Journal Logo

Immunobiology

Exogenous alpha-1-Acid Glycoprotein Protects against Renal Ischemia-Reperfusion Injury by Inhibition of Inflammation and Apoptosis

de Vries, Bart1; Walter, Sarah J.1; Wolfs, Tim G.A.M.1; Hochepied, Tino2; Räbinä, Jarkko3; Heeringa, Peter4; Parkkinen, Jaakko3; Libert, Claude2; Buurman, Wim A.1,5

Author Information
doi: 10.1097/01.TP.0000138096.14126.CA
  • Free

Abstract

Postischemic renal failure represents a major problem in clinical medicine, including transplantation; however, effective treatment for acute renal failure is currently not available (1). In a murine model for renal ischemia-reperfusion (I/R) injury, we recently showed that administration of the acute phase protein alpha-1-acid glycoprotein (AGP), also called orosomucoid, is protective against I/R-induced acute renal failure (2). In septic as well as hemorrhagic shock and acute liver failure models, exogenous AGP provided protection against organ injury and improved animal survival (3–5). These studies indicate that AGP is potentially a powerful therapeutic agent in a broad range of diseases mediated by ischemic injury.

AGP appears to have many different functions in vivo, and several mechanisms of action have been suggested. In vivo, the effects of AGP are thought to be mediated by fucosylated glycans expressed on the AGP protein (6, 7). These glycans inhibit the interaction between neutrophils and endothelium, probably by interfering with selectin-dependent cell adherence, thereby preventing neutrophil infiltration (8, 9). Indeed, Williams et al. (7) showed that fucosylated human AGP protected mice more effectively from intestinal ischemic injury than nonfucosylated AGP. In vitro studies have indicated that AGP inhibits superoxide and hydrogen peroxide generation by stimulated neutrophils, suggesting that AGP is also capable to inhibit neutrophil activation (10).

An interesting observation was made by Libert et al. (11) showing that transgenic mice constitutively over-expressing AGP are not protected against tumor necrosis factor (TNF)-α/galactosamine-induced acute liver failure, whereas exogenous AGP was shown to be protective in the very same model. This finding suggests that only acute administration of AGP, in contrast with endogenously produced AGP, has protective effects.

Thus, although exogenous AGP provides protection in various experimental models, the mechanism of protection remains to be established. The aim of the present study was to characterize the effects of endogenous as well as exogenous AGP in a renal I/R model, with special emphasis on the influence of fucosylation of the AGP protein. To get more insight in the mechanism of protection, we studied the effects of AGP on I/R-induced cytoskeletal derangements, apoptotic cell death, and inflammation.

MATERIALS AND METHODS

Reagents and Antibodies

Rabbit anti-human AGP was purchased from DAKO (St Louis, MO). NIMP-R14 (rat anti-mouse neutrophil monoclonal antibody [mAb]) was kindly provided by Dr. M. Strath (National Institute for Medical Research, London, UK). Polyclonal rabbit anti-active caspase-7 antibody was obtained from Biovision (Mountain View, CA), cross-reacting polyclonal rabbit anti-human ZO-1 antibody from Zymed Laboratories (San Francisco, CA), and Texas-red Phalloidin from Molecular Probes (Eugene, OR). Goat anti-mouse C3 polyclonal antibody was purchased from Cappel (ICN Biomedicals, Aurora, OH). Secondary antibodies, peroxidase conjugated goat anti-rat immunoglobulin (Ig)G, rabbit anti-goat IgG, as well as fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG were purchased from Jackson (West Grove, PA). All other reagents were purchased from Sigma (St. Louis, MO).

Purification of Human Plasma AGP and Fucose-Depleted AGP

Human AGP was purified from plasma of healthy blood donors according to the method of Hao and Wickerhauser (12). Fucose-depleted AGP was prepared by subjecting human plasma AGP to affinity chromatography with agarose-bound Aleuria aurantia lectin (Vector Laboratories, Burlingame, CA). Approximately 85% of plasma AGP was recovered in the nonbinding effluent fraction, and 15% was eluted from the lectin column with fucose. Fractions containing nonbinding AGP (designated as fucose-depleted AGP) were pooled, concentrated by ultrafiltration (10 kDa cut-off), and sterile filtered. Structural analysis of N-glycans in AGP preparations was performed by BioTie Therapies (Helsinki, Finland) by matrix-assisted laser desorption mass spectrometry with time-of-flight detection after enzymatic detaching and purification of glycans. The degree of N-glycan fucosylation in the original plasma AGP and fucose-depleted AGP was 21% and 12%, respectively. However, AGP molecules having two or more fucosylated glycan chains (proteins with high affinity for selectins) decreased more effectively, leading to an 80% reduction of highly fucosylated AGP, which is capable of interfering with selectin-lectin interactions.

