Renal cytokine profile in an endotoxemic porcine model

Acta Anaesthesiol Scand 2008; 52: 614–620 Printed in Singapore. All rights reserved

r 2008 The Authors Journal compilation r 2008 The Acta Anaesthesiologica Scandinavica Foundation ACTA ANAESTHESIOLOGICA SCANDINAVICA

doi: 10.1111/j.1399-6576.2008.01625.x

Renal cytokine profile in an endotoxemic porcine model A. GRANFELDT1, L. EBDRUP1, E. TØNNESEN1 and L. WOGENSEN2 1

Department of Anesthesiology and Intensive Care Medicine and 2Research Laboratory for Biochemical Pathology, Aarhus University Hospital, Aarhus, Denmark

Introduction: In animals exposed to acute endotoxemia with lipopolysaccharide (LPS), high levels of cytokines are found in the kidney. The objective of this study is to determine whether the high renal content of TNF-a, IL1b, IL-10 and IL-1 receptor antagonist (IL-1ra) is due to glomerular filtration and reabsorption, or whether the cytokines are produced locally in the kidney. Methods: Eighteen anesthetized and mechanically ventilated pigs (35–43 kg) were randomized into two groups: Group 1 (n 5 12) LPS infusion for 360 min and Group 2 (n 5 6) control pigs, no treatment. At 360 min, the pigs were euthanized and tissue samples from the kidneys were obtained. Localization of the cytokines was determined by immunohistochemistry and double immunofluorescence (dIF). Results: Pigs exposed to endotoxemia showed increased accumulation of leukocytes and increased protein expression of TNF-a and IL-1b when compared with controls. dIF showed that TNF-a-positive cells co-localized with both endothelial and mesangial cells in the glomeruli. Further-

more, the endothelial cells of the cortical arterioles were positive for IL-1b. TNF-a and IL-1b staining were absent in renal tubular cells. A positive signal for IL-10 was detected at the tubular brush border while IL-1ra was detected in the glomerulus and in the tubular cells. Conclusion: LPS-induced endotoxemia increased TNF-a and IL-1b protein expression and leukocyte accumulation in the kidneys. The results indicate that the increased levels of the pro-inflammatory cytokines TNF-a and IL1b are caused by a local production in the kidneys while the anti-inflammatory cytokines IL-10 and IL-1ra are filtrated and reabsorbed in the tubuli.

A

in renal blood perfusion is not the sole mechanism behind AKI. During sepsis, cytokines are released into the circulation. High amounts of the pro-inflammatory cytokines TNF-a and IL-6 have been detected in the kidney in a porcine model of endotoxemia (7). TNF-a and IL-1b stimulate the production of other cytokines like IL-6 and IL-8 and other inflammatory mediators such as endothelin, PAF, PGE2 and iNOS. TNF-a or IL-1b, when injected into animals, elicits a decline in the glomerular filtration rate (GFR) and induces renal damage (8). Furthermore, TNF-a and IL-1b activate endothelial cells and circulating leukocytes manifested as increased expression of adhesion molecules. This up-regulation contributes to the adherence of leukocytes to the vascular endothelium, followed by transmigration into the extravascular space with a release of inflammatory mediators contributing to organ damage (1, 9).

kidney injury (AKI) is a severe complication in critically ill patients and sepsis is the most important predisposing factor. AKI develops in up to 51% of patients with septic shock, and the mortality reaches 70% (1). Sepsis with AKI is associated with increased length of ICU stay and increased mortality when compared with non-septic AKI, suggesting a different pathophysiology behind septic and non-septic AKI (2, 3). The pathogenesis behind AKI is complex. One of the main theories is that renal vasoconstriction, followed by a decrease in blood flow results in ischemia and acute tubular necrosis (4). However, AKI still occurs in endotoxemic animals with a hyperdynamic circulation and increased renal blood flow (5). In a recent publication, the placebo group from the PROWESS trial was analyzed with the purpose of identifying the predictors of AKI. Neither mean arterial pressure, vasopressors nor the cardiovascular SOFA score were associated with AKI (6). These reports suggest that a decrease

614

CUTE

Accepted for publication 8 January 2008

Key words: Cytokines; endotoxemia; swine; acute kidney failure; immunohistochemistry. r 2008 The Authors Journal compilation r 2008 The Acta Anaesthesiologica Scandinavica Foundation

