Cholinergic agonists attenuate renal ischemia–reperfusion injury in rats

Abstract

Inflammation plays a significant role in the pathophysiology of renal ischemia-reperfusion injury. Local inflammation is modulated by the brain via the vagus nerve and nicotinic acetylcholine receptors such that electrical or pharmacologic stimulation of this cholinergic anti-inflammatory pathway results in suppression of proinflammatory cytokine production. We examined the effects of cholinergic stimulation using agonists, nicotine or GTS-21, given before or after bilateral renal ischemia-reperfusion injury in rats. Pretreatment of rats with either agonist significantly attenuated renal dysfunction and tubular necrosis induced by renal ischemia. Similarly, tumor necrosis factor-α protein expression and leukocyte infiltration of the kidney were markedly reduced following treatment with cholinergic agonists. We found functional nicotinic acetylcholine receptors were present on rat proximal tubule epithelial cells. Cholinergic stimulation significantly decreased tubular necrosis in vagotomized rats after injury, implying an intact vagus nerve is not required for this renoprotective effect.

Renal ischemia–reperfusion (I/R) injury, the leading cause of acute renal injury, is commonly encountered in clinical situations such as trauma, aortic bypass surgery, hemorrhagic shock, and renal transplantation.¹ I/R injury has a high morbidity and mortality in patients with native kidneys and contributes significantly to delayed graft function and acute rejection in the short-term and chronic rejection later after transplantation. Unfortunately, effective prophylactic/therapeutic modalities are non-existent.^1–4

The mechanisms underlying I/R injury are complex and incompletely understood.^5–6 Cytokine (for example, tumor necrosis factor, TNF) release from the damaged organ or tissue occurs early.^7–8 TNF then acts as a proximal mediator to stimulate the expression of other cytokines and chemokines. TNF also activates the endothelium to express adhesion molecules and additional proinflammatory molecules. Collectively, these agents then trigger the infiltration of proinflammatory cells into the tubulointerstitium to promote further inflammation, which adds to the tissue damage.^8–13 Renal tubular epithelial cells play an important role in the inflammatory response by producing cytokines.¹⁴ Strategies that inhibit macrophage and endothelial cell activation therefore may help prevent and treat I/R injury.^15–17

The cholinergic anti-inflammatory pathway (CAP) is a novel physiological mechanism through which the brain modulates inflammation at various locations in the body through the vagus nerve and nicotinic acetylcholine receptors (nAChRs) expressed by peripheral cells.^18–21 Experimental activation of the CAP by direct electrical stimulation of efferent vagus nerve inhibits the synthesis of TNF in the liver, spleen, and heart and attenuates serum TNF levels during endotoxemia, hemorrhagic shock, and other diseases associated with excessive cytokine release.^18–23 In addition, cholinergic stimulation blocks endothelial cell activation and leukocyte recruitment during inflammation.²⁴ Acetylcholine released by stimulation of the vagus nerve binds to nAChRs expressed by macrophages and other immunocompetent cells that modulate or participate in the inflammatory response to suppress proinflammatory cytokine production.^19,20 The nAChRs are comprised of five receptor subunits, including α1–10, β1–4, γ, δ, and ε, which form ligand-gated ion channels.²⁵ The subunits fall into two broad groups: neuronal nicotinic receptors (consisting of α2–10 and β2–4) and muscle nicotinic receptors (consisting of α1, β1, γ, δ, and ε subunits). The neuronal receptors are further subclassified into either homomeric nAChRs (for example, α7 or α9) or heteromeric nAChRs (consisting of combinations of α and β subunits, for example, α₃β₂). These neuronal nAChRs were first described in the nervous system where they mediate many physiological functions.²⁶ Many nAChRs have been demonstrated in non-neuronal tissues/cells, including placental epithelial cells,²⁷ the urothelium,²⁸ and endothelial cells.²⁹ Cholinergic agonists (including nicotine and GTS-21) produce effects similar to the electrical stimulation of the vagus nerve.^30–34 In this study, we investigated the effects of pharmacologic cholinergic stimulation in both normal and vagotomized rats after a standardized renal I/R injury.^16,35,36

RESULTS

Cholinergic agonists reduce renal dysfunction following I/R injury

Renal function was assessed by plasma creatinine levels. As expected, vehicle (saline)-treated rats subjected to 45 min of ischemia followed by 24 h of reperfusion experienced a significant increase in plasma creatinine levels compared to the sham-operated animals. Preischemic treatment with either nicotine or GTS-21 (a selective α7nAChR agonist) markedly reduced plasma creatinine levels 24 h after reperfusion compared to the saline-treated group (Figure 1a). Treatment with nicotine showed a dose-dependent improvement in I/R-induced renal dysfunction (Figure 1b). Delayed treatment with nicotine or GTS-21 (first dose given 2 h post-reperfusion) did not significantly improve renal function compared to saline-treated rats (data not shown).

