C.-C. Yang and M.-C. Ma contributed equally to this work.
Hypoxic preconditioning attenuates lipopolysaccharide-induced oxidative stress in rat kidneys
Article first published online: 25 JUN 2007
The Journal of Physiology
Volume 582, Issue 1, pages 407–419, July 2007
How to Cite
Yang, C.-C., Ma, M.-C., Chien, C.-T., Wu, M.-S., Sun, W.-K. and Chen, C.-F. (2007), Hypoxic preconditioning attenuates lipopolysaccharide-induced oxidative stress in rat kidneys. The Journal of Physiology, 582: 407–419. doi: 10.1113/jphysiol.2006.122747
- Issue published online: 25 JUN 2007
- Article first published online: 25 JUN 2007
- (Received 16 October 2006; accepted after revision 16 February 2007; first published online 22 February 2007)
Chronic hypoxic (CH) preconditioning reduces superoxide-induced renal dysfunction via the upregulation of superoxide dismutase (SOD) activity and contents. Endotoxaemia reduces renal antioxidant status. We hypothesize that CH preconditioning might protect the kidney from subsequent endotoxaemia-induced oxidative injury. Endotoxaemia was induced by intraperitoneal injection of lipopolysaccharide (LPS; 4 mg kg−1) in rats kept at sea level (SL) and rats with CH in an altitude chamber (5500 m for 15 h day−1) for 4 weeks. LPS enhanced xanthine oxidase (XO) and gp91phox (catalytic subunit of NADPH oxidase) expression associated with burst amount of superoxide production from the SL kidney surface and renal venous blood detected by lucigenin-enhanced chemiluminescence. LPS induced a morphologic-independent renal dysfunction in baseline and acute saline loading stages and increased renal IL-1β protein and urinary protein concentration in the SL rats. After 4 weeks of induction, CH significantly increased Cu/ZnSOD, MnSOD and catalase expression (16 ± 17, 128 ± 35 and 48 ± 21, respectively) in renal cortex, and depressed renal cortex XO (44 ± 16%) and renal cortex (20 ± 9%) and medulla (28 ± 11%) gp91phox when compared with SL rats. The combined effect of enhanced antioxidant proteins and depressed oxidative proteins significantly reduced LPS-enhanced superoxide production, renal XO and gp91phox expression, renal IL-1β production, and urinary protein level. CH also ameliorated LPS-induced renal dysfunction in the baseline and acute saline loading periods. We conclude that CH treatment enhances the intrarenal antioxidant/oxidative protein ratio to overcome endotoxaemia-induced reactive oxygen species formation and inflammatory cytokine release.
Sepsis-caused multiple organ failure remains the most frequent cause of death in patgents admitted to intensive care units, with a mortality rate exceeding 50% (Fry et al. 1989; Beal & Cerra, 1994). Once renal failure develops in septic course, the mortality can further exceed 70% (Baue et al. 1998; Wenzel, 2002). Endotoxaemia-induced sepsis has been frequently studied in the rat model by lipopolysaccharide (LPS) administration (Khan & Badr, 1999). LPS causes enhanced formation of reactive oxygen species (ROS), which are produced by the increased inflammatory infiltrates and damaged resident cells (Gullo, 1999; Wiesel et al. 2000), and contribute to multiple organ dysfunction syndrome (Leach et al. 1998).
In the kidney, two major oxidative enzymes, xanthine oxidase (XO) and nicotinamide adenine dinucleotide phosphate (NAD(P)H) oxidase, are found in the rat kidney responsible for intrarenal superoxide generation (Gwinner et al. 1998; Huang et al. 2002; Vaziri et al. 2003). The membrane-bound NAD(P)H oxidase in the kidney is similar to that in phagocytes and is composed of two glycosylated flavoproteins (gp91phox and p22phox) (Griendling et al. 2000). In contrast, copper–zinc superoxide dismutase (Cu/ZnSOD), manganese superoxide dismutase (MnSOD), and catalase are present in rat kidneys (Nakanishi et al. 1995; Gwinner et al. 1998; Huang et al. 2002; Vaziri et al. 2003; Chen et al. 2003). LPS has been reported to cause a decrease in Cu/ZnSOD levels in the kidney (Leech et al. 1998). Endotoxin-induced acute renal failure could be caused by intrarenal reactive oxygen species (ROS) production, cytokine TNF-α/IL-1β production, and renal apoptotic cell death and extrarenal Toll-like receptor 4 (TLR4) activation (Poltorak et al. 1998; Cunningham et al. 2002; Cunningham et al. 2004; Guo et al. 2004). The gp91phox-containing NAD(P)H oxidase had been reported to be pivotal in LPS-induced TNF-α expression (Peng et al. 2005).
ROS formation usually decreases during hypoxia (de Groot & Littauer, 1989). However, Nakanishi et al. (1995) showed that in the kidneys of rats exposed for 21 days to hypobaric hypoxia SOD and catalase expression was unchanged, but levels of the lipid peroxidation marker malondialdehyde were significantly higher than in control rats. We recently found higher SOD activity and mRNA in the kidneys of rats that had undergone similar hypoxic treatment for 4 weeks than in controls, but this was associated with similar amounts of malondialdehyde and isoprostane to those in control kidneys (Chen et al. 2003). We further showed that chronic hypoxic (CH) preconditioning is advantageous in protecting the kidney by attenuating ROS-induced kidney injury after cotreatment with xanthine and XO (Chen et al. 2003). During the response to various injuries, including ischaemic reperfusion, the protective effect of mild hypoxic pretreatment is seen not only in the kidney, but also in heart, liver, cultured cortical neurons and whole brain (Schizukuda et al. 1992; Lin et al. 2003; Ma et al. 2005). As far we know, little work has been carried out on the effect of CH on extrarenal endotoxaemia-mediated acute renal failure and the corresponding production of the inflammatory cytokine IL-1β and gp91phox-containing NAD(P)H oxidase activation.
