AT1-Receptor Blockade and the Kidney: Importance of Non-ACE Pathways in Health and Disease

AT1-receptor blockade and the kidney: importance of non-ACE pathways in health and disease

NK Hollenberg Brigham and Women’s Hospital and Harvard Medical School, Departments of Radiology and Medicine, Boston, Massachusetts, USA

Large-scale trials with angiotensin converting enzyme is increased in diabetic patients, and comparison of the

(ACE) inhibitors and angiotensin II type 1 (AT1)-receptor renal vascular responses to captopril and candesartan blockers have clearly shown that blockade of the renin- shows a strong correlation between the effects of ACE angiotensin system reduces the deterioration in renal inhibition and AT1-receptor blockade, indicating that the function associated with diabetes. AT1-receptor block- deleterious effects of renin-angiotensin system acti- ers represent a more rational approach to blockade of vation in diabetes are mediated largely through angio- this system than ACE inhibitors, due to the presence of tensin II. The presence of multiple risk factors, such as non-ACE pathways of angiotensin II formation. Studies genetic predisposition, hyperglycaemia, obesity and in healthy volunteers maintained on a low-salt diet indi- tissue damage, places diabetic patients at high risk of cate that such pathways account for approximately 30– disease related to activation of the renin-angiotensin 40% of total angiotensin II formation, and this ﬁgure system. Effective and early blockade of this system is increases to 60–70% in individuals maintained on a therefore an important aspect of management. high-salt diet (resembling the situation in most human Journal of Human Hypertension (2002) 16, S59–S63. populations). Activation of the renin-angiotensin system doi:10.1038/sj.jhh.1001441

Keywords: angiotensin II; angiotensin converting enzyme; candesartan; captopril; diabetes; renin-angiotensin system

Introduction vention in diabetic patients with microalbuminuria, and of using adequate doses to achieve effective The demonstration that treatment with angiotensin receptor blockade, and raise the question of how converting enzyme (ACE) inhibitors significantly 1 blockade of the renin-angiotensin system can achi- reduces the progression of diabetic nephropathy eve such substantial risk reductions in an essentially focused attention on the role of the renin- metabolic disease. angiotensin system in the kidney, and led to the widespread use of ACE inhibitor therapy in patients Therapeutic targets for blockade of the at risk of nephropathy. Moreover, recent large out- renin-angiotensin system come studies with angiotensin II type 1 (AT1)-receptor blockers have shown that these agents reduce the Blockade of the renin-angiotensin system was first deterioration in renal function, and delay the onset achieved with ACE inhibitors, which were of end-stage renal failure, in diabetic patients.2–4 In developed in the 1970s after the chance finding that the Irbesartan Diabetic Nephropathy Trial (IDNT)2 a component of pit viper venom had a powerful and the Reduction of Endpoints in NIDDM with the hypotensive effect. Arguably, however, a pharma- Angiotensin II Antagonist Losartan (RENAAL)3 cologist aiming to block the renin-angiotensin target study, treatment with an AT1-receptor blocker would not have selected ACE as a therapeutic target. reduced the incidence of end points by approxi- Renin is the rate-limiting step in the pathway lead- mately 20%. By contrast, in the Irbesartan in ing to angiotensin II formation (Figure 1), and thus patients with type 2 diabetes and Microalbuminuria would represent a rational target for intervention; (IRMA II) study,4 the onset of overt nephropathy was however, the development of suitable candidate reduced by 39% in patients receiving irbesartan, 150 compounds has proved difficult because of prob- mg, and by 70% in those receiving 300 mg. Such lems with synthesis and bioavailability.5 Moreover, findings highlight the importance of early inter- angiotensin II is produced by a number of pathways that do not involve ACE, such as chymase, trypsin and cathepsin G (Figure 1). Blockade of the AT1- Correspondence: Professor NK Hollenberg, Brigham and receptor thus represents a more rational therapeutic Women’s Hospital, 75 Francis St, Boston, MA 02115, USA target than ACE inhibition. AT1-receptor blockade in the kidney NK Hollenberg S60

Figure 1 The renin-angiotensin system.

