

lowerquadrantof the abdomen and the distalureterwasligated(UUO)

orleftundisturbed (Sham) (14).Theabdominal wall thenwasclosed in

onelayer andthe animalwasallowedtorecoverfromanesthesia.

Sub-sequently, the animalswerereturnedtotheir mothers untilweaningat

21d ofage,after which theratswerefedregularratchow(Purina 5012, Ralston-Purina,St.Louis, MO)andwereallowed freeaccessto

drink-ingwater.

Onday7of life(5d after UUOorshamoperation),guanethidineor

saline injections were begun. Guanethidine monosulfate (CIBA-GEIGYPharmaceuticals, Summit, NJ)wasdiluted in 0.9% saline (5-10_gl/gbodyweight)andwasgiveninadose of 50mg/kgperd,5 da

week for 3 wk by subcutaneous injections (a total of 15 injections) accordingtothe method of Johnsonetal.(15). Equalamountsof 0.9% salinewereadministered inasimilar fashiontosaline-treated animals.

ProtocolNo.2:mechanical renal denervation. 15ratswereusedfor

theseexperiments.TheratpupsweresubjectedtoleftUUOatday2 of

lifeasdescribedabove. At 25-27 d ofage,underpentobarbital

anesthe-sia, left renal denervation (n= 5), right renaldenervation(n=_6),or

sham renal denervation (n= 4)wereperformed. Renal denervation

wasaccomplishedasdescribedpreviously ( 16, 17).Undermicroscopic visualization, the renalarterywascarefullystrippedofitsadventitia andsubsequentlycoated witha solution of 10% phenolin absolute alcohol. Duringthis procedure, the kidney and surroundingtissues

werecarefullyprotected fromexposuretophenol. Sham renal

denerva-tionwasaccomplished byleftorrightabdominalincisions, visualiza-tion ofthekidneys,but the renalpedicleswereleft undisturbed. It has been shown that this denervationprocedureresults innearlytotal

elimi-nationof renal tissue catecholamines within 72 h(17).

At 4-5 wkofage,afteranovernight fastwith freeaccesstowater, the ratsin protocols 1 and 2wereanesthetized with intraperitoneal pentobarbitalsodium(Abbott Laboratories, Chicago, IL) (4 mg/100g

bw)andplacedon athermostatically-controlled heatingtableto

main-tainrectal temperatureat37±0.5°C.Therightcarotid arterywas

can-nulatedwithpolyethylene tubing (PE 50, Clay Adams, Parsippany, NJ)

and meanarterial pressure (MAP)was recorded continuously by

meansofaStatham23 IDpressuretransducer(Gould Inc., Oxnard,

CA) coupledtoaHewlett-Packard7754 Brecorder. Theleftureteral

diameterwasmeasured withcalipers midwaybetween therenalpelvis

and thebladder. Theratsweresacrificedwithsaturatedpotassium

chlo-ridesolutioninjected through the carotid catheter. The kidneyswere

removed,decapsulated, weighed, and processed for immunocytochem-istry andmRNAanalysisasdescribed below.

Renin,beta-tubulin, and tyrosinehydroxylase

immunocytochemis-try. Seven Sham+S,5 Sham+G,5 UUO+S,and 6UUO+Grats werestudied. Immunocytochemical localization of renin, beta-tubulin,

andtyrosine hydroxylasewasperformed using the ABC

immunoper-oxidasetechnique (4, 5, 13, 14). The primary antibodies used in the

presentstudyincluded (a)aspecific polyclonal antirat renin antibody

raised inthe rabbitwhich bindstoboth renin and prorenin (diluted 1:1,000, gift of Dr. T. Inagami, Department of Biochemistry,

Vander-biltUniversity) (18, 19), (b) aspecific monoclonal antibody against

class IIIbeta-tubulinisotype(neuron-associated), diluted 1:200 (gift of

Dr. A. Frankfurter, University of Virginia Health Sciences Center) (20), and (c)aspecific polyclonal antibody for tyrosine hydroxylase, the

enzymeresponsiblefor theconversion oftyrosinetophenylalaninein thecatecholaminesynthetic pathway (Eugene Tech International, Inc.,

Allendale, NJ, dilution 1:4,000). After adding the secondary antibody

(biotin-conjugatedanti-rabbit IgG raised in the goat), the sections were incubated with avidin horseradish peroxidase complex (Vectastain ABC Kits, Vector Laboratories, Inc., Burlingame, CA), exposed to 0.1% diaminobenzidine tetrahydrochloride and 0.02% H202 as a source ofperoxidase substrate, and counterstained with hematoxylin. Negative controls included omission of the primary antibody, replace-mentof primary antibody by nonimmune rabbit serum, and omission of secondaryantibody, andnostainingoccurred.

