Hypertension is an important cause of chronic kidney failure . Kidney damage is prominent in most experimental models of hypertension [2–4], and structural and functional changes of the vasculature are a hallmark of chronic hypertension . Two kidney, one clip (2K1C) is a model of renal hypertension in which a silver clip on the renal artery restricts perfusion to the clipped kidney. This induces renin release, which in turn causes systemic hypertension. Whereas the clipped kidney is protected, the nonclipped kidney is exposed to the high blood pressure (BP) and develops kidney damage . In other models of hypertension, vascular damage leads to reduced autoregulatory efficiency, which enables glomerular damage . Further, vascular structural changes have been closely tied to tubular atrophy and the progression of renal fibrosis . The interlobular artery (ILA) buffers perfusion pressure more efficiently to the outer than to the juxtamedullary cortex (JMC) , and the JMC develops kidney damage first. This has been shown in the spontaneously hypertensive rat (SHR) , as well as salt-loaded angiotensin II-infused Sprague–Dawley rats [9,10], diabetic Otsuka Long-Evans Tokushima Fatty rats , and in the Dahl salt-sensitive rat . The importance of small arteries as a first step in the progression of kidney damage was recently reviewed  and has previously focused on low-renin or suppressed-renin models such as the SHR or salt-loaded animals. 2K1C represents a conceptually different model because it is a high-renin model without salt loading. In addition, 2K1C rats develop morphological damage over a relatively long period, which is more comparable to the progression of hypertensive kidney damage in humans in whom this takes years to decades.
In a previous study, we investigated the development of kidney damage in the nonclipped kidney of 2K1C hypertensive rats , and, although we did not look for it specifically, vascular damage was prominent, whereas the outer cortex was relatively protected. This led to the hypothesis that chronic hypertension in 2K1C causes primary vascular damage in the nonclipped kidney, which with time results in injury to the JMC, although leaving the outer cortex protected.
The aim of the present study was to analyze the time course and morphological sequence of structural changes in the nonclipped kidney of hypertensive 2K1C rats after 16 and 24 weeks of high BP. The main hypothesis was that hypertensive kidney damage in the nonclipped kidney of 2K1C is a primary vascular damage with later progression of tubulointerstitial damage in the JMC. This would support the hypothesis that pressure-induced vascular damage is a general mechanism behind the development of hypertensive kidney damage.
MATERIALS AND METHODS
Animals and experimental groups
All experiments were performed in accordance with, and under the approval of, the Norwegian State Board for Biological Experiments with Living Animals. Male Wistar Hannover rats from Taconic (Ry, Denmark) were used. The rats had free access to tap water and standard rat chow. Animals underwent surgery when they were 6–7 weeks old and had a bodyweight around 170 g. In each group, 12 rats were randomly assigned to the 2K1C group, and eight rats were sham operated as controls. Hypertension was induced as described previously , in short; the left renal artery was dissected free, and a silver clip with an internal opening of 0.2 mm was placed on the artery. The sham-operated animals followed the same procedure but without the placing of the silver clip. Surgery was performed under isoflurane anesthesia (2% in oxygen), and 0.03 mg/kg buprenorphine was used for pain relief. The rats were followed for 16 weeks (16-week group), or for 24 weeks (24-week group) before sacrifice. The animals were sacrificed under isoflurane anesthesia.
Measurements of mean arterial blood pressure and bodyweight
After preheating the animals in a heating cabinet for 20–30 min at 32°C, BP was measured using the tail-cuff method (CODA-6, Kent Scientific). Mean arterial pressure (MAP) and bodyweight was measured before surgery and every fourth week until sacrifice.
Urinary protein excretion
Twenty-four weeks after clipping, urine was collected from the 24-week group for 24 h while the rats were kept individually in metabolic cages, with free access to food and water. Urinary protein and creatinine concentrations were analyzed in the Laboratories for Clinical Biochemistry at Haukeland University Hospital, Bergen (Roche/Hitachi 912: U/CSF Protein and CREA Plus assays).
