Renal Perfusion Pressure Determines Infiltration of Leukocytes in the Kidney of Rats with Angiotensin II Induced Hypertension

Abstract

The present study examined the extent to which leukocyte infiltration into the kidneys in angiotensin II (AngII)-induced hypertension is determined by elevation of renal perfusion pressure (RPP). Male Sprague-Dawley rats were instrumented with carotid and femoral arterial catheters for continuous monitoring of blood pressure (BP) and a femoral venous catheter for infusion. An inflatable aortic occluder cuff placed between the renal arteries with computer-driven servo-controller maintained RPP to the left kidney at control levels during 7-days of intravenous AngII (50 ng/kg/min) or vehicle (saline) infusion. Rats were fed a 0.4% NaCl diet throughout the study. AngII infused rats exhibited nearly a 50 mmHg increase of RPP (carotid catheter) to the right kidney while RPP to the left kidney (femoral catheter) was controlled at baseline pressure throughout the study. As determined at the end of the studies by flow cytometry, right kidneys exhibited significantly greater numbers of T cells, B cells, and monocytes/macrophages compared to the servo-controlled left kidneys and compared to vehicle treated rats. No difference was found between AngII servo-controlled left kidneys and vehicle treated kidneys. Immunostaining found that the density of glomeruli, cortical and outer medullary capillaries were significantly reduced in the right kidney of AngII infused rats compared to servo-controlled left kidney. We conclude that in this model of hypertension the elevation of RPP, not AngII nor dietary salt, leads to leukocyte infiltration in the kidney and to capillary rarefaction.

Graphical Abstract

Introduction

The importance of renal infiltrating leukocytes in the development of hypertension and end-organ damage is now well recognized. Currently under investigation are mechanisms responsible for the entry of the leukocytes into the kidney that stimulate pathways of inflammation and oxidative stress. In the present study, the importance of elevated renal perfusion pressure (RPP) upon leukocyte infiltration in angiotensin II (AngII) induced hypertension in rats was examined. Considerations of the injurious effects of elevated RPP upon the integrity of the glomeruli (barotrauma) evolved from studies by Brenner and colleagues ^{1, 2} and advanced by Bidani and Griffin^3–6 and studies in our laboratory^7–12. Direct evidence and the quantitative relevance of the contribution of RPP to renal injury in hypertension was obtained as it became possible to control (servo-control) RPP to a single kidney while the contralateral kidney was exposed to hypertension. Using a custom designed computer-controlled servo-control system, several rat models of hypertension have been studied in this manner including the AngII + high salt model^{8, 11}, the AngII + L-NAME model¹², the norepinephrine model¹⁰, and the Dahl salt-sensitive rat (Dahl S)^{7, 9}. The impact of RPP to influence renal injury is variable and dependent upon the model of hypertension that is studied.

As the importance of inflammation and immunity in hypertension was revealed using both immunosuppressive agents and genetic deletion approaches^13–19, flow cytometry experiments demonstrated the numbers and types of immune cells that infiltrated into the kidney in various models of hypertension. Renal infiltration of nearly equal numbers of infiltrating T helper (CD4⁺), and cytotoxic T cells (CD8⁺) was found in hypertensive Dahl S rats fed a high salt diet^{20, 21} and it was found that prevention of a rise of RPP by servo-control methods attenuated renal infiltration of immune cells and the accompanying renal damage⁷. Given these observations, the question arose as to whether these changes were unique to the Dahl S model and whether they could meaningfully be extrapolated to other models of hypertension. The present study was therefore undertaken to determine the extent to which leukocyte infiltration into the kidney would be reduced by protecting the kidney from experimental hypertension in a genetically different strain of rat. In the present study, AngII induced hypertension was induced in Sprague-Dawley (SD) rats fed 0.4% NaCl throughout the study while RPP to the left kidney was servo-controlled to remain a normal control RPP and compared to the right uncontrolled kidneys. This internally paired comparison of the left and right kidneys eliminated inter-rat variations and provided a model in which both kidneys were perfused with same circulating arterial blood. The number of leukocytes infiltrating to kidneys were measured with a flow cytometric-based analysis and the renal glomerular and peritubular capillary density was determined by morphologic observations.