We were therefore interested in exploring whether the enhancement of the renal antioxidative systems in rats seen after CH preconditioning protects the kidney against oxidative damage from LPS-induced endotoxaemia. The existence of endotoxin-induced oxidative stress in rat kidneys was demonstrated by direct measurement of ROS in vivo.
Female Wistar rats (200–250 g) were housed at a constant temperature on a light/dark cycle (light from 07.00 to 18.00 h). All surgical and experimental procedures were approved by National Taiwan University College of Medicine and College of Public Health Institutional Animal Care and Use Committee, and the animal care and experimental protocols were in accordance with the guidelines of the National Science Council of the Republic of China (NSC 1997). All efforts were made to minimize both animal suffering and the number of animals used throughout the experiment. At the end of each experiment, the animals were killed with an intravenous sodium pentobarbital (200 mg (kg body weight)−1) injection.
Induction of CH preconditioning
The test rats were exposed to CH preconditioning in an altitude chamber for 15 h (17.00–08.00 h) a day for 4 weeks, as previously described (Chen et al. 2003), while the controls were maintained at sea level (SL). The level of 5500 m (380 Torr) was selected for hypoxia because it represents the maximal altitude to which most rats can adapt successfully.
Induction of endotoxaemia
Rats received an intraperitoneal injection of 4 mg kg−1 LPS (Escherichia coli serotype 055:B5; Sigma) to induce endotoxaemia, and were termed the SL-LPS and CH-LPS groups, respectively. The control groups, SL and CH, received 0.15 mL NaCl. Basic body data, histological examination, and ROS detection from the kidney surface were evaluated at 6 (LPS6), 24 (LPS24) and 48 (LPS48) h after LPS treatment.
General surgical procedures
On the day of the experiment, the rats were anaesthetized with sodium pentobarbital (40 mg kg−1, i.p.) and underwent surgical preparations as previously described (Ma et al. 2002; Chen et al. 2003). The rat was then placed on its right side and the left kidney was exposed via a flank incision and dissection from the surrounding tissue. The left renal artery was cannulated by introducing a length of stretched PE10 tubing from the left femoral artery via the aorta. The left renal vein was cannulated via the inferior vena cava, as described by Chapman et al. (1981). Ninety minutes were allowed for stabilization after surgery. The maintenance of deep anaesthesia was determined by the persistence of miotic pupils as judged from frequent inspection and by the lack of heart rate and arterial blood pressure fluctuations in the absence of visceral stimuli. Body temperature was kept at 36.5–37°C by an infrared light and was monitored with a rectal thermometer.
Both the femoral artery and vein were cannulated for measurement of mean arterial blood pressure (MABP) and 0.15 m NaCl infusion, respectively. Arterial blood was sampled and centrifuged at 620 g to obtain plasma for subsequent determination of the concentration of creatinine using a commercial kit (Bio-Quant, San Diego, CA, USA). All efforts were made to minimize both animal suffering and the number of animals used throughout the experiment. At the end of each experiment, the animals were killed with an intravenous sodium pentobarbital (200 mg (kg body weight)−1) injection.
Measurement of renal function and renal haemodynamics
After general surgery, the rats were prepared for renal clearance studies (Chen et al. 2003) and for measurement of the renal blood flow (RBF; Transonic System, NY, USA) on left side of kidney. The bladder was cannulated via the urethra to collect urine. Saline containing inulin (Inutest, Laevosan-Gesellschaft, Austria) was administered intravenously (1.2 ml h−1) throughout the experiment. Arterial blood samples (0.2 ml) were obtained from the carotid arterial catheter at the middle of each clearance period (30 min). All blood withdrawn was replaced immediately with an equal volume of blood taken from a separate donor animal in order to maintain stable blood pressure and haematocrit. The urine volume was estimated gravimetrically and the haematocrit was determined using a Triac centrifuge (Clay-Adams, NJ, USA). Spectrophotometric methods were used to determine the urinary and plasma concentrations of inulin, and the glomerular filtration rate (GFR) was estimated as previously described (Chen, 1993). Urine concentrations of Na+ and K+ were determined using a flame photometer as previously described (FCM6341; Eppendorf, Hamburg, Germany) (Ma et al. 2002; Chen et al. 2003). Urine total protein was determined by the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules CA, USA), using a dye-binding quantitative microassay method. This involved the addition of an acidic dye to protein solution, and subsequent measurement at 595 nm with a spectrophotometer (Amersham Pharmacia U3100 PRO). The linear range of the assay for total protein is from 8.0 to 80.0 μg ml−1.
Acute saline loading
To challenge renal function after LPS stimulation, acute saline loading was performed by 10 min intravenous infusion of an amount of isotonic saline equal to 5% of the body weight as previously described (Ma et al. 2002). The MABP and RBF were monitored continuously. Urine and blood samples were collected at 5, 10, 20, 30, 45, 60 and 90 min after the start of saline infusion for determining GFR and urinary Na+ and K+ concentration, as described above. At the end of the experiment, the animal was perfused by a transcardiac method with cold phosphate-buffered saline (0.1 m PBS, pH 7.4) as previously described (Huang et al. 2002; Chen et al. 2003). Kidneys were excised and half was stored at −80°C for immunoblotting analysis and the other half was postfixed overnight in 4% paraformaldehyde (Sigma) solution containing 10% sucrose for histological examination.
Changes in lucigenin-enhanced chemiluminescence (CL) counts
The lucigenin-enhanced CL method provides a reliable assay for superoxide generation, as previously described (Chien et al. 2001). In the present study, we studied changes in CL counts from the left kidney surface and in venous blood in separate groups as described in the following.
