点击显示 收起
Division of Nephrology, University of Utah School of Medicine, and Salt Lake Veterans Affairs Medical Center, Salt Lake City, Utah
Brain injury in hemolytic-uremic syndrome (HUS) may be enhanced by inflammatory cytokine up-regulation of endothelial cell sensitivity to shigatoxin (Stx). The present study investigated whether inflammatory cytokine up-regulation of Stx toxicity could be ameliorated by inhibiting candidate signal transduction pathways. Exposure of human brain endothelial cells (HBECs) to tumor necrosis factor (TNF) greatly increased Stx-1 and Stx-2 cytotoxicity; this was reduced by inhibition of p38 mitogen-activated protein kinase (MAPK), but not c-Jun kinase. SB203580, a specific inhibitor of p38 MAPK, reduced TNF-stimulated Stx cytotoxicity in HBECs, TNF-stimulated 125Stx-1 binding to intact HBECs, the cellular content of Gb3 (galactose 1,4, galactose 1,4, glucose-ceramide) (the Stx receptor), and TNF-stimulated Gb3 synthase and glucosylceramide synthase activities but did not affect lactosylceramide synthase activities or mRNA content. Thus, inhibition of p38 MAPK substantially reduces inflammatory cytokine up-regulation of Stx-receptor synthesis and cell-surface expression, thereby decreasing Stx cytotoxicity. Inhibition of p38 MAPK may be of therapeutic benefit in HUS.
Hemolytic-uremic syndrome (HUS) is characterized by acute renal injury, microangiopathic hemolytic anemia, and thrombocytopenia [1, 2]. HUS is the major cause of childhood acute renal failure, and most patients recover completely except for residual renal insufficiency and/or hypertension [3]. However, 5% of patients with HUS die during the acute phase [4]. Notably, these deaths are not due to renal disease but are most commonly due to damage to the brain [5].
Central nervous system (CNS) injury in HUS is thought to be due, in large part, to shigatoxins (Stxs) derived from enteric Escherichia coli infection [1, 2]. There is good evidence that Stx causes brain injury in HUS. Intravenous (iv) inoculation of rabbits with Stx-2 causes severe ischemic CNS injury [6], whereas iv inoculation of piglets with Stx-2producing strains of E. coli 0157:H7 causes CNS arteriolar necrosis [7]. Inoculation of piglets with E. coli 0157:H7 causes convulsions and severe encephalopathy associated with pyknosis and karyorrhexis of brain endothelial cells [8]. Hence, Stx induces CNS dysfunction, which is characterized by predominant endothelial cell injury.
The mechanisms by which Stx damages brain endothelial cells have been investigated. Under normal circumstances, human brain endothelial cells (HBECs) are very resistant to Stx-1induced cytotoxicity [9, 10]. However, in the presence of inflammatory cytokines, these cells become highly sensitized to Stx. Mice inoculated intragastrically with E. coli 0157:H7 develop flaccid paralysis associated with brain edema and microhemorrhage; treatment with tumor necrosis factor (TNF) worsened the brain pathology in response to E. coli 0157:H7, whereas treatment with a TNF inhibitor markedly reduced brain injury [11]. In cultured HBECs, TNF and/or interleukin (IL)1 enhanced Stx toxicity by 1000-fold [9, 10, 12]. TNF and IL-1 up-regulation of sensitivity of HBECs to the cytotoxic effects of Stx-1 are due, at least in part, to modulation of expression of Stx receptor. Stx binds to glycosphingolipids terminating in galactose 14 galactose, whereupon the complex is internalized with resultant inhibition of peptide elongation [2, 13]. The major glycosphingolipid that binds Stx is Gb3 (galactose 1,4, galactose 1,4, glucose-ceramide) [13]. TNF and IL-1 have been shown to markedly increase expression of Gb3 and Stx binding in HBECs [9, 10, 12]. This effect of inflammatory cytokines is associated with increased activity of, and mRNA levels of, key enzymes in the Gb3 biosynthetic pathway (figure 1). Gb3 is synthesized from lactosylceramide and UDP-galactose by Gb3 synthase (GalT6) [14] and is metabolized to lactosylceramide by -galactosidase [15]. Lactosylceramide is derived from glucosylceramide (GlcCer) and UDP-galactose by lactosylceramide synthase (GalT2) [16], whereas GlcCer is derived from ceramide and UDP-glucose by GlcCer synthase [17]. Recent studies in our laboratory have shown that TNF and IL-1 increase GlcCer synthase, GalT2, and GalT6 activities and steady-state mRNA levels (mRNA half-lives are unchanged) but do not affect -galactosidase activity or mRNA levels [12]. Finally, Stx has been shown to stimulate production of TNF and IL-1 by several cell types, including endothelial cells [9, 10, 1821]. Thus, the above studies indicate that brain endothelial cell injury in HUS is highly dependent on cytokine up-regulation of Stx cytotoxicity and that this is due, at least in part, to marked increases in expression of Stx receptors in HBECs.
