INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ACUTE TUBULAR NECROSIS (ATN) and the ensuing renal failure induced by ischemia and reperfusion injury or sepsis remains a major cause of morbidity and mortality among patients in the intensive care unit (51). Ischemia-induced ATN has an attendant 30% mortality rate, and many survivors require dialysis (51). Indeed, this common clinical entity occurs during cardiopulmonary bypass (29), kidney transplantation (14, 47, 48), aortic bypass surgery (20), accidental or iatrogenic trauma (18), sepsis (49), hydronephrosis (45), and elective urological operations (11). Associated tangible and intangible costs are substantial. Although factors, such as ATP depletion, calcium dyshomeostasis, and free radical generation, undoubtedly contribute to the mechanisms of renal ischemia and reperfusion injury (32, 33, 39, 44, 50), it is becoming increasingly clear that local inflammation may also play an important role (7, 8, 14, 18, 19, 21, 27, 30, 32, 33, 38, 40, 48). Kelly and colleagues (21) recently demonstrated that intercellular adhesion molecule-1 (ICAM-1)-deficient mice are protected against ischemic renal injury. Takada and associates (47) showed that soluble P-selectin ligand attenuates postischemic neutrophil infiltration and injury. The kidney-initiated local/early signals that may be responsible for neutrophil infiltration and injury remain ill-defined; however, local production of tumor necrosis factor (TNF)- may be involved.

Peripheral monocytes infiltrating the kidney have traditionally been considered the primary source of renal TNF; however, recent evidence suggests that glomerular mesangial cells are an important additional source (13, 15). Lipopolysaccharide (LPS; endotoxin) induced isolated mesangial cells or glomeruli to produce TNF, even after rats were deprived of bone marrow-derived cells by whole body irradiation (11, 13). LPS-stimulated TNF production has been localized to glomerular mesangial cells (15, 19, 22, 23, 25, 44). Thus the kidney itself is capable of producing TNF in response to LPS, TNF, or interleukin (IL)-1. Oxidants released during the reperfusion of ischemic tissue stimulate transcription factors involved in TNF expression (26, 30, 35, 41). TNF is capable of upregulating its own production; promoting the synthesis of small inflammatory mediators, such as platelet activating factor and eicosanoids; as well as recruiting and stimulating various cells within the immune system (30). Although it is known that exogenous TNF induces renal cell apoptosis, glomerular endothelial damage, fibrin deposition, cellular infiltration, and renal failure (3, 6, 11, 21, 25, 27, 46, 47, 48), it remains unknown whether early endogenous renal TNF production contributes to neutrophil infiltration and injury after ischemia and reperfusion. Therefore, the purposes of this study were to 1) examine the time course of kidney TNF mRNA induction early after renal ischemia and reperfusion; 2) define the time course of renal tissue TNF protein expression after ischemia and reperfusion; 3) measure relative bioactivity of renal TNF after renal ischemia reperfusion injury; and 4) determine whether TNF neutralization has any salutary or deleterious effects on ischemia and reperfusion-induced TNF mRNA induction, protein expression and bioactivity, and/or renal neutrophil infiltration and injury.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Male Sprague-Dawley rats weighing 250-300 g were acclimated and maintained on a standard pellet diet for 1 wk before the initiation of experiments. Uninephrectomized rats (male, Harlan Sprague Dawley, Indianapolis, IN) were allowed to recover for 2 wk before use. Animals were anesthetized intraperitoneally with pentobarbital sodium, 30 mg/kg, before the experiments. The animal protocol was reviewed and approved by the Animal Care and Research Committee of the University of Colorado Health Sciences Center. All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals [DHEW Publication No. (NIH) 85-23, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20205].

Chemicals and reagents. Recombinant human TNF binding protein (BP) was kindly supplied by Dr. Carl Edwards (Amgen, Boulder, CO). TNF-BP is expressed in Escherichia coli as the four extracellular domains of the p55 TNF receptor linked together at the Fc portion of IgG (12). TNF-BP was diluted in normal saline containing 0.25% human serum albumin. Negative control kidneys received vehicle (0.25% human serum albumin) pretreatment with or without subsequent ischemia-reperfusion. Unless specifically mentioned, LPS and other chemicals were obtained from Sigma (St. Louis, MO).

