ology 213 (2006) 235–244www.elsevier.com/locate/ytaap

Toxicology and Applied Pharmac

p38 mitogen-activated protein kinase mediates IL-8 induction by theribotoxin deoxynivalenol in human monocytes

Zahidul Islam a,b, Jennifer S. Gray b,c, James J. Pestka a,b,c,⁎

a Department of Food Science and Human Nutrition, Michigan State University, 234 G.M. Trout Building,East Lansing, MI 48824-1224, USA

b Center for Integrative Toxicology, Michigan State University, 234 G.M. Trout Building, Michigan State University, East Lansing, MI 48824-1224, USAc Department of Microbiology and Molecular Genetics, Michigan State University, 234 G.M. Trout Building, East Lansing, MI 48824-1224, USA

Received 15 July 2005; revised 23 October 2005; accepted 1 November 2005Available online 20 December 2005

Abstract

The effects of the ribotoxic trichothecene deoxynivalenol (DON) on mitogen-activated protein kinase (MAPK)-mediated IL-8 expression wereinvestigated in cloned human monocytes and peripheral blood mononuclear cells (PBMC). DON (250 to 1000 ng/ml) induced both IL-8 mRNAand IL-8 heteronuclear RNA (hnRNA), an indicator of IL-8 transcription, in the human U937 monocytic cell line in a concentration-dependentmanner. Expression of IL-8 hnRNA, mRNA and protein correlated with p38 phosphorylation and was completely abrogated by the p38 MAPKinhibitor SB203580. DON at 500 ng/ml similarly induced p38-dependent IL-8 protein and mRNA expression in PBMC cultures from healthyvolunteers. Significantly increased IL-6 and IL-1β intracellular protein and mRNA expression was also observed in PBMC treated with DON (500ng/ml) which were also partially p38-dependent. Flow cytometry of PBMC revealed that DON-induced p38 phosphorylation varied amongindividuals relative to both threshold toxin concentrations (25–100 ng/ml) and relative increases in percentages of phospho-p38+ cells. DON-induced p38 activation occurred exclusively in the CD14+ monocyte population. DON was devoid of agonist activity for human Toll-likereceptors 2, 3, 4, 5, 7, 8 and 9. However, two other ribotoxins, emetine and anisomycin, induced p38 phosphorylation in PBMC similarly to DON.Taken together, these data suggest that (1) p38 activation was required for induction of IL-8 and proinflammatory gene expression in the monocyteand (2) DON induced p38 activation in human monocytes via the ribotoxic stress response.© 2005 Elsevier Inc. All rights reserved.

Keywords: Deoxynivalenol; Immunotoxicology; Chemokine; Cytokine; IL-8; IL-6; IL-1β; PBMC; Vomitoxin; Trichothecene; Mycotoxin

Introduction

Mitogen-activated protein kinases (MAPKs) are critical forsignal transduction in the immune response (Dong et al., 2002)and mediate numerous physiological processes including pro-liferation, differentiation and apoptosis in leukocytes (Rao,2001). The p38, extracellular signal-regulated kinase (ERK)and c-Jun N-terminal protein kinase (JNK) MAPK familiesare activated by a variety of stressors, toxins and receptoragonists (Widmann et al., 1999). Translational inhibitors thatbind ribosomes with high affinity can rapidly activate MAPKs

⁎ Corresponding authors. Department of Food Science and Human Nutrition,Michigan State University, East Lansing, 234 G.M. Trout Building,East Lansing, MI 48824-1224, USA. Fax: +1 517 353 8963.

E-mail address: pestka@msu.edu (J.J. Pestka).

0041-008X/$ - see front matter © 2005 Elsevier Inc. All rights reserved.doi:10.1016/j.taap.2005.11.001

in a process known as the “ribotoxic stress response” (Iordanovet al., 1997; Laskin et al., 2002a). A host can potentiallyencounter ribotoxins as a result of food contamination, envi-ronmental exposure or infection (Madsen, 2001; Henghold,2004). Activation of MAPKs by ribotoxins drives both inflam-matory gene expression and apoptosis in the macrophage, mak-ing the innate immune system a critical target for such agents(Hassoun and Wang, 1999, 2000; Mengeling et al., 2001;Cameron et al., 2003; Higuchi et al., 2003; Pestka et al.,2004; Rosser et al., 2004).