There are no effective therapies currently available that reduce brain injury in HUS. Since inflammatory cytokines are likely to be critically important in CNS injury, it seems that such therapies should be directed, at least in part, at limiting inflammatory cytokine up-regulation of Stx cytotoxicity on brain endothelial cells. To this end, the present study investigated the effect of inhibition of components of cytokine-activated signaling pathways on Stx-induced injury in cultured HBECs. We report that inhibition of p38 mitogen-activated protein kinase (MAPK), but not c-Jun kinase (JNK), ameliorates Stx cytotoxicity in HBECs, and we describe the mechanisms responsible, at least in part, for this protective effect.
MATERIALS AND METHODS
Reagents.
SB203580, SB220025, and SP600125 were purchased from Calbiochem. Inhibitors were dissolved in 100% DMSO and used at a 1 : 1000 dilution. TNF was purchased from PeproTech. All lipids and standards were purchased from Matreya and were dissolved in 2 : 1 chloroform : methanol.
Cell culture.
HBECs were obtained at primary culture from Cell Systems and were studied at passages 79. Cells were grown to confluence in EGM2-MV media (Clonetics) and were switched to serum-free Maintenance Formula media (Cell Systems) 24 h before all studies were initiated. In addition to the characterization performed by Cell Systems, we determined that these cells had uniformly positive immunofluorescence for von Willebrand factor and platelet endothelial cell adhesion molecule but were negative for cytokeratin. Primary cultures of human proximal tubule cells (Clonetics) were grown as described elsewhere [18], to obtain RNA (see below).
Stx-1 purification.
Stx-1 was purified from E. coli HB101 containing pNAS13, which encodes Stx-1 (gift from Alison D. O'Brian, Uniformed Services University of the Health Sciences, Bethesda, MD), as described elsewhere [22]. The crude toxin preparation from bacterial lysates was dialyzed against 50 mmol/L Tris-HCl (pH 8), subjected to CL-6B DEAE-Sepharose (Pharmacia) anion-exchange chromatography by use of the same buffer, and eluted with a 00.5-mol/L NaCl gradient in the same buffer. Fractions were pooled on the basis of their cytotoxicity to Vero cells and were dialyzed against PBS (pH 7.4). Crude toxin was concentrated and subjected to immunoaffinity chromatography with a monoclonal antibody (MAb) to the B-subunit of Stx-1 (13C4; American Type Culture Collection) linked to an AminoLink column (Pierce). Eluted samples were tested for cytotoxicity to Vero cells, and protein was directly visualized by electrophoresis on 15% native and denaturing polyacrylamide gels. Cytotoxic fractions were combined, concentrated, and dialyzed against PBS (pH 7.4). Lipopolysaccharide (LPS) contamination was determined to be minimal, by use of an E-Toxate assay (Sigma). Toxin used for all cell-culture experiments was purified a second time over the immunoaffinity column, concentrated, and dialyzed against PBS. Toxin concentration was based on the Bradford protein assay (optical density at 650 nm).
Cytotoxicity.
Cells grown in 96-well plates were analyzed for uptake of neutral red, as described elsewhere [22], after 24 h of exposure to varying concentrations of Stx-1. Cells were incubated in 50 g/mL neutral red in M199 plus 5% fetal bovine serum (FBS) for 3 h at 37°C and rinsed in 1% formaldehyde and 1% CaCl2, followed by addition of 50% ethanol and 1% acetic acid. Absorbance was determined by optical density at 450 nm.
Gb3 content.
