e 52, 20246 Hamburg, Germany. E-mail: Arterial hypertension is the most important factor determining progression to end-stage renal disease regardless of the underlying renal disease. The mechanisms by which hypertension aggravates progression of renal disease are not well understood. Hypertension may exert its adverse effects through vasoconstriction and intrarenal vascular sclerosis causing ischemia1,2. Other studies suggest that hypertension may damage the glomerulus directly by transmission of elevated systemic pressure3. Glomerular hypertrophy may enhance this injury4,5. In the model of 2-kidney, 1-clip Goldblatt hypertension, the effect of ischemia can be studied in the clipped kidney and the effect of elevated glomerular capillary pressure and glomerular hypertrophy in the nonclipped kidney. Nonhemodynamic trophic properties of angiotensin II promoting glomerular and tubular hypertrophy and accumulation of matrix can be studied under two different hemodynamic circumstances. A widely studied model of glomerulonephritis is the Thy-1 nephritis in which a single intravenous injection of an anti-Thy-1 antiserum induces complement dependent mesangiolysis. This lysis is subsequently followed by mesangial cell reappearance, proliferation and matrix deposition. The glomerular lesions of the Thy-1 nephritis resume and hypercellularity and extracellular matrix accumulation disappear within weeks6,7. We recently described that renovascular hypertension enhances impairment of renal function and structure in the acute phase of Thy-1 nephritis. Decreased renal function and increased number of glomerular microaneurysms were found in the nonclipped kidney five days after induction of the nephritis8. However, complete healing of the nephritis was found in the clipped and nonclipped kidney of renovascular hypertensive rats six weeks after induction of the nephritis7. In rats injected with one dose of anti-Thy-1 antibody, the glomerular disease resolves6,7, but if multiple doses are administered in the repetitive hit model, a progressive and irreversible glomerular damage develops9,10. The effects of hypertension and ischemia on this chronic process are unknown and the present study was therefore performed to gain insight whether hypertension has an impact on that process either by enhancing the hypertension specific lesions or by modifying the course of the nephritic changes.
METHODS
Two-kidney, one-clip hypertension
Studies were performed in male Sprague-Dawley rats (Charles River, Kisslegg, Germany). In rats weighing 120 to 140 g, 2-kidney, 1-clip hypertension was induced as described previously7,8,11,12. For this purpose a U-shaped silver clip (0.23 to 0.25 mm internal diameter) was placed around the right renal artery in 26 rats through a loin incision, while the animals were under ketamine/xylazin anesthesia (100/10 mg/kg IM; Parke-Davis/Bayer, Berlin/Leverkusen, Germany). Rats were studied six weeks after surgery. Only those rats with systolic blood pressure>160 mm Hg at six weeks were included in the protocol. One out of 26 clipped rats did not develop hypertension, 4 of 26 clipped rats died within six weeks after surgery due to severe hypertension. In a second study 18 rats were clipped. All rats developed arterial hypertension. The clip was removed six weeks after surgery under ketamine/xylazin anesthesia. Sham surgery of clip removal was performed in normotensive rats.
Induction of glomerulonephritis
Immune mediated mesangial cell injury was induced six weeks after initiation of 2-kidney, 1-clip hypertension in hypertensive and normotensive rats by four consecutive IV injection of 0.5 mL/100 g body weight of an antirat-thymocyte antiserum in weekly intervals. The antiserum was induced in rabbits by repeated immunization with thymocytes from Lewis rats as described13. Four groups of animals were studied: (1) normotensive control animals; (2) normotensive animals + nephritis (Thy-1); (3) 2-kidney, 1-clip hypertensive animals (2K1C); (4) 2-kidney, 1-clip hypertensive animals + nephritis (2K1C + Thy-1).
The study was carried out in two complete sets of experiments. Each set of experiments contained two to six animals in each group. Before injection of antiserum no significant difference was measured for systolic blood pressure in hypertensive controls (178 
10 mm Hg, N = 9) and in hypertensive rats receiving antiserum (193 6 mm Hg, N = 12). Blood pressure was 120 6 and 124 5 mm Hg, respectively, in both normotensive groups before injection of the antiserum. In the second protocol clips were removed six weeks after surgery. Blood pressure was 179 11 mm Hg in hypertensive rats and 190 8 in hypertensive rats receiving antiserum before and 110 7 mm Hg and 113 5 mm Hg, respectively, one to two days after removal of the clip. Blood pressure was 106 5 and 95 11 mm Hg in both normotensive groups, respectively. Induction of the repetitive model of the nephritis was started two to three days after removal of the clip.
Systolic blood pressure and albuminuria
Systolic blood pressure was measured by tail cuff plethysmography in awake rats as previously described7,8,11,12. Blood pressure data shown are the mean of two measurements performed on two consecutive days. The animals were placed in individual metabolic cages and 24-hour urine collections were made for determination of albuminuria. Albuminuria was measured with a commercial enzyme-linked immunosorbent assay (ELISA; WAK Chemie, Bad Soden, Germany) as recommended by the manufacturer.