Experimental Protocol

Male Swiss mice weighing 25 to 30 g were obtained from Charles River Breeding Laboratories (Heidelberg, Germany). Animals were housed individually in standard laboratory cages and were allowed free access to food and water. The studies were carried out under a protocol approved by the Institutional Animal Care Committee of the University of Maastricht. Mice were subjected to 45 minutes of unilateral ischemia of the left kidney followed by reperfusion and contralateral nephrectomy inducing reversible acute renal failure without mortality, as described in detail previously (13). The animals were killed 24 hours after reperfusion because at this time point, renal inflammation and organ failure are most pronounced. At the time of sacrifice, blood was collected, and the left kidney was harvested for analysis.

During ischemia, just before reperfusion, mice were administered 0.1, 0.5, or 5 mg of hAGP or fucose-depleted hAGP intraperitoneally dissolved in 0.5 mL phosphate-buffered saline (PBS) or 0.5 mL PBS as control treatment (n=8 per group).

In separate experiments, transgenic C57BL/6 mice over-expressing rat AGP were used (11). The serum concentration of rat AGP in transgenic mice is approximately 3.5 mg/mL, which is comparable with serum concentrations obtained after administration of 5 mg exogenous AGP (11). As described above, rat AGP–over-expressing mice or wild-type C57BL6 mice, as a control, were subjected to renal ischemia for 45 minutes followed by 24-hour reperfusion (n=6 per group).

Human AGP Measurement

To measure hAGP concentrations in mice, an hAGP sandwich enzyme-linked immunoadsorbent assay (ELISA) was used. This ELISA has been developed at our institute, as reported previously (14). The ELISA is specific for hAGP and does not detect mouse AGP.

Renal Histology

Cryostat sections (5 μm) of frozen tissue were cut and stained for active caspase-7, filamentous actin (F-actin), and ZO-1, as described in detail previously (15). After blocking aspecific antibody binding, slides were incubated for 1 hour at room temperature with the anti-active caspase-7 or the anti-ZO-1 primary antibody. After washing in PBS, slides were incubated for 30 minutes with the FITC-labeled secondary antibody with the addition of Texas red phalloidin, which specifically binds to F-actin. After washing, the slides were mounted using glycerol PBS with DAPI and viewed with an immunofluorescence microscope. No significant staining was detected in slides incubated with control serum instead of the primary antibody, indicating the absence of significant background staining. The number of caspase–7-positive tubules was quantified by counting 20 fields of vision per kidney section (3 sections per kidney, 4 kidneys per group) at ×200 magnification in a blinded fashion.

Cryostat sections were also stained for neutrophils as described previously (13). Neutrophils were counted by examining 10 fields of vision per kidney section (3–4 sections per kidney, 4 kidneys per group) at ×200 magnification in a blinded fashion.

Staining for complement factor C3 has been described previously (13). C3 staining was quantified using a computerized morphometry system (Quantimet 570; Leica, Cambridge, UK). To obtain representative data, four complete cross-sections of four kidneys per group were included. Per section, 30 fields of vision (10 cortical, 10 corticomedullary, and 10 medullary fields) at ×200 magnification were measured. Data are expressed as percentages of positive-stained tissue of the total tissue measured.

Myeloperoxidase ELISA

A new ELISA for the measurement of tissue myeloperoxidase (MPO) was used as a quantitative measure for tissue infiltration of neutrophils. Renal tissue was homogenized in a buffer containing 200 mM NaCl, 10 mM Tris HCl, pH 7.0, 5 mM EDTA, 10% glycerol, 1 mM PMSF, 0.1 μM aprotinin, and 1 μg/mL leupeptin. After centrifugation at 14,000g for 10 minutes, supernatants were stored at −20°C until analysis. An mAb directed against mouse MPO (8F4) was coated at a concentration of 3 μg/mL in a 0.1 M carbonate buffer (pH 9.6) for 1 hour at room temperature onto 96-well plates (Immuno-Maxisorp; Nunc, Roskilde, Denmark). Free binding sites were blocked with 1% bovine serum albumin (BSA) in PBS for 1 hour at room temperature. Wash buffer consisted of 150 mM NaCl, 10 mM Tris HCl, pH 8.0, containing 0.05% Tween-20. Plates were washed five times after each incubation step. Samples were diluted in 300 mM NaCl, 10 mM Tris HCl, pH 8.0 with 0.25% Tween-20, 2% BSA and 1% normal goat serum. Samples, as well as mouse MPO, used as a standard, were incubated for 1 hour at room temperature. Mouse MPO was purified from WEHI-3 cells as described previously (16). After washing, rabbit anti-human MPO (DAKO, Glostrup, Denmark) was incubated for 1 hour at room temperature. Next, after washing, peroxidase-labeled goat anti-rabbit IgG was incubated for 1 hour. After washing, color reaction was developed using 3,3′,5,5′-tetramethylbenzidine (Kirkegaard & Perry Laboratories, Gaithersburg, MD) and H2SO4 to stop the reaction. Color intensity was measured by determination of the absorbance at 450 nm using a micro-ELISA reader.

