Angiotensin Converting Enzyme Inhibition Prevents the Increase in Aortic Collagen in Rats

74

Angiotensin Converting Enzyme InhibitionPrevents the Increase in Aortic Collagen in Rats

Pierre Albaladejo, Herv6 Bouaziz, Micheline Duriez, Peter Gohlke, Bernard I. Levy,Michel E. Safar, Athanase Benetos

Abstract Four groups of 4-week-old spontaneously hyperten-sive rats were treated during 4 months with the angiotensinconverting enzyme inhibitor quinapril at 1 mg/kg per day (Ql) or10 mg/kg per day (Q10), hydralazine at 15 mg/kg per day (H), orplacebo (P). In the first set of experiments, blood pressure wasmeasured in conscious rats, and plasma and aortic angiotensinconverting enzyme activities were evaluated. In the second set ofexperiments, histomorphometric parameters of the thoracicaorta were evaluated. Mean blood pressure was lower in the Q10and H groups (136±16 and 149±11 mm Hg) compared with theP group (190±23 mm Hg) (P<.01). The Ql group showed meanblood pressure values (171 ±15 mm Hg) lower than the P group(P<.05) but significantly higher than the Q10 and H groups(P<.01 and P<.05, respectively). Aortic medial cross-sectionalarea was significantly lower in the H and Q10 groups (455 ±61and 487±57 x 103 jun2) than in the P group (636±72 x 103

lim2)

(/><.001). In the Ql group, medial cross-sectional area was lower(550±65 xlO3 /im2) than the P group (P<.01) but higher than

the Q10 and H groups. In the Q10 and Ql groups, the collagencontent of the aortic media was significantly lower than in the twoother groups (P<.01) (Ql, 15.3±4.6; Q10, 14.3±4.0; H,18.6±5.O, P, 21.9±4.7 xlO3 /im2/mm of aorta). Aortic angioten-sin converting enzyme activity was inhibited by approximately60% in both groups treated with quinapril, whereas plasmaangkrtension converting enzyme activity was reduced only in theQ10 rats. These results show that, whereas both hydralazine andquinapril prevented the development of aortic hypertrophy in apressure-dependent manner, the prevention of the increase inaortic collagen was observed only after quinapril treatment. Thislatter effect of angiotensin converting enzyme inhibition was notrelated to the blood pressure reduction but was associated withthe reduction of aortic and not plasma converting enzyme.(Hypertension. 1994;23:74-8Z)

Key Words • collagen • hypertrophy • hydralazine •rats, inbred SHR

Large-artery walls become thicker and stiffer inhypertension.1 Together with vascular smoothmuscle cell hypertrophy, an increase in medial

extracellular matrix involving principally collagen isalso a major pathological and biochemical finding inhypertension.2*3

A large amount of evidence shows that structuralalterations of the arterial wall in hypertension may bepartly attributed to elevated pressure and wall stress.4-5

Thus, many investigators suggest that hypertrophy is anadaptive mechanism to normalize wall stress duringhypertension.67 However, other authors have observedthat vascular hypertrophy can be induced by meansother than high blood pressure factors and may itself bethe cause of hypertension.8"10 Increasing evidence sug-gests a major role of the renin-angiotensin system andespecially angiotensin II (Ang II) in these nonpressuremechanisms9-11: (1) components of the renin-angioten-sin system are present in the vascular wall, where Ang IIgeneration has been demonstrated,12-13 (2) Ang II in-duces hypertrophy and collagen production in vascularsmooth muscle cell cultures,1416 and (3) in vivo admin-istration of nonpressor doses of Ang II induces arterialthickening.17

Received March 16, 1993; accepted in revised form September22,1993.

From INSERM U337, Broussais Hospital, Paris, France (A.B.,P.A., H.B., M.E.S.); the Department of Pharmacology, Universityof Heidelberg (Germany) (P.G.); and INSERM U141, Laribois-iere Hospital, Paris, France (M.D., B.I.L.).

Correspondence to Athanase Benetos, Internal Medicine I,Broussais Hospital, 96 rue Didot, 75015 Paris, France.

Angiotensin converting enzyme (ACE) inhibitorsseem to be more effective than other drugs in theregression or prevention of these arterial alter-ations.9-1820 These observations are consistent with thepossibility that ACE inhibitors have specific effects onvascular structure independent of their antihypertensiveaction.2123 However, the respective roles of blood pres-sure reduction and/or ACE inhibition on the preventionof arterial structural alterations remain unclear. Inaddition, smooth muscle cells, collagen, and elastin maybe influenced in different manners by blood pressurereduction, ACE inhibition, or both.

Thus, the aim of this study was to evaluate therespective roles of a decrease in blood pressure andACE inhibition on the prevention of aortic structuralalterations in spontaneously hypertensive rats (SHR).Therefore, hypertensive rats were treated for 4 monthswith two different doses of the ACE inhibitor quinapril(1 and 10 mg/kg), with only the higher dose having asubstantial antihypertensive effect. We have comparedthe effects of these treatments with those of hydralazineat antihypertensive doses equipotent to the higher doseof quinapril.