Renal morphology
At the end of the experimental protocol kidneys were removed and wet weight determined. Kidney slices were fixed in 4% buffered formalin or Carnoy solution, paraffin embedded, cut into 3- to 4-
m thick sections and stained with periodic acid-Schiff (PAS). Glomerular damage was evaluated using a modification of the method of Raij14 as described earlier by us7,8,11,12. Grade 1 represents involvement of up to 25% of the glomerulus and grade 4 represents injury of 75 to 100% of the glomerulus. Thirty glomeruli were evaluated from each kidney. In addition, the amount of tubulointerstitial damage in the kidney (tubular atrophy and dilation, interstitial fibrosis, proteinaceous casts) was estimated by examining 20 fields (magnification 
100) and semiquantitatively grading the degree of damage in each field using a 0 to 3+ scale as described previously7,12. Planimetric examinations of glomerular cross sectional area were performed by means of a Zeiss drawing tube in combination with a semiautomatic interactive image analysis system (Morphomat 30; Zeiss, Oberkochen, Germany) as described7,11,12. Using a serpentine movement from cortex to medulla and vice versa, the outlines of 30 consecutively encountered capillary tufts were traced manually and the mean glomerular random cross sectional area (AG) was determined. The average glomerular tuft volume (VG) was then calculated as VG = (
/k) (AG)3/2, where = 1.38 and k = 1.1 are shape and size distribution coefficients, respectively15. Tissues also were stained with antibodies against ED-1 (Chemikon International, Temecula, CA, USA). The tissue sections were developed with the APAAP complex using antibodies as described earlier7,8. The number of interstitial ED-1 positive cells was counted per high power field (15 fields/kidney, magnification 400).
Western blotting
Western blotting was performed as described recently by us7,13. Glomeruli were isolated by differential sieving and resuspended in Laemmli buffer. Samples were boiled and centrifuged. Protein concentration was determined with the protein DC-assay (Bio-Rad, Muenchen, Germany). To equal amounts of protein (100 g), Laemmli buffer 2 and 1/5 staining solution (42.5% glycerol, 0.5% bromphenol blue) were added. Samples were electrophoresed on SDS/polyacrylamide gels. Proteins were electroblotted onto nitrocellulose (Hybond-ECL; Amersham, Braunschweig, Germany). The membrane was blocked with 5% nonfat dry milk in washing buffer (1 PBS 0.1% Tween 20). The following antibodies were used in a 1:1000 dilution: rabbit anti-porcine TGF- antibody (R&D Systems, Minneapolis, MN, USA), mouse monoclonal anti-human alpha smooth muscle actin antibody (Sigma), cyclooxygenase (COX): goat anti-mouse COX-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and mouse anti-rat COX-2 antibody (Transduction Laboratories, Lexington, MA, USA). The second antibody was a rabbit anti-mouse or rabbit anti-goat or mouse anti-rabbit or mouse anti-porcine horseradish peroxidase-conjugated antibody in a 1:1000 dilution (Transduction Laboratories). Peroxidase labeling was detected with luminescence immunodetection (ECL; Amersham) according to the manufacturer's recommendations. Detection of rabbit IgG was performed by applying the mouse anti-rabbit IgG horseradish peroxidase-conjugated antibody without using a primary antibody (Transduction Laboratories). To control for small variations in protein loading and transfer, membranes were washed and re-incubated with a mouse monoclonal anti-beta-actin antibody (Sigma). Exposed films were scanned with Fluor-STM Multi-Imager (Bio-Rad Laboratories, Hercules, CA, USA), and data were analyzed with the computer program Multi-Analyst™ from Bio-Rad.
The American Physiological Society's guidelines of experimental animal research were followed and approval was obtained from the University animal care committee. A detailed analysis of the fate and the number of animals studied in the present series is given in the Methods and Results sections, as recently suggested in a correspondence to the journal Nature to avoid poorly reported animal tests16.
Statistical analysis
Results are expressed as means SEM unless stated otherwise. Statistical significance was defined as P < 0.05. To compare two distinct treatment groups, we applied the Student t test. For comparison of all groups one way analysis of variance was used as a screening test, if appropriate.
RESULTS
General characteristics
Table 1 shows body, kidney and heart weight of the animals studied. In hypertensive rats hypertrophy of the nonclipped kidney and atrophy of the clipped kidney occurred as determined by kidney weight. Induction of nephritis did not change this pattern. Heart weight was significantly increased in hypertensive animals versus normotensive. Survival after randomization was 100% in all four groups.
Blood pressure
The systolic blood pressure at the end of the study is demonstrated in Figure 1. Systolic blood pressure as measured by tail plethysmography was elevated in hypertensive rats compared with controls and was not statistically different between hypertensive rats and hypertensive rats with chronic Thy-1 nephritis.
Figure 1.
Systolic blood pressure measured in conscious rats two weeks after the last injection of the antiserum. Abbreviations are: 2K1C, 2-kidney, 1-clip hypertensive animals; Thy-1, nephritis. *P < 0.01 vs. controls.
Full figure and legend (10K)Albuminuria
Albuminuria two weeks after the last injection of the Thy-1 antiserum is depicted in Figure 2. Albuminuria was significantly increased in normotensive nephritic rats compared with controls. In addition, urinary albumin excretion was significantly increased in 2K1C rats. Moreover, a further and significant increase was found in hypertensive nephritic rats.
Figure 2.
