| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Prince Henry’s Institute of Medical Research (E.Y.M.L., J.W.F., P.J.F., M.J.Y.), Clayton, Victoria 3168, Australia; and Department of Nephrology (D.J.N.-P.), Monash University, Victoria 3800, Australia
Address all correspondence and requests for reprints to: Morag J. Young, Dr. Prince Henry’s Institute of Medical Research, Endocrine Genetics, P.O. Box 5152, Clayton, Victoria 3168, Australia. E-mail: morag.young{at}princehenrys.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Elevated plasma aldosterone in the context of a high sodium intake produces inflammation and oxidative stress in the blood vessel wall in the first instance, followed by tissue remodeling in both the heart and kidney (4, 7). Experimental models of mineralocorticoid administration [aldosterone or deoxycorticosterone (DOC)] plus salt may provide an insight into the mechanisms underlying the beneficial effects of MR blockade observed in recent large-scale clinical trials. In progressive cardiac failure, low-dose spironolactone was added to standard care in the Randomized ALdactone Evaluation Study (8); subsequently the selective MR antagonist eplerenone was similarly used in left ventricular dysfunction after myocardial infarction [Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and SUrvival Study (9). Recent preliminary data from clinical studies have shown that MR blockade protects the kidneys and in particular decreases proteinuria (9, 10). These effects appear independent of beneficial blood pressure effects and can protect patients from vascular injury associated with diabetes mellitus and hypertension (11, 12). In particular, eplerenone proved more effective than enalapril in reducing proteinuria in both patients with stage 1 or 2 congestive heart failure and essential hypertension (12).
Aldosterone has also been clearly shown to be a mediator of renal inflammation and fibrosis in rat models of mineralocorticoid/salt hypertension (7). Renal injury, as assessed by inflammatory markers and fibrosis, produces severe vascular and glomerular sclerosis, fibrinoid necrosis and thrombosis, interstitial leukocyte infiltration, and tubular damage with regeneration (13). Proinflammatory events are significantly reduced by coadministration of the selective MR antagonist eplerenone from d 0, indicating that inflammation is a key driver in aldosterone/salt-induced renal injury, as is the case in the heart. The progression of renal injury after an immune or nonimmune insult is closely associated with the accumulation of leukocytes and fibroblasts in the damaged kidney; macrophage accumulation is a prominent feature of most forms of human glomerulonephritis and correlates with renal dysfunction (14). Macrophages have also been shown to mediate acute and chronic renal injury in experimental kidney disease (15, 16).
Studies of the mineralocorticoid/salt model of tissue remodeling have predominantly addressed the preventive effects of MR blockade rather than reversal of established fibrosis. We have recently shown that the coronary vascular inflammation and cardiac fibrosis established by 4 wk of DOC treatment can be reversed by the MR selective blocker eplerenone but not mineralocorticoid withdrawal alone (17). The aim of the present studies was thus to determine whether the tissue inflammation and early tissue remodeling induced in the kidney by DOC/salt treatment can be similarly reversed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals were killed by CO2 in air at 8 wk, except for the DOC4 group, which was killed at 4 wk. Kidneys were excised, weighed, bisected, and half fixed in buffered 4% paraformaldehyde for 6 h at 4 C and then rinsed and stored overnight in PBS before processing for paraffin embedding. The remainder of the fresh tissue was snap frozen in liquid nitrogen and mRNA extracted for RT-PCR analysis. Serum was isolated from an arterial blood sample for determination of creatinine levels.
Histological analysis
Tissue blocks were sectioned at 5-µm thickness onto glass slides. Renal cortical collagen levels were determined by picrosirius red staining and analyzed by Analytical Imaging Systems software (Imaging Research, Inc., Piscataway, NJ). Immunochemistry for inflammatory cytokines and ED-1-positive macrophages was performed using antibodies as previously described for studies in the heart (9). Hemeoxygenase 1 (HO-1) immunostaining was performed on 5-µM sections and were dewaxed and rehydrated through graded alcohols before antigen retrieval by 10 min boiling in standard citrate retrieval buffer. They were then blocked sequentially in 0.3% H2O2 in Tris-buffered saline (TBS), avidin (0.001% in TBS), biotin (0.001% in TBS), and normal goat serum (10%) plus BSA (5%) before overnight incubation at 4 C with HO-1 primary antibody (1:2000 dilution in 5% goat and 10% rat serum; Stressgen, Ann Arbor, MI). On the second day, sections were washed 3 x 3 min in TBS and incubated with biotinylated goat antirabbit secondary antibody (1:200; Dako, Carpinteria, CA) for 45 min. Preincubated avidin biotin complex and filtered diaminobenzidine were applied before dehydration and mounting in Depex (BDH Merck, Poole, UK). Collagen content is expressed as percentage of area per field analyzed.