Endotoxemic porcine model

In animal studies, abrogating the effects of TNFa by using either receptor-deficient mice or by administrating neutralizing antibodies, kidney injury is reduced. In addition, the transplantation of TNF receptor-1-deficient kidneys into normal mice abrogated kidney injury. These results indicate that TNF-a is a major player in the pathophysiology of AKI (10, 11). The production of pro-inflammatory cytokines is accompanied by an almost simultaneous production of anti-inflammatory cytokines like IL-10 and IL-1 receptor antagonist (IL-1ra). IL-10 reduces the production of pro-inflammatory cytokines, neutrophil infiltration and the synthesis of iNOS, while the essential function of IL-1ra is competitive inhibition of the IL-1 receptor (12, 13). The kidneys are the most important catabolic organ in handling plasma cytokines. Cytokines and their degradation products are detectable in the urine, and in nephrectomized animals the half-life of cytokines is dramatically increased (14, 15). In vitro studies have shown that mesangial cells, glomerular endothelial cells and different types of leukocytes are capable of producing cytokines (16). However, the contribution from either glomerular filtration or intrarenal production in vivo is not clarified. Lipopolysaccharide (LPS)-induced endotoxemia in pigs elicits a pro- and anti-inflammatory cytokine response detectable in peripheral blood and at the organ level (7, 17). The objective of the present porcine study of acute endotoxemia was to determine the localization of TNF-a and IL-1b and to resolve whether the cytokines originate from glomerular filtration and reabsorption, or whether the cytokines are produced locally in the kidney. Furthermore, the aim was to investigate the presence and localization of the anti-inflammatory cytokines IL-10 and IL-1ra.

Methods Preparation of animals All pigs were fasted overnight, but were allowed free access to water. All pigs were premedicated with intramuscular s-ketamine (5 mg/kg) and midazolam (0.5 mg/kg) before intubation and mechanical ventilation. Anesthesia was induced with s-ketamine (5 mg/kg) and midazolam (0.5 mg/kg) intravenously and maintained thereafter by continuous infusion of midazolam (6 mg/kg/h) and fentanyl (60 mg/kg/h). Arterial, central venous and

Swan–Ganz catheters were inserted, and the pigs were monitored continuously. The study was approved by the National Committee on Animal Research Ethics.

Experimental protocol Eighteen anesthetized and mechanically ventilated pigs (35–43 kg) were randomized into two groups: Group 1 (n 5 12): LPS infusion for 360 min and Group 2 (n 5 6): Control pigs, no treatment. LPS (Escherichia coli 026:b6) infusion was initiated with increasing doses over 30 min (infusion rate 2.5 mg/ kg/h increasing with 2.5 mg every 5 min until 15 mg/ kg/h), followed by a sustained infusion of 2.5 mg/ kg/h for 330 min. All pigs received isotonic saline, corresponding to 16–20 ml/kg/h. At 360 min, pigs were euthanized by quickly eviscerating the heart.

Renal function Renal function was monitored by plasma creatinine, urinary output and GFR. GFR was measured using 51Crom-EDTA determined by a bolus injection of 5 MBq, followed by a sustained infusion of 1.5 MBq/h. The radioactivity in the samples was measured using a g-counter.

Tissue preparation Immediately after evisceration of the heart, tissue samples were taken from both the cortex and the medulla of the kidney and placed in 4% formaldehyde, embedded in paraffin and stained with hematoxylin and eosin, for histological evaluation. Leukocyte infiltration was evaluated in six animals per group. The number of leukocytes was counted in 10 randomly selected glomeruli and the area of each glomerulus was calculated by Olympus anas lySIS software in two tissue samples for each animal. Another set of tissue samples were embedded in optimal cutting temperature compound (Sakura Finetek Europe, BV, Zoeterwoude, Netherlands) and snap frozen in liquid nitrogen (
180 1C). Tissue samples were cut using a Frigocut 2800 cryostat (Reichert-Jung, Nussloch, Germany). The cryosections (4 mm thick) were then mounted onto slides and stored at
80 1C until analysis.