MethodsAnimals

Four-week-old male SHR (Iffa-Credo, L'Abresle, France)were randomly allocated into four groups. They were housed fiveto seven per cage in our animal room (temperature, 20° to 22°C;humidity, 55% to 65%; 12-hour light/dark cycle), fed a standarddiet (0.13 mEq/g Na+ and 0.205 mEq/g K+), and had free accessto tap water.

by guest on June 24, 2015http://hyper.ahajournals.org/Downloaded from

http://hyper.ahajournals.org/

Albaladejo et al Effects of Quinapril on Aortic Structure 75

TreatmentsThe different treatments were administered daily by gavage

during 16 weeks. Groups Q10 and Ql received, respectively, 10and 1 mg/kg per day of quinapril (Parke-Davis) diluted insaline solution; group H received 15 mg/kg per day of hydral-azine (Sigma Chemical Co, St Louis, Mo) diluted in salinesolution; and group P received saline solution (placebo). Thevolumes administered were similar in the four groups and didnot exceed 0.5 mL/100 g body wt per day.

The hydralazine dose was found in a short study to exert asimilar hypotensive effect as 10 mg/kg quinapril. For thatstudy, 10 SHR were divided into two groups of 5 rats each. Thefirst group received quinapril at a dose of 10 mg/kg for 3 days,and the second group received single doses of 5, 15, and 25mg/kg per day of hydralazine on days 1, 2, and 3, respectively.Blood pressure was noninvasively measured by a tail-cuffmethod 24 hours after each drug administration. After thisprocedure the dose of 15 mg/kg hydralazine was chosenbecause it had an antihypertensive effect similar to that of 10mg/kg quinapril. After the determination of this dose, ratswere treated for 2 weeks. At the end of this period the animalswere cannulated with a femoral catheter. Mean blood pressure(MBP) was then evaluated invasively in conscious rats 3 and 24hours after the last drug administration. The differencesobserved between the two groups did not exceed 10 mm Hg forboth peak and trough antihypertensive effect.

The two doses of quinapril (1 and 10 mg/kg) decrease bymore than 90% plasma and aortic ACE activities after a singleadministration,24-23 whereas only the higher dose has a signif-icant antihypertensive effect. In a pilot study, 1 mg/kgquinapril had a weak antihypertensive effect. Indeed, the peakand trough effect does not exceed 15 mm Hg compared withcontrol animals. The dose of 10 mg/kg quinapril is the lowerdose of this drug with a maximal antihypertensive effect atpeak and trough.18 The main aim of this study was to evaluatethe effect of different blood pressure levels on structuralarterial parameters. We chose to measure blood pressure 20 to24 hours after the last drug administration because the delayand duration of the peak effects especially after chronictreatment are variable and short compared with the globaleffect on blood pressure throughout the day. Therefore, thepressure levels evaluated 20 to 24 hours after the last treat-ment may better reflect the evolution of blood pressurethroughout the treatment period.

Study DesignDuring the treatment period, systolic blood pressure (SBP)

was measured noninvasively by a tail-cuff method at 8, 12, and16 weeks of age.26 Five consecutive measurements were per-formed each time, and the mean of these values was consid-ered. When the difference was higher than 15%, the measure-ments were started again.

At the end of the treatment period, two sets of experimentswere performed in separate animals. The first included eval-uation of invasive blood pressure, the pressor response toangiotensin I (Ang I) and Ang II, and plasma and aortic ACEactivities. In the second set of experiments, the animals wereused to evaluate the histomorphological parameters of thethoracic aorta.

The procedures followed in this study were in accordance withinstitutional guidelines for animal experiments.

Arterial Pressure and Heart Rate inConscious Rats

In the first set of experiments, animals were anesthetizedwith pentobarbital (60 mg/kg IP). A catheter (PE-50 fused toPE-10; Guerbet-Biomedical, France) was placed in the lowerabdominal aorta via the femoral artery. Another catheter wasinserted into the carotid artery for the administration of AngI and II. The catheters were filled with heparinized saline (50

U/mL) and were tunneled under the skin of the back andexited between the scapulae. The animals were then allowedto recover from anesthesia for 24 hours in individual cages.Treatments were continued during this period. Then arterialpressure measurements were performed in conscious, freelymoving rats in their own cages after at least 30 minutes of rest.MBP and heart rate were recorded by means of a Statham P23ID pressure transducer (Gould Instruments, Cleveland, Ohio)connected to a Gould Brush recorder G4133.

After 30 minutes of baseline blood pressure evaluation,maximum changes in MBP in response to the injection of 150ng/kg Ang II in the carotid catheter were measured. Thirtyminutes after the pressure returned to the preinjection level, thecatheter was thoroughly rinsed with saline. A single dose of 300ng/kg Ang I was then injected and the maximal response noted.

Angiotensin Converting Enzyme Activity in Plasmaand Thoracic Aorta

After the end of this procedure, the rats received their oraltreatment and were allowed to recover for 24 hours. Then 3mL of blood was drawn from the femoral catheter. Aftercentrifugation, plasma samples were stored at -20°C. The ratsthen were killed by a blow on the head, and the thoracic aortawas removed and rapidly cleaned of blood with saline. Afterfreezing in liquid nitrogen, the samples of aorta were stored at—20°C until the assay was performed. Plasma ACE activitywas evaluated in all rats included in the first set of experi-ments. For technical reasons, aortic ACE activity was mea-sured in 6, 7, 6, and 7 rats of the P, Ql, Q10, and H groups,respectively.

