Hypertensive target organ damage shows characteristics of accelerated cell turnover and aging. This might have developed during the evolution of hypertension. In the kidney, high cell turnover is mainly restricted to tubular cells. It was the aim of this study to investigate whether a transient intervention in spontaneously hypertensive rats (SHRs) leads to reduced tubular cell turnover and attenuates the renal aging process and tubulo-interstitial damage in the long-term.
SHRs (i) were prehypertensively (weeks 4–8) treated with losartan (ii) or hydralazine (iii) (20 and 4 mg/kg/day, respectively) and compared to Wistar-Kyoto (WKY) rats (iv). Groups were investigated at weeks 8 and 72 (except iii). At both time points tubular cell proliferation (proliferative cell nuclear antigen) and systolic blood pressure (SBP) were evaluated. At week 72, aging parameters such as telomere length were assessed. Renal damage was semiquantitatively assessed (scale: 0–4) by measuring the parenchyma (atrophy) and vasculature (media thickness).
Treatments equipotently reduced SBP in young SHRs (P < 0.01) but only losartan reduced renal proliferation (proliferative cell nuclear antigen: (i) 2.8 ± 0.8, (ii) 1.3 ± 0.3, (iii) 3.0 ± 0.6, (iv) 0.1 ± 0.1 cells/mm2). In SHRs treated with losartan(SHR-Los) tubular proliferation remained reduced and renal telomere length was significantly greater than in untreated SHRs (fold: (i) 1.0 ± 0.1, (ii) 2.8 ± 0.3, P < 0.01). Untreated SHRs (median 2.0, range 1–3; P < 0.007), but not SHR-Los (median 1.0, range 0–2; P = 0.06) demonstrated more tubular atrophy than WKY rats (median 0.5, range 0–1).
Transient losartan treatment reduces cell-turnover not only acutely but also for a prolonged period after drug withdrawal. This results in the long-term in reduced aging and attenuated tubulo-interstitial damage, suggesting there exists a modulating effect of angiotensin II (ANGII)-antagonism on long-term cell turnover.
Observational studies have demonstrated that pulse pressure increases with age.1 Therefore several authors have suggested that essential hypertension represents an age-related disorder.2,3,4 This was underlined by Jeanclos et al., demonstrating that an increased pulse pressure was associated with enhanced individual aging parameters such as reduced telomere length.5 Experimental studies in spontaneously hypertensive rats (SHRs) revealed an accelerated cell turnover of vascular smooth muscle cells as compared to normotensive rats,6 particularly at a young age.3 This was counterbalanced by an increased apoptosis at an older age.7 These findings in SHRs suggest not only a link between hypertension and aging but further indicate that this association might already be present at a young age.3
Angiotensin II (ANGII) modifies cell proliferation by inducing the transforming growth factor-β (TGF-β) pathway and direct modulation of the G1 phase by activation of cyclins and cyclin-dependent kinases; these as a complex lead to the phosphorylation of phosphorylated retinoblastoma protein, which reflects the “point of no return” in the G1 phase.8,9,10,11 In this context, recent studies demonstrated that chronic angiotensin-converting enzyme inhibition ameliorated aging-related vascular and renal function loss in a blood pressure independent manner.12,13
This highlights the renin-angiotensin system (RAS) as a potential mediator for the process of accelerated cell turnover in young SHRs. This idea is fuelled by the observation that the initial increase of cell proliferation in SHRs coincides with the time window in which ANGII is overactivated.14,15,16 Moreover, transient inhibition of the RAS during this time window leads to cardiovascular protection up to an advanced age without long-lasting blood pressure reduction.17 The latter raises such questions as whether a prolonged reduction of tubular cell turnover involves the TGF-β pathway, is blood pressure dependent or independent and might be associated with less tubulo-interstitial damage.
Therefore it was the aim of this study to investigate (i) whether early ANGII and TGF-β inhibition reduces cell turnover acutely and whether this effect is blood pressure dependent or independent and (ii) whether long-term effects on cell-turnover persist after cessation of ANGII antagonism, potentially resulting in attenuated aging and tubulo-interstitial protection in aged SHRs.
Three-week-old male Wistar-Kyoto (WKY) rats and SHRs were obtained from Charles River (Landgraaf, The Netherlands). The rats were fed a normal sodium diet and maintained on dry pellet diet and had free access to water. All experiments were approved by the animal ethics committee of the Maastricht University and were performed in accordance with institutional guidelines.
Four-week-old WKY rats and SHRs were given saline (placebo), losartan (SHR-Los: 20 mg/kg/day), hydralazine (SHRs-Hyd: 15 mg/kg/day), for a period of 4 weeks using osmotic minipumps (Alzet 2004, Durect, CA) that were implanted subcutaneously under isoflurane anaesthesia (1.5%). For TGF-β neutralization experiments, TGF-β1-antibody (anti-TGF-β1: 1 μg/kg/day, Biosource International, Camarillo, CA) or subclass-matched control antibody (anti-IgG-κ1: 1 μg/kg/day, R&D Systems, Minneapolis, MN) were administered to 4-week-old SHRs for a period of 4 weeks using osmotic minipumps (Alzet 2004). Each group consisted of 7–9 animals. Systolic blood pressure (SBP) was measured in the conscious unrestrained state through a heparanized indwelling polyethylene catheter that was introduced into the left femoral artery 2 days before measurement at the age of 8 weeks. Heparanized blood samples were drawn via the arterial catheter and were centrifuged at 4,000g for 10 min 4 °C. The serum was immediately stored at −20 °C until usage. Serum was used for determination of TGF-β1 by using commercially available ELISA (Genzyme, Wiesbaden, Germany), as previously described.
Placebo-treated WKY rats, SHRs, and SHR-Los were additionally investigated until 72 weeks of age. In these rats, oxidative stress-induced lipid peroxidation was longitudinally measured by quantification of thiobarbituric acid-reactive substances (TBARS) from 24 h urine samples using high-performance liquid chromatography.18 Kidneys were harvested at 8 (all treatment groups) and 72 weeks (WKY rats, SHRs, and SHR-Los), incubated for 24 h in formalin and embedded into paraffin. The paraffin-embedded specimens were cut into 4-μm thick slices for H&E, Sirius Red, proliferative cell nuclear antigen and α-smooth muscle actin staining as previously described by our group.19 Tubulointerstitial injury was defined as inflammatory cell infiltrates, tubular dilation or atrophy, or interstitial fibrosis by a nephropathologist in a blinded manner. Injury was graded on a scale of 0–4 (0 normal; 0.5 small focal areas of damage; 1 involvement of <10% of the cortex; 2 involvement of 10–25% of the cortex; 3 involvement of 25–75% of the cortex; and 4 extensive damage involving >75% of the cortex). proliferative cell nuclear antigen-, α-smooth muscle actin-, and Sirius-Red-stained sections were carefully scanned using a light microscope (DFC 280; Leica, Tokyo, Japan) connected to a computer using the NIH image-analysis system for histomorphometry at a original magnification of ×200. Proliferating tubular cells were quantified as cells/mm2, the cross-sectional area of renal afferent arterioles was assessed, the area of perivascular fibrosis was calculated as a ratio of the cross-sectional area, and interstitial fibrosis was calculated as a percentage of renal tubular tissue.
Renal cortical tissue was snap-frozen and RNA isolated with an RNeasy Mini Kit (QIAGEN, Hilden, Germany). RNA was purified (Ultraspec-II RNA, Biotec Laboratories, Houston, TX), quantified (Nanodrop ND-1000, WITEC, Switzerland) and transcribed into complementary DNA with Superscript III reverse-transcriptase, using 250 ng of random primers (Invitrogen Life Technologies, Strathclyde, UK). Primers and probes against the selected genes of cyclinD1 (up: CGCCTTCCGTTTCTTACTTCA; down: AACTTCTCGGCAGTCAGGGGA; 62 °C), cdk4 (up: AAGGATCTGATGCGCCAGTT; down: CAGGTCCCGGTGAACAATG; 60 °C) and type IV collagen (up: GAGGGTGCTGGACAAGCTCTT; down: TAAATGGACTGGCTCG GAATT; 65 °C) were investigated. Phosphoglucokinase-1 was used as housekeeping gene (up: CGGAGACACCGCCACTTG; down: AAGGCAGGAAATACTAAACAT; 60 °C).
Telomere measurement by quantitative PCR was used to determine cellular aging as described by Cawthon.20 Therefore, oligonucleotide primers were designed to hybridize to the TTAGGG and CCCTAA repeats which were used to measure telomeres (up: CGGTTTGTTTGGGTTTGGGTTTGGGTTTGGGTTTGGGTT; down: GGCTTGCCTTAC CCTTACCCTTACCCTTACCTTACCCT; 54 °C). This leads to results corresponding to the relative telomere length of their DNA. Telomere shortening is intrinsically related to cell turnover.21 In age-matched animals, shorter telomeres in one rat are related to increased cell turnover and thus shorter cell-cycle length as compared to the age-matched rat with longer telomeres.
Renal cortical tissue was homogenized and protein concentration quantified. Common western blot using phospho-phosphorylated retinoblastoma protein (110 kDa, 1:400, Cell Signaling Technology, Danvers, MA) was performed.
Data were given in mean ± s.d. We compared the groups with a one-way analysis of variance and a post-hoc Bonferroni test. Anti-TGF-β1 and control antibody treated SHRs were compared with Student's t-test. For these experiments we considered P < 0.05 to indicate statistical significance. Histological scoring data are given as median and range and analyzed using the Mann-Whitney Test with Bonferroni correction for non-parametric statistical analysis. According to the number of these groups an adjusted P < 0.016 was considered to indicate statistical significance.
In the young age rat, SBP was significantly reduced as a result of both antihypertensive treatments as compared to the untreated SHRs (P < 0.05). This effect was not apparent after anti-TGF-β1 treatment as compared to the placebo treatment with control IgG antibody (Figure 1a).
Cell cycle progression in spontaneously hypertensive rats (SHRs) is angiotensin II (ANGII), but not blood pressure dependent. (a) Intra-arterial blood pressure measurements in awake and unrestrained rats demonstrate blood pressure reduction in both antihypertensive treatment groups without reaching normotensive values. transforming growth factor-β1 (TGF-β1) antibody treatment showed no blood pressure effect. (b) TGF-β1 plasma levels were reduced in WKY rats and in losartan and anti-TGF-β1 treated SHRs. (c) Proliferating renal tubular cells were visualized using proliferative cell nuclear antigen (PCNA) immunohistochemistry. Positive cells were quantified per area. ANGII antagonism and TGF-β1 antibodies but not hydralazine treatment reduced proliferation of tubular cells. *P < 0.05 compared to WKY, **P < 0.05 SHR-Los compared to SHRs, ***P < 0.05 anti-TGF-β1 compared to control antibody treated SHRs (WKY: n = 8, SHRs: n = 8, SHR-Los: n = 8, SHR-Hyd: n = 7, SHR anti-TGF-β1: n = 9, SHR IgG n = 8).
TGF-β1 plasma levels were assessed as a marker of proliferation and were highest in SHRs. Losartan, but not hydralazine, significantly reduced TGF-β1 plasma levels to WKY levels, just as anti-TGF-β1 treatment did (Figure 1b). Tubular proliferation, as quantified by immunohistology, was highest in SHRs. ANGII-antagonism significantly attenuated the amount as anti-TGF-β1 treatment did. By contrast, the vasodilator hydralazine did not affect the rate of tubular cell proliferation (Figure 1c).
To assess the degree of tubular proliferation after drug withdrawal in SHR-Los, rats were followed up until 72 weeks of age. The SHR-Hyd group was not included in this setting, as hydralazine did not show any acute effects on tubular proliferation in the young SHRs. In a recent study we demonstrated that mean arterial pressure in SHR-Los remained significantly decreased until 48 weeks of age.17 At the advanced age of 72 weeks the SBP values of WKY rats were 129 ± 6 mm Hg, 221 ± 11 mm Hg in SHRs and 204 ± 16 mm Hg in SHR-Los. Thus, the SBP was not different in aged SHR-Los as compared to untreated SHRs.
Tubular proliferation was highest in untreated SHRs, but remained significantly reduced in SHR-Los despite drug withdrawal at 8 weeks of age. In parallel, TGF-β1 gene expression was attenuated in SHR-Los (Figure 2a,b). The G1 phase cell cycle progression was analyzed as depicted in Figure 2c,d,e. Major components of the G1 phase were reduced in mRNA levels in aged SHR-Los as compared to SHRs. Similarly, less phosphorylated retinoblastoma protein was apparent in the protein levels in SHR-Los as compared to SHRs (phosphorylated retinoblastoma protein (AU): WKY 30.6 ± 3.7, SHRs 78.3 ± 7.8, SHR-Los 52.5 ± 2.8).
Cell proliferation remained reduced after transient losartan treatment until 72 weeks of age and was associated with attenuated G1 signaling. Cell cycle progression after transient angiotensin II (ANGII) inhibition remained lower until advanced age. (a) Proliferating renal tubular cells were visualized using proliferative cell nuclear antigen (PCNA) immunohistochemistry. Positive cells were quantified per area. After 64 weeks of drug withdrawal in losartan treated spontaneously hypertensive rats (SHRs), the proliferation of tubular cell remained decreased by 50% as compared to untreated SHRs. (b) transforming growth factor-β1 (TGF-β1) gene expression was reduced in previously treated SHRs as compared to placebo treated SHRs. The same pattern was observed for (c) cdk4-expression. (d) Transient treatment reduced the cyclin D1-expression in SHRs below normotensive level. (d) The reduced gene expression of cdk4 and cyclin D resulted in reduced phosphorylation of phosphorylated retinoblastoma protein (pRb) as shown by western blot in arbitrary units (n = 5/group). Representative blots (2/group) are presented below the quantification. *P < 0.05 compared to WKY, **P < 0.05 SHR-Los compared to SHRs (number of observation excluding (e): WKY: n = 8, SHRs: n = 7, SHR-Los: n = 8).
As measured by total renal DNA, no significant hyperplasia was evident in any group (μg DNA/kidney: WKY 16.6 ± 1.8, SHRs 17.6 ± 3.1, SHR-Los 14.3 ± 3.2). However, the telomere length was different between treated and untreated SHRs (fold: SHRs 1.0 ± 0.1, SHR-Los 2.8 ± 0.3, P < 0.01). Moreover, oxidative damage, determined as renal TBARS, was significantly reduced in SHR-Los as compared to untreated SHRs during the whole observation period (except at 8 weeks of age); this may be related to the higher standard deviation at the first time point (Figure 3, P < 0.05). Irrespective of age, the analysis of variance showed a tendency for lower TBARS in WKY rats as compared to SHRs (P = 0.06), but not as compared to SHR-Los (P = 0.34).
Urinary excretion of TBARS. Urinary excretion of TBARS as marker for lipid peroxidation is reduced throughout the observation period in spontaneously hypertensive rats (SHRs)-Los as compared to untreated SHRs. *P < 0.05 SHR-Los compared to SHRs at the age of 12, 48, and 72 weeks. (WKY: n = 8, SHRs: n = 7, SHR-Los: n = 8).
The Mann-Whitney test demonstrated that only aged SHRs (median: 2.0, range 1–3; P < 0.007) but not SHR-Los (median: 1.0, range 0–2; P = 0.06) showed more tubular atrophy than 72-week-old WKY rats (median: 0.5, range 0–1; Bonferroni adjusted significance for repeated measures: P < 0.016). Characteristic pictures of the tubulo-interstitium at the juxtamedullary junction are depicted in Figure 4. Concomitantly, interstitial collagen deposition was reduced in these rats, which was confirmed by type IV collagen gene expression (Figure 5a,b,c). By contrast, renal vasculature did not demonstrate treatment effects. The cross-sectional area of afferent arterioles and perivascular matrix deposition were both not reduced in SHR-Los as compared to untreated SHRs (Figure 5d,e).
Tubulo-interstitial characteristics at the juxtamedullary junction in 72-week-old rats. (a) 72-week-old WKY rats demonstrate homogenous tubular tissue, while SHRs (b) are characterized by tubular atrophy and dilatation. SHR-Los (c) show these characteristics to less degree. Original magnification ×100.
Selective renal target organ damage in advanced age after transient losartan treatment. (a) Tubular atrophy (median and range), (b) tubulointerstitial collagen deposition and (c) type IV collagen deposition were reduced in transiently treated as compared to untreated spontaneously hypertensive rats (SHRs). In the renal vasculature no treatment effects were seen on (d) cross-sectional area (CSA) and (e) their perivascular collagen deposition. *P < 0.05 compared to WKY, **P < 0.05 SHR-Los compared to SHRs (WKY: n = 8, SHRs: n = 7, SHR-Los: n = 8).
Renal tubular cell proliferation is enhanced in young SHRs and reduced by ANGII and TGF-β inhibition, but not by general blood pressure reduction. Despite cessation of ANGII antagonism these effects continue up to an advanced age and are extended by a reduction of cellular senescence. Simultaneously, in aged SHRs, early and transient ANGII-antagonism resulted in a reduction of tubulo-interstitial damage, whereas no long-term renovascular effects were evident. These data suggest that accelerated cellular senescence is involved in tubulo-interstitial damage and that a reduction of the mitogenic effect might decelerate this process.
While studying the acute effects of transient treatment on tubular proliferation in SHRs, we demonstrated that ANGII-antagonism but not the lowering of blood pressure, reduced tubular proliferation in young SHRs. These results extend previous findings of pharmacological intervention in vascular smooth muscle cells to the tubular cell.6,22 Targeting the mitogenic effect of ANGII we addressed the TGF-β pathway as a potential pathway,9 particularly as the TGF-β plasma level revealed a similar pattern as the proliferation data. The central role of TGF-β as the mediator of ANGII-mediated tubular proliferation in young SHRs, is further clarified by the blood pressure independent attenuation of tubular proliferation after anti-TGF-β treatment. This demonstrates that already in young SHRs tubular proliferation is ANGII-dependent and aggregated, and suggests that the blood pressure–independent effects of TGF-β are involved in this process.
To address the question whether transient treatment can modulate long-term tubular proliferation we investigated aged SHRs, which had been transiently treated or untreated at a young age. In a previous study we demonstrated that transient treatment delays the onset of hypertension and albuminuria in SHRs.17 In this study we demonstrate that tubular proliferation remains reduced as compared to untreated SHRs. This suggests a sustained effect of the early and transient treatment on the mitogenic effects of tubular cells. The histological data are further strengthened by molecular data of reduced TGF-β expression and a reduced G1 phase pointing to the same direction. Again, a direct link to ANGII can be drawn as the gene expression of cyclins and cdk's is markedly modulated by ANGII.11
We can conclude from our data that the enhanced tubular cell proliferation in SHRs does not lead to hyperplasia but to an enhanced cell turnover, subsequently leading to cellular senescence as determined by reduced telomere length21 and increased TBARS in untreated SHRs.23 By contrast, reduced tubular proliferation due to the transient ANGII-antagonism is associated with longer telomeres and lower TBARS, suggesting an attenuated state of senescence as compared to untreated SHRs.
The question was addressed whether the reduced mitogenic effect of tubular cells after early and transient treatment results in reduced tubulo-interstitial damage of aged SHRs. Indeed, tubular atrophy and collagen deposition were reduced in transiently treated aged SHRs. However, several authors demonstrated protective effects after transient treatment and claimed that the prolonged blood pressure lowering effects were the reason for it.17,24 We therefore investigated the extent of renovascular hypertrophy in the same organ based on the idea that vascular smooth muscle cells show less proliferation than tubular cells. Thus, in the same organ high vs. low-turnover cells can be compared with respect to the local damage, which was defined for vSMC as media hypertrophy and perivascular collagen deposition.25 Based on this comparison, early and transient treatment showed protection of the “high”-cell turnover tissue, but not the “low”-cellturnover tissue. This does not clearly exclude the possibility that in areas of lower cell-turnover, cellular senescence has no effects on target organ damage.6,26 In particular, as Hadrava et al. demonstrated for vSMC of SHRs, there was an accelerated cell cycle in vitro which was mostly due to an earlier entry into the S phase.27 This seems to suggest that the extent of accelerated aging determines the biological age and cardiovascular target organ damage.28
The results have to be interpreted in context with their potential clinical relevance. Up to now two intervention studies have been published dealing with angiotensin II receptor blocker (ARB) in prehypertensive subjects (TROPHY)29 or subjects with hypertensive parents (DHyPP).30 In TROPHY a persistent effect of 2 mm Hg was found after one year ARB cessation, while DHyPP failed to find persistent effects on blood pressure. A recent study in SHRs found prolonged blood pressure lowering,17 but failed to find persistant long-term blood pressure effects similar to the DHyPP. However, results from this study suggest that the anti-aging effects of ARBs can be dissociated from antihypertensive effects. Thus, the intervention group in DHyPP may benefit from the persistant anti-aging effects of the transient ARBs. Until now, Skov et al.30 only provided cardiac and renal data at the end of the treatment period, but not at the end of the observation period after cessation of treatment. Therefore it remains uncertain whether our findings can be translated to the human situation.
In summary, we demonstrated that the degree of tubular proliferation can be continuously reduced by transient ANGII antagonism in young SHRs. This process is largely ANGII dependent while blood pressure is of minor importance. The decelerated proliferation is associated with attenuated tubular senescence and tubulo-interstitial damage. This phenomenon seems to focus on cells with high turnover as in parallel vascular smooth muscle cells do not show reduced damage.
The authors declared no conflict of interest.
This study was funded by the Dutch Kidney Foundation (C03.2064) and the Wellcome Trust (to Marcus Baumann).