Immunohistochemistry (IHC) The sections were air dried for 20 min, fixed in acetone (
20 1C) for 5 min and washed in phos-

615

A. Granfeldt et al.

phate-buffered saline (PBS). Endogen peroxidases were blocked with a 1% hydrogen peroxide solution dissolved in methanol, whereas endogenous biotin was blocked using a streptavidin/biotin blocking kit (Vector Laboratories, Peterborough, UK). The protein expression of TNF-a and IL-1b was determined by IHC using anti-porcine TNF-a (12.5 mg/ml) and IL-1b antibodies (17.5 mg/ml) (R&D Systems, cat. no AF690 and cat. no MAB681, respectively; Minneapolis, MN, USA), followed by incubation with biotinylated antigoat IgG (1 : 200) or biotinylated anti-mouse IgG (1 : 200) (Vector Laboratories) for 30 min. Vectastain ABC kit standard peroxidase from Vector Laboratories was used as an amplification system with 3,3 0 -diaminobenzidine tetrahydrochloride (SigmaAldrich, St. Louis, MI, USA) as a substrate. As negative controls, the primary antibody was substituted with PBS buffer and when relevant substituted with IgG from the same animal as the primary antibody. The protein expression of TNF-asand IL-1b was quantified using Olympus analySIS software. The threshold for a positive signal was defined by measuring the staining intensity (positive pixels) in four positive control slides. Pixel intensities equal to and above this level were regarded as positive. The software then calculated the positive staining area as well as the total area of 10 randomly selected glomeruli in six slides per group. The positive staining area was expressed as a percentage of the total area for each glomerulus. The investigator was blinded to the treatment.

Single immunofluorescence (IF) IL-10 and IL-1ra protein expressions were evaluated using single IF with anti-porcine IL-10 (25 mg/ ml) and IL-1ra (20 mg/ml) antibodies (R&D Systems, cat. no. MAB6932 and cat. no. AF780, respectively). These slides were thereafter incubated with biotinylated anti-mouse IgG (1 : 200) or biotinylated anti-goat IgG (1 : 200) (Vector Laboratories) for 30 min, followed by streptavidin-conjugated Alexa Fluor 488 (Invitrogen, Carlsbad, CA, USA).

of the mesangial cells (Biotechnology, Santa Cruz, CA, USA/Dako) (18, 19). Slides were incubated with the secondary antibodies, Rhodamine-labelled donkey anti-goat IgG (Santa Cruz), Texas Red-labelled goat anti-rabbit IgG (Santa Cruz) or Texas Red-labelled goat antimonkey IgG (Santa Cruz).

Statistics The leukocyte counts for each tissue sample fitted Poisson’s distributions with means proportional to the area of the glomeruli. The total count per tissue sample divided by the total area of glomeruli of the tissue sample was square root transformed (root normalized count) to approximate a normal distribution. The root-normalized counts were compared across treatment groups using a mixedmodel ANOVA with animal as a random factor. The expression of TNF-a and IL-1b was also square root transformed to become normally distributed. They were analyzed with a mixed-model ANOVA with animals as a random factor. One control pig was tested as an outlier with the Dean and Dixon test and excluded. All values are reported on original scale mean  SEM.

Results Hemodynamic data and renal function All pigs survived; only one received adrenalin for resuscitation. LPS infusion caused a hyperdynamic circulation with increased cardiac output, and renal dysfunction estimated as declining GFR, urine output and increased creatinine concentrations (data not shown).

Leukocyte infiltration Infiltration of leukocytes was observed in the glomeruli and in the peritubular spaces. Furthermore, leukocytes were attached to the endothelial surface of intertubular veins and capillaries. Leukocyte accumulation in the glomeruli (cells/ mm2) was significantly higher in the LPS-treated pigs (97  20) vs. the control group (23  10, Po0.05).

Double immunofluorescence (dIF) To identify the TNF-a- and IL-1b-containing cells, dIF analysis was performed, using the Von Willebrand factor (Dako, Glostrup, Denmark) as a marker of endothelial cells and Desmin as a marker

616

IHC TNF-a and IL-1b Infusion of LPS induced a significantly increased protein expression of TNF-a (4.5  1.1% vs. 1.1  0.6%, Po0.05) and IL-1b (7.5  0.8% vs.