Kidney surface After general surgery, rats were placed in a sample chamber (Model TLU-17, CLD-110; Tohoku Electronic Industrial, Japan) and the kidney was exposed for CL assay. To exclude photons from sources other than the kidney CL, the animals were housed in a dark box with a shielded plate. Only the renal window was left unshielded and positioned under a reflector, which reflected the photons from the exposed kidney onto the photosensitive area of the detector. To record superoxide production optically, 0.1 ml of 0.1 mm lucigenin was injected directly into the kidney via the renal artery. Six rats from each group were studied at time points of 6, 24 and 48 h after LPS treatment.
Renal venous blood
A heparinized 0.2 ml sample of whole blood was taken from the left renal vein of each animal, and 0.1 ml PBS was added to the sample. CL levels were measured as described above. The model CLD-110 is extremely sensitive, being able to detect as little as 10–15 W radiant energy. The CL was measured in the absolutely dark chamber of the system. After 100 s, lucigenin in PBS was mixed with the blood, then the CL of the blood sample was counted continuously for a total of 600 s. The assay was performed in duplicate for each sample, and total CL counts in 600 s were calculated by integrating the area under the curve.
Other venous blood Blood samples taken from left jugular vein (after cannulation) and femoral vein were sampled to measure CL counts as above to see whether ROS generation by LPS could be detected in venous bloods other than kidney.
Assessment of renal tissue IL-1β concentration
On the day of renal tissue IL-1β estimation, the entire kidney cortex or medulla was homogenized with 10% (w/v) cold PBS (0.1 mol l−1, pH 7·4) using a homogenizer. The homogenates were used to estimate the IL-1β concentration of the kidney cortex or medulla after centrifugation at 3500 g for 15 min at 4°C to separate the nuclear debris. The supernatant obtained was used to assay IL-1β concentration and total protein contents. Assays for IL-1β were performed by solid-phase enzyme-linked immunoassay (ELISA) using the supernatant and a commercially available cytokine assay kit (R&D Systems, Inc., Minneapolis, USA) according to the manufacturer's instructions. The ELISA was sensitive to 5 pg ml−1 of the cytokine released. Absorbance was measured at 450 nm. The renal tissue IL-1β concentration (picograms per milligram) was obtained by the IL-1β concentration in the supernatant (picograms per millilitre) divided by total protein contents in the supernatant (milligrams per millilitre).
Immunoblotting of antioxidative and oxidative enzymes in renal tissues
Kidneys were placed on the ice and separated into the renal cortex and medulla to prepare protein samples as previously described (Ma et al. 2002, 2005). Proteins (20–60 μg) were separated on 10% polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes (Amersham Biosciences, Buckingham, England, UK). After blocking, the membranes were incubated overnight at 4°C with rabbit antisera specific for Cu/ZnSOD (Calbiochem, Darmstadt, Germany), MnSOD (Upstate, NY, USA), or catalase (The Binding Site, Birmingham, England, UK) or sheep antisera specific for XO (Santa Cruz, CA, USA) or gp91phox (Santa Cruz), diluted 200- to 2000-fold. After washes, the membrane was incubated for 1 h at room temperature with either donkey antirabbit IgG antibodies or rabbit antisheep IgG antibodies conjugated to horseradish peroxidase (Vector, Burlingame, CA, USA), then washed, and the bound antibody was visualized on film using a commercial ECL kit (Amersham Biosciences). The density of the band with the appropriate molecular mass was determined semiquantitatively by densitometry using an image analysing system (Alpha Innotech, San Leandro, CA, USA).
Preparation of renal sections for histological and immunohistochemistrical examination of MnSOD and gp91phox production
The kidney sections were embedded in OCT compound and flash-frozen in a liquid nitrogen bath. Sectioning and mounting of the kidney tissue was performed by Leica cryostat (model CM3050 S, Wetzlar, German). Cryosections (5 μm thick) were prepared and stained with haematoxylin and eosin for histological examination. Other sections were rehydrated by PBS. Endogenous peroxidase was blocked by 0.3% H2O2/methanol treatment, and then the sections were blocked by incubation with 1% bovine serum albumin (Sigma) in PBS. The sections were then incubated for overnight at 4°C with antisera against rat MnSOD or gp91phox, as above, diluted 200-fold in blocking solution, and biotinylated goat antirabbit IgG antibodies diluted 1:200 in blocking solution. Bound antibody was visualized using a commercial diaminobenzidine peroxidase substrate kit (Vector), and the sections counterstained with haematoxylin. Sections were examined in different high power fields at 200- or 400-fold magnification (Huang et al. 2002).
All data are expressed as the means ±s.e.m. Statistical analysis was performed using analysis of variance for multiple comparisons and linear regression between groups. A significance level of 5% was adopted.
After 4 weeks of CH, all rats were in apparent good health, except that the haematocrit level of the CH rats was significantly higher than that of SL rats (Table 1). LPS treatment did not significantly affect the body weight, kidney weight, or haematocrit in either SL or CH rats. At 24 and 48 h after LPS treatment, plasma creatinine level was significantly increased in both SL and CH rats. However, the LPS-enhanced plasma creatinine levels in CH group were less than those in SL group, suggesting a protective role of CH in endotoxin-induced renal failure.
