|Home | About | Journals | Submit | Contact Us | Français|
Renal artery stenosis (RAS) causes renovascular hypertension and renal damage, which may result from tissue inflammation. We have previously shown that the kidney in RAS exhibits increased expression of monocyte chemoattractant protein (MCP)-1, but its contribution to renal injury remained unknown. This study tested the hypothesis that MCP-1 contributes to renal injury and dysfunction in the stenotic kidney.
Kidney hemodynamics, function, and endothelial function were quantified in pigs after 10 weeks of experimental RAS (n=7), RAS supplemented with the MCP-1 inhibitor bindarit (RAS+bindarit, 50mg/kg/day PO, n=6), and normal controls (n=8). Renal inflammation was assessed by the immunoreactivity of MCP-1, its receptor CCR2, and NFkB, and oxidative-stress by NADPH-oxidase expression and in-situ superoxide production. Renal microvascular density was evaluated by micro-CT, and fibrosis by trichrome staining, collagen-I immunostaining, and hydroxyproline content.
After 10 weeks of RAS, blood pressure was similarly elevated in RAS and RAS+bindarit. Compared with normal, stenotic RAS kidneys had decreased renal blood flow (5.4±1.6 vs. 11.4±1.0 mL/min/kg, p<0.05) and glomerular filtration rate, and impaired endothelial function, which were significantly improved in bindarit-treated RAS pigs (to 8.4±0.8 mL/min/kg, p<0.05 vs. RAS). Furthermore, bindarit markedly decreased tubulointerstitial (but not vascular) oxidative-stress, inflammation, and fibrosis, and slightly increased renal microvascular density. The impaired renovascular endothelial function, increased oxidative-stress, and fibrosis in the contralateral kidney were also improved by bindarit.
MCP-1 contributes to functional and structural impairment in the kidney in RAS, mainly in the tubulointerstitial compartment. Its inhibition confers renoprotective effects by blunting renal inflammation and thereby preserving the kidney in chronic RAS.
Renal artery stenosis (RAS) may cause renovascular hypertension (RVH) and ischemic nephropathy, and may lead to end-stage renal disease, which is a major health problem in Western Society. The mechanisms by which RAS causes renal fibrosis and deterioration of renal function have not been fully elucidated, but inflammation is likely an important component in this process.
One of the hallmarks of RAS is activation of the renin-angiotensin system. Angiotensin II can then increase macrophage infiltration, possibly by upregulation of monocyte chemoattractant protein-1 (MCP-1), a chemokine that increases monocyte infiltration into inflamed tissues and an important inflammatory mediator. MCP-1 has been implicated in a variety of renal diseases[3, 4] and progressive renal damage, mediated through cognate receptors like CCR2 . Activated macrophages and fibroblasts in the RAS kidney may directly induce NAD(P)H oxidase activity, stimulating TGF-β1 production and triggering fibroblast proliferation and differentiation into collagen-secreting myofibroblasts. Indeed, we have previously shown that swine RAS kidneys exhibit increased levels of inflammation mediators[7, 8]. Furthermore, we demonstrated that RAS may lead to microvascular remodeling and rarefaction in the stenotic kidney, probably consequent to ischemia, increased oxidative stress, and inflammation[7, 9]. However, the role of MCP-1 in renal dysfunction and remodeling in RAS has not been fully resolved. Its central role in the pathogenesis of renal diseases lends itself to the opportunity to block MCP-1/CCR2 in order to prevent macrophage-induced tissue damage. Supporting this notion, neutralization of MCP-1 reduces macrophage infiltration and progressive kidney damage in rat tubulointerstitial nephritis.
