(Hypertension. 2001;38:243.)
© 2001 American Heart Association, Inc.
Scientific Contributions |
From the Department of Medicine-Nephrology, University of Erlangen-Nürnberg (K.F.H., R.V., H.K., A.H., C.S., R.E.S.), Erlangen, Germany; Department of Pediatrics, University of Virginia (K.F.H., S.R.B., R.A.G.), Charlottesville; and Franz-Volhard-Clinic, Humboldt University (D.N.M.), Berlin-Buch, Germany.
Correspondence to Karl F. Hilgers, MD, Nephrology Research Laboratory, Loschgestrasse 8, D-91054 Erlangen, Germany. E-mail karl.hilgers{at}rzmail.uni-erlangen.de
| Abstract |
|---|
|
|
|---|
Key Words: renin angiotensin angiotensinogen mRNA vessels perfusion endothelium
| Introduction |
|---|
|
|
|---|
The role of the endothelium and endothelium-dependent mediators has been a focus of research in the last decade,8,9 but its contribution to the local vascular generation of Ang I and Ang II is still controversial. ACE is present on endothelial cells, and some authors have described an obligatory role of the endothelium for the conversion of Ang I to Ang II.10 Contradictory findings, however, were also reported.11 Attempts to localize immunoreactive renin or Ang II in extrarenal vessels are scarce and have yielded conflicting results.1215 In cell culture, endothelial cells and smooth muscle cells are both capable of generating renin and angiotensin peptides.16,17
Therefore, the present study was carried out to clarify the role of the endothelium in the local, vascular formation of Ang I and Ang II. In an isolated, perfused resistance vessel bed, we tested the hypotheses that the endothelium is necessary for (1) the synthesis and/or uptake of renin, (2) the generation of Ang I, and (3) the conversion of Ang I to Ang II. The synthesis and localization of renin in large vessels also were investigated by reverse-transcription polymerase chain reaction (RT-PCR) and immunohistochemistry, respectively.
| Methods |
|---|
|
|
|---|
Endothelium Denudation
The endothelium was destroyed by perfusion with 2.6% CHAPS for 225 seconds. Fifteen minutes later, bolus doses of acetylcholine to elicit endothelium-dependent vasodilatation were applied after preconstriction with methoxamine.18 Bolus doses of norepinephrine were applied to constrict vessels. Some hindquarters were perfusion fixed with 4% paraformaldehyde and processed for electron microscopy.
Generation of Ang I and Ang II
The spontaneous release of Ang I was measured by "online" peptide extraction, high-performance liquid chromatography, and radioimmunoassay.18,19 Captopril (10 µmol/L) was added to the perfusate to inhibit ACE activity.
Porcine renin (final concentration in the perfusate, 300 mU/L), prepared as described previously,18 was infused for 15 minutes in the presence of captopril. The renin preparation did not contain prorenin, angiotensin peptides, or angiotensinogen. Ang I (final concentrations, 1 and 10 nmol/L) and Ang II (final concentration, 5 nmol/L) were infused for 20 minutes; captopril was added during the last 5 minutes. Peptides were also infused into perfusion channels without hindquarter. Ang I and Ang II were measured by radioimmunoassay, and the conversion and disappearance rates of the peptides were calculated as described previously.19
Identical concentrations of renin, angiotensin peptides, and saline (vehicle) were infused for 20 minutes into endothelium-denuded or -intact hindquarters preconstricted with methoxamine.18 During the second 10 minutes, 10 µmol/L captopril was added. All infusions were performed at a flow rate of 100 µL/min. Only 1 infusion of an angiotensin peptide or renin per preparation was performed to avoid excessive edema formation.20
Results are expressed as mean and SEM. Students t test for independent samples was used to assess significant (P<0.05) differences between intact and endothelium-denuded preparations, and the t test for paired samples was used to assess significant responses to captopril.