Apoptosis Assay

Presence of internucleosomal DNA cleavage in kidneys was investigated with a commercial ligase-mediated (LM)-polymerase chain reaction (PCR) assay kit (Apoalert, Clontech, Palo Alto, CA) enabling semiquantitative measurement of the extent of apoptosis. DNA was isolated from tissue samples using a commercially available DNA purification kit (Promega, Madison, WI). DNA purity and concentration were determined by electrophoresis through a 0.8% agarose gel containing ethidium bromide followed by visualisation under ultraviolet illumination as well as by measuring absorbance at 260/280 nm. Dephosphorylated adaptors were ligated to 5′-phosphorylated blunt ends with T4 DNA ligase (during 16 hours at 16°C) and served as primers in a LM-PCR. Amplified DNA was subjected to gel electrophoresis on a 1.2% agarose gel containing ethidium bromide.

Renal Function

Blood urea nitrogen (BUN) was measured in serum obtained at the time of sacrifice using a Urea 25 Kit (ABX Diagnostics, Eindhoven, Holland) in a Cobas Fara autoanalyzer (Roche, Basle, Switzerland).

Statistical Analysis

Data are expressed as the mean±SEM, except for data obtained from renal histology (C3 deposition and neutrophil influx) and MPO-ELISA, which are expressed as the median with interquartile ranges. Statistical significance was evaluated by one-way analysis of variance with Bonferroni’s post hoc test. P<0.05 was taken to denote statistical significance.

RESULTS

Treatment with Exogenous Human AGP Protects against Renal I/R Injury in a Dose-Dependent Manner, Independently from the Degree of Fucosylation

First, we determined whether intraperitoneal treatment with AGP results in detectable circulating levels of hAGP. Whereas in healthy mice and mice subjected to renal I/R no hAGP could be detected, treatment with hAGP resulted in dose-dependent levels of circulating hAGP as measured 24 hours after renal I/R (concentrations of 3.05±0.37, 0.45±0.05, and 0.11±0.03 μg/mL hAGP after 5, 0.5, and 0.1 mg hAGP treatment, respectively).

Next, we examined whether exogenous hAGP provides functional protection against renal I/R injury. The data presented here show that hAGP protects against the I/R-induced loss of renal function, as measured by BUN levels, in a dose-dependent manner (Fig. 1). Whereas 0.1 mg hAGP already provided a significant protection as reflected by reduced BUN levels, maximal protection was provided using 5 mg of hAGP.

FIGURE 1.
FIGURE 1.:
Human alpha-1-acid glycoprotein (hAGP) reduces loss of renal function after ischemia-reperfusion (I/R) independently from the degree of fucosylation. Compared with control-treated animals, hAGP dose dependently reduced blood urea nitrogen (BUN) values as measured 24 hours after renal I/R. Also fucose-depleted hAGP and bovine AGP (bAGP) significantly reduced BUN levels in a dose-dependent manner. There were no significant differences in BUN values of hAGP, fucose-depleted hAGP, or bAGP-treated animals. Statistical significance as compared with control-treated animals denoted at P<0.05* and P<0.01‡. Data are means±SEM.

To unravel the mechanism of protection by AGP, we determined whether the functional protection against renal I/R injury is dependent on the degree of fucosylation of AGP. Therefore, mice were treated with hAGP, fucose-depleted hAGP, or bovine AGP (bAGP). The data show that AGP protects against loss of renal function independently from its degree of fucosylation (Fig. 1). Equally effective, hAGP and fucose-depleted hAGP provided protection against renal-function loss as compared with control-treated mice in a dose-dependent manner. These data indicate that the protective effects of hAGP against renal I/R injury are not mediated by fucose groups but rather by other properties.

AGP–Over-Expressing Mice are not Protected against Renal I/R Injury

Next, we examined whether high levels of endogenous AGP could protect against renal I/R injury. Therefore, transgenic mice over-expressing rat AGP and their wild-type controls were subjected to renal I/R injury. Libert et al. (11, 17) previously showed that AGP–over-expressing mice are not protected against TNF-α/galactosamine-induced hepatitis; in contrast, AGP–over-expressing mice showed protection against lethal Gram-negative infection. Here, we show that mice with high levels of endogenous AGP, as compared with wild-type mice and control Swiss mice, are not protected against the loss of renal function after renal I/R (Fig. 2). There was no significant difference in postischemic BUN levels between Swiss mice, C57BL/6 wild-type mice, and C57BL/6 transgenic (AGP–over-expressing) mice.