Experimental groups and operative technique. The left renal pedicle was isolated and occluded for periods of 30 or 60 min. Reperfusion was then allowed to occur for 0, 1, 2, 4, 24, or 48 h. Sham-operated animals underwent identical surgical treatment, including isolation of the renal pedicle; however, subsequent occlusion of the pedicle was not performed. After reperfusion, the kidneys were removed and frozen in liquid nitrogen. The samples were stored at 70°C until further testing could be performed. All animals were killed after completion of the experiment. The animals were divided into the following experimental groups: 1) animals undergoing a sham operation (negative control; n = 8); 2) animals receiving a sublethal dose of LPS (0.5 mg/kg ip) 2 h before kidney removal (positive control; n = 6); 3) 15 min of ischemia alone (n = 5); 4) 30 min of ischemia alone (n = 5); 5) 45 min of ischemia alone (n = 5); 6) 60 min of ischemia alone (n = 5); 7) 30 min of ischemia followed by 1 h of reperfusion (n = 5); 8) 60 min of ischemia followed by 1 h of reperfusion (n = 5); 9) 60 min of ischemia followed by 2 h of reperfusion (n = 5); 10) 60 min of ischemia followed by 4 h of reperfusion (n = 5); and 11) 60 min of ischemia followed by 24 and 48 h of reperfusion (n = 5). To determine the biological significance of ischemia and reperfusion-induced TNF expression, TNF was neutralized (TNF-BP) in groups that demonstrated an increase in TNF protein expression and activity (n = 5/group). TNF-BP (160 µg) suspended in 0.5 ml PBS was administered intraperitoneally 15 min before injury. This dose was based on previous dose-response experiments that demonstrated that equivalent or lesser doses prevented LPS-induced or ischemia-induced myocardial TNF production and contractile suppression (30, 36, 38, 42, 43).

Tissue homogenization. A portion of each kidney was homogenized for the TNF- bioactivity and ELISA assays. Homogenization was performed after the samples had been diluted in four volumes of homogenate buffer [in mM: 10 HEPES (pH 7.9), 10 KCl, 0.1 EGTA, 1 DTT, and complete protease inhibitor tabs (Boehringer Mannheim, Indianapolis, IN)] using a vertishear tissue homogenizer. Renal homogenates were centrifuged at 3,000 g for 15 min, supernatant total protein concentration was quantitated using the Lowry assay, and supernatants were stored at 70°C until the TNF bioactivity and ELISA assays could be performed.

RT-PCR. Semiquantitative RT-PCR was used to assess renal TNF gene expression. Renal tissue was obtained from sham-operated controls and animals exposed to either LPS or an early time course of graded ischemia and reperfusion. Total RNA was extracted from the tissue by homogenization in Tri-Reagent (MRC, Cincinnati, OH), then isolated by precipitation with chloroform and isopropanol. Two micrograms of the isolated RNA was subjected to RT-PCR with reverse transcriptase, using random hexaoligonucleotides as primers (Perkin Elmer, Norwalk, CT). RT-PCR was carried out for 30 min at 42°C followed by enzyme inactivation at 99°C for 5 min. PCR SuperMix containing Taq DNA polymerase was used to amplify the cDNA obtained from the PCR. Primers previously described for the detection of rat TNF were used at an annealing temperature of 55°C. Glyceraldehyde phosphodehydrogenase (GAPDH) gene expression served as the loading control. The amplified products were separated in a 1.5% agarose gel containing 0.5× Tris-borate-EDTA, pH 8.3. PCR amplification products were quantified by scanning the gel photographs with an Apple Color One Scanner connected to a Apple Power MacIntosh using Ofoto 2.0 scanning software. Density of the the TNF and GAPDH mRNA bands were determined using National Institutes of Health image analysis program 1.59b4f (41). The data are presented as the ratio of the densitometric units of the TNF mRNA band to the densitometric units of the GAPDH mRNA band.

