Loss of Prolyl Carboxypeptidase in Two-Kidney, One-Clip Goldblatt Hypertensive Mice

It is well documented that angiotensin (Ang) II contributes to kidney disease progression. The protease prolyl carboxypeptidase (PRCP) is highly expressed in the kidney and may be renoprotective by degrading Ang II to Ang-(1-7). The aim of the study was to investigate whether renal PRCP protein expression and activity are altered in two-kidney, one-clip (2K1C) Goldblatt hypertensive mice. Left renal artery was constricted by using 0.12 mm silver clips. Blood pressure was measured using telemetry over the eleven weeks of study period and revealed an immediate increase in 2K1C animals during the first week of clip placement which was followed by a gradual decrease to baseline blood pressure. Similarly, urinary albumin excretion was significantly increased one week after 2K1C and returned to baseline levels during the following weeks. At 2 weeks and at the end of the study, renal pathologies were exacerbated in the 2K1C model as revealed by a significant increase in mesangial expansion and renal fibrosis. Renal PRCP expression and activity were significantly reduced in clipped kidneys. Immunofluorescence revealed the loss of renal tubular PRCP but not glomerular PRCP. In contrast, expression of prolyl endopeptidase, another enzyme capable of converting Ang II into Ang-(1-7), was not affected, while angiotensin converting enzyme was elevated in unclipped kidneys and renin was increased in clipped kidneys. Results suggest that PRCP is suppressed in 2K1C and that this downregulation may attenuate renoprotective effects via impaired Ang II degradation by PRCP.


Introduction
There are 26 million adults with chronic kidney disease (CKD) in the United States and the number of those affected continues to increase [1]. The activation of the renin angiotensin system (RAS) and the elevated formation of the vasoconstrictor angiotensin (Ang) II both contribute to renal pathophysiology by stimulating pathways that lead to aldosterone release, water retention, vasoconstriction, fibrogenesis, and inflammation [2][3][4]. There is emerging evidence that Ang II and Ang-(1-7) have counter-regulatory roles. While Ang II functions as a potent vasoconstrictor and is implicated in the pathophysiology of various kidney diseases, Ang-(1-7) protects against renal damage and cardiovascular disease (5)(6)(7)(8). Indeed, treatment Blood pressure, heart rate and locomotor activity measurements using radiotelemetric probes Mice were anesthetized with 2% isoflurane. An incision was made along the neck. Using blunt dissection the left common carotid artery was located followed by clamping and cannulation. A sterile radiotelemetric catheter, model TA11PA-C10 (Data Sciences International, St Paul, MN, USA), was inserted into the carotid artery and tied in place. The body of the transmitter was inserted into a pouch below the skin on the right flank. The skin was closed using 5.0 sterile sutures (Arosurgical, Newport Beach, CA, USA). Carprofen (5 mg/kg; Sigma-Aldrich, St. Louis, MO, USA) was given SC immediately and 24 hour later for post-operative analgesia. Animals were allowed to recover for one week before blood pressure, heart rate and locomotor activity were recorded.

Goldblatt Two-Kidney, One-Clip (2K1C) Surgery
Mice were anesthetized with 2% isoflurane and the kidney was visualized through a flank incision. A u-shaped silver clip (Exidel SA, Delemont, Switzerland) was inserted onto the left renal artery leaving a 0.12 mm gap for ischemic blood flow. Carprofen (5 mg/kg, SC) was given immediately and 24 hour later for post-operative analgesia.

Renal PRCP and renin enzyme activity
Ang-(1-7) formation from Ang II was assessed using matrix assisted laser desorption ionization (MALDI) time-of-flight (TOF) based enzyme assays as previously described with some modifications [37,38]. For the in situ PRCP enzyme assays, consecutive tissue sections (12 μm) were prepared from fresh frozen kidneys and incubated with 1 mM Ang II at 37°C for 5 min. For the in vitro PRCP enzyme assays, kidney homogenate from whole kidney (20 μg protein) was incubated at 37°C in 50 mM sodium citrate buffer pH 5 containing 5 mM ZnCl 2 , 150 mM NaCl 2 , 10 μM lisinopril and 0.1 mM Ang II. The enzyme reaction was terminated by addition of 1% trifluoroacetic acid (TFA). Renin activity was measured as described before with some modifications [59]. Kidney homogenate from whole kidney (20 μg protein) was incubated at 37°C in 50 mM MES buffer pH 6.75 containing 1 mM EDTA, 100 μM peptidase inhibitors (thiorphan, 4-aminophosphonobutyric acid, Cpp-AAF-pAb) and 40 μM tetradecapeptide. Internal Ang peptide standards (New England Peptide, Gardner, MA or Anaspec, San Jose, CA, USA) were added, the mixture was diluted with 90% acetonitrile containing 0.3% TFA and spotted onto an MTP 384 ground steel target plate (Bruker Daltonics, Billerica, MA, USA). Mass spectra were obtained using an Autoflex III smartbeam MALDI TOF/TOF instrument (Bruker Daltonics, Billerica, MA, USA). The spectral analysis was performed with proprietary Bruker Flex Analysis and Imaging software.