FIGURE 2.
FIGURE 2.:
Transgenic mice over-expressing rat AGP are not protected against ischemia-induced loss of renal function. BUN levels 24 hours after renal I/R of transgenic mice over-expressing rat-AGP were not different from their wild-type controls (C57BL/6) or Swiss mice controls. Statistical significance as compared with control animals (C57BL/6 and Swiss) denoted at P<0.05*. Data are means±SEM.

Human AGP Inhibits I/R-Induced Inflammation Independently from the Degree Of Fucosylation

AGP inhibits the infiltration of neutrophils in postischemic tissue (2, 7). This inhibitory effect has been reported to be dependent on the presence of fucose groups on the AGP protein in intestinal I/R injury (7). Because the influx of neutrophils is a prominent feature of renal I/R injury, we determined whether AGP prevents neutrophil influx after renal I/R injury, and moreover, we examined whether the effects on neutrophil infiltration are dependent on the degree of fucosylation. Our immunohistochemical data show that AGP dose-dependently reduces the influx of neutrophils induced by renal I/R injury (Fig. 3A). Interestingly, fucose-depleted AGP appeared to have equal inhibitory effects on neutrophil infiltration as naturally fucosylated AGP (Fig. 3A). Moreover, the data show that a high level of constitutively expressed AGP in AGP-transgenic mice, as compared with wild-type mice, does not influence the influx of neutrophils (Fig. 3A).

FIGURE 3.
FIGURE 3.:
hAGP inhibits I/R neutrophil-influx, independently from the degree of fucosylation. Infiltrating neutrophils were counted after immunohistochemical staining (3–4 sections per kidney, 4 kidneys per group) (A). Postischemic infiltration of neutrophils was also quantified using a myeloperoxidase-enzyme-linked immunoadsorbent assay (MPO-ELISA) (B). Neutrophils were scarcely present in kidneys obtained from healthy mice, whereas ischemia followed by reperfusion induced a strong influx of neutrophils. Treatment with hAGP dose dependently reduced the influx of neutrophils at 24 hours reperfusion. Fucose-depleted hAGP was as effective as natural hAGP. Transgenic mice over-expressing rat AGP were not protected against ischemia-induced influx of neutrophils as compared with their wild-type littermates. Statistical significance as compared with control-treated animals denoted at P<0.05*. Data expressed as median number of neutrophils per field of vision with interquartile ranges (A) or median concentration of MPO (μg/g) with interquartile ranges (B).

Next to immunohistochemistry, we used a new MPO-ELISA as a quantitative measure of renal neutrophil-influx. The data show that, as compared with renal tissue from healthy control mice, renal I/R induces an evident increase in renal MPO levels (Fig. 3B). Treatment with hAGP dose-dependently reduces renal MPO levels as compared with control-treated animals. Moreover, fucose-depleted AGP is as effective as naturally fucosylated hAGP in reducing renal MPO levels. Finally, AGP over-expression in transgenic mice does not reduce renal MPO levels as compared with their wild-type controls (Fig. 3B).

Taken together, using semiquantitative immunohistochemistry as well as a quantitative MPO-ELISA, we show that the influx of neutrophils into postischemic renal tissue is inhibited by exogenous hAGP in a dose-dependent manner independently from the degree of fucosylation. In contrast, endogenous over-expression of AGP does not influence the ischemia-induced influx of neutrophils.

We recently showed that complement activation, as a part of the inflammatory response, is a crucial event in I/R injury (13, 18). In particular, because the complement system plays an important role in postischemic neutrophil infiltration, we determined whether treatment with hAGP could reduce the activation of the complement system. The present data show that hAGP strongly inhibits the ischemia-induced deposition of complement factor C3 (Fig. 4). Whereas renal I/R led to significant renal deposition of C3, treatment with hAGP dose-dependently inhibited C3 deposition (Fig. 4).

FIGURE 4.
FIGURE 4.:
hAGP significantly reduces complement activation after renal I/R. Renal C3 deposition was assessed by immunohistochemistry and quantified using the Leica Quantimet 570 system (3–4 kidneys per group, 30 fields of vision per kidney). As compared with healthy control mice (A), renal ischemia followed by 24 hours of reperfusion-induced C3 deposition, which was most prominent in the corticomedullary region (B). Treatment with hAGP dose dependently reduced renal complement deposition(C, 0.5 mg hAGP; D, 5 mg hAGP). Quantification of C3 staining showed that indeed hAGP significantly inhibited renal complement deposition (E). Data expressed as median percentages of positive stained tissue with interquartile ranges. Statistical significance as compared with control treated mice denoted at P<0.05*.

All together, these data indicate that exogenous AGP, in contrast with high levels of endogenous AGP, protects against I/R-induced inflammation and renal function loss. Surprisingly, these protective effects of AGP appear to be independent from its fucosylation and are thus mediated by other as yet unknown mechanisms.

