Donor kidneys are cooled to 4°C after procurement to prevent injury. Early experiments in kidney preservation indicated that maintenance of a removed organ at body temperature resulted in significant injury, whereas cooling to 4°C prevented injury for long enough to permit a transplant (1). The beneficial effects of cooling the kidney to be transplanted are limited, however, and deceased-donor kidneys cannot be maintained at 4°C indefinitely. The risk of delayed graft function (DGF) increases progressively as cold ischemia time (CIT) increases (2). For every 6-hr increment of cold ischemia (CI), the risk of DGF increases by 23% (2). DGF is much more common in kidneys stored with a CIT of greater than 36 hr (3, 4) and is associated with worse long-term graft survival compared with kidneys that function immediately (3). Tubular cell death during human kidney transplantation is apoptotic during cold preservation (5, 6). During warm ischemia-reperfusion at implantation, both apoptotic and necrotic cell death occur (7). Tubular apoptosis, necrosis and brush border injury (BBI) during stages of hibernation is a focus of the present study. A number of investigators have suggested that hibernation represents a unique natural model of organ preservation and kidney transplantation (8, 9), where torpor can be viewed as a natural equivalent of donor kidney storage at 4°C, whereas arousal is a natural equivalent of the warm reperfusion that occurs during implantation of a transplant organ into a recipient. The caveat to these comparisons is that hibernators do not undergo ischemic brain death and there is obviously no alloimmune component during arousal when reperfusion occurs.
Mammalian hibernators undergo extreme reductions in core body temperature (CBT) every winter during torpor (Fig. 1). CBT can fall for several days from 38°C to as low as −3°C in arctic hibernators (10) or 2°C to 10°C in temperate-zone hibernators (11). During torpor, hibernators undergo profound physiological changes, reducing their heart rate from a summer-time level of 200 to 300 beats/min to 3 to 5 beats/min (12) and their respiratory rate from 100 to 200 breaths/min to 4 to 6 breaths/min (8). Torpor is periodically interrupted by rewarming to 36°C for approximately 12 hr, a period termed interbout arousal (IBA). The functional reason for arousal is not known (11, 13). During IBA, organs undergo reperfusion and rewarming to near normal levels (11). Thereafter, the hibernator reenters torpor (Fig. 1). Hibernators typically undergo cycles of torpor and arousal several times during the winter (11). We hypothesized that there would be differences in the kidney function and histology when kidneys from different stages of hibernation are exposed to cold and warm ischemia.
Caspases are a novel group of cysteine proteases that are major mediators of cell death and inflammation (14, 15). Both human and animal studies suggest that the adverse impact of CI may be associated with caspase-3 activation and apoptosis. Caspase-3 is the “executioner caspase” that is centrally important in apoptotic cell death in vivo (16). Apoptosis rate has been shown to correlate significantly with cold-ischemia time in human deceased-donor renal transplants (17). Preimplantation biopsies of human donor kidneys, which subsequently develop postoperative acute tubular necrosis, demonstrate increased apoptosis in renal tubular epithelial cells (5). Our previous work demonstrates that CI results in caspase-3 activation, tubular injury, and apoptosis in a model of CI in mice (6). Preservation of kidneys with a pancaspase inhibitor prevented caspase-3 activation and significantly reduced tubular injury and apoptosis. The effect of a prolonged reduction in CBT on tubular apoptosis and caspase activity in the kidney of hibernators is a focus of this study.
Complete protection from apoptotic or necrotic cell death would have major implications for organ preservation and transplantation (9, 11). A number of studies have examined ultrastructural changes that occur in the kidney at various stages of hibernation (9, 18–20) but none has examined the occurrence of apoptosis or necrosis. Whether renal function returns in a spontaneously aroused animal is not known and has important implications regarding the effect of torpor and arousal on renal glomerular and tubular function.
We sought to address these questions by examining kidneys taken from ground squirrels (GSs) at various stages of hibernation for apoptosis, necrosis, caspase activity, and BBI. We also determined the effect of hibernation on serum creatinine (SCr) and urinary concentrating ability. Finally, we hypothesized that kidneys removed from hibernating animals would tolerate ex vivo CI, whereas kidneys removed from nonhibernating summer animals would demonstrate apoptotic cell death and caspase activation after ex vivo CI similar to that observed in our previously published mouse model (6).