Gb3 content was determined as described elsewhere [22]. Cells were extracted in chloroform : methanol : water and separated on high-performance thin-layer chromatographysilica plates (Baker Si-HPF TLC Plate-Silica Gel with 200 m of analytical layer; Mallinckrodt Baker). The plates were dried, immersed in 0.5% polyisobutylmethacrylate in acetone, and sequentially incubated with Stx-1, antiStx-1 MAb (purified from a hybridoma cell line, 13C4 [American Type Culture Collection]), and 125I-goat antimouse IgG (DuPont NEN). Gb3 concentrations were calculated by densitometry and standardized to total protein. Before centrifugation, a cell aliquot was solubilized in 0.1 N NaOH and mixed with Bradford reagent (Bio-Rad), and protein concentration was determined by measuring absorbance at 590 nm.
Stx-1 binding.
Cells grown in 96-well plates were used for 125I-Stx-1 binding assays, as described elsewhere [22]. 125I-Stx-1 (17,000 cpm; Stx-1 was iodinated according to the Iodobead manufacturer's protocol ) in 100 L of M199 containing 5% FBS and 25 mmol/L HEPES plus varying concentrations of unlabeled Stx-1 was added for 24 h at 4°C. Cells were rinsed with ice-cold Hanks' balanced salt solution and solubilized in 0.1 N NaOH, and counts per minute were determined.
GalT6 activity.
GalT6 activity was determined as described elsewhere [22]. Briefly, cells were homogenized in 500 L of 50 mmol/L MES (2-[N-morpholino]ethanesulfanic acid) (pH 6.5). Dried LacCer (25 nmol; Matreya) was added to sodium cholate in water (250 g) and dried under vacuum, and the dried mixture was incubated for 60 min at 4°C. A total volume of 100 L of 50 mmol/L MES (pH 6.5) containing 10 mmol/L MnCl2, 100 mol/L 5-adenylimidodiphosphate, 250 mol/L cold UDP-galactose, 44 mol/L UDP-14C galactose (150,000400,000 cpm; Amersham Biosciences), and 125 g of total cellular protein was added to the dried LacCer/sodium cholate, the samples were vortexed and incubated for 1 h at 37°C, and the reaction was stopped by adding 1 mL of 2 : 1 chloroform : methanol. A Folch partition was established by adding 200 L of 0.1 mol/L KCl, and the upper phase was reextracted by adding 500 L of 2 : 1 chloroform : methanol. The lower phase was reextracted by adding 500 L of 1 : 1 methanol : 0.1 mol/L KCl. The lower phases (containing the neutral lipids) were combined, dried under vacuum, and chromatographed as described above for Gb3.
GalT2 activity.
Cells were sonicated in 500 L of 50 mmol/L HEPES (pH 6.8). Protein content was determined by use of the Bradford assay. Dried GlcCer (100 nmol; Matreya) was vortexed with 2.5 L of 10% Triton X-100 and incubated for 30 min at room temperature. A total volume of 100 L of 50 mmol/L HEPES (pH 6.8) containing 5 mmol/L MgCl2, 5 mmol/L MnCl2, 55.7 mol/L UDP-galactose, 44 mol/L UDP-14C galactose (150,000400,000 cpm), and 150 g of cell sonicate was added to the GlcCer mixture, and the samples were vortexed and incubated for 1.5 h at 37°C. A Folch partition was established, and the neutral lipids were chromatographed as described above for GalT6.
GlcCer synthase activity.
Cells were sonicated in 200500 L of 50 mmol/L HEPES (pH 7.4). Protein content was determined by use of the Bradford assay. Dried ceramide (100 nmol; Matreya) was vortexed with 2.5 L of 10% Triton X-100 and incubated for 30 min at room temperature. A total volume of 100 L of 50 mmol/L HEPES (pH 7.4) containing 5 mmol/L MgCl2, 5 mmol/L MnCl2, 55.7 mol/L UDP-glucose, 44 mol/L UDP-14C glucose (150,000400,000 cpm; Amersham), and 150 g of cell sonicate was added to the ceramide mixture, and the samples were vortexed and incubated for 1.5 h at 37°C. A Folch partition was established, and the neutral lipids were chromatographed as described above for GalT6.
Northern analysis.