Endotoxemic porcine model

Fig. 1. Immunohistochemical (a, b) and single immunofluorescence (c, d) staining on frozen kidney sections. Immunohistochemical positive staining appears brown and IF positive staining appears green. (a) TNF-a positive staining in the glomerulus ( 100); (b) IL-1b positive staining in the glomerulus ( 100); (c) IL-10 positive staining at the brush border of the cortical tubulus ( 250); (d) IL-1ra positive staining in the cortical tubular cells ( 100).

Fig. 2. Double immunofluorescence on frozen kidney sections, glomerulus. (a) TNF-a Alexa s Fluor 488 (green) ( 100); (b) Desmin Rhodamine (red) ( 100); (c) Composit image, co localization appears yellow ( 100); (d) IL-1b s Alexa Fluor 488 (green) ( 100); (e) VWF Texas Red (red) ( 100); (f) Composit image, co localization appears yellow ( 100).

1.1  0.3%, Po0.05) when compared with the control group. As evluated by IHC, positive signals for TNF-a were primarily detected in the glomeruli whereas immunopositive signals for IL-1b were found both in the glomeruli and in the cortical arterioles (Fig. 1a and b). Furthermore, a few TNFa- and IL-1b-positive cells were located in the peritubular spaces corresponding to infiltrating leukocytes. TNF-a and IL-1b staining were absent in the medulla and in the renal tubular cells.

Single IF IL-10 was only detected at the brush border of the tubular cells, while IL-1ra was detected sporadically scattered in the glomeruli and in the cortical tubuli (Fig. 1c and d).

dIF dIF showed that TNF-a-positive cells co-localized primarily with mesangial cells but also with endothelial cells in the glomeruli (Fig. 2a–c). In con-

617

A. Granfeldt et al.

Fig. 3. Double immunofluorescence on frozen kidney sections, cortical s arteriole. (a) IL-1b Alexa Fluor 488 (green) ( 100); (b) VWF Texas Red (red) ( 100); (c) Composit image, co localization appears yellow ( 100).

trast, no co-localization between IL-1b-positive cells and endothelial or mesangial cells in the glomeruli was found (Fig. 2d–f). The endothelial cells of the cortical arterioles were positive for IL-1b (Fig. 3a–c).

Discussion To our knowledge, this is the first study investigating the localization of the four major pro- and antiinflammatory cytokines TNF-a, IL-1b, IL-10 and IL-1ra in the kidney in an endotoxemic porcine model. Endotoxemia was accompanied by inflammatory changes at organ level with accumulation of leukocytes, and increased protein expression of TNF-a and IL-1b in the kidney. Impaired renal dysfunction was manifested as an increase in plasma creatinine concentration and a decrease in the GFR and urinary output according to the RIFLE criteria (20). TNF-a protein expression was primarily found in the glomerulus, while IL-1b protein expression was found in both the glomerulus and the cortical arterioles. TNF-a co-localized with both mesangial and endothelial cells in the glomerulus, while the IL-1b-positive staining pattern in the glomerulus suggested presence in the podocytes. Furthermore, IL-1b was also found to co-localize with the endothelial cells of the cortical arterioles. In contrast, IL-10 and IL-1ra were found in the renal tubular cells, suggesting tubular reabsorption. The finding that TNF-a was localized in the mesangial and endothelial cells is consistent with other studies. TNF-a expression was evaluated by electron microscopy in rats injected with LPS (21)

618

and was found to be present in the lysosomes of the mesangial cells, suggesting a role in synthesis and storage. IL-6 was also found to be localized in the mesangial cells in an endotoxemic mice model, consistent with the finding of TNF-a in the mesangial cells in our study (22). The role of mesangial cells in TNF-a production is also consistent with in vitro studies demonstrating that mesangial cells are capable of synthesizing TNF-a (16, 23). Furthermore, LPS-treated rats also showed TNF-a mRNA expression in the mesangial cells (24). In both, the present study and in the study by Kita et al., TNF-a was also found to be localized in the endothelial cells. This is in accordance with the capability of activated endothelial cells to produce cytokines (25). However, endothelial cells also express cytokine receptors and some of the TNF-a, and IL-1b found localized to the endothelial cells may be due to receptor binding. The production of IL-1b differs from most cytokines like TNF-a and IL-6 by its conversion to active IL-1b by caspase-1. The histological location of TNF-a and IL-1b differed in this present study. As evaluated by IHC IL-1b, protein was present in the cytoplasm of cells consistent with podocytes, a known source of IL-1b in human studies (26). However, currently, no sufficient antibodies against the porcine podocytes are available to confirm this by dIF. In agreement with other studies, we found that LPS infusion increased the accumulation of leukocytes in the glomeruli (27). Leukocytes play an important role in the pathogenesis of septic AKI (28) through their production of cytokines, soluble mediators, reactive oxygen species (ROS)