We have shown previously that CI increases caspase activity, apoptosis, and BBI in a mouse model of CI (6). Kidneys taken from hibernating animals during IBA had significantly increased caspase-3-like activity versus summer animals, whereas kidneys from hibernating torpid animals were not significantly different (Fig. 2A).
To determine whether the increase in caspase-3 results in apoptosis, kidneys were examined histologically. There was no significant difference in the number of apoptotic cells/high-power fields seen in the three groups. Apoptotic cell death was rare (Fig. 2B, C). Kidneys taken from hibernating animals during IBA had significantly increased BBI than summer animals (Fig. 3A). Representative kidney histology is shown in Figure 3(C). In addition to an increased number of tubules with BBI (Fig. 3A), the severity of BBI was also significantly worse in IBA (Fig. 3B). In contrast, hibernating torpid animals had virtually no BBI and displayed virtually no injury (Fig. 3C). In summary, caspase-3-like activity was significantly increased in IBA animals and was associated with BBI rather than apoptosis.
To determine the functional consequence of the BBI, we measured SCr and urine osmolality (Uosm) at number of different stages throughout the hibernation cycle. Mean SCr was significantly elevated in early arousal after prolonged CI. Mean SCr returned to baseline levels during IBA (Fig. 4A). Mean Uosm was highest during the summer, was significantly reduced during late torpor, and was subsequently increased during IBA (Fig. 4B). These findings suggest that despite BBI, IBA kidneys were able to clear SCr and concentrate urine.
Next, we hypothesized that kidneys from hibernating animals would be less susceptible to CI than those from summer animals. Kidneys from summer animals, hibernating torpid animals, and hibernating animals in IBA were perfused with cold University of Wisconsin (UW) solution at 4°C, removed, and then maintained in cold UW solution at 4°C ex vivo for 72 hr. The contralateral kidney perfused with cold UW solution at 4°C and processed immediately served as a control. Kidneys removed from summer, torpor, and IBA animals all tolerated ex vivo CI and no significant difference was found in the number of tubular apoptotic cells between ex vivo cold ischemic kidneys and controls (Fig. 5A–C). BBI significantly worsened only in kidneys removed from torpid animals and stored ex vivo (Fig. 5E). BBI was not significantly worse in kidneys removed from summer or IBA animals compared with in vivo kidneys (Fig. 5D, F). However, IBA kidneys had BBI at baseline (Fig. 3A).
To determine the mechanism of protection against severe CI both in vivo and ex vivo, we examined kidneys for caspase-3-like activity. Caspase-3-like activity was significantly increased in kidneys removed from summer animals (Fig. 5G) and hibernating IBA animals when stored ex vivo (Fig. 5I). The increase in caspase-3-like activity was much less than the 100-fold increase we previously described in a mouse model of CI (6). Caspase-3-like activity did not differ significantly between torpid control kidneys and torpid kidneys subjected to ex vivo CI for 72 hr (Fig. 5H).
In an analysis of 6465 deceased-donor kidneys, Salahudeen et al. (21) demonstrated that SCr levels at first hospital discharge and graft loss at 6-years posttransplant were significantly worse for kidneys with CIT more than 30 hr vs. CIT less than 20 hr. Prolonged CI is associated with DGF, apoptosis, BBI, and caspase-3 activation (5, 6). In comparison, torpid GS kidneys are naturally subjected to prolonged cold exposure far in excess of that tolerated by human deceased-donor kidneys. Studies of the protective mechanisms used by GSs may suggest novel clinical approaches for the treatment of prolonged CI and DGF.
We studied whether caspase activation, tubular cell apoptosis, and necrosis occurred in hibernating squirrel kidneys at two time points: in torpor, after prolonged CI, as a corollary of human organ preservation; and during IBA, when CBT returned to approximately 37°C and the kidney was reperfused, as occurs in human transplantation. Summer kidneys were used as controls. Despite prolonged cold preservation followed by warm reperfusion, frank necrosis and apoptotic cell death were not features of hibernating kidneys. The lack of tubular apoptosis after up to 12 days of a CBT of 4°C is remarkable considering multiple studies, which show extensive apoptosis after CI in kidneys (5, 6, 22) and other organs (23–25). BBI was absent in torpid kidneys, but was more prevalent, significantly more severe, and was associated with increased caspase-3 activity in IBA kidneys. BBI during IBA was likely sustained during the arousal process, when oxidative stress is known to occur (26).