Total RNA was isolated from confluent cells, electrophoresed on 0.9% formaldehyde gel, transferred to a nylon membrane, and prehybridized for 3 h at 60°C in 50% formamide, 5× standard saline citrate (SSC), 5× Denhardt's solution, 1% sodium dodecyl sulfate, and 100 g/mL salmon sperm DNA. Fresh solution, along with radioactively labeled probe, was added for hybridization. For probes, cDNA was made from human proximal tubule cell total RNA by use of oligo(dT) mRNA primer and SuperScript II reverse transcriptase (Invitrogen). The cDNA was then used as a template for polymerase chain reaction amplification of the coding region of the gene, by use of specific primers: (1) GalT6 (GenBank accession no. AB037883), 5-GATCTGGGGATACCATGTCCAAG-3 (forward) and 5-CAGTAGCGGGCATGCAGCTGG-3 (reverse), which yields a product size of 1040 bp; (2) GalT2 (GenBank accession no. AF097159), 5-AACGGTACAGATTATCCCGAAGG-3 (forward) and 5-TGGAGCTAACTCTGGCATGAGG-3 (reverse), which yields a product size of 912 bp; and (3) GlcCer synthase (GenBank accession no. D50840), 5-GCTGTGGCTGATGCATTTCATGG-3 (forward) and 5-CAGTTCTCCAGCTTATAGTTGGG-3 (reverse), which yields a product size of 1070 bp. All products were purified, sequenced, and cloned into pGEM-T cloning vector (Promega). The inserts were again sequenced, to ensure cloning fidelity and to confirm orientation. Antisense strand riboprobes were made by use of 32P-UTP incorporation with either T7 or SP6 RNA polymerase (Invitrogen). The probe was added to hybridization solution at 10 ng/mL, with a specific activity 109 dpm/g, and incubated overnight at 60°C. Blots were washed in decreasing concentrations of SSC and increasing temperature until background was removed. Labeled blots were subjected to autoradiography and densitometry. Glyceraldehyde-3-phosphate dehydrogenase mRNA was used as the loading control.
Total and phosphorylated mitogen-activated protein kinaseactivated protein (MAPKAP) kinase-2 protein determination.
After incubation with TNF and/or SB203580, HBECs were lysed with 62.5 mmol/L Tris base (pH 6.8), 2% (wt/vol) SDS, 10% glycerol, and 0.005% bromophenol blue. Samples were sonicated for 10 s to shear DNA and to reduce viscosity. A 15-L aliquot was heated to 100°C for 5 min and then placed on ice. Samples were then centrifuged for 5 min and loaded onto a denaturing NUPAGE 4%12% Bis-Tris mini-gel by use of the MOPS buffer system (Invitrogen). Proteins were transferred to polyvinylidene fluoride plus nylon membrane by electroelution. The blots were blocked with 5% nonfat dry milk in TBST (10 mmol/L Tris base, 150 mmol/L NaCl, and 0.5% Tween 20 [pH 7.5]) for 1 h at room temperature. Blots were washed in TBST and then incubated overnight at 4°C with either rabbit antihuman MAPKAP kinase-2 or rabbit antihuman phosphorylated MAPKAP kinase-2 (both antibodies from Cell Signaling Technology) in 5% bovine serum albumin in TBST. Blots were then washed in TBST, followed by incubation with a horseradish peroxidaseconjugated donkey antirabbit IgG (Amersham International) for 1 h at ambient temperature in blocking solution. Blots were washed in TBST and visualized by use of the enhanced chemiluminescence system (Amersham).
Statistics.
Data were analyzed by 1-way analysis of variance with the Bonferroni correction. In cases in which only 2 data points were compared, the unpaired Student's t test was used. P < .05 was considered to be significant. All data are expressed as mean ± SE.