The trichothecene mycotoxins are a group of ribotoxinsthat cause anorexia, decreased nutritional efficiency, nausea,vomiting, diarrhea, hemorrhage, leukopenia and shock inanimals and humans (Joffe, 1983; Ueno, 1984; Lorenzana etal., 1985). Deoxynivalenol (DON or vomitoxin), a trichothe-cene that commonly contaminates cereal-based foods, has

236 Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

been linked to human and animal illnesses worldwide (Pestkaand Smolinski, 2005). Trichothecenes are also of concernbecause they might contribute to diseases associated withindoor air mold (Revankar, 2003; Ammann, 2003; Sudakin,2003) and have the potential for use as chemical weapons(Etzel, 2002; Bennett and Klich, 2003; Henghold, 2004;Zapor and Fishbain, 2004).

Depending on exposure frequency and timing of adminis-tration, DON and other trichothecenes stimulate or suppressleukocyte proliferation, cell-mediated immunity, immunoglob-ulin production and host resistance in experimental animals(Pestka, 2003). DON induces inflammation- and immune-related genes in vivo and in vitro. In mice, DON inducesexpression of numerous chemokines and cytokines within 2h in lymphoid tissues (Kinser et al., 2004; Azcona-Olivera etal., 1995; Zhou et al., 1997, 1998; Islam and Pestka, inpress). Of relevance to these in vivo findings, DON upregu-lates expression of chemokines including IL-8 and MIP-2(Chung et al., 2003; Kinser et al., 2004; Sugita-Konishi andPestka, 2001), proinflammatory cytokines including IL-6 andIL-1β (Miller and Atkinson, 1986; Ji et al., 1998; Wong etal., 1998; Kinser et al., 2004; Sugita-Konishi and Pestka,2001) and cyclooxygenase-2 (COX-2) (Moon and Pestka,2002) in cloned murine and human macrophage cultures.Besides affecting cloned cell lines, DON reportedly altersfunction of human primary blood mononuclear cells(PBMC) (Froquet et al., 2003; Sun et al., 2002; Pestka andForsell, 1988; Forsell et al., 1985). Recently, Meky et al.(2001) observed increases in several cytokines in phytohe-magglutinin-stimulated PBMC after exposure to DON.

The capacity of DON to upregulate chemokines mightcontribute to its toxicity. IL-8 is a CXC chemokine thathas multiple functions including initiation of the acute in-flammatory response, chemotaxis and shape change in neu-trophils and induction of histamine and leukotriene release inbasophils (Linevsky et al., 1997; Wu et al., 1997). Celltypes that produce IL-8 include monocytes (Roebuck,1999), macrophages (Sugita-Konishi and Pestka, 2001), neu-trophils (Laskin et al., 2002b), endothelial cells (Pestka etal., 2005) and natural killer cells (Wu et al., 1997). Inducersof IL-8 include tumor necrosis factor (TNF)-α, IL-1β, H2O2,as well as viral and bacterial infections. Extensive damagecan result from IL-8 recruitment of neutrophils to specifictissues thereby triggering activation of an inflammatory re-sponse. Elevated IL-8 concentrations have been seen inserum, urine and at sites of inflammation in a number ofdiseases. Administration of an antibody to IL-8 lessens dis-ease severity in several inflammation models, including lungperfusion injury and lipopolysaccharide/IL-1 induced arthritis(Akahoshi et al., 1994).

The capacity of DON to activate p38, ERK and JNK invitro (Shifrin and Anderson, 1999; Yang et al., 2000; Moonand Pestka, 2002; Zhou et al., 2003b, 2005) and in vivo(Zhou et al., 2003b; Islam and Pestka, in press) likely con-tributes to chemokine and cytokine induction. p38 and ERKare involved in DON-induced transactivation of cytokine andCOX-2 expression with additional involvement of p38 in

DON-mediated mRNA stability (Moon and Pestka, 2002;Moon et al., 2003; Chung et al., 2003). DON markedlyinduces phosphorylation of p38, JNK and ERK as well ascytokine expression and apoptosis in Jurkat cells (Pestka etal., 2005) and notably, SB203580, a specific inhibitor of p38,suppresses DON-induced IL-8 production in this cloned T cellline.

