Journal of Human (2001) 15, 185–190  2001 Nature Publishing Group All rights reserved 0950-9240/01 $15.00 www.nature.com/jhh ORIGINAL ARTICLE Polymorphisms of the - system in patients with multifocal renal arterial fibromuscular dysplasia

A Bofinger, C Hawley, P Fisher, N Daunt, M Stowasser and R Gordon Hypertension Unit, University of Queensland Department of Medicine, Greenslopes Private Hospital, Brisbane, Queensland, Australia

Fibromuscular dysplasia (FMD) is an important cause of quencies did not differ significantly between groups, renal , particularly in young females. except that MF-FMD patients had a significantly higher Polymorphisms of the renin-angiotensin (RA) system frequency of the ACE I allele than control subjects (0.62 Since the ACE I allele is associated .(0.026 ؍ have been implicated in the pathogenesis of hyperten- vs 0.47, P sion and atherosclerotic , and may play with lower circulating ACE levels and possibly lower a role in the development of FMD. Examination of poly- tissue levels of angiotensin II (Ang II), and since Ang morphisms by PCR for angiotensin-converting enzyme II modulates vascular smooth muscle cell growth and

(ACE) I/D, angiotensin II type 1 receptor (AT1R) A1166C synthetic activity, the I allele might predispose to defec- and angiotensinogen (AGT) M235T and T174M was tive remodelling of the arterial media, and thus to the undertaken in 43 patients with typical multifocal renal development of MF-FMD. This contrasts with athero- arterial FMD (MF-FMD) and in 89 controls. The age of sclerotic stenosis, coronary resten- MF-FMD patients at the time of diagnosis of hyperten- osis and carotid intimal thickening, which are diseases sion did not differ (38.6 ؉ 11.1 years vs 35.5 ؎ 10.3 affecting the arterial intima, and which are associated .from controls and the proportion (95% with increased frequency of the D allele (0.12 ؍ years, P of females was similar. Allele fre- Journal of Human Hypertension (2001) 15, 185–190 (0.14 ؍ vs 86%, P

Keywords: fibromuscular dysplasia; renin-angiotensin system; renovascular hypertension; renal artery; gene polymor- phisms; aetiology

Introduction angiotensin I (Ang I), which is then cleaved by angi- otensin converting enzyme (ACE) to form Ang II.14 The most common cause of renovascular hyperten- Ang II is not only a powerful vasoconstrictor, but sion in patients presenting under 40 years of age, also modulates vascular smooth muscle cell (VSMC) fibromuscular dysplasia (FMD), is three times more 15,16 1 growth and synthetic activity. These latter common in females than males. Suggested patho- effects are markedly suppressed or abolished by Ang physiological factors have included mechanical II type 1 receptor (AT R) antagonism.15,16 Studies 2 3,4 1 arterial stretching, female sex hormones, genetic suggest that neointimal hyperplasia following factors,5,6 vascular spasm,7 defects in arterial elastic 3 8–11 arterial injury can be significantly reduced by ACE tissue and smoking. 17–20 19,21 inhibition and by selective AT1R blockade. There has been much interest in the role of poly- Hence, not only is the RA system important in morphisms of the renin-angiotensin (RA) system, hypertension directly by vasoconstriction and not only in hypertension but in cardiovascular dis- VSMC growth, but also in failure of interventions ease in general. Angiotensin II (Ang II), an octapep- which increase the arterial lumen but damage the tide hormone, is the main effector peptide of RA sys- 12,13 vessel wall. Thus the RA system affects the behav- tem. Angiotensinogen (AGT), synthesised in the iour of the lining as well as the muscular wall of liver, is cleaved by renin to form the decapeptide, , and these effects may be achieved by quite different mechanisms. The RA system has the capacity to influence predominantly endothelial Correspondence: Professor Richard D Gordon, University Depart- pathology, such as , but also ment of Medicine, Greenslopes Private Hospital, Newdegate Street, Brisbane Q 4120, Australia hypertrophy of the muscular arterial wall, as in 1,22 E-mail: med.gslopesȰmailbox.uq.edu.au some forms of FMD. As