inhibition of convertin g enzyme of the renin...

Inhibition of Converting Enzymeof the Renin-Angiotensin Systemin Kidneys and Hindlegs of Dogs

By James W. Aiken and John R. Vane

ABSTRACTThe activity of angiotensin-converting enzyme in hindlegs and kidneys was

compared in anesthetized dogs. Intra-arterial injections of angiotensin I orangiotensin II to one kidney or one hindleg caused dose-dependent decreases inblood flow in that vascular bed. An inhibitor of angiotensin-converting enzymeactivity did not affect the vasoconstrictor activity of angiotensin II butsubstantially reduced that of angiotensin I. The reduction in vasoconstrictoractivity of angiotensin I during enzyme inhibition was used to calculateconversion of angiotensin I to angiotensin II. There was 40% conversion inhindleg but only 2.1% in kidney. After intra-arterial injection of angiotensin I,bioassay of angiotensin II in the renal or iliac venous blood indicated that60-90% of freshly formed angiotensin II was inactivated before leaving thekidney or hindleg. The results show that converting enzyme activity is muchgreater in hindlegs than in kidneys and that the endogenously formedangiotensin II following intra-arterial injection of angiotensin I is destroyed tothe same extent as intra-arterially injected angiotensin II. Intra-arterial infusionof the enzyme inhibitor increased hindleg blood flow in half of the dogs; renalblood flow increased in only one dog. The inhibitor did not affect vascularresponses produced by norepinephrine, histamine, vasopressin, or prostaglandinF2a but the effects of bradykinin were potentiated.

KEY WORDS angiotensin I angiotensin IIiliac blood flow converting enzyme inhibitorBothrops jararaca peptides

renal blood flowbradykinin

• During intra-arterial infusions of angio-tensin I (AI) or angiotensin II (All), 60-90%of the biological activity of the peptidesdisappears in one passage through renal orhindlimb vascular beds of dogs (1, 2). Ng andVane (1) discussed the possibility that someof the AI which disappeared in kidneys orhindlegs may first have been converted to All

From the Department of Pharmacology, Institute ofBasic Medical Sciences, Royal College of Surgeons ofEngland, Lincoln's Inn Fields, London WC2A 3PN,England.

This study was supported by grants from theMedical Research Council and the Wellcome Trust.Dr. Aiken is a Fellow of the American ThoracicSociety, Medical Section of the National Tuberculosisand Respiratory Disease Association. His presentaddress is Pharmacological Research Division, LillyResearch Laboratories, Eli Lilly and Company,Indianapolis, Indiana 46206.

Received April 26, 1971. Accepted for publicationDecember 20, 1971.

and the All then destroyed before its activitycould be detected in the venous output.However, because AI was much less potentthan All in reducing hindlimb or renal bloodflow when injected intra-arterially, they con-cluded that there was little or no angiotensin-converting enzyme activity in these two pe-ripheral vascular beds and that, as shownpreviously (3), the lungs were the mostimportant site for converting blood-borne AIto AIL Franklin et al. (4), however, foundintra-arterial AI to be more potent in reducingrenal blood flow of dogs than originallyreported by Ng and Vane (1), and furtherstudies (5) suggest that this activity of AI inthe kidney was due to intrarenal conversion ofAIto AIL

Discrete regions of the kidney have convert-ing enzyme activity—for example, juxtaglomu-lar apparatus (6), lymph (7), and isolated

Circulation Research, Vol. XXX, March 1972 263

by guest on May 18, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

264 AIKEN, VANE

strips of renal arteries (8) . Converting en-zyme activity, including that extracted fromkidney (9) and that responsible for theintramural generation of All in isolated stripsof renal and other arteries (8) is inhibited bythe pentapeptide, pyrrolidone carboxylic acid-Lys-Trp-Ala-Pro (10-13). We have, therefore,examined in vivo the vascular effects of AI byusing the inhibitor of converting enzyme tocompare in the kidney and hindleg of dogs theparticipation of local converting enzymeactivity in the vascular responses produced byAI.

MethodsNineteen dogs (14-30 kg) of either sex were

anesthetized with sodium pentobarbital (30mg/kg> iv) a n d ventilated with a positive pressurepump via an endotracheal tube. A midlineincision was made in the abdomen. Blood flow tothe kidneys or hindlegs was measured bynoncannulating electromagnetic flow probesplaced around either the left and right renalarteries or the left and right external iliac arteries.The probes were connected to a two-channelflowmeter. Mean blood pressure was measuredfrom a cannula in a carotid artery with a pressuretransducer. All responses were displayed onmultichannel rectalinear pen recorders.

Blood flow to the kidneys was studied in onegroup of dogs and to the hindlegs in another.Intra-arterial injections into one renal artery orone external iliac artery were made through nyloncatheters with tip diameters of 1 mm. When renalblood flow was studied, a femoral artery wascannulated and the catheter was passed retro-grade up the abdominal aorta. The tip of thecatheter was bent and was directed by hand intothe left renal artery. When external iliac bloodflow was studied, an internal iliac artery wascannulated and the catheter was passed retro-grade to the junction of the abdominal aorta andthe two external iliac arteries. The bent tip of thecatheter was then manipulated into one of theexternal iliac arteries. Thus substances injectedthrough the intra-arterial catheter mixed withblood supplying one hindleg (or kidney) whileblood flow in the contralateral hindleg (orkidney) was monitored as a control for responsesmediated by reflexes or recirculation of intra-arterially injected drugs.