Albuminuria in the various groups. Abbreviations are: 2K1C, 2-kidney, 1-clip hypertensive animals; Thy-1, nephritis. *P < 0.01, **P < 0.001 vs. controls.
Full figure and legend (10K)Glomerular binding of the antiserum
Following the injection of rabbit anti-thymocyte-serum, selective binding of rabbit IgG to the mesangium can be detected8. Figure 3 shows a Western blot of glomerular proteins using an anti-rabbit IgG antibody. Bands of similar intensity were found in normotensive nephritic rats and the clipped as well as the nonclipped kidney of nephritic 2K1C rats after four injections of the antiserum. This finding indicates no differences in delivery and binding of antiserum in normotensive and hypertensive rats8.
Figure 3.
Western blots for rabbit IgG from glomerular proteins. (A) Similar amounts of rabbit IgG were detected in glomeruli from normotensive nephritic and both kidneys of hypertensive nephritic rats. The blot was washed and re-incubated with an antibody against -actin to proof similarity in protein loading and transfer. The glomerular lysate used in this Western blot was pooled from 3 to 6 kidneys in each group. A very similar pattern was found in an another Western blot from an independent second set of experiments again using pooled proteins from 2 to 6 kidneys. (B) Densitometric analysis of the IgG bands from both blots.
Histologic finding
Quantitative evaluation of PAS stained sections obtained two weeks after the last of the four injections of the antiserum was performed. Intact glomeruli were found in normotensive controls Figure 4a. In contrast, chronic glomerulonephritis with matrix deposition, hypercellularity and scarring could be detected in normotensive nephritic rats Figure 4b. The clipped Figure 4c as well as the nonclipped kidney Figure 4d showed changes consistent with earlier reports by us and others of kidney morphology in rats with renovascular hypertension7,8,11,12,17,18. However, no chronic glomerulonephritis could be detected in the clipped kidney after four injections of antiserum Figure 4e. In contrast, dramatically enhanced glomerular damage was found in the nonclipped nephritic kidney Figure 4f. Scoring of the described changes is summarized in Figure 4g.
Figure 4.
Glomeruli from thfigf4ge different groups. Glomeruli from controls revealed a normal appearance (A). Chronic glomerulonephritis with matrix deposition, hypercellularity and increased glomerular scarring was found in normotensive nephritic rats (B). The glomeruli of the clipped kidney generally had a normal appearance (C). A light micrograph of glomeruli from the nonclipped kidney demonstrated segmental sclerosis (D). Surprisingly, light microscopy of glomeruli in the clipped kidney from the hypertensive and nephritic group revealed almost no increased damage (E). In contrast, a dramatic increase in glomerular damage was detected in the nonclipped kidneys from the same group (F; original magnification 400). Scoring of the glomerular damage is shown in G. **P < 0.001, *P < 0.01 versus normotensive controls, #P < 0.05 vs. nonclipped kidney of non-nephritic 2K1C.
Alpha-smooth muscle actin (
-SMA) is a very sensitive marker of mesangial cell activation and injury19. To further explore the changes in glomerular injury detected by light microscopy we measured glomerular levels of alpha-smooth muscle actin by Western blotting. As shown in Figure 5, -SMA expression was up-regulated in normotensive nephritic rats as well as in the clipped and nonclipped kidney of hypertensive nephritic rats indicating activation of the glomerular cells in both kidneys of 2K1C nephritic rats.
Figure 5.
Western blotting for 
-smooth muscle actin from glomerular proteins. (A) -Smooth muscle expression was heavily induced in glomeruli from normotensive nephritic rats as well as in the clipped and in the nonclipped kidney from nephritic 2K1C rats. The blot was washed and re-incubated with an antibody against -actin. The glomerular lysate used in this Western blot was pooled from 3 to 6 kidneys in each group. A very similar pattern was found in an another Western blot from an independent second set of experiments using again pooled proteins from 2-6 kidneys. (B) Densitometric analysis of the actin bands from both blots.
Tubulointerstitial damage and glomerular volume
It is known that tubulointerstitial damage precedes glomerular injury in the nonclipped kidney of Goldblatt hypertensive rats, suggesting a dominant pathophysiologic role for the tubulointerstitial damage in nephrosclerosis17,18. We therefore evaluated the pattern of tubulointerstitial injury. In accordance with the glomerular data, tubulointerstitial damage was increased in the nonclipped kidney of hypertensive nephritic rats compared with the nonclipped kidney of 2K1C rats (P < 0.02). Tubulointerstitial damage was increased in the clipped kidney. However, no significant difference could be detected between the clipped kidneys of hypertensive rats with and without nephritis Figure 6a. Light micrographs of the tubulointerstitial changes found in the different kidneys are shown in Figure 6b-g. Intact tubuli were found in normotensive controls Figure 6b, while tubulointerstitial damage could be detected in normotensive nephritic rats Figure 6c. The clipped Figure 6d as well as the nonclipped kidney Figure 6e showed changes consistent with earlier reports by us7,8,11,12 and others17,18 of kidney morphology in rats with renovascular hypertension. Tubulointerstitial injury could be detected in the clipped kidney after four injections of antiserum Figure 6f. In contrast, dramatically enhanced tubulointerstitial damage was found in the nonclipped nephritic kidney Figure 6g.
Figure 6.