Despite the extensive research conducted on molecular tar-gets of trichothecene in cloned cells and animal models, rela-tively little is known on the effects of these toxins on MAPKsand gene expression in primary human leukocytes. The goal ofthis research was to test the hypothesis that p38 mediatesinduction of IL-8 in human monocytes. Specifically, the roleof MAPKs in DON-induced IL-8 expression was addressed incloned monocytes and related to IL-8 and cytokine productionin PBMC cultures. The results indicate that p38 was critical toDON-induced expression of IL-8 as well as IL-6 and IL-1β inPBMC with the monocyte being a primary target for p38induction. The translational inhibitors emetine and anisomycinsimilarly induced p38 phosphorylation in PBMC, suggestingthat DON's effects are mediated through the ribotoxic stressresponse.

Materials and methods

Chemicals. All reagents were purchased from Sigma Chemical Co. (St.Louis, MO) unless otherwise noted. Chemical inhibitors of p38 (SB203580,2.5 μM), ERK (PD98059, 10 μM) and JNK (SP600125, 1.0 μM) were obtainedfrom Calbiochem (San Diego, CA).

Cultures. U937 cells were originally isolated from the pleural effusion ofan individual with diffuse histiocytic lymphoma and display many character-istics of monocyte (Sundstrom and Nilsson, 1976). This line was obtainedfrom American Type Culture Collection (ATCC; Manassas, VA) and grownin RPMI-1640 medium supplemented with 10% (w/v) heat-inactivated fetalbovine serum albumin (FBS) (Atlanta Biologicals, Lawrenceville, GA) and100 U/ml penicillin and 100 μg/ml streptomycin (Gibco BRL; Rockville,MD) at 37 °C with 6% CO2. Prior to toxin treatment, cells were cultured for24 h to achieve “resting” condition and minimize background stressactivation.

For PBMC, blood (10 ml) was collected in heparin-coated vacutainer(Beckton Dickinson, Franklin Lakes, NJ) from seven healthy volunteers asapproved by the MSU University Committee on Research Involving HumanSubjects. PBMC were immediately isolated from blood using Accuspin Sys-tem-Histopaque-1077 (Sigma, St. Louis, MO). Cells were counted using aCoulter Particle Counter (Coulter Co. Miami, FL) and 0.5 to 1 × 106 cells/mlcultured at 37 °C in a humidified 6% CO2 atmosphere in L-glutamine supple-mented RPMI 1640 medium (Sigma) with 10% (w/v) heat-inactivated FBS, 100U/ml penicillin plus 100 μg/ml streptomycin. Prior to any treatment, cells werecultured for 24 h to minimize background stress resulting from isolationprocedure.

Human embryonic kidney (HEK) 293 cells (InvivoGen) expressing variousToll-like receptor (TLR) and reporter constructs were used in a TLR ligandbinding assay as described below.

MTT viability assay. Cells (1 × 106/ml) were plated (200 μl/well) in 96-wellplates. MTT reagent (3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazoliumbromide) dissolved in Dulbecco's phosphate-buffered saline (PBS) (5 mg/ml)was added (25 μl) and cells incubated at 37 °C in 6% CO2 for 3 h (Marin et al.,1996). Plates were centrifuged at 450 × g for 10 min, and supernatant wasremoved. DMSO (125 μl) was added, and the absorbance was read at 690–570nm on a Vmax Kinetic Microplate Reader (Molecular Devices; Menlo Park,CA).

237Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

Real-time PCR. RNAwas isolated from U937 cells using an RNAqueous Kit(Ambion; Austin, TX) and from human PBMC using Qiagen RNeasy ProtectMini Kit (Valencia, CA) according to the manufacturer's instructions. TotalRNAwas treated with RNAse-free DNAse (Ambion) to remove DNA contam-ination. Reverse transcription real-time PCR was performed using One-stepPCR Master Mix (Applied Biosystems; Foster City, CA) and Pre-developedAssay Reagents (PDAR; Applied Biosystems) for IL-8, IL-6 and IL-1β with18S separately or multiplexed. For heteronuclear RNA (hnRNA), Primer Ex-press software v 1.5 (Applied Biosystems) was used to make primer (basenumbers 3113351–3113369 and 3113431–3113452) and a probe (base numb-ers 3113411–3113429) from the NM_006216 (Genbank; IL-8 DNA). Theprimer/probe set was selected to include an exon–intron junction.

ABI 7700 (96-well) and 7900HT (384-well) thermocyclers at the MSUGenomics Technology and Support Facility were used for real-time PCRusing reaction conditions and PCR program recommended by manufacturer.Fold change was determined using the relative quantitation method. Standardcurves are created using dilutions of total RNA from LPS-treated U937 cells.An equation for the trend line of the data was used to convert the Ct valuesobtained in the assay to nanogram amounts of target. IL-8 values were normal-ized by dividing by the 18S values (endogenous control). Relative expressionwas obtained by dividing all normalized values by the average of the controlnormalized value.