ACE activity in plasma and aortic homogenates was deter-mined by a modified fluorometric assay as described byDepierre and Roth27 using carbobenzoxyphenylalanyl-histidyl-leucine as substrate. The assay conditions have been describedpreviously in detail.28

Morphological StudyIn a second set of experiments, histomorphological vascular

parameters were measured at 20 weeks of age29 according tothe following procedure. The animals were anesthetized withpentobarbital, a catheter was placed in the abdominal aortathrough the femoral artery, and MBP was measured. Aftermedian thoracotomy, the animals were exsanguinated bymeans of a catheter placed in the right atrium, and saline wasinjected into the femoral catheter. When the liquid comingfrom the atrium was clear, the circulatory tract was rinsed witha 4% formaldehyde solution. The animals died within secondsafter the beginning of the formaldehyde infusion. After 1 or 2minutes, a clamp was positioned on the atrium, and thefixation liquid was infused for 3 hours at a pressure equal tothe MBP of each animal.30 At the end of the perfusion, thethoracic aorta was dissected and preserved in a 4% formalde-hyde solution until the histological study was performed.

The different structures of the aortic media were studied ina segment of thoracic aorta longitudinally embedded in paraf-fin. Three successive sagittal sections of 5 ^m thickness weretreated by specific staining to obtain a monochromatic colorassociated with the various structures studied in the aorticmedia. Sirius red was used for collagen staining, orcein forelastin, and hematoxylin after periodic acid oxidation fornucleus staining. Morphometric analysis was performed with aspecialized automated image processor (NS 15000, Nachet-Vision, Paris, France). This processor is based on morpholog-ical mathematic principles and is software controlled. Differ-ent algorithms were developed to analyze each of the threestructures shown by the staining in each of the three successivesections. For image processing, the image is sent to theprocessor via a video camera and can be viewed on the TVmonitor. The control of luminosity is automatically adjusted bythe software to obtain similar contrasts, taking into accountthe total luminosity transmitted by the video camera. This



76 Hypertension Vol 23, No 1 January 1994

Body weight

grams4S0 -i

350-

250-

150-

— • PLACEBO

• — QUNAP1UL 1 mg/kg

O — QUMAPRIL 10 mg/kg

D — HYMtALAZINE 15 mg/kg

FIG 1. Line graph shows changes In body weight In the four rat groups during the treatment period.

analog image is then digitized as follows: Each elementarypoint (pixel) is automatically compared with a threshold; if thegray level of a pixel exceeds this threshold, the pixel is giventhe numeric value 1, otherwise it is given the numeric value 0.Threshold determination is a complex operation mainly involv-ing pixel ensembles. Threshold parameters are the size of suchpixel groups and their local contrast (top-hat transformation).The threshold is determined using the top-hat algorithm tominimize variations in nuclear staining and background. Fordata processing, this binary image is then processed to (1)eliminate background and artifacts, (2) delineate the zones ofinterest and reference zone, and (3) extract and measure theparameters from the various zones of interest.

The first algorithm permitted the analysis of the mean mediathickness by measurement of the distance between the internaland external elastic laminae. The medial elastin network wasanalyzed in terms of relative area and mean thickness of elastinlamella and lamina; the measurements and calculations weremade in 10 fields in each section. The second algorithmanalyzed the collagen matrix by measurement of relative areadensity and mean thickness of collagen fibers in 20 contiguousfields in each Sirius red-stained section. Elastin and collagendensities were defined as the ratio of the surface stained byorcein or Sirius red, respectively, to the surface of the studiedfield. The third algorithm counted the number of nuclei within20 fields of a 7442-jun2 area of measurement in each sectionand measured the mean area of each nucleus. A two-stepprocedure (conditional opening then conditional closing) leadsto the elimination of all particles under a predetermined size.The final result is retention of the images of the nuclei withoutany "holes" or deformations because of the structuring element(hexagon). The image processor automatically eliminates "bor-derline" nuclei before making the measurement of the numberof nuclei per unit of surface. Repetitive measurements wereperformed, pooled, and averaged for the three algorithms in thecorresponding stained sections of the aortic wall media of eachanimal. Results were stored on a floppy disk.

The measurement of media and lumen cross-sectional area(CSA) was performed in samples of the same thoracic aortaincluded in a gel used for the low-temperature sections

(medium of inclusion, ISOSYSTEM) and cooled at -20°C.When the gel was solidified, some transverse sections of thearterial rings were realized in each sample. The sections wereexamined with a microscope and photographed at a knownmagnification (x48). When the films were developed, thepictures obtained were projected on a digitizer so that thesurface of the vascular lumen and the external surface of thevessel could be measured. The final magnification used (mag-nification of the microscope x enlargement of the projector)was 160. For each sample, the collected results were the CSAof the vascular media and lumen. This last parameter reflectsthe degree of arterial hypertrophy better than medial thick-ness. Different studies have shown that the CSA of the arterialmedia is the most reliable constant of the vessel wall becauseit is not influenced by the variations of perfusion pressure.31-32

Morphological analyses were performed twice in a blindfashion by two independent workers.

Statistical AnalysisResults are expressed as mean±l SD. Data were analyzed

with a one-way analysis of variance. When F was less than 0.05,a Fisher test was performed for intergroup comparison.