Renal tissue TNF protein expression and bioactivity. Kidney homogenate TNF content was determined by ELISA, and bioactivity was determined by WEHI-164 clone cytotoxicity assay. ELISA was performed by adding 100 ml of each sample (equal protein and tested in duplicate) to wells in a 96-well plate of a commercially available ELISA kit (R&D Systems). The antibodies used in this ELISA are not influenced by either the type 1 or type 2 TNF- receptors. According to the manufacturer, the detection limit of this assay was determined to be 15 pg/ml after statistical analysis of the ELISA results. Furthermore, the manufacturer has determined that the ELISA is highly specific for TNF-. Concentrations as high as 10⁶ pg/ml of IL-1, IL-3, IL-4, IL-6, IL-7, TNF-, IFN-, and granulocyte monocyte colony-stimulating factor (mGM-CSF) did not yield detectable cross-reactivity. Similarly, concentrations as high as 10⁴ pg/ml IL-1 or 10⁵ pg/ml IL-2 did not yield detectable cross-reactivity. TNF- ELISA was performed according to the manufacturer's instructions. Final results were expressed as picograms of TNF- per gram of protein. TNF- bioactivity in the kidney samples was determined by using the WEHI-164 cell clone cytotoxicity assay as previously described (1, 2, 31-34). Final results were expressed as units of TNF- activity per gram of total protein.

Renal tissue myeloperoxidase. Myeloperoxidase (MPO) is an enzyme specific for neutrophils and is an accepted index of neutrophil infiltration. Renal samples obtained after sham operation, LPS administration, and 60 min of ischemia followed by varying hours of reperfusion were evaluated. The tissue was homogenized for 30 s in 4 ml of 20 mM potassium phosphate buffer, pH 7.4. The samples were centrifuged for 30 min at 40,000 g at 4°C (Beckman L-80 Ultracentrifuge, Beckman Instruments, Palo Alto, CA). The supernatant was discarded, and the pellet was resuspended in 4 ml of 50 mM potassium phosphate buffer, pH 6.0, with 0.5 g/dl cetrimonium bromide. The samples were then sonicated for 90 s (ultrasonic homogenizer, Cole-Parmer Instruments, Chicago, IL) and incubated for 2 h at 60°C. Homogenates were centrifuged at 14,000 g for 10 min. The supernatant was decanted, and 25 µl was added to 725 µl of 50 mM phosphate buffer, pH 6.0, containing 0.167 mg/ml o-dianisidine and 5 × 10⁴% hydrogen peroxide. The change in absorbance was measured spectrophotometrically (Beckman DU7 spectrophotometer, Beckman Instruments) at 460 nm. One unit of MPO activity was defined as the quantity of enzyme degrading 1 mmol peroxide/min at 25°C.

Renal functional analysis. To avoid the confounding, compensatory creatinine clearance of the contralateral kidney, a separate group of rats, having undergone prior right nephrectomy, were used. Renal function was determined by measuring serum creatinine immediately before and 48 h after the onset of reperfusion. The rats were killed after the second serum creatinine determination. The experimental groups included 1) animals undergoing a sham procedure; 2) animals given 0.5 ml NaCl iv 30 min before injury (30 min of ischemia followed by reperfusion); 3) animals given 0.5 ml NaCl iv 15 min before injury (60 min of ischemia followed by reperfusion); 4) animals given 160 µg TNF-BP suspended in 0.5 ml PBS iv before injury (30 min of ischemia followed by reperfusion); and 5) animals given 160 µg TNF-BP suspended in 0.5 ml PBS 15 min before injury (60 min of ischemia followed by reperfusion).

