Inhibition of α v β 5 Integrin Attenuates Vascular Permeability and Protects against Renal Ischemia-Reperfusion Injury

Abstract

Ischemia-reperfusion injury (IRI) is a leading cause of AKI. This common clinical complication lacks effective therapies and can lead to the development of CKD. The αvβ5 integrin may have an important role in acute injury, including septic shock and acute lung injury. To examine its function in AKI, we utilized a specific function-blocking antibody to inhibit αvβ5 in a rat model of renal IRI. Pretreatment with this anti-αvβ5 antibody significantly reduced serum creatinine levels, diminished renal damage detected by histopathologic evaluation, and decreased levels of injury biomarkers. Notably, therapeutic treatment with the αvβ5 antibody 8 hours after IRI also provided protection from injury. Global gene expression profiling of post-ischemic kidneys showed that αvβ5 inhibition affected established injury markers and induced pathway alterations previously shown to be protective. Intravital imaging of post-ischemic kidneys revealed reduced vascular leak with αvβ5 antibody treatment. Immunostaining for αvβ5 in the kidney detected evident expression in perivascular cells, with negligible expression in the endothelium. Studies in a three-dimensional microfluidics system identified a pericyte-dependent role for αvβ5 in modulating vascular leak. Additional studies showed αvβ5 functions in the adhesion and migration of kidney pericytes in vitro Initial studies monitoring renal blood flow after IRI did not find significant effects with αvβ5 inhibition; however, future studies should explore the contribution of vasomotor effects. These studies identify a role for αvβ5 in modulating injury-induced renal vascular leak, possibly through effects on pericyte adhesion and migration, and reveal αvβ5 inhibition as a promising therapeutic strategy for AKI.

Keywords: acute renal failure; adhesion molecule; renal ischemia; vascular.

Figure 1.

Treatment with αvβ5 mAb protects kidneys in a rat model of renal IRI. To establish the dose-response efficacy of an αvβ5-blocking antibody (αvβ5 antibody), rats were treated with (A) 10, 3, or 1 mg/kg or (B) 3, 1, 0.3, or 0.1 mg/kg at 18 hours pre- and 30 hours post-clamp. Doses of 10 and 3 mg/kg were equally effective in attenuating ischemia-induced serum creatinine increases at 24, 48, and 72 hours post-injury. *P<0.05; **P<0.01 versus control antibody; error bars = SEM.

Figure 2.

Single dose of αvβ5 antibody improves kidney function and reduces tubular injury. (A) Rats were treated with a single dose (3 mg/kg) of αvβ5 antibody 6 hours before ischemic injury. This single dose was effective in attenuating ischemia-induced serum creatinine increases at 24, 48, and 72 hours postinjury. *P<0.05; **P<0.01 versus control antibody; error bars = SEM. (B) Kidneys were harvested at 24 and 72 hours postclamp from the study in A. Histopathologic examination of hematoxylin and eosin–stained kidney sections utilized several morphologic features to generate a composite lesion score (see Concise Methods for details). Rats treated with αvβ5 antibody had a significantly lower score at 72 hours post-IRI. *P<0.05 compared with control antibody.

Figure 3.

αvβ5 inhibition decreases ischemia-induced apoptosis. Kidneys were harvested at 24 and 72 hours postclamp after a single dose (3 mg/kg) of αvβ5 antibody 6 hours preclamp. Activated capsase-3 was used as a measure of apoptosis in post-IRI kidney sections. The degree of caspase-3 immunoreactivity was quantified and calculated as a percent area of medulla. Rats treated with αvβ5 antibody had significantly reduced apoptosis in the renal medulla at 24 hours postinjury. *P<0.05 compared with control antibody.

Figure 4.

Profiling of kidney reveals previously identified genes in rodent and human AKI are altered with αvβ5 inhibition. Heat map depicts transcripts differentially expressed at 24 or 72 hours post-IRI with αvβ5 antibody treatment compared with control antibody treatment. Table includes genes previously identified as altered in rodent IR models and in human AKI that were oppositely regulated with αvβ5 antibody treatment.

Figure 5.

αvβ5 inhibition reduces vascular leak after injury. (A) Texas Red-conjugated albumin and a 150 kD fluorescein-dextran were intravenously coinjected into rats at 24 hours post-ischemia to evaluate microvascular permeability after injury. Intravital imaging of the cortex revealed leakage of both the albumin and the high molecular weight dextran from the renal microvasculature in the control antibody-treated animals (A and B), and this leak was substantially reduced in the αvβ5 antibody-treated animals (C and D). The images were scored for degree of extravasation and the average leak scores for albumin (E) and the 150 kD dextran (F) were calculated. *P<0.05; **P<0.01 versus control antibody; error bars = SEM; image dimensions = 212×212 μm.

Figure 6.

Characterization of αvβ5 integrin expression in the kidney. Frozen sections of mouse kidney were stained for αvβ5 (red), CD31 (green), and PDGFRβ (white) with nuclei in blue. Representative images (63× magnification) from the glomerulus, cortex, medulla as well as an artery are shown. αvβ5 is expressed in podocytes, mesangial, epithelial, and perivascular cells of the kidney and within interstitial cells, and appears to be present within PDGFRβ-positive cells, but not in CD31-positive endothelial cells.

Figure 7.

αvβ5 regulates vascular permeability in a pericyte-dependent manner. (A) Human kidney endothelial cells were seeded into a three-dimensional microfluidic device and allowed to form a tube in the presence of flow. Cells were treated with control or αvβ5 antibody 1 hour before challenge with thrombin (3 U/ml). 70 kD Texas Red albumin was perfused into the vessel and the diffusion was measured by confocal microscopy to calculate the diffusive permeability coefficient (P_d). No significant difference was observed with αvβ5 antibody with endothelial cells alone (n=6). (B) All conditions are the same as described above except human kidney pericytes are first seeded into the microfluidics device, followed by seeding with the endothelial cells. In this condition, the αvβ5 antibody significantly inhibited the thrombin-induced vascular permeability. *P<0.05 versus thrombin + control antibody; error bars = SEM.

Figure 8.

αvβ5 regulates pericyte adhesion and migration on vitronectin. Human kidney pericytes were seeded on plates coated with collagen or vitronectin. (A) Fluorescently labeled cells are seeded in the presence of the αvβ5 antibody, an αvβ3 antibody, or a control antibody and were counted after washing the adherent cells. The αvβ5 antibody significantly reduced adhesion of cells on vitronectin compared with control antibody treatment. ***P<0.001 versus control antibody; error bars = SEM. (B) Cells were treated with antibody 1 hour before scratch wound, and images were taken every 2 hours for 2 days and wound confluence was calculated. The αvβ3 antibody had no significant effect on migration, the JNK inhibitor SP600125 (included as a positive control) inhibited migration on both substrates and the αvβ5 antibody significantly inhibited migration of pericytes on vitronectin. *P<0.05 versus control antibody.

Figure 9.

Treatment with αvβ5 antibody after ischemia provides protection from injury. (A) Rats were treated with a single dose (3 mg/kg) of αvβ5 antibody 8 hours postclamp release. This single dose was effective in attenuating ischemia-induced serum creatinine increases at 48 hours postinjury. *P<0.05 versus control antibody; error bars = SEM. (B) Kidneys were harvested at 72 hours postclamp from the study in A for quantitative PCR analysis. At 8 hours postclamp, treatment significantly reduced the expression of two biomarkers induced by injury (havcr1, lcn2) and increased the expression of two biomarkers diminished by injury (umod, egf). *P<0.05; **P<0.01 versus control antibody; error bars = SEM.