Tubulointerstitial damage evaluated by scoring. (A) Tubulointerstitial damage was increased in the nonclipped kidney of hypertensive nephritic rats compared with the nonclipped kidney of 2K1C rats (P < 0.02). (B–G) Light micrographs of the tubulointerstitial changes are shown in (B) controls, (C) Thy-1, (D) clipped kidney, (E) nonclipped kidney. (F) clipped kidney + Thy-1, (G) nonclipped kidney + Thy-1 (original magnification 200). (H) The number of interstitial ED-1 positive cells per high power field (15 fields/kidney). The number of interstitial monocyte/macrophage infiltration was increased in the clipped kidneys. The highest number was observed in the nonclipped kidney of nephritis 2K1C rats. However, due to the high variability the found differences were not statistically significant. (I) Glomerular volume measured by planimetry of glomerular cross sectional areas. All nephritic glomeruli revealed a significant increase in glomerular size over their respective controls. *P < 0.05, **P < 0.01 vs. normotensive controls.
Monocyte/macrophages are important mediators of renal injury. We therefore examined the interstitial infiltration of monocytes. As shown in Figure 6h, the number of monocytes was closely associated with the degree of tubulointerstitial injury. An increased number was found in the clipped kidney of nephritic and nonnephritic hypertensive rats. The highest number was detected in the nonclipped kidney of nephritic 2K1C rats. Glomerular hypertrophy is another important parameter contributing to the progression of chronic renal disease4,5. At least to our knowledge, no data are available on the changes of glomerular size in the repetitive model of the Thy-1 nephritis. As shown in Figure 6i glomerular hypertrophy occurred in the nonclipped kidney and a decrease of glomerular size was noted in the clipped kidney compared with controls. Interestingly, glomerular size was significantly increased in normotensive nephritic rats as well as in the clipped and nonclipped kidney of hypertensive nephritic rats compared with nonnephritic rats.
TGF-
We and others have demonstrated that TGF- plays a pivotal role in the Thy-1 nephritis20,21 as well as in Goldblatt hypertension7,22. To further quantify glomerular TGF- levels Western blotting of glomerular lysates was performed. Similar to our previous finding in rats with short- and long-term 2K1C hypertension7,22, hypertensive rats in the present study 12 weeks after clipping showed increased levels of TGF- in the nonclipped kidney compared with the clipped kidney and controls Figure 7. Glomerular expression of TGF- was modestly increased in normotensive nephritic rats and the clipped kidney of nephritic 2K1C rats. However, a strong increase of TGF- abundance was found in the nonclipped kidney of nephritic 2K1C rats matching the increased glomerular damage found in these kidneys.
Figure 7.
Expression of glomerular transforming growth factor- (TGF-) protein levels. (A) The 21- and 30-kD molecular marker is shown. A single band of around 25 kD is visible reflecting the TGF- dimer. Increased expression was visible in normotensive nephritic rats. Hypertensive rats showed increased levels of TGF- in the nonclipped kidney compared with the clipped kidney and controls. The expression of TGF- in the clipped and nonclipped kidney of hypertensive nephritic rats was increased. The blot was washed and re-incubated with an antibody against -actin. The glomerular lysate used in this Western blot was pooled from three to six kidneys in each group. A very similar pattern was found in an another Western blot from an independent second set of experiments using again pooled proteins from two to six kidneys in each group. (B) Densitometric analysis of the TGF- bands from both blots.
COX-1 and COX-2
Recent evidence from models of inflammatory kidney disease revealed that COX products are important in inflammation and might participate in the healing process23. Moreover, we have demonstrated different glomerular levels of prostaglandins in rats with 2K1C hypertension24. Therefore, a differential induction of COX-1 and COX-2 might cause the different development of damage in the Goldblatt rats. The glomerular abundance of COX-1 and COX-2 was evaluated by Western blotting as an initial screening test. Modestly increased levels of COX-1 and COX-2 were found in the nonclipped kidney compared with the clipped kidney and controls Figure 8a, c. Glomerular expression of both proteins was further increased in normotensive nephritic rats, in the clipped kidney and in the nonclipped kidney of nephritic 2K1C rats.
Figure 8.
Western blotting for COX-1 (A) and COX-2 (C) from glomerular proteins. Both enzymes were induced in glomeruli from normotensive nephritic rats as well as in the clipped and in the nonclipped kidney from nephritic 2K1C rats. The blots were washed and re-incubated with an antibody against -actin. The glomerular lysate used in both Western blots was pooled from 3 to 6 kidneys in each group. A very similar pattern was found in two Western blots from an independent second set of experiments using again pooled proteins from 2 to 6 kidneys in each group. (B and D) Densitometric analyses of the COX bands from both blots are shown below.