Human AGP Inhibits I/R-Induced Apoptosis Independently from the Degree of Fucosylation

Apoptosis plays a crucial role in the initiation of renal I/R injury (19). Therefore, we questioned whether hAGP would inhibit renal apoptosis. Here, we show that kidneys of animals treated with hAGP showed a reduction in internucleosomal DNA cleavage, a specific hallmark of apoptosis, as compared with control treatment (Fig. 5). Fucose-depleted hAGP inhibited renal apoptosis as effectively as natural hAGP and bAGP, indicating that the inhibitory effects of the AGP protein on apoptotic cell death are independent from its fucosylation (Fig. 5).

FIGURE 5.
FIGURE 5.:
Treatment with hAGP abrogates I/R-induced apoptosis. The extent of renal apoptosis was reflected by fragmentated DNA amplified by ligase-mediated polymerase chain reaction (PCR) and visualized on ethidium-bromide-stained gel. In control-treated animals, I/R-induced evident internucleosomal DNA cleavage. Natural hAGP, fucose-depleted hAGP, as well as bovine AGP reduced renal apoptosis. Data are representative for three independent assays on different renal samples (3 per group). M, 100 bp molecular weight marker.

AGP Prevents Cytoskeletal Derangements Induced by Renal I/R Injury

Actin cytoskeletal changes are reported to precede the induction of apoptosis in (renal) epithelial cells (20–22). Indeed, we recently showed that apoptotic cell death in the course of renal I/R injury is localized to tubular cells, which show severe actin cytoskeletal derangements (15). Thus, we questioned whether the protective, among others anti-apoptotic, effects of hAGP could be mediated by cytoskeletal protection. Therefore, we stained renal tissue for active caspase-7 as a marker of apoptosis and F-actin, the main component of the cytoskeleton, mainly present in the brush border of tubular epithelial cells. The data show that, whereas in control kidneys no apoptotic cell death could be detected, after renal I/R, a significant number of injured tubuli stained positive for active caspase-7 (Fig. 6A and C). Apoptosis mainly localized to tubular epithelial cells with severe cytoskeletal derangements (Fig. 6A and C). Treatment with hAGP significantly reduced the number of caspase–7-positive tubuli (Fig. 6, E and G). Most striking was the protective effect of hAGP on tubular cytoskeletal structure; in contrast with control-treated mice, hAGP-treated animals showed minimal injury to the tubular brush border and a well-preserved tubular structure (Fig. 6E).

FIGURE 6.
FIGURE 6.:
hAGP abrogates I/R-induced apoptosis and disruption of epithelial integrity. (A, C, and E) Filamentous actin (F-actin) was stained red (Texas-Red), and active caspase-7 was stained green (fluorescein isothiocyanate). Nuclei were stained blue (4,6-diamidino (2)phenylindole) in all sections. Renal I/R induced activation of caspase-7 in tubular epithelial cells (C, in green) as compared with the normal kidney where active caspase-7 is not present (A). AGP treatment prevented caspase-7 activation (E). Quantification of the number of caspase-7 positive tubuli shows that AGP significantly inhibited apoptotic cell-death induced by renal I/R (G). In healthy animals, tubular epithelial brush-borders are intact as determined by filamentous actin (F-actin) staining (A, red). After I/R, tubular epithelial cells lost brush-border integrity, and caspase-7 activation is mainly present in tubular epithelial cells with complete disrupted brush-border (C). Treatment with AGP evidently preserved tubular epithelial brush-border integrity (E). (B, D, an F) Tight-junction protein ZO-1 is stained in green (fluorescein isothiocyanate). In the normal kidney, ZO-1 is mainly located at the apical side of tubular epithelial cells (B, green). Renal I/R-induced a severe disintegration of epithelial tight-junctions (D); tubular cells with complete disruption of ZO-1 showed nuclear condensation and fragmentation, two hallmarks of apoptotic cell-death (D, white arrows). AGP treatment completely prevented tight-junction disintegration (F). (magnification, ×600)

An important determinant of cell survival, next to maintenance of cell structure, is the maintenance of cell-matrix and cell-cell interactions such as tight junctions (23). Therefore, we stained tissue sections for ZO-1, an intracellular protein that takes part in the tight junction. The present data show that in the normal kidney, the ZO-1 protein localizes to the apical side of the tubular epithelial cells (Fig. 6B). Renal ischemia induced a severe disruption of tight junctions, as demonstrated by the disappearance of the normal ZO-1 organization (Fig. 6D). In addition, ZO-1 is found in the tubular lumen, probably in tubular cellular debris (Fig. 6D). Interestingly, in particular, tubular cells with complete disruption of their tight-junctional complex showed nuclear condensation and fragmentation, two hallmarks of apoptotic cell death (Fig. 6D, white arrows). Treatment with hAGP prevented this I/R-induced destruction of tight junctions; the ZO-1 protein remained well organized in kidneys from AGP-treated mice, comparable with control kidneys (Fig. 6F). These data show that AGP treatment preserves cellular structure and cell-cell interactions after renal I/R.

