Abstract
The purpose of this study was to evaluate an intracellular solution with polyethylene glycol (PEG, molecular weight 20,000) as an impermeant, compared with University of Wisconsin (UW) and Euro-Collins (EC) solutions, after a 48-h cold storage (CS). The normothermic isolated perfused rat kidney (IPK) technique was used to assess renal function after CS. Five groups were studied: a control group (immediately reperfused, n = 10); one that received EC (n = 16); one that received UW (n = 16); and two that each received an intracellular (IC) solution, one with PEG (ICPEG, n= 16) and one without PEG (IC, n = 16). The perfusion flow rate was significantly greater in the PEG group and correlated with less significant cellular and interstitial edema and lower renal vascular resistance than in the IC, EC, and UW groups. Glomerular filtration rate was significantly higher in the PEG group during reperfusion than in the IC, EC, and UW groups. Proximal tubular functions were more efficient with PEG: fractional sodium reabsorption and total sodium reabsorption were significantly greater during reperfusion in the PEG group than in the IC, EC, and UW groups. Of greater interest is the protective effect of PEG on lipid peroxidation, which reflects ischemia/reperfusion damage. The second major effect is the dramatic ATP restoration during reperfusion, which outlines the preservation of oxidative phosphorylation after preservation by ICPEG. These results are supported by histological studies, particularly concerning brush border and mitochondrial preservation. Our results indicate that PEG is promising for cold ischemia and reperfusion injury protection.
The preservation of the harvested organ constitutes a prerequisite for organ transplantation. For kidney preservation, hypothermia storage remains the most common technique in use. However, hypothermic organ preservation is associated with oxygen deprivation, which inevitably leads to some degree of ischemia-reperfusion injury upon transplantation (Lehr and Messner, 1996). During the renal storage before transplantation, hypothermic swelling of the medullary thick ascending tubules results in a mechanical constriction of the peritubular capillaries and vasa recta (Yamamoto et al., 1984). During the reperfusion, a large amount of reactive oxygen agents (superoxide anions, hydroxyl radicals, and hydrogen peroxides) is produced by the re-entry of oxygenated blood in the ischemic tissue. Euro-Collins (EC) and University of Wisconsin (UW) solutions are the mainstay of therapy for hypothermic storage protection (Belzer et al., 1967; Collins et al., 1969; Collins and Halasz, 1975; Bonventre and Weinberg, 1992;Toledo-Pereyra and Rodriguez, 1994; Southard and Belzer, 1995). Nevertheless, the delayed graft function related to acute tubular necrosis (ATN) still remains an important complication after transplantation (Koning et al., 1995; Troppmann et al., 1995). Recent studies have demonstrated that ischemia and reperfusion contribute to the nonimmunological damage that complicates transplantation (Shoskes et al., 1990; Shackleton, 1998). ATN that develops as a result of warm and cold ischemia increases both post-transplant morbidity and the early loss of transplanted kidneys. The incidence of ATN following renal transplantation is often quite high, occurring in 10 to 40% of the recipients of cadaveric donor kidneys (Koning et al., 1995). Consequently, the improvement of quality preservation and the development of new preservation strategies could be an important therapeutic challenge.
The use of PEG in a preservation solution has been associated with a decreased incidence of rejection in clinical and experimental heart transplantation (Wicomb et al., 1990; Collins et al., 1991). In addition, the mode of action includes the prevention of osmotic swelling (Kober et al., 1996). The aim of this study was to assess the protective effect of PEG (molecular weight 20,000) on hypothermic swelling, on lipid peroxidation during reperfusion, on ATP content, and on ATP/Pi ratio as a reflection of the energetic status, as well as on recovery of an initial renal function and structure after a 48-h CS preservation when compared with EC and UW solutions. The IPK model was performed to avoid extrarenal influences. The method allows the study of immediate functional parameters, and the cold preservation and reperfusion effect can be isolated.
Materials and Methods
Animals and Surgical Technique.