The internal volume of the arterial catheterswas 0.35-0.45 ml. Drugs were injected into thecatheter in less than 0.3 ml saline (0.9% NaClw/v). This was then washed from the dead spaceof the catheter into the arterial blood with 0.5 ml

saline. The inhibitor of converting enzyme wasinfused into the kidney or hindleg through thesame catheter used for intra-arterial injections.During these infusions, drugs were washed inwith saline containing the converting enzymeinhibitor. Intravenous injections were madethrough the cannula in a jugular vein. The dogswere given heparin (300 IU/kg, iv) at thebeginning of the experiment.

In five experiments, the amount of All in therenal or iliac venous blood after injection of AI orAll into the homolateral renal or iliac artery, wasestimated by bioassay using the blood-bathedorgan technique (14). These dogs were givenheparin (1000 IU/kg, iv). Blood from the renalvein was sampled through a catheter passed froma femoral vein up the vena cava. The bent tip ofthe catheter was directed by hand so that it layin, but did not occlude, the left renal vein. Bloodfrom the iliac vein was sampled through acatheter inserted into the femoral vein of theopposite leg and passed to the junction of the leftand right common iliac veins. The bent tip of thecatheter was then directed upstream so that it layin, but did not occlude, the common iliac vein ofthe leg to which intra-arterial injections of AI andAll were made. Blood was withdrawn continu-ously through the venous catheter at a rate of 10ml/min with a roller pump and used to superfuse(15) a rat colon (16), a rat stomach strip (17),and chick rectum (18). Changes in lengths of theassay tissues were detected by transducers fittedwith auxotonic levers. Blood which had super-fused the isolated tissues collected in a smallreservoir and flowed by gravity back into the dogthrough a cannula in the jugular vein.

The combination of assay tissues selected wassuited for a specific assay of All in circulatingblood. The rat colon is contracted by lowconcentrations of All (16, 19); the rat stomachstrip is also contracted by All, but higherconcentrations are needed (20). Both tissues arerelatively insensitive to AI, and since the activityof AI is due mostly to intramural conversion toAll (8), they can be rendered even less sensitiveto AI by infusing the inhibitor of convertingenzyme activity into the superfusion fluid. Thechick rectum was included in the assay systembecause it is insensitive to AI or All, but is moresensitive to prostaglandin E2 than the rat colon(21). Thus, any release of prostaglandin E2 fromthe kidney (22) (which would add to thecontraction of the rat colon and interfere withaccurate assay of All) would be detected by thechick rectum. Angiotensin II is the only knownnaturally occurring substance which contracts therat colon and rat stomach strip but not the chickrectum. However, even if some as yet unidentifiedrat-colon-contracting substance (23) was released

Circulation Research, Vol. XXX, March 1972



ownloaded from


CONVERTING ENZYME IN KIDNEY AND HINDLEG 265

during these experiments, it would not invalidateor significantly alter the interpretation (seeDiscussion).

The All used was l-Asp-/3-amide-5-Val-angio-tensin II.1 The AI used was 5-Ileu-angiotensin I.2

For every mg of AI there should be 0.77 ^imoles.On the basis of optical density measurements at280 nm (reference, ethyl tyrosinate) the prepara-tion contained 0.59 /imoles/mg and by assay onthe rat blood pressure (reference, AI fromSkeggs), 0.46 ^imoles/mg. Thus the preparationcontained, at best, 76% AI and by rat pressorassay, 60%. When we compared it with l-Asp-^3-amide-5-Val-AII for pressor effect after intrave-nous injection in 12 dogs, the AI had 65 ± 4%(mean ± SE) of the activity of the All, afterallowing for the difference in molecular weights.

The activity of AI was compared with that ofAll by calculating the equipotent molar ratio.This is the ratio between the molar doses of AIand All required to produce equal reductions inblood flow.3 Molar amounts of AI were calculatedon the assumption that the AI preparation wasonly 65% pure. All calculations of percentconversion of AI to All are based on molarcomparisons. However individual doses and dose-response curves are expressed as weight of the AI(or All) preparation, with no correction factorapplied. Suitable precautions (24) were taken toprevent bacterial activation or inactivation of thepeptides.

Other substances used were Z-norepinephrine-D-bitartrate, 8-Arg-vasopressin, histamine acidphosphate, prostaglandin F2 a and bradykinin.Doses of norepinephrine and histamine areexpressed as base.

The converting enzyme inhibitor was thesynthetic pentapeptide pyrrolidone carboxylicacid-Lys-Trp-Ala-Pro. This pentapeptide is one ofseveral peptides responsible for the pharmacologi-cal activities of "bradykinin potentiating factor"(BPF) extracted from Bothrops jararaca venom(25). It was first known as V-3-A (11, 26) butthe name BPP5a was suggested for the syntheticsubstance (27). The sample of pentapeptide usedin this work was synthesized at the SquibbInstitute for Medical Research and called SQ20,475. In view of the lack of agreement innaming the substance, we shall refer to it as "thepentapeptide." It was infused at a rate calculatedto give a blood concentration of 1-3 ju.g/ml,

aHypertensin, Ciba.2Schwartz BioResearch.3On the isolated rat uterus, which has little or no

converting enzyme activity (8), our molar dose ratioof AI to All for producing equal size contractions was50:1.


previously shown (8) to produce near maximuminhibition of converting enzyme activity.