Endotoxemic porcine model

and proteolytic enzymes, which all are mediators of tissue injury. Leukocytes were observed in glomeruli, in the peritubular spaces and attached to the endothelial surface of intertubular veins and capillaries. However, we only found a few TNF-a and IL-1b protein-expressing cells in the peritubular space corresponding to the localization of infiltrating leukocytes. This finding suggests that the leukocytes are not responsible for the large amount of cytokines detected in the kidney. Instead, they may cause tissue damage by the production of ROS and/or proteolytic enzymes (29). The anti-inflammatory cytokines IL-10 and IL1ra are synthesized and released to counteract the effect of the pro-inflammatory mediators. The finding of IL-10 at the tubular brush border and IL-1ra scattered in the glomerulus and in the tubular cells suggests that they are produced systemically and subsequent undergo glomerular filtration and reabsorption. The kidney is the major catabolic organ in handling plasma cytokines. Several studies using labelled recombinant cytokines have shown accumulation in the kidney cortex and, in nephrectomized animals, plasma cytokines half-lives are increased, proving the kidneys’ role in clearing cytokines (30–32). It is unclear whether the cytokines are filtrated and excreted in the urine or whether they are reabsorbed and degraded. In a study by Reimers et al., (15) recombinant IL-1b was detected in the proximal tubulus of the kidney while both degraded and intact IL-1b were found in the urine. Other studies only found either intact or degraded IL-1b in the urine (14, 33). Increased levels of pro-inflammatory cytokines are also found in the urine from patients with septic shock and AKI. However, it is impossible to discriminate whether they are produced locally in the kidney or filtrated (34). A prevailing theory is that the anti-inflammatory cytokines are less readily filtrated in the glomerulus while the pro-inflammatory cytokines are easily filtrated and reabsorbed due to a difference in molecular weight. Molecules with a molecular weight around 5–10 kDa are freely filterable and large molecules around the size of albumin (69 kDa) are restricted although GFR and the electrostatic properties also influence filtration (35, 36). However, only minor differences exist between the molecular weights of the pro- and anti-inflammatory cytokines, implying that both anti- and proinflammatory cytokines are capable of undergoing glomerular filtration (36). Also, animal studies

using injection of labelled recombinant IL-10 and IL-1ra show that the elimination time is increased in nephrectomized animals and that the clearance is dependent primarily on glomerular filtration (37–39). Our porcine model is highly reproducible and the systemic inflammatory response provoked is analogous to the human response (17, 40). Furthermore, the cardiovascular and renal responses are equivalent to what is seen clinically in septic patients with a hyperdynamic circulation. The porcine kidney is multipapillate with a calyceal structure similar to humans, making it more anatomically relevant than commonly used laboratory animal species (41). LPS-induced AKI has similar histological damages as in human AKI (9). However, septic patients often have comorbidities and infections with both Gram-negative and Gram-positive organisms and interpretation of the data must therefore be made with caution. In conclusion, this study demonstrates that the kidneys are highly active during acute endotoxemia and that the pro-inflammatory cytokines TNFa and IL-1b are produced locally in the kidney whereas the anti-inflammatory cytokines IL-10 and IL-1ra are produced systemically, filtrated and reabsorbed in the tubuli. However, these findings should be supported by using in situ hybridization. Therapies targeting the AKI such as furosemide and low-dose dopamine have been used, but without evidence of a beneficial effect. The results from this study, together with other studies, suggest that mediators produced locally in the kidney may be of major importance for the development of AKI. Therefore, future therapies should target the production of these mediators.