For human PBMC, relative quantification of IL-8, IL-6 and IL-1β expres-sion was carried out using the arithmetic formula method (Applied BiosystemsUser Bulletin 2) as described by Audigé et al. (2003). In brief, the averagethreshold cycle (CT) value for 18S was subtracted from the average CT valuesfor IL-8, IL-6 and IL-1β for each sample to obtain ΔCT values. For each groupof samples, the average ΔCT values for IL-8, IL-6 and IL-1β of the controlsamples were determined and used as calibrators. For each target gene, theΔΔCT values were calculated by subtracting the respective calibrator from theindividual ΔCT value. The relative amount of mRNA for IL-8, IL-6 and IL-1βfrom treatment groups compared with the controls was calculated by using theformula 2−ΔΔCT.

IL-8 evaluation by ELISA. Cell cultures (U937 and human PBMC) werecentrifuged at 450 × g for 10 min, and supernatant was collected and stored at−20 °C. An OptELISA IL-8 kit (BD-Pharmingen; San Diego, CA) was usedwith two modifications. First, the highest standard utilized was 1600 pg/ml,instead of 400 pg/ml. Second, to economize on reagents, 50 μl of dilutedantibody and samples was used per well instead of 100 μl. All samples wereread at 450 nm in a Vmax Kinetic Microplate Reader (Molecular Devices).

p38 Evaluation by Western analysis. Total protein was isolated by lysingU937 cells with hot 1% (w/v) SDS buffer (1 mM sodium ortho-vanadate in 10mM Tris, pH 7.4), sonicating for 30 s and centrifuging at 12,000 × g for 15 min.Supernatant protein was quantitated with a DC Protein Quantitation kit(BioRad, Hercules, CA). Total protein (10 μg) was loaded on a 10% (w/v)SDS-PAGE gel and run at 100 V. The gel was electrotransferred to a PDVFmembrane (Amersham; Piscataway, NJ) overnight at 4 °C. The membrane wasthen blocked in 5% (w/v) bovine serum albumin in 20 mM Tris–HCl, pH 8, and137 mM NaCl containing 0.1% (w/v) Tween 20 (TBST) washed and incubatedwith rabbit anti phospho-p38 (Cell Signaling; Beverly, MA) at 25 °C for 1 h.The membrane was washed and incubated with an anti-rabbit whole IgG (CellSignaling) antibody conjugated with horseradish peroxidase for 1 h at roomtemperature. Bands were visualized with the ECL detection system (Amer-sham). To verify loading, membranes were stripped, blocked and incubatedwith rabbit anti-p38 (IgG) (Cell Signaling) overnight and analyzed as describedabove.

Flow cytometry. For all flow cytometry studies, data from 30,000 cells werecollected in list mode using a Becton Dickinson FACS Vantage (San Jose, CA).FITC and R-PE were excited with a 488 nm argon laser and emission detectedat 530 ± 15 and 570 ± nm, respectively.

For the flow cytometric measurement of intracellular IL-6 and IL-1β pro-duction, human PBMC were treated with DON for 20 h and then Golgi-stop(BD Biosciences) for 2 h prior to cell harvest. Cells were fixed and permeabi-lized (Elson et al., 1995; Jung et al., 1993) and then labeled with FITC-conjugated IL-6 and PE-conjugated IL-1β antibodies (BD Biosciences) using

manufacturer's recommended protocol and dilutions. Scatter gating was used toremove aggregates and debris. Gated unstained cells were placed in the firstfluorescent decade and used to determine the cut off for positive fluorescence.Background fluorescence was determined using vehicle controls stained withIL-6 or IL-1β antibodies stained samples and subtracted from experimental datato determine percentages of IL-6+ and IL-1β+ cells.