Expression of osteopontin, cyclooxygenase (COX)-2, and HO-1 was determined by unbiased systematic sampling to determine the average number of positively stained structures (i.e. tubules, collecting ducts) per field (osteopontin and HO-1) or the percent area of positively stained tissue (COX-2). Changes in ED-1 expression were estimated by determining the average number of positively stained macrophages per field. In the case of osteopontin, COX-2, and HO-1, the cells comprising positively stained tubules and collecting ducts were stained uniformly; negative structures showed no staining. At least 60 fields for osteopontin, COX-2, and HO-1 were analyzed and at least 80 ED-1-positive macrophages were counted for each rat to ensure sufficient numbers were sampled for the correct calculation of statistical significance. The counting frame used for analysis was 826,890 µm2. Mean values for each group were compared by one-way ANOVA with Tukey’s post hoc test.
RT-PCR
Total RNA was prepared from freshly isolated rat renal tissues (cortex) with Ultraspec (Fisher Biotec, West Perth, Australia) following the manufacturer’s instructions. First-strand cDNA synthesis from 500 ng total RNA was performed after DNase treatment with avian myeloblastosis virus reverse transcriptase (Roche, Stockholm, Sweden) primed by random hexamers. PCRs were carried out using the following primer sets (all 5' to 3'): p22phox sense, CCC CCG GGG AAA GAG GAA AA, antisense, GCA GGC GAC AGC ACT AAG; gp91phox sense, CCA TTC GGA GGT CTT ACT TTG, and antisense, CTG GGC ACT CCT TTA TTT TTC; NOX4, the nicotinamide adenine dinucleotide phosphate (reduced) [NAD(P)H] oxidase subunit 4, sense, GAA CCT CAA CTG CAG CCT GAT C and antisense, CCT TTG TCC AAC AAT CTT CTT GTT CTC. Expression levels were normalized to 18S rRNA levels: (18S sense, CGG CTA CCA CAT CCA AGG AA, antisense, GCT GGA ATT ACC GCC GCT). To validate the real-time PCR protocol, gene-specific standard curves for p22phox and ribosomal 18S were generated from 10-fold serial dilutions of pre-prepared standards. Standards were diluted as follows: from 10 to 0.1 pg/µl for 18S and from 500 fg/µl to 0.5 fg/ßl for all other transcripts. Real-time PCR amplification was performed on the Lightcycler (Roche Molecular Biochemicals, Mannheim, Germany) using SYBR Green reaction mix (Roche Molecular Biochemicals) and the primers described above. cDNA samples were diluted 1:20 in water immediately before use for p22phox and 18S; gp91phox and NOX-4 were analyzed undiluted. Relative amount of mRNA was calculated using the comparative crossing threshold method. All specific amplicons were normalized against housekeeping gene 18S whose expression was determined as an internal control using predeveloped assay reagents (Applied Biosystems, Foster City, CA). In addition, TGFß1 mRNA levels were analyzed by multiplex real-time RT-PCR on the Corbett Rotor-Gene using the following primers and probes: TGFß1 sense, GGA CAC ACA GTA CAG CAA; antisense, GAC CCA CGT AGT AGA CGA T; FAM-labeled MGB probe, ACA ACC AAC ACA ACC C; and the Vic-labeled 18S TaqMan rRNA Control (Applied Biosystems). Reactions were performed using the RealMasterMix (Eppendorf AG, Hamburg, Germany) and analyzed using the 
Ct ( crossing threshold) method.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). After steroid withdrawal, the extent of fibrosis was not significantly different from that seen with DOC treatment at either the 4- or 8-wk time points. The addition of eplerenone to continuing DOC administration from 5 to 8 wk reduced collagen to levels not different from control. Importantly, addition of eplerenone substantially and significantly reduced levels of fibrosis to half those seen in the DOC404 steroid withdrawal group.