Statistical analysis. Data are presented as mean values ± SE (n = 5-8 animals/group). Differences at the 95% confidence level were considered significant. The experimental groups were compared using ANOVA with post hoc Bonferroni-Dunn (StatView 4.0, Berkeley, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Time course of renal TNF mRNA induction during ischemia and reperfusion. Renal samples were obtained after an early time course of graded ischemia and reperfusion. Sham-operated animals did not demonstrate any TNF mRNA induction (Fig. 1, A and B). TNF mRNA was detectable during early ischemia (30 and 45 min) alone, yet undetectable at 60 min of ischemia alone. After 60 min of ischemia and during reperfusion, TNF mRNA was detectable at 30 min, but not 15 or 60 min of reperfusion. Densitometric analysis of TNF mRNA expressed as percent of GAPDH mRNA are shown in Fig. 1B. After 30 and 45 min of ischemia, the TNF mRNA expressed represented 12 ± 2.7 and 6 ± 1.4% of GAPDH mRNA (P < 0.05 vs. sham), respectively. After 60 min of ischemia and 30 min of reperfusion, the TNF mRNA expressed represented 15 ± 4% GAPDH mRNA (P < 0.05 vs. sham).

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Time course of renal tissue tumor necrosis factor (TNF) mRNA induction during ischemia and reperfusion. A: gel photograph depicting TNF and glyceraldehyde phosphodehydrogenase (GAPDH) mRNA bands at various time points. B: densitometric analysis of mRNA at bands of corresponding time points, represented as TNF mRNA percent of GAPDH mRNA. During ischemia, renal tissue TNF mRNA is detectable at 30 and 45 min of ischemia alone. At 60 min of ischemia, TNF mRNA was not detectable; however, during reperfusion, TNF message was again induced and detectable at 30 min, but not at 15 or 60 min, of reperfusion. * P < 0.05 vs. sham.

Time course of renal TNF protein expression during ischemia and reperfusion. The time course of renal tissue TNF protein expression during renal ischemia and reperfusion is shown in Fig. 2A and are represented as picograms TNF per milligram total protein. Untreated and sham-treated animals demonstrated very low renal tissue TNF expression (14 ± 2.2 and 19 ± 1.1 pg/g protein, respectively). LPS (0.5 mg/kg ip) served as the positive control, and TNF levels in this group measured 57 ± 9.9 pg/mg protein (P < 0.05 vs. sham). At 30 and 60 min ischemia alone renal tissue TNF did not increase and measured 18 ± 6 and 24 ± 7.8 pg/mg protein, respectively. Furthermore, 30 min of ischemia followed by 1 h of reperfusion did not increase renal tissue TNF levels, which measured 28 ± 3.5 pg/mg protein. However, 1 h of ischemia followed by 1 h of reperfusion resulted in a significant increase (P < 0.05 vs. sham) in renal tissue TNF (41 ± 2.5 pg/mg protein). This increase was sustained at 2 h of reperfusion (48 ± 13 pg/mg protein) and declined after 4 h of reperfusion (19 ± 3 pg/mg protein).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Time course of renal tissue TNF protein expression (A) and bioactivity (B) during ischemia and reperfusion. Ischemia alone did not result in significant TNF protein expression or bioactivity (cytotoxicity); however, after 1 h ischemia and 1 or 2 h of reperfusion there was a significant increase in TNF protein expression and bioactivity. TNF protein expression returned to baseline levels at 4 h reperfusion. TNF-binding protein (TNF-BP) significantly attenuated () maximal cytotoxicity observed after 1 h ischemia and 2 h reperfusion (B). Lipopolysaccharide (LPS)-induced TNF protein expression and bioactivity served as positive controls. * P < 0.05 vs. sham.  P < 0.05 vs. 1 h ischemia and 2 h reperfusion.

Effect of TNF-BP on renal TNF protein bioactivity (cytotoxicity). In parallel to the TNF protein levels determined by ELISA, TNF bioactivity (cytotoxicity) increased in the LPS (19.5 ± 4 U/mg protein), 1 h ischemia-1 h reperfusion (7 ± 2.3 U/mg protein), and 1 h ischemia-2 h reperfusion (11 ± 2.4 U/mg protein) groups (P < 0.05 vs. sham = 2.3 ± 0.9 U/mg protein). However, TNF-BP significantly decreased (P < 0.05) renal TNF bioactivity in the 1 h ischemia-2 h reperfusion group to 4.9 ± 1 U/mg protein (Fig. 2B).