The aortic ACE activity was analyzed by a nonparametrictest of variance (Kruskal-Wallis) followed by a Wilcoxon test.This test was chosen because of the number of studied samplesin two of four groups (less than seven per group).

ResultsFig 1 shows the evolution of body weight in the four

groups. QlO-treated animals showed a trend towardlower body weight, but this difference did not reach asignificant level.

Fig 2 shows the longitudinal changes in SBP evalu-ated with a tail-cuff method in the four groups at 8, 12,and 16 weeks of age. SBP increase was significantlyprevented in the Q10 and H groups from the 12th weekof age. The lack of difference at the eighth week may berelated to the important variability of the values inyounger rats.




Systolic blood pressure

mmHg 300 -i

250-

200-

15010 12

AGE ( m b )

16

*p<0.01 vs. QIOandH

• — PLACEBO

• QJUMAPML 1 mg/kg

O QUNAPRfl. 10 mg/kg

—d— HYDRALAZME 15 mg/kg

FIG 2. Une graph shows longitudinal changes in systolic blood pressure evaluated with a tall-cuff method in the four rat groups duringthe treatment period. Q10 indicates 10 mg/kg quinapril; H, hydralazlne.

Table 1 shows MBP and heart rate measured inconscious animals. In the P group, hypertension wasfully developed (190±23 mm Hg), whereas in the H andQ10 groups, hypertension was substantially preventedin a similar way (149±11 and 136±16 mm Hg, respec-tively). In the Ql group, MBP was slightly but signifi-cantly lower (171±15 mmHg) than in the P group(/><.O5) but higher than in the H and Q10 groups(P<.01 and P<.05, respectively). There were no signif-icant differences among groups concerning heart ratemeasured 24 hours after the last drug administration.

As shown in Table 1, the pressure response to Ang Iinjection was significantly (P<.01) decreased in the Q10rats compared with the three other groups. In the Qlrats, the response was slightly decreased compared withthe P group, but this difference did not reach a signifi-cant level. The response to Ang II injection was similarin all groups (Table 1).

Aortic media CSA was lower in the H and Q10 groups(455±61 and 487±57 xlO3 fim2) compared with the Pand Ql groups (636±72 and 550±65 xlO3 fim2)

(P<.001). There was a significant difference betweenthe P and Ql groups (P<.0l) (Fig 3). There were nosignificant differences in internal lumen CSA among thefour rat groups (P, 2544±278; Ql, 2474±335; Q10,2274±326; H, 2499±417 xlO3 ^m2).

Table 2 summarizes the histomorphometric data.Collagen content was similar in both groups treatedwith quinapril and significantly lower (f<.01) than inthe two other groups (Table 2 and Fig 4). A differenceof approximately 18% was observed for the elastinmedia content between the Q10 and P groups (Table 2).This difference was not significant (F=0.10). The nucleidensity (number per field) was increased in all treatedgroups compared with the placebo-treated animals. Theincrease was more pronounced in the hydralazine-treated animals (H>Q1O=Q1>P) (Table 2). Nucleicontent was not modified by the different treatments.

Plasma ACE activity was inhibited by 47% in the Q10group (P<.01), whereas in the Ql group, no significantdifference was observed compared with the P and Hgroups (Fig 5).

TABLE 1. Blood Pressure, Heart Rate, and Pressure Response to Angiotensin I and II InConscious Rats After 4 Months of Treatment

n

MBP, mm Hg

HR, bpm

Ang I: AMBP, mm Hg

Ang II: AMBP, mm Hg

P

11

190±23

373±47

26±7

43±7

Q1

9

171 ±15*

387±38

19±10

44±17

Q10

10

136±16tt

374 ±28

10±7t§U

39±12

H

8

149±11t§

363±35

25±7

44±8

P Indicates placebo; Q1,1 mg/kg quinapril; Q10,10 mg/kg quinapril; H, 15 mg/kg hydralazlne; MBP, meanblood pressure; HR, heart rate; bpm, beats per minute; Ang I, angiotensin I; Ang II, angiotensin II; and AMBP,maximum change In mean blood pressure. Values are mean±SD; n is number of animals.

*P<.05, tP<.01 vs P.$P<.01,§P<.05vsQ1.||P<.01 vs H.




TABLE 2. Hlstomorphometrtc Parameters of Thoracic Aorta After 4 Months of Treatment

n

Media thickness, nm

Elastin content, xio3 /urrVmm

Collagen density, %

Thickness of collagen fiber, ^m

Collagen content, xio3 jirrrVmm

Nuclei density, No. per field

Nuclei content, per mm

P

12

125±19

29.1 ±6.5

17.6±3.9

1.75±0.2

21.9±4.7

10.3±1.8

240±58

Q1

12

115±9

26.2±4.9

12.7±3.6*§

1.6±0.2

15.3±4.6*

14.6±4.81

292±59

Q10

12

101±12*t

23.8±3.4

14.3±4.1||

1.7±0.2

14.3±4.0*||

15.5±3.6||1

291 ±87

H

12

99±13**

27.2±4.1

18.8±4.9

1.75±0.2

18.6±5.0

19.7±6.0*t

322±58

P indicates placebo; Q1, 1 mg/kg quinapril; Q10, 10 mg/kg quinapril; and H, hydralazine. Values aremean±SD; n Is number of animals.