Methods

Detailed experimental methods and description of antibodies are available in the online-only Data Supplement.

Male SD rats (9 weeks of age) were purchased from Envigo (Indianapolis, IN) and housed in environmentally controlled rooms under a 12-h light/dark cycle. Rats had free access to 0.4% NaCl AIN-76A diet and water ad libitum. All protocols were approved by the Medical College of Wisconsin Institutional Animal Care and Use Committee. At 10-11 weeks of age, chronic servo-controlling surgeries were done as previously described⁸ (Figure S1). After 7 days of recovery, baseline blood pressure was measured for 3 days, then the rats were infused with AngII (50 ng/kg/min) intravenously for 7 days to induce hypertension or saline as a vehicle control. We note that in previous manuscripts^{8, 11}, we infused 25 ng/kg/min AngII to rats fed 4% NaCl diet, but that dose of AngII was not sufficient to induce hypertension in rats fed 0.4% NaCl so the dose was increased to 50 ng/kg/min. A computer-driven servo-control system was used to inflate the aortic cuff to maintain left RPP at baseline levels while right RPP was uncontrolled. Following 7 days of Ang II infusion, kidneys were flushed by saline and removed. The upper pole of each kidney was removed for flow cytometry. A thin slice, 2 mm around the papilla, was cut and fixed in neutral-buffered formalin for histological analyses. The bottom pole was divided into the cortex and medulla and snap frozen and stored in −80°C for RNA analyses. Flow cytometry^{7, 22}, qPCR and immunohistochemistry were performed.

Histological analysis

More than 100 glomeruli in each slide were automatically and randomly selected throughout the entire renal cortex using a previously described program²³. Glomeruli were scored as 0-4. Percent protein cast, α-smooth muscle actin (αSMA) positive area was semi-quantified using Metamorph analysis as we have described previously²⁴. As measurement of micro vessel density is widely used for quantitative evaluation of the angiogenic or rarefaction response²⁵, glomerular and peritubular capillary densities were determined by image analysis of anti-endothelial cell antibody (RECA1) staining using commercially available NiE microscope with Elements software (Nikon). Glomerular capillary density was calculated by dividing the RECA1 positive staining area with the total glomerular area in at least 50 glomeruli in each slide. The density of peritubular capillaries in the renal cortex and the capillary bundles in the outer medulla were calculated the same way after masking glomerulus and vessel area in 20 random frames in each slide.

Statistical analysis

Rats were randomly allocated to each group. All rats that were completed experimental period was analyzed. Sample size was chosen based on previously published experiments performed under similar conditions^{7, 9, 11}. Analysis of tissue samples were performed blindly. Continuous values are presented as the means ± standard error of the means (SEM). Statistical comparisons were made using a t-test for two-group comparisons, and analysis of variance (ANOVA) followed by Holm-Sidak’s post-hoc test for multiple between-group comparisons. A p < 0.05 was considered significant.

Results

Chronic servo-control of renal perfusion pressure to the left kidney of conscious SD rats

RPP of the present study is shown in Figure 1. Carotid blood pressure, the surrogate of RPP in right kidney, rapidly increased from 117.6 ± 3.2 mmHg to 142.0 ± 3.2 mmHg during the first day following the start of AngII infusion (50 ng/kg/min) reaching a plateau elevation by day 4 of infusion and remained elevated at levels nearly 50 mmHg above control values throughout the 7 days of AngII infusion (Figure 1A). Femoral blood pressure, the surrogate of RPP in left kidney, was servo-controlled to remain within 5 mmHg of the levels of RPP measured during the three days preceding the infusion of AngII. RPP of vehicle treated rats, instrumented in the same manner but with no inflation of the balloon occluder, exhibited no changes in RPP throughout the experiment (Figure 1B) and no significant differences were observed between the left and right kidneys of vehicle treated rats. Of relevance, the RPP to the right and left kidneys in vehicle treated rats was not significantly different from the RPP of the servo-controlled kidneys of the AngII infused rats.