well, Ang II is the princi- Received 29 December 1999; revised 26 June 2000; accepted 18 pal regulator of aldosterone levels, and aldosterone August 2000 has a trophic effect on myocardial cells and the RA system polymorphisms in FMD A Bofinger et al 186 VSMC, as well as promoting fibrosis.23–26 Polymor- Exploring a possible role for the RA system in the phisms associated with differing levels of Ang II are development of FMD, we examined polymorphisms therefore of interest in FMD. of the ACE, AT1R and AGT genes in 43 patients with An increased prevalence of the ACE D allele has angiographically typical MF-FMD and in 89 con- been reported in patients with atherosclerotic renal trol subjects. artery stenosis when compared with controls.27 While the atherosclerotic process is very different Methods and materials histologically and, probably, aetiologically, from FMD, both are examples of vascular occlusive dis- A total of 110 hypertensive patients referred ease, and perhaps the components of the RA system between 1970 and 1997 were diagnosed with FMD might also modulate the development of FMD. of the renal arteries. Angiograms of 66 patients were Circulating levels of the enzyme renin and its sub- available for review by at least two of the authors. strate AGT are important determinants of circulating Typical microaneurysmal MF-FMD was identified Ang I and, after conversion of Ang I to Ang II by in 43 subjects. ACE, of Ang II levels. In general, the higher the level Eighty-nine normotensive volunteers (‘controls’) of AGT, the higher the level of Ang II. Certain poly- were recruited from staff of the Greenslopes Private morphisms of the AGT gene are associated with Hospital and the University of Queensland Depart- higher AGT levels.28 ments of Medicine and Surgery. Subjects were The ACE D allele is associated with higher circul- included as controls if they met the following cri- ating levels of ACE29 which, other things being teria: (1) able to provide informed consent, (2) no equal, will also be associated with higher circulating personal history of hypertension nor cardiovascular levels of Ang II. Increased ACE activity and the ACE disease, (3) had never taken antihypertensive medi- D allele have both been associated with an increase cations, (4) no family history of FMD or early (aged in carotid intimal-medial thickness, the distance Ͻ50) cardiovascular events, and (5) normal blood between the media-adventitia interface and the pressure, defined as a seated blood pressure reading lumen-intima interface,30 in both diabetic31 and of р140/90 mm Hg (average of two seated blood non-diabetic32–34 subjects. Coronary stent , pressure readings taken approximately 2 min apart). which is primarily a result of neointimal hyper- Venous blood was collected from each subject plasia,35 was significantly increased in patients with into lithium heparin tubes and stored at −70°C until the ACE DD genotype.36,37 On the other hand, resten- DNA extraction. DNA was extracted from whole osis following coronary without stent, blood using a commercial kit (Nucleon BACC2; which appears to be primarily due to vessel Amersham International, Little Chalfont, Buckingh- remodelling,38 was not associated with ACE geno- amshire, UK). type.39–41 The ACE I/D polymorphism was identified by Percutaneous transluminal renal angioplasty polymerase chain reaction using the primers (PTRA) has become the treatment of first choice for described by Lindpaintner et al47 (hace3s renal artery FMD. Therefore, current diagnosis and 5ЈGCCCTGCAGGTGTCTGCAGCATGT3Ј; hace3as classification of FMD is based on the angiographic 5ЈGGATGGCTCTCCCCGCCTTGTCTC3Ј). Preferen- rather than histologic appearance.6,9,42 Lesions are tial amplification of the D allele occurs in ID hetero- classified as multifocal FMD (MF-FMD), in which zygotes, leading to misclassification of ID as DD in there are multiple points of stenosis within the one 4% to 5% of cases.47 Therefore, all DD samples were lesion, and unifocal FMD (UF-FMD) in which there subjected to a second, separate PCR amplification is a single short or elongated stenosis. with a primer pair