*P<.01,V<.05vsP.*P<.01,tP<.05vsQ1.§P<.01,|]f><.05vsH.

Aortic ACE activity was significantly (P<.01) re-duced in both groups treated with quinapril: 59% in Qland 61% in Q10 rats (Fig 6). Hydralazine treatment didnot affect this parameter.

DiscussionIn this study, both hydralazine and 10 mg/kg per day

quinapril administered daily for 4 months prevented thedevelopment of high blood pressure in SHR. In thesetwo groups, MBP was approximately 40 mm Hg lowerthan in the P gToup. The MBP values observed with 10mg/kg quinapril were comparable to those reported inother experimental studies using quinapril or otherACE inhibitors.518-21-29 In contrast, treatment with 1mg/kg quinapril had a small but significant antihyper-tensive effect compared with the P group, but this effectwas lower than seen with hydralazine and high-dosequinapril. Studying the same ACE inhibitor at this dose,

Ando et al33 reported a similar decrease in MBP after a3-week treatment in SHR.

The principal goal of our study was to evaluate theeffects of preventive treatment on the histomorpholog-ical parameters of the aorta. The pressure perfusionfixation technique was used as a physiological methodenabling the evaluation of the physiological medial andluminal dimensions.29-31 A 3-hour fixation with formal-dehyde was performed to avoid tissue deformationwhen transmural pressure is released.30 For reasonsexplained in "Methods," the CSA rather than thethickness of the aortic media was chosen for evaluationof arterial hypertrophy. In the present study, the place-bo-treated animals developed, as expected, significantmedial aortic hypertrophy, with CSA values comparableto those observed in SHR of the same age.34-35

Based on CSA measurements, aortic hypertrophy wasprevented to a similar degree in the groups treated with

Aortic Medial CSA2

800000-ili m

600000-

400000*: p<0.01 vs. P; *: p<0.01 vs. Ql§: [xO.05 vs. Ql

D PLACEBO

• QUNAPIUL 1 mg/kgD QUINAPRIL 10 mg/kg

• HYDRALAZINE 15 mg/kg

FIG 3. Bar graph shows cross-sectional area (CSA) of thoradc aorta in the four treated rat groups at 20 weeks of age. P indicatesplacebo; Q1, 1 mg/kg quinapril.




Aortic Collagen contentJ

10 /mm

30

20-

1 0 •*-

• PLACEBO

• QUINAPRIL Img/kg

E3 QUINAPRL 10 mo/kg• HYDRALAZINE 15 mg/kg

•: p<0.01 vs. P; f: p<0.05 vs. H

FIG 4. Bar graph shows collagen content of aortic mediaevaluated in the four treated rat groups at 20 weeks ofage. P indicates placebo; H, hydralazine.

hydralazine or a high dose of quinapril, whereas in theQl group the effect on these parameters was lesssignificant. Thus, the differences among the four groupswere parallel to those observed for MBP. These findingsstrongly suggest that in this experimental model ofhypertension, elevated blood pressure is the main factorinvolved in the development of aortic medial hypertro-phy. Similar findings have been shown previously byother authors.20-34

Hydralazine has been reported to be less effectivethan ACE inhibitors in preventing arterial hyperto-phy.36-37 In SHR, hydralazine was not able to preventthe development of mesenteric or renal artery hyper-trophy despite a significant decrease in blood pres-sure.37-38 However, Owens34 showed that in the aorta theefficacy of hydralazine and captopril in preventing thedevelopment of medial hypertrophy was the same as

their efficacy in lowering blood pressure. These discrep-ancies are consistent with the observation that theaorta, a rather elastic artery, is mainly sensitive to bloodpressure changes,2 whereas in peripheral arteries, whichare rather muscular, hypertrophy may be modulatedmainly by nonhemodynamic factors37-38 such as auto-nomic nervous system activity. Therefore, stimulation ofthis system after hydralazine administration could beresponsible for the lack of an antihypertrophic effect atthe side of the peripheral arteries. One may further addthat medial thickening in the aorta is mainly due tocellular hypertrophy, whereas in mesenteric arterieshyperplasia rather than hypertrophy occurs. These ob-servations can explain the differences reported betweenhydralazine and other antihypertensive drugs20-34-37-38

and suggest that caution is required when results areextrapolated from one artery to another.

Plasma ACE activity

nMol His-Leu/mln/ml

1 6 0 T

120 -

8 0 -

40

FIG 5. Bar graph shows plasma angiotensin con-verting enzyme (ACE) activity in four treated ratgroups at 20 weeks of age. P indicates placebo;Q1, 1 mg/kg quinapril; and H, hydralazine.

• PLACEBO

• QUNAPRIL 1 mg/kg

D QUNAPRH. 10 mg/kg

• HYDRALAZME IS mg/kg

*: p<0.01 v». P; ': p<0.01 vs. Ql$: p<0.01 vs. H




Aortic ACE activity

nHol Hii-Leu/mln/mi

80 -i

40- FIG 6. Bar graph shows aortic angiotensin con-verting enzyme (ACE) activity in four treated ratgroups at 20 weeks of age. P indicates placebo; H,hydralazine.

t3J<0.05 vs. P; l:p<0.05 vs. H

• PLACEBO

• QUINAPRIL Img/kg

D OJUNAPRL 10 mg/kg

• HYDRALAZINE 15 mg/kg

In the present study the antihypertrophic effect of thedrugs also was shown by the significant increase in aorticwall nuclei density. It is known that chronic experimen-tal hypertension is associated with a significant decreasein the relative number of nuclei, reflecting an increasedsize of vascular smooth muscle cells.18-29 In contrast, theincrease in nuclei density is consistent with a decreasein smooth muscle cell size. A similar increase in thenumber of nuclei after treatment with quinapril hasbeen reported by Richer et al.18

Taken together, these findings provide clear evidencethat in SHR, when treatment started before the devel-opment of genetic hypertension, hydralazine andquinapril prevented aortic hypertrophy in a mannerparallel to prevention of high blood pressure develop-ment. This effect was principally due to a reduction inarterial smooth muscle cell size.