Removal of the clip
We next asked whether the aggravation of glomerular damage in the nonclipped and sparing of the clipped kidney depends on the presence of the stenosis and high blood pressure or is due to irreversible structural changes in both kidneys at the time when repetitive injections of the Thy-1 antiserum started. Therefore, nephritis was induced in rats immediately after removal of the clip. Systolic blood pressure was normotensive six weeks after removal of the clip in all four groups studied [controls 133 4, Thy-1 108 6, 2K1C(ex) 131 5, 2K1C(ex) + Thy-1 135 5 mm Hg]. Survival was 100% in all four groups. As shown in Figure 9 glomerular damage increased in all nephritic kidneys and no significant difference could be detected between the formerly clipped kidney and the nonclipped kidney in nephritic 2K1C rats. In addition, albuminuria was not significant different between 2K1C(ex) (1.38 0.38 mg/24 h) and 2K1C(ex) + Thy-1 (5.29 2.15 mg/24 h). Albuminuria was 0.26 0.10 mg/24 h in controls and 2.72 0.88 mg/24 h in normotensive Thy-1 rats. As shown in Figure 10, glomerular binding of IgG and induction of TGF- occurred to a similar extent in normotensive nephritic rats and both kidneys of the unclipped nephritic rats.
Figure 9.
Glomerular damage evaluated by scoring. No more damage was found in the formerly nonclipped kidneys 6 weeks after removal of the clip. Glomerular damage was significantly increased in normotensive nephritic rats and in the formerly clipped and nonclipped nephritic kidneys. *P < 0.05, **P < 0.01 vs. normotensive controls.
Full figure and legend (14K)Figure 10.
Western blotting from glomerular proteins for rabbit IgG (A) and TGF- (B) is shown. The glomerular lysate used in the Western blot was pooled from 4 rats in controls and Thy-1, respectively, and 9 rats in both clipped (ex) groups, respectively.
DISCUSSION
Systemic hypertension is a major risk factor that determines the rate of progression of renal disease. We and others have recently studied and characterized the model of anti-thymocyte serum induced glomerulonephritis6,7,8,9,13,20,21,23,25. The Thy-1 nephritis is a model of immune mediated glomerular injury, in which hypertension normally does not occur. This makes it a good model to study the effects of superinduced hypertension on glomerular and tubulointerstitial injury because hypertension and nephritis can be experimentally separated. In agreement with the detrimental effect of high blood pressure on renal disease, we described aggravation of the glomerular lesion five days after induction of the Thy-1 nephritis in rats with 2-kidney, 1-clip hypertension8. In contrast, we found complete healing of the glomerular lesions of the Thy-1 nephritis in renovascular hypertensive rats after six weeks, indicating that neither glomerular hypertension and hypertrophy nor ischemia inhibits healing of this form of glomerulonephritis7.
The mechanisms by which hypertension damages the kidney in renal disease are incompletely understood. Our present studies document accelerated development of glomerular lesions in the kidney exposed to hypertension compared with the clipped kidney in the same rat or to kidneys of normotensive nephritic animals. Furthermore, the clipped kidneys were almost completely protected from sclerotic nephritic lesions. The most straightforward explanation for these findings is nephron hemodynamics. Normal to high glomerular pressure is an absolute requirement for injury3. Elevated glomerular capillary pressure and glomerular blood flow as well as increased single nephron glomerular filtration rate have been found in the nonclipped kidney of Goldblatt hypertensive rats26,27. Thus, the advanced glomerular damage in the nonclipped kidney might be caused by altered glomerular hemodynamics. Although single nephron hemodynamics are less well documented in the clipped kidney26, it is reasonable to assume that glomerular hemodynamics in these kidneys may be altered in a direction that protects the glomerulus from the chronic immunological damage. Angiotensin-converting enzyme (ACE) inhibition lowers glomerular pressure, which is beneficial in several models of renal disease. Therefore, the protection against glomerular damage may be explainable by the altered hemodynamics in the clipped kidney that somehow mimic the hemodynamic effects of ACE inhibition. Additional experiments will be clearly required to address these issues. However, in a previous study we did not find a protective effect of the clipped kidney on the early phase of the Thy-1 nephritis, since proliferation and matrix deposition were not different between kidneys of normotensive nephritic rats and the clipped kidney of Goldblatt hypertensive nephritic rats8. Goldblatt hypertension is at least in part an angiotensin II-dependent model of hypertension28. Especially the clipped kidney is rich in renin29. Although the harmful effects of angiotensin are well established30, angiotensin II is also a potent inducer of prostaglandins in isolated glomeruli31. We have recently demonstrated that administration of prostaglandin E2 reduces collagen deposition in rats with Thy-1 nephritis32. Therefore, it is reasonable to speculate that increased production of prostaglandins or other antifibrogenic substances in the clipped kidney might have contributed to the reduced glomerular damage. We evaluated glomerular expression of COX-1 and COX-2 as a screening test to determine whether different induction of prostaglandins may influence glomerular damage. However, the pattern of induction of COX-1 and COX-2 was similar in the clipped as well as in the nonclipped kidney of nephritic 2K1C rats, making this possibility unlikely. In addition, angiotensin II receptors are down-regulated in the clipped kidney29. Therefore, the increased glomerular and tubulointerstitial damage in the nonclipped kidney and complete protection of the clipped kidney could be at least in part explained by the fact that the angiotensin II receptors are down-regulated in the clipped kidney. However, the function of the clipped kidney depends on angiotensin II, since prolonged ACE-inhibition induces severe tubulointerstitial damage in the clipped kidney as shown by us and others11,33.
