Abstract
Ischemia-reperfusion injury (IRI) after transplantation is a major cause of delayed graft function, which has a negative impact on early and late graft function and improve acute rejection. We have previously shown that polyethylene glycol (PEG) and particularly PEG 20M has a protective effect against cold ischemia and reperfusion injury in an isolated perfused pig and rat kidney model. We extended those observations to investigate the role of PEG using different doses (30g or 50g/l) added (ICPEG30 or ICPEG50) or not (IC) to a simplified preservation solution to reduce IRI after prolonged cold storage (48-h) of pig kidneys when compared with Euro-Collins and University of Wisconsin solutions. The study of renal function and medulla injury was performed with biochemical methods and proton NMR spectroscopy. Histological and inflammatory cell studies were performed after reperfusion (30–40 min) and on days 7 and 14 and weeks 4, 8, and 12. Peripheral-type benzodiazepine receptor (PBR), a mitochondrial protein involved in cholesterol homeostasis, was also studied. The results demonstrated that ICPEG30 improved renal function and reduced medulla injury. ICPEG30 also improved tubular function and strongly protect mitochondrial integrity. Post-IRI inflammation was strongly reduced in this group, particularly lymphocytes TCD4+, PBR expression was influenced by IRI in the early period and during the development of chronic dysfunction. This study clearly shows that PEG has a beneficial effect in renal preservation and suggests a role of PBR as a marker IRI and repair processes.
Delayed graft function (DGF), which manifests many characteristics of acute renal failure, remains an important complication affecting kidney allograft after transplantation. DGF is usually the result of initial IRI, which is the sum of the transient warm ischemic interval before or during removal from donor, cold ischemia associated with preservation and storage, ischemia occurring during the period of revascularization, and the events of postischemic reperfusion. Other manifestations such as a higher incidence of acute rejection have been suggested to link directly to IRI (Boom et al., 2000). Several studies supported the increasing evidence that indicates that rejection could be initiated by the response to injury (for a review, see Lu et al., 1999). There is now a growing recognition that long-term graft failure is at least in part influenced by the consequence of transplantation-related IRI.
Using an autotransplant pig kidney model, we previously demonstrated that polyethylene glycol (PEG) reduced inflammation and improved renal function 7 days after a 48 h-preservation and reperfusion (Hauet et al., 2000). Considering these findings, we have sought to further assess whether PEG can influence long-term renal function and long-term consequences of DGF after preservation. PEG, and particularly PEG 20M, has been found to improve tissue and organ preservation. The addition of PEG to the cooling solution preserved actin filament and microtubule morphology in hepatocytes during cold treatment at 4°C for up to 24 h (Stefanovich et al., 1995). The mode of action includes the prevention of osmotic swelling and formation of malondialdehyde (Mack et al., 1991; Ganote et al., 1997; Hauet et al., 2001). In addition, it has been shown that PEG 20M improved the viability of kidney and liver slices stored in a phosphate-buffered saline solution (Daniel and Wakerley, 1976). PEG was then shown to improve endothelial cell preservation and viability when compared with UW and EC (Killinger et al., 1992). Inclusion of PEG in a simplified UW solution (no pharmacological additives) gave superior results to standard UW in models of cardiac and pancreas transplantation (Wicomb et al., 1990;Zheng et al., 1991). More recently, it has been demonstrated that the microperfusion method associated with PEG provided excellent protection in long-term hypothermic heart preservation (Ferrera et al., 2000).Murad et al. (1999a,b) have shown that camouflaging antigens by cell surface pegylation of red cells impairs the activation of signals initiating T cell activation, cytokine production, and T cell proliferation. These authors have demonstrated that methoxy PEG-derivatization of peripheral blood mononuclear cells from human leukocyte antigen class II human donors abolished the allospecific MHC class II-mediated lymphocyte T cell activation (Murad et al., 1999a). In addition, it has been reported that PEG 20M seems to mitigate the immune response in small bowel, heart, and liver transplantation (Collins et al., 1991; Tokunaga et al., 1992; Itasaka et al., 1994).