ResultsIntra-arterial injections of AI or AH to the

hindleg or kidney produced transient, dose-dependent decreases in the rate of blood flow.The response always began within 5 secondsand lasted from 1 to 6 minutes. Blood flow inthe opposite hindleg or kidney was unaffected,except with high doses of AI and All, whichoften caused a delayed increase or decrease inflow beginning 18 to 25 seconds after theinjection. Infusion of the pentapeptide inhibi-tor of converting enzyme into either the renalor the iliac artery did not significantly affectresponses produced by All, but the effects ofAI were always substantially reduced.

Figure 1 illustrates these effects in the iliacvascular bed. Injections of All at 0.01, 0.03,and 0.1 jug and AI at 0.1 and 0.5 fig into theright external iliac artery caused dose-depen-dent decreases in flow without substantiallyaffecting blood flow in the contralateralhindleg or the blood pressure. Higher doses ofAI (1.5 /Jig) and All (0.3 /ig) injected intra-arterially caused even greater decreases inflow in the right iliac artery and, after a 25-second delay, a transient increase in bloodflow on the left side and in blood pressure.Similar increases in blood pressure and bloodflow were obtained when AI (0.5 /xg) and AH(0.3 jxg) were injected intravenously. Duringintra-arterial infusion of the pentapeptide at250 /Ag/min, doses of AI (0.5 and 1.5 //.g)which previously had caused large reductionsin blood flow now produced only smalldecreases, whereas the effects of AH at 0.01,0.03, and 0.1 jxg were the same as, or slightlygreater than, during the control period.Although local vasoconstriction in the hindlegproduced by intra-arterially injected AI wasantagonized during close intra-arterial infu-sion of the converting enzyme inhibitor, theeffects of an intravenous injection of AI wereunchanged. This suggested that the smallamount of pentapeptide infused directly intothe blood supply of the hindleg had little orno effect on converting enzyme activity in thelungs after being diluted by the total venous



ownloaded from


266 AIKEN, VANE

hp.

Moodflow

inl/irin

flow

left iliac arteryleft iliac artery | ,

right iliac artery

All001

Al01

All

Alas

An0.1

Al1.5

All03

Al05

All0.3i.v.

. S

All003

pentapeptide

All0.1

Al Al0.5 t.5

AH001

•

AlQ5

•

A:O5

Al01

FIGURE 1

At the dots, injections of Al or AH were made through a cannula into the right external iliacartery or (when designated i.v.) through a cannula in the jugular vein. Doses are shown intig. At s saline was infused intra-arterially at 0.5 ml/tnin to the right iliac artery. Pentapeptidewas infused where shown at a rate of 0.5 ml/min and a dose of 250 iig/min. At the break inthe traces 20 minutes of the record is omitted.

return. Twenty minutes after the pentapeptideinfusion had been stopped, Al (0.5 fxg)injected intra-arterially had regained most ofits initial activity.

These results showed that the decreases inblood flow produced by intra-arterial injec-tions of either Al or All were caused by alocal vasoconstriction and not by an actionoccurring after their recirculation. Further-more, the selective antagonism of the effectsof intra-arterially injected Al by the pentapep-tide strongly suggested that Al caused reduc-tion in blood flow in the iliac vascular bedpredominantly through a rapid, local conver-sion to All.

Similar results were obtained when renalblood flow was measured, except that muchhigher intra-arterial doses of Al had to beused to elicit a comparable degree of vasocon-striction. Also contrary to the results in thehindlegs, intravenous Al or All usuallyproduced a bilateral decrease in renal bloodflow. Likewise, after large intra-arterial dosesof Al or All, there was a delayed decrease inflow in the contralateral kidney.

Both in renal and iliac vascular beds, therate of decline in blood flow produced by Alwas slower than that produced by All. The

latency from injection to the onset of the fallin flow was similar for both peptides(2-5 seconds), but the peak reduction in flowafter All was reached within 10 to 20 seconds,whereas that for an equiactive dose of Al took3-6 seconds longer. Also, the duration of theresponse to equiactive doses of Al and Allwas usually greater for the decapeptide (inFig. 1 compare Al at 0.5 fxg to All at 0.1

The effects of Al and All were comparedquantitatively in these two vascular beds bydetermining, in each experiment, the equipo-tent molar ratios of Al to All before andduring inhibition of converting enzyme. Thepeak vasoconstrictor effects of different sub-maximal intra-arterial injections of Al wereeach bracketed between smaller and largereffects of doses of All. In this way, severaldeterminations of relative activity of Al to Allwere made in each experiment. The equipo-tent molar ratios were calculated and themeans of these ratios are shown in Table 1.The changes in the equipotent molar ratiosinduced by inhibition of converting enzymeactivity were used to estimate the apparentconversion of Al to All. For example, in thehindleg a mean of 2.3 moles of Al were




ownloaded from



TABLE 1

Equipotent Molar Ratio (Angiotensin I: Angiotensin II)

Vascularbed

Control(Mean ± BE)

Duringpentapeptide(Mean ± SE)

Calc percentAI converted to All

Mean (range)

RenalIliac

27 ± 2.82.3 ± 0.2

6128

124.6

2.1 (1.4-2.8)39.9 (31-51)

equivalent to 1 mole of All (or 100 molesAI = 43.5 moles All). During inhibition ofconverting enzyme activity, 28 moles of AIwere equivalent to 1 mole of All (or 100moles AI = 3.6 moles All). Thus, for every100 moles of AI, 39.9 (43.5-3.6) wereconverted to All; i.e., there was 39.9%conversion. In the kidney, there was only 2.1%conversion (100 moles of AI = 3.7 moles Allbefore, and 1.6 moles during, convertingenzyme inhibition). In one dog, flow wassimultaneously measured in a renal and aniliac artery and catheters were inserted so thatintra-arterial injections could be made toeither vascular bed. Similar results wereobtained.