Plasma collection
Mice were sacrificed by decapitation using a rodent guillotine. Trunk blood was collected in ice-chilled heparinized tubes. Blood was centrifuged at 10,000 x g for 10 minutes at 4°C. Plasma was separated, aliquoted, and stored at -80°C.

Urine collection and analysis of creatinine and albumin
Mice were placed individually in metabolic cages for 24h-urine collection with free access to food and water. Protease inhibitors (Roche Diagnostics, Indianapolis, IN, USA) were added to prevent protein degradation. Twenty four-hour urine samples were centrifuged at 3,000 x g for 5 min at 4°C to remove debris. Urinary creatinine was measured using a kit purchased from Quidel (San Diego, CA, USA). Urinary albumin excretion was determined using a Mouse Albumin ELISA Quantitation Set (Bethyl Laboratories, Montgomery, TX, USA). The optical density was measured at a wavelength of 450 nm using a Fusion Packard plate reader (Packard BioScience, Meriden, CT, USA).

Data Analysis
The relative mesangial matrix area (RMMA), relative glomerular fibrosis area (RGFA) and tubular or glomerular PRCP staining were quantified using MetaMorph image analysis software (Molecular Devices, Inc., Sunnyvale, CA, USA). All data were expressed as mean ± SEM. Unpaired student's t-test was used to evaluate the differences between two groups. One-way ANOVA and post hoc Bonferroni's Multiple Comparison Test compared the difference between more than two groups. A value of p<0.05 was considered statistically significant.

Metabolic and cardiovascular parameters
After eleven weeks of 2K1C, the kidneys were removed and weighed. Clipped kidneys were significantly smaller and atrophic compared to the contralateral unclipped kidneys and kidneys from control mice (Fig. 1A). As shown in Fig. 1B, a significant rise in blood pressure was observed at one and two weeks after clip placement compared with baseline. The blood pressure returned to baseline levels from weeks three through eight after clip placement, which was followed by a significant increase in blood pressure from baseline at weeks nine, ten, and eleven. There was a significant circadian change of blood pressure at baseline and 11 weeks after 2K1C while at 2 weeks of 2K1C the blood pressure in the 12 h light period was not different from the 12 h dark period (Fig. 1C). A significant increase in heart rate was detected after one week of clip placement compared to baseline (Fig. 1D). Similar to the circadian changes of blood pressure, there was a significant difference of heart rate in the 12 h light period compared to the 12 h dark period at baseline and 11 weeks after 2K1C (Fig. 1E). However, at 2 weeks after 2K1C, the heart rate in the 12 h light period was not different from the 12 h dark period (Fig. 1E). Locomotor activity was not affected by clip placement (Fig. 1F). Circadian changes of activity indicate a significant higher locomotor activity in the 12 h dark period compared to the 12 h light period (Fig. 1G). There was a trend of increased urinary output after clip placement (Fig. 1H) and daily urinary albumin excretion was significantly increased one week after clip placement and returned back to baseline levels during the course of the study (Fig. 1I). The 2-week and 11-week time points were selected for further studies.

Renal Morphology
Mesangial expansion was significantly increased in both 2K1C kidneys at 2 weeks and 11 weeks of 2K1C compared to the control kidneys; however, the unclipped kidneys had a significantly greater mesangial expansion compared to the clipped kidneys at both 2 and 11 weeks post-clipping ( Fig. 2A). The clipped and contralateral unclipped 2K1C kidneys also exhibited significantly greater glomerular fibrosis at 2 weeks and 11 weeks of 2K1C compared to the control kidneys. The clipped kidneys showed greater glomerular fibrosis than the unclipped kidneys (Fig. 2B).