Together with vascular smooth muscle cell hypertro-phy, an increase in arterial medial extracellular matrixproduced by vascular smooth muscle cells is one of themajor findings of hypertensive arterial structure. Agingand chronic hypertension induce a significant increasein extracellular matrix, especially in its collagen con-tent.18-29-39 In the normotensive rat, accumulation ofelastin and collagen in the aorta occurs mainly duringthe first 10 to 12 weeks of life.39 Although the increasein arterial collagen content seems to be partly related toelevated blood pressure, other factors have been foundto stimulate collagen synthesis in hypertensive animals.More specifically, Ang II receptor stimulation is able toincrease collagen synthesis by acting directly on smoothmuscle cells.40 In hypertensive animals, arterial collagenaccumulation is prevented by ACE inhibitors when usedat antihypertensive doses.18-29 In normotensive animals,early treatment with ACE inhibitors was able to slowthe age-related collagen accumulation in the arterialwall.39

In our study, there is a clear dissociation betweenblood pressure levels and collagen content in the differ-ent groups. Whereas hydralazine and high-dosequinapril substantially prevented development of high

blood pressure, the accumulation of collagen in thearterial wall was hindered only in rats treated withquinapril and not in those treated with hydralazine. Inthe two groups treated with quinapril, similar values ofarterial collagen content were found, despite differentblood pressure profiles. In this respect, two hypothesescan be made: (1) hydralazine increased the collagencontent by a direct effect or through sympathetic stim-ulation counteracting the beneficial effect of loweringblood pressure, and (2) quinapril has a specific actionon collagen accumulation. Although the first hypothesiscannot be eliminated, the observed results with a lowdose of quinapril suggest an effect of ACE inhibitors onaortic collagen turnover.

In the present study we cannot conclude whether theobserved effects on collagen accumulation are relatedto an inhibition of the normal maturation process or tothe hypertension-induced collagen increase because wedid not study normotensive controls. However, thispoint is very difficult to elucidate because hypertensionand aging have synergic effects on cardiovascular struc-ture, and one cannot separate the effects of these twofactors. Moreover, important genetic differences be-tween SHR and their normotensive controls wouldimpair the interpretation of such a comparison. Takinginto consideration these limitations, the present studyshows that factors other than pressure are involved inthe prevention of hypertension-induced arterial colla-gen accumulation.

Recently, Himeno et al41 showed that a nonhypoten-sive dose of quinapril was able to reverse the fibronectinmRNA expression induced by Ang I. Also, ACE inhib-itors have been shown to be able to decrease arterialcollagen content in prevention and regression stud-ies.18J529 However, our study is the first to our knowl-edge to clearly demonstrate that in hypertensive ratsthis effect may be mediated through blood pressure-independent factors.

An important finding of the present study was thatboth doses of quinapril decreased aortic ACE activity tothe same extent, whereas only the high dose of quinapril




blocked plasma ACE activity 20 to 24 hours after thelast drug administration. This result differs from previ-ously reported data in which ACE activity was evalu-ated after a single dose of ACE inhibitor and not afterchronic treatment.24 Fabris et al42 studied the timecourse of plasma and tissue ACE inhibition after asingle dose of 0.3 mg/kg quinapril. The highest ACEinhibition in plasma and aorta occurred 2 hours afterthe drug administration. After 24 hours, ACE was stillinhibited by 25% in plasma and by 30% in the aorta. Inour study, the lack of inhibition of plasma ACE in theQl group was confirmed by the absence of changes inthe Ang I pressure response compared with the P group.In Q10-treated rats, ACE activity and response to AngI injections were decreased by approximately 50% com-pared with the P group. Unger et al43 showed that afterchronic treatment with antihypertensive doses of enal-april or ramipril, tissue ACE activity was significantlyinhibited several days after the discontinuation of thedrugs, whereas plasma ACE activity and the pressorresponse to Ang I returned to pretreatment levels 24hours after drugs were stopped. Our study shows asignificant inhibition of aortic ACE activity even atnonhypotensive doses of quinapril. This effect may berelated to the high affinity of this compound for thetissular ACE as shown in binding studies.25'44 Takentogether, the results of our study strongly suggest thatthe prevention of aortic collagen accumulation is notrelated to blood pressure levels but rather to theinhibition of ACE activity. Our study cannot clarifywhether tissue Ang II inhibition, bradykinin activation,or other mechanisms are responsible for these effects onarterial structure. Further investigations are needed todefine the exact mechanisms of this action.