The animals were sacrificed as described before . The kidneys were removed, decapsulated and weighed, before they were cut in 1–2 mm thick transversal slices that were either put directly in RNA, later, frozen or fixed in 4% buffered formaldehyde for morphology. Renal damage was determined in the nonclipped kidney of 2K1C hypertensive animals and the equivalent kidney in sham-operated animals.
The border between the medulla and cortex was identified by stereomicroscopy, and pieces for mRNA and protein analysis were taken from the outer cortex and JMC. The pieces were collected as strips from transversal slices of about 1 mm thickness, giving a total size per piece of around 1 mm by 1 mm by 5–10 mm.
Transversal slices from nonclipped kidneys and corresponding kidney in controls were fixed with 4% buffered formaldehyde and embedded in paraffin by standard procedures. Three-micrometre-thick sections were stained with periodic acid Schiff (PAS) and trichrome. Microscopic investigations were performed in a blinded manner on coded sections. Tubulointerstitial damage was determined by amount of tubular dilatation, tubular atrophy, cast formation or interstitial expansion with inflammation or fibrosis, and was assessed in at least 10 microscopic fields in each section from outer cortex and JMC, respectively (PAS stain, objective 10 times) using a semiquantitative scoring system (0–4): grade 0 no lesions, grade 1; lesions in less than 25%, grade 2; damage in 25–50%, grade 3; damage in 50–75%, and grade 4; damage in more than 75% of the tubulointerstitial area. The tubulointerstitial damage score was the mean damage score of all investigated microscopic fields of outer or JMC.
Collagen determination using sirius red staining and polarization microscopy
Seven micrometre thick sections were stained with 1% Sirius red F3B in a saturated solution of picric acid for 1 h and investigated under polarized light in a Leica DMLB microscope connected to a CCD ColorView IIIu camera. CellD 2.4 software was used for image acquisition and automatic image analysis. Colour images from outer cortex and JMC were obtained by taking consecutive samples parallel to the kidney capsule and to the corticomedullary border, avoiding glomeruli and large vessels. Images were acquired with a 40 times objective under constant identical illumination and polarization settings. Image resolution was 480 × 480 pixels corresponding to an area of 240 × 240 μm and a resolution of 0.25 μm2/pixel. The HSI colour space was used, and the image was separated into the intensity component resulting in a 256 greyscale image. Thresholds for recognition of birefringent collagen were set on the grayscale image, and the same thresholds were used for all images. For quantification of interstitial collagen with Sirius red staining, an average of 20 images per kidney from 16-week-old rats (10 images from the JMC and 10 images from the outer cortex) and 43 images per kidney from 24-week-old rats (20 images from the juxtamedullary cortex and 23 images from the outer cortex) were used. Altogether, 391 images were analyzed from kidneys from 16-week-old rats and 732 images from kidneys from 24-week-old rats. Collagen content was expressed as percentage of the detected area.
Measurement of periarterial collagen
Arteries in Sirius red-stained sections were imaged using visible light. The images were imported into Adobe Photoshop CS4 Extended. The outline of the periarterial collagen and arterial media were traced using the lasso and used to measure the total area of the perivascular connective tissue. Then the magic wand (sensitivity 50, noncontiguous) was used to deselect the nonstained area to measure the collagen. Several vessels were measured in each section and the results were averaged for each animal. The results are expressed as the percentage collagen-positive area out of the total perivascular area.
Distribution of col1a2 mRNA synthesis was studied by nonradioactive mRNA in-situ hybridization using a probe covering nucleotides 2256–2468 of rat col1a2 cDNA (NM_053356) as previously described . In-situ hybridization was performed as previously described . Hybridized RNA probes were visualized using 4-nitroblue tetrazolium chloride.
Primary antibody was applied (monoclonal mouse anti-human α-smooth muscle actin, Clone 1A4; Dako, Glostrup, Denmark), and bound antibody was visualized using a horseradish peroxidase-labeled swine anti-mouse secondary antibody and diaminobenzidine (Dako). Vimentin (monoclonal, mouse, M0725; DakoCytomation, Glostrup, Denmark) antibody binding was detected by peroxidase-conjugated polymer, carrying antibodies to rabbit and mouse immunoglobulins (K5007; DakoCytomation).