Figure 1. Cholinergic agonists reduce plasma creatinine levels following renal I/R injury.

(a) Rats underwent I/R surgery following pretreatment with either saline (vehicle), nicotine (Nic, 1 mg kg⁻¹), or GTS-21 (10 mg kg⁻¹). (b) The effect of increasing doses of nicotine on renal function after I/R injury is shown. The lowest dose used (0.2 mg kg⁻¹) did not produce a significant improvement. Values are expressed as mean±s.e.m. (n = 5–6 rats per group). *P<0.001; **P<0.0001.

Cholinergic agonists ameliorate acute tubular damage following renal I/R injury

Although no morphological changes were observed in the kidneys of sham-operated rats (Figure 2a), severe and extensive acute tubular damage was observed in the cortex and outer medulla in saline-treated rats 24 h after reperfusion. This was manifested as tubular epithelial cell necrosis, tubular dilatation, and proteinaceous cast formation in the cortex and outer medullary zones (Figure 2b). Pretreatment with nicotine or GTS-21 reduced the extent and severity of injury (Figure 2c and d). Semiquantitative scoring of the acute tubular necrosis was performed by an experienced renal pathologist (blinded to the experimental groups). Preischemic treatment with either nicotine or GTS-21 significantly attenuated the acute tubular necrosis scores compared to the saline-treated group (Figure 3a). Treatment with nicotine showed a dose-dependent improvement in renal tubular damage (Figure 3b). Delayed treatment with the cholinergic agonists did not significantly protect the kidneys from acute tubular necrosis (data not shown). A strong correlation between plasma creatinine levels and tubular necrosis scores (the two main measures of renal injury in this model) was observed (Pearson correlation coefficient = 0.73), suggesting that the plasma creatinine can be used as a substitute for the tubular necrosis score.

Figure 2. Cholinergic stimulation reduces renal damage following renal I/R injury.

Kidneys were processed and stained with hematoxylin and eosin. Representative photomicrograghs are shown. (a) Essentially normal histology (sham-operated rats). (b) Saline-treated rats following I/R injury with severe tubular damage, including vascular congestion, tubular dilatation (TD), proteinaceous casts (PC), and tubular necrosis (TN). Preischemic treatment with (c) nicotine (1 mg kg⁻¹) or (d) GTS-21 (10 mg kg⁻¹) reduced the extent of tubular damage. Original magnification × 200.

Figure 3. Cholinergic agonists improve tubular necrosis scores following renal I/R injury.

(a) Pretreatment with nicotine (Nic, 1 mg kg⁻¹) or GTS-21 (10 mg kg⁻¹) significantly attenuated the extent of renal injury. (b) The effect of increasing doses of nicotine on tubular necrosis scores after I/R injury is shown. Values are expressed as mean±s.e.m. (n = 5–7 rats per group). *P<0.05; **P<0.001.

Cholinergic agonists inhibit neutrophil infiltration following renal I/R injury

Neutrophils mediate/promote the acute inflammatory response after I/R injury. Infiltration of neutrophils following I/R injury was assessed using Leder-stained kidney sections. I/R injury significantly increased the number of kidney-infiltrating neutrophils in saline-treated animals compared to sham-operated animals. Pretreatment with either nicotine or GTS-21 significantly blocked neutrophil accumulation within the kidney following I/R compared to the saline-treated animals (Figure 4a).

Figure 4. Cholinergic agonists inhibit renal leukocyte infiltration following I/R injury.