Non-ACE pathways of angiotensin II formation Non-ACE pathways in the kidney Evidence for the contribution of non-ACE pathways The relative contributions of ACE and non-ACE to angiotensin II formation in the human pathways of angiotensin II formation in the human cardiovascular system comes from a study in which kidney have been investigated in experiments in angiotensin II was measured in plasma and myocar- healthy volunteers fed a low-salt diet.11 If the dial tissue from normal and failing hearts obtained increase in renal plasma flow that occurs during at transplantation.6 In plasma, conversion of blockade of the renin-angiotensin system were angiotensin I to angiotensin II can be completely solely attributable to blockade of angiotensin II pro- blocked by ACE inhibitors; by contrast, in myocar- duced via ACE, then renin inhibitors, ACE inhibi- dial membrane preparations, ACE inhibitors have tors, and AT1-receptor blockers should produce only a limited effect on angiotensin II formation similar results. By contrast, if the renal vascular whereas serine protease inhibitors reduce angioten- response is partly due to a reduced breakdown of sin formation by approximately 80%. Similar results vasodilator kinins such as bradykinin following were obtained in studies with human gastroepiploic ACE inhibition, the effect of ACE inhibitors would artery preparations, in which chymase inhibitors be expected to be greater than that of renin inhibi- reduced the contractile response to angiotensin II, tors or AT1-receptor blockers. formed locally following addition of angiotensin I, In one study, participants received 90-min to a greater extent than ACE inhibitors, and the two infusions of the renin inhibitor enalkiren, 256 ␮g/kg inhibitors in combination produced almost com- or 512 ␮g/kg, at 2-day intervals.12 The lower dose plete abolition of the response.7 In these was at the top of the dose-response curve for inhi- experiments, approximately 30–40% of angiotensin bition of angiotensin II formation. The increase in II formation was attributable to ACE, and the renal plasma flow following administration of enal- remainder to non-ACE pathways. There are, how- kiren was substantially greater than that seen in pre- ever, marked species differences; in humans and vious studies with ACE inhibitors, and was dose- other primates, angiotensin I is the sole substrate for dependent; a greater response was achieved with the chymase and angiotensin II is the sole product, higher dose, even though plasma angiotensin II con- whereas in rodents and rabbits this enzyme is prim- centrations were not further reduced (Figure 2). This arily responsible for degradation of angiotensin II, suggests that the response to the renin inhibitor was rather than formation. For this reason, studies with attributable to inhibition of non-ACE pathways animal models have often been misleading. An within the kidney. example of this is provided by early studies in rats, This study was repeated using different renin which showed that ACE inhibitors were strong inhibitors and three ACE inhibitors, each at the top inhibitors of neointima formation after vascular of their dose-response ranges.13 The combined data injury.8,9 However, large clinical trials, notably the from these studies showed that the renal vascular MERCATOR Study,10 failed to show significant response to renin inhibition was markedly greater beneficial effects of ACE inhibitors on restenosis than the response to ACE inhibition (approximately rates after percutaneous transluminal coronary 140 mL/min/1.73 m2 vs 90–100 mL/min/1.73 m2). In angioplasty. It is now widely accepted that this dis- further studies, the AT1-receptor blockers crepancy between animal studies and clinical trial candesartan, irbesartan and eprosartan were found outcome is largely due to the presence of non-ACE to produce comparable increases in renal plasma pathways of angiotensin II formation in humans. flow to renin inhibitors.11,14,15 Together, these data

Journal of Human Hypertension AT1-receptor blockade in the kidney NK Hollenberg S61

Figure 2 Changes in plasma angiotensin II and aldosterone concentrations, and renal plasma ﬂow, following administration of the renin inhibitor enalkiren in healthy volunteers maintained on a low-salt diet.12 Reprinted with permission from Cordero P et al. Renal and endocrine responses to a renin inhibitor, enalkiren, in normal humans. Hypertension 1991; 17: 510–516.