To determine whether BBI affected kidney function, SCr and Uosm were measured at multiple time points throughout the hibernation cycle. Mean SCr peaked during early arousal and subsequently decreased to baseline levels in IBA, despite the presence of BBI. Uosm was significantly reduced during late torpor and then increased significantly in IBA animals. These findings indicate that IBA kidneys were able to clear SCr and concentrate urine despite the BBI. Return of renal function during IBA suggests that rewarming during hibernation might occur to permit renal clearance of waste products. Improvement of SCr to baseline indicates a functional resistance to prolonged CI and warm reperfusion that is not seen in human kidneys with DGF.
Hibernators tolerate dramatic reductions of CBT, heart, and respiratory rate that are lethal in nonhibernators. How this tolerance occurs has been speculated on since the late 1800s. The phenotypic change from summer animals to torpid hibernators suggests that a molecular genetic mechanism is operational (11). Plasma levels of putative antioxidants, such as ascorbic acid (27) and β-hydroxybutyrate (28), and the protective osmolytes glutamine and betaine (29) are increased during hibernation, suggesting the possibility of circulating factors that provide systemic protection. Fasting is associated with protection during hibernation. Fasting causes a shift from use of carbohydrates to fat as the primary fuel source, resulting in increased ketones (11). Increased levels of beta hydroxybutyrate are associated with increased hypoxic survival in GSs (30), stabilization of HIF-1α, and up-regulation of the antiapoptotic BCl-2 (31). Whether organ protection during hibernation requires perfusion for delivery of circulating factors or is intrinsic to the organ itself is unknown.
To assess whether the protection could be applied clinically, we subjected kidneys removed from GSs to ex vivo cold storage in UW solution (Fig. 5). Cold-stored kidneys also demonstrated a resistance to apoptotic cell death, which was associated with reduced caspase-3 activity.
For these protective mechanisms to be clinically relevant, they would have to be operational in nonhibernating conditions. Clinical application of hibernation studies includes the possibility of creating a hibernation-like condition in a human donor. There are numerous reports of humans having survived accidental hypothermia with full functional recovery, even after being declared clinically dead with a CBT as low as 13.7°C (32, 33). Therefore, the potential for cooling organs without incurring injury in human donors exists. There have been successful experimental attempts at creating hibernation-like states involving hypothermia. Safar and coworkers (34) induced cardiac death in dogs by complete exsanguination, followed by perfusion with normal saline at 2°C for up to 72 hr, and subsequent reinfusion of their blood. All animals made a full recovery despite having had no heartbeat or respiration. A further set of dogs subjected to trauma and hypothermia demonstrated a 50% survival rate. Induction of a hibernation-like state may be achieved biochemically (35, 36). Lee (36) induced hypothermia in mice using 5′ AMP, a pivotal metabolic signal involved in regulation of peripheral organ energy supply, and suggested the possibility that 5′ AMP could be used to manage CBT in humans during major surgery or trauma. Blackstone et al. (37) found that nonlethal doses of hydrogen sulfide in mice reduced CBT to 15°C. At this CBT, O2 consumption was approximately 10% of normal and respiratory rate decreased from approximately 120 to less than 10 breaths/minute. CBT and BMR normalized when the mice were returned to room air (37). Bos et al. (38) found that subtoxic doses of hydrogen sulfide significantly improved survival, renal function, apoptosis, and inflammation in a model of renal ischemia/reperfusion injury. Based on this evidence, investigators have hypothesized that creating a hibernation-like state in a human donor (35, 36) or donor-organ (39) is possible.
Although the specific genes responsible for inducing hibernation have yet to be determined, several mRNAs have been identified which are up-regulated during hibernation (11, 40), including cytochrome-c oxidase subunit 1 (41). Mitochondrial cytochrome-c oxidase subunit 1 may prevent damage to the mitochondrial electron transport chain due to CI, a finding of particular relevance to clinical transplantation because CI is known to induce mitochondrial injury in human kidney epithelial cells (42). Furthermore, Castaneda et al. (17) observed activation of mitochondrial apoptotic pathways in deceased-donor allografts. Therefore, it is possible that turning-on specific “hibernating” genes in a human donors or donor organs to create a hibernation-like state may prevent injury from CI and warm reperfusion. These possibilities demonstrate the potential for studies of hibernating mammals to yield improvements in clinical organ preservation and transplantation.