Dose-response curves were plotted from thecombined results of these experiments. The

reductions in blood flow induced by AI or Allwere all expressed as percent decreases, takingrate of flow just before injection as 100%. Theresults are summarized in Figure 2. In kidneyand hindleg, the dose-response curves for AIand All were parallel during the controlperiod. During pentapeptide infusion, theactivity of AI was decreased, but that of Allwas not; indeed, there may have been slightpotentiation of the effects of the smaller dosesof All.

Although the effect of inhibiting activity ofconverting enzyme on the activity of AI in thekidney was qualitatively similar to that in thehindleg, there were important quantitativedifferences. Whereas All had the samepotency in hindleg and kidney (51% and 53%reduction in flow at 0.1 /xg), AI was ten times

100 renal

•r

001 pg 1.0 KX 0.01 pg CM 1.0 10.

dose (log scale)

FIGURE 2

Dose-response relationship from AH controls (u), AI controls (•), All during infusion ofpentapeptide (1-3 ng/ml) (o), and Al during pentapeptide (o). Mean and SE for 3-7 responsesat each dose level are shown. When no SE is shown, the point is the mean of 2 responses.In calculating the means, no more than one response at each dose level was obtained fromeach dog (iliac, 7 dogs; renal, 6 dogs). The lines were drawn to fit the computed best linethrough the points.




ownloaded from


268 AIKEN, VANE

more potent in hindleg than in kidney (50%reduction in hindleg flow at 0.5 fig comparedwith 50% reduction in renal flow at 5 fig).

Thus, both from the molar ratios calculatedfrom the individual experiments and from thedose-response curves, it is evident that AI ismuch more potent in the hindleg than in thekidney and that this difference in potency ofAI in the two vascular beds is due to the muchhigher activity of converting enzyme inhindlegs than in kidneys.

VENOUS OUTPUT OF ANGIOTENSIN II-LIKE ACTIVITY

Ng and Vane (1) found that only 10^0% ofthe decapeptide's biological activity survivedpassage through kidneys or hindlegs. How-ever, they could only roughly estimate wheth-er the activity remaining in the venousblood was due to AI or a mixture of AI andAll. The availability of the inhibitor ofconverting enzyme activity made it possiblefor us to clarify this point.

In four other dogs, All-like activity in theiliac vein, after intra-arterial injection of AI orAll, was estimated by the blood-bathed organtechnique. Pentapeptide (15 fig/min) wasinfused into the blood superfusing the assaytissues to give a concentration of 1.5 fig/ml.This prevented any conversion of AI to Alleither in the superfusing blood or in the assaytissues themselves. The amount of pentapep-tide recirculating to the iliac vascular bed wasinsufficient to have any detectable effect on AIactivity.

Injection of All into the iliac artery (0.1 or0.3 fig) led to contractions of the rat colonsbathed in the venous outflow. Assay of the Alloutput confirmed previous results (1) andshowed that 69 ± 6.3% of the lower dose and62 ± 5% of the higher dose was inactivated inthe hindleg. During pentapeptide infusion,there was no change in inactivation of All(disappearance of 71 ± 5% of the lower and62 ± 11% of the higher dose).

AI was injected in amounts (0.5 and 1.5fig) calculated to give vascular effects equiva-lent to these two doses of AIL The activityemerging from the iliac vein was equivalent to0.06 fig of AH at the smaller dose and 0.18 figat the higher one. During pentapeptide

infusion, this activity was substantially re-duced; none was detectable after the lowerdose (limit of assay: 16 ng of All) and 0.04fig of All after the higher dose. These resultsshowed that a substantial portion of the AII-like activity in the iliac vein after intra-arterialinjection of AI was, indeed, due to All formedin the hindleg, and this All represented13.5-15% conversion of the total AI dose. Therewas 60-70% disappearance of AI in the hindleg,estimated by comparing the effects induced bythe remaining AI on the blood pressure (afterrecirculation and conversion in the lungs)with intravenous All injections. This waswithin the range (60-90%) found previouslyusing infusions rather than injections (1). Innone of these experiments was there evidencefor release of prostaglandinlike activity.

In the renal vein of one dog All-like activitywas measured in the same way. There was 75%inactivation of All injected into the renalartery. This degree of inactivation was un-affected by the pentapeptide. After injectionof 2 fig AI, the activity emerging in the renalvein was equivalent to only 30 ng of All or1.5% conversion of the injected dose. Infusionof pentapeptide reduced the activity whichappeared in the renal vein after intra-arterialinjection of AI, confirming that some of theactivity was due to All formed in the kidney.

Further experiments on output of All-likeactivity in the renal vein were not attempted,for the amounts detectable were so small as tobe near the limits of the assay procedure.Furthermore, contractions of the other assaytissues suggested that there was a release ofprostaglandinlike material after AI injec-tion.