The neutrophil count was obtained from Leder-stained kidney sections, and the combined neutrophil and macrophage infiltration into the kidney was assessed by the MPO activity. Pretreatment with either nicotine (Nic, 1 mg kg⁻¹) or GTS-21 (10 mg kg⁻¹) significantly reduced the renal (a) neutrophil count and (b) MPO activity. Values are expressed as mean±s.e.m. (n = 5–7 rats per group). *P<0.05; **P<0.001; ***P<0.0001.

Cholinergic agonists suppress myeloperoxidase activity following renal I/R injury

Accumulation of leukocytes in renal tissue at 24 h after reperfusion was assessed by the determination of renal myeloperoxidase (MPO) activity. MPO activity reflects both macrophage and neutrophil infiltration.⁸ Renal tissue MPO levels were elevated after I/R injury in saline-treated rats compared to sham-operated rats. Pretreatment of rats with either nicotine or GTS-21 significantly reduced renal MPO activity compared to saline treatment (Figure 4b).

Cholinergic agonists inhibit renal TNF production following I/R injury

To explore the mechanism by which the cholinergic agonists attenuate renal I/R injury in this model, we measured kidney TNF levels. TNF is a proximal mediator of inflammation following I/R injury and it promotes tissue injury at the site of inflammation. TNF was markedly elevated 24 h after reperfusion in the saline-treated group compared to sham-operated rats. Preischemic treatment with either nicotine or GTS-21 significantly reduced kidney TNF levels compared to saline treatment (Figure 5).

Figure 5. Cholinergic agonists suppress renal TNF protein production following I/R injury.

Renal I/R injury significantly elevated TNF protein levels. Pretreatment with either nicotine (Nic, 1 mg kg⁻¹) or GTS-21 (10 mg kg⁻¹) significantly reduced TNF levels. Values are expressed as mean±s.e.m. (n = 4–7 rats per group). *P<0.05; **P<0.0001.

Nicotine attenuates renal injury in vagotomized animals following I/R injury

Next, we examined the effect of cholinergic stimulation in kidneys lacking vagal innervation, which occurs in clinical renal transplants; previously vagotomized animals underwent renal I/R injury with or without previous treatment with nicotine. Pretreatment of vagotomized animals with nicotine significantly attenuated renal damage after I/R injury, as determined by the tubular necrosis scores, compared to saline treatment (Figure 6).

Figure 6. Nicotine attenuates renal injury in vagotomized animals following I/R injury.

Previously vagotomized rats underwent I/R surgery following pretreatment with either saline (vehicle) or nicotine (Nic, 1.5 mg kg⁻¹). Pretreatment with nicotine significantly reduced the tubular necrosis scores. The nicotine dose was chosen based on previous dose–response experiments. Values are expressed as mean±s.e.m. (n = 6 rats per group). **P<0.0001.

Tubular epithelial cells constitutively express nAChR subunit mRNA and protein

We investigated the expression of nAChR subunit mRNA in rat tubular epithelial cells. We observed α1–7, α9, α10, and β1–4 subunit mRNA expression (Figure 7a). The primers for each subunit yielded products of expected size. On the basis of the nAChR subunit mRNA expression, we determined that most of the possible nAChRs in the cells would contain one or more of the α2, α3, or α7 subunits. By using antibodies to these three subunits, flow cytometry showed that the tubular epithelial cells constitutively express α2, α3, and α7 subunit proteins (Figure 7b).

Figure 7. NRK52E cells constitutively express functional nAChR subunits.

(a) nAChR mRNA expression determined by reverse transcription-PCR (GAPDH was used as an internal control). The δ and γ transcripts were not expressed (M, molecular weight marker; G, GAPDH). (b) Expression of the α2, α3, and α7 nAChR subunits on the cell membrane of NRK52E cells determined by flow cytometry. (c) Calcium flux after nicotine (10⁻⁶ M) stimulation. This effect was abolished by preincubation with hexamethonium.

Tubular cell nAChRs form functional receptors that are inhibited by cholinergic antagonists

To assess whether nAChR subunits expressed by tubular epithelial cells form functional receptors, we examined the ability of nicotine to induce intracellular calcium ion flux. Stimulation with nicotine (10⁻⁶ M) increased intracellular calcium concentrations (Figure 7c). Calcium flux mediated by nicotine was prevented by preincubation with the nAChR antagonist, hexamethonium (10⁻³ M).