Kidneys were obtained from male Wistar rats (250–300 g body weight). The animals were anesthetized by i.p. injection of pentobarbital (30 mg/kg). After a midline incision, the right ureter was cannulated for the collection of urine with a short polyethylene tubing (PE-10 tubing, 5 mm) connected to a larger polyethylene catheter (PE-50, 80 mm) to prevent ureteral back pressure. The right adrenal artery and the small lumbar arteries were tied off. After an i.v. injection of heparin (1000 U), a polyethylene catheter was inserted into the infrarenal aorta. The kidneys were immediately flushed in situ with 80 ml of a cold preservation solution (4°C) at a maximum pressure of 100 mm Hg. After excision, the kidneys were placed in a small container with 100 ml of the preserving solution and stored at 4°C for 48 h.
Preservation Solution and Experimental Groups.
Standard EC (Pharmacie Centrale des Hôpitaux de Paris), UW (DuPont Pharmaceuticals, Paris, France), and PEG combined with an intracellular solution (ICPEG) prepared in the biochemical department (DuPont Pharmaceuticals) were used. The composition of the three solutions used is shown in Table 1. The animals were divided into four groups: the EC group (EC preservation,n = 16), the UW group (UW preservation,n = 16), the ICPEG group (ICPEG preservation,n = 16) and the IC group (IC without PEG preservation,n = 16). The control group (n = 10), with neither cold ischemia nor flushing solution, was studied (immediate normothermic reperfusion of the right kidney).
The IPK Technique.
At the end of the hypothermic storage, the left renal artery was tied up, and the left kidney was removed for a lipid peroxidation or metabolic study. The right kidney was placed in a moist temperature chamber (Hugo Sachs Elektronic, March-Hugstetten, Germany). The moist chamber temperature was controlled during the perfusion. The kidney was perfused via the renal artery with warmed (5°C), freshly prepared Krebs-Henseleit solution containing bovine serum albumin (Sigma, St. Louis, MO) 5 g/100 ml, and (in mM): sodium, 140.0; potassium, 5.0; chloride, 123.0; bicarbonate, 25.0; calcium, 2.2; ionic calcium, 1.2; magnesium, 1.2; inorganic phosphates, 2.0; sulfates, 1.2; d-glucose, 11; pyruvate, 2.0; oxaloacetate, 1.0; l-sodium lactate, 5.0. To improve the stability of the preparation, an amino acids mixture was added using a commercial solution (g/100 ml: l-isoleucine, 0.7;l-leucine, 1.4; l-lysine, 1;l-methionine, 0.7; l-phenylalanine, 0.9;l-threonine 0.55, l-tryptophan 0.25,l-valine 0.7, l-arginine 1.5,l-histidine 0.4, l-aspartate, 0.3;l-cysteine, 0.2; l-glutamate, 0.5; glycine, 0.9; l-proline, 1.1; l-serine, 0.3;l-tyrosine, 0.04 (Clintec Nutrition Clinique, Sèvres, France). Polyfructosan (Inutest, Laevosan, Linz, Austria) was added (1 g/l) to determine the glomerular filtration rate (GFR). The pH was adjusted to 7.40 to 7.45 and the medium was continuously gassed with a prewarmed and moistened gas mixture (95% O2, 5% CO2). The pH in the prewarmed medium did not change during perfusion. The solution was perfused in a recirculatory perfusion system by draining back the venous effluent into the reservoir (Universal Perfusion System UP-100 Type 834, Hugo Sachs Elektronic). Two in-line filters (5-μm pore size) were used during the reperfusion. Perfusion was provided by an Ismatec peristaltic pump (Glattbrugs, Zürich, Switzerland). The perfusion flow rate (PFR) was adjusted to maintain renal arterial perfusion pressure (PP) at 100 mm Hg. PP was measured by a pressure transducer (Plugsys Equipment for constant perfusion pressure control, Hugo Sachs Elektronic). After the initiating perfusion, when a steady state had been attained, urine and perfusate samples were collected after 30, 60, and 90 min of reperfusion. PFR (ml/min/g) was measured (flowmeter module flow probe type 20N; Transonic, Ithaca, NY); GFR (μl/min/g) was evaluated from inulin clearance using this formula: GFR = urine flow rate × urine inulin/perfusate inulin. FRNa+ and TNa+ were calculated using this formula: FRNa+ = 100 × (1 − urine Na+/perfusate Na+ × perfusate inulin/urine inulin); TNa+ = (perfusate Na+ × GFR − urine Na+ × urine flow rate/1000). Polyfructosan was measured after acid hydrolysis by including a glucose 6-phosphate isomerase fraction into the assay. Na+ was analyzed by flame photometry.