OTHER EFFECTS OF THE PENTAPEPTIDE

Infusion of the pentapeptide at 250 fig/mininto the renal artery (rate of blood flow 80-210ml/min) had no direct effect on blood flow in6 of the 7 dogs. In one dog there was agradual increase in flow, amounting to 20%over a period of 25 minutes. In the iliacvascular bed, blood flow varied from 90 to 250ml/min and in 5 of the 12 dogs thepentapeptide gradually increased flow by13-24%. In the other 7 dogs there was either no




ownloaded from



change in flow or no increase greater than thatproduced by an intra-arterial infusion of salineat the same rate (0.5 ml/min). The experi-ment of Figure 1 was on one of the 5 dogs inwhich the pentapeptide gradually increasedblood flow. During the first half of the tracethe base-line flow in both the left and rightiliac arteries followed the same pattern.However, during pentapeptide infusion, theleft iliac flow gradually increased, whereas theright iliac flow maintained a constant baseline.

Pentapeptide in amounts (1-3 fig/ml) suffi-cient to antagonize substantially the vasculareffects of AI did not affect the changes iniliac blood flow produced by intra-arterialinjections of histamine, norepinephrine, vaso-pressin, prostaglandin F2a, or AIL Figure 3shows some of these effects. Intra-arterialinjections of histamine (0.01, 0.03, and 0.1jug) produced dose-dependent increases inflow and All (0.05 and 0.3 fig), AI (0.5 and1.5 fig), and norepinephrine (0.1 and 0.3 fig)all produced dose-dependent decreases in iliacblood flow. During infusion of pentapeptide,

350

•| 300

^ 2 5 0

J 200- 150

8 ioo

* 50

0

350

E 300

^ 2 5 0

* 200

= 150

| 100

* 50

0

H H H.01 .03 .1

A I.5

AI1.5

All.3

NE.1

NE

pentapeptide

H H.03 .1

All.05

AI AI.5 1.5

NE NE.1 .3

AI.5

FIGURE 3

Blood flow in an external iliac artery of a 17-kg femaledog. At the dots injections of histamine (H), AI, AH,and norepinephrine (NE) were made into the externaliliac artery. Doses are shown in ng. Bottom trace is acontinuation of the top trace. Pentapeptide was infusedat 250 ng/min i.a. to the hindleg.


only the effects of AI were reduced; all theother substances retained their activity.

Prostaglandin F2a, when injected intra-arterially to the hindlegs (0.3-3 fig), produceda transient increase in flow which lasted to 5-10seconds, followed by a prolonged decrease inflow lasting from 3 to 5 minutes. Both theinitial vasodilatation and secondary vasocon-striction increased with increasing doses. Thepentapeptide (1-3 /Ag/ml blood) did notchange these effects of prostaglandin F2a.Similarly, decreases in blood flow produced byvasopressin (10 and 30 mU ia) wereunaffected by the pentapeptide.

The pentapeptide potentiates several of theactions of bradykinin, including the contrac-tile effect on isolated guinea pig ileum (26),the increased capillary permeability in rat skinand the decrease in the blood pressure of rats(27). We found that bradykinin (1-30 ng)injected into the iliac artery caused dose-dependent increases in flow. In the presenceof the pentapeptide (1-4 jug/ml ia), brady-kinin was five times more potent than beforepentapeptide. Figure 4 shows that 2 ngbradykinin produced only a slight increase inflow but 5 and 20 ng produced larger ones.Thirty ng histamine produced an increase inflow approximately equivalent to 3 ng brady-kinin. Neither the histamine nor the higher

350

£ 250

o

150

8n

50

JL5min

Xi pentapeptide

Bk Bk

2 5

Bk

20

H Bk

30 2

FIGURE 4

Bk Bk

2 1

H

30

Bk

2

Blood flow in an external iliac artery of a 17-kg fe-male dog. At the dots, intra-arterial injections ofbradykinin (Bk) and histamine (H) were made towardthe hindleg. Doses are expressed in ng. Pentapeptidewas infused at 250 ng/min into the external iliacartery.



ownloaded from


270 AIKEN, VANE

doses of bradykinin altered the effects of asubsequent dose of 2 ng bradykinin. In thepresence of pentapeptide (250 fig/min) theeffect of 2 ng bradykinin was substantiallypotentiated (now 2 ng bradykinin was equiva-lent to a response produced by approximately10 ng before pentapeptide), but the effect of30 ng histamine was unchanged.

Discussion

Previous work from this laboratory hasshown that in the pulmonary vascular bedthere is high conversion of AI and noinactivation of All. This has been extensivelyconfirmed (8, 12, 23, 28-32) in several species.In other vascular beds, there is an overallinactivation of about 70% of AI or AH (1-3,28). Because of the efficient inactivatingmechanisms in peripheral vascular beds, it hasbeen difficult to determine in vivo how muchconversion of AI to All takes place prior todestruction. However, the availability of aninhibitor of converting enzyme has madepossible detailed analysis of local conversionof Alto All (8).

Our present results show that AI injectedinto the arterial blood supplying either hind-leg or kidney of the dog is partially convertedto All and that the All generated induceslocal vasoconstriction. Converting enzyme inplasma has too low an activity to account forthese results in the hindleg (1, 3, 29) so theenzyme must be fixed in the tissue. Isolatedrenal and femoral arteries have convertingenzyme activity (8), so the enzyme may belocated within the vessel wall. Conversion inthe kidney, however, is so low that it may allbe due to converting enzyme in plasma; in thedog, Oparil et al. (29) found 3% conversion inwhole blood within 10 seconds. Conversion inthe hindleg and kidney has been demonstrat-ed by the reduction in local vasoconstrictoractivity of AI when converting enzymeactivity is inhibited and by the reductionduring enzyme inhibition of the venous outputof AH-like activity after intra-arterial injectionof AI. Both methods indicate that conversionof AI to All in the hindleg is 10 to 20 timesgreater than that in the kidney.