|
shows values for ED-1 (Fig. 2A, top panel), COX-2 (Fig. 2, B and C), and osteopontin for each of the five groups of animals. Eplerenone showed different effects on each of the three inflammatory markers examined.
|
COX-2 expression showed similar responses in both the outer and inner cortex (Fig. 2
, B and C) to those in the heart (17). Significant increases were seen after 4 and 8 wk of DOC treatment and in the steroid withdrawal group in the inner and outer cortex of the kidney, with eplerenone reducing expression back to control levels. The renal medulla was also evaluated and showed no significant changes in COX-2 expression in tubules or collecting ducts in any of the groups (data not shown).
For osteopontin (Fig. 2D), the values at 4 and 8 wk of DOC and after steroid withdrawal were all significantly elevated above control, in a pattern similar to that seen in the cardiac vasculature; no significant differences were seen between these groups. Whereas both the steroid withdrawal group and the eplerenone treatment group appeared to be also elevated above control, this did not reach significance. However, these groups were also not different from the DOC treatment groups.
In the present study, we also determined mRNA expression of TGFß1 in whole kidney by quantitative RT-PCR. DOC administered for 4 wk markedly and significantly elevated mRNA expression above control (Fig. 3D); levels of TGFß1 were not different from control in the other treatment groups. Thus, eplerenone treatment appears to reverse the DOC-induced increase in some but not all markers of inflammation in the renal cortex.
|
, A–C, shows mRNA expression of components of the NADPH oxidase system implicated in the generation of reactive oxygen species in the five experimental groups, showing a commonality of response. For the widely expressed p22phox (Fig. 3, top panel), mean values are similarly increased at 4 and 8 wk and at 4 wk after DOC withdrawal (DOC404), with those after 8 wk DOC not further elevated by ANOVA; values in animals receiving eplerenone over the last 4 wk (DOC8E4) returned to control levels. Very similar patterns of response are seen for gp91phox (center panel), whereas the NOX-4 subunit (third panel), which is expressed at high levels in the kidney, is markedly elevated at 8 wk of DOC and shows small but commensurate increases in mean levels at 4 wk and in the steroid withdrawal group.
HO-1.
Expression of HO-1 was restricted to the outer cortex and was found predominantly in distal and proximal tubules. However, whereas staining was also seen in the blood vessels and medulla region of DOC-treated animals, the level of staining was not sufficient to be quantified by stereological methods. Figure 4 shows HO-1 staining in the cortex for each of the treatment groups. DOC treatment for either 4 or 8 wk significantly (P < 0.05) increased HO-1 staining; levels were sustained after steroid withdrawal but returned to control values with MR blockade.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, we showed that 8 wk of DOC treatment in conjunction with a high-salt diet produces increased collagen deposition in the tubulointerstitial spaces. Moreover, when DOC is withdrawn at 4 wk, collagen levels remain significantly elevated. MR activation in the presence of a high-salt intake has been shown to cause renal injury and fibrosis in several experimental models, including stroke-prone, spontaneously hypertensive rats (19), angiotensin II infused/saline drinking rats (20), and N-omega-nitro-L-arginine methyl ester/angiotensin II-treated rats (13). In these models MR blockade or adrenalectomy reduces proteinuria and prevents fibrinoid necrosis of renal vessels and interstitial tissues. Similarly, MR activation has been shown to contribute to the tubulointerstitial damage and fibrosis that accompanies unilateral ureteral obstruction, in that the MR antagonist spironolactone can prevent tissue remodeling in this model (21).
These models have been shown to involve common inflammatory responses such as macrophage infiltration and cytokine up-regulation (7). In particular, osteopontin, which is expressed at low levels in the kidney under normal conditions, is clearly up-regulated and expressed more broadly. We found osteopontin expression, as determined by immunohistochemistry, in the thick ascending loop of Henle, proximal tubules, and some glomeruli in response to DOC/salt administration for 8 wk. Osteopontin expression remained elevated after steroid withdrawal and, in contrast with the heart, was not reversed by eplerenone (17). Given the clear increase in osteopontin expression in a range of renal disease models, it is possible that increased expression was exacerbated by uninephrectomy and thus that our results reflect not only MR activation but also non-MR responses.