Threesections per kidney were immunostained for renin and exam-ined by light microscopy for quantitative analysis of intrarenal renin distribution. The number ofrenin-stained and total number

ofjuxtaglo-merular apparatuses (JGA) were counted in each section. With the aid ofanocularmicrometer, the total length oftheafferent arteriolar seg-ments(AA) and the lengthof renin immunostaining along eachAA were measured. All measurements were done on coded slides such that theinvestigatorwas unawareof thetreatmentgroups. Because the total numberof JGA variedamongsections, thefollowing immunohisto-chemical ratiosweredeterminedandcompared among groups (5, 13, 14): %JGA=(number of stained JGA/total numberofJGA)X 100; %LAA =_{(length of stained AA/total length of AA)} x 100.Theratios thusobtained from each slidewerepooled and averaged for each kid-ney. Comparisons were made using the final ratios obtained for each kidney. Only JGAs witha vasafferens enteringtheglomeruluswere used forcomputation.

Oneto twosections of each kidneywerestained for beta-tubulin andfortyrosine hydroxylaseas a meansofdocumentingthesuccessof renaldenervation.

ReninmRNAanalysis.TotalRNAfrom the left andright kidneys of sixratsin each group in protocol 1 and from the three groups in protocol 2 was extracted accordingtothemethodof Chirgwinetal. (21) and pooled for each group. 15ggofprecipitatedRNA was sepa-ratedbyelectrophoresison a 1.2%agarose-formaldehydegelandthen transferred from the gelto anylon membrane(Zetabind, Cuno,Inc., Meriden, CT) by capillary action(Northern blot).Tocontrol for the differentefficienciesatwhichmRNAspeciesmightbe extracted rela-tiveto totalRNA, each RNAsample wasanalyzed bothfor renin mRNAandforthe mRNAof the constitutively expressed cytoskeletal protein, beta actin,as aninternalstandard.

The Northernblots werehybridizedtoafull-length renin cDNA, and tobeta-actincDNA(giftsofDr. K. R.Lynch andDr.G.Owens, University of Virginia) labeled with phosphorus-32 by a random primer technique (22)orby nick translation (23)to aspecificactivity of > 1.6x 108cpm/,ug, isolated using Sephadex G-50 columns (Pharma-cia FineChemicals, Piscataway, NJ) and eluted bycentrifugation. 32p_

cDNAsweredetectedbycontactautoradiography andquantifiedby

densitometry.

Renal norepinephrine content. Kidney norepinephrinecontents were assayed by HPLC with electrochemical detection (Bioanalytical Systems Inc.,WestLafayette, IN).

Plasmarenin activity was measured by radioimmunoassay (24). Statisticalanalysis. Comparisons among the groups were per-formed by one-way analysisof variance followed byNewman-Keuls test; and between the left and right kidneys in each group by pairedt

test.Statisticalsignificanceisdefinedas P <0.05, andthe results are presented as means±SEM.

Results

Animal survival

In protocol 1, eightratsdied duringthe study resulting in an

overallsurvivalrateof87%. Forindividualgroups, the survival

ratewas 100% in Sham +S,92% in Sham+G, 92% in UUO+

S, and 73% in UUO+G.GuanethidineorUUO alone didnot

TableI.Characteristics ofAnimalsinthe ChemicalSympathectomyProtocol

Body LKW/I00 g RKW/I00 g Left ureteral

Group weight bodyweight body weight diameter MAP

g g g mm mmHg

SHAM+S

(n= 10) 85.0±5.2 0.50±0.01 0.52±0.01 1.00±0.03 97.0±6.5

SHAM+G

(n= 11) 76.3±1.5 0.47±0.01 0.49±0.01 0.90±0.02 92.0±2.6

UUo+S

(n=7) 78.0±4.0 0.33±0.061w

0.77±0.0211

2.20±0.200 101.0±3.1

UUO+G

(n= 16) 70.0±2.5* 0.45±0.02' 0.67±0.0211 1.80±0.10t 82.0±5.0

LKW,left kidney weight;RKWrightkidney weight.Meanarterial pressure (MAP) was measured in 7 Sham +S, 10 Sham + G, 4 UUO +S,and 11 UUO+Grats. *P<0.05vs.Sham+S; tP <0.01vs. Sham + S and Sham + G;I_{P < 0.05 vs. UUO + G;1"P<0.01 vs. other groups;}'P

<0.01 vs.right kidney; mean±SEM.

rats, whereas thecombinationof ureteral obstruction and

sys-temic sympathectomywasassociated with increased mortality.

Inprotocol 2,survivalrate was 100%.

Somatic

and renal growth

Comparedto saline-treated rats, chemical sympathectomy

with chronic guanethidine treatment did not affect body

weight insham-operated rats. Similarly, saline-treated animals

subjected tochronic neonatal UUOdid not suffer from any

growth deficit. However, compared to the Sham + S group

only, UUOratswithsympathectomy (group UUO + G) had a

significantlylowerbodyweightby the endofthethirdweekof guanethidinetreatment(Table I). Bodyweightswere not

differ-entamong the three groupssubjectedtosham ormechanical denervation (Table II).