DISCUSSION

In this report, we demonstrate that cholinergic stimulation using nicotine or GTS-21 (a selective α7nAChR agonist) attenuates I/R-induced acute renal injury in a well-established rat model. Rats that underwent renal ischemia followed by reperfusion showed characteristic signs of renal dysfunction and inflammation. These signs that include elevated plasma creatinine, histologic changes consistent with severe renal damage, and cellular infiltration are all typical of severe I/R injury. Pretreatment with the cholinergic agonists significantly reduced the tubular injury and renal dysfunction. The renoprotective effect was also seen in vagotomized animals. In addition, we found a marked reduction in renal TNF levels in animals that were pretreated with the cholinergic agonists. Furthermore, for the first time we show that tubular epithelial cells, the most vulnerable cells during renal I/R injury, constitutively express functional nAChRs. Together, our data identify a novel role for the CAP in renal I/R injury and suggest a local effect of cholinergic stimulation.

The pivotal role of inflammation as a key mechanism in renal I/R injury is well accepted.^{15–17,37,38} The recently discovered CAP is a mechanism through which the central nervous system regulates excessive inflammation and limits self-damage. This pathway acts via the vagus nerve, which innervates the viscera including the kidneys.³⁹ The intracellular mechanisms by which cholinergic stimulation blunts cytokine production are not completely understood but current knowledge suggests that the CAP acts at both the transcriptional and post-transcriptional levels.

In this study, we used nicotine and GTS-21 to elucidate the role of the CAP in renal I/R injury. GTS-21 is a selective α7nAChR agonist; therefore, the effectiveness of GTS-21 suggests a specific role of the α7nAChR in mediating the renoprotective effects. Although both significantly improved the outcome as noted above, nicotine tended to have a superior effect (Figures 1a and 3a). One reason may be because GTS-21 is a partial agonist. Also, nicotine may act via other nAChRs to augment its anti-inflammatory effects. Matsunaga et al.⁴⁰ demonstrated that nicotine may act through a non-α7nAChR mechanism(s). Another explanation for the apparent superior effects of nicotine may be due to its effect on other cellular mechanisms, for instance through the increased availability of nitric oxide (NO) in the endothelium. NO release is impaired during I/R injury.⁴¹ The reduced NO availability leads to vasoconstriction, which worsens the outcome. Preliminary experiments from our laboratory show that nicotine may increase NO in endothelial cells. It is therefore possible that nicotine improves the outcome in renal I/R injury through increased early NO production.^41–43

Our results are consistent with what is known about cholinergic stimulation, including suppression of TNF, reduced inflammation and leukocyte infiltration, and attenuated organ dysfunction. The observed renoprotective effects of nicotine in vagotomized animals suggest that cholinergic agonists act on the cholinergic receptors locally within the kidney and that an intact vagus nerve is not required for the renoprotection. This finding is consistent with what van Westerloo et al.³¹ found in experimental peritonitis, where nicotine was still effective in controlling peritonitis in vagotomized mice. After submitting our paper for publication, Sadis et al.⁴⁴ published a similar study showing the renoprotective effects of nicotine during I/R injury in mice. Using α7nAChR-knockout mice, they demonstrated a role for the α7nAChR in mediating nicotine’s protective effects.⁴⁴ Their study, however, did not rule out the potential role of other nAChRs, and they did not explore local expression of nAChRs within the kidney. In addition, that study did not examine the effect of nicotine in vagotomized animals, which is important when considering the clinical utility of such a treatment in renal transplant recipients. Our data clearly suggest a role for cholinergic agonists both in native and transplanted kidneys.

Delayed treatment with agonists 2 h after the onset of reperfusion did not significantly protect the kidneys. This is probably due to the fact that some preformed TNF-α is released earlier during the ischemic phase and although the (pre)transcriptional component of TNF-α release may be attenuated, this may be too late to salvage the kidney. This result has potential clinical implications especially for transplanted kidneys concerning the timing of treatment to achieve optimal results. For example, patients could be pretreated with cholinergic agonists just before transplantation. The same approach could be used to prevent I/R-induced renal failure during abdominal vascular surgeries. Following intravenous administration to adult rats (5 mg kg⁻¹), the elimination half-life of GTS-21 is approximately 3.7 h, indicating the need for multiple doses per day.⁴⁵

The role of the CAP in various inflammatory conditions continues to be explored. It is hoped that as we continue to learn more about it and its signaling mechanisms, we will be able to develop potent and selective agonists that lack the many known side effects of nicotine.