The novel drug 2-Methyl-2-[[1-(phenylmethyl)-1H-indazol-3yl]methoxy]propanoic acid (bindarit) selectively inhibits MCP-1 production in vitro and in vivo[12, 13] in mice by reducing its mRNA transcript levels[14–16], and without affecting production of other chemokines, such as IL-8, MIP-1α, or RANTES[11, 17]. Its anti-inflammatory activity is also achieved without immunosuppression or interference with arachidonic acid metabolism [13, 18]. Furthermore, unlike small peptides and neutralizing antibodies, bindarit can be administered orally in large animals and in humans, to investigate the involvement of MCP-1 in chronic renovascular disease.
Therefore, the objective of this study was to test the hypothesis that MCP-1 plays an important role in mediating renal injury in RAS. A corollary of this hypothesis is that inhibition of the MCP-1 pathway could potentially improve renal function, attenuate microvascular injury, and reduce fibrosis in the pig kidney. Furthermore, we also evaluated potential effects of bindarit on the contralateral kidney, since in rat RVH MCP-1 was also upregulated in the non-stenotic kidney.
All procedures were approved by the Institutional Animal Care and Use Committee of the Mayo Clinic.
Twenty one domestic pigs (45–55kg) were randomized into three groups and studied after 10 weeks of experimental RAS (n=7), RAS supplemented with the MCP-1 inhibitor bindarit (RAS+bindarit, 50mg/kg/day PO for 10 weeks, starting 1 day after induction of RAS, n=6), and age- and body-weight matched normal controls (n=8). At baseline, animals were anesthetized with 0.5 g of intra-muscular ketamine and xylazine, and anesthesia then maintained with intravenous ketamine (0.2 mg/kg/min) and xylazine (0.03 mg/kg/min). RAS was then induced by placing a local-irritant coil in the main renal artery to induce gradual development of unilateral RAS, as previously described. A PhysioTel® telemetry system (Data Sciences) was also implanted in the left femoral artery to measure mean arterial pressure (MAP) [9, 20]. The average MAP over the last 2 weeks before the in vivo study was calculated. Bindarit treatment was then initiated, at a dose based on a previous study, in which bindarit was shown to inhibit renal MCP-1 synthesis and inflammation.
After 10 weeks of observation, the pigs were again similarly anesthetized, and renal hemodynamics and function in each kidney assessed by multi-detector computed tomography (MDCT). The degree of stenosis was determined by angiography. Venous blood samples were collected for creatinine and plasma renin activity measurements (GammaCoat™ PRA kit, DiaSorin Inc., MN).
Pigs were then euthanized with a lethal intravenous dose of 100mg/kg of sodium pentobarbital (Sleepaway®, Fort Dodge Laboratories, Inc, Fort Dodge, IA). Both the stenotic and contralateral kidneys were removed, and prepared in-vitro studies. Inflammation was assessed in the stenotic kidney by the expression of MCP-1 and its receptor CCR2, macrophage infiltration, nuclear-factor kappa B (NFkB), the chemokines RANTES and interleukin (IL)-8, and the cytokine interferon (IFN)-γ. Renal oxidative-stress was evaluated by the expression of the NADPH-oxidase (p47 and p67phox subunits), and in-situ production of superoxide anion detected by fluorescence microscopy using dihydroethidium (DHE), as previously described. Renal fibrosis was evaluated by trichrome staining, transforming-growth-factor (TGF)-beta, collagen I immunostaining, and the hydroxyproline assay. Renal microvascular remodeling was evaluated by vessel density using micro-CT, vascular media/lumen ratio, and expression of vascular-endothelial-growth-factor (VEGF). Several of these pathways were also assessed in the contralateral kidney.
MDCT flow studies were performed, as previously detailed, for assessment of basal renal blood flow (RBF), glomerular filtration rate (GFR), and renal vascular resistance (RVR), and intra-aortic infusion of acetylcholine (Ach, 5 μg/kg/min) to test endothelium-dependent response in each stenotic and contralateral kidney. Ach was administered for 15 minutes through a catheter placed in the aorta superior to both renal arteries, and renal scanning performed towards the end of this infusion, which is a period of hemodynamic stability and no change in arterial pressure.