Effect of TNF-BP on neutrophil infiltration (MPO). Ischemia and reperfusion-induced neutrophil infiltration into the kidney was determined by renal tissue MPO assay (Fig. 3). MPO levels of TNF-BP-treated and vehicle-treated sham-operated animals were 5.2 ± 1.4 and 4 ± 1 U/g, respectively. MPO levels remained unchanged at 1 h ischemia alone, 1 h ischemia-1 h reperfusion, and 1 h ischemia-2 h reperfusion (3.9 ± 1.2, 6 ± 2.7, and 7 ± 1.9 U/g, respectively). TNF-BP pretreatment did not effect MPO levels in these three groups (4.9 ± 1.6, 4.4 ± 1.7, and 6.5 ± 1.3 U/g, respectively). However, 1 h ischemia-4 h reperfusion resulted in a significant increase in renal tissue MPO levels to 24 ± 4 U/g (P < 0.05 vs. sham), which was significantly decreased (P < 0.05 vs. sham) when animals received prior TNF-BP (14+2.5 U/g). Extended reperfusion times (1 h ischemia-24 and 48 h reperfusion) resulted in a decrease in MPO levels in both the TNF-BP-treated and untreated groups.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Renal tissue myeloperoxidase (MPO), quantitative measure of kidney neutrophil content, during ischemia and reperfusion with/without prior TNF neutralization with TNF-BP. After 1 h of ischemia and 4 h of reperfusion there was a significant increase in infiltration of neutrophils into kidney, which waned after 48 h. TNF-BP significantly reduced neutrophil infiltration observed after 1 h of ischemia and 4 h of reperfusion. * P < 0.05 vs. sham.  P < 0.05 vs. ischemia and reperfusion group of corresponding time point.

Effect of ischemia and reperfusion, with and without TNF-BP, on renal function. To determine the biological significance of kidney TNF production and neutrophil infiltration, the effect of TNF-BP on serum creatinine was determined after two different degrees of injury severity (Fig. 4, A and B). Forty-eight-hour serum creatinine levels in sham-operated animals were no different than untreated controls (0.36 ± 0.024 mg/dl; Fig. 4, A and B). TNF-BP did not affect renal function of sham-operated or untreated animals (0.39 ± 0.027 mg/dl; Fig. 4A). Thirty minutes of ischemia and 48 h of reperfusion (reversible injury model; Ref. 47) impaired renal function (Fig. 4A), resulting in a significant increase (P < 0.05 vs. preischemia/reperfusion or sham) in serum creatinine from 0.34 ± 0.024 to 1.17 ± 0.198 mg/dl; Fig. 4A). TNF-BP completely abolished renal dysfunction in this group (P < 0.05 vs. 30 min ischemia-48 h reperfusion) as demonstrated by the serum creatinine level of 0.513 ± 0.04 mg/dl (P > 0.05 vs. preischemia/reperfusion or sham; Fig. 4A). Renal function after a more severe model (irreversible injury; Ref. 47) of renal ischemia and reperfusion injury is shown in Fig. 4B (1 h of ischemia and 48 h of reperfusion). After this injury, serum creatinine levels increased from 0.38 ± 0.02 to 5.2 ± 1.137 mg/dl (P < 0.05 vs. preischemia/reperfusion or sham; Fig. 4B). In contrast to the less severe (more reversible) model of ischemia and reperfusion, TNF-BP failed to improve renal function after this injury (serum creatinine = 4.5 ± 1.186 mg/dl; Fig. 4B).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Serum creatinine after ischemia and reperfusion in uninephrectomized rats. Serum creatinine in uninephrectomized rats reflects creatinine clearance capability of remaining kidney. Serum creatinine was determined, with and without prior TNF-BP administration, in either a "reversible" (A; 30 min ischemia-48 h reperfusion) or an "irreversible" (B; 60 min ischemia-48 h reperfusion) model of ischemia and reperfusion injury. TNF-BP reduced postischemic renal insufficiency in reversible, but not irreversible, model of injury. * P < 0.05 vs sham or pre-ischemia-reperfusion.  P < 0.05 vs. 1 h ischemia and 48 h reperfusion.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The results of this study constitute the initial demonstration that 1) renal ischemia, with or without reperfusion, induces renal tissue TNF mRNA and protein production and increases TNF bioactivity in the kidney; 2) TNF-BP limits ischemia and reperfusion-induced renal tissue TNF bioactivity; 3) renal TNF contributes to neutrophil infiltration and injury early after ischemia and reperfusion; 4) TNF-BP attenuates reversible, but not irreversible, injury.

Significant renal dysfunction and cellular degeneration occurs after short periods of renal ischemia (21, 27, 46, 47, 48). Although the exact mechanisms of ischemia-reperfusion injury remain undefined, accumulating evidence suggests that local, early TNF production may play a role in the pathogenesis of this injury (5, 14, 40, 47). In this paradigm, resident renal macrophages and renal parenchymal cells may produce TNF early after the onset of ischemia and reperfusion. Early local TNF production has paracrine and autocrine effects, which may serve to upregulate local adhesion molecules that contribute to neutrophil rolling, sticking, infiltration, and, ultimately, renal injury (21, 46, 47). Although recent studies have demonstrated that even cold ischemia and reperfusion induces neutrophil infiltration and renal dysfunction (21, 46, 47), early signaling mechanisms remain undefined. The present study provides evidence that clearly implicates TNF in early neutrophil infiltration and renal dysfunction after ischemia and reperfusion injury.

TNF is a proinflammatory cytokine capable of upregulating its own expression as well as the expression of other genes pivotal to the inflammatory response (4, 10, 11, 15, 30, 49). Furthermore, exposure of renal tissue to TNF causes significant cellular damage and dysfunction (3, 11, 14). Although increased serum TNF levels have been detected after renal ischemia and reperfusion injury (14), renal tissue TNF bioactivity or mRNA levels early after ischemia have not been reported. TNF mRNA was induced within 30 min of the onset of renal ischemia (Fig. 1, A and B), and significant protein levels and bioactivity occurred at 1 and 2 h of reperfusion (Fig. 2, A and B). Ischemia alone may induce TNF mRNA and protein via the generation of reactive oxygen species (30, 37, 39). Indeed, key signaling elements involved in TNF mRNA and protein production, i.e., nuclear factor kappa B and p38 mitogen-activated protein kinase, are activated by oxidant stress (26, 30, 37, 39, 41). Reperfusion of ischemic tissue imposes an oxidant burden in which the reduction product of molecular oxygen, hydrogen peroxide, contributes to injury (8, 11, 17, 30, 37, 39, 41, 44, 45, 48). Activation of oxidant-sensitive enzymes involved in TNF production represents an additional mechanism by which oxidant stress induces cellular damage. The results of this study also confirm early findings regarding ischemia and reperfusion-induced neutrophil infiltration (9, 24, 28). Similar to the findings of Takada and colleagues (47), neutrophil infiltration peaked at 4 h after renal pedicle unclamping and reperfusion (Fig. 3). Because ATN clearly occurs in the absence of neutrophils, their role in the development of acute renal failure has remained controversial. However, evidence obtained from several recent investigations suggests that neutrophil activation and infiltration contribute to renal injury after ischemia and reperfusion (6, 9, 21, 24, 25, 28, 46-48). Indeed, reperfusion of ischemic kidneys with blood containing neutrophils worsens ischemic injury (24). Convincing evidence implicating the neutrophil in this injury was provided by Kelly and colleagues (21), who demonstrated that prior neutrophil depletion with antineutrophil serum protected against ischemic renal failure in mice. The results of this study further substantiate the association between neutrophil infiltration and injury; i.e., when renal neutrophil infiltration was attenuated with TNF-BP, postischemic renal function was improved. Furthermore, they indicate that TNF may be a key mediator of this process. Recent investigations suggest that, in other models of injury, TNF plays a key role in adhesion molecule, ICAM-1 and vascular cell adhesion molecule-1 (VCAM-1), upregulation, and neutrophil influx (6, 21, 25, 46-48). TNF-deficient mice are protected against the crescentic glomerulonephritis induced by antiglomerular membrane antibody (25). LeHir and colleagues (25) reported that TNF-deficient mice did not demonstrate the ICAM-1 or VCAM-1 upregulation or the neutrophil infiltration characteristic of this injury (25). Chana and Wheeler (6) showed that the administration of exogenous TNF had a profound effect on renal mesangial cell ICAM-1 and VCAM-1 expression. TNF upregulated ICAM-1 expression by 505% and upregulated VCAM-1 expression by 179%, thereby enhancing mesangial cell-monocyte binding by 335% (6). These investigations provide further support for the proposed model of injury in which early renal TNF expression induces neutrophil infiltration and injury.