Because mitochondria is a major target during IRI, any interference with mitochondrial function may have important consequences. The 18-kDa PBR protein is a ubiquitous high affinity cholesterol binding protein (Li et al., 2001) and is involved in numerous biological functions, including steroid synthesis, mitochondrial respiration, cell proliferation, and apoptosis (for a review, see Brown and Papadopoulos, 2001). Cholesterol is a critical component of plasma membranes, and within renal tubular cells, it exists in an approximate 1:5 ratio with plasma membrane phospholipids. Dysregulated cholesterol metabolism is a hallmark of the maintenance phase of experimental acute renal failure (for a review, see Zager et al., 2001). Recent studies have demonstrated the possible role of PBR in the organism response to tissue damage associated with inflammatory process (Trincavelli et al., 2002). PBR was also involved in proximal tubular injury/repair process in an angiotensin II-induced hypertension rat model (Bribes et al., 2002). Considering these functions of PBR, we were interested to assess renal (particularly the outer strip of the outer medulla) PBR protein expression in cold ischemia reperfusion condition.
Materials and Methods
Experimental Design and Operative Technique.
As previously described, we used an established model of autotransplant pig kidney model (Hauet et al., 2000). Briefly, following nephrectomy, kidneys were immediately flushed with preservation solution and stored at 4°C. After preservation, heterotopic autotransplantation was accomplished via the midline incision, using the Aorta and Vena cava. Ureteroneocystostomy was performed and a double pigtail stent was inserted to avoid urine retention, and the contralateral kidney was removed. This study was carried out as a prospective randomized trial with strict exclusion criteria, animals that died from causes other than renal failure during the 2-week follow-up period were excluded, and any animals that developed renal artery, vein, or ureteric problems were also excluded. All experiments were performed in accordance with the Guidelines of the French Agricultural Office and the legislation governing animal studies.
Preservation Solutions and Experimental Groups.
Standard EC (Pharmacie Centrale des Hôpitaux de Paris, Paris, France), UW (DuPont Pharmaceuticals, Paris, France), and PEG combined with an intracellular solution (ICPEG) or not (IC) were used. The composition of the preservation solutions used was as follows. EC: 15 mM KCl, 10 mM NaHCO3, 15 mM KH2PO4, 42,5 mM K2HPO4, 195 mM glucose, 10 mM Na+, 115 mM K+, pH 7.0, 355 mOsm; UW: 5 mM MgSO4, 25 mM KH2HPO4, 30 mM raffinose, 100 mM lactobionate, 5 mM adenosine, 3 mM glutathione, 1 mM allopurinol, 50 g/l hydroxyethyl starch, 25 mM Na+, 125 mM K+, pH 7.4, 320 mOsm; ICPEG30: 118 mM KCl, 5 mM NaCl, 25 mM NaHCO3, 1.20 mM MgCl2, 1.75 mM CaCl2, 11 mM glucose, 30 g/l PEG 20M, 118 mM K+, 30 mM Na+, pH 7.3, 340 mOsm; ICPEG50: 118 mM KCl, 5 mM NaCl, 25 mM NaHCO3, 1.20 mM MgCl2, 1.75 mM CaCl2, 11 mM glucose, 50 g/l PEG 20M, 118 mM K+, 30 mM Na+, pH 7.3, 395 mOsm; and IC: 118 mM KCl, 5 mM NaCl, 25 mM NaHCO3, 1.20 mM MgCl2, 1.75 mM CaCl2, 11 mM glucose, 118 mM K+, 30 mM Na+, pH 7.3, 300 mOsm. All kidneys were stored at 4°C for 48 h before being retransplanted. The pigs were assigned to one of seven groups: control and uninephrectomized (Nef) age-matched group (n = 5, respectively), EC (transplanted kidneys cold-flushed with the EC solution; n = 14), UW (transplanted kidneys cold-flushed with the UW solution; n = 14), IC (transplanted kidneys cold-flushed with the IC solution;n = 8), ICPEG30 (transplanted kidneys cold-flushed with the ICPEG solution; n = 14), and ICPEG50 (transplanted kidneys cold-flushed with the ICPEG solution; n = 12).
Renal Function.