RESULTS
TNF (100 U/mL) markedly increased the cytotoxic effect of Stx-1 in HBECs (figure 1). This concentration of TNF was used in all subsequent studies since we have previously shown it to maximally up-regulate Stx-1 cytotoxicity without being directly toxic itself [12]. To determine whether p38 MAPK is involved in TNF up-regulation of Stx-1 cytotoxicity, HBECs were incubated with SB203580, a specific inhibitor of p38 MAPK [2325]. When given 1 h before TNF, SB203580 (10 mol/L) reduced TNF-stimulated Stx-1 cytotoxicity, shifting the LD50 by 1000-fold, from 10-8 g/L (140 fmol/L; TNF alone) to 10-5 g/L (140 pmol/L; TNF plus SB203580) (figure 2). Interestingly, SB203580 also reduced Stx-1 cytotoxicity in the absence of TNF, although Stx-1 alone had a relatively modest cytotoxic effect on HBECs (figure 2). A dose-response analysis determined that SB203580 dose-dependently inhibited TNF-stimulated Stx-1 cytotoxicity and that 10 mol/L SB203580 was maximally protective (figure 3); higher concentrations were not given, because these have been shown to start inhibition of non-p38 MAPK pathways [24]. The protective effect of SB203580 required that it be given concurrently with TNF. If the agent was given after 24 h of exposure to TNF, no protective effect was evident. The protective effect of SB203580 was not dependent on how long it was given before TNFprotective effects of the same magnitude were evident when 10 mol/L SB203580 was given 0, 1, 4, 7, or 24 h before TNF. These latter findings also indicate that SB203580 in solution is stable for at least 24 h. The protective effect of SB203580 also did not depend on the duration of exposure to TNF. Previous studies have demonstrated that TNF up-regulation of Stx-1 cytotoxicity in HBECs began after 8 h of exposure to the cytokine and was maximal after 24 h [12]. No difference in amelioration of TNF-stimulated Stx-1 cytotoxicity by 10 mol/L SB203580 was evident when TNF was given 8, 12, 24, 48, or 120 h before Stx-1. To demonstrate that protection was not exclusive to SB203580, the effect of another p38 MAPK inhibitor, SB220025 [26], was assessed. SB220025 (1 mol/L) conferred protection against Stx-1 cytotoxicity identical to that of 10 mol/L SB203580. To determine whether inhibition of p38 MAPK affected Stx-1 and Stx-2 cytotoxicity equally, HBECs were sequentially exposed to 10 mol/L SB203580 for 1 h, then TNF for 24 h, followed by Stx-1 or Stx-2 for 24 h (Stx-2 was provided by Tom Obrig, University of Virginia, Charlottesville, VA). The 2 toxins were equipotent in killing HBECs in the presence or absence of TNF, whereas SB203580 had a similar protective effect on Stx-1 and Stx-2 cytotoxicity in the presence or absence of TNF. Finally, the effect of a JNK inhibitor, SP600125 (1 mol/L), on Stx-1 and Stx-2 cytotoxicity in the presence or absence of TNF in HBECs was assessed. SP600125 had no effect on Stx-1 or Stx-2 toxicity, with or without preincubation with TNF (given simultaneously with or up to 7 h before TNF). Higher concentrations of SP600125 were not used, because they caused significant cell death.
To determine whether inhibition of p38 MAPK was protective by virtue of reducing expression of Stx receptor (Gb3), the effect of SB203580 on TNF-stimulated 125I-Stx-1 binding and Gb3 content in HBECs were assessed. TNF (100 U/mL for 24 h) increased 125I-Stx-1 binding (715% ± 152% of control), whereas SB203580 greatly decreased Stx-1 binding (241% ± 29% of control) (figure 4). TNF also increased Gb3 content (955% ± 152% of control), and SB203580 significantly inhibited TNF-stimulated accumulation of Gb3 (697% ± 78% of control) (figure 5). Thus, TNF-stimulated Stx-1 binding and Gb3 content in HBECs are reduced by inhibition of p38 MAPK.
To determine the mechanism of SB203580-induced down-regulation of levels of TNF-stimulated Gb3, the effect of SB203580 on TNF modulation of enzymes involved in Gb3 biosynthesis was assessed. TNF (24 h of exposure) increased GlcCer synthase (291% ± 40% of control), GalT2 (synthesizes lactosylceramide) (171% ± 21% of control), and GalT6 (synthesizes Gb3) (289% ± 44% of control) activities (figure 6). SB203580 did not significantly affect baseline enzyme activity, but it did reduce TNF-stimulated GlcCer synthase and GalT6 activity. SB203580 did not alter GalT2 activity (figure 6). TNF (24 h of exposure) increased steady-state mRNA levels of all 3 enzymes (figure 7). SB203580 decreased steady-state mRNA levels of GlcCer synthase and GalT6, but it did not affect mRNA levels of GalT2 (figure 7).
Finally, to demonstrate that SB203580 did indeed inhibit p38 MAPK activity, the phosphorylation of a specific downstream target of p38 MAPK, MAPKAP kinase-2, was assessed [25]. SB203580 given simultaneously with TNF completely prevented TNF-induced MAPKAP kinase-2 phosphorylation after 1060 min of exposure (figure 8 shows data for the 30-min time point).