Measurement of the Percentage of Tissue Water Content.
The tissue water content was determined in fresh tissue (n= 6) soon after laparotomy; in the control group after reperfusion; at 12, 24, and 48 h of CS, and after reperfusion. The kidney tissues were measured initially for the wet weight and then after 48 h in an oven at 100°C for the dry weight, thus allowing the determination of the water content. The percentage of renal water content was calculated as follows: renal water content = (1 − dry weight/wet weight) × 100 (%).
Metabolic Study and Measurement of Malondialdehyde (MDA) Tissue Level.
ATP levels were determined in the whole kidney by a bioluminescence technique (left kidney from control group and experimental groups after CS, and right kidney after reperfusion). After rapid removal of the kidney and immediate homogenization in 5% trichloroacetic acid, the acid-insoluble material was removed by cold centrifugation and the trichloroacetic acid extract carefully neutralized with NaOH. The homogenate was frozen until assay. ATP was measured by the bioluminescent firefly luciferase as described previously (Thore, 1979). ATP was measured in fresh kidneys cut off immediately after anesthesia, in kidneys cold-stored for 48 h, in kidneys from control group after reperfusion, and in cold-stored kidneys after normothermic reperfusion.
Fatty acid peroxidation was evaluated by renal tissue levels of MDA. Renal tissue levels of MDA were evaluated and measured with a thiobarbituric acid test. An aliquot of renal tissue homogenate was heated with thiobarbituric acid under acidic conditions and a pink color formation measured at 530 nm. To measure the formation of lipid-soluble Schiff bases, aliquot was removed after incubation and lipids extracted into chloroform:methanol and monitored for a fluorescence between 400 and 450 nm when excited at 360 nm. The protein content of the homogenates was determined by Lowry's method (Lowry et al., 1951). Lipid peroxidation markers and ATP levels were determined in fresh left kidneys cut off immediately after anesthesia, in kidneys cold-stored for 48 h, in control kidneys after 90 min of normothermic perfusion, and in cold-stored kidneys for 48 h after being reperfused for 90 min.
NMR Measurement.
After the equilibration period, isolated perfused kidneys were placed in 20-mm NMR tubes and introduced into the vertical magnet. During the experiment, the venous effluent was removed from the NMR tube by an aspiration line and recycled.31P NMR experiments were performed on a Bruker (Paris, France) AM400WB spectrometer at 92 MHz. Field homogeneity was achieved by shimming the water signal. Spectra were acquired with 4000 data points for a spectral width of 6000 Hz and 20-μs (90°) pulses before Fourier transformation. In each experiment, spectra were acquired during reperfusion at 30, 60, and 90 min (total acquisition time 6 min). Intracellular pH (pHi) was estimated by using the chemical shift of the Pi according to the formula: pHi = 6.7 + log[(δPi − 3.148)/(5.695 − δPi)] (Kost, 1990). The signal intensities of ATP and Pi were estimated by measuring the peak intensity on plotted spectra. The pHi was also determined after 12, 24, and 48 h of cold ischemia. The ATP/Pi ratio was calculated as a bioenergetic index during perfusion.
Histopathological Examination.