diet, the responses to both agents were attenuated (Figure 3). In this situation, however, the response to candesartan was even more markedly greater than the response to captopril (97 ± 20 mL/min/1.73 m2 vs 30 ± 15 mL/min/1.73 m2, respectively, P Ͻ 0.01; Figure 3), since the response to captopril was dimin- ished to a greater extent than the response to candesartan. This suggests that the high-salt diet had less effect on non-ACE pathways of angiotensin II formation than on ACE; in this situation, non-ACE pathways accounted for approximately 60–70% of renal angiotensin II formation, compared with 30–40% in volunteers maintained on low-salt diets. This finding may be clinically relevant since the daily salt Figure 3 Increase in renal plasma flow following treatment with intake of most people is closer to that in the high- candesartan cilexetil, 16 mg, or captopril, 25 mg, in healthy vol- salt diet used in this study than to the low salt-diet. unteers maintained on low-salt or high-salt diets.16 Reproduced Since non-ACE pathways appear to make a greater with permission from Hollenberg NK et al. Salt intake and non- ACE pathways for intrarenal angiotensin II generation in man. contribution to angiotensin II production when salt JRAAS 2001; 2:14–18. intake is high, AT1-receptor blockers might be expected to produce a greater therapeutic effect than ACE inhibitors. suggest that the renal vascular response to ACE inhibition systematically under-estimates the contri- Activation of the renin-angiotensin bution of angiotensin II to renovascular tone during activation of the renin-angiotensin system.11 system in diabetic nephropathy Subsequent studies used the relative renal vascu- In contrast to diabetic retinopathy, the incidence of lar response to ACE inhibitors and AT1-receptor which increases progressively with time, the inci- blockers as an index of non-ACE formation of angio- dence of diabetic nephropathy reaches a plateau tensin II. In one series of experiments, 65 healthy after 20–25 years, by which time approximately 30– volunteers maintained on low-salt or high-salt diets 40% of patients with type 1 diabetes have clinical received either captopril, 25 mg, or candesartan nephropathy. The risk of developing frank nephro- cilexetil, 16 mg, and renal haemodynamic responses pathy is determined by both glycaemic control17 and were measured over 4 h.16 In volunteers maintained genetic factors. For example, there is a strong associ- on a low-salt diet, the increase in renal plasma flow ation between ACE insertion/deletion (I/D) poly- following candesartan (mean 165 ± 14 mL/min/1.73 morphism and the risk of nephropathy.18,19 m2) was significantly greater than that following cap- In subjects maintained on a high-salt diet, the topril (118 ± 12 mL/min/1.73 m2, P Ͻ 0.01), as renal vascular response to captopril is markedly expected. In volunteers maintained on a high-salt increased in patients with type 1 diabetes, compared

Journal of Human Hypertension AT1-receptor blockade in the kidney NK Hollenberg S62 mediated primarily via suppression of angiotensin II production. In contrast to type 1 diabetes, type 2 diabetes is normally characterised by low levels of plasma renin activity. Despite this, the renal vascular response to captopril is markedly greater in patients with type 2 diabetes than in those with type 1 diabetes, indicating greater activation of the renin- angiotensin system (unpublished observations). Comparison of the renal vascular responses to captopril and candesartan in patients with type 2 diabetes showed an even stronger correlation than in type 1 diabetes (r = 0.886, P = 0.001). Such ﬁndings indi- cate that angiotensin II is a major risk factor for tissue damage in type 2 diabetes.