DISCUSSION
The present study was performed to identify pathways involved in TNF up-regulation of Stx toxicity in brain microvascular endothelial cells. Three major TNF signaling pathways were consideredp38 MAPK, JNK, and NF-B [27]. At the outset, we decided to not attempt inhibition of NF-B, since TNF activation of NF-B has been shown to be protective against apoptosis [27]. Inhibition of JNK failed to ameliorate either baseline or TNF-stimulated Stx cytotoxicity in HBECs, suggesting that this pathway is not critical to TNF-augmented CNS injury. In contrast, inhibition of p38 MAPK substantially ameliorated Stx toxicity in HBECs, raising the possibility that inhibition of this pathway may be of therapeutic value.
Although the role that p38 MAPK plays in modulating cytokine activation of cell sensitivity to Stx cytotoxicity was unknown before the present study, there was evidence that p38 MAPK is directly involved in mediating Stx effects on cell toxicity as well as Stx-stimulated production of cytokines. Inhibition of p38 MAPK reduces Stx-2stimulated TNF promoter activity in a human adenocarcinomaderived renal tubular epithelial cell line (ACHN) [28], Stx-mediated IL-8 secretion in a human colonic epithelial cell line [29], and Stx-1induced secretion of TNF by human macrophages [30]. In addition, inhibition of p38 MAPK partially inhibits Stx-induced cell death in Vero cells [31], prevents Stx-1mediated intestinal epithelial cell death [32], and decreases Stx-2stimulated apoptosis and caspase activity in ACHN [33]. Our data also suggest that p38 MAPK partially mediates Stx-induced cytotoxicity in HBECs; however, the magnitude of this effect is relatively small.
The present study found that TNF increased expression of Gb3 in HBECs, Stx binding associated with increased activities, and mRNA levels of GlcCer synthase, GalT2, and GalT6 in HBECs. Inhibition of p38 MAPK reduced expression of Gb3 and Stx binding associated with decreased TNF stimulation of GlcCer synthase and GalT6 activity, as well as steady-state mRNA content. In contrast, inhibition of p38 MAPK did not affect mRNA levels of GalT2 or GalT2 activity. We assume, therefore, that p38 activation is important in mediating TNF up-regulation of expression of Gb3 in HBECs and that this occurs, at least in part, through stimulation of GlcCer synthase and GalT6. p38 MAPK stimulation of GalT6 gene transcription could be related to the presence of an activating protein (AP)1 domain 160 bp upstream of the transcription start site. The human GlcCer synthase promoter also contains an AP-1 site; however, it is 1100 bp upstream of the transcription start site. In addition, the human GalT2 promoter has an AP-1 site at position -880. Hence, although p38 MAPK may act through AP-1 to increase GalT6 gene transcription, how p38 MAPK modulates GlcCer synthase gene transcription and how GalT2 gene expression is regulated are unclear. Studies are needed to define regions of the GlcCer synthase, GalT2, and GalT6 genes responsible for mediating increased transcription of these genes in response to inflammatory cytokines.
Despite several lines of in vitro and in vivo evidence demonstrating that TNF is an important mediator of Stx cytotoxicity (discussed in the introduction), recent studies have called the role of TNF into question. Wolski et al. observed that Stx-1mediated death in mice was unaffected by administration of a TNF- neutralizing antibody or if TNF- knockout mice were used [34]. There are, however, important considerations involved in this study. First, Stx-1 caused minimal induction of production of TNF in wild-type mice. Second, no significant brain injury occurred. Third, the mice most likely died from severe renal failure. Although these studies suggest that TNF may be uninvolved in Stx-induced death, there are several important factors that must be considered. Unlike that in humans, Stx mortality in mice is primarily due to renal disease, not CNS injury, although other mouse models have shown significant Stx-induced CNS injury [11]. Also, Stx induces severe CNS damage in rabbits and piglets [6, 7]. Another important consideration is that, in some circumstances, Stx alone may be insufficient to elicit brain injury. For example, LPS, which is likely to be present in the serum of patients with HUS, markedly stimulates production of both TNF and IL-1 by a variety of cell types [35], which may, in turn, sensitize cells to the toxin. Consequently, one would have to consider the effect of all factors coexisting with Stx that might affect cellular injury.
In summary, the present study has demonstrated that inhibition of p38 MAPK ameliorates Stx-1 and Stx-2induced cytotoxicity in HBECs. It remains to be seen whether inhibition of p38 MAPK will be protective in vivo. It is relevant to note that inhibition of p38 MAPK reduces renal injury in experimental antiglomerular basement membrane glomerulonephritis [36], a model of renal disease that is not dissimilar from HUS-associated renal damage. The inhibitor, NPC 31145, is a newly developed orally active blocker of p38 phosphorylation of the downstream targets of p38 MAPK [36]. Although this agent is not generally available (we were unable to obtain it), the emergence of such compounds, together with the findings of the present study, suggests that there is promise for therapeutic approaches that include targeting p38 MAPK in HUS.
References
1. Remuzzi G, Ruggenenti P, Bertani T. Thrombotic microangiopathy. In: Tisher CC, Brenner BM, eds. Renal pathology with clinical and functional correlations. 2nd ed. Philadelphia: JB Lippincott, 1994:115484. First citation in article
2. Robson WLM, Leung AKC, Kaplan BS. Hemolytic-uremic syndrome. Curr Probl Pediatr 1993; 23:1633. First citation in article
3. Siegler RL. The hemolytic uremic syndrome. Pediatr Clin North Am 1995; 42:150529. First citation in article
4. Siegler RL, Pavia AT, Christofferson RD, Milligan MK. A 20-year population-based study of postdiarrheal hemolytic uremic syndrome in Utah. Pediatrics 1994; 94:3540. First citation in article
5. Siegler RL. Hemolytic uremic syndrome in children. Curr Opin Pediatr 1995; 7:15963. First citation in article
6. Fujii J, Kinoshita Y, Kita T, et al. Magnetic resonance imaging and histopathological study of brain lesions in rabbits given intravenous verotoxin 2. Infect Immun 1996; 64:505360. First citation in article
7. Francis D, Moxley R, Andraos C. Edema diseaselike brain lesions in gnotobiotic piglets infected with Escherichia coli serotype O157:H7. Infect Immun 1989; 57:133942. First citation in article
8. Tzipori S, Chow C, Powell H. Cerebral infarction with Escherichia coli O157:H7 in humans and gnotobiotic piglets. J Clin Pathol 1988; 41:1099103. First citation in article
9. Eisenhauer PB, Chaturvedi P, Fine RE, et al. Tumor necrosis factor alpha increases human cerebral endothelial cell Gb3 and sensitivity to Shiga toxin. Infect Immun 2001; 69:188994. First citation in article
10. Ramegowda M, Samuel J, Tesh V. Interaction of shiga toxins with human brain microvascular endothelial cells: cytokines as sensitizing agents. J Infect Dis 1999; 180:120513. First citation in article
11. Isogai E, Isogai H, Kimura K, et al. Role of tumor necrosis factor alpha in gnotobiotic mice infected with an Escherichia coli 0157:H7 strain. Infect Immun 1998; 66:197202. First citation in article
12. Stricklett PK, Hughes AK, Ergonul Z, Kohan DE. Molecular basis for up-regulation by inflammatory cytokines of Shiga toxin 1 cytotoxicity and globotriaosylceramide expression. J Infect Dis 2002; 186:97682. First citation in article
13. Lingwood CA. Verotoxin-binding in human renal sections. Nephron 1994; 66:218. First citation in article
14. Kojima Y, Fukumoto S, Furukawa K, et al. Molecular cloning of globotriaosylceramide/CD77 synthase, a glycosyltransferase that initiates the synthesis of globo series glycosphingolipids. J Biol Chem 2000; 275:151526. First citation in article
15. Bishop DF, Calhoun DH, Bernstein HS, Hantzopoulos P, Quinn M, Desnick RJ. Human -galactosidase A: nucleotide sequence of cDNA clone encoding the mature enzyme. Proc Natl Acad Sci USA 1986; 83:485963. First citation in article
16. Takizawa M, Nomura T, Wakisaka E, et al. cDNA cloning and expression of human lactosylceramide synthase. Biochim Biophys Acta 1999; 1438:3014. First citation in article