We have previously described a quantitative assessment, that is, a numerical score of ischemic proximal tubule injury (Goujon et al., 1999). Using several basic morphological patterns, without knowledge of the protocols of CS used, we found that such a method of grading system analyses allowed scores that correlated well with the functional study. At the end of the cold flush, after 48 h of CS and the reperfusion of the kidneys, samples were collected for light microscopy. Samples were fixed with Dubosq-Brazil (Laboratoire d'Anatomie Pathologie, Poitiers, France) and 10% formalin in 0.01 mM phosphate buffer (pH 7.42) and embedded in paraffin. Conventional stains were applied (hematoxylin and eosin and periodic acid, both from Schiff). Analysis of cellular ultrastructure using electron microscopy was also performed at the same time. Small pieces of renal tissue were fixed in 2.5% glutaraldehyde, washed and post-fixed in 2% osmium tetroxide for 2 h at 4°C. After dehydration with ethanol, they were embedded in araldite. Ultrathin sections were stained with uranyl acetate and lead citrate and then examined under an electron microscope (JEOL, Tokyo, Japan). Microscopic study was reviewed by a pathologist who did not know the preservation solution used. Light microscopic sections were examined for vacuolization, denuded basement membrane, necrosis, tubular dilatation, and cell detachment. Electron-microscopic sections were examined for mitochondria and tubular cell brush border integrity and interstitial and cell edema. Nine basic morphological patterns (apical cytoplasm vacuolization, tubular necrosis, denuded basement membrane, tubular dilatation, cell detachment, intracellular edema, interstitial edema, brush border integrity, mitochondria integrity) were graded on a five-point scale as follows: 1, no abnormality; 2, mild lesions affecting 10% of kidney samples; 3, lesions affecting 25% of kidney samples; 4, lesions affecting 50% of kidney samples; and 5, lesions affecting more than 75% of kidney samples.
Statistical Analysis of Results.
Results are expressed as means ± standard error and compared by using variance analysis and the Student-Newman-Keuls test for multiple comparison. When a nonparametric test was needed, the Kruskal-Wallis analysis was used. Significance was accepted at the 0.05 level of probability. Control group values are shown as reference values but are not used for statistical analysis.
Results
Perfusate Flow Rate.
After preservation, the kidneys from the ICPEG solution group showed significantly greater PFR during normothermic perfusion than in the UW and EC groups, and particularly the IC group (Fig. 1).
Glomerular Filtration and Proximal Tubular Function.
After 48 h of cold ischemia, renal function was reduced in all experimental groups. Kidneys from EC and UW showed significantly lower GFR than kidneys from ICPEG group (Fig.2A). TNa+ and FRNa+ was significantly reduced in the EC group when compared with UW and ICPEG groups (Fig. 2, B and C). The kidneys preserved with ICPEG solution had significantly higher FRNa+ and TNa+ than kidneys preserved with UW solution. Because the urine flow rate was insignificant, GFR, FRNa+, and TNa+ were not determined in the IC group.
Change in the Percentage of Renal Water Content.
The water content increased during preservation and reperfusion in the EC and particularly in the IC group. The water content decreased during preservation and increased rapidly during reperfusion in the UW group. However, the water content in the ICPEG was significantly lower than that in the IC and EC groups, and particularly the UW group (Fig.3, A and B).
Change in Renal Malondialdehyde Production and ATP Levels.
After 48 h of cold ischemia and normothermic reperfusion, MDA tissue levels showed a dramatic decrease in ICPEG when compared with other experimental groups, particularly the IC group (Fig.4, A and B). Values of energy compounds in fresh tissue and control isolated perfused kidneys was 1.39 ± 0.1 μmol/g, and 0.72 ± 0.1 μmol/g, respectively. After the cold preservation, ATP was 0.07 ± 0.01 μmol/g in the EC group, 0.10 ± 0.02 μmol/g in the UW group, 0.12 ± 0.01 μmol/g in the ICPEG group, and 0.04 ± 0.01 μmol/g in the IC group. After CS followed by reperfusion, ATP was significantly greater in the ICPEG group (0.28 ± 0.02 μmol/g) than in the IC group (0.07 ± 0.01 μmol/g; p < 0.01), in the EC group (0.12 ± 0.02 μmol/g; p < 0.01) and in the UW group (0.19 ± 0.02 μmol/g; p < 0.05).
Change in ATP/Pi Ratio and in pHi in the Kidneys.