Pigs were placed in a cage to allow specific 24-h urine collections. Endogenous creatinine clearance (CLCr; milliliters per minute), protein excretion in urine and fractional excretion of sodium (FENa+; percentage) were measured before kidney preservation and on postoperative days 1, 3, 5, 7, 11, and 14 (D1-D14) and weeks 4, 8, and 12. CLCr was calculated using the formula CLCr = urine flow rate × urine creatinine/plasma creatinine. FENa+ was calculated using the formula FENa+ = (urine Na+ × urine flow rate/plasma Na+)/glomerular filtration rate × 100. The Na+ level was measured by flame photometry, and creatinine was measured enzymatically with an automatic analyser (Kodak Ektachem 700 XR; Ortho, Paris, France). Urinary proteins were determined by a photometry method (Laboratoire Biorea, Talant, France).
NMR Experiments.
As previously described, urine and plasma samples from control and preserved kidneys were collected and frozen at −20°C until NMR measurements were performed (Hauet el al., 2000b). For urine NMR spectra, the ratio of trimethylamine-N-oxide (TMAO) was calculated and expressed in millimoles per mole of Cr. To assess oxidative metabolism and citrate synthase activity, the ratio of the citrate to Cr was also measured in urine and expressed in millimoles per mole of Cr. The occurrence of the TMAO resonance in plasma spectra (TMAOp) was noted only when it was intense enough to be separated from glucose resonances.
Histological Studies.
Kidneys were processed for light and electron microscopy. Tissue sections (5 μm) were examined by two pathologists blinded to the experimental conditions. After 40 min of reperfusion and on postoperative days 3, 7, and 14 and weeks 4, 8, and 12, biopsy tissue samples from the deep cortex-outer medulla region of the kidney were fixed in Dubosq-Brazil and 10% formalin in phosphate-buffered saline (PBS), embedded in paraffin and stained with hematoxylin and eosin. Small pieces of renal tissue were fixed in 2.5% glutaraldehyde, washed and postfixed in 2% osmium tetroxide for 2 h at 4°C, dehydrated in graded series of ethanols, and embedded in araldite for transmission electron microscopy. Ultrathin sections were cut and stained with uranyl acetate and lead citrate and were examined under an electron microscope (JEOL 100 CX; JEOL, Tokyo, Japan). Between 40 min and D14, light microscopic sections were examined for tubular necrosis and cells detachment, and using electron microscopy, mitochondria were also studied (membrane damage and crest reduction). The degree of histological lesioning was determined using a semiquantitative graded scale: 0, no abnormality; 1, mild lesions affecting less than 25% of kidney samples; 2, lesions affecting 25 to 50% of kidney samples; 3, lesions affecting 50 to 75% of kidney samples; and 4, lesions affecting over 75% of kidney samples. Between D14 and W12, the degree of interstitial fibrosis stained with Picro sirius was determined by a semiquantitative imaging technique. The percentage of Picro sirius stained surface were measured on five different tissue sections viewed at 100× magnification for each experimental condition and expressed as a percentage of the total surface area examined.
Immunohistochemical Studies.
Frozen and paraffined kidney biopsy sections (5 μm) from biopsies were processed for indirect immunocytochemistry using several mouse monoclonal antibodies (dilution 1:20). Sections were deparaffined, rehydrated, and heated in pressure cooker containing citrate buffer, pH 6, to boiling point for 2 min. Thereafter, sections were cooled down, rinsed in PBS, and processed for indirect immunofluorescence on sections incubated for 30 min at room temperature using the mouse anti-porcine MCA1218 macrophage/monocyte and neutrophils marker (IgG2b; Serotec Products, Oxford, UK), the mouse anti-porcine CD4 (MCA1749, IgG2b; Serotec Products), or the mouse anti-porcine CD8 (MCA1223, IgG2a; Serotec Products). We also used mouse monoclonal anti-porcine MHC class II (MCA1335, IgG2b; Serotech Products), mouse monoclonal anti-human VCAM1 (CD106, MCA 907B, IgG1; Serotec Products), and mouse monoclonal anti-human E-selectin (ELAM1, CD62E, IgG1; Chemicon, Temecula, California). VCAM-1 and E-selectin cross-react with pig. In all cases, the sections were rinsed in PBS and incubated with biotinylated anti-mouse Ig antibodies (DAKO, Copenhagen, Denmark) for 20 min (1:100) at room temperature, rinsed again, and incubated with alkaline phosphatase-conjugated streptavidin (DAKO). Alkaline phosphatase activity was revealed by staining with substrate solution of Fast Red (Sigma-Aldrich, St. Louis, MO) in Tris-buffered saline. In some cases, the sections were counterstained with hematoxylin. All sections were examined under blind condition and photographed. The number of