Theeffects of UUO on renal growth weredifferent inthe

saline-treated

and the

sympathectomized

animals (Table I). Compared with sham-operated kidneys, a significant growth arrestoccurredinthe obstructedkidneys of UUO+S but not

UUO+Grats. In

addition,

theleft ureteral dilationwas signifi-cantlygreaterinsaline-treated than in sympathectomizedrats (P <0.05)(TableI). Compensatory hypertrophyoccurred in

theintact opposite kidneys of both UUO+ SandUUO+G

Table II. Characteristics ofAnimalsintheMechanical

DenervationProtocol

Body

Group weight LKW RKW MAP

g g g mmHg

UUO+LMD

(n=5) 64.0±9.1 0.14±0.02* 0.59±0.08 123.0±11.0 UUO+RMD

(n=6) 74.0±6.3 0.23±0.05* 0.64±0.03 124.0±4.4 UUO+SMD

(n=4) 80.0±5.1 0.19±0.05* 0.68±0.07 115.0±9.0

LMD, left mechanical denervation. RMD, right mechanical denerva-tion.SMD,sham-mechanicaldenervation.*P<0.05vs.right kid-ney.Mean±SEM.

animals. However,thehypertrophicresponse wassignificantly

greater intheintact kidneys of UUO+ S than of UUO+ G

groups (Table I).

Unlike sympathectomy, neither ipsilateral nor contralat-eral renaldenervation had aneffect onthedegree of growth

arrestin the obstructed

kidneys

as compared with

sham-me-chanical denervationrats.Themagnitude ofcompensatory hy-pertrophy in the intact kidney with contralateral UUO was

similarinthe three mechanicaldenervation groups(Table II).

Mean

arterial pressure

There were no statistically significant differences among the groupsinMAP, although the UUO+ G rats tended to have

slightly

lower blood pressures (Table I). Similarly, MAP was

similarin all threemechanical denervationgroups(Table II).

Kidney norepinephrine content, beta-tubulin and tyrosine

hydroxylase immunoreactivity

Guanethidine administration resulted in an80-90% decrease in renal norepinephrinecontent as compared to the valuesin saline-treatedrats(Table III).Asshownin Table IV,

mechani-cal denervation resulted ina

significant suppression

of renal

norepinephrine

content in both UUO and sham groups. In

saline-treatedratsof eithersham or UUO groups,beta-tubulin

andtyrosine hydroxylaseimmunoreactivity wasdistributed mainly along the adventitia ofthemajor intrarenal vessels

(ar-cuate,interlobular) with occasional stainingalongtheafferent

Table III. Guanethidinedenervation kidney norepinephrine

content

Group Leftkidney Right kidney jug/gkidneywt ug/gkidneywt

Sham+S 0.240±0.030 ND

Sham+G 0.030±0.010* ND

UUO+s 0.120±0.020* 0.130±0.010 UUO+G 0.019±0.003* 0.017±0.004*

*_{P<0.01 vs.thecorrespondingsaline-treatedgroup;}tP<0.05vs.

sham+S.ND,notdone.n=5 in eachgroup.Mean±SEM.

Table IV.Mechanicaldenervation kidneynorepinephrinecontent

Denervated Nondenervated

Group kidney kidney

jug/gkidney wi pg/gkidney wi

UUO+LMD 0.071 0.118

UUO+RMD 0.001 0.142

SHAM+LMD 0.004 0.150

SHAM+RMD 0.032 0.137

Mean* 0.027 0.137

SEM 0.016 0.007

*_{P <0.02.}

arterioles (Fig.1,AandB).Beta-tubulin andtyrosine hydroxy-laseimmunoreactivitywas notdetectedinthe

kidneys

of

gua-nethidine-treated animals of eithersham orUUOgroups.

Intrarenal

renin

distribution

Therewere no statistically

significant

differences in the

num-ber

(mean±SEM) of

total JGAsperratused for

computations

amongthe

kidneys

of the

experimental

groups

(Sham

+S: left

kidney 66±15.9, right kidney 70±8; Sham + G: left kidney

55±7.7, right kidney 61±5.7; UUO + S: left kidney 61±4.7, rightkidney68±9;UUO + G:left kidney

74+22.7,

right

kid-ney59±10.8).

Effect

ofsympathectomy

in

sham-operated

rats. No

differ-ences werenotedinthedistribution of immunoreactiverenin

in the kidneys of saline-treated (Sham + S) versus guanethi-dine-treated (Sham +G)rats. Inboth groups,reninwas con-finedtothose cellsoftheafferentarteriolesthat were juxtaglo-merular in location. Renin

immunoreactivity

was not ob-servedbeyond 45-50

,m

upstreamfrom the glomerulus along theAA. In

addition,

thepercentageof

renin-containing

JGA (%JGA)andthepercentagelength of theAA

containing

renin (%LAA)weresimilar in thesetwogroupsofrats(Figs.2and3).