In summary, we observed complete protection from necrosis, apoptosis, caspase-3 activation in hibernating kidneys during prolonged CI, and normal kidney function during IBA. Our study suggests that a factor intrinsic to the kidney, perhaps related to the action of pro-apoptotic caspases, may be responsible for the protection against CI and warm reperfusion in hibernating squirrel kidneys. Inhibition of caspase activity, apoptosis, and necrosis in the human transplanted kidney, using novel and powerful caspase inhibitors or siRNA technology may be beneficial. An understanding of how protection occurs during hibernation may yield significant improvements in human organ preservation.
MATERIALS AND METHODS
Laboratory-bred 13-lined GSs (Ictidomys tridecemlineatus) of both sexes were purchased from D. Vaughan, University of Wisconsin, Oshkosh, in August. Squirrels were housed individually with free access to water and food (Teklad Global Cat Diet 2060, Harlan Laboratories, supplemented with sunflower seeds) at 20°C with a 14:10 hr light-dark cycle. Squirrels were held in these conditions for at least 5 days before use in experiments. In early September, squirrels were surgically implanted with temperature-sensitive radio telemeters (VM-FH disks; Mini Mitter, Bend, OR) for continuous sampling of CBT, allowed to heal for at least 2 weeks, and then transferred to a cold room maintained at 4°C after a stepwise lowering of ambient temperature over 1 week. The cold room was dark, and water and food were removed after squirrels began regular bouts of torpor as determined by remote CBT monitoring (every 2 min; MetaMeter system, Sable Systems International, Las Vegas, NV). The University of Colorado Institutional Animal Care and Use Committee approved all animal procedures.
Squirrels were euthanized by cardiac exsanguination under deep isoflurane anesthesia. During cardiac puncture, blood was drawn up into tubes containing 10 μL filter sterilized ACD (75 mM Na citrate, 38 mM citric acid,124 mM glucose) for each milliliter of blood. Summer squirrels were euthanized in early August when Tb was 37°C. Activity states during nonhibernation (October-March) were as follows: (1) entrance, entering torpor with Tb=23°C to 27°C; (2) early torpor, after Tb = 4°C for 5% to 10% of the time of the previous torpor bout; (3) late torpor, in torpor with Tb = 4°C for 80% to 95% of the length of the previous torpor bout; (4) early arousing, Tb between 7.0°C and 12.8°C during spontaneous rewarming from torpor; (5) late arousing, Tb between 18°C and 25°C during spontaneous rewarming from torpor; (6) IBA, 3 hr after the Tb inflection point indicated a return to euthermia; and (7) late IBA (Late—IBA), 5 hr after the inflection point with euthermic Tb.
Paraformaldehyde (4%)-fixed and paraffin-embedded kidneys were sectioned at 4 μm and stained with hematoxylin-eosin and periodic acid-Schiff (PAS) by standard methods. Histological scoring of BBI was performed by a renal pathologist in a blinded fashion as BBI per high power on PAS-stained fields (×400). The score was derived by quantifying the percent of tubules that displayed BBI per 10 high-power fields. Scoring was as follows: 0=no changes to brush border, 1=less than 10% of tubules affected, 2=11% to 25% of tubules affected, 3=26% to 45% of tubules affected, 4=46% to 75% of tubules affected, and 5=more than 75% tubules involved. Severity of BBI was scored on 10 randomly selected tubules. A score of 1 signifies less than 30% of the circumference of the analyzed tubule had BBI; a score of 2 indicates 30% to 60% of tubule had BBI; and a score 3 indicates more than 60% of the circumference of the tubule has BBI. The total score was calculated as mean BBI score per 10 tubules.
Morphologic criteria were used to count apoptotic tubular epithelial cells on hematoxylin-eosin and PAS staining as previously described (6). Apoptotic tubular epithelial cells were quantitatively assessed per high-power field (×400) by the renal pathologist in a blinded fashion. At least 10 high-power fields per slide were counted.
Ex Vivo CI Experiments
Caspase-3-like activity was determined by using fluorescent substrates as we have previously described (43, 44). Ac-Asp-Glu-Val-Asp-7-amido-4-methyl coumarin in 100% DMSO was used as a susceptible substrate for caspase-3-like activity. The reaction was then initiated by the addition of substrate.