MCA1218-, CD4+-, and CD8+-labeled cells per surface area (104/μm2) was counted on five different tissue sections in each experimental condition. Paraffined kidney sections (5 μm) from biopsies were processed for immunochemistry using a rabbit polyclonal antibody (Biotrim International, Dublin, Ireland; dilution 1/200) for 30 min at room temperature. The resultant intensity was quantified on a 0 to 4+ scale (4+ = strongest intensity). Immunolocalization of PBR was determined as previously described using an affinity purified anti-PBR peptide anti-serum raised against an amino acids sequence (amino acids 7–19; VGLTLVPPSLGGFMGAYFVR) conserved across species (Li et al., 2001). Paraffin-embedded sections were incubated with rabbit anti-PBR (1:400, dilution with 10% fetal bovine serum-PBS) for 1 h at room temperature. After rinsing the sections in PBS, horseradish peroxidase conjugated goat anti-rabbit IgG (Transduction Laboratory, Lexington, KY), diluted 1:500. PBR staining was determined as follows; 100 tubule sections were examined for positive staining, and the intensity of immunostaining was also quantified on a 0 to 4+ scale (0 = no staining to 4+ = dense). The degree of MHC class II, VCAM-1, and E-selectin staining was also determined semiquantitatively, estimated using a scale from 0 to 4+ (4+ = strongest intensity), and the number of MCA1218-, CD4+-, and CD8+-labeled cells per surface area (104/μm2) was counted on five different tissue sections for each of the experimental conditions.
Statistical Analysis.
Mean values were calculated for each group (mean ± S.E.M.) and compared for statistical significance by the unpaired t test or variance analysis and Student-Newman-Keuls for multiple comparison test. The unpairedt test was used for cellular infiltration in immunohistochemical analysis. The Mann-Whitney U test was used for histologic analysis and immunohistochemical data on the expression of adhesion molecules and cytokines. Differences at aP value of less than 0.05 were considered to be significant. Control group values are shown in the figures and tables as reference values, but they were not systematically used for statistical analysis.
Results
Effect of PEG on Prevention of Delayed Graft Function, Functional Results, and Survival after Reperfusion.
Total weights (control, 39 ± 2.6 kg; Nef, 42.5 ± 3 kg; EC, 43.2 ± 3.1 kg; UW, 44.5 ± 3.4 kg; ICPEG30, 42.5 ± 4.1 kg; ICPEG50, 43.6 ± 3.7 kg) and kidney weight (EC, 136 ± 6.2 g; UW, 139 ± 5.6 kg; ICPEG30, 137 ± 6.1 kg; ICPEG50, 135 ± 6.7 kg) were not significantly different between the control, Nef, and experimental groups. Survival was 100% in the control, Nef, and ICPEG30 groups. Five pigs in the EC group and four pigs in the ICPEG50 and UW groups developed acute renal failure confirmed by autopsy and histology (data not shown). They died on postoperative day 4 and 6 in the EC group, on postoperative day 6 and 10 in the UW group, and on day 6 and 9 in ICPEG50. All animals in IC group died between day 7 and 12 after autotransplantation. Diuresis was measurable in the control, Nef, and ICPEG30 group 24 h after autotransplantation. In contrast, prolonged anuria was noted in the EC, UW, ICPEG50, and particularly IC groups (<100 ml/24 from day 1 and 5 in the EC group and from day 1 and 3 in the UW and ICPEG50 groups; <30 ml in IC group). Consequently, the assessment of renal function was not very efficient before day 3 and 5 in the ICPEG50, UW, and EC groups, respectively. CLCr dramatically decreased, and FENa+ was higher in preserved kidneys in the EC, UW, and ICPEG50 groups than in the ICPEG30 group after cold preservation and autotransplantation, consistent with the decreased glomerular filtration rate and reabsorption of Na+ in renal tubules of damaged kidneys (Fig. 1, A and B). These findings suggested that PEG improved the glomerular function and Na+-transporters activity in reperfusion injury. However, 50g/l of PEG was not as efficient as 30g/l of PEG. During the first 2 weeks postsurgery, protein excretion was not significantly different between preserved groups. However, protein excretion was significantly higher in these groups than in the Nef and control groups (P < 0.01). Protein excretion became significantly higher in EC, UW, and ICPEG50 compared with ICPEG30-preserved kidneys between W4 and 12 (Fig. 1C). Protein excretion, CLCr, and FENa+ were not determined in the IC group because there was not production of urine from these kidneys after autotransplantation related to a primary nonfunctioning of these organs. These findings demonstrated that the addition of 30g/l of PEG to a simplified intracellular solution is associated with an improvement of tubular function in a more efficient manner than the preservation with EC, UW, or in the presence of 50g/l of PEG.