Effect of

UUO insaline-treated rats. In contrast to sham-operatedkidneys in which reninwasconfinedto

ajuxtaglomer-ular position, renin was found along most ofthe identified length ofthe AA in the obstructed kidneys ofsaline-treated UUOrats(UUO + S)(Fig. 4A).Although lessdramatic,

in-creaseddistribution of reninwas seenalso intheintact

oppo-site kidneys of UUO+Srats

(Fig.

4B).Inaddition,the%JGA

and %LAA were

significantly higher

in boththe obstructedand

intact opposite kidneys of UUO + Sgroupthan in thoseof sham-operatedrats(Figs. 2and3).

Effect

of

UUOin

sympathectomized

rats.Renin immunore-activity of the AA was greatly diminished in both the ob-structed and intact opposite

kidneys

of UUO+ Grats when compared to that of UUO + Skidneys(Fig. 4, C and D). Thus,

comparedto sham-operated kidneys,renindistribution(both %JGA and %LAA) wasincreasedin the obstructed kidneys of UUO+ S but not UUO+G rats(Figs. 2 and3).

Sympathec-tomy also prevented theincreasein thedistributionof

immuno-reactive renin in the intact kidneys with contralateral UUO

whichwasseen insaline-treatedUUO rats (Figs. 2 and 3). No significant differences in plasma reninactivitywere

foundbetween UUO + S and UUO + G rats (mean±SEM;

UUO + S[n = 4]: 5.7±1.8 ng/ml perh; UUO + G [n = 4]: 5.8±1.1 ng/mlperh).

Renin mRNAstudies

Asshown in the Northern blot in Fig. 5, guanethidinegiven chronicallytosham-operatedmaturing rats did not appearto

haveanappreciable effect on renin mRNA levels relativeto

those of saline-treated rats. Renin mRNA levels quantitatedby densitometrywereeightfold higherin the obstructed kidneys of UUO + S rats than in the intact opposite kidneys and the kidneys of the remaining groups. Fig. 6 shows the results of

hybridizationofthe Northern blot with beta actin cDNA. Mechanical denervationofthe obstructed kidneys was

as-sociated with a fivefold decrease of renin mRNA in these kid-neys, whereas contralateral denervation of the intact kidney

did not affect renin gene expression in either kidney as

com-pared

tosham denervated obstructedrats

(Fig.

7).

Discussion

To ourknowledge theinterrelationships between the sympa-thetic nervous system and the renin-angiotensin system at the level of gene expression have not been reported previously in the developing animal. In this respect, two mainquestions

arise: (a) Are renal nerves important in thedevelopmental regu-lation of renin gene expression? (b) What role do renal nerves

playinthe modulation of renin geneexpressioninthe setting of enhanced intrarenal renin production? The purpose of the present study was to investigate the importance of the renal nerves inmodulatingtheexpressionof renin and its mRNA in the developing kidney with chronic ureteral obstruction.

The results of the present studyindicate that renin gene expression is activated inchronicallyobstructedkidneysof

ma-turingratsand that an intactsympatheticrenalinnervationis

requiredfor thatactivationto occur. In contrast, sympathec-tomy did not influence the distribution of renin or the amount of its mRNA in the kidneys of nonobstructed rats.

The chemical denervationprotocol employed inthe pres-entstudywasadaptedfrom that of Johnsonetal.( 15).These

investigators demonstratedthat chemical sympathectomy of newborn rats with chronic guanethidine administration re-ducesperipheral tissue norepinephrinecontentby>90%, abol-ishes ganglionic tyrosine hydroxylase activity, and results in destruction ofperipheral noradrenergic neurons, producinga permanent sympathectomy. The present study confirms the

findings of Johnson et al.Guanethidine treatment markedly reduced kidney

norepinephrine

content aswellascompletely

abolishedintrarenal tyrosine hydroxylaseandbeta-tubulin

im-munoreactivity (15). Weadapted thisprotocolfor the follow-ingreasons: First, the experiments of Johnsonet al. were per-formed on ratsduringthe newborn period, the same period

during which our experiments were performed. Second, we

performed

UUO atday2oflifewhenmechanicaldenervation would betechnicallyimpractical due to the small animal size.

Third, thismethod does not affectcentralautonomicrelay centers(15),permittingthetransmissionofimpulsesbetween any

remaining

intact afferentandefferentlimbs. However,to becertain that thechangesinintrarenalrenininduced by

*- ::::.. Ni:: a

t.~~~~~~

~

4;..1;.*.. _IN

.

*..!..^ ... .! A... a.

Sb

,.'

%

i, w.

4we

!AW_. ....

-id

* _

_.k... ..

t..

0:

.

i

W

....

..

0

a II

9

vi: '

*, :.~~~~~~~~~~~~~~~~4

' S ,,,

t$rst t ~~~~~*V

"A

ir_

4v$r~~~*

er4~

bS a2J5*

4V

d

46

_4''o.

'LA!

* _s o~~~~~Iv _

W.

_0.