DISCUSSION

AGP is an acute-phase protein, and as for other acute-phase proteins, the function of AGP is not known. In general, AGP is considered to be an important drug-binding protein, binding a wide variety of drugs as well as endogenous substances (24). Moreover, AGP has been implicated to have various immunomodulating effects, such as inhibition of lymphocyte proliferation, inhibition of platelet aggregation, activation of mononuclear cells, and inhibition of neutrophil activation (8).

Exogenous AGP has been shown to be protective against lethal inflammatory hepatitis induced by galactosamine in combination with either TNF-α or lipopolysaccharide(LPS) (5, 25). This in contrast with LPS-induced systemic shock in absence of liver toxicity, in which AGP treatment was not protective (4, 5). In the same study, AGP provided protection against septic peritonitis and hypovolemic shock (4). It was suggested that the protective effects of AGP were mediated by preservation of capillary barrier function (4). In line, Johnsson, Haraldsson, and Rippe (26, 27) showed that AGP is essential for the maintenance of glomerular capillary permselectivity.

Highly fucosylated AGP has been shown to be protective against ischemic injury in a rat intestinal I/R model (7). Local as well as remote (liver and lung) injury were inhibited by recombinant hAGP obtained by cotransfection of Chinese Hamster Ovary (CHO) cells with a human fucosyltransferase more efficiently than by nonfucosylated recombinant hAGP (7). Fucose groups present on neutrophils form an essential part of the ligand for endothelial selectin molecules, promoting neutrophil adhesion and migration. It is therefore suggested that AGP protects against I/R injury by way of fucose-mediated interference of neutrophil-endothelium interaction (7, 9). In contrast, in the present study, we show that the protective effects of AGP against renal I/R injury are independent from the degree of fucosylation, indicating that the effects of AGP are not mediated by direct interference with neutrophil-selectin interactions. Natural fucosylated and fucose-depleted AGP equally reduced the postischemic influx of neutrophils. For quantification of neutrophil influx, we used a newly developed MPO-ELISA. Our semiquantitative immunohistologic data correlated well with the renal MPO levels as measured with this ELISA. The classical enzymatic MPO assays, which are often used to quantify neutrophil influx, are hard to interpret, and in particular in renal tissue, have high background values and are difficult to reproduce, probably because of high endogenous peroxidase levels (28, 29). These difficulties are circumvented applying an ELISA using antibodies specifically recognizing murine MPO.

AGP has been shown to inhibit apoptotic cell death in vivo (2, 3, 30). In TNF-α/galactosamine and TNF-α/actinomycin D induced hepatitis as well as renal I/R injury, AGP prevented apoptotic DNA cleavage (2, 3). We previously showed that the inhibition of apoptosis by a broad-spectrum caspase-inhibitor (Z-VAD-fmk) is protective against I/R injury, by inhibition of neutrophil influx among other mechanisms (19). These data would suggest that the effects of AGP on renal function and inflammation are primarily mediated by the inhibition of apoptotic cell death. Indeed, the present study shows that AGP reduces tubular epithelial apoptosis, an effect independent from the presence of fucose groups. However, a direct inhibition of caspase activation is not the case (Libert, unpublished results), suggesting that the effects of AGP on apoptosis are most likely indirect (30).

Cytoskeletal derangements have been reported to initiate apoptosis in epithelial cells in vitro (20). Also, disruption of cell-cell interaction has been implicated in the induction of apoptosis in vitro (23). Recently, we showed in vivo that apoptotic cell death colocalizes with cytoskeletal derangements (e.g., F-actin breakdown) after renal I/R (15). The present study shows for the first time that AGP prevents this ischemia-induced breakdown of the actin cytoskeleton. AGP also prevents the disruption of tight junctions. We therefore hypothesize that the protective effects of AGP observed in the present study are mediated by the preservation of cellular structure and cell-cell interaction of renal epithelial cells. The mechanism by which AGP protects against ischemia-induced loss of cellular structure, cell-cell interaction, apoptotic cell-death, and subsequent inflammation still remains to be established.

In vitro studies by Sörensson et al. (31) demonstrated that AGP increases the intracellular cyclic adenosine monophosphate (cAMP) concentration in endothelial cells. Interestingly, enhancement of cellular cAMP levels has been reported to increase number of tight junctions and decrease epithelial and endothelial permeability in vitro (32, 33). In addition, cAMP enhancement inhibits apoptosis in stimulated epithelial cells (34). Therefore, an intriguing hypothesis would be that the protective effects of AGP observed in the present study are mediated by up-regulation of cAMP. This is a particularly tempting idea because enhancement of cAMP levels has already been shown to protect against renal I/R injury (35). Of course, additional studies are required to establish the potential role of cAMP in AGP-mediated effects in vivo.