Figure 1. AngII infused servo-control model and immune cell infiltration.

(A) Blood pressure data of AngII infused group (n=9). (B) Blood pressure data of vehicle infused group (n=6, 1 carotid and 1 femoral catheters failed). ● indicates right RPP and ○ indicates left RPP. Data are presented as the means ± SEM of each day. *p < 0.05 versus baseline (Control 3), †p < 0.05 versus left RPP ; Holm-Sidak’s test.

(C) Flow cytometry data for T cell (CD3⁺), (D) B cell (CD45R⁺) and (E) Monocyte/Macrophage (CD11bc⁺). The data show individual values and means ± SEM. *p < 0.05 versus left servo-controlled kidney, †p < 0.05 versus vehicle infused group; Holm-Sidak’s test. AngII n=9, Vehicle n=6. L: Left kidney, R: Right kidney, RPP: Renal perfusion pressure

Servo-control of RPP to the left kidney prevented increased leukocyte infiltration observed in the right kidney that was exposed to elevated renal perfusion pressure during AngII induced hypertension.

The number of infiltrated leukocytes in each kidney was counted by flow cytometric analysis. CD3⁺ T cells and those subsets of helper (CD3⁺CD4⁺) T-cells, and cytotoxic (CD3⁺CD8⁺) T-cells were all significantly increased in the right kidney compared to the servo-controlled left kidney (Figure 1C). Similarly, both CD45R⁺ B cells and CD11bc⁺ monocytes or macrophages were also significantly higher in the right kidney (Figure 1D, E). No differences were observed in infiltrated leukocytes between the left and right kidneys of vehicle treated rats (Figure 1C, D, E). Importantly, the average numbers of leukocytes of the right and left kidneys in vehicle treated rats were similar to those observed in the left servo-controlled kidney of the AngII infused rats.

Increased renal perfusion pressure caused glomerular injury, but tubular damage was pressure independent.

Glomerular injury was quantified by scoring glomeruli on a scale of 0-4 (Figure S2). Glomerular injury of AngII infused rats was higher compared to either kidney of vehicle treated rats. The additional injurious effects of the elevated RPP to the glomeruli in the AngII infused rats was apparent. Glomerular scores were significantly higher in the AngII infused uncontrolled right kidneys than in the servo-controlled left kidneys (Figure 2A). In contrast, tubular casts did not differ significantly between the servo-controlled left and uncontrolled right kidney but were in both kidneys higher than observed in the vehicle treated rats (Figure 2B–D). Similar patterns were observed using another marker of kidney injury, Hepatitis A virus cellular receptor (Havcr)1 (also known as Kidney injury molecule1) as mRNA expression levels tended to increase in AngII infused rats compared to vehicle infused rats in both cortex and medulla, but no differences were detected between the right and servo-controlled left kidneys (Figure 2E, Figure S3A). Also assessed were renin mRNA expression levels which were suppressed in AngII infused uncontrolled right kidney exposed to the elevated perfusion pressure, but remained at levels similar to the vehicle infused rats in the servo-controlled left kidney (Figure 2F).

Figure 2. Histology.

(A) Glomerular score data. (B) Periodic acid–Schiff stained kidney. Scale bar 250 μm. Percent cast positive area (C) per cortex and (D) per outer medulla. mRNA expression level of (E) Havcr1 and (F) Ren in cortex. The data show individual values and means ± SEM. *p < 0.05 versus left servo-controlled kidney, †p < 0.05 versus vehicle infused group; Holm-Sidak’s test. AngII n=9, Vehicle n=6. L: Left kidney, R: Right kidney, RPP: Renal perfusion pressure

Elevated renal perfusion pressure decreased microvascular density.