17. Ichikawa S, Sakiyama H, Suzuki G, Hidari KI, Hirabayashi Y. Expression cloning of a cDNA for human ceramide glucosyltransferase that catalyzes the first glycosylation step of glycosphingolipid synthesis. Proc Natl Acad Sci USA 1996; 93:463843. First citation in article
18. Hughes A, Stricklett P, Kohan D. Shiga toxin1 regulation of cytokine production by human proximal tubule cells. Kidney Int 1998; 54:1093106. First citation in article
19. Hughes A, Stricklett P, Kohan D. Shiga toxin1 regulation of cytokine production by human glomerular epithelial cells. Nephron 2001; 88:1423. First citation in article
20. Tesh VL, Ramegowda B, Samuel JE. Purified shiga-like toxins induce expression of proinflammatory cytokines from murine peritoneal macrophages. Infect Immun 1994; 62:508594. First citation in article
21. van Setten PA, Monnens LA, Verstraten RG, van den Heuvel LP, van Hinsbergh VW. Effects of verocytotoxin-1 on nonadherent human monocytes: binding characteristics, protein synthesis, and induction of cytokine release. Blood 1996; 88:17483. First citation in article
22. Hughes A, Stricklett P, Kohan D. Cytotoxic effect of Shiga toxin1 on human proximal tubule cells. Kidney Int 1998; 54:42637. First citation in article
23. Cuenda A, Rouse J, Doza YN, et al. SB 203580 is a specific inhibitor of a MAP kinase homologue which is stimulated by cellular stresses and interleukin-1. FEBS Lett 1995; 364:22933. First citation in article
24. Davies SP, Reddy H, Caivano M, Cohen P. Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochem J 2000; 351:95105. First citation in article
25. Young PR, McLaughlin MM, Kumar S, et al. Pyridinyl imidazole inhibitors of p38 mitogen-activated protein kinase bind in the ATP site. J Biol Chem 1997; 272:1211621. First citation in article
26. Casanovas O, Miro F, Estanyol JM, Itarte E, Agell N, Bachs O. Osmotic stress regulates the stability of cyclin D1 in a p38SAPK2-dependent manner. J Biol Chem 2000; 275:350917. First citation in article
27. Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell Death Differ 2003; 10:4565. First citation in article
28. Nakamura A, Johns EJ, Imaizumi A, Yanagawa Y, Kohsaka T. Activation of 2-adrenoceptor prevents shiga toxin 2induced TNF- gene transcription. J Am Soc Nephrol 2001; 12:228899. First citation in article
29. Thorpe CM, Hurley BP, Lincicome LL, Jacewicz MS, Keusch GT, Acheson DW. Shiga toxins stimulate secretion of interleukin-8 from intestinal epithelial cells. Infect Immun 1999; 67:598593. First citation in article
30. Foster GH, Tesh VL. Shiga toxin 1induced activation of c-Jun NH2-terminal kinase and p38 in the human monocytic cell line THP-1: possible involvement in the production of TNF-. J Leukoc Biol 2002; 71:10714. First citation in article
31. Ikeda M, Gunji Y, Yamasaki S, Takeda Y. Shiga toxin activates p38 MAP kinase through cellular Ca2+ increase in Vero cells. FEBS Lett 2000; 485:948. First citation in article
32. Smith WE, Kane AV, Campbell ST, Acheson DW, Cochran BH, Thorpe CM. Shiga toxin 1 triggers a ribotoxic stress response leading to p38 and JNK activation and induction of apoptosis in intestinal epithelial cells. Infect Immun 2003; 71:1497504. First citation in article
33. Nakamura A, Imaizumi A, Yanagawa Y, Niimi R, Kohsaka T, Johns EJ. 2-adrenoceptor activation inhibits Shiga toxin2induced apoptosis of renal tubular epithelial cells. Biochem Pharmacol 2003; 66:34353. First citation in article
34. Wolski VM, Soltyk AM, Brunton JL. Tumour necrosis factor alpha is not an essential component of verotoxin 1induced toxicity in mice. Microb Pathog 2002; 32:26371. First citation in article
35. Ruggenenti P, Noris M, Remuzzi G. Thrombotic microangiopathy, hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura. Kidney Int 2001; 60:83146. First citation in article
36. Stambe C, Atkins RC, Tesch GH, et al. Blockade of p38 MAPK ameliorates acute inflammatory renal injury in rat anti-GBM glomerulonephritis. J Am Soc Nephrol 2003; 14:33851. First citation in article