Ischemia and reperfusion-induced renal TNF expression may result in renal cell injury via at least two distinct mechanisms: 1) direct cytotoxicity (induction of dysfunction and/or apoptosis; Ref. 27) and 2) neutrophil-mediated tissue injury (11, 44). Ischemia and reperfusion-induced renal TNF production was associated with impaired renal function (Fig. 4). In the reversible injury model (47), 30 min ischemia and 48 h reperfusion, TNF-BP restored creatinine clearance (Fig. 4A); however, in the irreversible injury model, 1 h ischemia and 48 h reperfusion (47), TNF-BP failed to affect creatinine clearance (Fig. 4B). These data suggest that TNF may be more involved in those early mechanisms of injury that mediate reversible dysfunction, rather than mechanisms such as calcium dyshomeostasis, which mediate necrosis and permanent injury (50).

The results of this study should be interpreted with several important caveats. This model of ischemia and reperfusion does not directly replicate clinical situations in which the kidney is rendered ischemic by combinations of low, and no, flow. Furthermore, protective strategies, which, in some instances, represent clinical standards, were not employed. Hypothermia and cytoprotective agents (e.g., Euro-Collins solution) may provide a degree of protection that is not advanced by TNF-BP. However, it should be noted that recent studies by Takada and colleagues (46-48) have demonstrated neutrophil infiltration and injury despite hypothermic preservation during ischemia. The warm ischemia and reperfusion model was used to 1) limit treatment variables and 2) relate our findings to those clinical situations (e.g., trauma, cardiopulmonary bypass, aortic bypass surgery, renal angioplasty) in which hypothermic preservation is not employed. Moreover, it should be noted that these findings do not indicate that the well-established mechanisms of renal cell injury are superseded by TNF-mediated injury. Other inflammatory signaling mechanisms are likely involved. Other mediators, such as IL-1, were not examined in this study; however, Haq and colleagues (16) reported that IL-1 may not be involved in renal injury after ischemia and reperfusion. Finally, the effects of posttreatment were not examined. It remains to be determined whether TNF-BP's protective effects will be limited to pretreatment situations. Nevertheless, these results may 1) help define the mechanistic role of TNF production in neutrophil infiltration and renal failure after ischemia and reperfusion and 2) suggest that the aforementioned clinical situations, which allow pretreatment, may be putative therapeutic targets for TNF-BP adjunctive therapy.

Renal ischemia and reperfusion injury is encountered in many clinical situations that are uniquely suited to pretreatment; i.e., kidney transplantation, partial nephrectomy, renal artery angioplasty, cardiopulmonary bypass, aortic bypass surgery, accidental or iatrogenic trauma, sepsis, hydronephrosis, and elective urological operations. Renal failure after these conditions remains a devastating problem. Improved understanding of cellular and molecular mechanisms of injury may enhance therapy. The findings presented in this report demonstrate that renal tissue TNF mRNA, TNF protein, TNF bioactivity, and renal neutrophil content are increased early after ischemia and reperfusion. Furthermore, these results reveal that TNF is partly responsible for the renal dysfunction that occurs during reversible injury. TNF-BP, a specific inhibitor of TNF activity, prevents the development of renal dysfunction after moderate degrees of warm ischemia. As the signaling cascade for renal TNF production becomes elucidated, the development of an anti-TNF therapy that attenuates ischemia-reperfusion-induced renal injury may become a realistic clinical possibility.