In-situ hybridization and immunohistochemistry were examined using a microscope equipped with an interference contrast module (Leica, Wetzlar, Germany). Images were obtained using a SPOT RT 2.3.0 digital camera (Diagnostic Instruments, Sterling Heights, Michigan, USA) and the MetaVue imaging System (Molecular Devices, Downingtown, Pennsylvania, USA).
Quantitative reverse transcriptase PCR
Tissues in RNA later were used to extract total RNA using the RNeasy Mini kit (Qiagen Hilden, Germany). Reverse transcription was performed with RT-core kit (Eurogentec, Seraing, Belgium) using random nonamers as primers. Quantitative PCR (qPCR) was performed on an ABI prism (Applied Biosystems Inc., Carlsbad, California, USA) using a qPCR Mastermix for SYBR Green I (Eurogentec Seraing, Belgium). Probes for collagen type I a1 (col1a1), procollagen-n-peptidase (n-pep), procollagen-c-peptidase (c-pep), 18S ribosomal RNA (18S), were constructed with Primer Express (Applied Biosystems Inc. Carlsbad, California, USA) as described previously . Readymade Gene Expression Assays (Applied Biosystems Inc.) were used for matrix metalloprotease 2 (MMP-2) and MMP-9, tissue inhibitors of metalloproteases 1 (TIMP-1), and TIMP-2. In all quantitative measurements, the same three-step dilution of cDNA standard was used as reference, and the results were expressed as the quantity of the relevant mRNA in relation to the quantity of 18S ribosomal RNA in the same cDNA sample.
Western blot analysis
Proteins were extracted as described previously . Protein concentrations were determined using Bradford protein assay reagent (BioRad, Hercules, California, USA). The protein expression of MMP-2, MMP-9, TIMP-1, and TIMP-2 was investigated using western blot analysis by electrophoresis of 60 μg of extracted proteins through 12% SDS-PAGE. The proteins were transferred to polyvinylidene difluoride membrane using the iBlot dry blotting system (Invitrogen, Carlsbad, California, USA), and the membrane was blocked with blocking mix [Tris-buffered saline (137 mmol/l NaCl, 20 mmol/l Tris) 0.1% Tween 20, 0.2% I-block, 3.08 mmol/l NaN3 and 10.72 mmol/l MgCl2 *6 H2O] before incubation with primary antibodies (MMP-2, MMP-9, TIMP-1, TIMP-2; Chemicon international, Billerica, Massachusetts, USA) mixed with blocking buffer over night at 4°C. After washing, the blots were incubated with secondary antibody in Tris-buffered saline with tween for 1–3 h in room temperature. The bands were visualized using enhanced chemiluminescence and quantified using the BioRad Quantity one program (BioRad, Hercules, California, USA). Protein expressions are given as normalized to sham outer cortex, and all other groups are presented relative to this group.
All data are given as mean ± SEM. Analysis of variance (ANOVA) is a statistical analysis performed using R version 2.14.0 . Multiway ANOVA was used to test for significance using Fisher's test with a priori contrasts to test between groups. The semi-quantitative tubulointerstitial damage score was tested with the nonparametric Kruskal–Wallis test and followed by Wilcox's test with a priori contrasts. A probability of chance difference, P < 0.05 was considered significant. Comparison between outer cortex and JMC in the same animal was paired.
All animals survived and were available for analysis in the 16-week group and in both sham groups. In the 24-week group, three rats developed severe proteinuria and died before sacrifice. Two more rats developed proteinuria before 24 weeks and were available for analysis. Bodyweights were similar in the sham and the 2K1C group at start and end of the experiment (Table 1). The clipped kidney was smaller than the nonclipped kidney in 2K1C animals, both after 16 and 24 weeks (P < 0.05), and the difference was larger at 24 weeks than at 16 weeks (P < 0.05). No difference in kidney weight was seen between kidneys in sham at 16 and 24 weeks (Table 1). Arterial hypertension developed 2–3 weeks after clipping and increased during the experiment (Fig. 1a and b). At sacrifice, both 16-week and 24-week 2K1C animals had a higher MAP compared with the corresponding sham group (P < 0.05), and the 24-week 2K1C animals developed a higher MAP than the 16-week group (Table 1). Urinary protein excretion was higher in the 24-week 2K1C group compared with sham animals, but due to large variation, this was not statistically significant (Table 1).