|Grouping||n||BW (g)||KW (g)||Hct (%)||Plasma Cr (mg dl−1)|
|SL||18||226 ± 8||2.11 ± 0.08||41.0 ± 1.1||0.46 ± 0.12|
|SL-LPS6||12||213 ± 5||2.20 ± 0.06||39.5 ± 1.7||0.52 ± 0.13|
|SL-LPS24||12||220 ± 7||2.19 ± 0.07||40.7 ± 1.8||0.92 ± 0.17*|
|SL-LPS48||12||228 ± 5||2.13 ± 0.08||42.7 ± 0.6||0.94 ± 0.14*|
|CH||18||245 ± 8||1.95 ± 0.06||58.2 ± 3.9*||0.28 ± 0.09|
|CH-LPS6||12||240 ± 7||2.10 ± 0.05||52.5 ± 4.5‡||0.24 ± 0.06‡|
|CH-LPS24||12||229 ± 5||1.99 ± 0.06||56.8 ± 4.4‡||0.39 ± 0.11‡|
|CH-LPS48||12||228 ± 9||1.92 ± 0.08||60.3 ± 2.6‡||0.58 ± 0.14†, ‡|
Figure 1 shows representative micrographs in the SL (Fig. 1A) and CH (Fig. 1B) kidney with and without LPS treatment. No significant morphological changes in the kidney sections were found in the LPS-treated groups compared with the corresponding control groups (n= 3 in each group). These data are consistent with previous observation that endotoxaemia induces functional, not morphological, change in renal tubules (Guo et al. 2004).
CH effect on haemodynamics and renal functional changes after LPS treatment
To evaluate the protective effect of CH treatment on LPS-induced renal dysfunction, we compared haemodynamics and renal excretory function after 48 h of LPS treatment between CH and SL groups. As shown in Table 2, most of the basal renal functional indices were similar in the control SL and CH groups. LPS treatment resulted in a reduction in the RBF and GFR and an increase in urinary protein content in the SL group. In the CH-LPS48 group, the RBF level was reduced, but the reduced extent was lesser than that in the SL-LPS48 group. There were no significant changes in MABP, GFR, urinary Na+ and K+ concentration, and urinary protein level in the CH-LPS48 group.
|SL||114 ± 8||4.8 ± 1.2||1.36 ± 0.08||11.3 ± 1.5||0.68 ± 0.11||1.26 ± 0.10||4.26 ± 0.13|
|SL-LPS48||107 ± 7||3.6 ± 0.9*||0.88 ± 0.13*||10.8 ± 1.4||0.63 ± 0.18||0.95 ± 0.10||6.19 ± 0.55*|
|CH||118 ± 10||4.9 ± 0.8||1.39 ± 0.32||11.4 ± 2.8||0.70 ± 0.09||1.19 ± 0.37||3.87 ± 0.26|
|CH-LPS48||112 ± 9||4.2 ± 1.2†,‡||1.28 ± 0.11||11.1 ± 2.7||0.65 ± 0.15||1.25 ± 0.15||4.11 ± 0.20|
Changes in haemodynamics and renal excretion in response to acute saline loading
Figure 2 shows the haemodynamic and excretory responses to acute saline loading after 48 h LPS treatment. As shown in the first panel, acute saline loading lowered MABP in all four groups, but LPS treatment further decreased the MABP level in the SL group, but not the CH group, at 45–90 min. The basal RBF and GFR were lower in the SL-LPS48 group. In the SL-LPS group, RBF persistently increased during or after saline loading, but this was not affected in either of the CH groups. No significant changes in GFR in response to saline loading were seen in any of the groups, except a low basal GFR was found in the SL-LPS group. Acute saline load induced diuresis and natriuresis in all groups. However, the peak responses of the urine volume and urinary Na+ excretion at the time points of 10 and 20 min were attenuated in the SL-LPS48 rats, and the cumulative urine output and Na+ excretion values were 74 ± 10 and 75 ± 13% of the those in SL rats (P < 0.05). In contrast, LPS had no effect on acute saline-load-induced diuretic and natriuretic responses in the CH rats, except for an increase in urinary Na+ excretion at the time point of 20 min. The cumulative urine output and Na+ excretion in the CH-LPS48 group were 102 ± 18 and 132 ± 21% of those in the CH group. LPS reduced the percentage changes of urinary volume and urinary Na+ excretion in the SL-LPS48 group. However, CH treatment prevented LPS-reduced renal excretory responses to acute saline load in the CH-LPS48 group.
CH reduces LPS-enhanced superoxide production in the kidney
To evaluate the renal superoxide formation in response to LPS, a CL assay based on the oxidation of lucigenin was used (Chien et al. 2001). Lucigenin can be excited by oxygen radicals and it emits light when returning to the ground state (Okuda et al. 1992).
Figure 3A shows superoxide generation detected at the rat kidney surface in vivo. In the left panels, LPS caused an increase in superoxide formation in SL kidneys at 6 h, peaked at 24 h, and maintained to 48 h. The lower panel shows these changes are all statistically significant when compared with the SL kidneys without LPS stimulation. LPS also enhanced renal superoxide formation in CH rats at 24 and 48 h when compared with the CH group (right and lower panels). However, the increased superoxide amounts in the CH-LPS groups were less than those in the SL-LPS groups.
Figure 3B shows increased superoxide production in the renal venous blood after LPS treatment. The renal venous blood superoxide counts in the CH group were lower than that in the SL group (0.33 ± 0.06 in SL versus 0.12 ± 0.05 in CH (×104 counts), P < 0.05). The insert in Fig. 3B shows that LPS increased the superoxide amount by 8440 ± 357% in SL rats, but by only 1426 ± 90% in CH rats (P < 0.05). We were also interested in knowing whether superoxide generation could be detected in venous blood not drawn from the renal vein.
Figure 3C shows that there was an increase in superoxide amount in both femoral and jugular venous blood samples on LPS treatment. However, the superoxide level in femoral vein sample was higher than that in the jugular vein. However, the CL counts of the femoral vein sample were still lower than those in the renal venous blood (16 ± 3 in femoral vein versus 393 ± 47 in renal vein (×104 counts), P < 0.05). Increased ROS in the femoral vein may come from the dysfunction of vascular endothelial cells or skeletal muscle in the lower limb. However, the possibility cannot be ruled out that the femoral venous blood drawn in this study was a blood mixture of renal vein and vena cava because the cannula stops the distal blood flowing back to the inferior vena cava.