which recognises the insertion- Typical microaneurysmal MF-FMD is usually specific sequence (hace5a, 5ЈTGGGACCACAGCG associated with medial fibroplasia on histological CCCGCCACTAC3Ј; hace5c, 5ЈTCGCCAGCCCTCCC examination.1,22,43 This pattern usually affects the ATGCCCATAA3Ј).47 Reactions were performed in a mid- to distal portions of the renal artery and is usu- final volume of 20 ␮l containing 1 ␮M sense primer, ally right-sided or bilateral.1,6,22,44 Other arteries, 1 ␮M antisense primer, 200 ␮M dNTPs (Promega, particularly the internal carotid and external iliac Madison, WI, USA), 0.35 U Taq DNA polymerase arteries, may be affected.1,6 Approximately 90% of with 2 ␮l of the supplied PCR reaction buffer patients with medial fibroplasia are females between (QIAGEN, Clifton Hill, Victoria, Australia), 4 ␮lQ- 25 and 50 years of age.1,6,22,44,45 UF-FMD usually Solution (QIAGEN, Clifton Hill, Victoria, Australia) affects the proximal segments of the renal artery. and 1 ␮l genomic DNA. Q-Solution is a non-DMSO This pattern is characteristic of intimal fibroplasia. solution which ‘facilitates amplification of difficult However, all histological FMD subtypes may pro- templates by modifying the melting behavior of duce unifocal lesions.46 In UF-FMD, the left and DNA’ (QIAGEN Product Guide 1999). Thermocy- right renal arteries are affected with similar fre- cling was performed for 35 cycles with denaturation quency and bilateral involvement is less common. at 94°C for 30 sec, annealing for 45 sec (at 56°C for Patients with UF-FMD tend to be younger than those the hace3 primers and 69°C for the hace5 primers), with MF-FMD, and do not display the female pre- and extension at 72°C for 2 min with a final exten- dominance seen in MF-FMD.1,22,44 sion at 72°C for 7 min. The amplified products were

Journal of Human Hypertension RA system polymorphisms in FMD A Bofinger et al 187 electrophoresed on 1% agarose gels, stained with homa, USA). Parametric data was compared using ethidium bromide and viewed under UV illumi- the t-test for independent samples. The Fisher exact nation. The hace3 primers resulted in 319-bp and test was used to compare allele frequencies between 597-bp amplicons, representing, respectively, the D groups. A two-sided probability value of Ͻ0.05 was and I alleles. The hace5 primers resulted in a 335- considered significant. bp fragment produced only in the presence of the I allele. Results Determination of the ATR A1166C polymorphism is facilitated by the presence in the C allele of an Demographic and clinical characteristics of the 43 additional recognition site for the restriction endon- MF-FMD subjects and the 89 control subjects are uclease Dde I.48 Amplification using the primers 5Ј- presented in Table 1. At the time of diagnosis of AATGCTTGTAGCCAAAGTCACCT-3Ј and 5Ј- FMD, subjects with MF-FMD were significantly GGCTTTGCTTTGTCTTGTTG-3Ј results in an 856- older than controls (47.7 ± 10.9 years vs 35.5 ± 10.3 bp product.49 Subsequent digestion with DdeI pro- years, P Ͻ 0.0001). However, there was no difference duces two bands of 602-bp and 254-bp for the A in the ages of the MF-FMD group at the time of diag- allele, with the C allele having two smaller bands nosis of hypertension and controls (38.6 ± 11.1 years (143-bp and 111-bp) in place of the 254-bp band. vs 35.5 ± 10.3 years, P = 0.12). Both groups had a The PCR reaction mixture was the same as for the similar proportion of females (95% vs 86%, ACE I/D polymorphism. DNA was amplified for 40 P = 0.14). cycles with denaturation at 94°C for 1 min, Histological diagnoses were available for five sub- annealing at 60°C for 1 min, and extension at 72°C jects: three with unilateral right-sided MF-FMD and for 2 min with a final extension at 72°C for 10 mins. two with bilateral MF-FMD. Each of the five subjects The amplified PCR product was subjected to diges- had a histological diagnosis of medial fibroplasia. tion with DdeI (New England BioLabs, Beverly, MA, Genotype distributions and allele frequencies for ␮ ␮ USA). Each 15 l digestion contained 8 l PCR pro- the ACE I/D, AT1R A1166C, AGT T174M and AGT duct, 5 U DdeI and 