For the measurement of intracellular phospho-p38 (Krutzik and Nolan,2003), cultures (5 × 105 cells/ml/well) were treated with an equal volume (1ml) of 3% (v/v) formaldehyde to obtain 1.5% final concentration for 10 minat 25 °C. Cells were removed by centrifugation at 450 × g for 15 min,washed twice with staining buffer (1% [v/v] FBS, 0.09% [w/v] sodiumazide in PBS, pH 7.4) and then permeabilized by resuspending in 700 μl ofice-cold 90% (v/v) methanol and incubating overnight at −20 °C. Cellswere washed twice with staining buffer, incubated with PE-conjugatedphospho-p38 (BD Biosciences) antibodies or PE-conjugated isotype controlusing manufacturer's recommended protocol and dilutions. Isotype stainedcontrol was used to identify the positive regions. PE-positive phospho-p38+

cell populations and mean fluorescence intensities were then determined.For the detection of monocytes containing phosphorylated-p38, dual

color flow cytometry was used (Krutzik and Nolan, 2003). Followingformalin fixation, PBMC (5 × 105) were stained first with FITC-conjugatedanti-mouse CD14 antibody (marker against monocyte) or FITC-conjugatedisotype antibody control (BD Biosciences) according to the manufacturer'sprotocol. Cells were then permeabilized by resuspending in 700 μl of ice-cold 90% methanol and incubated overnight at −20 °C. Cells were stainedwith PE-conjugated phospho-p38 (BD Biosciences) antibody. Isotypestained controls were used to identify the positive regions. FITC- andPE-positive populations and relative fluorescence were then determined.

Toll-like receptor screening. DON was assessed for potential TLR ligandactivity using the InvivoGen TLR Ligand Screening Service (San Diego, CA).TLR stimulation was assessed in HEK293 cells stably transfected with plasmidscontaining human TLR 2, 3, 4, 5, 7, 8 and 9 as well as an NF-κB induciblepromoter plasmid expressing the secreted embryonic alkaline phosphatase(SEAP) reporter gene. TLR gene expression was verified in each cell line byRT-PCR and by activation with authentic TLR ligands and assessing IL-8 pro-duction and/or NF-κB activation. For screening assays, 10 μl aliquot of vehicle,DON (250 ng/ml final concentration) or positive control ligand was added to96-well plates containing 100 ml of reporter cells (5 × 104/ml). Cultures wereincubated for 24 h and analyzed for NF-κB-driven SEAP expression by addi-tion of substrate and measuring absorbance at 650 nm on a Beckman CoulterAD 340C absorbance detector. The positive controls employed were TLR2(heat-killed Listeria monocytogenes, 108 cells/ml), TLR3 (Poly(I:C) 1 μg/ml),TLR4 (Escherichia coli K12 LPS, 100 ng/ml), TLR5 (S. typhimurium flagellin,1 μg/ml), TLR7 (Loxoribine, 1 mM), TLR8 (ssPolyU/LyoVec, 20 μg/ml) andTLR9 (CpG ODN 2006, 1 μg/ml).

Statistics. Data were analyzed with SigmaStat v 1.0 (Jandel Scientific;San Rafael, CA) using one-way analysis of variance (ANOVA) with Stu-dent–Newman–Keuls method for pairwise comparisons. Treatment groupswith a P value of b0.05 were considered significantly different fromcontrol.

Results

DON induced IL-8 gene expression in cloned human U937monocytes is p38-dependent

The effects of DON exposure for 3 h on IL-8 hnRNA andmRNA expression in U937 cells were measured by real-timePCR. DON significantly induced IL-8 mRNA expression in aconcentration-dependent fashion over a range of 250 to 1000ng/ml (Fig. 1A). IL-8 hnRNA, an indicator of IL-8 transcription,was similarly affected, suggesting that DON induced IL-8 tran-scription in this cloned cell line (Fig. 1B).

Fig. 1. Concentration-dependent induction of IL-8 mRNA and hnRNA ex-pression by DON in U937 cells. Cells were treated with 0, 250, 500 and 1000ng/ml DON for 3 h, and then (A) IL-8 mRNA and (B) hnRNAwere measuredby real-time PCR. Bars with different letters are significantly different(P b 0.05). Data are mean ± SEM (n = 3). Results are representative of twoseparate experiments.

Fig. 2. MAPK inhibitors suppress IL-8 gene expression in U937 cells. (A)Cells were treated with SB203580 (2.0 μM), SP600125 (1.0 μM) andPD98059 (10 μM) for 30 min and then with DON. IL-8 hnRNA andmRNA expression were measured by real-time PCR 3 h after DON treatment.Supernatant IL-8 protein was measured by ELISA 12 h after DON treatment.Data are mean ± SEM (n = 11, RNA; n = 5, protein) pooled from two separateexperiments. Bars with different letters are significantly different (P b 0.05).(B) Western analysis of phosphorylated and nonphosphorylated p38 in U937cells treated with DON for 30 min. Cells were treated with DON (500 ng/ml)for 30 min and phospho-p38 and total p38 protein analyzed by Westernanalysis. Results are representative of two separate experiments.