Protein expression of renal PRCP, PREP, and ACE
Renal PRCP protein expression was significantly reduced in the clipped kidneys but not in the contralateral unclipped kidneys compared to the control kidneys at 2 and 11 weeks after 2K1C (Fig. 3A,B). Expression of renal PREP, an alternative peptidase capable of producing Ang-(1-7), was not changed in both clipped and contralateral unclipped 2K1C kidneys compared to the control (Fig. 3C). Renal ACE expression was significantly increased in unclipped 2K1C kidneys and decreased in clipped 2K1C kidneys compared to controls (Fig. 3D,E).

Renal renin activity
Renin activity was determined using a MALDI-TOF-based enzyme assays in which kidney homogenates were incubated with the renin substrate tetradecapeptide. Formed Ang I (m/z 1296) was absolutely quantified using a stable-isotope labeled peptide standard. Fig. 3F shows that renin activity was significantly increased in clipped 2K1C kidneys compared to control and unclipped kidneys.

Renal, urinary and plasma PRCP activity
Renal enzymatic activity of PRCP was determined using MALDI-TOF-based enzyme assays that allow for the incubation of kidney sections or homogenates with the natural substrate Ang II. PRCP enzymatic activity in tissue sections was determined in situ by detecting the formed product Ang-(1-7) (Fig. 4A). Renal PRCP activity was detected predominantly in the cortex of the unclipped, clipped, and control kidneys and was significantly reduced in the clipped kidneys compared to the control and unclipped kidneys at 2 weeks of 2K1C (Fig. 4A). The in vitro enzyme assay confirmed the results obtained with the in situ enzyme assay showing a significantly reduced PRCP activity in the clipped kidneys compared to contralateral unclipped and control kidneys after 11 weeks of 2K1C (Fig. 4B). There was no difference in PRCP activity between the contralateral unclipped and control groups (Fig. 4A,B). Plasma and urinary PRCP activities after clip placement were not different from baseline levels (Fig. 4C,D) although there was a trend toward reduced urinary PRCP levels one week after clip placement and a sustained recovery of PRCP levels during the following ten weeks.

Immunolocalization of renal PRCP
Immunofluorescence of PRCP revealed expression of the enzyme in the apical membrane of the renal proximal tubules (Fig. 5A). Furthermore, PRCP co-localized with nephrin in the glomeruli of all groups (Fig. 5A-C). Tubular PRCP expression was reduced in clipped kidneys while glomerular PRCP and nephrin expression were not affected by placement of the renal artery clip as compared with the control (Fig. 5A-C). Overall, glomerular expression of PRCP in the clipped kidneys was unchanged compared to control kidneys and unclipped kidneys while total cortical expression and tubule staining for PRCP was decreased (Fig. 5D).