SCr was measured using the alkaline picrate method (Jaffe reaction; Alfa Wasserman ACE, West Caldwell, NJ). Uosm was measured using the freezing point depression method with a freezing point osmometer (Model no. 3300; Advanced Instruments, Norwood, MA).
All values are expressed as mean±SD. For multiple comparisons, data were analyzed by ANOVA. For single comparisons, normally distributed data were evaluated using unpaired, two-tailed Student t tests, and nonnormally distributed data were analyzed by the nonparametric unpaired Mann-Whitney U test. A P value of less than 0.05 was considered statistically significant.
1.Calne RY, Pegg DE, Pryse-Davies J, et al. Renal preservation by ice-cooling: An experimental study relating to kidney transplantation from cadavers. BMJ
1963; 2: 651.
2.Ojo AO, Wolfe RA, Held PJ, et al. Delayed graft function: Risk factors and implications for renal allograft survival. Transplantation
1997; 63: 968.
3.Halloran PF, Hunsicker LG. Delayed graft function: State of the art, November 10–11, 2000. Summit meeting, Scottsdale, Arizona, USA. Am J Transplant
2001; 1: 115.
4.Lee CM, Carter JT, Alfrey EJ, et al. Prolonged cold ischemia
time obviates the benefits of 0 HLA mismatches in renal transplantation. Arch Surg
2000; 135: 1016.
5.Oberbauer R, Rohrmoser M, Regele H, et al. Apoptosis of tubular epithelial cells in donor kidney biopsies predicts early renal allograft function. J Am Soc Nephrol
1999; 10: 2006.
6.Jani A, Ljubanovic D, Faubel S, et al. Caspase inhibition prevents the increase in caspase-3, -2, -8, and -9 activity and apoptosis in the cold ischemic mouse kidney. Am J Transplant
2004; 4: 1246.
7.Solez K, Racusen LC, Marcussen N, et al. Morphology of ischemic acute renal failure, normal function, and cyclosporine toxicity in cyclosporine-treated renal allograft recipients. Kidney Int
1993; 43: 1058.
8.Deavers DR, Musacchia XJ. Water metabolism and renal function during hibernation
and hypothermia. Fed Proc
1980; 39: 2969.
9.Zancanaro C, Malatesta M, Mannello F, et al. The kidney during hibernation
and arousal from hibernation
. A natural model of organ preservation during cold ischaemia and reperfusion. Nephrol Dial Transplant
1999; 14: 1982.
10.Barnes BM. Freeze avoidance in a mammal: Body temperatures below 0 degree C in an Arctic hibernator. Science
1989; 244: 1593.
11.Carey HV, Andrews MT, Martin SL. Mammalian hibernation
: Cellular and molecular responses to depressed metabolism and low temperature. Physiol Rev
2003; 83: 1153.
12.Zatzman ML. Renal and cardiovascular effects of hibernation
and hypothermia. Cryobiology
1984; 21: 593.
13.Yan J, Barnes BM, Kohl F, et al. Modulation of gene expression in hibernating arctic ground squirrels. Physiol Genomics
2008; 32: 170.
14.Barinaga M. Death by dozens of cuts. Science
1998; 280: 32.
15.Fraser A, Evan G. A license to kill. Cell
1996; 85: 781.
16.Liu X, Zou H, Slaughter C, et al. DFF, a heterodimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis. Cell
1997; 89: 175.
17.Castaneda MP, Swiatecka-Urban A, Mitsnefes MM, et al. Activation of mitochondrial apoptotic pathways in human renal allografts after ischemiareperfusion injury. Transplantation
2003; 76: 50.
18.Anderson DG. Changes in renal morphology and renin secretion in the golden-mantled ground squirrel (Spermophilus lateralis
) during activity and hibernation
. Cell Tissue Res
1990; 262: 99.
19.Zimny ML, Franco EE, St Onge M, et al. Ultrastructure of juxtaglomerular cells correlated with biochemical parameters in a hibernator. Comp Biochem Physiol A Comp Physiol
1984; 78: 229.
20.Zimny ML, Rigamer E. Glomerular ultrastructure in the kidney of a hibernating animal. Anat Rec
1966; 154: 87.