The ATP/Pi ratio, in the control group, during reperfusion was 1.30 ± 0.15 after 30 min, 1.29 ± 0.2 after 60 min, and 1.28 ± 0.2 after 90 min. After 48 h of cold ischemia, the ATP/Pi ratio was greater in the ICPEG group than in the IC, EC, and UW groups during reperfusion (Fig. 5). Cold ischemia reduced pHi in all preserved groups (Table 2). However, in ICPEG, pHi was significantly greater during cold ischemia than in the IC, EC, and UW groups. During normothermic reperfusion, the kidneys preserved with ICPEG showed a significantly higher pHi after 48 h of CS than in kidneys preserved with IC, EC, or UW solution.
Histological Analysis.
Results are presented in Table3. In optical microscopy, after 48 h of CS, kidneys preserved with PEG solution presented localized vacuolization and necrosis. Cell detachment was moderate. Ultrastructural study showed no disruption of brush border and localized interstitial edema. After reperfusion, changes in tubular structure, cell detachment, and interstitial and cell edema were less significant with PEG solution than with UW solution and particularly EC solution. Disruption of brush border was more pronounced with EC solution. Mitochondrial depletion and disintegration were more significant with UW and particularly with EC solutions.
Discussion
Reactive oxygen species, which are generated during ischemia and reperfusion, play a major role in microvascular dysfunction and exert direct tissue damage, leading to a lipid peroxidative attack on fatty acids, denaturation of protein, polysaccharide depolymerization, and deoxyribonucleotide degradation. Although renal failure following cold ischemia and reperfusion can be controlled with dialysis, such injury influences the outcome of renal transplantation. Recent studies have demonstrated that the lipid peroxidation products, MDA and Schiff bases, are detectable in cold-stored kidneys (Gower et al., 1992). Moreover, it has been suggested that ischemia and reperfusion up-regulate major histocompatibility complex class I and class II expression (Shoskes et al., 1990; Lehr and Messmer, 1996; Shackleton, 1998.) Thus, postischemic allografts are more immunogenic (Bishop et al., 1988). PEG, particularly PEG molecular weight 20,000, has been found to improve tissues and organ preservation. Previous studies have demonstrated the efficiency of PEG in heart, liver, pancreas, and small bowel preservation in experimental models and clinical studies (Wicomb et al., 1990; Zheng et al., 1991; Tokunaga et al., 1992; Itasaka et al., 1994). PEG can strongly protect isolated hepatocytes and renal tubules against cold injury in in vitro models (Marsh et al., 1989;Southard et al., 1990). Otherwise, it is of interest that PEG strongly suppressed the formation of MDA that was induced by the addition of iron to isolated hepatocytes at 37°C (Mack et al., 1991). In addition, PEG appears to mitigate the immune response in both clinical heart and experimental liver transplantation (Collins et al., 1991,1992). It has been known for several years that antigens combined with PEG frequently manifest reduced antigenicity. These antigens become tolerogenic and may even reduce the immune response in previously sensitized animals (Lee et al., 1978; Savoca et al., 1984; Katre, 1990).
The medullary thick ascending tubules and the pars recta segments of the proximal tubules are particularly susceptible to hypoxia and CS (Bonventre and Weinberg, 1992). In this study, results show an improvement of glomerular and proximal tubular functions with a greater FRNa+ and GFR during reperfusion. GFR and PFR are dramatically increased with PEG solution during perfusion closely correlated with lesser renal vascular resistance. These data are related to the osmotic prevention of cell swelling and lower vascular endothelial damage. Otherwise, the tubuloglomerular feedback is less activated in the kidneys preserved with PEG solution. It is also possible that tubular alterations increase tubular pressure that restrains GFR. FRNa+ and TNa+, as indexes of tubular function, were significantly different among the cold-stored kidneys; it is evident that the PEG solution resulted in a higher FRNa+. We can speculate that this is related to increased Na+-K+-ATPase activity. These results are supported by histological data which show that medullary and cortical edema is reduced by PEG. Brush border is also better preserved after cold ischemia and reperfusion when compared with kidneys preserved in UW and particularly EC solution. Therefore, this improvement could not simply be attributed to prevention of osmotic swelling. Previous studies demonstrated that interaction between PEG and lipids in cell membranes might stabilize the membrane and render it less permeable to extracellular solution (Robinson, 1971; Daniel and Wakerley, 1976).