In conclusion, the present study clearly demonstratesdifferences in the factors influencing aortic structuralparameters during chronic hypertension. Thus, aortichypertrophy and smooth muscle cell size were mainlydependent on blood pressure levels, whereas collagenaccumulation occurred even when hypertension devel-opment was prevented, as was the case in the hydral-azine group. On the other hand, chronic ACE inhibitionprevented collagen accumulation even when hyperten-sion developed, as shown in the low-dose quinaprilgroup.

AcknowledgmentsThis study was performed with grants from the Institut

National de la Sant6 et de la Recherche M6dicale (INSERM)and from the Parke-Davis Pharmaceutical Co. We thank MrsMonique Legrand, Bernadette Lucet, Claudine Perret, and MrPierre Poitevin for their technical assistance and skillfuladvice.

References1. Wiener J, Loud AV, Giacomelli F, Anversa P. Morphometric

analysis of hypertension induced hypertrophy of rat thoracic aorta.AmJPathol. 19T7;88:619-634.

2. Wolinslcy H. Response of the rat aortic media to hypertension:morphological and chemical studies. Circ Res. 1971^6:507-522.

3. Wolinsky H. Long term effects of hypertension on the rat aorticwall and their relation to concurrent aging changes. Circ Res.1972;30:301-309.

4. Leung DYM, Glagov S, Mathews MB. Cyclic stretching stimulatessynthesis of matrix components by arterial smooth muscle cells invitro. Science. 1976;191:475-477.

5. Christensen KL. Reducing pulse pressure in hypertension maynormalize small artery structure. Hypertension. 1991;18:722-727.

6. Mulvany MJ, Hansen PK, Aalkjaer C. Direct evidence that thegreater contractility of resistance vessels in spontaneously hyper-tensive rats is associated with a narrowed lumen, a thickenedmedia, and an increased number of smooth muscle cell layers. CircRes. 1982;43:854-864.

7. Warshaw DM, Root DT, Halpern W. Effects of antihypertensivedrug therapy on the morphology and mechanics of resistancearteries from spontaneously hypertensive rats. Blood Vessels. 1980;17:257-270.

8. Folkow B, Hallback M, Lundgren R, Sivertsson R, Weiss L.Importance of adaptive changes in vascular design for estab-lishment of primary hypertension, studied in man, and in sponta-neously hypertensive rats. Circ Res. 1973;32-33(suppl I):I-2-I-16.

9. Dzau VJ, Safar ME. Large conduit arteries in hypertension: role ofthe renin-angiotensin system. Circulation. 1977;77:947-953.

10. Bevan RD. Effect of sympathetic denervation on smooth musclecell proliferation in the growing rabbit ear artery. Ore Res. 1975;37:14-19.

11. Bunkenburg B, van Amelsvoort T, Rogg H, Wood JM. Receptor-mediated effects of angiotensin II on growth of vascular smoothmuscle cells from spontaneously hypertensive rats. Hypertension.1992;20:746-754.

12. Campbell DJ. Tissue renin-angiotensin system: sites of angiotensinformation. J Cardiovasc Pharmacol. 1987;10(suppl 7):S1-S8.

13. Gohlke P, Banning P, Unger T. Distribution and metabolism ofangiotensin I and II in the blood vessel wall. Hypertension. 1992;20:151-157.

14. Geisterfer AAT, Peach MJ, Owens GK. Angiotensin II induceshypertrophy, not hyperplasia, of cultured rat aortic smooth musclecells. Ore Res. 1988;62:749-756.

15. Naftilan AJ, Pratt RE, Dzau VJ. Induction of c-fos, c-myc andPDGF A chain gene expression by angiotensin II in cultured ratvascular smooth muscle cells. / Clin Invest. 1989;83:1419-1424.

16. Hiroshi K, Hiromichi S, Shingo T, Yasuhide O, Takehiko T,Atsuhisa S, Takao S. Angiotensin II stimulates collagen synthesisin cultured vascular smooth muscle cells. / Hypertens. 1991;9:17-22.

17. Griffin SA, Brown WCB, MacPherson F, MacGrath JC, WilsonVG, Korsgaard N, Mulvany MJ, Lever AF. Angiotensin II causesvascular hypertrophy in part by a nonpressor mechanism. Hyper-tension. 1991;17:626-635.

18. Richer C, Mulder P, Fornes P, Richard V, Camillcri JP, GiudicelliJF. Hemodynamic and morphological effects of quinapril duringgenetic hypertension development. / Cardiovasc Pharmacol. 1991;18:631-642.

19. Safar ME. Therapeutics trials and large arteries in hypertension.Am Heart J. 1988;115:702-710.

20. Christensen KL, Jespersen LT, Mulvany MJ. Development ofblood pressure in spontaneously hypertensive rats after withdrawalof long-term treatment related to vascular structure. J Hypertens.1989;7:83-90.

21. Linz W, Schdlkens BA. A specific B2-bradykinin receptor antagonistHOE 140 abolishes the antihypertrophic effect of ramipril. Br JPharmacol. 1992;105:771-772.

22. Linz W, Schdlkens BA, Ganten D. Converting enzyme inhibitionspecifically prevents the development and induces regression ofcardiac hypertrophy in rats. Clin Exp Hypertens A. 1989;11:1325-1350.

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

24. Kaplan HR, Taylor DG, Olson SC, Andrews LK. Quinapril: apreclinical review of the pharmacology, pharmacokinetics and tox-icology. Angiology. 1989:40:335-350.