In our experiments, assay of All was easier

in hindleg venous blood, where it was presentin higher concentrations, than in renal venousblood. Even though the immediate vasocon-strictor activity of AI in the hindlegs indicatedup to 40% conversion (Table 1), the venousoutput of All-like activity represented only14% of the injected dose of AL This agreeswell with a 12% apparent conversion in venousblood from dog hindlegs found by Oparil etal. (29), although they believed this to be dueto plasma conversion. Thus conversion of AIto All probably takes place at a site whichallows local vasoconstrictor activity, but there-after there is about 70% inactivation of thenewly formed All before it reaches the iliacvein. This interpretation is supported by thefact that the same percent of injected All isinactivated on passage through the hindlegs.A similar process may also occur in the kidney,where there is a qualitatively similar disparity:the immediate vasoconstrictor activity of AIrepresents 2.1% conversion, but less All-likeactivity appeared in renal venous blood. Theinactivation of newly formed All in thekidney was probably greater than our experi-ments indicate, since the observed release ofprostaglandinlike material from the kidneywould have led to an overestimation of theamount of All in the renal venous blood.Hence, there was probably even less than 1.5%of the intra-arterially injected AI in the formof All in the renal venous blood. Noprostaglandin release was seen in the experi-ments on hindlegs. Although the mechanismof inactivation of AI and All in dog kidneyand hindleg is unknown, inactivation of All inrat kidney is due mostly to rapid metabolismand not to tissue sequestration (33).

Although an intrarenal role for the renin-angiotensin system is not excluded by ourresults, it seems unlikely that the system has alocal role in the kidney for regulating bloodflow. The very small proportion of blood-borne AI converted to All in the dog kidney(2.1%) compared to the hindleg (40%), theintestine (25%) (34) and the lungs (50-100%)suggests that the kidneys may be exceptionalfor their low converting enzyme activity.Release of renin from the kidney could then

Circulation Research, Vol. XXX. March 1972



ownloaded from



take place without an immediate generation ofAIL The relative absence of convertingenzyme in the kidney blood vessels would be aprotective mechanism, and the renal vascularbed, which is highly sensitive to All, wouldonly receive All formed by converting en-zyme in the lungs, after the renin content ofrenal venous blood had been diluted with therest of the cardiac output. Indeed, the kidneymay have a dual mechanism for protectingitself from the vasoconstrictor effects of AILFirst, the paucity of converting enzymeactivity in the renal vasculature would mini-mize local generation of All during reninrelease. Second, infusions of low concentra-tions of All into dog kidney release aprostaglandin E2-like substance (22) whichantagonizes the local vasoconstrictor action ofAll (35).

The activity of AI cannot be completelyabolished by the converting enzyme inhibitor(8), suggesting that AI has a small (1^1%,relative to All) but significant inherentactivity. Others have assumed that AI iscompletely inactive until converted to All(4). Whichever view is correct, whenever AIhas more than 5% (on a molar basis) of theactivity of All in any organ or isolated tissue itis likely to be due to local conversion to AIL

In dog kidney, even when the inherentactivity is disregarded, the maximum possibleconversion of AI to locally active AH de-scribed here and by Di Salvo et al. (36) andNg and Vane (1) is very much lower than the19% claimed by Franklin et al. (4). There areseveral possible explanations for the discrep-ancy between the 19% of Franklin et al. andthe lower figures. First, implicit in theircalculation of the relative potency of AI to Allis the assumption that the effects of a drug arelinearly proportional to the dose. It is acornerstone of bioassay that dose-responsecurves are sigmoid (37); a linear relationshipexists between logio and the response. How-ever, even if one plots the reductions in renalblood flow produced by AI and All in Figure1 of their published experiments (4) againstlog dose of the angiotensins, and assumes thedose response curves for AI and All to be


parallel (which they are in our experiments),then the percent conversion found by Franklinet al. in dog kidney is still relatively high(14-17%).

Another contribution to the discrepancymay be the difference in potency of the AIand All of Franklin et al. compared to ours.From the data in their Figure 1 (4) it appearsthat their AI was more potent and All lesspotent than ours. These differences in potencywould be cumulative in any calculation ofconversion based on relative potency of AIand AIL They did not state the relativeactivity of their AI and All solution, either onisolated organs or after intravenous injection;so the size of this contribution to thediscrepancy is difficult to assess. However, anydifferences in results due to contamination ofAI or All with impurities could be avoided bymeasuring and publishing the purity of thepeptides (24) and by comparing their activityto that of international reference standardswhich are now available (38).

Our results on the kidney are also somewhatat variance with those of Oparil et al. (29).They showed that following the intra-arterialinjection of AI, renal venous blood contained7-9% of the injected dose in a form whichreacted with antibodies for AIL However, it ispossible that this does not represent 7-9%conversion, for the degree of reactivity of theAH antibodies with other substances invenous blood was not known, and this cansometimes be as high as 70% (39).