A significant increase of albuminuria was found in hypertensive nephritic rats in the present study. We have not separately assessed albuminuria of the clipped and nonclipped kidney because we know from our previous work in 2-kidney, 1-clip hypertensive rats that most of the proteinuria is derived from the nonclipped kidney12. Moreover, 1-kidney, 1-clip hypertensive rats do not develop significant albuminuria (unpublished data). From these observations we believe that the increased albuminuria in the nephritic 2K1C rats originates from the nonclipped kidney.
We have recently studied in detail the increased glomerular size of the nonclipped kidney in rats with 2-kidney, 1-clip hypertension7,11,12. According to the law of Laplace an increased glomerular size causes an increased wall tension. The increased glomerular size in the nonclipped kidney may therefore enhance the damage caused by the immunologic events of the nephritis. Very similarly, the increased glomerular size in the normotensive nephritic rats may have facilitated the development of glomerular damage. However, glomerular hypertrophy by itself is not sufficient to induce sclerosis since the increased glomerular size of the clipped kidney in nephritic 2K1C rats did not enhance glomerular damage.
It has been described that tubulointerstitial changes precede glomerular damage in the nonclipped kidney of Goldblatt hypertensive rats17,18. Glomerular and tubulointerstitial damage in the nonclipped kidney correlates well in the present study; however, a dissociation was found in the clipped kidney. Whereas the glomerular structure is well preserved, tubulointerstitial damage occurs in all studied groups. This dissociation clearly needs to be investigated further. We are currently attempting to understand the properties by studying separately tubular and glomerular expression of potential modulators of damage in Goldblatt hypertensive rats (abstract; Wenzel et al, J Am Soc Nephrol 11:449, 2000).
One of the most intensively studied profibrogenic cytokines is TGF-34. We have recently described up-regulation of TGF- in the nonclipped kidney of rats with 2K1C hypertension7,22. In the present study there was a modest up-regulation of TGF- in the clipped kidney and in normotensive rats after induction of chronic nephritis. Interestingly, a strong overexpression of TGF- was found in the nonclipped kidney after induction of the nephritis that matched the increased glomerular damage and matrix deposition in this kidney. However, despite increased levels of TGF- no glomerular damage was found in the clipped kidney of hypertensive nephritic rats, indicating that increased levels of TGF- itself are not sufficient to induce glomerular damage. Glomerular damage and induction of TGF- of similar magnitude were found in the clipped and nonclipped kidneys after removal of the clip, indicating that the clip as well as arterial hypertension are necessary for the different induction of damage in both kidneys. It shows that the enhanced damage in the nonclipped kidney and the protection of the clipped kidney are not caused by the structural changes like glomerular hypertrophy or atrophy six weeks after clipping of the renal artery.
To our knowledge, the present study is the first demonstrating enhanced glomerular damage in the nonclipped kidney and sparing of the glomeruli in the clipped kidney of Goldblatt hypertensive rats with the repetitive hit Thy-1 nephritis. Our data are in line with the finding of Mauer and colleagues, who described less diabetic lesions in the glomeruli of the clipped kidney of diabetic rats with Goldblatt hypertension35. Moreover, there is an observation in a patient with unilateral renal artery stenosis and diabetes that the stenotic kidney had only ischemic changes while the nonclipped kidney demonstrated advanced diabetic glomerulopathy36. Palmer, Eversole and Stamey also reported unilateral nephritis in patients with renal artery stenosis37. In addition, Dikman et al described in a clinical report a patient with unilateral glomerulonephritis, in whom a hydronephrotic kidney was spared from nephritic damage38. Our present observation and these clinical reports are in line with a growing body of literature that certain models of renal disease induce a cross tolerance to a subsequent injury. It was recently shown that the remnant kidney is protected against ischemia/reperfusion injury39. In addition, endotoxin-induced cross tolerance to subsequent renal ischemia/reperfusion injury has been described40. We even found that angiotensin II infusion surprisingly ameliorated the early phase of the Thy-1 nephritis41. Although it is most likely that reduced glomerular pressure in the clipped kidney explains that no glomerular damage developed in this kidney, another possibility is that mild ischemia in the clipped kidney induced cross tolerance against the inflammatory injury of the chronic Thy-1 nephritis.
Three major conclusions can be drawn from these data and our previous studies7,8. First, renovascular hypertension aggravates the glomerular damage in the early phase of the Thy-1 glomerulonephritis. Second, whereas renovascular hypertension does not prevent healing of the one shot Thy-1 nephritis, an increased glomerular damage is found in the nonclipped kidney after induction of the repetitive hit model of this nephritis. The third and particularly interesting observation is that despite activation of glomerular cells and an increased expression of TGF- and COX-1 and -2 as well as an increased glomerular volume, no glomerular damage develops in the clipped kidney of nephritic 2K1C rats. The mechanisms protecting the clipped kidney remain to be determined.
In summary, induction of the repetitive hit model of the Thy-1 nephritis in rats with 2-kidney, 1-clip hypertension induces severe glomerular damage in the nonclipped kidney. In contrast, no glomerular damage develops in the clipped kidney after induction of this chronic glomerulonephritis.