V.

p(0.05 p(0.05

P(0.01 D(O.01 =LEFT z

8100 0D(.05 _RIGH

80

~~~~~~~~~p(0.05

z

z

W 40)

t 20

W

W0

a. 0

SH+S UUO+S SH+G UUO+G

P(0.01

p(0.01 p(O.01 D(0.05

P(0.01 P(0.01

_LEFRT

IIIIIRIGHT

Figure 2. Percent ofrenin-containing juxtaglomerularapparatuses in thekidneys of the differenttreatmentgroups. Sham+ S (n =7),

Sham+G (n= 5), UUO+S (n=5), UUO+G (n=6).

studies inwhich ipsilateral orcontralateral mechanical renal denervationwereaccomplished. Due to thesmall size ofthe animals, mechanical denervationwasdelayed until 3.5wkof

age,when therenal nerves could beidentifiedreliably.

Theresultsofourimmunohistochemical studiesin the

ob-structedkidneysareconsistentwith the reninmRNA

accumu-lation.Chronic UUO in saline-treatedratsresulted ina

signifi-cantincrease in both %JGAand %LAAwithimmunoreactive renin,consistent withourprevious findings(14).Thiswas

as-sociated withaneightfold increaseinreninmRNAlevels

com-pared to the intact

opposite kidneys

and the kidneys of the remaininggroups. In contrast, inchronically denervated

hy-dronephrotic

kidneys, intrarenal renin distribution andgene

expressionweresuppressedtolevelsnotsignificantly different from those ofsham-operated control rats. The specificity of

these results was testedbythemechanicaldenervationstudies: Renin gene expression in the obstructed kidneys was

sup-pressed byipsilateralbut notcontralateral renaldenervation. Comparedto

guanethidine

treatment

(Fig.

5),the lesscomplete

suppression

of reningene

expression

in the obstructed

kidneys

subjected to mechanical denervation (Fig. 7) may be due to

incomplete denervation. This is suggested by the variation in norepinephrine content of mechanically denervated kidneys (Table 4).

These results provide novel evidence thatrenin gene

ex-pression in chronicallyobstructedkidneys ofthematuring rats

is facilitated by the chronic or long-term presence of intact

sympathetic nervous system and renal nerves. On the other hand, the lackofan effect ofguanethidine in sham-operated

animalsonkidney renin synthesisand distribution, despite a

comparable degree of denervation, argues against a primary role of the sympathetic nervous system in the regulation of

reningene expression innormalanimals. It should be noted that chemical sympathectomy was associated with a greater

Figure3. Percent of distal identified afferent arteriolar length with renin immunostaining. Sham+S (n=7),Sham+G (n=5),UUO +S(n=5),UUO+G(n=6).

decrease in %LAA than %JGA in the obstructed kidneys of UUO + G than in those

of

UUO+ Srats. Itis

possible

that after denervation ofthe obstructed kidneys, the first cellsto

switchoff reninproductionare thosewhichwererecruited dur-ingUUO toexpressrenin, locatedupstreamfrom the

glomeru-lusalong theafferent arteriole.

The mechanism(s) underlying the

suppression

of renin

geneexpression by

sympathetic

denervation in the UUO

kid-neys is unknown. Several

possibilities

exist. Recent

prelimi-narystudies havedemonstrateda transient increase in renin

gene

expression

inresponse to acute

beta-adrenergic

receptor

stimulation with

isoproterenol (25).

Theincreaseinreningene

expressionwaspreceded byadecrease inrenalrenincontent,

suggesting that isoproterenol increased renin secretionrateand

depleted renin stores, thereby

stimulating

renin gene

expres-sion (25). Stimulation ofrenal mechanoreceptors and in-creased activity of afferent renal nerves have been shown to

occurinresponseto acuterisesinureteralpressure inadultrats (26).Itisnotknownwhether theseevents occuralso inchronic UUO inthe young

animal,

and whether anincrease in

ipsilat-eral efferent renal nerve

activity

also occurs. Should these

events occur,

norepinephrine

released from

noradrenergic

nerveendingsmaycontributetoincreased reningene

expres-sion intheobstructed

kidneys, presumably

viaits interaction with the JG cell

B1

receptor-cAMP system. Supporting this

possibility,

renal

adrenergic

axons and terminals have been demonstratedtobeintact in the

severely hydronephrotic

kid-ney(27).Moreover, stimulation of the

hydronephrotic

kidney microvessels with beta-adrenergicagonists increases renin

re-lease,indicatingthatbeta-adrenergicreceptors arefunctionally intact(27). Another