Discussion
In this study, the induction of renovascular hypertension caused a significant decrease in clipped kidney weight and an immediate hypertensive response at week one after 2K1C. After one week of clip placement, 24 h blood pressure gradually returned to baseline, showed normal levels at week four post-clipping and increased again toward the end of the study at week 11 after 2K1C. The 24 h heart rate also increased one week after 2K1C and returned to baseline levels during the following ten weeks of 2K1C. Circadian changes were observed at baseline and at 11 weeks after 2K1C. However, at two weeks after 2K1C the 12 h light period did not significantly dip compared with the 12 h dark period although the animals maintained circadian locomotor activity. This finding suggests that, in the early phase, renovascular hypertension disrupts the circadian blood pressure and heart rate rhythm causing a non-dipping effect of the light period.
Previous studies described a gradual increase and full development of hypertension in mice over two weeks and sustained hypertension until week four post-clipping [60][61][62][63]. Gradual and sustained hypertension was also observed in rats after clip placement [64,65]. The present study is one of the few studies that monitored the progression of renal artery constriction in mice for longer than four weeks. Consistent with our results, two other studies in mice describe that blood pressure peaked at one week after 2K1C and tended to decline in the following four weeks after 2K1C surgery [66,67] suggesting that progressive stenosis, severe atrophy and complete renal ischemia cause resolution of the hypertension. Similarly, one study in rats demonstrated an immediate increase of blood pressure in the first week of 2K1C that was followed by a decline over the following 4 weeks although levels were still significantly different from baseline at the end of the study [68]. The discrepancies between these studies could be explained by different degrees of renal blood flow restriction, use of other methods to measure blood pressure, the experimental approach, use of different clips and sizes, or the genetic background of the mice.
The 2K1C treatment caused renal injury in both the clipped and the contralateral unclipped kidneys. This was seen with the increased urinary albumin excretion after clip placement although albuminuria normalized within one week of 2K1C following the same trend as observed for blood pressure and heart rate measurements. Indeed, recent studies have shown that many patients with renal impairment are normoalbuminuric, therefore urging a search for underlying pathologies and new markers of CKD [69][70][71][72]. The initial increase of blood pressure and urinary albumin excretion in our model may be directly dependent on hyperreninemia and elevated Ang II production especially in the early phase when renin secretion is activated [73] while in the chronic phase, after nine weeks of clip placement, other factors might be responsible for causing the hypertension.
Renal injury was also demonstrated by mesangial expansion and glomerular fibrosis in both kidneys of the clipped animals; however mesangial expansion was exacerbated in the contralateral unclipped group. These renal pathologies are similar to those found in humans with renal artery stenosis where the stenotic kidney showed ischemic changes while the contralateral kidney showed signs of glomerulosclerosis [58]. A separate clinical study involving patients with atherosclerotic renal artery stenosis also demonstrated advanced glomerulosclerosis and similar renal morphology in the affected kidneys [74].
The progression of renovascular hypertension was associated with changes of renal PRCP expression and activity as analyzed using Western blot and mass spectrometry. Renal PRCP expression levels were significantly lower in the clipped kidney compared to both the unclipped and control kidneys. Similar to PRCP, renal ACE2 expression and Ang-(1-7) levels were decreased in 2K1C rats while Ang II was increased suggesting that a differential regulation of the RAS may contribute to hypertension in this model [64]. In addition, ACE was increased in the unclipped kidney while renin was elevated in the clipped kidney confirming previous findings of the RAS in animals with renovascular hypertension [75]. However, protein expression of renal PREP, an alternative Ang-(1-7) forming peptidase, was not different between all groups. These findings suggest that ACE, renin, PRCP and ACE2 participate in renal changes in the 2K1C model and indicate that PREP protein expression is not affected by renal pathologies induced in this model. Indeed, a previous study conducted in our laboratory using genetic deletion mouse models confirmed that the contribution of PREP to renal Ang II processing is negligible compared to that of PRCP and ACE2 [37].
Enzymatic activity studies using established MALDI-TOF mass spectrometric methods [37,38] support the Western blot data. PRCP enzymatic activity was significantly decreased in the clipped kidney compared to the control and unclipped kidney. Although a trend for altered PRCP levels in whole urine was observed, the changes were not significant most likely due to a predominant localization of PRCP in urinary exosomes [76]. Plasma PRCP activity was not changed. Accumulation of Ang II in the 2K1C model could be responsible for decreased renal PRCP activity as Ang II has been shown to inhibit PRCP activity at higher concentrations [77].
To understand and characterize the role of PRCP in the renal cortex during injurious conditions protein expression patterns in the kidney were analyzed using immunofluorescence. PRCP was localized in the renal cortex in the apical membrane of tubules and also within the glomerulus. PRCP expression was significantly reduced in the tubules in the clipped group, perhaps due to the severe tubulointerstitial atrophy present. Severe tubulointerstitial atrophy was also seen in the nephrectomized, affected kidney in patients with atherosclerotic renal artery stenosis [74]. PRCP co-localized with nephrin in all groups, suggesting its expression within podocytes. Expression of nephrin was not affected by the 2K1C approach.
Localization of PRCP to renal tubules and glomeruli fits well with previous studies identifying the renal cortex as the major region in the kidney that generates Ang-(1-7) from Ang II and expresses the Ang-(1-7) binding site [38,78]. We propose that renal Ang II can be degraded to Ang-(1-7) by PRCP in the renal cortex thereby preventing activation of cell signaling through Ang II type I receptors. Therefore, loss of renoprotective PRCP in renovascular hypertension may contribute to progression of renal injury and exacerbate damage in the affected kidney. As one of the primary functions of PRCP in the vasculature is the activation of prekallikrein leading to activated kallikrein and bradykinin release [50,79,80], loss of PRCP in 2K1C could also cause decreased synthesis and secretion of renal kallikrein and thus diminished kinin release. Indeed, the kallikrein-kinin system is reduced in patients and animals with renovascular hypertension [81][82][83]. Changes in kallikrein and kinin levels could be an alternative explanation for the observed progression of hypertension and renal damage as kinins exhibit vasodilatory effects and may afford renoprotection through AT2 and B2 receptors. Therefore, PRCP may functionally connect the kallikrein-kinin system with the RAS and the combined differential regulation of both systems could contribute to the pathogenesis of renovascular hypertension.