Glomerular and peritubular capillary density was semi-quantified by the positive area staining of RECA1 to the rat endothelial cell antigen (mAb RECA-1)²⁶ (Figure 3A). The yellow arrows denote a glomerular capillary and the red arrows denote a peritubular capillary. RECA1 positive area per glomerulus was significantly reduced in high RPP right kidney of AngII infused rats compared to servo-controlled left kidney (26.0% ± 2.9% vs 19.5% ± 2.3%) (Figure 3B). Similarly, the density of peritubular capillaries in the renal cortex and the capillary bundles in the outer medulla (Figure 3C, D) were significantly reduced in the right kidney compared to the servo-controlled left kidney (Cortex: 4.1% ± 0.5% vs 3.0% ± 0.4%, Medulla: 1.3% ± 0.2% vs 0.8% ± 0.2%). AngII tended to increase peritubular vascular density even in the servo-controlled left kidney compared to vehicle treated left kidney (Cortex: 4.1% ± 0.5% vs 2.9% ± 0.4%, Medulla: 1.3% ± 0.2% vs 0.6% ± 0.1%) although these did not reach statistical significance. Cortical tissue mRNA expression of Cd31, also an endothelial cell marker tended to be higher in the servo-controlled left kidney compared to the right kidney of AngII infused rats but did not reach significance (Figure S4A). The left kidney of the servo-controlled AngII infused rats was moderately but significantly higher than the vehicle left kidneys. The von Willebrand factor (Vwf) mRNA, made in endothelial cells was also tended to increase, but was not significantly different in either kidney of the AngII infused rats or the vehicle infused rats (Figure S4B). Cd31 and Vwf did not show statistically significant difference in medullary tissue mRNA expression (Figure S3B, C). Interestingly, Prostaglandin-Endoperoxide Synthase 2 (Cox2) mRNA expression in cortex was significantly elevated in the left kidney of AngII infused rats compared to vehicle (Figure S4C). It was also significantly lower in the right kidney reaching levels similar to vehicle infused rats. On the other hand, Cox2 mRNA expression in medulla was elevated in the right kidney of AngII infused rats (Figure S3D)

Figure 3. Microvascular density.

(A) RECA1 staining. Yellow arrows denote positive area in glomerulus and red arrows denote positive area in peritubular capillaries. Scale bar 100 μm. (B) RECA1 positive area per glomerulus. (C) RECA1 positive area per cortex and (D) outer medulla. The data show individual values and means ± SEM. *p < 0.05 versus left servo-controlled kidney, †p < 0.05 versus vehicle infused group; Holm-Sidak’s test. AngII n=9, Vehicle n=6. L: Left kidney, R: Right kidney, RPP: Renal perfusion pressure

Regional kidney injury.

Anti-CD68 antibody (ED1)⁺ macrophages were found around damaged glomeruli and some arteries in the right kidney of AngII infused rats but extensive infiltration throughout the kidney was not observed (Figure 4A). The marker of tissue fibrosis α-smooth muscle actin (αSMA) stained more densely in the AngII infused rats in both kidneys compared to vehicle infused rats but no significant differences were observed between uncontrolled right kidney and servo-controlled left kidney. (Figure 4B–C). Cortical tissue mRNA expression levels of αSma was significantly higher in the AngII infused uncontrolled right kidneys compared to the servo-controlled left kidneys (Figure 4D). αSma did not show statistically significant difference in medullary tissue mRNA expression (Figure S3E).

Figure 4. Immunohistochemistry.

(A) ED1 staining. Arrows denote positive area. Scale bar 100 μm. (B) αSMA staining. Arrows denote positive area. Scale bar 100 μm. (C) αSMA positive area per cortex. (D) mRNA expression level of αSma. The data show individual values and means ± SEM. *p < 0.05 versus left servo-controlled kidney, †p < 0.05 versus vehicle infused group; Holm-Sidak’s test. AngII n=9, Vehicle n=6. L: Left kidney, R: Right kidney, RPP: Renal perfusion pressure