RT-PCR for Renin mRNA in Vascular Tissue
RNA was extracted from aortic, mesenteric, and kidney tissue.21 The RNA quality was checked by gel electrophoresis and by RT-PCR for a housekeeping gene.21,22 One to 5 µg aortic or mesenteric RNA or 5 to 500 ng kidney RNA underwent RT-PCR with the upstream primer 182 (5'-GTCAAACTTGGCCAGCATGA-3') or 204 (5'-ACTTCCTCCTTTAGCACCTC-3') and the downstream primer 181 (5'-ATGCCTCTCTGGGCACTCTT-3') or 203 (5'-AGGATGC-CTCTCTGGGCACT-3') as described.21 Southern hybridization of PCR products21 was performed with a full-length rat renin cDNA probe. Some PCR products were cloned and sequenced.21
Renin Immunohistochemistry
Immunohistochemistry for renin of saline-perfused aortic tissue was performed as previously described,22 except the dilution of the rabbit anti-renin antiserum was 1:200.
| Results |
|---|
|
|
|---|
|
Endothelium denudation reduced the release of Ang I from perfused rat hindquarters (Figure 2). After bilateral nephrectomy, Ang I release was no longer detectable (Figure 2). Infusion of exogenous renin for 15 minutes induced Ang I release, which was sustained in intact preparations for
30 minutes after cessation of renin infusion (Figure 3). In endothelium-denuded preparations, the initial release of Ang I during renin infusion was higher than in intact preparations (Figure 3), whereas there was no sustained phase of Ang I release.
|
|
The conversion of Ang I to Ang II was virtually abolished by endothelium denudation, regardless of whether Ang I was infused at 10 nmol/L (Figure 3) or at 1 nmol/L (n=5 each, data not shown). At all time points during Ang I infusion, levels of immunoreactive Ang II in the venous effluent from endothelium-denuded preparations were close to the detection limit (10 pmol/L). In endothelium-intact preparations, the conversion of Ang I was completely suppressed by captopril (Figure 3). In contrast, endothelium denudation had no effect on the disappearance of infused Ang II: 64±4% of infused Ang II was degraded and/or taken up during 1 pass through endothelium-intact preparations (n=6), and 66±5% was degraded or taken up during 1 pass through endothelium-denuded preparations (n=6, P>0.1). Captopril did not affect the disappearance rate of Ang II. The nonACE-mediated disappearance of Ang I in the presence of captopril was not affected by endothelium denudation (72±8% in endothelium-denuded versus 68±6% in intact preparations for the higher dose of Ang I; n=5 each; P>0.1) .
Methoxamine increased the perfusion pressure from 23.3±0.5 to 118.2±0.9 mm Hg in endothelium-denuded preparations (n=24) and from 22.4±0.5 to 112.7±0.8 mm Hg in intact rat hindquarters (n=24). During infusion of saline for 20 minutes, perfusion pressure increased by 0.9±1.6 mm Hg in intact preparation (n=6) and by 20.6±5.1 mm Hg in endothelium-denuded hindquarters (n=6), accompanied by visible edema formation. The Table shows the net effects of agonists on perfusion pressure after subtraction of the pressure changes during saline infusion. Endothelium denudation abolished the pressor responses to renin and Ang I but also diminished the response to Ang II. Captopril abolished the pressor effects of renin and Ang I without affecting the response to Ang II (Table).
|
With the use of RT-PCR to detect renin mRNA, all primer combinations yielded PCR products of the expected size if 5 to 500 ng kidney RNA was used in the RT reaction (Figure 4). With 0.5 to 5 µg aortic or mesenteric RNA subjected to RT-PCR, however, 3 of 4 primer combinations did not generate a specific product. Only 1 primer combination (181 and 182) yielded a product of the expected size from aortic RNA (Figure 4) but not from mesenteric RNA (data not shown). This RT-PCR product from aortic RNA exhibited weak hybridization with a renin cDNA compared with the strong hybridization of the same cDNA with a RT-PCR product from kidney RNA (Figure 4). We cloned the PCR products and sequenced 5 clones containing the RT-PCR product from aortic RNA and 5 clones containing the RT-PCR product from kidney RNA. The product from aortic RNA exhibited the correct size and contained both primers but did not contain the renin cDNA sequence, whereas all 5 clones from kidney RNA contained the expected renin cDNA sequence. The weak hybridization of the aortic RT-PCR product with a renin cDNA was most likely due to the presence of the primers, although we cannot fully exclude the possibility that rare renin cDNA fragments may have been responsible. Our experiments, however, did not provide evidence for a local synthesis of renin in vascular tissue within the detection limits of our method.