Expressions of IL-1β proteins
Figure 4 shows IL-1β protein production detected at the rat kidney tissues. In the left panels, LPS caused an increase in IL-1β protein production in SL kidneys at 6 h, which peaked at 6 h in medulla and in cortex, and at 24 h in cortex, and persisted to 48 h in both medulla and cortex. CH significantly decreased cortex IL-1β protein production when compared with the SL-LPS24 and SL-LPS48 cortex and decreased medulla IL-1β protein production when compared with SL-LPS6 and SL-LPS48 medulla.
Expression of antioxidative and oxidative enzymes
The above results clearly show that LPS increase oxidative stress at kidney level. We therefore examined the changes in expression of the major oxidative enzymes NAD(P)H oxidase (gp91phox) and XO, and of the antioxidative enzymes Cu/ZnSOD, MnSOD and catalase, in renal tissues of groups with and without 48 h LPS treatment. Figure 5 shows that LPS upregulated the expression of XO by 46 ± 13% and that of gp91phox by 59 ± 22% in the renal cortex of SL rats compared with untreated controls (as 100%), but had no effect on antioxidative enzymes in both parts of the kidney or on oxidative enzymes in the renal medulla, except for an increase of 48 ± 19% in catalase. Interestingly, the expression of all three antioxidative enzymes in the cortex was significantly higher in CH rats than in SL rats (increases of 16 ± 17, 128 ± 35 and 48 ± 21%, respectively, for Cu/ZnSOD, MnSOD and catalase) and, with the exception of Cu/ZnSOD, in the medulla (increases of 132 ± 34 and 77 ± 25% for MnSOD and catalase, respectively). This was associated with lower expression of XO in the renal cortex (44 ± 16%) and gp91phox in both the cortex (20 ± 9%) and medulla (28 ± 11%). Levels of most antioxidative enzymes in the CH-LPS48 group were similar to (Cu/ZnSOD and catalase in the cortex) or higher (MnSOD in both tissue parts and catalase in the medulla) than those in the SL group. In terms of oxidative enzymes, no apparent change in XO and gp91phox expression was seen in the renal cortex in the CH-LPS48 group, in which the levels were similar to those in the SL group. These immunoblotting findings support our in vivo observations that the LPS-induced excessive ROS formation is probably due to the upregulation of oxidant proteins.
Expression of MnSOD and gp91phox in the LPS-treated kidneys
Using immunohistochemical staining, we further examine the changes in the antioxidant MnSOD and oxidant gp91phox proteins in specific cell type (n= 3 in each group). Because of widespread distribution, the results of positive cell counting between subgroups were difficult to compare, but we were able to determine qualitatively. As showed in Fig. 6A, E and I, the MnSOD-positive cells in the SL kidney were found in tubular cells and vessels nearby. LPS treatment decreased MnSOD expression in renal cortex (Fig. 6B) and inner medulla (Fig. 6F), but not in outer medulla (Fig. 6J). Compared with the SL kidney, CH largely increased the intracellular abundance of MnSOD in renal cortex and outer medulla (Fig. 6C, K and N). Though LPS lowered the number of MnSOD-positive cells in the CH kidneys (Fig. 6D, H and L), it maintained a level similar to or higher than that in the SL kidneys.
gp91phox was abundant in the proximal tubules and vessels of renal cortex in the SL kidneys (Fig. 7A). For the renal medulla, most of the gp91phox-positive cells were located in tubular cells of inner medulla (Fig. 7E), but rarely in the outer medulla (Fig. 7I). LPS strikingly increased tubular expression of gp91phox in the renal cortex and the positive cells could be found in glomeruli (Fig. 7B). LPS also increased gp91phox levels in tubular cells of the outer medulla (Fig. 7J), but not in the inner medulla (Fig. 7F). The gp91phox-positive cells in the renal cortex of the CH kidneys were not only located in proximal tubules but also existed in glomeruli (Fig. 7C). Compared with the SL kidneys, low abundance of gp91phox was found in inner medulla of CH kidneys (Fig. 7G). CH showed no effect on gp91phox expression in the outer medulla (Fig. 7K) but prevented its increase (Fig. 7L) when compared with the SL kidney. Most of the immunohistochemical findings were consistent with the results of Western blot analysis and strengthened our observations that CH might enhance intrarenal expression of antioxidant MnSOD to counteract, at least, the effects of LPS-induced upregulation of oxidant gp91phox.
Using the rat endotoxaemic model, the present study demonstrated that (1) LPS impaired renal haemodynamic and excretory function; (2) renal oxidative injury due to LPS was associated with excessive superoxide formation, which was demonstrated by two approaches detecting an increase in lucigenin-enhanced CL and inflammatory cytokine (IL-1β); (3) the underlying mechanism for LPS-induced oxidative stress might be related to increases in the oxidative enzymes, XO and NAD(P)H oxidase (gp91phox); (4) CH upregulated antioxidative enzymes, but may prevent the increment of oxidative enzymes to reduce LPS-induced ROS generation.
Our study indicated LPS treatment was associated with a decline in renal function and haemodynamics, consistent with previous findings in rat and mice models (Lugon et al. 1989; Wang et al. 2003). Cohen et al. (2001) showed that LPS has no effect on GFR and Na+ reabsorption in the isolated perfused rat kidney, but caused a significant decrease in renal function when given in vivo, suggesting that kidney injury was due to an extrarenal effect. Cunningham et al. (2004) provided support for this observation and showed that LPS acts on an extrarenal LPS receptor, Toll-like receptor 4 (TLR4), to induce cytokine release, which resulted in functional decrement in a kidney from a TLR4-lacking mouse transplanted into a wild-type mice. However, TLR4 has been reported to be localized exclusively in the kidney (Wolfs et al. 2002; Anders et al. 2004; Kim et al. 2005). Based on the above findings, it seemed that LPS might affect renal function via an extrarenal source or a directly intrarenal effect. Our present results showed that CH prevented the loss of renal function caused by LPS, raising the possibility that CH affects TLR4 signalling to achieve renoprotection. However, a further study is required.