Discussion

The present study determined the extent to which leukocyte infiltration into the kidneys would be reduced by protecting the kidney from hypertension in the AngII induced model of hypertension in SD rats. This study was prompted by our observations in the Dahl S rat demonstrating that there was a significantly higher abundance of T cells, B cells, and monocytes and macrophages in the hypertensive uncontrolled right kidney compared with the servo-controlled left kidney⁷. Remarkably, the present study that examined hypertension produced by AngII in a rat of a very different genetic background produced very similar results to that of the Dahl S rat fed high salt (Figure 5A). The robust level of hypertension (> 165 mmHg) induced by 7 days of AngII infusion resulted in significantly greater numbers of total infiltrated leukocytes, including T cells, B cells, and macrophages/monocytes, in the right kidney, which had elevated RPP, compared to the servo-controlled left kidney. This was not a consequence of a high salt diet which is often given with AngII to enhance the hypertension since the rats were maintained on a low-normal (0.4% NaCl) diet throughout the study. The data also demonstrates that the rise of arterial pressure must therefore precede the infiltration of lymphocytes and renal injury and that immune cell infiltration is not determined by elevations of AngII directly.

Figure 5. Comparison with the other model.

(A) Comparison of the ratios of T and B cell infiltration in the servo-controlled left kidneys (LK) to uncontrolled right kidneys (RK) of salt induced hypertension in Dahl Salt Sensitive (SS) rat (black) and AngII induced hypertension in SD rats (white). Nearly identical results were found in the two models. Comparison of (B) glomerular score and (C) cortical and (D) medullary protein cast salt induced hypertension in Dahl Salt Sensitive rats (SS + High Salt) and AngII induced hypertension in SD rats (SD + AngII). The data show individual values and means ± SEM. *p < 0.05 versus left servo-controlled kidney, †p < 0.05 versus SS + High Salt; Holm-Sidak’s test. SS+high salt n=6, SD+AngII n=9. L: Left kidney, R: Right kidney, RPP: Renal perfusion pressure

Data of salt induced hypertension in SS rat is adapted from our previous report⁷.

Varying patterns of renal injury related to specific models of hypertension.

Although a similar pattern of RPP dependent leukocyte and monocyte infiltration was observed comparing the Dahl S and present AngII model of hypertension, differences were observed in the nature of the associated renal injury. It is interesting that despite the remarkable similarities of the ratios for infiltrating immune cells between right and left kidney in both the Dahl S and AngII models, notable differences in the severity of renal injury were apparent between the two models. In both models, the glomerular injury score was higher in the uncontrolled right kidneys exposed to the elevated RPP although the overall glomerular damage was higher in Dahl S compared with AngII model (Figure 5B) despite similar levels of hypertension. Both AngII-infused and Dahl S rats exhibited greater levels of intersititial fibrosis in the uncontrolled right kidneys as indicated by αSMA staining but AngII-infused rats exhibited far less medullary tubular protein casts than the Dahl S rats (Figure 5C, D). It is interesting to speculate on the possible reasons for these differences. They could of course be ascribed to salt diet since the AngII infused rats of the present study were maintained on a low-normal salt diet. They could in part be attributed to strain differences of the rats. More likely, the differences were a consequence of the intrarenal vasoconstrictor actions of AngII. Polichnowski et al. reported that the renal injury with AngII chronic infusion is relatively mild despite the high RPP, and they suggested that this is related the vasoconstriction of preglomerular vasculature²⁷. Whatever accounts for these differences, it is apparent that the effects of RPP upon renal injury can differ substantially between different forms of hypertension. Such differences of individuals to antihypertensive therapies and various models of hypertension has suggested by Bidani and Griffin^28,29, and our own studies^10–12.

Renin-angiotensin responses.