Acknowledgements The explicit skills of technicians Lene Vestergaard, Karin Vestergaard and Lotte Arentoft are gratefully acknowledged.

References 1. Klenzak J, Himmelfarb J. Sepsis and the kidney. Crit Care Clin 2005; 21: 211–22. 2. Rangel-Frausto MS, Pittet D, Costigan M et al. The natural history of the systemic inflammatory response syndrome (SIRS). A prospective study. JAMA 1995; 273: 117–23. 3. Bagshaw SM, Uchino S, Bellomo R et al. Septic acute kidney injury in critically ill patients: clinical characteristics and outcomes. Clin J Am Soc Nephrol 2007; 2: 431–9. 4. Schrier RW, Wang W. Acute renal failure and sepsis. N Engl J Med 2004; 351: 159–69.

619

A. Granfeldt et al. 5. Langenberg C, Bellomo R, May C et al. Renal blood flow in sepsis. Crit Care 2005; 9: R363–74. 6. Chawla LS, Seneff MG, Nelson DR et al. Elevated plasma concentrations of IL-6 and elevated APACHE II score predict acute kidney injury in patients with severe sepsis. Clin J Am Soc Nephrol 2007; 2: 22–30. 7. Brix-Christensen V, Vestergaard C, Andersen SK et al. Evidence that acute hyperinsulinaemia increases the cytokine content in essential organs after an endotoxin challenge in a porcine model. Acta Anaesthesiol Scand 2005; 49: 1429–35. 8. Kohan DE. Role of endothelin and tumour necrosis factor in the renal response to sepsis. Nephrol Dial Transplant 1994; 9 (Suppl. 4): 73–7. 9. Cunningham PN, Wang Y, Guo R et al. Role of toll-like receptor 4 in endotoxin-induced acute renal failure. J Immunol 2004; 172: 2629–35. 10. Cunningham PN, Dyanov HM, Park P et al. Acute renal failure in endotoxemia is caused by TNF acting directly on TNF receptor-1 in kidney. J Immunol 2002; 168: 5817–23. 11. Knotek M, Rogachev B, Wang W et al. Endotoxemic renal failure in mice: role of tumor necrosis factor independent of inducible nitric oxide synthase. Kidney Int 2001; 59: 2243–9. 12. Scumpia PO, Moldawer LL. Biology of interleukin-10 and its regulatory roles in sepsis syndromes. Crit Care Med 2005; 33: S468–71. 13. Irikura VM, Lagraoui M, Hirsh D. The epistatic interrelationships of IL-1, IL-1 receptor antagonist, and the type I IL-1 receptor. J Immunol 2002; 169: 393–8. 14. Poole S, Bird TA, Selkirk S et al. Fate of injected interleukin 1 in rats: sequestration and degradation in the kidney. Cytokine 1990; 2: 416–22. 15. Reimers J, Wogensen LD, Welinder B et al. The pharmacokinetics, distribution and degradation of human recombinant interleukin 1 beta in normal rats. Scand J Immunol 1991; 34: 597–610. 16. Baud L, Oudinet JP, Bens M et al. Production of tumor necrosis factor by rat mesangial cells in response to bacterial lipopolysaccharide. Kidney Int 1989; 35: 1111–8. 17. Brix-Christensen V, Andersen SK, Andersen R et al. Acute hyperinsulinemia restrains endotoxin-induced systemic inflammatory response: an experimental study in a porcine model. Anesthesiology 2004; 100: 861–70. 18. Muller AM, Hermanns MI, Skrzynski C et al. Expression of the endothelial markers PECAM-1, vWf, and CD34 in vivo and in vitro. Exp Mol Pathol 2002; 72: 221–9. 19. Holthofer H, Sainio K, Miettinen A. The glomerular mesangium: studies of its developmental origin and markers in vivo and in vitro. APMIS 1995; 103: 354–66. 20. Hoste EA, Kellum JA. Acute kidney injury: epidemiology and diagnostic criteria. Curr Opin Crit Care 2006; 12: 531–7. 21. Kita T, Tanaka N, Nagano T. The immunocytochemical localization of tumour necrosis factor and leukotriene in the rat kidney after treatment with lipopolysaccharide. Int. J. Exp. Pathol. 1993; 74: 471–479. 22. Kayama F, Yoshida T, Kodama Y, Matsul T, Matheson JM, Luster MI. Pro-inflammatory cytokines and interleukin 6 in the renal response to bacterial endotoxin. Cytokine 1997; 9: 688–95. 23. Fouqueray B, Boutard V, Philippe C et al. Mesangial cellderived interleukin-10 modulates mesangial cell response to lipopolysaccharide. Am J Pathol 1995; 147: 176–82. 24. Noiri E, Kuwata S, Nosaka K et al. Tumor necrosis factoralpha mRNA expression in lipopolysaccharide-stimulated rat kidney. Chronological analysis of localization. Am J Pathol 1994; 144: 1159–66.