In the Table 1, the plasma creatinine of CH-LPS48 group was increased with respect to CH group, but the values of GFR (based on inulin clearance) in both groups were equal in the Table 2. Comparison of the plasma creatinine levels in the groups of SL with SL-LPS48 and CH with CH-LPS48 indicated that LPS administration elevated the plasma creatinine level about twofold after 48 h in both comparisons. However, CH potentially but not significantly reduced the basal level of plasma creatinine. Additionally, CH depressed the degree of the LPS-induced rise in plasma creatinine when compared with that of the SL group. The results for GFR in the Table 2 show that CH prevented the LPS-induced fall in GFR. It seems that there is an inconsistency between the results of plasma Cr and GFR. It is suggested that this discrepancy could be due to a differential protection in the glomerular and tubular structure (intrarenal factors) after CH treatment.
On the other hand, the results of this manoeuvre show that LPS did not affect the excretory renal response to Na+ loading in SL-LPS48 group, but it affected some extrarenal factors which resulted in fall of MABP after Na+ loading to stimulate autoregulation of RBF and to reduce excretion of urine and Na+. The pathogenesis of septic acute renal failure induced by LPS involves systemic vasodilatation by an increase in vasodilator CO production and a decrease in MABP. A compensatory upregulation of vasoconstrictors in the kidney can lead to renal vasoconstriction and acute renal failure (Poole et al. 2005). Septic shock is characterized by hypotension and decreased systemic vascular resistance and impaired vascular reactivity. Renal vasoconstriction markedly contrasts with sepsis-induced generalized systemic vasodilatation, which is strongly dependent on nitric oxide (Boffa & Arendshorst, 2005). High haematocrit levels after chronic hypoxia inhibit endothelium-dependent vasodilatation in response to acetylcholine, possibly through inactivation of endothelial-derived NO by haemoglobin (Defouilloy et al. 1998). Additionally, CH increased 40% in total blood volume and vascular angiogenesis (Chien et al. 1995), and altered several regulatory hormones, including endothelin (Chen et al. 1992), norepinephrine (Johnson et al. 1983), atrial natriuretic factor (Raffestin et al. 1990), renin (Gould & Goodman, 1970), antidiuretic hormone (Raff et al. 1983) and an alteration in sympathetic nervous system (Chien et al. 1995). We suggest that CH-induced alteration in the humoral and neural factors (extrarenal factors) could ameliorate LPS-induced MABP reduction.
The reasons that no significant morphological changes in the kidney sections were found in the LPS-treated groups compared with the corresponding control groups in our study may be due to gender differences, species differences and the dose of LPS used. Andrea Fekete reported that post-ischaemic blood urea nitrogen and creatinine were higher, and renal histology showed more rapid progression, in male versus female (P < 0.05), and in female kidneys the higher basal and pos-ischaemic levels of HSP72 and different colocalization with neurokinin A might contribute to the gender differences in renal ischaemic renal injury (Fekete et al. 2006). Sex differences in the alterations of Na+,K+-ATPase following ischaemia–reperfusion injury in the rat kidney were also reported (Fekete et al. 2004). In the article of Guo et al. (2004), LPS was administered by a single intraperitoneally injection of 0.3 mg (approximately 7.5 mg kg−1) of E. coli LPS (Sigma) to male C57BL/6 mice (body weight (BW), around 40 g). We used female Wistar rats (200–250 g) with an intraperitoneal injection of 4 mg (kg BW)−1 of LPS (E. coli serotype 055:B5; Sigma). Therefore, a lower dosage of LPS in our study possibly did not induce significant morphological change in the kidney.
It is well known that LPS triggers immunity activation or cytokine production during the inflammatory processes that lead to enhanced oxidative stress by excessive formation of free radicals (Wong et al. 2000). In this study, increased superoxide production caused by LPS detected by lucigenin-enhanced approaches was consistent with previous observation. However, the underlying mechanism might be dependent on the increase in oxidative protein expression. Overproduction of superoxide or ROS due to increased activity of NAD(P)H oxidase and XO accounts for the pathogenesis of various nephropathies, such as diabetes, different hypertensive models, and chronic renal failure induced by 5/6 nephrectomy, nephrolithiasis, or membranous nephropathy (Huang et al. 2002; Vaziri et al. 2003; Jung et al. 2004; Zhan et al. 2004; Asaba et al. 2005). Consistent with the results of the study of Brandes et al. (1999) on aortic segments, the present study demonstrated that intrarenal expression of NAD(P)H oxidase and XO was upregulated by LPS and might consequently contribute to excessive superoxide formation. Since ROS generation appeared earlier than kidney injury (Fig. 3A, Table 1), we speculated that the ROS may cause the injury. However, the later ROS production may be due to the injury or inflammation induced by LPS and that CH preconditioning delays and reduces ROS formation. CH pretreatment partially prevented the LPS-induced increase in oxidative proteins and reduced excessive superoxide generation, although CH alone lowered the abundance of both oxidative enzymes. The underlying mechanisms involved in the LPS-induced increase in oxidases and their counteraction by CH are not known. However, previous studies have shown that, of the molecules involved in LPS signalling, TLR4 is the most effective in stimulating tumour necrosis factor-α/IL-1β production and subsequently stimulating NAD(P)H oxidase-dependent superoxide generation (Li et al. 2002; Fujii et al. 2003).