Effect of PEG on Renal Medulla Injury and Oxidative Metabolism.
For urine 1H NMR spectra, TMAO to Cr ratios were significantly higher in urine from kidneys preserved with EC, UW, and ICPEG50 than in the urine from kidneys preserved with ICPEG30 (Fig. 2A). This ratio was not determined in the IC group because the urine production was not measurable. In addition, according to the very low urine output before day 5 in the EC group and day 3 in the UW group, the determination of renal medullary cell injury was not strictly conceivable. In contrast, TMAOp peak intensity was significantly greater in the EC, UW, ICPEG50, and particularly the IC groups than in ICPEG30-preserved kidneys and the Nef and control groups (Fig. 2B). These findings demonstrated that PEG reduced, in a dose-dependent manner, the damage to the renal medulla. In addition, renal medulla cell injury was measurable by the determination of TMAOp by1H NMR spectroscopy in the absence of urine production.
Citrate was not detected during the 1st postoperative week in urine from kidneys preserved in EC. Its excretion was detected significantly earlier in urine from the ICPEG30-preserved group than from the UW and ICPEG50 groups (Fig. 2C). These findings suggested that PEG associated to a simplified intracellular solution strongly improves oxidative metabolism and reduces renal medulla injury in a concentration-dependent manner.
Effect of PEG on Preservation of Graft Morphologic Features and PBR Expression throughout the Follow-Up Period.
After a 40-min reperfusion, kidneys flushed and preserved with EC, UW, and ICPEG50 solutions showed a significantly higher graded score than kidneys flushed and preserved with ICPEG30 solution (Table1). At D14, tubular alterations remained more pronounced in the EC, UW, and ICPEG50 groups than in the ICPEG30 group (Table 1). The more intense histological alteration was noted in the IC group at sacrifice day.
Electron microscopy demonstrated that mitochondrial integrity was remarkably preserved in ICPEG30 after 40 min of reperfusion and at D7 (Table 1). The loss of mitochondrial integrity was marked in the IC group between the reperfusion and the sacrifice day. In the control and Nef groups, PBR staining was detected particularly in the different tubule segments located in the deep cortex and outer medulla (Fig.3, A and B), and mitochondria integrity was well preserved with a normal aspect of the membranes and crests (Fig. 3, C and D). Dramatically, the rarefaction of the PBR-positive tubules occurred near the inner stripe/inner medulla boundary (Fig. 3, E, F, and G) and continued in the inner medulla (Fig. 3H). After reperfusion, PBR immunoreactivity was not seen in the EC and IC groups (Fig. 3, I and K), and this absence of immunostaining was related to severe mitochondrial damage (Fig. 3, J and L). At the same time, PBR staining was slightly expressed in UW and ICPEG50 (Fig. 3, M and O), and mitochondrial integrity was improved when compared with EC and IC groups (Fig. 3, N and P). In contrast, PBR was strongly detected in ICPEG30 in the different tubule segments (Fig. 3, Q and R) and related with the increased number of mitochondria and the preservation of membranes and crests (Fig. 3, S and T). Evolution of PBR expression was marked by an increased of PBR staining intensity in EC, UW, ICPEG50, and particularly ICPEG30 by D7 postsurgery (Fig. 3, U to X). This increased staining was located within regenerating tubules. PBR staining remained absent in the IC group (data not shown). PBR staining intensity was improved in ICPEG30 (+++) when compared with the other experimental groups (+/++).