620

25. Volk T, Kox WJ. Endothelium function in sepsis. Inflamm. Res. 2000; 49: 185–198. 26. Niemir ZI, Stein H, Dworacki G et al. Podocytes are the major source of IL-1 alpha and IL-1 beta in human glomerulonephritides. Kidney Int 1997; 52: 393–403. 27. Cunningham PN, Holers VM, Alexander JJ et al. Complement is activated in kidney by endotoxin but does not cause the ensuing acute renal failure. Kidney Int 2000; 58: 1580–7. 28. Wu X, Guo R, Wang Y et al. The role of ICAM-1 in endotoxin-induced acute renal failure. Am J Physiol Renal Physiol 2007; 293: 1262–71. 29. Linas SL, Whittenburg D, Repine JE. Role of neutrophil derived oxidants and elastase in lipopolysaccharidemediated renal injury. Kidney Int 1991; 39: 618–23. 30. Ferraiolo BL, McCabe J, Hollenbach S et al. Pharmacokinetics of recombinant human tumor necrosis factor-alpha in rats. Effects of size and number of doses and nephrectomy. Drug Metab Dispos 1989; 17: 369–72. 31. Bocci V. Metabolism of protein anticancer agents. Pharmacol Ther 1987; 34: 1–49. 32. Kudo S, Mizuno K, Hirai Y et al. Clearance and tissue distribution of recombinant human interleukin 1 beta in rats. Cancer Res 1990; 50: 5751–5. 33. Newton RC, Uhl J, Covington M et al. The distribution and clearance of radiolabeled human interleukin-1 beta in mice. Lymphokine Res 1988; 7: 207–16. 34. Mariano F, Guida G, Donati D et al. Production of plateletactivating factor in patients with sepsis-associated acute renal failure. Nephrol Dial Transplant 1999; 14: 1150–7. 35. Brenner BM, Bohrer MP, Baylis C et al. Determinants of glomerular permselectivity: insights derived from observations in vivo. Kidney Int 1977; 12: 229–37. 36. Chang RL, Ueki IF, Troy JL et al. Permselectivity of the glomerular capillary wall to macromolecules. II. Experimental studies in rats using neutral dextran. Biophys J 1975; 15: 887–906. 37. Andersen SR, Lambrecht LJ, Swan SK et al. Disposition of recombinant human interleukin-10 in subjects with various degrees of renal function. J Clin Pharmacol 1999; 39: 1015–20. 38. Chiu PJ, Radwanski E, Tetiloff G et al. Interleukin-10 pharmacokinetics in intact and nephrectomized mice. Eur Cytokine Netw 1996; 7: 67–9. 39. Kim DC, Reitz B, Carmichael DF et al. Kidney as a major clearance organ for recombinant human interleukin-1 receptor antagonist. J Pharm Sci 1995; 84: 575–80. 40. Nielsen JS, Larsson A, Brix-Christensen V, Nyengaard JR, Ledet T, Tonnesen E. Endotoxemia-induced lymphocyte apoptosis is augmented by a hyperinsulinemic-euglycemic clamp. Anesthesiology 2005; 102: 768–73. 41. Dalmose AL, Hvistendahl JJ, Olsen LH, Eskild-Jensen A, Djurhuus JC, Swindle MM. Surgically induced urologic models in swine. J Invest Surg 2000; 13: 133–45.

Address: Asger Granfeldt Department of Anesthesiology and Intensive Care Medicine Aarhus Sygehus Building 1C Noerrebrogade 44 Aarhus Denmark e-mail: [email protected]