Given that CH showed wide-ranging effects on the animal, one may argue that it becomes impossible to know the source of free radical generation responsible for the renal functional deteriorations in endotoxaemia. For example, the excessive ROS emanating from kidney might come from leucocytes that were present in the renal venous blood and activated by LPS. However, this could be ruled out by the similar extent of superoxide generation in the leucocytes isolated from renal venous blood of the SL and CH rats stimulated by a protein kinase C (PKC) activator, phorbol myristate acetate (our unpublished observation). Moreover, high haematocrit in the CH blood undoubtedly increased the antioxidant proteins present in erythrocytes and might possibly scavenge the LPS-induced superoxide formation even further. When the haematocrit in CH blood samples was reduced to the same level same as in the SL blood, diluted CH blood still could scavenge more superoxide that generated by cotreatment of xanthine and XO (Chen et al. 2003). Therefore, changes in antioxidant/oxidant proteins in renal tissues and erythrocytes might protect CH rats against LPS-induced renal failure.
Two SOD isoforms, Cu/ZnSOD and MnSOD, are found in the kidney and are responsible for scavenging superoxide (Gwinner et al. 1998; Vaziri et al. 2003; Chen et al. 2003). Except for the increase in catalase expression in the renal medulla, our present findings showed that LPS had no effect on antioxidative enzyme expression in the kidney. However, Brandes et al. (1999) showed that MnSOD expression was increased for up to 30 h in aortic segments after LPS treatment. We speculated that an increase or lack of change in the levels of antioxidative proteins may, at least, protect against the oxidative stress caused by LPS and subsequent endotoxaemic injury. Unfortunately, these proteins did not appear to overcome the enhanced function of oxidative enzymes. Whether LPS has an effect, such as protein modification, on these antioxidative proteins is not known.
In the present study, levels of the intrarenal antioxidative enzymes, the SODs and catalase, were markedly increased in kidneys preconditioned with CH. In the case of the SODs, the upregulation seen in the present study was similar to our previous finding of an increase in SOD mRNA levels in kidney tissues, resulting in increased protein and activity (Chen et al. 2003). Strengthening of the antioxidative defence should prove advantageous to the kidney by counteracting detrimental stimuli, such as the LPS-induced injury used in this study or the exogenous superoxide damage caused by intrarenal cotreatment with xanthine and XO used in our previous study (Chen et al. 2003). Furthermore, hypoxic preconditioning (HPC) not only increased levels of antioxidative enzymes, but also lowered levels of two specific oxidative enzymes, gp91phox and XO. We therefore conclude that the benefit of CH-induced HPC is to prevent excessive generation of ROS by shifting the intrarenal redox status towards antioxidative defence. Though LPS decreased the expression of SODs and catalase in HPC kidneys, their levels were still higher than, or similar to, those in kidneys without HPC treatment. Moreover, no increase in XO and gp91phox expression was seen in CH-LPS kidneys. These results showed that HPC maintained the balance of the intrarenal redox state that would otherwise be disrupted by LPS.
In conclusion, significant induction of gp91phox and XO, associated with increased superoxide generation, was seen in the rat kidney following intraperitoneal administration of LPS, which caused acute renal failure. In this model, CH-mediated HPC increased levels of renal antioxidative proteins (SODs and catalase) to suppress the LPS-induced oxidative stress, as judged by decreased superoxide production, IL-1β release and functional preservation. These results showed that preconditioning with CH protects rats against LPS-induced acute renal failure, which is part of the severe syndrome seen in endotoxaemia.
- 2004). Signaling danger: Toll-like receptors and their potential roles in kidney disease. J Am Soc Nephrol 15, 854–867. , & (
- 2005). Effects of NADPH oxidase inhibitor in diabetic nephropathy. Kidney Int 67, 1890–1898. , , , , , & (
- 1998). Systemic inflammatory response syndrome (SIRS), multiple organ dysfunction syndrome (MODS), multiple organ failure (MOF): are we winning the battle? Shock 10, 79–89. , & (
- 1994). Multiple organ failure syndrome in the 1990s. Systemic inflammatory response and organ dysfunction. JAMA 271, 226–233. & (
- 2005). Maintenance of renal vascular reactivity contributes to acute renal failure during endotoxemic shock. J Am Soc Nephrol 16, 117–124. & (
- 1999). Role of increased production of superoxide anions by NAD(P)H oxidase and xanthine oxidase in prolonged endotoxemia. Hypertension 33, 1243–1249. , , , , , & (
- 1993). Renal functional response to ischaemic renal failure in chronic hypoxic rats. Clin Sci 85, 123–127. (
- 1992). Direct renal effects of endothelin in chronic hypoxic spontaneously hypertensive rats. Clin Exp Pharmacol Physiol 19, 809–813. , & (
- 2003). Hypoxic preconditioning enhances renal superoxide dismutase in rats. J Physiol 552, 561–569. , , & (
- 1995). Role of renal nerves in volume expansion in chronic hypoxic rats. Ren Physiol Biochem 18, 153–160. , , & (
- 2001). De novo demonstration and co-localization of free radical production and apoptosis formation in rat kidney subjected to ischemia/reperfusion. Journal of the American Society of Nephrology 12, 973–982. , , , , & (