In summary, our study demonstrates for the first time that pretreatment with cholinergic agonists attenuates acute tubular injury and renal dysfunction following renal I/R injury in rats. This protective effect is seen even in vagotomized animals and appears to involve the α7nAChR and an anti-inflammatory pathway centered on suppression of TNF. The presence of functional nAChRs in rat tubular epithelial cells (which are actively involved in I/R-mediated inflammatory response) suggests a local effect of the cholinergic agonists. CAP can be harnessed for the treatment and prevention of I/R injury and other inflammatory conditions.

MATERIALS AND METHODS

Reagents

GTS-21, a selective α7nAChR agonist also known as 3-(2,4-dimethoxybenzylidene) anabaseine dihydrochloride, was provided by Dr Yousef Al-Abed (The Feinstein Institute for Medical Research, Manhasset, NY, USA). GTS-21 has been used in several in vivo models at doses similar to that used in this study.⁴⁶ Nicotine (1-methyl-2-[3-pyridyl] pyrrolidine) and hexamethonium were purchased from Sigma-Aldrich (St Louis, MO, USA). A transient, mild limb twitching was observed with all the doses of nicotine (0.2–1.5 mg kg⁻¹), otherwise no severe clinical side effects were noted. All other reagents were commercial products of the highest quality.

Cell culture

The well-characterized normal rat proximal tubular epithelial cell line, NRK52E, was obtained from the American Type Culture Collection (Manassas, VA, USA). Cells were cultured in Dulbecco’s modified Eagle’s medium (with 4.5 g l⁻¹ glucose; Invitrogen, Grand Island, NY, USA) containing 5% fetal bovine serum, 100 IU ml⁻¹ penicillin–streptomycin, and 2 mmol l⁻¹ glutamine in 5% CO₂ at 37 °C.

Animals

Male Sprague–Dawley rats (250–350 g) purchased from Taconic Farms (Germantown, NY, USA) were housed in light-controlled room with a 12-h light–dark cycle and allowed free access to water and standard rat chow. Rats were acclimatized for 1 week before experimentation. All animals received humane care in compliance with the National Research Council’s Guide for the Care and Use of Laboratory Animals. The animal protocol was approved by the Institutional Animal Care and Use Committee of the North Shore—LIJ Health System.

Experimental groups

The following experimental groups were studied. Group 1 (n = 5) received an intraperitoneal injection of nicotine (1 mg kg⁻¹ in saline) 20 min before the renal vessel clamp. Group 2 (n = 5) received an intraperitoneal injection of GTS-21 (10 mg kg⁻¹ in saline) 20 min before renal vessel clamp. Group 3 (n = 7) received an intraperitoneal injection of saline (vehicle) 20 min before the renal vessel clamp. Groups 4 and 5 (n = 5 each) received the first injection of either nicotine or GTS-21 2 h after reperfusion. Group 6 (n = 5) underwent sham surgery without clamping of the renal vessels. In all the groups, the treatment drug was repeated 6–8 h after the initial dose. Subsequently, the effects of two lower doses of nicotine (0.2 and 0.4 mg kg⁻¹, n = 5 in each case) were studied. In a separate set of experiments, 12 previously vagotomized animals received either 1.5 mg kg⁻¹ nicotine (n = 6) or saline (n = 6) 20 min before the induction of I/R injury.

Flow cytometry

Confluent NRK52E cells were harvested using an enzyme-free dissociation buffer. Flow cytometry analysis was performed using polyclonal antibody against the α2, α3, and α7 nAChR subunits (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or an irrelevant Ig isotype control. Staining was visualized with anti-goat Ig-PE (for α2 and α7 subunits; Rockland Immunochemicals, Gilbertsville, PA, USA) or anti-rabbit Ig-FITC (for α3 subunit; Jackson Laboratory, Bar Harbor, Maine, USA), and assessed for fluorescence using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA). For each analysis, 4000 cells were evaluated and data were obtained using CellQuest software.

Calcium flux studies

NRK52E cells were loaded with 2 μM Indo-1AM (Molecular Probes, Eugene, OR, USA). Cells were then washed twice with Dulbecco’s modified Eagle’s medium containing 1% fetal bovine serum and resuspended at 5 × 10⁶ cells ml⁻¹. Changes in intracellular calcium were monitored by flow cytometry (BD LSR II; BD Biosciences) after stimulation with nicotine (10⁻⁶ M) with or without 15 min prior incubation with the cholinergic antagonist, hexamethonium (10⁻³ M). Data analysis was performed with FlowJo software.

Reverse transcription-PCR

RNA was extracted from NRK52E cells using RNeasy mini kit (Qiagen, Valencia, CA, USA), following the manufacturer’s protocol. cDNA synthesis and PCR were performed in a single tube using gene-specific primers and total RNA by SuperScript One-Step reverse transcription-PCR with a Platinum Taq kit (Invitrogen, Carlsbad, CA, USA) using a GeneAmp PCR system 9600 (Perkin-Elmer Life and Analytical Sciences, Waltham, MA, USA) as follows: 1 cycle at 50 °C for 30 min, 1 cycle at 94 °C for 3 min, 35 cycles at 94 °C for 30 s, 55–57 °C for 1 min, for individual transcripts (see Table 1), 72 °C for 1 min, with a final extension step at 72 °C for 7 min. Primer sequences were selected from the unique cytoplasmic domain region of each nAChR subunit and are essentially those used by Liu et al.⁴⁷ Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control to verify the efficiency of RNA isolation and cDNA synthesis. PCR products were run on a 1.5% agarose gel containing ethidium bromide and visualized under ultraviolet illumination.

Table 1.

Primer sequences

Sequence	Annealing temperature	Product size
α1
5′-TGGAAGCACTGGGTGTTTTA-3′	53	288
5′-AACATGTACTTCCCGATCAGG-3′
α2
5′-TGCCCAGGTGGCTGATGATGAACC-3′	57	300
5′-GCTTTCTGTATTTGAGGTGACAGC-3′
α3
5′-AACCTGCTCCCCAGGGTCATGTTT-3′	55	300
5′-CACTTTGGATGGCTTCTTTGATTT-3′
α4
5′-GTCAAAGACAACTGCCGGAGACTT-3′	57	300
5′-TGATGAGCATTGGAGCCCCACTGC-3′
α5
5′-GTGGATTTAGTGAGCAGTCATGCA-3′	55	300
5′-TTTGGGGGGAGTTTTAAATAGTCT-3′
α6
5′-CAGGTCTTCCCCTCGATTCTGATG-3′	57	300
5′-CATTGTGGCTTTTCATGTTTTCTG-3′
α7
5′-AACTGGTGTGCATGGTTTCTGCGC-3′	57	300
5′-AGATCTTGGCCAGGTCGGGGTCCC-3′
α9
5′-ATCCTGAAGTACATGTCCAGGATC-3′	57	300
5′-TGGCCTTGTGGTCCTTGAGGCACT-3′
α10
5′-TAGCCAGTCTCTCCCCAAA-3′	57	209
5′-GCTGGAATTACCGTGCTCA-3′
β1
5′-ATAGCTCAGTAAGGCCGGCG-3′	57	355
5′-TAGGTGACCTGGATGCTGCA-3′
β2
5′-ACGGTGTTCCTGCTGCTCATC-3′	57	507
5′-CACACTCTGGTCATCATCCTC-3′
β3
5′-GAAGATGTGGATACATCGTTTCCA-3′	57	300
5′-GAGCAGAGGGAGTAGTTCAGGAAC-3′
β4
5′-ATGAAGCGTCCCGGTCTTGAAGTC-3′	57	300
5′-GGTCATCGCTCTCCAGATGCTGGG-3′
δ
5′-CAGCCGTCTACAGTGGGATG-3′	55	235
5′-CTGCCAGTCGAAAGGGAAGTA-3′		291
γ
5′-GATGCAATGGTGCGACTATCGC-3′	55	360
5′-GCCTCCGGGTCAATGAAGATCC-3′		244

Bilateral subdiaphragmatic vagotomy

Rats were anesthetized with ketamine (100 mg kg⁻¹) and xylazine (10 mg kg⁻¹) given intramuscularly. Access to the abdominal cavity was gained through a midline abdominal incision. By applying traction on the stomach with a cotton applicator, the esophagus was brought into view without manipulation of the liver. At the gastroesophageal junction, the two vagal trunks were identified on either side of the esophagus, and were carefully separated and ligated with the help of forceps. The incision was closed with 4-0 sutures. Postoperatively, the animals were monitored daily for 2 days before induction of renal I/R injury.