Figure 4 Correlation between the change in renal plasma flow fol- Dose issues for suppression of the lowing administration of captopril, 25 mg, and candesartan cilex- renin-angiotensin system etil, 16 mg, in patients with type 1 diabetes maintained on a high- salt diet.20 Reproduced with permission from Lansang MC et al. Studies such as those described above show that the Renal vascular responses to captopril and to candesartan in renin-angiotensin system is activated in diabetes, patients with type 1 diabetes mellitus. Kidney Int 2001; 59: 1432–1438. and it is widely accepted that inhibition of this system reduces the progression of nephropathy in diabetic patients. These renoprotective effects appear to with non-diabetic individuals.15 Glomerular hyper- be independent of, and additional to, effects on filtration, increased glycosylated haemoglobin, and blood pressure. However, most outcome trials with high plasma renin activity, were all factors inhibitors of the renin-angiotensin system have used predisposing to an increased response to captopril the doses needed to reduce blood pressure, and (unpublished observations). there is increasing evidence that optimal renoprotec- In a study involving 12 patients with type 1 dia- tive effects may require higher doses. betes who were maintained on a high-salt diet, In studies using the remnant kidney model in rats, patients received captopril, 25 mg, and candesartan high-dose treatment with ACE inhibitors produced cilexetil, 16 mg, on separate occasions during the significantly greater reductions in glomerulo- same week.20 Renal plasma flow and glomerular fil- sclerosis following subtotal nephrectomy than low- tration were measured before, and for 4 h after, dose treatment.21 Similarly, dose-response studies administration. Both drugs produced a significant in humans have shown that ACE inhibitors exert increase in renal plasma flow, compared with base- renoprotective effects at doses higher than those line, with no change in glomerular filtration rate. needed to control blood pressure. In one such study There was a strong correlation between the with lisinopril, for example, the maximal reduction responses to the two agents (r = 0.86, P Ͻ 0.001; in blood pressure was seen with the lowest dose (5 Figure 4), indicating that the effect of captopril is mg); however, increasing reductions in proteinuria

Figure 5 Twenty-four hour urinary protein excretion in normotensive patients with substantial proteinuria (Ͼ1.56 g/day) at baseline, treated with increasing doses of candesartan.24 Data are presented as mean ± s.e.m., n=4–10. Reproduced with permission from Weinberg MS et al. The effect of high dose angiotensin II receptor blockade beyond maximal recommended doses in reducing urinary protein excretion. JRAAS 2001; 2 (Suppl 1): S196–S198.