|
Staining of aortic sections for renin (Figure 5) demonstrated renin immunoreactivity in the endothelial layer. Attempts to localize renin immunoreactivity more precisely by immuno-gold electron microscopy were not successful, presumably because of the low amounts of renin protein. Aortic sections from bilaterally nephrectomized rats did not exhibit renin staining in the endothelial or smooth muscle layers (Figure 5).
|
| Discussion |
|---|
|
|
|---|
It has been widely acknowledged that the endothelium contributes importantly to the conversion of Ang I to Ang II,10 but others described additional extraendothelial ACE activity in the smooth muscle or adventitial layers11 of the vessel wall. Our data show that the endothelium is responsible for most if not all of the conversion of Ang I to Ang II in skeletal muscle resistance vessels. In contrast to endothelium denudation by CHAPS, mechanical disruption of the endothelium of large vessels spares the vasa vasorum. Thus, ACE in the endothelium of vasa vasorum might explain part of this controversy.11,23 The disadvantage of CHAPS is its capacity to damage the vascular smooth muscle layer. CHAPS did not decreases the response to norepinephrine, but we cannot exclude the possibility that CHAPS had some effect on vasoconstriction, eg, at higher doses of norepinephrine24 or after Ang II infusion.
Furthermore, part of Ang II formed from Ang I in deeper layers of the vascular wall may not be released from the tissue. In coronary vascular and myocardial tissue of the heart, little if any Ang II generated locally from Ang I reaches the vascular lumen.25,26 In the rat hindquarter, most of the locally formed Ang I and II (estimated 50% to 90%) is not released but metabolized within the tissue.19 Nevertheless, the absence of a pressor response to renin and Ang I after endothelium denudation argues against a major extraendothelial conversion of Ang I to Ang II, in agreement with the peptide measurements.
Interpretation of these data is difficult, however, because the response to Ang II was also significantly diminished by endothelium destruction. This unexpected observation might be explained by increased edema formation in endothelium-denuded hindquarters, which may induce an apparent increase in vascular resistance,20 thereby masking vasoconstrictor effects of peptides. Alternatively, an intact endothelium may be a prerequisite for the full expression of the pressor actions of Ang II.27,28 Whether endothelial mediators like endothelin-1 or vasoconstrictor prostanoids might contribute to the pressor response to Ang II29,30 requires further study.
Our results extend previous observations by pointing out the role of the endothelium in the uptake of renin and formation of Ang I. Both the spontaneous release of Ang I during perfusion and the sustained Ang I release after renin infusion were greatly diminished in the absence of the endothelium. In addition, renin immunoreactivity was localized to the endothelial layer. Whereas a previous report14 localized immunoreactive renin to the media of vessels after infusion of heterologous renin, Okamura et al13 and we stained for endogenous renin and detected immunostaining in the endothelial layer. Of note, our perfusion experiments, which were performed with heterologous (porcine) renin, support the same conclusion as the immunostainings for homologous (rat) renin. We are not aware of perfusion experiments with rat renin, but perfusion of rat vessels with renin of human31 or murine32 origin yielded results comparable to those obtained with porcine renin,18,19 indicating that renin uptake by these vessels does not depend on the species of origin.