Renin mRNA expression was significantly reduced in the uncontrolled right kidneys as would be generally expected in face of high circulating levels of infused AngII and elevated RPP. However, suppression of renin was observed only in the kidney exposed to high RPP and not in the servo-controlled left kidney. If signaling for reduction of renin expression by high levels of circulating AngII was occuring via negative feedback this should have occurred in both kidneys^{30, 31} Such feedback may have been overridden by a reduction of juxtaglomerular wall tension of the preglomerular vasculature of the servo-controlled kidney²⁷. However, recent studies in mice with conditional deletion of AngII-AT1 receptors on renin-producing cells found no evidence for existence of a direct negative feedback of AngII on the AT1 receptors³². This would indicate that suppression of renin in the right kidney would most likely be attributed to the elevated RPP. However, even in the absence of direct feedback AngII suppression in the servo-protected left kidney, high intrarenal AngII levels would be present in both kidneys due to the high levels of circulating AngII and thereby elevating vascular wall tone and perhaps reduced renal blood flow despite a constancy of RPP. This may explain why similar levels of tubular injury were observed in both right and left kidneys in AngII infused rats even though the RPP of these kidneys were different. The intrarenal actions of AngII such as preglomerular arteriolar vasoconstriction could also explain why differences in markers of renal injury were substantially less than was observed in the Dahl S servo-controlled model of hypertension.

Interestingly, an opposite expression pattern of Cox2 was observed in our study between the renal cortex and medulla. The mechanism for this is not apparent. Gonzalez et al. reported that Cox2 mRNA expression was elevated only in the renal medulla of SD rats infused for 7 days with AngII compared to control but not in the cortex ³³. These differences and those observed in the present study between cortical and medullary expression of Cox2 may be a consequence of interactions between renin and Cox2. There is evidence that Cox2 expression is increased in high renin states with salt restriction, angiotensin converting enzyme inhibition and renovascular hypertension³⁴ .

Glomerular and peritubular capillary density.

The loss of renal capillary density is considered one of the most important mechanisms in progression of kidney disease³⁵. Both glomerular and peritubular capillary density were significantly reduced in the right kidneys exposed to the elevated RPP as revealed by RECA1 endothelial cell staining. The glomerular injury is consistent with the elevated glomerular injury score of non-servo-controlled right kidneys and the concept of “barotrauma” on glomerular injury and hyperfiltration³⁶. Less readily explained by physical factors is the microvessel rarefaction observed in the cortical peritubular capillaries. Both red cell flux and peritubular capillary pressures of the superficial cortical microvessels have been found to be well autoregulated in the classic sense that they remain unchanged in response to acute step changes of RPP³⁷. And yet, we and others have observed renal microvessel rarefaction in both the Dahl S and the AngII high salt models of hypertension in the cortex and especially prominent in the renal outer medulla^{38, 39}. The mechanisms responsible for rarefaction of the cortical microvessels is unclear. It is possible that since AngII has been shown to reset autoregulation of renal blood flow to a lower level⁴⁰, the post-glomerular microvessels may be subjected to higher levels of flow and pressure⁴⁰. The more prominent capillary rarefaction in the renal outer medulla could be attributed to the well-recognized lack of blood flow autoregulation of blood flow to the outer medulla^41–43.

Potential role of microvascular injury in leukocyte infiltration.

The mechanism(s) whereby elevations of RPP lead to increased infiltration of lymphocytes is poorly understood. However, it is relevant that in two very different models of hypertension, the Dahl S plus high salt diet (a low renin model) and the SD high AngII low salt model, the protection of the left kidney from the elevated RPP reduced the infiltration of all types of T-cells, B-cells and macrophages. This lack of selectivity clearly indicates that physical factors are responsible for enabling the infiltration of these cells, albeit to a different extent in two models. In a recent RNAseq analysis we have reported that increases of RPP result in the overexpression of the leukocyte transendothelial migration pathway in Dahl S rats fed a high salt diet⁴⁴ indicative of endothelial damage in the kidney under high RPP.

Earlier studies have observed that peritubular microvascular rarefaction was present in both Dahl S rats fed a high salt diet, AngII infused SD rats, and in DOCA-salt hypertension^{38, 39, 45}. The extent to which this was related to RPP could not be ascertained in those studies. Iwazu et al.⁴⁵ found apoptosis and inhibition of proliferation of peritubular endothelial cells in the rat DOCA-salt model which was inhibited by spironolactone treatment indicating that rarefaction of microvessels in the kidneys was mediated via mineral corticoid receptor activation. However, the spironolactone also reduced hypertension in those studies which may have accounted for much of the protection from rarefaction.