References
| 1. | Wenzel U & Helmchen U. Effects of hypertension on renal vasculature and structure. inOxford Textbook of Clinical Nephrology 1997; edited by Davison AM, Cameron JS, Grünfeld JP, et al Oxford, Oxford Medical Publications. |
| 2. | Helmchen U & Wenzel UO. Benign and malignant nephrosclerosis and renovascular hypertension. inRenal Pathology. With Clinical and Functional Correlations 1994; (2nd ed), edited by Tisher CC, Brenner BM Philadelphia, JB Lippincott Company pp 1201−1238. |
| 3. | Brenner BM. Nephron adaption to renal injury or ablation. Am J Physiol 1985; 249: F324−F337. | PubMed | ISI | ChemPort | |
| 4. | Miller PL, Rennke HG & Meyer TW. Glomerular hypertrophy accelerates hypertensive glomerular injury in rats. Am J Physiol 1991; 261: F459−F46543. | PubMed | ISI | ChemPort | |
| 5. | Yoshida Y, Fogo A & Ichikawa I. Glomerular hemodynamic changes vs. hypertrophy in experimental glomerular sclerosis. Kidney Int 1989; 35: 654−66010. | PubMed | ISI | ChemPort | |
| 6. | Bagchus WM, Jeunik MF & Elema JD. The mesangium in anti-Thy-1 nephritis. Am J Pathol 1990; 137: 215−223. | PubMed | ISI | ChemPort | |
| 7. | Wenzel UO, Wolf G & Thaiss F et al. Renovascular hypertension does not influence repair of the glomerular lesions induced by the anti-thymocyte glomerulonephritis. Kidney Int 2000; 58: 1135−1147 10.1046/j.1523-1755.2000.00270.x. | Article | PubMed | ISI | ChemPort | |
| 8. | Wenzel UO, Thaiss F & Panzer U et al. Effect of renovascular hypertension on experimental glomerulonephritis in rats. J Lab Clin Med 1999; 134: 292−303. | Article | PubMed | ISI | ChemPort | |
| 9. | Stahl RAK, Thaiss F, Wenzel U & Helmchen U. Morphologic and functional consequences of immune mediated mesangiolysis: Development of chronic glomerular sclerosis. J Am Soc Nephrol 1992; 2 Suppl: S144−S148. | PubMed | ISI | ChemPort | |
| 10. | Harendza S, Schneider A, Helmchen U & Stahl RAK. Extracellular matrix deposition and cell proliferation in a model of chronic glomerulonephritis in the rat. Nephrol Dial Transplant 1999; 14: 2873−2879. | PubMed | ISI | ChemPort | |
| 11. | Wenzel UO, Helmchen U, Schoeppe W & Schwietzer G. Combination treatment of enalapril with nitrendipine in rats with renovascular hypertension. Hypertens 1994; 23: 114−122. | ISI | ChemPort | |
| 12. | Wenzel UO, Helmchen U & Troschau G et al. Adverse effect of the calcium channel blocker nitrendipine on nephrosclerosis in rats with renovascular hypertension. Hypertens 1992; 20: 233−241. | ISI | ChemPort | |
| 13. | Wenzel U, Schneider A & Valente AJ et al. Monocyte chemoattractant protein 1 mediates monocyte/macrophage influx in anti-thymocyte antibody-induced glomerulonephritis. Kidney Int 1997; 51: 770−776. | PubMed | ISI | ChemPort | |
| 14. | Raij L, Azar S & Keane W. Mesangial injury, hypertension and progressive glomerular damage in Dahl rats. Kidney Int 1984; 26: 137−275. | PubMed | ISI | ChemPort | |
| 15. | Nyengaard JR. Stereologic methods and their application in kidney research. J Am Soc Nephrol 1999; 10: 1100−1123. | PubMed | ISI | ChemPort | |
| 16. | Hampshire V. Poorly conducted (or reported) animal tests put humans at risk (letter). Nature 2000; 407: 671 10.1038/35037798. | Article | PubMed | ISI | ChemPort | |
| 17. | Mai M, Geiger H & Hilgers KF et al. Early interstitial changes in hypertension-induced renal injury. Hypertens 1993; 22: 754−765. | ISI | ChemPort | |
| 18. | Eng E, Veniant M & Floege J et al. Renal proliferative and phenotypic changes in rats with two-kidney, one-clip Goldblatt hypertension. Am J Hypertens 1994; 7: 177−185. | PubMed | ISI | ChemPort | |
| 19. | Johnson RJ, Iida H & Alpers CE et al. Expression of smooth muscle cell phenotype by rat mesangial cells in immune complex nephritis. J Clin Invest 1991; 87: 847−858. | PubMed | ISI | ChemPort | |
| 20. | Schneider A, Panzer U & Zahner G et al. Monocyte chemoattractant protein 1 mediates collagen deposition in experimental glomerulonephritis by transforming growth factor beta. Kidney Int 1999; 56: 135−144 10.1046/j.1523-1755.1999.00543.x. | Article | PubMed | ISI | ChemPort | |
| 21. | Border WA, Okuda S & Languino LR et al. Suppression of experimental glomerulonephritis by antiserum against transforming growth factor beta 1. Nature 1990; 346: 371−374. | Article | PubMed | ISI | ChemPort | |
| 22. | Wolf G, Schneider A & Wenzel U et al. Regulation of glomerular TGF-Beta expression in the contralateral kidney of two-kidney, one-clip hypertensive rats. J Am Soc Nephrol 1998; 9: 763−772. | PubMed | ISI | ChemPort | |