A further possibility is that convertingenzyme activity varies substantially fromlaboratory to laboratory and from time to timewith factors as yet uncontrolled. Many exter-nal factors can influence enzyme activity,including such chemicals as laboratory insecti-cides (40). A low salt diet reduces convertingenzyme activity (5). Some uncontrolled factormay also be responsible for the discrepanciesbetween the relatively high conversion inhindlegs found here and the lack of conver-sion in hindlegs found by Ng and Vane (1).

We have previously suggested that convert-ing enzyme is located in vascular tissue (8),and fractionation studies on lung tissues (10,



ownloaded from


272 AIKEN, VANE

41) are consistent with this view. A reninlikeenzyme has been extracted from vasculartissue (42, 43). The contiguity of convertingenzyme and renin in vascular tissue mightallow intramural generation of All from reninsubstrate, thereby selectively contributing tothe local control of vascular tone in thosevessels sensitive to AIL

The rise in hindleg blood flow, which wasoften associated with infusions of convertingenzyme inhibitor, might also indicate a localrole for the renin-angiotensin system, inhibi-tion of local formation of All leading tovasodilatation. However, the inhibitor ofconverting enzyme activity also potentiatesthe actions of bradykinin, mainly by prevent-ing its destruction (27). Thus the gradualvasodilatation might also have been due topotentiation of circulating or locally formedbradykinin.

Despite the presence of converting enzymeactivity in some other vascular beds, that inthe lung must play a predominant role in theconversion of blood-borne AI. The lack ofinactivation of AH in the lungs and highinactivation elsewhere, also suggests that lungconversion, leading to arterial concentrationsof All, has a physiological function. Mostestimates of lung conversion vary from50-100%. Recently, Barrett and Sambhi (30)have shown that lung conversion depends onthe concentration of AI, the lower theconcentration the more the conversion. At thelowest dose which they used, conversion was80% and the curve was quickly approaching100%. Thus, when there is renin in thecirculation, the pulmonary vein delivers pre-dominantly AH to the left ventricle. Whetherconverting enzyme in vascular beds such asthe hindlegs or intestine has an important rolein contributing to the total amount of AHavailable for local receptors will depend onthe rate at which renin generates AI in thearterial circulation between the left ventricle,and the vascular bed concerned.

AcknowledgmentsWe would like- to thank Mr. M. Peck for skilled

technical assistance and the Squibb Institute, forMedical Research, the Upjohn Company and Parke-Davis & Company for gifts of compounds.

References1. Nc, K.K.F., AND VANE, J.R.: Fate of angiotensin I

in the circulation. Nature 218:144-150, 1968.2. HODGE, R.L., Nc, K.K.F., AND VANE, J.R.:

Disappearance of angiotensin from the circula-tion of the dog. Nature 215:138-141, 1967.

3. Nc, K.K.F., AND VANE, J.R.: Conversion ofangiotensin I to angiotensin II. Nature 216:762-766, 1967.

4. FRANKLIN, W.G., PEACH, M.J., AND GILMORE,

J.P.: Evidence for the renal conversion ofangiotensin I in the dog. Circ Res 27:321-324,1970.

5. GILMORE, J.P., MERRILL, J., AND PEACH, M.J.:

Further evidence for the renal conversion ofangiotensin (abstr.). Fed Proc 30:449, 1971.

6. GRANGER, P., DAHLHEIM, H., AND THURAU, K.:

Renin-Angiotensinase- und converting enzym-Bestimmungen an einzelnen mikrodisseziertenjuxtaglomerularen Apparaten und Teilen besNephrons. Pfluegers Arch 312:R87-88, 1969.

7. HORKEY, K., ROJO-ORTEGA, J.M., RODRIGUEZ, J.,

BOUCHER, R., AND GENEST, J.: Renin, reninsubstrate, and angiotensin I converting enzymein the lymph of rats. Am J Physiol 220:307-311, 1971.

8. AKEN, J.W., AND VANE, J.R.: Renin-angiotensin

system: Inhibition of converting enzyme inisolated tissues. Nature 228:30-34, 1970.

9. CUSHMAN, D.W., AND CHEUNG, H.S.: Concentra-

tions of angiotensin-converting enzyme intissues of the rat. Biochim Biophys Acta250:261-265, 1971.

10. BAKHLE, Y.S.: Conversion of angiotensin I toangiotensin II by cell-free extracts of dog lung.Nature 220:919-921, 1968.

11. FERREIHA, S.H., GREENE, L.J., ALABASTER, V.A.,

BAKHLE, Y.S., AND VANE, J.R.: Activity of

various fractions of bradykinin potentiatingfactor against angiotensin converting enzyme.Nature 225:379-388, 1970.

12. NG, K.K.F., AND VANE, J.R.: Some properties ofangiotensin converting enzyme in the lung invivo. Nature 225:1142-1144, 1970.

13. KRIEGER, E.M., SALCADO, H.C., ASSAN, C.J.,

GREENE, L.J., AND FERREIRA, S.H.: Potential

screening test for detection of overactivity ofthe renin-angiotensin system. Lancet 1:269-271, 1971.

14. VANE, J.R.: Use of isolated organs for detectingactive substances in the circulating blood. Br JPharmacol Chemother 23:360-373, 1964.

15. GADDUM, J.H.: Technique of supervision. Br JPharmacol Chemother 8:321-326, 1953.'

16. RECOLI, D., AND VANE, J.R.: Sensitive methodfor the assay of angiotensin. Br J PharmacolChemother 23:351-359, 1964.




ownloaded from



17. VANE, J.R.: Sensitive method for the asasy of 5-hydroxytryptainine. Br J Pharmacol Chemother12:344-349, 1957.

18. MANN, M., AND WEST, G.B.: The nature ofhepatic and splenic sympathin. Br J PharmacolChemother 5:173-177, 1950.

19. REGOLI, D., AND VANE, J.R.: Continuousestimation of angiotensin formed in thecirculation of the dog. J Physiol (Lond)183:513-531, 1966.

20. HODGE, R.L., LOWE, R.D., AND VANE, J.R.:Effects of alteration of blood-volume on theconcentration of circulating angiotensin inanaesthetized dogs. J Physiol (Lond) 185:613-626, 1966.

21. FERREIRA, S.H., AND VANE, J.R.: Prostaglandins:Their disappearance from and release into thecirculation. Nature 216:868-873, 1967.

22. MCGIFF, J.C., CROWSHAW, K., TERRAGNO, N.A.,AND LONIGRO, A.J.: Release of a prostaglandin-like substance into renal venous blood inresponse to angiotensin II. Circ Res 27 (suppl.I ) : 1-121-130, 1970.

23. TURKER, R.K., YAMAMOTO, M., BUMPUS, F.M.,AND KHAIRALLAH, P.A.: Lung perfusion withangiotensin I and II. Circ Res 28:559-567,1971.

24. VANE, J.R.: Purity and stability of syntheticpeptides such as angiotensins and kinins.Nature 230:382, 1971.

25. FERREIRA, S.H.: Bradykinin-potentiating factor(BPF) present in the venom of Bothropsjararaca. Br J Pharmacol Chemother 24:163-169, 1965.

26. FERREIRA, S.H., BARTELT, D.C., AND GREENE,L.J.: Isolation of bradykinin-potentiating pep-tides from Bothrops jararaca venom. Biochem-istry 9:2583-2593, 1970.

27. STEWART, J.M., FERHEIRA, S.H., AND GREENE,L.J.: Bradykinin potentiating peptide PCA-LYS-TRP-ALA-PRO: Inhibitor of the pulmo-nary inactivation of bradykinin and conversionof angiotensin I to II. Biochem Pharmacol20:1557-1567, 1971.

28. B A K H L E , Y.S., REYNARD, A.M., A N D V A N E , J.R.:

Metabolism of the angiotensins in isolatedperfused tissues. Nature 222:956-959, 1969.

29. OPARIL, S., SANDERS, C.A., AND HABER, E.: In

vivo and in vitro conversion of angiotensin I toangiotensin II in dog blood. Circ Res26:591-599, 1970.

30. BARRETT, J.D., AND SAMBHI, M.P.: Pulmonaryactivation and degradation of angiotensin I:Dual enzyme system. Res Comm Chem PatholPharmacol 2:128-145, 1971.

31. BIRON, P., CAMPEAU, L., AND DAVID, P.: Fate ofangiotensin I and II in the human pulmonarycirculation. Am J Cardiol 24:544-547, 1969.

32. BIRON, P., AND HUGGINS, C.G.: Pulmonaryactivation of synthetic angiotensin I. Life Sci7:965-970, 1968.

33. LEARY, W.P., AND LEDINGHAM, J.G.: Removal ofangiotensin by isolated perfused organs of therat. Nature 222:959-960, 1969.

34. DISALVO, J., AND MONTEFUSCO, C.B.: Conversionof angiotensin I to angiotensin II in the caninemesenteric circulation. Am J Physiol, 221:1576-1579, 1971.

35. AIKEN, J.W., AND VANE, J.R.: Blockade ofangiotensin-induced prostaglandin release fromdog kidney by indomethacin (abstr.). Pharma-cologist 13:293, 1971.

36. Di SALVO, J., MENTA, M., MONTEFUSCO, C , ANDPETERSON, A.: Intrarenal conversion of angio-tensin I to angiotensin II in the dog. Circ Res29:398-406, 1971.

37. BURN, J.H.: Biological Standardization. Oxford,England, Oxford Medical Publications, 1937.

38. BANGHAM, D.R., CALAM, D.H., PARONS, J.A., ANDROBINSON, C.J.: Purity of synthetic peptidepreparations. Nature 232:631, 1971.

39. CAIN, M.D., CATT, K.J., AND COUGHLAN, J.P.:Immunoreactive fragments of angiotensin II inblood. Nature 223:617-618, 1969.

40. CONNEY, A.H.: Pharmacological implications ofmicrosomal enzyme induction. Pharmacol Rev19:317-366, 1967.

41. SANDER, G.E., AND HUGGINS, C.G.: Subcellulardistribution of angiotensin I converting enzymein rabbit lung. Nature 230:27-29, 1971.

42. DENGLER, J.: Uber einen reninartigen Wirkstoffin Arterienextrakien. Arch Exp Pathol Pharm227:481-487, 1956.

43. GOULD, A.B., SKECGS, L.T., JR., AND KAHN, J.R.:Presence of renin activity in blood vessel walls.J Exp Med 119:389-399, 1964.




ownloaded from


JAMES W. AIKEN and JOHN R. VANEof Dogs

Inhibition of Converting Enzyme of the Renin-Angiotensin System in Kidneys and Hindlegs

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1972 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.30.3.2631972;30:263-273Circ Res.

http://circres.ahajournals.org/content/30/3/263World Wide Web at:

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

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research 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 about this process is available in the

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

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



ownloaded from

http://circres.ahajournals.org/content/30/3/263

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


inhibition of convertin g enzyme of the renin...

Documents