possible

mechanism for suppressed renin

gene

expression

by

sympathetic

denervation in UUO involves hemodynamic factors. The role of renalnervesinthe

vasocon-striction ofthekidney with chronic UUOhas notbeenstudied

Figure 1.(A) Immunocytochemical detection of beta-tubulin in a kidney from a saline-treated sham-operated rat. Beta tubulin immunoreactivity (arrows) isfoundalong the outer wall of the afferent arteriole. (X200).Betatubulin immunoreactivity was absent in the kidneys of

A

B

.A..C,. ::.

Figure 4. Renin immunocytochemistry. (A) Section from an obstructed kidney of a saline-receiving rat (UUO + S group); renin

immunoreactivity occupies the entire length of the afferent arteriole. (B) Section from the intact opposite kidney of the same animal in A; similar totheobstructed kidneys, renin immunoreactivity is found along most of the length of the afferent arteriole. (C) Section from an obstructed

...

.9 .0

1.

'4r

I.

4 C~~~~~~~~~~~~~~~~~A

4 i4

1*'

* ₄

*

I

i%

Ati.

xi.

I,

I...

`.

40 -::.S.' .A .: %-. ..'N.

ON<.1 /~~~~~~

4~~~~~~~~~~~~~~

I~~~~~~0

dtoth Vk

#4~~~~~~~~~~~~~P

t<. V

A

f,4

.41

,P"I"..

f

.0 fl,

i&

*

a

$1

A

4Mb

. _4~I

4

,.r: 1:1* -.p,::

aw

-.:. 'Y 'te

)..1

GUANETHIDINE

SHAM uuo

acutecontralateral UUO (30), it isnotclear whether renorenal reflexes account for the parallel increase in immunoreactive renin in both the obstructed and intact opposite kidneys. It is

importanttonotethatalarge fraction of the immunoreactive renin contained in the intact kidney in the presence of

contra-lateral UUO may represent inactive renin(prorenin), because thepolyclonalreninantibodyrecognizes both renin and prore-nin and because active reprore-nin content was depressed in the

non-obstructed hypertrophied kidneys (14). These findings may

helpto explain why vasodilation occurs in these kidneys de-spite an increase in renin distribution. The lack of increase in renin mRNA in the intact opposite kidney of UUO + Srats

despitean increase in renin immunoreactive renin has been addressed previously (14). Possible explanations include the longer half-life of the stored protein (prorenin vs. renin), de-creased conversion of prorenin to renin, enhanced transla-tionalefficiencyof renin mRNA,orrenin uptake by the renal

microvasculature.

Theresultsof the presentstudyalsoindicate that

compensa-toryrenal hypertrophy wasproportionatelygreaterinthe

in-L R L R _{L R L R} tact _{opposite kidney} ofUUO + Sthan UUO + G rats. An interesting findingwasthat guanethidinetreatmentattenuated the degree ofgrowth arrest and ureteral dilation in the

ob-structedkidneysof UUO+Gvs.UUO+Srats. It ispossible

Figure5. Northern blot. 15Atgof total RNA whichwaspooled from the left (L)andright (R) kidneysof sixanimalsin eachgrouphave been loaded into each lane and hybridized witha 1.43-kbratrenin cDNA that has been labeled with 32p (seeMethods).Bydensitometric analysis kidney renin mRNA levelswereatleast eightfold higher in the obstructedkidneysof saline-treated (0.89 absorbance units) than the obstructedkidneysof guanethidine-treatedrats(0.08 absorbance units). Denervation by itself did nothaveasignificant effectonrenin

geneexpressioninsham-operatedkidneys (0.11 absorbance units).

innewbornanimals.Althoughin adultdogs, chemical denerva-tion with6-hydroxydopamineoralpha-adrenergic blockadedo not attenuaterenal vasoconstriction (28), aprimary effect of

guanethidineoripsilateralmechanical denervation onthe

va-soconstrictiveresponsetoUUOcannotbe ruledout.Another

possiblemechanism is the effect of renal denervation in

upre-gulatingintrarenal angiotensin II receptor number and

en-hancing itsintrarenal action (17). These effects ultimately wouldresult inanegative feedback inhibitiononrenin

biosyn-thesis(29).

We have previously found that 4-wk-oldratssubjectedto neonatal UUOdisplayanincrease in the renovascular

distribu-tion ofimmunostainingrenin in both the obstructed and intact

oppositekidneys comparedtosham-operated kidneys (14).In thepresent study, bilateral renal denervation with

guanethi-dine resulted ina significantdecrease in renin distribution in both the obstructed and intactopposite kidneys. Although ef-ferent renalnerveactivityis decreased in the intactkidneywith