Even in the absence of the endothelium, infusion of renin caused Ang I formation. This observation rules out the possibility that CHAPS itself might interfere with the renin-angiotensinogen reaction. Furthermore, the data demonstrate that angiotensinogen is still available after endothelium denudation. Vascular angiotensinogen could be synthesized locally33 or taken up from the circulation.34 We can only speculate why endothelium denudation led to an initially higher Ang I release after renin infusion. For example, an intact endothelial layer might limit the equilibration of tissue and luminal compartments, thereby reducing the spillover of tissue Ang I into the lumen.19
Endothelium denudation affected neither the disappearance rate of Ang II nor the disappearance of Ang I after ACE inhibition. The molecular mechanisms of the disappearance of these peptides are unclear (except for the important role of ACE in the metabolism of Ang I), but the disappearance rate during perfusion with an artificial medium closely mirrors the metabolism of these peptides in the same vascular beds in vivo.57 Surprisingly, neither Ang II receptors32 nor angiontensinases19 play a major role. Therefore, we cannot exclude the possibility that the apparently unaltered disappearance rate after CHAPS treatment is the result of multiple complex alterations. We want to emphasize, however, that this observation excludes the possibility that the observed changes in Ang I and II formation after CHAPS treatment are mimicked by an altered angiotensin disappearance rate. The formation of these peptides, not the disappearance, is endothelium dependent.
The sustained Ang I release after cessation of renin infusion was abolished by endothelium denudation, pointing to retention of renin by the endothelial layer. The molecular mechanism of renin uptake is unknown. Some authors postulated a receptor specific for renin35 or for prorenin and renin.36 Our recent results in perfused rat hindquarters mitigate against the latter notion.31 Relatively nonspecific macromolecule carriers, however, may also be involved in endothelial renin uptake.37 Regardless of the precise mechanism, renin uptake by the endothelium seems to be the main pathway for local vascular Ang I formation in our preparation. We did not detect renin mRNA in vascular tissue by an RT-PCR method that readily detected a signal from a 1000-fold-lower amount (5 ng) of kidney RNA. Previous studies on renin mRNA in vascular tissue reported positive3841 and negative32,42 findings. Some of these authors3941 may have used more sensitive methods than we did, and even very low amounts of local renin RNA could lead to physiologically important renin synthesis. Furthermore, local renin synthesis in vascular tissue may occur in some forms of hypertension43 or in damaged vessels,41 conditions not reflected in our experiments. On the other hand, illegitimate transcription may account for some positive findings obtained by high amounts of RNA in the RT or by the use of many PCR cycles.1
We did not rely on negative RT-PCR findings alone; our results on Ang I release and renin staining agree with the RT-PCR findings. Our data point to the endothelial uptake of renin, which appears to be the predominant source of vascular angiotensin formation.
| Acknowledgments |
|---|
Received October 4, 2000; first decision November 1, 2000; accepted February 20, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Pignone, A. D. Rosso, K B. Brosnihan, F. Perfetto, R. Livi, G. Fiori, S. Guiducci, M. Cinelli, V. Rogai, A. Tempestini, et al. Reduced circulating levels of angiotensin-(1 7) in systemic sclerosis: a new pathway in the dysregulation of endothelial-dependent vascular tone control Ann Rheum Dis, October 1, 2007; 66(10): 1305 - 1310. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Paul, A. Poyan Mehr, and R. Kreutz Physiology of local Renin-Angiotensin systems. Physiol Rev, July 1, 2006; 86(3): 747 - 803. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Hartner, M. Porst, B. Klanke, N. Cordasic, R. Veelken, and K. F. Hilgers Angiotensin II formation in the kidney and nephrosclerosis in Ren-2 hypertensive rats Nephrol. Dial. Transplant., July 1, 2006; 21(7): 1778 - 1785. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.H. J. Danser and J. Deinum Renin, Prorenin and the Putative (Pro)renin Receptor Hypertension, November 1, 2005; 46(5): 1069 - 1076. [Full Text] [PDF] |
||||
![]() |
A. J. Danser and J. Deinum Spotlight on Renin: Renin, Prorenin and the Putative (Pro)renin Receptor Journal of Renin-Angiotensin-Aldosterone System, September 1, 2005; 6(3): 163 - 165. [PDF] |
||||
![]() |
T. F. Pedersen, O. B. Paulson, A. H. Nielsen, and S. Strandgaard Effect of nephrectomy and captopril on autoregulation of cerebral blood flow in rats Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H1097 - H1104. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. M. Carey and H. M. Siragy Newly Recognized Components of the Renin-Angiotensin System: Potential Roles in Cardiovascular and Renal Regulation Endocr. Rev., June 1, 2003; 24(3): 261 - 271. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |