Pathways of renal injury in systemic gram-negative sepsis.

Acute renal failure is a grave complication of systemic gram-negative sepsis. The pathophysiological mechanisms of sepsis leading to kidney injury result in part from systemic inflammatory and haemodynamic alterations. These are triggered by the interaction of endotoxin with Toll-like receptor 4 (TLR4) on cells of the immune system. Recently, TLR4 and other co-effector molecules were identified on renal tubular and vascular cells. Furthermore, it was demonstrated that systemic endotoxin has direct access to renal sites where these receptors are expressed. Therefore, we review data in support of this novel pathway of renal injury in sepsis, whereby systemic endotoxin causes direct injury through interactions with local epithelial and endothelial TLR4.

TNF-alpha mediates obstruction-induced renal tubular cell apoptosis and proapoptotic signaling.

Obstruction of the upper urinary tract induces a progressive loss in renal mass through apoptotic renal cell death. Although TNF-alpha has been implicated in ischemia-reperfusion-induced apoptotic renal cell death, its role in obstructive renal cell apoptosis remains unknown. To study this, male Sprague-Dawley rats were subjected to left unilateral ureteral obstruction vs. sham operation. Twenty-four hours before surgery and every 84 h thereafter, rats received either vehicle or a pegylated form of soluble TNF receptor type 1 (PEG-sTNFR1). The kidneys were harvested 1, 3, or 7 days postoperatively, and tissue samples were subsequently analyzed for TNF-alpha (ELISA, RT-PCR), Fas ligand (RT-PCR), apoptosis (TUNEL, ELISA), and caspase 8 and 3 activity (Western blot). Renal obstruction induced increased tissue TNF-alpha and Fas ligand mRNA levels, TNF-alpha protein production, apoptotic renal tubular cell death, and elevated caspase 8 and 3 activity, whereas treatment with PEG-sTNFR1 significantly reduced obstruction-induced TNF-alpha production, renal tubular cell apoptosis, and caspase activity. PEG-sTNFR1 did not significantly alter Fas ligand expression. These results demonstrate that TNF-alpha mediates obstruction-induced renal tubular cell apoptosis and proapoptotic signaling and identify TNF-alpha neutralization as a potential therapeutic option for the amelioration of obstruction-induced renal injury.

Minocycline inhibits apoptosis and inflammation in a rat model of ischemic renal injury.

Tetracyclines exhibit significant anti-inflammatory properties in a variety of rheumatologic and dermatologic conditions. They have also been shown to inhibit apoptosis in certain neurodegenerative disorders. Because ischemic renal injury is characterized by both apoptosis and inflammation, we investigated the therapeutic potential of tetracyclines in a rat model of renal ischemia-reperfusion. Male Sprague-Dawley rats underwent bilateral renal artery clamp for 30 min followed by reperfusion and received either minocycline or saline for 36 h before ischemia. Minocycline reduced tubular cell apoptosis 24 h after ischemia as determined by terminal transferase-mediated dUTP nick end-labeling staining and nuclear morphology. It also decreased cytochrome c release into the cytoplasm and reduced upregulation of p53 and Bax after ischemia. The minocycline-treated group showed a significant reduction in tubular injury and cast formation. In addition, minocycline reduced the number of infiltrating leukocytes, decreased leukocyte chemotaxis both in vitro and ex vivo, and downregulated the expression of ICAM-1. Serum creatinine 24-h postischemia was significantly reduced in the minocycline-treated group. We conclude that minocycline has potent antiapoptotic and anti-inflammatory properties and protects renal function in this model of ischemia-reperfusion. Tetracyclines are among the safest and best-studied antibiotics. They are thus attractive candidates for the therapy of human ischemic acute renal failure.

Guanine nucleosides and Jurkat cell death: roles of ATP depletion and accumulation of deoxyribonucleotides.

Guanine nucleosides are toxic to some forms of cancer. This toxicity is pronounced in cancers with upregulated guanine nucleotide synthesis, but the mechanisms are poorly understood. We investigated this toxicity by measuring the effects of guanine nucleosides on nucleotides in Jurkat cells using HPLC. We also measured proliferation and cell death with microscopy and fluorescence-activated cell sorting. Guanosine increased GTP to 600% and reduced ATP to 40% of control. This resulted in cell death with a predominance of necrosis. Deoxyguanosine caused similar increases in GTP but at earlier time points. Cell death was severe with a predominance of apoptosis. Deoxyguanosine but not guanosine increased dGTP to 800% of control. Adenosine inhibited the effects of guanosine, in part by competing for uptake. In stimulated leukocytes, guanosine and deoxyguanosine altered the nucleotide pools in a way qualitatively similar to that observed in Jurkat cells. However, proliferation was enhanced rather than impaired. In conclusion, guanosine and deoxyguanosine are toxic to Jurkat cells through two mechanisms: ATP depletion, causing necrosis, and the accumulation of dGTP, resulting in apoptosis.