Macrophages are the predominant infiltrating inflammatory cells recruited to the kidney in models of renal injury and may also produce proinflammatory cytokines to increase tissue damage and fibrosis. Macrophage infiltration is regarded as an indicator of tissue damage in experimental animals and clinical renal biopsies (14). Several authors have demonstrated that increased renal production of chemoattractants, such as osteopontin and monocyte chemoattractant protein-1, plays a functional role in renal macrophage accumulation and tissue damage (22, 23, 24). In the current study, ED-1-positive macrophages were detected in the interstitial space, around blood vessels, and within glomeruli of animals treated with DOC. Macrophage numbers were significantly up-regulated in the kidneys of DOC-treated animals and followed a similar pattern of expression to that of osteopontin (Fig. 2), in that MR activation increased levels but neither steroid withdrawal nor eplerenone treatment produced a significant reduction, supporting the causal link between osteopontin and macrophage recruitment. These findings again differ, however, from the ED-1 expression pattern determined in the heart (17), in which expression was significantly elevated after both 4 and 8 wk of DOC and returned to control values after eplerenone treatment for wk 5–8. These differences may reflect the underlying cellular changes that occur in the remaining kidney independent of administered DOC. One other possibility is that the infiltrating macrophages seen in the eplerenone-treated kidney are involved in the resolution of renal damage. It is known that macrophages can actively promote tissue repair as well as cause damage (25). Therefore, MR blockade may have changed the activation status of the infiltrating renal macrophages from a state of promoting renal fibrosis to one of promoting renal repair. Further studies are required to investigate this possibility.
COX-2 is an inducible enzyme and under normal conditions is expressed at low or undetectable levels. It is, however, normally expressed in the kidney, predominantly in the macula densa but also in the thick ascending loop of Henle. Increased COX-2 expression has been associated with a number of renal disease states in which expression is detected in a wide range of tissue structures. We identified renal tubule and collecting duct changes in COX-2 expression and found that significant changes were detected only in the renal cortex. The cortical pattern of response in both the inner and outer regions was similar to that in heart, with expression levels clearly elevated after steroid withdrawal and eplerenone treatment returning levels to control values (Fig. 2, B and C). The observation that steroid withdrawal alone does not effectively reverse the observed changes in the tubules and collecting duct of the kidney cortex suggests that sustained renal damage allows continued activation of the MR; a similar pattern of COX-2 expression was found in the heart. Given that the pattern of response for renal interstitial fibrosis is similar to that for COX-2 but not for the other inflammatory markers, COX-2, and the production of prostaglandins may thus be a tissue-specific signaling mechanism for the induction of renal damage and fibrosis. The beneficial effects of COX-2 inhibitors in the kidney under some circumstances (26) provide further support for this hypothesis and, given that similar protection is not seen in the heart (27), further demonstrate that overlapping but not identical mechanisms may operate in the development of fibrosis in these tissues.
In this study we also determined for the first time changes in expression of mRNA levels for specific markers of increased NAD(P)H oxidase activity in the kidney. Given that isoforms of the membrane bound NOX subunit of the NAD(P)H oxidase complex are expressed in a tissue-specific manner, our results indicate that there may also be cell-specific regulation of the oxidative stress response. The gp91phox isoform (NOX-2) is predominantly expressed in endothelial cells and infiltrating macrophages and is markedly up-regulated by DOC from 4 wk. Expression of this subunit did not fall after steroid withdrawal, suggesting ongoing production of superoxide ions in these cells types, which may contribute to the sustained inflammatory response. In contrast, expression of the NOX-4 isoform is only slightly increased after DOC for 4 wk but markedly elevated at 8 wk of treatment, with levels after steroid withdrawal comparable with those after 4 wk of steroid treatment. The NOX-4 isoform is expressed not only in the vasculature but also extensively in the renal epithelium; the higher abundance of epithelial cells in the kidney than the heart implies that the values recorded are likely to reflect changes in the renal tubules rather than the vascular wall as is the case in the heart. For example, MR activation has recently been shown to produce oxidative stress in tissues other than the vessel wall (28, 29), and in cultured rat mesangial cells, aldosterone increases localization of the cytosolic NAD(P)H subunits to the membrane (30). Similar patterns of subunit mRNA expression were detected in the heart, suggesting that along with COX-2 expression, oxidative stress may be a common factor in translating MR signaling into vascular inflammation and tissue remodeling in these different tissues.
HO-1 is increased by oxidative stress and is a well-accepted marker of oxidative stress in the kidney. It has been used in the current study to both support the mRNA oxidative stress data and provide an indication of the cell types involved in the oxidative stress response. Staining was predominantly seen in the proximal and distal tubules of the cortex and was confined to the outer cortex in control and eplerenone-treated animals. The HO-1 staining supports the NOX-4 in particular, given that the tubular epithelium is a predominant site of NOX-4 expression. It is also important to note that the pattern of staining for the immunostaining is different from that for mRNA expression of the NAD(P)H oxidase subunits. Whereas the mRNA expression is significantly reduced after both steroid withdrawal and MR blockade, the HO-1 staining remained significantly elevated in the steroid withdrawal group, indicating that whereas the NAD(P)H oxidase subunit expression is reduced, oxidative stress is still present. Without protein data for NAD(P)H oxidase, it is unclear as to the underlying cellular mechanisms that are responsible for the sustained oxidative response in this study.
An important consideration in understanding MR signaling is the maintenance of MR specificity by the enzyme 11ß-hydroxysteroid dehydrogenase type 2 (11ßHSD2) (31). Whereas vascular sites of MR expression are protected by 11ßHSD2 activity, cardiac myocytes are not, indicating that several levels of regulation of MR signaling may exist in the cardiovascular system (17, 32). We have shown that glucocorticoids can activate vascular MR when 11ßHSD2 is blocked, and this results in similar inflammatory and fibrotic responses to aldosterone excess (4). Similarly, Ward et al. (33) showed that eplerenone blocked coronary fibrosis and constriction after angioplasty in normal pigs, which suggests that the MR may be activated in the absence of administered salt and/or mineralocorticoid in the context of tissue damage. Tissue damage or 11ßHSD2 blockade leading to altered intracellular redox state has thus been proposed to activate glucocorticoid occupied MR, promoting the progression of vascular damage, even when plasma aldosterone levels are not elevated (34).
The current study compares tissue responses with either receptor blockade or steroid withdrawal in a model of renal fibrosis and thus differs from many other studies that investigate models of prevention. That established pathology can be reversed by MR blockade has been recently shown by the 4E Left Ventricular Study (35), which demonstrated the ability of eplerenone, enalapril, or the two in combination to reverse left ventricular hypertrophy in essential hypertensives. Whereas there was also a reduction in blood pressure, this correlated poorly with hypertrophy, indicating the involvement of other factors in progression and maintenance of cardiac hypertrophy. Taken together these studies strongly suggest that tissue remodeling in the kidney and heart is a dynamic process that may be reversed if MR activation is a part the pathologic stimulus.
|
Footnotes |
|---|
First Published Online April 20, 2006
Abbreviations: COX, Cyclooxygenase; DOC, deoxycorticosterone; HO-1, hemeoxygenase 1; 11ßHSD2, 11ß-hydroxysteroid dehydrogenase type 2; MR, mineralocorticoid receptor; NAD(P)H, nicotinamide adenine dinucleotide phosphate (reduced); NOX-4, NAD(P)H oxidase subunit 4; TBS, Tris-buffered saline.
Received December 5, 2005.
Accepted for publication April 12, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. J. Brown Aldosterone and Vascular Inflammation Hypertension, February 1, 2008; 51(2): 161 - 167. [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Rickard, J. W. Funder, J. Morgan, P. J. Fuller, and M. J. Young Does Glucocorticoid Receptor Blockade Exacerbate Tissue Damage after Mineralocorticoid/Salt Administration? Endocrinology, October 1, 2007; 148(10): 4829 - 4835. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.05 vs. DOC4, DOC8, DOC404.