Journal of Human Hypertension AT1-receptor blockade in the kidney NK Hollenberg S63 were seen with doses of between 5 mg and 20 mg.22 enzyme inhibition. J Cardiovasc Pharmacol 1990; 16 (Suppl 4): S42–S49. Similar findings have been reported with AT1-receptor blockers. In one study, for example, hypertensive 9 Powell JS et al. Inhibitors of angiotensin-converting patients with non-diabetic renal disease received enzyme prevent myointimal proliferation after vascu- placebo, losartan, 50 mg and 100 mg, and placebo lar injury. Science 1989; 245: 186–188. 23 10 The MERCATOR Study Group. Does the new angioten- for 1 month each in sequence. Both blood pressure sin converting enzyme inhibitor cilazapril prevent and urinary protein excretion were reduced follow- restenosis after percutaneous transluminal coronary ing treatment with losartan, 50 mg; the higher dose angioplasty? Results of the MERCATOR study: a multi- produced a further reduction in proteinuria but had center, randomized, double-blind placebo-controlled no additional effect on blood pressure. trial. Circulation 1992; 86: 100–110. In a recent study, 10 patients with proteinuria 11 Hollenberg NK. The renin-angiotensin-aldosterone (Ͼ1.5 g/day; mean 4.4 g/day) were treated with can- system, blockade and diabetic nephropathy. JRAAS desartan cilexetil at initial doses of 16–32 mg/day, 2001; 2 (Suppl 1): S185–S187. and after 1–2 months the doses were titrated 12 Cordero P et al. Renal and endocrine responses to a upwards in 32 mg increments to a maximum of 96 renin inhibitor, enalkiren, in normal humans. Hyper- 24 tension 1991; 17: 510–516. mg/day. Candesartan, 16–32 mg, produced a 13 Hollenberg NK, Fisher NK. Renal circulation and marked reduction in proteinuria, and further blockade of the renin-angiotensin system. Is angioten- decreases were seen when doses were increased sin-converting enzyme inhibition the last word? above the normal recommended range (Figure 5); in Hypertension 1995; 26: 602–609. patients receiving 96 mg/day, mean urinary protein 14 Price DA, De’Oliveira JM, Fisher NDL, Hollenberg NK. excretion was reduced to 1.1 g/day. High-dose treat- Renal hemodynamic response to an angiotensin II ment appeared to be safe and well tolerated, although antagonist, eprosartan, in healthy men. Hypertension further investigation is necessary before specific rec- 1997; 39: 240–246. ommendations can be made for clinical practice.24 15 Price DA et al. The paradox of the low-renin state in diabetic nephropathy. J Am Soc Nephrol 1999; 10: 2382–2391. Conclusions 16 Hollenberg NK et al. Salt intake and non-ACE pathways for intrarenal angiotensin II generation in man. Diabetic patients are at high risk of renal (and other) JRAAS 2001; 2:14–18. disease related to activation of the renin-angiotensin 17 The Diabetes Control and Complications Trial system because of the presence of multiple risk fac- Research Group. The effect of intensive treatment of tors. These include genetic predisposition, hypergly- diabetes on the development and progression of long- caemia leading to activation of the renin- term complications in insulin-dependent diabetes angiotensin system,25 obesity and tissue injury. Early mellitus. N Engl J Med 1993; 329: 977–986. blockade of the renin-angiotensin system is therefore 18 Marre M et al. Contribution of genetic polymorphism in the renin-angiotensin system to the development of an important element of diabetes management. renal complications in insulin-dependent diabetes: Genetique de la Nephropathie Diabetique (GENEDIAB) References study group. J Clin Invest 1997; 99: 1585–1595. 19 Hadjadj S et al. Prognostic value of angiotensin-I con- 1 Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect verting enzyme I/D polymorphism for nephropathy in of angiotensin-converting enzyme inhibition on dia- type 1 diabetes mellitus: a prospective study. J Am Soc betic nephropathy. N Engl J Med 1993; 329: 1456– Nephrol 2001; 12: 541–549. 1462. 20 Lansang MC et al. Renal vascular responses to capto- 2 Lewis EJ et al. Renoprotective effect of the angiotensin pril and to candesartan in patients with type 1 diabetes receptor antagonist irbesartan in patients with nephro- mellitus. Kidney Int 2001; 59: 1432–1438. pathy due to type 2 diabetes. N Engl J Med 2001; 345: 21 Ikoma M et al. Cause of variable therapeutic efficacy 851–860. of angiotensin converting enzyme inhibitor on glom- 3 Brenner BM et al. Effects of losartan on renal and car- erular lesions. Kidney Int 1991; 40: 195–202. diovascular outcomes in patients with type 2 diabetes 22 Palla R et al. Effect of increasing doses of lisinopril on and nephropathy. N Engl J Med 2001; 345: 861–869. proteinuria of normotensive patients with IgA nephro- 4 Parving H-H et al. The effect of irbesartan on the devel- pathy and normal renal function. Int J Clin Pharmacol opment of diabetic nephropathy in patients with type Res 1994; 14:35–43. 2 diabetes. N Engl J Med 2001; 345: 870–878. 23 Gansevoort RT et al. Effects of the angiotensin II antag- 5 Fisher ND, Hollenberg NK. Is there a future for renin onist losartan in hypertensive patients with renal dis- inhibitors? Expert Opin Investig Drugs 2001; 10: 417–426. ease. J Hypertens 1994; 12 (Suppl): S37–S42. 6 Urata H et al. Angiotensin II-forming pathways in nor- 24 Weinberg MS, Weinberg AJ, Cord R, Zappe DH. The mal and failing human hearts. Circ Res 1990; 66: effect of high-dose angiotensin II receptor blockade 883–890. beyond maximal recommended doses in reducing uri- 7 Okunishi H et al. Marked species-difference in the vas- nary protein excretion. JRAAS 2001; 2 (Suppl 1): cular angiotensin II-forming pathways: humans versus S196–S198. rodents. Jap J Pharmacol 1993; 62: 207–210. 25 Osei SY et al. Hyperglycemia and angiotensin- 8 Powell JS et al. The proliferative response to vascular mediated control of the renal circulation in healthy injury is suppressed by angiotensin-converting humans. Hypertension 1999; 3: 559–564.

Journal of Human Hypertension