1.5 ␮l of the supplied buffer. M235T polymorphisms are presented in Table 2. Digestion was performed at 37°C for 2 hours. The The frequency of the ACE I allele was significantly resulting digestion product was electrophoresed on higher in the MF-FMD group than in controls (0.62 2% agarose gels, stained with ethidium bromide and vs 0.47, P = 0.026). No significant differences were viewed under UV illumination. Determination of the AGT T174M and M235T Table 1 Demographic characteristics of MF-FMD subjects and polymorphisms was performed according to the controlsa method of Rutledge et al.50 A single pair of primers is used to amplify a 338-bp fragment of DNA con- MF-FMD Control P value taining the two polymorphisms: sense 5Ј-GATGCGCACAAGGTCCTGTC-3Ј, n 43 89 antisense 5Ј-CCGCCCGCCCCGCCCGCCGCCCGCC Age at (years) 47.7 ± 10.9 CCGCCCGCCGCCCGC TGCTGTCCACACTGGCT Age at diagnosis of 38.6 ± 11.6 35.5 ± 10.3 0.12 Ј hypertension or CGC-3 . normotension (years) The antisense primer includes a 40-bp GC-clamp Females (%) 41 (95%) 77 (86%) 0.14 to ensure visualisation of the fragments after diges- tion. The T174M mutation involves a C→T base sub- aValues expressed as n or mean ± s.d. The age given for the con- stitution, introducing an Nco I restriction site trol subjects is the age at recruitment into the study. CٙCATGG) in the 174 M allele. The M235T) mutation involves a T→C base substitution. The Table 2 Genotype distributions and allele frequencies in 43 MF- mispairing primer method creates a BstU I restric- FMD subjects and 89 controls .tion site (CGٙCG) in the presence of the 235T allele The PCR reaction mixture was the same as for the MF-FMD Control P values ACE I/D polymorphism. DNA was amplified for 40 cycles with denaturation at 94°C for 30 sec, n 43 89 annealing at 64°C for 30 sec, and extension at 72°C ACE I/D for 2 min with a final extension at 72°C for 10 min. Genotype II: ID: DD 17: 19: 7 20: 43: 26 To detect the T174M polymorphism, 8 ␮l of PCR Allele I: D 0.62: 0.38 0.47: 0.53 0.026 ° ATR1 A1166C product was digested with 10 U Nco Iat37C for 2 Genotype AA: AC: CC 16: 23: 4 46: 37: 6 hours. To detect the M235T polymorphism, 8 ␮lof Allele A: C 0.64: 0.36 0.72: 0.28 0.20 PCR product was digested 10 U BstU Iat60°C for 2 AGT M235T hours. The resulting digestion products were elec- Genotype MM: MT: TT 13: 20: 10 37: 40: 12 Allele M: T 0.53: 0.47 0.64: 0.36 0.11 trophoresed on 2% agarose gels, stained with ethid- AGT T174M ium bromide and viewed under UV illumination. Genotype TT: MT: MM 34: 8: 1 70: 16: 3 Statistical analysis was performed using Statistica Allele T: M 0.88: 0.12 0.88: 0.12 1.00 for Windows (Release 5.0) (StatSoft Inc, Tulsa, Okla-

Journal of Human Hypertension RA system polymorphisms in FMD A Bofinger et al 188 found between groups for any of the other polymor- but a genotype which predisposes to defective phisms. remodelling in the muscular layer of arteries. While MF-FMD is principally a disease of females 1,22 Discussion of reproductive age, atherosclerotic arterial dis- ease tends to spare this group. The gender-associa- Major difficulties in any study of FMD are: (1) the ted predisposition to MF-FMD has not been necessity of classification by angiographic appear- explained. However, female sex hormones have ance since success of PTRA reduces the availability been proposed as possible pathogenic factors.1 In of tissue for histological, histochemical or genetic oophorectomised rats, chronic administration of examination, and (2) the imperfect correlation estrogens was associated with lower plasma and between angiographic and histological diagnoses. tissue ACE activities by approximately 40%, com- Yet accurate phenotypic determination is essential pared to oophorectomised animals without estrogen for meaningful genetic analysis. In this study we replacement.52,53 In similar studies in cynomolgus have attempted to define a group of subjects with monkeys54 oestrogen replacement reduced plasma angiographic appearances absolutely typical of ACE activity by 17%. Furthermore, a significant medial fibroplasia by selecting subjects with typical reduction in ACE mRNA