S

... A L I N

E

.:

GU

:S

H AM uu0 S H.

..

%

Ml ~ ~ ~ .

AN ET H I DI NE

A M UUO

Ro

.'

...l-....

..I...

,

Figure6.Northernblothybridizedwith 32P-labeledbeta-actincDNA

afterdehybridizationof the reninprobe.Chemicalsympathectomy protocol.

808 El-Dahretal.

SALINE

SHA

M

uuo

0,2"'IL'..

" _"a

SMD

RMD

LMD

L

R

L

R

L

R

In summary, chronic UUO in newborn animalswas

asso-ciated with activation of renin gene expression in the ob-structedkidneys. Chronic sympathectomy with guanethidine or ipsilateral mechanical renal denervation suppressed renin gene expression in the hydronephrotic kidneys. Sympathec-tomywith guanethidine also was associated with a decrease in renindistribution in the obstructed kidney as well as the intact

kidney withcontralateral UUO. In control animals, intrarenal renin distribution and gene expression were not affected by chronic denervation. We conclude that the sympathetic renal nerves modulatereningeneexpressionof developing kidneys

inthesetting of chronicureteralobstruction.

Acknowledgments

Guanethidine monosulfate (Ismelin) has been generously provided by CIBA-GEIGY Pharmaceuticals, Summit, NJ. The authors are grateful toDr. J. C. Pelayo (University of Colorado School of Medicine) for his help in themechanical denervation procedure.

Dr.S. El-Dahr is a research fellow of theAmericanHeart

Associa-tion,VirginiaAffiliate.This workwassupported byNationalInstitutes of Health grantDK40558.

References

Figure7. Northemblothybridized with32P-labeledrenin cDNA. Mechanicaldenervation protocol. 15,gof totalRNAwhichwas pooled from theleft(L) andright(R)kidneysof each group have been loaded into each lane. LMD, left mechanical denervation (n= 5); RMD,right mechanical denervation (n=6); SMD,sham-denervation

(n=4).Comparedtothe obstructed kidneys of SMD rats, renin mRNAlevels werefivefold suppressed in the obstructed kidneys with ipsilateral (LMD) but not contralateral (RMD) mechanical denervation.

thatalowerMAPin UUO+Gratsmay havecontributedto

the enhancedpreservation ofrenalmassin the obstructed

kid-neys.Preservation ofrenal massdidnot occurinthe

mechani-cally denervated kidneys, but this could be related to the

shorter duration of denervation in these kidneys (5-7 d) as

compared withthechemically-denervated kidneys(3 wk).

Re-gardless of the mechanism involved, preservation of mass in theobstructed kidneysmayhavecontributedto the reduced compensatory hypertrophy in theintact opposite kidneys of

UUO+Grelativetothoseof UUO+Srats.Similar preserva-tionof renalmassin the obstructedkidneys also has been

re-ported in normotensive captopril-treated ratswith chronic UUO(1 1). Takentogether, these findingssuggest that

capto-pril,by blocking angiotensin II formation, and guanethidine,

by decreasing renin synthesisand/orreducing MAP contrib-utedto thepreservation ofrenalmassofthechronically ob-structedkidneys. This may have occurred by limiting the

ef-fects of angiotensininproducingglomerular hypertension (3 1)

orintrarenal vasoconstriction (8-10).

1.Barajas,L.1978.Innervation ofthe renalcortex.Fed.Proc.37:1192-1201.

2.Barajas, L., andP.Wang. 1979.Localizationoftritiatednorepinephrinein therenalarteriolarnerves.Anat. Res. 195:525-534.

3.Osborn, J. L., G.F.DiBona,and M. D.Thames. 1981. Beta-I receptor mediation of renin secretion elicited by lowfrequencyrenalnervestimulation. J.

Pharmacol.Exp.Ther.216:265-269.

4.Gomez, R. A.,R. L.Chevalier,B.C.Sturgill,D.W.Johns,M. J.Peach,and

R. M.Carey. 1986. Maturation of the intrarenal renin distribution in Wistar-Kyoto rats.J.Hypertens. 4(Suppl. 5):S3 1-S33.

5.Gomez,R.A.,K. R.Lynch,R.L.Chevalier,N.Wilfong,A.D.Everett,

R.M.Carey, and M. J. Peach. 1988. Renin andangiotensinogengeneexpression

inmaturingratkidney. Am.J. Physiol. 254(RenalFluidElectrolytePhysiol.

23):F582-F587.

6.Robillard,J.E.,K. T.Nakamura,M. K.Wilkin,0. J.McWeeny,and G.F.

DiBona. 1987.Ontogenyofrenalhemodynamicresponsetorenalnerve

stimula-tionin sheep.Am.J. Physiol.252(RenalFluidElectrolyte Physiol.

21):F605-F612.

7.Carmines,P. K., and G.A.Tanner. 1983.Angiotensinin the

hemody-namic responsetochronicnephron obstruction.Am.J.Physiol.245:F75-F82. 8.Chevalier,R.L., and M. J. Peach. 1985.Hemodynamiceffectsofenalapril