Tubular damage and interstitial fibrosis in the nonclipped kidney, indicated by vimentin positivity and blue colour on trichrome stain, increased considerably from the 16-week to the 24-week groups (P < 0.05), it was markedly more pronounced in JMC compared with outer cortex in 24-week 2K1C (P < 0.05), and markedly more pronounced as compared with 24-week sham (P < 0.05) (Figs. 2 and 3). Both groups of sham animals had no or minor damage in JMC and outer cortex. The nonclipped kidney from 24-week animals showed some glomerular damage, whereas the other groups did not.
Interstitial collagen (measured using Sirius red) was statistically significantly increased in the JMC compared with outer cortex when analyzed using multiway ANOVA (P < 0.05) (Fig. 4). However, none of the individual comparisons reached statistical significance. Periarterial collagen density increased from 25 ± 1% in 16-week sham rats (exemplified in Fig. 5a) to 34 ± 2% in 24-week sham rats (P < 0.05, Fig. 4, and exemplified in Fig. 5b). In the nonclipped kidney of 2K1C, periarterial collagen was increased to 42 ± 2% at 16 weeks and 67 ± 3% at 24 weeks (P < 0.05) (Fig. 4, and exemplified in Fig. 5c and d).
mRNA and protein expression showed relatively modest changes (Tables 2 and 3). Col1a1 mRNA was increased in nonclipped JMC of 16-week 2K1C. In the 24-week 2K1C mRNA for col1a1, n-pep, TIMP-1, TIMP-2, and MMP-2 were increased in JMC compared with age-matched sham animals (Table 2). A significant effect of age in sham animals was seen as a reduced expression (P < 0.05) of TIMP-1 and TIMP-2 mRNA, and a reduced expression of MMP-2 mRNA in JMC in 24-week compared with 16-week animals (Table 2). In the nonclipped kidney, col1a1mRNA expression increased with age in both outer cortex and JMC. MMP-9 mRNA was higher in outer cortex and n-pep, and TIMP-2 was lower in outer cortex of the 24-week animals compared with the 16-week group (P < 0.05).
No differences were found in the cortical protein expression of TIMP-1, TIMP-2, or MMP-2, but lower MMP-9 expressions were found in the outer cortex of 16-week 2K1C compared with the JMC (P < 0.05, Table 3, representative blots in Fig. 6).
In-situ hybridization for col1a2 showed no activated fibroblasts in sham animals at 16 or 24 weeks (Fig. 7a, d and g). The 16-week 2K1C showed discreet but consistent staining of cells around interlobar and arcuate arteries (Fig. 7b) but not in the cortical interstitium (Fig. 7e). 2K1C at 24 weeks showed activated fibroblasts in the perivascular space of interlobar, arcuate and interlobular arteries (Fig. 7c, 7i), as well as in the interstitium in areas with tissue damage (Fig. 7f). α-SMA staining was consistent with the results of in-situ hybridization. As expected, strong staining was found in vascular smooth muscle cells, but only the 24-week 2K1C group showed consistent staining of activated fibroblasts in the perivascular space and interstitium (Fig. 8).
The present study confirms the hypothesis that hypertensive renal damage in the nonclipped kidney of 2K1C rats starts around the arteries after 16 weeks of hypertension and develops further into tubulointerstitial and glomerular damage in the JMC after 24 weeks. This conclusion is supported by identification of activated fibroblasts in the periarterial space using both in-situ hybridization and immunohistochemistry. Further, periarterial collagen was increased already 16 weeks after clipping, before substantial histological damage or interstitial fibrosis could be identified. Taken together, this shows that glomerular and interstitial damage in the nonclipped kidney of 2K1C only develops after vascular damage has been established, which validates this as a general mechanism behind the development of hypertensive kidney damage in this high-renin model, as previously shown in genetic, and salt-sensitive hypertension. This indicates that the ILA buffers perfusion pressure and protects outer cortex compared with JMC.
The 16-week group developed less severe hypertension and had shorter exposure to hypertension compared with the 24-week group. The two distinct levels of damage are well suited for distinguishing the sequence of progression in hypertensive kidney damage in the nonclipped kidney. In the present study, proteinuria was used to identify the advent of interstitial damage. From a clinical perspective, this is too late, and experimentally, it results in large variation. Albuminuria has been suggested to be an early marker for vascular injury  and studies of the sequential development of vascular damage, albuminuria, and proteinuria in hypertension will be of major future interest.
The deposition of periarterial collagen will likely affect the elastic properties of the vessels, and thereby impair their ability to react to rapid changes in perfusion pressure, which would then be transmitted to the glomeruli. Autoregulation normally buffers any increase in renal perfusion pressure by vasoconstriction of the preglomerular vasculature in order to maintain constant renal blood flow and glomerular hydrostatic pressure , thereby protecting the glomeruli . In SHR, JMC loses autoregulatory efficiency before glomerular damage develops . In our study, activated fibroblasts were found almost exclusively in the direct vicinity of the arterial media. This may indicate that the vascular smooth muscle cells mediate the migration and proliferation of fibroblasts in hypertensive kidney damage. Although this has not been shown directly, smooth muscle cells have been shown to participate in the regulation of renal matrix metabolism .
Hypertensive injury in 2K1C has been suggested to result from a variety of mechanisms such as ischemia , reactive oxygen species , and proteinuria . All of which can be directly affected by angiotensin II. The pioneering study by Mori and Cowley  using servo-controlled renal perfusion pressure under combined salt and angiotensin II treatment showed that the pressure-dependent damage primarily affects the vasculature and the juxtamedullary cortex. This is consistent with the present findings that arteries are affected first in 2K1C hypertension. This indicates that although the above-mentioned factors are important for the development of kidney damage and fibrosis, they appear to do this primarily wherein perfusion pressure is high. The progression of damage from the JMC toward the outer cortex is a strong indication that the damage is pressure dependent.
The gene expression is consistent with the histological data. Both mRNA and protein were measured in cortical slices including mostly tubuli, and only col1a1 was significantly upregulated in the 24-week JMC of the nonclipped kidney. Collagen deposition is regulated at the mRNA level, by degradation of newly synthesized procollagen, and by its conversion into collagen by the removal of C-terminal and N-terminal propeptides by n-pep and c-pep . In the present study, we found that n-pep was upregulated, which validated the earlier finding in SHR , and confirms that it is the more important enzyme for rate regulation.
In the present study, neither MMP-2 nor MMP-9 showed consistent changes. This is probably explained by the relatively slight injury compared with earlier studies . MMPs were originally believed to protect the kidney from fibrosis, however, recent studies suggest that elevated levels of MMP-2 and MMP-9 can disrupt the tubular basement membrane, promote epithelial–mesenchymal transition (EMT), and thereby play a part in the progression of fibrosis [24,25]. If EMT was active in 2K1C, one would expect collagen I-positive cells in the tubuli and elevated levels of MMP-2 and MMP-9 protein. However, in-situ hybridization for collagen I was only positive around vessels and in the interstitium. Thus, at these timepoints, we find no evidence for EMT in the nonclipped kidney of 2K1C, and the importance of MMP-2 and MMP-9 in the early phases of the development of kidney damage in the nonclipped kidney is difficult to infer.
TIMP-1 and TIMP-2 showed a lower expression in the 24-week groups than in the 16-week groups. The interaction of TIMP and MMP in collagen metabolism is complex and involves regulation at many levels. First, TIMPs bind and inhibit MMP activity . In addition, they affect the membrane type-MMP-mediated activation of MMP-2 on the cell surface  so that low TIMP-2 concentration facilitates activation [28,29]. The present finding may indicate that decreased inhibition of MMP activity is a more important mechanism in aging-induced changes than in hypertensive kidney damage per se, as we have found previously .
In conclusion, renal damage starts around the preglomerular vessels in the juxtamedullary cortex of the nonclipped kidney in 2K1C hypertensive rats. This suggests that vascular buffering of perfusion pressure is impaired in the juxtamedullary cortex, and that pressure-induced damage is a necessary component for the development of glomerular and tubulointerstitial damage in this model. Further, perivascular collagen and myofibroblasts are potentially useful early markers for hypertensive damage and could have a place in the histological evaluation of renal biopsies.
The authors wish to acknowledge the skilled technical assistance of Kajsa Bramer.
The study was funded by the Western Norway Regional Health Authority (Helse-Vest) and the Strategic Research Program of Haukeland University Hospital.
Conflicts of interest
There are no conflicts of interest.
Reviewer's Summary Evaluations Referee 2
Skogstrand et al. sought to address the hypothesis that perivascular collagen deposition precedes the development of structural damage, such as tubulointerstitial fibrosis in the two-kidney, one-clip hypertensive (2K1C) model. The major finding was that glomerular and interstitial damage in nonclipped kidney of 2K1C developed after vascular damage. The manuscript is well written, interesting, and presents high quality histology and immunohistology images. A limitation of the study is that it is mostly descriptive and there is a lack of insight into the molecular mechanism of the injury.
The authors studied the mechanisms of kidney injury in the non-clipped kidney in a 2K1C model and provided useful clues for further investigations. The model used in this study has a high renin level whereas the renin level is relatively low in primary hypertension in humans, therefore, the applicability of this study may be limited to certain disease types.
1. Hanratty R, Chonchol M, Havranek EP, Powers JD, Dickinson LM, Ho PM, et al. Relationship between blood pressure and incident chronic kidney disease in hypertensive patients. Clin J Am Soc Nephrol 2011; 6:2605–2611.
2. Hultstrom M, Leh S, Skogstrand T, Iversen BM. Upregulation of tissue inhibitor of metalloproteases-1 (TIMP-1) and procollagen-N-peptidase in hypertension-induced renal damage. Nephrol Dial Transpl 2008; 23:896–903.
3. Gudbrandsen OA, Hultstrom M, Leh S, Bivol LM, Vagnes O, Berge RK, Iversen BM. Prevention of hypertension and organ damage in 2-kidney, 1-clip rats by tetradecylthioacetic acid. Hypertension 2006; 48:460–466.
4. Ochodnicky P, Henning RH, Buikema HJ, de Zeeuw D, Provoost AP, van Dokkum RP. Renal vascular dysfunction precedes the development of renal damage in the hypertensive Fawn-Hooded rat. Am J Physiol Renal Physiol 2010; 298:F625–633.
5. Folkow B, Grimby G, Thulesius O. Adaptive structural changes of the vascular walls in hypertension and their relation to the control of the peripheral resistance. Acta Physiol Scand 1958; 44:255–272.
6. Roald AB, Ofstad J, Iversen BM. Attenuated buffering of renal perfusion pressure variation in juxtamedullary cortex in SHR. Am J Physiol Renal Physiol 2002; 282:F506–F511.
7. Leh S, Hultström M, Rosenberger C, Iversen B. Afferent arteriolopathy and glomerular collapse but not segmental sclerosis induce tubular atrophy in old spontaneously hypertensive rats. Virchows Archiv 2011; 459:99–108.
8. Heyeraas KJ, Aukland K. Interlobular arterial resistance: influence of renal arterial pressure and angiotensin II. Kidney Int 1987; 31:1291–1298.
9. Polichnowski AJ, Cowley AW Jr. Pressure-induced renal injury in angiotensin ii versus norepinephrine-induced hypertensive rats. Hypertension 2009; 54:1269–1277.
10. Mori T, Cowley AW. Role of pressure in angiotensin II-induced renal injury: chronic servo-control of renal perfusion pressure in rats. Hypertension 2004; 43:752–759.
11. Ihara G, Kiyomoto H, Kobori H, Nagai Y, Ohashi N, Hitomi H, et al. Regression of superficial glomerular podocyte injury in type 2 diabetic rats with overt albuminuria: effect of angiotensin II blockade. J Hypertens 2010; 28:2289–2298.
12. Mori T, Polichnowski A, Glocka P, Kaldunski M, Ohsaki Y, Liang M, Cowley AW Jr. High perfusion pressure accelerates renal injury in salt-sensitive hypertension. J Am Soc Nephrol 2008; 19:1472–1482.