An intravascular contrast agent (Microfil MV122, Flow Tech, Inc., Carver, MA) was perfused into the stenotic kidney , and samples then scanned and images analyzed as previously described. The spatial density and average diameter of microvessels (diameters 20–500μm) in the inner, middle, and outer thirds of the renal cortex were calculated [7, 9].
standard immunostaining was performed in 5μm paraffin-embedded mid-hilar renal cross-sections with antibodies against MCP-1 (1:50, Biovision, CA), collagen I (1:100, Cosmo Bio, Tokyo), and macrophages (mouse anti-macrophage CD163, 1:50, AbD Serotec, Oxford, UK). To explore the spatial relationship between infiltrating macrophages and production of oxidative stress, double immunostaining of macrophages and nitrotyrosine was performed, using SIGMAFAST™ Fast Red TR/Naphthol AS-MX tablets (red) and DAB (brown), respectively. Slides were examined using MetaMorph® (Meta Imaging Series 6.4), as we described previously. Macrophages were manually counted under 60× fields, and averaged from 10 fields in each sample. In addition, the microvascular media/lumen ratio (vessels under 500μm, representing branching orders of the interlobular arteries) was measured using conventional techniques.
standard blotting protocols were followed, as previously described, using specific antibodies against MCP-1 (1:2000, MyBioSourse.com), p47phox, p67phox, TGF-b, NFkB, and VEGF (Santa Cruz Biotechnology Inc., CA; 1:200 for all), RANTES, IL-8, IFN-γ (MBL International, MA, all 1:1000), CCR2 (1:500, ABR Affinity BioReagents CO), and its downstream effector protein kinase C (PKC, 1:1000, Cell Signaling). GAPDH (Covance, CA, 1:1000) were used as loading controls. Intensities of the protein bands were determined using densitometry, and expressed as ratio to GAPDH.
The stenotic kidney cortex of each animal was used for a hydroxyproline assay as described previously. Briefly, cortex was weighed, minced, homogenized, and diluted in PBS to 100mg cortex weight/ml. One-hundred-microliter samples were then hydrolyzed in 12M HCl, and duplicate samples analyzed by OH-proline assay and expressed as micrograms OH-proline content per 100mg cortex.
Results are expressed as mean±SEM. Multiple groups comparisons using one way ANOVA, followed by the post-hoc Tukey test. Statistical significance was accepted for p≤0.05.
Compared with normal, after 10 weeks RAS and RAS+bindarit pigs showed a similar increase in MAP and angiographic degree of RAS (Table 1). There were no significant differences in serum creatinine levels among the groups (Table 1).
Stenotic kidneys showed decreased basal RBF and GFR, and increased RVR, which were significantly improved after bindarit treatment. Ach induced a significant increase in RBF and decrease in RVR in normal, but not in RAS kidneys, indicating impaired endothelial function, which was also improved by bindarit (Table 1).
Stenotic kidneys showed increased expression of MCP-1, which was observed mainly in tubules and the interstitium (Figure 1&2). This was accompanied by interstitial macrophage infiltration (Figure 1) and upregulation of the inflammatory mediator NFkB, CCR-2, and its downstream effector PKC (Figure 2). Bindarit decreased tubulointerstitial MCP-1 expression and macrophage infiltration (Figure 1), and downregulated NFkB, CCR-2, and PKC expression (Figure 2), but had less effect on perivascular macrophages. Double immunostaining of macrophages and nitrotyrosine showed that macrophage infiltration and nitrotyrosine expression were highly spatially correlated. Nitrotyrosine expression in epithelial, endothelial and vascular smooth muscle cells was also increased in the stenotic kidney and attenuated by bindarit.
In addition, RAS pigs showed increased renal expressions of the chemokines RANTES and IL-8, which were not decreased by bindarit, while IFN-γ was similar among the three groups (Figure 3). These results support the amplified inflammation in the stenotic kidney and indicate that bindarit is a relatively selective inhibitor of MCP-1 in our pig model.