After D14, consistent with the results of the renal function study, ICPEG30 remarkably reduced the histological changes of chronic graft injury, above all tubular atrophy and interstitial fibrosis. As shown in table 2, by W4 and W8, the percentage of interstitial fibrosis areas from deep cortex-outer medulla was significantly reduced in the ICPEG30 group, and tubular atrophy was significantly less severe than in the EC, UW, and ICPEG50 groups. By W12, interstitial fibrosis (Fig.4, A, B, C, and D) and tubular atrophy were more intense in UW, ICPEG50, and particularly EC than in ICPEG30 (Fig. 4, E, F, G, and H). After W4, PBR staining in the EC, UW, and ICPEG50 groups was detected in regenerating tubules (Fig. 4, I, J, and K), and the intensity was strongly improved (++ to +++). PBR immunostaining was even more intense in collapsed proximal tubule (++++). PBR expression was detected in different tubule segments in ICPEG30 (Fig. 4L). PBR immunostaining was also detected in infiltrating mononuclear cells particularly in EC, UW, and ICPEG50 (Fig. 4, M, N, and O). In contrast, mononuclear cells staining was reduced in ICPEG30 (Fig. 4P). These results suggest that PEG (30g/l) has a remarkable protective effect on morphologic features throughout the follow-up period. In addition, these results suggest that PBR could be involved in tubular injury/repair process.
Influence of PEG on Inflammatory Response and MHC-class II, E-Selectin, and VCAM-1 Expression.
The number of CD4+ cells increased rapidly for 2 weeks after autotransplanting kidneys cold-stored in the EC, UW, and ICPEG50 solutions and then slowed to plateau up to 8 weeks following surgery (Fig. 5A). Only a few infiltrating CD4+ cells were detected in the kidneys from the ICPEG30 group. As a result, the number of CD4+cells was always 3 to 5 times greater in week-2 postoperative kidneys from the EC, UW, and ICPEG50 groups than in those from the ICPEG30 group. Thereafter, the number of CD4+ cells remained significantly higher (p < 0.05) between W8 and W12 post-transplanted kidneys from the EC, UW, and ICPEG50 groups than in those from the ICPEG30 (Fig. 5A). In contrast, the number of CD8+ cells was much lower and gradually increased to reach a plateau 4 weeks after surgery (Fig. 5B). Like the CD4+ cell count, the number of CD8+ cells was always significantly (1.5 to 2 times lower) in W2 and W12 post-transplanted kidneys cold-flushed with the 30-g PEG-supplemented preservation solutions than in kidneys flushed with the EC, UW, and ICPEG50 solutions. This means that concomitantly to reducing the expression of MHC class II, PEG almost completely prevented the rapid influx of CD4+cells and, to a lesser extent, the influx of CD8+infiltrating cells. Almost no infiltrating CD4+and CD8+ cells were detected in the remaining kidney from the Nef group, and no cell was detected in the control group (data not shown). A biphasic period of macrophage/monocyte infiltration occurred between 1 and 8 weeks after surgery. There were significantly fewer MCA1218-labeled cells in kidneys from the ICPEG30 group than in those from the EC, UW, and ICPEG50 groups (Fig. 5C). Thereafter, the number of MC1218-labeled cells decreased rapidly during the 2nd week after surgery and then gradually increased again in the EC, UW, and ICPEG50. The number of MC1218-labeled cells was reduced in the ICPEG50 and Nef groups. Like for the CD4+ and CD8+ cells, the number of MCA1218-labeled cells was significantly lower in post-transplanted kidneys cold-flushed with the ICPEG30 solution than with the EC, UW, and ICPEG50 solutions (Fig.5C). Consistent with these findings, E-selectin staining was strongly reduced in ICPEG30 from D7 to W12 (Fig.6, A–D; Table 3). Consistent with the E-selectin expression and cell infiltration, the expression of VCAM-1 in interstitial cells and/or tubular basement membranes gradually increased in post-transplanted kidneys over the 12 weeks following surgery. At D7, the expression was more important at the basal side of tubule cell sections in EC groups and reduced in UW and ICPEG50 (Fig. 6, E–H). VCAM-1 expression was reduced in ICPEG50. The evolution of VCAM-1 expression demonstrated that VCAM-1 expression is more pronounced from 2 weeks to 8 to 12 weeks postsurgery in the EC, UW, and ICPEG50 solutions compared with those stored in ICPEG30 (Table3). Kidneys cold-flushed with the EC, UW, and ICPEG50 solutions exhibited intense positive staining in epithelial tubule MHC class II positive cells (Fig. 6, I to K). In sharp contrast, fewer MHC class II positive infiltrative cells were detected in the kidneys cold-flushed with the 30g/l PEG-supplemented IC (Fig. 6L). Kidneys of the EC, UW, and ICPEG50 groups also displayed intense staining in the peritubular infiltrative and endothelial cells sections, and kidneys from ICPEG30 exhibited a reduced staining (Table 3). Because there were severe lesions of acute tubular necrosis, expression of E-selectin, VCAM-1, and MHC class II was not determined in the IC group between reperfusion and sacrifice.