During normothermic reperfusion, there is no participation of leukocytes and the reperfusion injury in the kidneys is only due to the endothelial cells. In this study, the strong inhibition on MDA production is consistent with the effect of PEG as a powerful antioxidant and inhibitor of cyclooxygenase (Mack et al., 1991). Thus, adverse effects on the metabolism of arachidonic acid via cyclooxygenase and reactive oxygen species generation are limited by PEG solution in this model. Moreover, the extent of edema correlated well with the rate of lipid peroxidation as determined by MDA tissue levels.
Molecular weight of 20,000 is important. Recent study has demonstrated that PEG was less efficient as an impermeant than mannitol. However, in this study, PEG molecular weight was 600 (Lane et al., 1996). Benefits of PEG for heart, pancreas, liver, and small bowel transplantation have been reported with PEG 20,000 only (Wicomb et al., 1990; Collins et al., 1991; Zheng et al., 1991; Tokunaga et al., 1992; Itasaka et al., 1994). Among available colloids that can be used, PEG 20,000 has been previously shown to be more effective than albumin, dextran, gelatin, or PEG with a lower molecular weight (Daniel and Wakerley, 1976). Since multiorgan harvesting has emerged, UW solution has replaced EC as kidney preservation solution. In the present study, PEG solution is superior to EC and particularly to UW. In this solution, previous studies suggest a role for two of the pharmacologic additives, adenosine and reduced glutathione (Bizugas et al., 1990; Southard et al., 1990). However, during revascularization, these additive agents are rapidly depleted by the high amount of reactive oxygen agents produced in the ischemic tissue. In addition, as Evans et al. (1996)demonstrated recently, the t1/2 of reduced glutathione added to the UW fluid was about 1 day, which may be insufficient for a 48-h preservation. On the other hand, hydroxylethyl starch is biologically inert and would not appear to confer an identical advantage when compared with PEG (Wicomb et al., 1990). Finally, the high efficacy of PEG may be related to its capacity to act as an impermeant colloid. In addition, PEG can act as a scavenger of reactive oxygen metabolites, probably because the high molecular weight might protect endothelial cells.
To conclude, the results of the present study can be taken as evidence of the cytoprotective effect of the PEG solution, particularly against MDA production. It is also efficient to use this colloid to reduce the expansion of the interstitial space. During preservation, the influx of Na+ and Ca2+ induces a consecutive flow of water from interstitial space into the cells and results in translocation of water from the vessels to the interstitium. Consequently, the swelling of tubular epithelial and endothelial cells reduces the vascular lumen, and the local hemodynamic is disturbed particularly at the moment of reperfusion. The protective effect of PEG to prevent this cell swelling is correlated with the improvement of renal perfusion and function. However, renal function is still poor compared with control. These results are probably related to the effect of cold ischemia per se and the early period of evaluation when the impact at the biochemical level is more relevant than the study of the renal function. This colloid might be useful for kidney preservation when compared with EC and particularly UW solutions. Moreover, PEG solution is much cheaper than UW solution for kidney transplantation. However, these findings require confirmation in the renal transplant model.
Footnotes
-
Send reprint requests to: Dr. Thierry Hauet, INRA, Laboratoire de Transplantation Experimentale, Domaine du Magneraud, BP 52, 17700 Surgeres, France. E-mail:hauet{at}magneraud.inra.fr
-
This work was supported in part by the Ministère de la Recherche and the Conseil Régional Poitou Charente.
- Abbreviations:
- EC
- Euro-Collins solution
- ATN
- acute tubular necrosis
- CS
- cold storage
- FRNa+
- fractional sodium reabsorption
- GFR
- glomerular filtration rate
- IC
- intracellular solution without PEG
- ICPEG
- intracellular solution with PEG
- IPK
- isolated perfused rat kidney
- MDA
- malondialdehyde
- PEG
- polyethylene glycol
- pHi
- intracellular pH
- PFR
- perfusate flow rate
- PP
- perfusion pressure
- UW
- University of Wisconsin solution
- TNa+
- total sodium reabsorption
- Received August 21, 2000.
- Accepted October 24, 2000.
- The American Society for Pharmacology and Experimental Therapeutics