RAS kidneys also showed increased oxidative stress (Figure 4) as reflected in increased superoxide anion production (DHE staining) and expression of NAD(P)H oxidase p67phox (albeit not p47). Furthermore, a decrease in eNOS expression in RAS kidney was also improved by MCP-1 inhibition (Figure 4). Interestingly, bindarit decreased mainly tubular superoxide production, while vascular oxidative-stress was less affected (Figure 4).
Trichrome staining and collagen I immunofluorescence (Figure 5), as well as TGF-β expression (Figure 2), were all increased in the RAS kidney, as was total collagen content assessed by the hydroxyproline assay (Table 1), indicating increased fibrosis in the stenotic kidney. Importantly, all were attenuated by inhibition of MCP-1, suggesting that this chemokine mediates renal fibrosis.
Furthermore, the increased fibrosis in the stenotic kidney was accompanied by decreased microvascular density in all cortical zones (Figure 6). Inhibition of MCP-1 increased the density of larger microvessels (diameters 200–500μm), which are usually located in the inner and middle regions of the cortex of the stenotic kidney, but had no effect on smaller microvessels (<200μm), or on microvessels in the outer cortex. The media/lumen ratio was also increased in RAS and improved by bindarit (Table 1), indicating improved microvascular remodeling. The expression of VEGF was similar among the groups (Figure 6).
Compared with normal, the contralateral kidney in RAS pigs with RVH had slightly greater volume and basal renal perfusion, and unchanged GFR. However, RBF response to Ach was significantly impaired, indicating renovascular endothelial dysfunction in RVH, which was improved in bindarit-treated RVH pigs. Compared with the stenotic kidney, the RVH non-stenotic kidney showed higher hemodynamics and function and lower RVR. However, these differences with the stenotic kidneys largely disappeared in bindarit-treated animals, indicating the improvement in the stenotic kidney. RVR in the contralateral kidney of RVH+bindarit is significantly higher than RVR in untreated RVH (p=0.04), and tends to be higher than in normal (p=0.10) (Table 2).
In addition, although to a lesser degree than the stenotic RAS kidneys, the RVH contralateral kidneys showed renal interstitial recruitment of macrophages (Table 2), as well as upregulation of MCP-1 and CCR2 (Figure 7), suggesting that RVH imposes renal inflammation. These were accompanied by increased fibrosis, TGF-β expression, and oxidative stress, but all were attenuated by inhibition of MCP-1 (Figure 7). The expression of RANTES remained similar among the groups (Figure 7).
This study demonstrated that functional impairment and structural damage in the stenotic pig kidney are partly mediated by inflammation. Attenuation of MCP-1 expression improved renal function and decreased renal injury by decreasing inflammation and oxidative-stress, especially in the tubulointerstitial compartment. The novel renoprotective effects suggest a role for MCP-1 inhibition in preserving the kidney in chronic RAS.
In experimental models of RAS the contribution of inflammation to the pathogenesis of renal injury has been demonstrated [8, 24]. We have previously shown in a pig model that RAS induces a significant reduction in RBF and GFR and an increase in inflammation and oxidative stress[7, 8] in the stenotic kidney. The current study extends our previous observations and shows that these changes in RAS are accompanied by increased protein expression of MCP-1, and its inhibition not only attenuated renal macrophages recruitment, but also improved renal hemodynamics and function in both stenotic and contralateral kidneys. Therefore, blockade of the MCP-1 pathway might be a useful tactic to preserve both kidneys in RVH. The observation that inhibition of MCP-1 effectively improved many aspects of renal dysfunction and damage suggests that MCP-1 plays a pivotal upstream role in the cascade of events that ultimately results in renal injury in RAS. Indeed, previous studies have implicated this chemokine as a critical mediator of angiotensin II effects in vascular disease[25, 26].