Discussion
The impairment of organ function derived from IRI is still an important problem in solid organ transplantation. This critical alloantigen-independent risk factor in chronic graft dysfunction occurs during organ retrieval, cold storage, and transplantation. Furthermore, it has been demonstrated that an initial ischemic insult may increase the up-regulation of class II MHC molecules, thus increasing the immunogenicity of the ischemic organ (Shoskes et al., 1990). Because preservation techniques have their limitations in the prevention of reperfusion injury, various attempts have been made to decrease IRI. However, many of these interventions have failed to show any beneficial effect in human trials (Sheridan and Bonventre, 2000). The present model, using one the most attractive species for the future, could be crucial in understanding the complex interrelationship between IRI and graft dysfunction. A recent study has also demonstrated that cold ischemic trauma exerted a permanent detrimental effect on the graft (Tullius et al., 2000).
The results from this study clearly demonstrate that ICPEG30 reduces cold ischemia and reperfusion injury in a dose-dependent manner. Reductions of glomerular filtration have been attributed to activation of tubuloglomerular feedback as a result of enhanced delivery of solute to the macula densa, persistent vasoconstriction, and an increase in paracellular permeability resulting in “back-leak” of glomerular filtrate. In addition, the inability to retain sodium is a well described feature after postischemic acute renal failure and is associated with a down-regulation of the transcription of several Na+ exchanger transporters (Wang et al., 1998). In this study, renal function is significantly improved for kidney preservation in ICPEG30 when compared with ICPEG50.
Another line of evidence for the beneficial effect of PEG is the limitation of renal medulla damage, which leads to the reduction of TMAO excretion in urine. This osmolyte was precisely determined by1H NMR spectroscopy in experimental and clinical setting (for a review, see Neild et al., 1997). In the present investigation, the abnormal excretion of TMAO was particularly noted in the EC group after 48 h of cold ischemia and correlated with the morphologic damage. Citrate is usually considered a component of the tricarboxylic acid cycle and is a significant renal metabolic substrate. Urine citrate excretion is consistent with renal citrate synthase activity, which is a key enzyme of the Krebs' tricarboxylic acid cycle. Citrate synthase catalyzes the stereospecific synthesis of citrate from acetyl coenzyme A and oxaloacetate. The reduced citrate excretion in EC, UW, and ICPEG50 is correlated with a decreased renal parenchymal citrate synthase activity. This is correlated with the loss of mitochondrial integrity. Consequently, the early detection of citrate in urine from ICPEG30-preserved kidneys is related to an efficient functional recovery of the citric acid cycle and the oxidative metabolism compared with EC and UW solution.
Preservation of mitochondrial integrity in the ICPEG30 group is also correlated with the PBR immunostaining. Our data demonstrated clearly that PBR protein is involved in IRI as an indicator of kidney viability and mitochondria integrity after preservation in different solutions and is also involved in long-term tubule repair process. A previous study has observed that in hematopoietic cells, the ability of the cells to resist H2O2cytotoxicity correlated with the expression level of PBR (Carayon et al., 1996). Moreover, the resistance of the cells to H2O2 could be significantly increased by transfection with PBR. In addition, considering the function of PBR in cholesterol uptake and intra mitochondrial movement, it has been proposed that decreased PBR expression may lead to reduced levels of mitochondrial membrane cholesterol, which could define the ability of the cells to undergo apoptosis (for a review, see Brown and Papadopoulos, 2001).