was demonstrated in lung, microaneurysmal MF-FMD. In the current study, aorta, renal cortex and renal medulla in estradiol- angiographic diagnosis of MF-FMD accurately pre- treated oophorectomised rats.52 Reduction of serum dicted the histological diagnosis of medial fibro- ACE activity has also been reported in the first and plasia in the five subjects where this was available. second trimesters in normotensive pregnant females The fibromuscular ridges of MF-FMD are charac- with elevated plasma oestrogen compared to non- terised by degeneration of medial elastic fibres and pregnant controls,55 and following combined oestra- replacement of the normal media by loose collagen diol valerate/norethisterone hormone replacement and plump disoriented smooth muscle cells.22 The therapy for 6 months in healthy normotensive post- mural between these fibromuscular menopausal females.56 Therefore, by the mechanism ridges consist of thinning or loss of smooth muscle of lowered tissue Ang II, the combined effects of with deficiency of the IEL.22 This is in contrast to estrogens and the ACE I allele may predispose to the the other pathological forms of FMD, with either development of MF-FMD, particularly in females of extensive fibroplasia of the outer half of the media reproductive age. in ‘subadventitial fibroplasia’ or the intimal changes While FMD and atherosclerotic arterial stenosis of intimal fibroplasia.22 It is also in marked contrast are very different diseases, it is of interest that smok- to the pathology of atherosclerosis and stent resten- ing, which has a marked effect on the development osis. and progression of , is also associated with In the current study, patients with MF-FMD had an increased severity of MF-FMD.8 Just as severe a significantly higher frequency of the ACE I allele atheroma can occur in non-smokers, smoking is not than controls. The increased prevalence of the ACE essential for the development of FMD.8 I allele in MF-FMD is in contrast of studies of coron- To the best of our knowledge, this is the first ary stent restenosis,36,37 carotid intimal thicken- report investigating the relationship between the ing,32–34 and atherosclerotic renal artery stenosis27 in polymorphisms of the RA system and FMD. In view each of which an association between increased risk of the paucity of knowledge regarding the aetiology and the ACE D allele has been reported, and to a and pathogenesis of FMD, the possible effects of dif- large prospective study seeking an association fering levels of activity of the renin-angiotensin sys- between ACE genotypes and the risk of ischaemic tem on the development of FMD deserve further heart disease or myocardial infarction47 in which no investigation. association was found with either allele. This is not surprising, given that MF-FMD has a completely dif- Acknowledgements ferent morphological lesion from that of atheroma- tous disease. This work was supported by grants from the Aus- The ACE I allele is associated with lower circulat- tralian Kidney Foundation and the Sir Edward Dun- ing ACE levels than the ACE D allele.29 Muller et lop Medical Research Foundation. AB received a al,51 using perfused isolated rat hindquarters in rats Medical Postgraduate Research Scholarship from with and without two-kidney one clip (2K1C) hyper- the National Health and Medical Research Council tension, showed that increased ACE gene expression of Australia. MS was a Postdoctoral Fellow with the and ACE levels led to an increase in conversion of National Heart Foundation of Australia. Ang I to Ang II. Therefore, the ACE I allele is prob- ably associated with lower local concentrations of References Ang II. Could the ACE I allele, by an association 1Lu¨ scher TF et al. Arterial fibromuscular dysplasia. with lower local concentrations of Ang II, impede Mayo Clin Proc 1987; 62: 931–952. remodelling, and thus promote the development of 2 Palubinskas AJ, Wylie EJ. Roentgen diagnosis of fib- MF-FMD? Perhaps we should not be looking for a romuscular hyperplasia of the renal arteries. Radiology genotype which predisposes to intimal hyperplasia 1961; 76: 634–639.