onneonatalchronicpartialureteral obstruction.KidneyInt. 28:891-898. 9.Chevalier,R.L., and C.E.Jones. 1986. Contribution ofendogenous

va-soactivecompoundstorenal vascular resistance in neonatal chronicpartial

ure-teralobstruction. J.Urol. 136:532-535.

10.Chevalier,R.L., and R.A.Gomez. 1988.Responseofthe

renin-angioten-sin systemtorelief ofneonatal ureteralobstruction.Am.J.Physiol. 255(Renal

Fluid ElectrolytePhysiol. 24):F1070-1077.

11.McDougal,W.S. 1982.Pharmacologicpreservationof renalmassand function in obstructiveuropathy.J.Urol. 128:418-421.

12. Yarger,W. E., D. D. Schocken,and R. H. Harris. 1980. Obstructive nephropathy in therat:possiblerolesfortherenin-angiotensinsystem, prostaglan-dins,and thromboxanes inpostobstructive renalfunction. J.Clin. Invest. 65:400-412.

13.El-Dahr, S. S., R.A.Gomez, G. Khare,M.J.Peach,R. M.Carey,and

R. L.Chevalier. 1990.Expressionof renin and itsmRNAin the adultratkidney

withchronic ureteral obstruction. Am. J. Kidney Dis. 15:575-582.

14.El-Dahr, S. S.,R. A.Gomez,M.S.Gray,M.J.Peach,R.M.Carey,and

R. L.Chevalier. 1990.Insitu localization of reninanditsmRNAin neonatal ureteralobstruction.Am.J. Physiol. 258(Renal Fluid ElectrolytePhysiol.

27):F854-F862.

15.Johnson,E.M., Jr.,F.O'Brien,andR.Werbitt. 1976. Modification and characterization of the permanentsympathectomyproduced by the administra-tion ofguanethidinetonewbornrats.Eur. J.Pharmacol. 37:45-54.

17.Pelayo, J. C., and R. C. Blantz. 1984. Analysis of renal denervationinthe hydropenic rat:interactions with angiotensin II.Am.J.Physio!. 246(Renal Fluid Electrolyte Physiol. 15):F87-F95.

18. Naruse, K., Y. Takii, andT.Inagami. 1981. Immunohistochemical local-ization of renin inluteinizinghormone-producing cells ofratpituitary.Proc.

Natl.Acad. Sci. USA. 78:7579-7583.

19. Takii, Y., A. F. Figueiredo, and T. Inagami. 1985. Application ofimmuno-histochemical methods to the identification and characterization of rat kidney inactive renin. Hypertension (Dallas). 7:237-243.

20.Caccamo, D., C. D. Katsetos, M.M. Herman, A. Frankfurter, V. P. Collins,and L.J.Rubinstein. 1989.Immunohistochemistry ofaspontaneous

murine ovarianteratomawithneuroepithelial differentiation. Lab. Invest. 60:390-398.

21.Chirgwin, J. M.,A.E.Przbyla, R. J. MacDonald, and W. J. Rutter. 1979. Isolation ofbiologicallyactive ribonucleic acid from sources enriched for ribonu-cleases.Biochemistry. 18:5294-5299.

22.Feinberg,A.P., and B.Vogelstein. 1983.Atechnique for radiolabeling DNArestriction endonucleasefragments to highspecific activity. Anal. Biochem. 132:6-13.

23. Rigby, P. W. J., M. Dieckmann, C. Rhodes, and P. Berg. 1977. Labeling

deoxyribonucleicacidtohighspecificactivityin vitro by nick translation with DNA polymerase I. J.Mol.Biol. 113:237-251.

24. Sealy, J. E., and J. H. Laragh. 1977. How todoplasma renin assay.

Cardiovasc. Med.2:1079-1086.

25.Dzau, V. J., J. E. Carlton, and T. Brody. 1987. Sequential changes in renin secretion-synthesis coupling in responseto acutebetaadrenergic stimulation. Clin. Res. 35:604A.(Abstr.)

26. Kopp, U. C., L.A.Olson, and G. F. DiBona. 1984. Renorenal reflex

responses tomechano-and chemoreceptor stimulation in the dog andrat.Am.J. Physiol.246(Renal Fluid Electrolyte Physiol. 15):F67-F77.

27.Buhrle, C. P., E. Hackenthal, U. Helmchen, K. Lackner, R. Nobiling, M.

Steinhausen,and R.Taugner. 1986. The hydronephrotic kidneyas atool for intravitalmicroscopyand in vitroelectrophysiological studies of

renin-contain-ingcells. Lab. Invest. 54:462-471.

28.Huland, H., H. P. Leichtweib,and H.J.Augustin.1980.Effectof angio-tensin II antagonist, alphareceptorblockage, anddenervationon bloodflow reduction in experimental chronic hydronephrosis.Invest.Urol. 18:203-206.

29.Ice,K.S.,K. M.Geary, R.A.Gomez, D. W. Johns, M. J. Peach, andR.M. Carey. 1988. Cell and molecular studies of renin secretion. Clin. Exp.Hyper. A10:1169-1187.

30.Francisco, L. L., L. G. Hoversten, and G. F. DiBona. 1980. Renalnerves

in thecompensatoryadaptationtoureteralocclusion.Am. J.Physiol. 238(Renal FluidElectrolyte Physiol. 7):F229-F234.

31. Anderson, S., T. W. Meyer,H.G. Rennke, and B. M. Brenner. 1985. Control of glomerular hypertension limits glomerular injury in rats withreduced

renalmass.J. Clin. Invest. 76:612-619.