| 23. | Schneider A, Harendza S & Zahner G et al. Cyclooxygenase metabolites mediate glomerular monocyte chemoattractant protein-1 formation and monocyte recruitment in experimental glomerulonephritis. Kidney Int 1999; 55: 430−441 10.1046/j.1523-1755.1999.00265.x. | Article | PubMed | ISI | ChemPort | |
| 24. | Stahl RA, Helmchen U & Paravicini M et al. Glomerular prostaglandin formation in two-kidney, one-clip hypertensive rats. Am J Physiol 1984; 247: F975−F981. | PubMed | ISI | ChemPort | |
| 25. | Floege J, Eng E & Lindner V et al. Rat glomerular mesangial cells synthesize basic fibroblast growth factor. Release, upregulated synthesis, and mitogenecity in mesangial proliferative glomerulonephritis. J Clin Invest 1992; 90: 2362−2369. | PubMed | ISI | ChemPort | |
| 26. | Schwietzer G & Gertz KH. Changes of hemodynamics and glomerular ultrafiltration in renal hypertension of rats. Kidney Int 1979; 15: 134−143. | PubMed | ISI | ChemPort | |
| 27. | Azar S, Johnson MA, Hertel B & Tobian L. Single-nephron pressures, flows and resistance in hypertensive kidneys with nephrosclerosis. Kidney Int 1977; 12: 28−40. | PubMed | ISI | ChemPort | |
| 28. | Ploth DW. Angiotensin-dependent renal mechanisms in two-kidney, one clip renal vascular hypertension. Am J Physiol 1983; 245: F131−F141. | PubMed | ISI | ChemPort | |
| 29. | Della Bruna R, Bernhard I & Gess B et al. Renin gene and angiotensin II AT1 receptor gene expression in the kidneys of normal and two-kidney/one-clip rats. Eur J Physiol 1995; 430: 265−272. | ChemPort | |
| 30. | Wolf G. Angiotensin II: A pivotal factor in the progression of renal diseases. Nephrol Dial Transplant 1999; 14 Suppl 1: 42−44. | PubMed | ISI | ChemPort | |
| 31. | Stahl RAK, Paravicini M & Schollmeyer P. Angiotensin II stimulation of prostaglandin E2 and 6-keto-F1alpha formation by isolated human glomeruli. Kidney Int 1984; 26: 30−34. | PubMed | ISI | ChemPort | |
| 32. | Schneider A, Thaiss F & Rau HP et al. Prostaglandin E1 inhibits collagen expression in anti-thymocyte antibody induced glomerulonephritis: Possible role of TGF-beta. Kidney Int 1996; 50: 190−199. | PubMed | ISI | ChemPort | |
| 33. | Jackson B, Franze C, Sumithran E & Johnston I. Pharmacologic nephrectomy with chronic angiotensin converting enzyme inhibitor treatment in renovascular hypertension in the rat. J Lab Clin Med 1990; 115: 21−27. | PubMed | ISI | ChemPort | |
| 34. | Border WA & Noble NA. Transforming growth factor beta in tissue fibrosis. N Engl J Med 1994; 331: 1286−1292. | Article | PubMed | ISI | ChemPort | |
| 35. | Mauer SM, Steffes MW & Azar S et al. The effects of Goldblatt hypertension on development of the glomerular lesions of diabetes mellitus. Diabetes 1978; 27: 738−744. | PubMed | ISI | ChemPort | |
| 36. | Berkman J & Rifkin H. Unilateral nodular diabetic glomerulosclerosis (Kimmelstiel-Wilson). Report of a case. Metab Clin Exp 1973; 22: 715−722. | PubMed | ChemPort | |
| 37. | Palmer JM, Eversole SL & Stamey TA. Unilateral glomerulonephritis. Virtual absence of nephritis in a kidney with partial occlusion of the main renal artery. Am J Med 1966; 40: 816−822. | Article | PubMed | ISI | ChemPort | |
| 38. | Dikman SH, Strauss L & Berman LJ et al. Unilateral glomerulonephritis. Arch Pathol Lab Med 1976; 100: 480−483. | PubMed | ISI | ChemPort | |
| 39. | Vercauteren SR, Ysebaert DK & De Greef KE et al. Chronic reduction in renal mass in the rat attenuates ischemia/reperfusion injury and does not impair tubular regeneration. J Am Soc Nephrol 1999; 10: 2551−2561. | PubMed | ISI | ChemPort | |
| 40. | Heemann U, Szabo M & Hamar P et al. Lipopolysaccharide pretreatment protects from renal ischemia/reperfusion injury. Possible connection to an interleukin-6-dependent pathway. Am J Pathol 2000; 156: 287−293. | PubMed | ISI | ChemPort | |
| 41. | Wenzel UO, Thaiss F & Helmchen U et al. A novel role for angiotensin II: Angiotensin II infusion ameliorates the early phase of a mesangioproliferative glomerulonephritis. Kidney Int 2002; 61: 1020−1029. | Article | PubMed | ISI | ChemPort | |
Acknowledgments
This study was supported by the Deutsche Forschungsgemeinschaft (We 1688/4-3 and 4-6). Parts of this study were presented at the Annual Meeting of the American Society of Nephrology, Toronto, 2000 and published in abstract form (J Am Soc Nephrol 11:501A, 2000).