Journal of Human Hypertension RA system polymorphisms in FMD A Bofinger et al 189 3 Wylie EJ, Wellington JS. Hypertension caused by fib- 26 Young M. Adrenal steroids and cardiac fibrosis. Ster- romuscular hyperplasia of the renal arteries. Am J Surg oids 1995; 60: 133–136. 1960; 100: 183–193. 27 Missouris CG et al. Genetic risk for renal artery sten- 4 Hunt JC et al. Idiopathic fibrous and fibromuscular osis: association with deletion polymorphism in angi- stenoses of the renal arteries associated with hyperten- otensin 1-converting enzyme gene. Kidney Int 1996; sion. Mayo Clin Proc 1962; 37: 181–216. 49: 534–537. 5 Rushton AR. The genetics of fibromuscular dysplasia. 28 Jeunemaitre X et al. Molecular basis of human hyper- Arch Intern Med 1980; 140: 233–236. tension: role of angiotensinogen. Cell 1992; 71: 169– 6 Pannier-Moreau I et al. Possible familial origin of 180. multifocal renal artery fibromuscular dysplasia. J Hy- 29 Rigat B et al. An insertion/deletion polymorphism in pertens 1997; 15: 1797–1801. the angiotensin I-converting enzyme gene accounting 7 Fievez ML. Fibromuscular dysplasia of arteries: a spas- for half the variance of serum enzyme levels. J Clin tic phenomenon? Med Hypotheses 1984; 13: 341–349. Invest 1990; 86: 1343–1346. 8 Bofinger A et al. Increased severity of multifocal renal 30 Pignoli P et al. Intimal plus medial thickness of the arterial fibromuscular dysplasia in smokers. J Hum arterial wall: a direct measurement with ultrasound Hypertens 1999; 13: 517–520. imaging. Circulation 1986; 74: 1399–1406. 9 Sang CN et al. Etiologic factors in renovascular fibro- 31 Hosoi M et al. Angiotensin-converting enzyme gene muscular dysplasia. A case-control study. Hyperten- polymorphism is associated with carotid arterial wall sion 1989; 14: 472–479. thickness in non-insulin-dependent diabetic patients. 10 Nicholson JP et al. Cigarette smoking and renovascular Circulation 1996; 94: 704–707. hypertension. Lancet 1983; 2: 765–766. 32 Bonithon-Kopp C et al. Plasma angiotensin-converting 11 Mackay A et al. Smoking and . enzyme activity and carotid wall thickening. Circu- BMJ 1979; 2: 770. lation 1994; 89: 952–954. 12 Levy BI. The potential role of angiotensin II in the vas- 33 Castellano M et al. Angiotensin-converting enzyme I/D culature. J Hum Hypertens 1998; 12: 283–287. polymorphism and arterial wall thickness in a general 13 Moeller I et al. Bioactive angiotensin peptides. J Hum population. The Vobarno Study. Circulation 1995; 91: Hypertens 1998; 12: 289–293. 2721–2724. 14 Poulsen K, Jaconsen J. Enzymatic reactions of the 34 Pujia A et al. Deletion polymorphism in angiotensin renin-angiotensin system. In: Robertson JI, Nicholls converting enzyme gene associated with carotid wall MG (eds). The Renin-Angiotensin System. Gower thickening in a healthy male population. Coron Artery Medical Publishing: New York, 1993, pp 5.1–5.12. Dis 1996; 7: 51–55. 15 Kato H et al. Angiotensin II stimulates collagen syn- 35 Hoffmann R et al. Patterns and mechanisms of in-stent thesis in cultured vascular smooth muscle cells. J Hy- restenosis. A serial intravascular ultrasound study. Circulation 1996; 94: 1247–1254. pertens 1991; 9: 17–22. 36 Amant C et al. D allele of the angiotensin I-converting 16 Daemen MJAP, Lombardi DM, Bosman FT, Schwartz enzyme is a major risk factor for restenosis after coron- SM. Angiotensin II induces smooth muscle prolifer- ary stenting. Circulation 1997; 96: 56–60. ation in the normal and injured rat arterial wall. Circ 37 Ribichini F et al. Plasma activity and insertion/ Res 1991; 68: 450–456. deletion polymorphism of angiotensin I-converting 17 Powell JS et al. Inhibitors of angiotensin-converting enzyme: a major risk factor and a marker of risk for enzyme prevent myointimal proliferation after vascu- coronary stent restenosis. Circulation 1998; 97: 147– lar injury. Science 1989; 245: 186–188. 154. 18 Clozel J-P et al. Inhibition of converting enzyme and 38 Mintz GS et al. Arterial remodeling after coronary neointima formation after vascular injury in rats and angioplasty: a serial intravascular ultrasound study. guinea pigs. Hypertension 1991; 18: II-55–II-59. Circulation 1996; 94: 35–43. 