Guanosine supplementation reduces apoptosis and protects renal function in the setting of ischemic injury.

Ischemic injury to the kidney is characterized in part by nucleotide depletion and tubular cell death in the form of necrosis or apoptosis. Recently, we linked anoxia-induced apoptosis in renal cell cultures specifically to the depletion of GTP. We therefore hypothesized that enhancing GTP repletion in vivo might protect function by reducing apoptosis in postischemic tubules. Male C57 black mice (the "I" group of animals) underwent bilateral renal artery clamp for 32 minutes to induce ischemia and then received either normal saline ("NS") or guanosine ("G"). After 1 hour of reperfusion, renal GTP levels in NS/I were reduced to nearly half of those in sham operated mice, whereas these levels were nearly unchanged in G/I mice. Morphologic examination of tubular injury revealed no significant differences between the two groups. However, there was a significant reduction in the number of apoptotic tubular cells in the medulla in the G/I group as compared with the NS/I group. At 24 hours, creatinine was significantly elevated in the NS/I group, compared to the G/I group. We conclude that guanosine protects against renal ischemic injury by replenishing GTP stores and preventing tubular apoptosis.

Modeling ischemia in vitro: selective depletion of adenine and guanine nucleotide pools.

Intracellular ATP depletion is a hallmark event in ischemic injury. It has been extensively characterized in models of chemical anoxia in vitro. In contrast, the fate of GTP during ischemia remains unknown. We used LLC-PK proximal tubular cells to measure GTP and ATP changes during anoxia. In 45 min, antimycin A decreased ATP and GTP to 8% and 2% of controls, respectively. Ischemia in vivo resulted in comparable reductions in GTP and ATP. After 2 h of recovery, GTP levels in LLC-PK cells increased to 65% while ATP increased to 29%. We also investigated steady-state models of selective ATP or GTP depletion. Combinations of antimycin A and mycophenolic acid selectively reduced GTP to 51% or 25% of control. Similarly, alanosine selectively reduced ATP to 61% or 26% of control. Selective GTP depletion resulted in significant apoptosis. Selective ATP depletion caused mostly necrosis. These models of ATP or GTP depletion can prove useful in dissecting the relative contribution of the two nucleotides to the ischemic phenotype.

Modulation of chloride secretion in the rat ileum by intracellular bicarbonate.

Increasing intracellular bicarbonate concentration ([HCO3-]i) inhibits calcium-mediated Cl- secretion in rat distal colon and T84 cells. We investigated the effect of [HCO3-]i on Cl- secretion in rat ileum. Segments of intact ileum from Sprague-Dawley rats were studied in Ussing chambers and villus and crypt intracellular pH and [HCO3-]i were determined using BCECF. A range of crypt and villus [HCO3-]i from 0 to 31 mM was obtained by varying Ringer's composition. Basal serosal-to-mucosal Cl- flux (JsmCl) averaged 8.5 +/- 0.2 mu eq.h-1.cm-2 and was unaffected by changing [HCO3-]i or serosal bumetanide. Carbachol increased JsmCl by 3.9 +/- 0.5 mu eq.h-1.cm-2 at [HCO3-]i = 0 mM but only by 1.0 +/- 0.3 mu eq.h-1.cm-2 at high crypt and villus [HCO3-]i. Dibutyryl-cAMP increased JsmCl by 2.5 +/- 0.2 mu eq.h-1.cm-2 at all [HCO3-]i. Carbachol and db-cAMP showed mutual antagonism at low [HCO3-]i and near-additivity at high [HCO3-]i. We conclude that like rat colon and T84 cells, calcium-mediated but not cAMP-mediated Cl- secretion in the ileum is inhibited by increasing [HCO3-]i. Mutual antagonism between carbachol and db-cAMP at low [HCO3-]i was present in ileum and distal colon but not in T84 cells.

Short-chain fatty acids inhibit cAMP-mediated chloride secretion in rat colon.