- 2001). Renal effects of nitric oxide in endotoxemia. Am J Respir Crit Care Med 164, 1890–1895. , , , , & (
- 2002). Acute renal failure in endotoxemia is caused by TNF acting directly on TNF receptor-1 in kidney. J Immunol 168, 5817–5823. , , et al. (
- 2004). Role of Toll-like receptor 4 in endotoxin-induced acute renal failure. J Immunol 172, 2629–2635. , , , & (
- 1998). Polycythemia impairs vasodilator response to acetylcholine in patients with chronic hypoxemic lung disease. Am J Respir Crit Care Med 157, 1452–1460. , , , , , , & (
- 2006). Sex differences in heat shock protein 72 expression and localization in rats following renal ischemia-reperfusion injury. Am J Physiol Renal Physiol 291, F806–F811. , , , , , , & (
- 2004). Sex differences in the alterations of Na+,K+-ATPase following ischaemia–reperfusion injury in the rat kidney. J Physiol 555, 471–480. , , , , , , , & (
- 1989). Multiple system organ failure: The role of uncontrolled infection. Arch Surg 89, 517–523. , , & (
- 2003). l-Arginine reverses p47phox and gp91phox expression induced by high salt in Dahl rats. Hypertension 42, 1014–1020. , , & (
- 1970). The effect of hypoxia on the renin-angiotensinogen system. Lab Invest 22, 443–447. & (
- 2000). NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res 86, 494–501. , & (
- 1989). Hypoxia, reactive oxygen, and cell injury. Free Radic Biol Med 6, 541–551. & (
- 1999). Sepsis and organ dysfunction/failure. An overview. Minerva Anestesiol 65, 529–540. (
- 2004). Acute renal failure in endotoxemia is dependent on caspase activation. J Am Soc Nephrol 15, 3093–3102. , , , & (
- 1998). Antioxidant-oxidant balance in the glomerulus and proximal tubule of the rat kidney. J Physiol 509, 599–606. , , , , , & (
- 2002). Changes in the oxidant–antioxidant balance in the kidney of rats with nephrolithiasis induced by ethylene glycol. J Urol 167, 2584–2593. , , & (
- 1983). Sympathoadrenal responses to acute and chronic hypoxia in the rat. J Clin Invest 71, 1263–1272. , & (
- 2004). gp91phox-containing NADPH oxidase mediates endothelial dysfunction in renovascular hypertension. Circulation 109, 1795–1801. , , , , & (
- 1999). Endotoxin and renal function. perspectives to the understanding of septic acute renal failure and toxic shock. Nephrol Dial Transplant 14, 814–818. & (
- 2005). Ischemia-reperfusion injury activates innate immunity in rat kidneys. Transplantation 79, 1370–1377. , , , , , , , & (
- 1998). Decline in the expression of copper/zinc superoxide dismutase in the kidney of rats with endotoxic shock. effects of the superoxide anion radical scavenger, tempol, on organ injury. Br J Pharmacol 125, 817–825. , , , & (
- 2002). Essential role of the NADPH oxidase subunit p47phox in endothelial cell superoxide production in response to phorbol ester and tumor necrosis factor-α. Circ Res 90, 143–150. , , , , , & (
- 2003). Hypoxic preconditioning prevents cortical infarction by transient focal ischemia-reperfusion. Ann N Y Acad Sci 993, 168–178. , , , & (
- 1989). Renal function and glomerular hemodynamics in male endotoxemic rats. Kidney Int 36, 570–575. , , , & (
- 2002). Impaired renal sensory responses after renal ischemia in the rat. J Am Soc Nephrol 13, 1872–1883. , , , & (
- 2005). Oxygen sensitive δ-opioid receptor-regulated survival and death signals. Novel insights into neuronal preconditioning and protection. J Biol Chem 280, 16208–16218. , , , & (
- 1995). Effects of hypobaric hypoxia on antioxidant enzymes in rats. J Physiol 489, 869–876. , , , , , , , & (
- 1992). Oxygen radical generation during ischemia-reperfusion in the isolated perfused rat liver monitored by enhanced chemiluminescence. Cir Shock 38, 228–237. , , & (
- 2005). Pivotal role of gp91phox-containing NADH oxidase in lipopolysaccharide-induced tumor necrosis factor-α expression and myocardial depression. Circulation 111, 1637–1644. , & (
- 1998). Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085–2088. , , et al. (
- 2005). Role of heme oxygenase-1 in endotoxemic acute renal failure. Am J Physiol Renal Physiol 289, F1382–F1385. , , , , , & (
- 1983). Vasopressin, ACTH, and corticosteroids during hypercapnia and graded hypoxia in dogs. Am J Physiol Endocrinol Metab 244, E453–E458. , , & (
- 1990). Synthesis and secretion of atrial natriuretic factor during chronic hypoxia: a study in the conscious instrumented rat. Clin Sci (Lond) 78, 597–603. , , , , , , & (
- 1992). Hypoxic preconditioning of the ischemic canine myocardium. Cardiovasc Res 26, 534–542. , , & (
- 2003). Oxidative stress and dysregulation of superoxide dismutase and NADPH oxidase in renal insufficiency. Kidney Int 63, 179–185. , , , & (
- 2003). Interaction among nitric oxide, reactive oxygen species, and antioxidants during endotoxemia-related acute renal failure. Am J Physiol Renal Physiol 284, F532–F537. , , , , , , , , , & (
- 2002). Treating sepsis. N Engl J Med 347, 966–968. (
- 2000). Endotoxin-induced mortality is related to increased oxidative stress and end-organ dysfunction, not refractory hypotension, in heme oxygenase-1-deficient mice. Circulation 102, 3015–3022. , , , , , , , , , , , & (
- 2002). In vivo expression of Toll-like receptor 2 and 4 by renal epithelial cells. IFN-gamma and TNF-alpha mediated up-regulation during inflammation. J Immunol 168, 1286–1293. , , , , , & (
- 2000). Genes, receptors, signals and responses to lipopolysaccharide endotoxin. Scand J Immunol 51, 123–127. , & (
- 2004). Upregulation of kidney NAD(P)H oxidase and calcineurin in SHR: reversal by lifelong antioxidant supplementation. Kidney Int 65, 219–227. , & (