The unaltered elevation in MAP in RAS+bindarit pigs agrees with previous observations that organ protection by MCP-1 inhibition is blood pressure independent[27, 28]. We and others [26–28] have also shown that in RAS plasma renin activity returns to basal levels after about 10 weeks of RAS, yet the tissue renin-angiotensin system likely remains activated. MCP-1 expression in tubules of the stenotic kidney might be upregulated by angiotensin II as well as by albumin, which is often increased in the renal tubules in kidney injury. Indeed, this study shows robust expression of MCP-1 in the renal tubules and in the vicinity of peritubular, perivascular, and interstitial macrophage infiltration. This may implicate MCP-1 and inflammatory cells in tubular injury that characterizes ischemic renal disease and in development of renal fibrosis. Furthermore, activated macrophages stimulate NAD(P)H oxidase to generate superoxide anion in their vicinity, as well as in epithelial, endothelial, and vascular smooth muscle cells (as shown in this study by DHE and nitrotyrosine staining). Indeed, we found that macrophage infiltration and nitrotyrosine expression were spatially correlated. Superoxide produces several vasoactive and fibrogenic factors, and increased oxidative stress in turn is a strong mediator of MCP-1 expression, thereby amplifying a positive feedback loop of renal injury. Moreover, the interaction between macrophages and renal epithelial cells could be central to RAS-induced renal injury. MCP-1/CCR2-dependent activation of NFkB in tubular epithelial cells might amplify local inflammation and fibrosis. The reduction of MCP-1 expression may, therefore, interrupt this cascade, and thereby attenuate inflammation and oxidative stress in renal epithelial cells. This notion is supported by our findings that bindarit reduced expression of MCP-1, NFkB, and superoxide in tubular epithelial cells. NFkB may also regulate MCP-1 and RANTES expression, but both may also be differentially regulated by infiltrating inflammatory cells, independently of NFkB[33, 34]. The attenuation of the expression of MCP-1 or CCR-2 may have contributed to the decline in macrophage infiltration in the diseased kidneys. Bindarit significantly decreased interstitial macrophage infiltration and oxidative stress (DHE staining), but had lower efficacy in decreasing vascular oxidative stress or inflammation. Bindarit treatment also reduced CCR2 and its downstream mediator PKC, suggesting that MCP-1 can regulate the expression of its receptor, and confirming that the drug is effective in decreasing MCP-1/CCR2 signaling in diseased kidneys.
Increased renal inflammation may directly lead to endothelial dysfunction and loss of microvessels, and consequently a decrease in both basal RBF and its responses to challenge. Furthermore, vascular remodeling (e.g. increased media/lumen ratio) may also impair vascular function in RAS. Indeed, the decrease in the number of vessels in the stenotic kidneys was associated with a decline in GFR and RBF. In addition to inflammation, reactive oxygen species like superoxide can also act as direct vasoconstrictors, mediate fibrosis and scarring in the ischemic kidney, and thereby restrict and interfere with vessel formation, leading to decreased microvascular density in RAS[9, 35]. Interestingly, chronic inhibition of MCP-1 improved RBF and GFR, and improved media/lumen ratio, but the number of small vessels was only partly restored in the stenotic kidney, possibly because bindarit attenuated oxidative damage in microvessels less effectively than in tubules, as shown by DHE staining. Speculatively, inflammatory changes play a greater role in regression of inner cortical (and thus larger) microvessels than in smaller and more superficial microvessels. Moreover, this observation may imply that vascular oxidative stress in RAS may be mediated by alternative or additional mechanisms other than MCP-1, such as direct activation of NAD(P)H oxidase by angiotensin II[36, 37]. Notably, MCP-1 may contrarily increase tumor angiogenesis in a VEGF dependent manner. Hence, the slight improvement in microvascular density, particularly in the face of unchanged VEGF expression, suggests that this might have been achieved indirectly by decreased renal fibrosis and remodeling that attenuated microvascular rarefaction, rather than a direct effect on angiogenesis. Along the same line, the improvement in RBF and GFR achieved by inhibition of MCP-1 was likely mediated mainly by attenuated renal inflammation and consequently improved vascular endothelial function in both the stenotic and non-stenotic kidneys.