13. Ito S, Nagasawa T, Abe M, Mori T. Strain vessel hypothesis: a viewpoint for linkage of albuminuria and cerebro-cardiovascular risk. Hypertens Res 2009; 32:115–121.
14. Helle F, Vagnes OB, Iversen BM. Angiotensin II-induced calcium signaling in the afferent arteriole from rats with two-kidney, one-clip hypertension. Am J Physiol Renal Physiol 2006; 291:F140–F147.
15. Hirose T, Nakazato K, Song H, Ishii N. TGF-beta1 and TNF-alpha are involved in the transcription of type I collagen alpha2 gene in soleus muscle atrophied by mechanical unloading. J Appl Physiol 2008; 104:170–177.
16. Paliege A, Rosenberger C, Bondke A, Sciesielski L, Shina A, Heyman SN, et al. Hypoxia-inducible factor-2 alpha-expressing interstitial fibroblasts are the only renal cells that express erythropoietin under hypoxia-inducible factor stabilization. Kidney Int 2010; 77:312–318.
17. R Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://http://www.R-project.org
18. Navar LG. Renal Autoregulation: perspectives from whole kidney and Single Nephron Studies. Am J Physiol 1978; 234:F357–F370.
19. Bidani AK, Schwartz MM, Lewis EJ. Renal autoregulation and vulnerability to hypertensive injury in remnant kidney. Am J Physiol 1987; 252:F1003–F1010.
20. Dubey RK, Jackson EK, Rupprecht HD, Sterzel RB. Factors controlling growth and matrix production in vascular smooth muscle and glomerular mesangial cells. Curr Opin Nephrol Hypertens 1997; 6:88–105.
21. Lerman LO, Textor SC, Grande JP. Mechanisms of tissue injury in renal artery stenosis: ischemia and beyond. Prog Cardiovasc Dis 2009; 52:196–203.
22. Garcia-Saura MF, Galisteo M, Villar IC, Bermejo A, Zarzuelo A, Vargas F, Duarte J. Effects of chronic quercetin treatment in experimental renovascular hypertension. Mol Cell Biochem 2005; 270:147–155.
23. Leung MK, Fessler LI, Greenberg DB, Fessler JH. Separate amino and carboxyl procollagen peptidases in chick embryo tendon. J Biol Chem 1979; 254:224–232.
24. Yang JW, Shultz RW, Mars WM, Wegner RE, Li YJ, Dai CS, et al. Disruption of tissue-type plasminogen activator gene in mice reduces renal interstitial fibrosis in obstructive nephropathy. J Clin Invest 2002; 110:1525–1538.
25. Cheng SF, Lovett DH, Gelatinase A. (MMP-2) is necessary and sufficient for renal tubular cell epithelial-mesenchymal transformation. Am J Pathol 2003; 162:1937–1949.
26. Eddy AA, Giachelli CM, McCulloch L, Liu E. Renal expression of genes that promote interstitial inflammation and fibrosis in rats with protein-overload proteinuria. Kidney Int 1995; 47:1546–1557.
27. Shimada T, Nakamura H, Ohuchi E, Fujii Y, Murakami Y, Sato H, et al. Characterization of a truncated recombinant form of human membrane type 3 matrix metalloproteinase. Eur J Biochem 1999; 262:907–914.
28. Strongin AY, Collier I, Bannikov G, Marmer BL, Grant GA, Goldberg GI. Mechanism of cell-surface activation of 72-Kda type-Iv collagenase: isolation of the activated form of the membrane metalloprotease. J Biol Chem 1995; 270:5331–5338.
29. Butler GS, Butler MJ, Atkinson SJ, Will H, Tamura T, Schade van Westrum S, et al. The TIMP2 membrane type 1 metalloproteinase ‘receptor’ regulates the concentration and efficient activation of progelatinase A. A kinetic study. J Biol Chem 1998; 273:871–880.
30. Hultstrom M, Leh S, Paliege A, Bachmann S, Skogstrand T, Iversen BM. Collagen-binding proteins in age-dependent changes in renal collagen turnover: microarray analysis of mRNA expression. Physiol Genomics 2012; 44:576–586.