25. Nakajima T, Yamada T, Setoguchi M. Prolonged inhibition of localangiotensin converting enzyme after single or repeated treatmentwith quinapril in spontaneously hypertensive rats. J CardiovascPharmacol. 1992;19:102-107.

26. Bunag RD. Validation in awake rats of a tail-cuff method formeasuring systolic pressure. JAppl PhysiM. 197334:279-282.

27. Depierre D, Roth M. Fluorometric determination of dipeptidylcarboxypeptidase (angiotensin I converting enzyme). Enzyme.1975:19:65-70.

28. Unger Th, Schull B, Rascher W, Lang RE, Ganten D. Selectiveactivation of the converting enzyme inhibitor MK421 and com-parison of its diacid form with captopril in different tissues of therat. Biochem Pharmacol. 1982^1:3063-3070.

29. Levy BI, Michel JB, Salzmann JL, Azizi M, Poitevin P, Safar M,Camilleri JP. Effects of chronic inhibition of converting enzyme on




mechanical and structural properties of arteries in rat renovascularhypertension. Ore Res. 1988;63:227-239.

30. Kratky RG, Lo DK, Roach MR. Quantitative measurement offixation rate and dimension changes in the aldehyde/pressure fbtedcanine carotid artery. Blood Vessels. 1991;28:386-395.

31. Lee RMKW. Preservation of in vivo morphology of blood vesselsfor morphometric studies. Scanning Microsc. 1987;1:1287-1293.

32. Walmsley JG, Gore RW, Dacey RG Jr, Damon DN, Duling BR.Quantitative morphology of arterioles from the hamster cheek pouchrelated to mechanical analysis. Microvasc Res. 1982;24:249-271.

33. Ando K, Frohlich ED, Chien Y, Pegram BL. Effects of quinaprilon systemic and regional hemodynamics and cardiac mass in spon-taneously hypertensive and Wistar-Kyoto rats. J Vase Med Biol.1991^:117-123.

34. Owens GK. Influence of blood pressure on the development ofaortic medial smooth muscle hypertrophy in spontaneously hyper-tensive rats. Hypertension. 1987;9:178-187.

35. Henrichs KJ, Unger Th, Berecek KH, Ganten D. Is arterial mediahypertrophy in spontaneously hypertensive rats a consequence ofor a cause for hypertension? Clin Sci. 198O;59:331-333.

36. Freslon JL, Giudicelli JF. Compared myocardial and vasculareffects of captopril and dihydralazine during hypertension devel-opment in spontaneously hypertensive rats. Br J Pharmacol. 1983;80:533-543.

37. Jespersen LT, Nyborg NCB, Lederballe Pedersen O, MikkelsenEO, Murvany MJ. Cardiac mass and peripheral vascular structure

in hydralazine-treated spontaneously hypertensive rats. Hyper-tension. 1985;7:734-741.

38. Smeda JS, Lee RMKW, Forrest JB. Prenatal and postnatalhydralazine treatment does not prevent renal vessel wall thickening inSHR despite the absence of hypertension. Ore Res. 1988;63:534-542.

39. Keeley FW, Elmoselhi A, Leenen FHH. Enalapril suppressesnormal accumulation of elastin and collagen in cardiovasculartissues of growing rats. Am J Physiol. 1992;262:H1013-H1021.

40. Kato H, Suzuki H, Tajima S, Ogata Y, Tominaga T, Sato A, SarutaT. Angknensin II stimulates collagen synthesis in cultured vascularsmooth muscle cells. J Hypertens. 1991;9:17-22.

41. Himeno H, Hosoi M, Brecher P, Chobanian AV. Effect of angio-tensin converting enzyme inhibition with quinapril on aorticfiionectin expression. Hypertension. 1992^3:50. Abstract.

42. Fabris B, Chen B, Pupic V, Perich R, Johnston CI. Inhibition ofangiotensin-converting enzyme in plasma and tissue. / CardiovascPharmacol. 199O;15(suppl 2):36-41.

43. Unger T, Ganten D, Lang RE, Sch61kens BA. Persistent tissueconverting enzyme inhibition following chronic treatment withHoe498 and MK421 in spontaneously hypertensive rats.J Cardiovasc Pharmacol. 1985;7:36-41.

44. Johnston CI, Jandelei K, Mooser V, Katapothis A, Perich R,Paxton D, Murohara Y, Jackson B. Angiotensin-convertingenzyme and its inhibition in the heart and blood vessels./ Cardiovasc Pharmacol. 1992;20(suppl B):S6-S11.



P Albaladejo, H Bouaziz, M Duriez, P Gohlke, B I Levy, M E Safar and A BenetosAngiotensin converting enzyme inhibition prevents the increase in aortic collagen in rats.

Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 1994 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/01.HYP.23.1.74

1994;23:74-82Hypertension.

http://hyper.ahajournals.org/content/23/1/74World Wide Web at:

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

http://hyper.ahajournals.org//subscriptions/

is online at: Hypertension 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 process is available in the

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

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


http://hyper.ahajournals.org/content/23/1/74

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

http://www.lww.com/reprints

http://hyper.ahajournals.org//subscriptions/


Documents

Angiotensin Converting Enzyme Inhibition Prevents the Increase in Aortic Collagen in Rats