The other major finding is the reduced infiltration of T cells within preserved kidneys with ICPEG30. A previous study demonstrated that the chemokine RANTES could act as an antigen-independent activator of T cells (Bacon et al., 1995). Oxygen free radicals, released during IRI, may provide the signal for T cell activation in the absence of antigen and several receptor-ligands interactions (e.g., CD28-B7 and CD40-CD40L) and may provide costimulatory signals to T cells (for a review, see Waaga et al., 2000). Recent evidence indicates that endothelial cells and leukocytes (particularly CD4+ T) are major pathogenic mediators in IRI (Burne et al., 2001; Dragun et al., 2001). Using T cell-deficient mice and an in vitro stimulation of the IRI condition model, Burne et al. (2001) demonstrated a reduction of tubular necrosis score in T cell deficient mice and an increased interaction between T cells and renal tubular epithelial cells in in vitro model. The present results are also consistent with the pathogenesis role of T lymphocytes in renal IRI and demonstrate in an in vivo model that adhesion between T cells and renal tubular epithelial cells may mediate or amplify postischemic renal damage. In a recent study, the positive impact of PEG on porcine activated endothelial, where PEG reduced E-selectin expression, was shown (Stuhlmeier and Lin, 1999). Using an in vivo model, we demonstrated the reduction of E-selectin and VCAM-1 expression. Consequently, PEG is efficient specifically on the cytokine-activated endothelial cells. Otherwise, ICPEG30, which preserves the morphology and integrity of preserved kidney and reduces endothelial activation, reduced expression of MHC class II in renal tubule cells. These MHC class II-positive cells are particularly important in IRI and related to an increased immunogenicity of the inflamed organ (Shoskes et al., 1990; Azuma et al., 1997). Consequently, ICPEG30 reduces one of the major negative effects of IRI that can influence the early and late outcome of the grafts. The limitation of inflammation is correlated with this effect. Further studies are necessary to assess the consequence on rejection in allotransplant model.
In summary, the major finding of the present study is that PEG is efficient in limiting endothelial damage and, probably, the activation of cytokine/adhesion molecules cascade. PEG is efficient when used at a dose of 30 g/l when compared with ICPEG50. This result outlines the impact of preservation solution viscosity during organ preservation. Our results also suggest that PBR could be a valuable marker of mitochondria and kidney viability in the 1st days after preservation. PBR could be also involved in long-term tubule injury/repair processes, particularly for the tubule segments located in the deep cortex and outer medulla via its involvement in cholesterol homeostasis and compartmentalization (Li and Papadopoulos, 1998). In addition, the involvement of PBR in the inflammation within damaged tissue could also demonstrate that PBR protein is a key factor in IRI processes, which could trigger and control renal regeneration.
Footnotes
-
This work was supported by grants from the Conseil Régional Poitou-Charente, the University Hospital and Faculty of Medicine of Poitiers, and the Etablissement Francais des Greffes (Paris, France) and from the National Institute of Child Health and Human Development (HD-37032) (National Institutes of Health, Bethesda, MD). Jean Pierre Faure and Thierry Hauet have contributed equally to this work.
-
DOI: 10.1124/jpet.102.033688
- Abbreviations:
- DGF
- delayed graft function
- IRI
- ischemia-reperfusion injury
- PEG
- polyethylene glycol
- EC
- Euro-Collins solution
- UW
- University of Wisconsin solution
- MHC
- major histocompatibility complex
- PBR
- peripheral-type benzodiazepine receptor
- ICPEG30
- polyethylene glycol added to a simplified intracellular solution (30g/l)
- ICPEG50
- polyethylene glycol added to a simplified intracellular solution (50g/l)
- IC
- simplified intracellular solution without polyethylene glycol
- Nef
- control and uninephrectomized
- CLCr
- creatinine clearance
- FENa+
- fractional excretion of sodium
- TMAO
- trimethylamine-N-oxide
- TMAOp
- TMAO resonance in plasma spectra
- PBS
- phosphate-buffered saline
- W
- week
- D
- day
- VCAM-1
- vascular cell adhesion molecule-1
- Received February 12, 2002.
- Accepted April 5, 2002.
- The American Society for Pharmacology and Experimental Therapeutics