19 Osterrieder W et al. Role of angiotensin II in injury- 39 Hamon M et al. Dual determination of angiotensin- induced neointima formation in rats. Hypertension converting enzyme and angiotensin-II type 1 receptor 1991; 18: II60–II64. genotypes as predictors of restenosis after coronary 20 Capron L, Heudes D, Chajara A, Bruneval P. Effect of angioplasty. Am J Cardiol 1998; 81: 79–81. ramipril, an inhibitor of angiotensin converting 40 Hamon M et al. Relation between the deletion poly- enzyme, on the response of rat thoracic aorta to injury morphism of the angiotensin-converting enzyme gene with a balloon catheter. J Cardiovasc Pharmacol 1991; and late luminal narrowing after coronary angioplasty. 18: 207–211. Circulation 1995; 92: 296–299. 21 Kauffman RF et al. Losartan, a nonpeptide angiotensin 41 Samani NJ et al. Insertion/deletion polymorphism in II (Ang II) receptor antagonist, inhibits neointima for- the angiotensin-converting enzyme gene and risk of mation following balloon injury to rat carotid arteries. restenosis after coronary angioplasty. Lancet 1995; Life Sci 1991; 49: L223–L228. 345: 1013–1016. 22 Harrison EG, Jr., McCormack LJ. Pathologic classi- 42 Grimbert P et al. Genetic study of renal artery fibro- fication of renal arterial disease in renovascular hyper- muscular dysplasia. Arch Mal Coeur Vaiss 1998; 91: tension. Mayo Clin Proc 1971; 46: 161–167. 1069–1071. 23 Weber KT, Brilla CG. Pathological hypertrophy and 43 Sheps SG, Kincaid OW, Hunt JC. Serial renal function cardiac interstitium. Fibrosis and renin-angiotensin- and angiographic observations in idiopathic fibrous aldosterone system. Circulation 1991; 83: 1849–1865. and fibromuscular stenoses of the renal arteries. Am J 24 Weber KT et al. Vascular remodeling and mineralocor- Cardiol 1972; 30: 55–60. ticoids. J Endocrinol Invest 1995; 18: 533–539. 44 Foster JH et al. Hypertension and fibromuscular dys- 25 Weber KT. Aldosterone and spironolactone in heart plasia of the renal arteries. Surgery 1969; 65: 157–181. failure. N Engl J Med 1999; 341: 753–755. 45 Schreiber MJ, Pohl MA, Novick AC. The natural his-

Journal of Human Hypertension RA system polymorphisms in FMD A Bofinger et al 190 tory of atherosclerotic and fibrous renal artery disease. 51 Muller DN et al. Vascular angiotensin-converting Urol Clin North Am 1984; 11: 383–392. enzyme expression regulates local angiotensin II. 46 Kincaid OW, Davis GD, Hallermann FJ, Hunt JC. Fibro- Hypertension 1997; 29: 98–104. muscular dysplasia of the renal arteries. Arterio- 52 Gallagher PE et al. Estrogen regulation of angiotensin- graphic features, classification, and observations on converting enzyme mRNA. Hypertension 1999; 33: natural history of the disease. Am J Roentgenol Rad- 323–328. ium Ther Nucl Med 1968; 104: 271–282. 53 Brosnihan KB, Li P, Ganten D, Ferrario CM. Estrogen 47 Lindpaintner K et al. A prospective evaluation of an protects transgenic hypertensive rats by shifting the angiotensin-converting-enzyme gene polymorphism vasoconstrictor-vasodilator balance of RAS. Am J Phy- and the risk of ischemic heart disease. N Engl J Med siol 1997; 273: R1908–R1915. 1995; 332: 706–711. 54 Brosnihan KB et al. Effects of chronic hormone 48 Doria A, Ji L, Warram JH, Krolewski AS. Ddel poly- replacement on the renin-angiotensin system in cyno- morphism in the AGTR1 gene. Hum Mol Genet 1994; molgus monkeys. J Hypertens 1997; 15: 719–726. 3: 1444. 55 Li J, Hu H-Y, Zhao Y-N. Serum angiotensin-converting 49 Hunley TE et al. Angiotensin converting enzyme gene enzyme activity in pregnancy-induced hypertension. polymorphism: potential silencer motif and impact on Gynecol Obstet Invest 1992; 33: 138–141. progression in IgA nephropathy. Kidney Int 1996; 49: 56 Proudler AJ et al. Hormone replacement therapy and 571–577. serum angiotensin-converting-enzyme activity in post- 50 Rutledge DR, Browe CS, Kubilis PS, Ross EA. Analysis menopausal women. Lancet 1995; 346: 89–90. of two variants of the angiotensinogen gene in essen- tial hypertensive African-Americans. Am J Hypertens 1994; 7: 651–654.

Journal of Human Hypertension