21.Salahudeen AK, Haider N, May W. Cold ischemia
and the reduced long-term survival of cadaveric renal allografts. Kidney Int
2004; 65: 713.
22.Regner KR, Nilakantan V, Ryan RP, et al. Protective effect of Lifor solution in experimental renal ischemia-reperfusion injury. J Surg Res
2010; 164: e291.
23.Pileggi A, Ribeiro MM, Hogan AR, et al. Impact of pancreatic cold preservation on rat islet recovery and function. Transplantation
2009; 87: 1442.
24.Qian HJ, Du XJ, Zhang C, et al. Cold ischemia
time influences spermatogenesis in a testicular ischemia/reperfusion injury model. Transplant Proc
2010; 42: 1610.
25.Zheng S, Feng X, Qing D, et al. The tolerance time limits of biliary tracts of liver grafts subjected to warm ischemia and cold preservation: An experimental study in swine. Transplant Proc
2008; 40: 1629.
26.Ma YL, Zhu X, Rivera PM, et al. Absence of cellular stress in brain after hypoxia induced by arousal from hibernation
in Arctic ground squirrels. Am J Physiol Regul Integr Comp Physiol
2005; 289: R1297.
27.Drew KL, Osborne PG, Frerichs KU, et al. Ascorbate and glutathione regulation in hibernating ground squirrels. Brain Res
1999; 851: 1.
28.D'Alecy LG, Lundy EF, Kluger MJ, et al. Beta-hydroxybutyrate and response to hypoxia in the ground squirrel, Spermophilus tridecimlineatus
. Comp Biochem Physiol B
1990; 96: 189.
29.Serkova NJ, Rose JC, Epperson LE, et al. Quantitative analysis of liver metabolites in three stages of the circannual hibernation
cycle in 13-lined ground squirrels by NMR. Physiol Genomics
2007; 31: 15.
30.Caprette DR, Senturia JB. Isovolumetric performance of isolated ground squirrel and rat hearts at low temperature. Am J Physiol
1984; 247(4 Pt 2): R722.
31.Puchowicz MA, Zechel JL, Valerio J, et al. Neuroprotection in diet-induced ketotic rat brain after focal ischemia. J Cerebral Blood Flow Metab
2008; 28: 1907.
32.Lloyd EL. Accidental hypothermia. Resuscitation
1996; 32: 111.
33.Gilbert M, Busund R, Skagseth A, et al. Resuscitation from accidental hypothermia of 13.7 degrees C with circulatory arrest. Lancet
2000; 355: 375.
34.Nozari A, Safar P, Wu X, et al. Suspended animation can allow survival without brain damage after traumatic exsanguination cardiac arrest of 60 minutes in dogs. J Trauma
2004; 57: 1266.
35.Szabo C. Hydrogen sulphide and its therapeutic potential. Nature Rev Drug Discov
2007; 6: 917.
36.Lee CC. Is human hibernation
possible? Annual Rev Med
2008; 59: 177.
37.Blackstone E, Morrison M, Roth MB. H2S induces a suspended animation-like state in mice. Science
2005; 308: 518.
38.Bos EM, Leuvenink HG, Snijder PM, et al. Hydrogen sulfide-induced hypometabolism prevents renal ischemia/reperfusion injury. J Am Soc Nephrol
2009; 20: 1901.
39.Roth MB, Nystul T. Buying time in suspended animation. Sci Am
2005; 292: 48.
40.Epperson LE, Martin SL. Quantitative assessment of ground squirrel mRNA levels in multiple stages of hibernation
. Physiol Genomics
2002; 10: 93.
41.Hittel DS, Storey KB. Differential expression of mitochondria-encoded genes in a hibernating mammal. J Experiment Biol
2002; 205 (Pt 11): 1625.
42.Salahudeen AK, Joshi M, Jenkins JK. Apoptosis versus necrosis during cold storage and rewarming of human renal proximal tubular cells. Transplantation
2001; 72: 798.
43.Melnikov VY, Ecder T, Fantuzzi G, et al. Impaired IL-18 processing protects caspase-1-deficient mice from ischemic acute renal failure. J Clin Invest
2001; 107: 1145.
44.Melnikov VY, Faubel S, Siegmund B, et al. Neutrophil-independent mechanisms of caspase-1- and IL-18-mediated ischemic acute tubular necrosis in mice. J Clin Invest
2002; 110: 1083.