Renal ischemia model

An established model of renal I/R injury in rats was used.^16,35,36 In brief, male Sprague–Dawley rats were anesthetized by intraperitoneal injection of ketamine (100 mg kg⁻¹) and xylazine (10 mg kg⁻¹). A midline abdominal incision was made to expose the kidneys. Blood supply to the kidneys was interrupted by the application of non-traumatic microvascular clamps (Fine Science Tools, Foster City, CA, USA) around both renal arteries. Ischemia was confirmed by blanching of the kidneys. After 45 min, the clamps were removed and reperfusion was confirmed visually. The wound was then closed in two layers with a 4-0 silk suture and the animals were allowed to recover with free access to food and water. During the experiment, the animals were kept hydrated with normal saline instilled intraperitoneally and were kept on warm heating pads to maintain body temperature. Sham operations were performed in a similar way, except for clamping of the renal arteries. All the animals were euthanized by carbon dioxide inhalation at 24 h post-reperfusion. Blood was immediately collected in heparinized syringes via cardiac puncture and the kidneys were rapidly excised.

Determination of plasma creatinine

Heparinized blood was centrifuged at 1000 g for 10 min at 4 °C to separate the plasma, which was subsequently removed and stored at −80 °C until creatinine measurement. Plasma creatinine (mg per 100 ml) was measured on a Roche Modular automated system (Roche, Nutley, NJ, USA) by the Core Laboratory of the North Shore—LIJ Health System.

MPO assay

Portions of kidneys were frozen immediately after removal and stored at −80 °C until measurement of MPO activity. Tissue samples were homogenized in 5% (wt/vol) 0.05 M potassium phosphate buffer (pH 6.0) containing 0.5% hexadecyltrimethylammonium bromide and sonicated for 30 s. After two freeze and thaw cycles, with sonication between cycles, the samples were centrifuged at 11 000 g for 15 min at 4 °C and the supernatant was collected. A 7 μl portion of aliquots of supernatant was added to 203 μl of reaction mixture containing 0.05 M potassium phosphate buffer (pH 6.0), 0.167 mg ml⁻¹ o-dianisidine, and 0.05% H₂O₂. Change in absorbance over 4 min was measured at 450 nm using a spectrophotometer. One unit of enzyme activity was defined as change in absorbance (abs) of 1 per min. The MPO results are expressed as abs per min per g of wet tissue.

Determination of renal levels of TNF

Kidney samples were homogenized in phosphate-buffered saline containing a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA). The homogenate was first centrifuged at 2000 g for 10 min at 4 °C, and the supernatant was removed and centrifuged at 10 000 g for 10 min at 4 °C. The supernatant was stored at −80 °C until measurement of TNF-α. Renal levels of TNF-α were measured using a rat TNF-α sandwich enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN, USA), according to the manufacturer’s instructions. All samples were tested in duplicate. Results were expressed as TNF (ng) per mg protein (determined by the Bio-Rad protein assay; Bio-Rad, Hercules, CA, USA).

Histopathological studies of the kidneys

Formalin (10%)-fixed kidney samples were embedded in paraffin. Sections (4 μm) were prepared and stained with hematoxylin and eosin. The slides were reviewed and scored using a semiquantitative scale designed to assess the degree of tubular necrosis by a pathologist unaware of the experimental groups, as previously described.³⁶ The scoring ranged between 0 and 4; it is based on the percentage of tubules affected (0: <10%; 1: 10–25%; 2: 25–50%; 3: 50–75%; 4: >75%). Neutrophil infiltration into the interstitium was quantified by a naphthol-AS-D-chloroacetate esterase staining method (Leder staining). Ten high-power fields (magnification × 400) of each variable were reviewed in each slide and the mean was recorded.

Statistical analysis

All data are expressed as means±s.e.m. Multiple-group comparisons were performed using analysis of variance followed by Bonferroni post hoc testing and Student’s t-test for experiments with only two subgroups. Statistical significance was set at P<0.05. Correlation testing was performed between plasma creatinine levels and tubular necrosis scores using Pearson’s correlation coefficient testing.