An interesting new finding in the present study is the observation that transgenic mice constitutively over-expressing rat AGP, in contrast with mice treated with exogenous AGP, are not protected against renal I/R injury. This is in line with studies by Libert et al. (11) showing that transgenic mice over-expressing AGP are not protected against TNF-α/galactosamine-induced acute liver failure, whereas exogenous AGP was shown to be protective in the very same model. Apparently, endogenous circulating AGP is essentially different from acute administered AGP. As hypothesized by Libert et al. (11) an important difference between endogenous and exogenous AGP might be related to the binding properties of the AGP protein; endogenous circulating AGP might be bound to a binding factor making it inactive, whereas exogenous AGP, unbound, is protective as described in several in vivo models (11).

The present study describes the protective effects of hAGP against renal I/R injury. Treatment with hAGP has been shown to be without adverse effects even in high doses in rats (36). Clinical treatment with hAGP, which is an endogenous protein, would have several interesting applications. On the basis of the present study, hAGP could be used preventively if acute renal failure is to be expected, for instance after cardiovascular surgery or transplantation of marginal kidneys. In the setting of organ transplantation, one could also consider donor treatment and even the addition of AGP to cold preservation solution in particular when using marginal donors.

In summary, the present study shows that hAGP protects against renal I/R injury by preservation of epithelial integrity, the inhibition of apoptosis, and reduction of inflammation. Moreover, the protective effects were shown to be independent from the degree of fucosylation of the AGP protein. We conclude that hAGP, which has been shown to be nontoxic and safe, might be a potential therapeutic agent in the treatment of clinical acute renal failure as seen after renal transplantation.