The improvement in renal function by inhibition of MCP-1 may also account for its effect to decrease renal fibrosis. Mononuclear cells attracted through MCP-1/CCR2 signaling may be an important source of fibrogenic mediators, such as TGF-β and fibroblast growth factors. In addition to its chemotactic properties, MCP-1 enhances the fibrogenic potential of macrophages by inducingTGF-β1 and collagen synthesis.
Compared to the stenotic kidney, the functional impairment in the contralateral kidney was relatively mild, likely because of the short duration of the increase in blood pressure. Basal RBF was higher than normal, suggesting that in some kidneys the increase in blood pressure exceeded the range of RBF autoregulation and compensated for the decreased function of the stenotic kidney. Indeed, renal volume, RBF, and GFR all increased compared with the stenotic kidney. The impaired renovascular endothelial function (blunted RBF response to Ach) in the contralateral kidney, on the other hand, may have resulted from RVH-induced inflammation and oxidative stress. In fact, while in the stenotic kidney angiotensin II may directly induce renal MCP-1 expression and macrophage infiltration, in the contralateral kidney RVH per se can stretch mesangial cells, stimulating their monocyte chemoattractant activity via an MCP-1-dependent pathway. Furthermore, stretch can also increase oxidative stress, and thereby upregulate MCP-1 secondary to renal cell injury. Therefore, upregulation of MCP-1 expression in the contralateral kidney might be secondary to cellular injury, stretch, as well as oxidative stress, rather than to angiotensin II. Inhibition of MCP-1 at an early phase may hence attenuate its progression towards permanent renal injury. Our data show that decreased expression of MCP-1 can reduce the number of infiltrating macrophages in the contralateral kidney, improve its function, and decrease fibrosis.
MCP-1/CCR2 have been previously and successfully targeted in an attempt to decrease renal injury using gene transfer[42, 43], MCP-1 neutralizing antibody , or a CCR-2 antagonist in rodent models. However, most of these approaches are too invasive or impractical to use in large animals or humans, and furthermore, to date their functional consequences remained unclear. In contrast, bindarit is a well-tolerated oral drug, and this study demonstrates that its use for MCP-1 inhibition is relatively selective (as no other tested chemokine was affected), and this inhibition not only decreases renal inflammation and fibrosis, but also improves renal hemodynamics and function. It is possible that this method for inhibition of MCP-1 is incomplete, and that a higher dose might be needed to achieve complete blockade of MCP-1. However, being non-physiological, complete absence of MCP-1 activity may be associated with adverse effects (like susceptibility to infections) with long term administration. In addition, since the bindarit regimen was initiated at the same time we induced RAS, further studies are needed to determine if this approach is able to reverse pre-established or more advanced renal injury. The incomplete improvement in microvascular oxidative stress, rarefaction, and remodeling also argues against a direct role for MCP-1 in these processes in the stenotic kidney. Because the focus in this study was on the stenotic kidney, additional studies are also needed to characterize the involvement of MCP-1 in hypertensive renal damage in greater detail. Nevertheless, this study demonstrated that MCP-1 is involved in regulation of renal hemodynamics and function in RVH, and that its inhibition may have potential clinical utility.
Bindarit was kindly donated by Angelini Research Center – ACRAF, Italy.
Sources of Funding: This study was partly supported by NIH grant numbers DK73608, DK77013, HL77131, and HL085307.
Conflict(s) of Interest/Disclosure(s): Dr. Angelo Guglielmotti is employed by Angelini Research Center – ACRAF. Dr. Lerman provided a one-time consultation to the company regarding this study.