In the present study, microvessel rarefaction and leukocyte infiltration were found only in the hypertensive right kidneys of AngII infused rats. The low RPP might directly protect from the mechanical pressures in the capillaries which lacked the autoregulation because of the high circulating AngII. Another possibility is that the higher expressed levels of renin in the servo-controlled kidney may have attenuated the rarefaction since high levels of both renin⁴⁶ and COX2⁴⁷ have been found to stimulate angiogenesis. The close association of elevated RPP with microvessel rarefaction in our study indicates that mechanical factors were responsible for both the capillary loss and leukocyte infiltration although the molecular signaling mechanisms remain to be determined. It has been found that increased vascular wall tension can stimulate endothelial ROS production which is known to induce the endothelial dysfunction^48,49. It has also been found that elevations of transmural vascular pressure can detach pericytes⁵⁰, and cause capillary loss^{51, 52} and that endothelial damage induces an increase in vascular permeability⁵³. Endothelial permeability is enhanced by oxidative injury, which increases entry of lipoproteins to the subendothelial space, where they are oxidized and enhance inflammation⁵⁴. The results of the present study are consistent with the view that increased capillary permeability plays an important role in recruiting circulating leukocytes to the tissue⁵⁵.

Perspectives.

In the study, we have used a chronic servo-control technique to identify the role of RPP in renal immune cell infiltration in AngII infused SD rats. The result was consistent in both Dahl S high-salt induced hypertension and SD AngII-induced hypertension, which indicates that the importance of RPP for recruitment of immune cells into the kidneys. We also suggested the relationship between capillary rarefaction and immune cell infiltration. Further study is necessary to elucidate the mechanism of the development of capillary rarefaction and vascular permeability.

As both Dahl S high-salt induced hypertension and Sprague Dawley AngII-induced hypertension are models exhibiting reduced kidney autoregulation, it would be interesting to carry out similar studies in a rat model with a greater autoregulatory capacity such as the Spontaneously Hypertensive Rat.

Novelty and Significance.

1). What Is New

• The quantitative importance of renal perfusion pressure (RPP) related to the infiltration of leukocytes and related kidney injury in an angiotensin II induced model of hypertension was assessed by a custom-designed servo-control technique.

• In this model of hypertension, the elevation of RPP, not AngII nor dietary salt, was responsible for leukocyte infiltration in the kidney.

• The study shows that elevation of RPP must precede Immune cell infiltration.

• Capillary density of the glomeruli, renal cortex, and the outer medulla was found to be highly dependent upon increased RPP.

2). What Is Relevant?

• We previously reported that high RPP in Dahl S rats when fed a high salt diet is responsible for immune cell infiltration into the kidneys. In the present study we confirmed the importance of high RPP for the recruitment of immune cells into the kidneys in very different model of hypertension (e.g., AngII with a low salt diet) and in a different strain of rat (e.g., SD).

• Capillary rarefaction has been reported to occur in several hypertensive models and is considered to be an important factor for leukocyte infiltration. The present results find that the rarefaction is highly dependent on RPP. Together, the findings emphasize the great importance of protecting of the kidneys from chronic elevations of arterial pressure and elucidate the consequences of failure to do so.

3). Summary

The present study finds that elevation of RPP determines leukocyte infiltration in AngII induced hypertension in SD rats in the presence of a low-normal level of salt intake. This infiltration parallels microvascular rarefaction of both renal glomerular and peritubular capillaries which we propose is importantly involved in the process of infiltration of leukocytes. We conclude that renal leukocyte infiltration and microvascular rarefaction is importantly determined by RPP and not directly by the effects of AngII nor dietary salt. Comparisons with the Dahl salt-sensitive rat model fed a high salt diet indicate that the impact of these physical forces can be quantitatively modified under varying conditions that determine intrarenal distributions of pressures and flows.