REFERENCES

1. Thadhani R, Pascual M, Bonventre JV. Acute renal failure. N Engl J Med 1996; 334(22): 1448.
2. Daemen MA, Heemskerk VH, van’t Veer C, et al. Functional protection by acute phase proteins alpha(1)-acid glycoprotein and alpha(1)-antitrypsin against ischemia/reperfusion injury by preventing apoptosis and inflammation. Circulation 2000; 102(12): 1420.
3. Van Molle W, Denecker G, Rodriguez I, et al. Activation of caspases in lethal experimental hepatitis and prevention by acute phase proteins. J Immunol 1999; 163(10): 5235.
4. Muchitsch EM, Auer W, Pichler L. Effects of alpha 1-acid glycoprotein in different rodent models of shock. Fundam Clin Pharmacol 1998; 12(2): 173.
5. Libert C, Brouckaert P, Fiers W. Protection by alpha 1-acid glycoprotein against tumor necrosis factor-induced lethality. J Exp Med 1994; 180(4): 1571.
6. De Graaf TW, Van der Stelt ME, Anbergen MG, et al. Inflammationinduced expression of sialyl Lewis X-containing glycan structures on alpha 1-acid glycoprotein (orosomucoid) in human sera. J Exp Med 1993; 177(3): 657.
7. Williams JP, Weiser MR, Pechet TT, et al. alpha 1-Acid glycoprotein reduces local and remote injuries after intestinal ischemia in the rat. Am J Physiol 1997; 273: G1031.
8. Hochepied T, Berger FG, Baumann H, et al. alpha(1)-Acid glycoprotein: an acute phase protein with inflammatory and immunomodulating properties. Cytokine Growth Factor Rev 2003; 14(1): 25.
9. Fournier T, Medjoubi NN, Porquet D. Alpha-1-acid glycoprotein. Biochim Biophys Acta 2000; 1482: 157.
10. Vasson MP, Roch-Arveiller M, Couderc R, et al. Effects of alpha-1 acid glycoprotein on human polymorphonuclear neutrophils: influence of glycan microheterogeneity. Clin Chim Acta 1994; 224(1): 65.
11. Libert C, Hochepied T, Berger FG, et al. High-level constitutive expression of alpha 1-acid glycoprotein and lack of protection against tumor necrosis factor-induced lethal shock in transgenic mice. Transgenic Res 1998; 7(6): 429.
12. Hao YL, Wickerhauser M. Development of large-scale fractionation methods. IV. A simple method for the large-scale preparation of alpha1-acid glycoprotein. Biochim Biophys Acta 1973; 322(1): 99.
13. De Vries B, Matthijsen RA, Wolfs TG, et al. Inhibition of complement factor C5 protects against renal ischemia-reperfusion injury: inhibition of late apoptosis and inflammation. Transplantation 2003; 75(3): 375.
14. Van Dielen FM, van’t Veer C, Schols AM, et al. Increased leptin concentrations correlate with increased concentrations of inflammatory markers in morbidly obese individuals. Int J Obes Relat Metab Disord 2001; 25(12): 1759.
15. De Vries B, Matthijsen RA, Van Bijnen AA, et al. Lysophosphatidic Acid prevents renal ischemia-reperfusion injury by inhibition of apoptosis and complement activation. Am J Pathol 2003; 163(1): 47.
16. Xiao H, Heeringa P, Hu P, et al. Antineutrophil cytoplasmic autoantibodies specific for myeloperoxidase cause glomerulonephritis and vasculitis in mice. J Clin Invest 2002; 110(7): 955.
17. Hochepied T, Van Molle W, Berger FG, et al. Involvement of the acute phase protein alpha 1-acid glycoprotein in nonspecific resistance to a lethal gram-negative infection. J Biol Chem 2000; 275(20): 14903.
18. De Vries B, Kohl J, Leclercq WK, et al. Complement factor C5a mediates renal ischemia-reperfusion injury independent from neutrophils. J Immunol 2003; 170(7): 3883.
19. Daemen MA, van ’t Veer C, Denecker G, et al. Inhibition of apoptosis induced by ischemia-reperfusion prevents inflammation. J Clin Invest 1999; 104(5): 541.
20. White SR, Williams P, Wojcik KR, et al. Initiation of apoptosis by actin cytoskeletal derangement in human airway epithelial cells. Am J Respir Cell Mol Biol 2001; 24(3): 282.
21. Van de Water B, Kruidering M, Nagelkerke JF. F-actin disorganization in apoptotic cell death of cultured rat renal proximal tubular cells. AmJ Physiol 1996; 270: F593.
22. van de Water B, Nagelkerke JF, Stevens JL. Dephosphorylation of focal adhesion kinase (FAK) and loss of focal contacts precede caspase-mediated cleavage of FAK during apoptosis in renal epithelial cells. J Biol Chem 1999; 274(19): 13328.
23. Bergin E, Levine JS, Koh JS, et al. Mouse proximal tubular cell-cell adhesion inhibits apoptosis by a cadherin-dependent mechanism. Am J Physiol 2000; 278(5): F758.
24. Israili ZH, Dayton PG. Human alpha-1-glycoprotein and its interactions with drugs. Drug Metab Rev 2001; 33(2): 161.
25. Moore DF, Rosenfeld MR, Gribbon PM, et al. Alpha-1-acid (AAG, orosomucoid) glycoprotein: interaction with bacterial lipopolysaccharide and protection from sepsis. Inflammation 1997; 21(1): 69.
26. Haraldsson BS, Johnsson EK, Rippe B. Glomerular permselectivity is dependent on adequate serum concentrations of orosomucoid. Kidney Int 1992; 41(2): 310.
27. Johnsson E, Haraldsson B. Addition of purified orosomucoid preserves the glomerular permeability for albumin in isolated perfused rat kidneys. Acta Physiol Scand 1993; 147(1): 1.
28. Hillegass LM, Griswold DE, Brickson B, et al. Assessment of myeloperoxidase activity in whole rat kidney. J Pharmacol Methods 1990; 24(4): 285.
29. Laight DW, Lad N, Woodward B, et al. Assessment of myeloperoxidase activity in renal tissue after ischemia/reperfusion. Eur J Pharmacol 1994; 292(1): 81.
30. Van Molle W, Libert C, Fiers W, et al. Alpha 1-acid glycoprotein and alpha 1-antitrypsin inhibit TNF-induced but not anti-Fas-induced apoptosis of hepatocytes in mice. J Immunol 1997; 159(7): 3555.
31. Sorensson J, Ohlson M, Bjornson A, et al. Orosomucoid has a cAMPdependent effect on human endothelial cells and inhibits the action of histamine. Am J Physiol 2000; 278(5): H1725.
32. Adamson RH, Liu B, Fry GN, et al. Microvascular permeability and number of tight junctions are modulated by cAMP. Am J Physiol 1998; 274: H1885.
33. Friedman GB, Taylor CT, Parkos CA, et al. Epithelial permeability induced by neutrophil transmigration is potentiated by hypoxia: role of intracellular cAMP. J Cell Physiol 1998; 176(1): 76.
34. Nakamura A, Imaizumi A, Yanagawa Y, et al. Beta2-adrenoceptor activation inhibits Shiga toxin2-induced apoptosis of renal tubular epithelial cells. Biochem Pharmacol 2003; 66(2): 343.
35. Riera M, Torras J, Cruzado JM, et al. The enhancement of endogenous cAMP with pituitary adenylate cyclase-activating polypeptide protects rat kidney against ischemia through the modulation of inflammatory response. Transplantation 2001; 72(7): 1217.
36. Keyler DE, Pentel PR, Haughey DB. Pharmacokinetics and toxicity of high-dose human alpha 1-acid glycoprotein infusion in the rat. J Pharm Sci 1987; 76(2): 101.
Keywords:

Ischemia-reperfusion; AGP; Inflammation; Neutrophil; Apoptosis

© 2004 Lippincott Williams & Wilkins, Inc.