Butyrate stimulates salt absorption in mammalian colon. We examined whether butyrate also affects Cl- secretion. Mucosal segments of distal colon of male Sprague-Dawley rats and T84 cells were studied in Ussing chambers. In control colon, 1 mM dibutyryl adenosine 3',5'-cyclic monophosphate (DBcAMP) increased short-circuit current (Isc) and serosal-to-mucosal Cl- flux (JsmCl) by 3.2 +/- 0.8 and 2.9 +/- 0.8 mueq.cm-2.h-1, respectively. Mucosal or serosal 25 mM butyrate prevented DBcAMP-induced increases in Isc and JsmCl. Four and eight millimolar butyrate caused half-maximal inhibition of the increases in JsmCl and Isc, respectively. Butyrate also inhibited basal JsmCl (by 2.0 +/- 0.4 mueq.cm-2.h-1) but not carbachol-mediated Cl- secretion. The relative inhibitory potency at 25 mM of other short-chain fatty acids (SCFA) paralleled their degree of cellular metabolism: butyrate > acetate = propionate > isobutyrate. At 25 mM, all SCFA reduced mucosal intracellular pH (pHi) transiently by 0.1 pH unit. In intact T84 cells, 50 mM butyrate inhibited the DBcAMP-induced rise in Isc by 55%. In T84 cells with nystatin-permeabilized basolateral membranes, butyrate inhibited the increase in Isc by 82%. We conclude that butyrate inhibits basal and cAMP-mediated Cl- secretion by a mechanism independent of pHi, possibly located at the apical membrane.

Dissociation of colonic apical Na/H exchange activity from bulk cytoplasmic pH.

Intracellular acidification by stimuli rather than CO2 fails to stimulate colonic apical Na/H ex-change and Na absorption. We examined whether Na absorption could be stimulated in the absence of changes in cytoplasmic pH (pHi). Distal colon of male Sprague-Dawley rats was used for pHi measurements with 2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein and for flux measurements in Ussing chambers. In 21 mM HCO3-Ringer, increasing PCO2 from 20 to 70 mmHg decreased pHi from 7.51 to 7.03 and increased net Na flux (JnetNa) from 4.2 +/- 0.4 to 6.8 +/- 0.6 mu eq.cm-2.h-1. Similar increases in JnetNa occurred in the absence of mucosal CI and in the presence of phalloidin to inhibit microfilaments or penzolamide to inhibit membrane-bound carbonic anhydrase. sohydric increases in Pco2 did not alter pHi but stimulated JnetNa from 5.1 +/- 0.6 to 7.2 +/- 0.8 mu eq.cm-2.h-1. Carbonyl cyanide m-chlorophenylhydrazone (CCCP) decreased pHi from 7.45 to 7.35 but did not stimulate JnetNa. Butyrate (25 mM) decreased pHi from 7.15 to 7.02 with recovery to baseline within 6 min; however, JnetNa increased by 2.2 mu eq.cm-2.h-1 for 60 min. We conclude that apical Na/H exchange activity is unresponsive to changes in bulk pHi and is independent of Cl/HCO3 exchange, microfilaments, and membrane-bound carbonic anhydrase. The presence of an H-tight, CO2, and butyrate-permeable subapical domain is postulated.

Modulation of secretagogue-induced chloride secretion by intracellular bicarbonate.

We have previously demonstrated inhibition of basal Cl- secretion by intracellular bicarbonate concentration ([HCO3-]i) in rat distal colon. We now examined whether secretagogue-induced Cl- secretion is inhibited by [HCO3-]i as well. Stripped segments of distal colon from male Sprague-Dawley rats and the colon tumor cell line T84 were studied. Flux measurements were performed in the Ussing chamber under short-circuit conditions. [HCO3-]i was calculated from intracellular pH (pHi) values that were estimated with the pH-sensitive dye 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein. Dibutyryl adenosine 3',5'-cyclic monophosphate (cAMP) and carbachol were used as secretagogues. In both distal colon and T84 cells, [HCO3-]i did not affect cAMP-induced Cl- secretion. However, carbachol-induced secretion was inhibited by [HCO3-]i; in rat colon, Cl- secretion decreased from 2.3 to 1.5 mueq.cm-2.h-1 when [HCO3-]i was increased from 15.0 to 28.4 mM (P < 0.05). In T84 cells, the change in short-circuit current decreased from 8.1 to 1.1 microA/cm2 over a range of [HCO3-]i from 0 to 15.6 mM (P < 0.001). We conclude that [HCO3-]i is an important modulator of carbachol-stimulated Cl- secretion in both rat distal colon and the T84 cell line. cAMP-mediated secretion is not affected by [HCO3-]i.

Mechanism of bicarbonate secretion in rat (Rattus rattus) colon.

1. Colonic HCO3 secretion was measured as the residual flux in male Sprague-Dawley rats (Rattus rattus). 2. Basal HCO3 secretion was increased by 1 mM dibutyryl cyclic AMP (dbcAMP) and was reduced to baseline by 0.1 mM methazolamide (Mtz) but not by SITS (1 mM), DIDS (1 mM) or amiloride (1 mM). 3. In vivo, intravenous vasoactive intestinal peptide increased HCO3 secretion and prior perfusion with 1 mM Mtz prevented this increase. 4. These results suggest that the source of basal and cAMP-stimulated HCO3 secretion is, in part, intracellular and requires the action of carbonic anhydrase.

Effect of intracellular acidification on colonic NaCl absorption.

CO2 stimulates Na+ and Cl- absorption in rat distal colon. This is most likely due to intracellular generation of H+ and HCO3- and stimulation of apical Na(+)-H+ and Cl(-)-HCO3- exchangers. We examined whether intracellular acidification by means other than CO2 would also stimulate Na+ absorption. Stripped segments of distal colon from male Sprague-Dawley rats were studied under short-circuit conditions in Ussing chambers. Identically prepared tissues were used for intracellular pH (pHi) measurements with the pH-sensitive dye 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein. When the Ringer PCO2 was increased from 20 to 34 mmHg, pHi decreased from 7.50 +/- 0.04 to 7.35 +/- 0.04 and net Na absorption increased from 2.4 +/- 0.7 to 3.7 +/- 0.7 mu eq.cm-2 x h-1. A similar degree of intracellular acidification was obtained with 2.6 microM nigericin, but no stimulation of Na+ absorption was seen. When Ringer-HCO3- concentration was reduced from 39 to 11 mM at constant PCO2 = 35 mmHg, pHi decreased from 7.55 +/- 0.02 to 7.11 +/- 0.02 with no effect on net Na+ absorption. A similar reduction in pHi in a CO2-HCO3(-)-free, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid-buffered Ringer also did not stimulate Na+ absorption. Methazolamide had no effect on steady-state pHi at any given PCO2 but caused marked reductions in net Na+ absorption (9.6 +/- 2.4 to 5.2 +/- 1.2 mu eq.cm-2 x h-1 at PCO2 = 70 mmHg). We conclude that Na+ absorption in rat distal colon is not stimulated by intracellular acidification per se but rather has an absolute requirement for CO2 and carbonic anhydrase activity.

Modulation of chloride secretion in the rat colon by intracellular bicarbonate.

Extracellular HCO3- stimulates colonic net Cl- absorption in part by inhibiting basal Cl- secretion. This inhibition was investigated by measuring serosal-to-mucosal Cl- flux across short-circuited colonic segments from Sprague-Dawley rats. Mucosal intracellular pH and bicarbonate were estimated using the pH-sensitive dye BCECF. When extracellular [HCO3-] ([HCO3-]e) was increased from 0 to 39 mmol/L at PCO2 33 mm Hg, mucosal intracellular [HCO3-] ([HCO3-]i) increased to 25.3 mmol/L and serosal-to-mucosal Cl- flux decreased from 13.0 to 7.1 microEq.cm-2.h-1. When PCO2 was increased to 72 mm Hg at [HCO3-]e 39 mmol/L, [HCO3-]i increased to 29.8 mmol/L and serosal-to-mucosal Cl- flux decreased to 5.9 microEq.cm-2.h-1. In Ringer's solution containing 21 mmol/L HCO3- and 20 mmol/L Cl- (but not 100 mmol/L Cl-), increasing PCO2 from 21 to 70 mm Hg increased [HCO3-]i to 22.6 mmol/L and decreased serosal-to-mucosal Cl- flux from 3.0 to 1.7 microEq.cm-2.h-1. Overall, serosal-to-mucosal Cl- flux was inversely related to [HCO3-]i on either side of an [HCO3-]i plateau of 9-18 mmol/L at which flux was stable. These data suggest that [HCO3-]i is an important modulator of basal Cl- secretion in rat distal colon.