238 Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

Chemical inhibitors for p38 (SB203580), JNK (SP600125)and ERK (PD98059) were used to assess the role of MAPKs inIL-8 expression induced by 500 ng/ml DON. An MTT assayconfirmed that the MAPK inhibitors were not toxic at theconcentrations utilized (data not shown). p38 inhibition com-pletely suppressed DON-induced IL-8 protein, mRNA andhnRNA (Fig. 2A). In contrast, JNK and ERK inhibitors causedmodest inhibition of DON-induced IL-8 hnRNA and IL-8 mRNA expression but not IL-8 protein production. Westernanalysis indicated that DON rapidly induced p38 phosphoryla-tion in U937 cells after 30 min (Fig. 2B). Thus, DON-inducedp38 activation was critical for upregulation of IL-8 in thiscloned human monocyte line.

DON-induced IL-8 in PBMC is p38-dependent

To verify that DON-induced IL-8 in U937 monocytes waspredictive of this toxin's action in primary human cells, theconcentration-dependent effects of DON on IL-8 productionwere assessed in PBMC cultures after 12 h. Exposure to 500ng/ml DON for 12 h induced IL-8 production, whereas lowerDON concentrations (25 and 100 ng/ml) were without effect(Fig. 3A). The maximum IL-8 levels induced ranged from0.4 to 1.4 ng/ml among the individuals which were compa-rable to upregulated levels observed in U937 cells. Consistentwith U937 cells, inclusion of p38 inhibitor completely sup-pressed DON-induced IL-8 production in PBMC (Fig. 3B).

RT-PCR similarly revealed that IL-8 mRNA gene expressionwas induced 70–100-fold by DON (500 ng/ml) among theindividuals tested, and that this was suppressed by the p38inhibitor (Fig. 3C).

Fig. 3. DON induction of IL-8 production in PBMC is p38-dependent. (A) IL-8 production was measured by ELISA 12 h after DON treatment in PBMCcultures from 7 different individuals (S1-S7). Following 30-min treatment withSB (2.5 μM) and 12-h incubation with DON (500 ng/ml), (B) IL-8 protein wasmeasured by ELISA and (C) IL-8 mRNA assessed by RT-PCR. Data aremeans ± SEM (n = 6–9 for ELISA; n = 3 for RT-PCR) pooled from twoexperiments. Bars with different letters within a subject are significantly differ-ent (P b 0.05).

Fig. 4. p38 mediates DON-induced IL-6 expression in PBMC. Cells weretreated with or without SB203580 (2.5 μM) for 30 min and then with DONfor (A) 20 h prior to measurement of intracellular IL-6 flow cytometric analysisor (B)12 h prior to IL-6 mRNA assessment by RT-PCR. Data are means ± SEM(n = 6 for flow cytometry; n = 3 for RT-PCR) pooled from two experiments.Bars without same letters within an individual panel, differ (P b 0.05).

Fig. 5. p38 mediates DON-induced IL-1β expression in PBMC. Cells weretreated with or without SB203580 (2.5 μM) for 30 min and then with DON for(A) 20 h prior to measurement of intracellular IL-1β flow cytometric analysis or(B)12 h prior to IL-1β mRNA assessment by RT-PCR. Data are means ± SEM(n = 6 for flow cytometry; n = 3 for RT-PCR) pooled from two experiments.Bars without same letters within an individual panel, differ (P b 0.05).

239Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

DON also induces p38-dependent IL-6 and IL-1β expression inPBMC

The effects of DON on two proinflammatory cytokines oftenupregulated with IL-8 in mononuclear phagocytes were alsoassessed. Flow cytometry and real-time PCR indicated thatDON at 500 ng/ml also induced expression of IL-6 (Fig. 4)and IL-1β (Fig. 5) in PBMC. Upregulation of intracellular IL-6 (Fig. 4A) and IL-6 mRNA (Fig. 4B) by DON was inhibited bySB203580 in cultures from all individuals except S7. In S7,DON also did not induce significant IL-6 protein and mRNAexpression. DON induction of both intracellular IL-1β protein(Fig. 5A) and IL-8 mRNA (Fig. 5B) was inhibited by SB203580in cultures from all individuals tested. These latter results sug-

240 Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

gested that, as with IL-8, both IL-6 and IL-1β expression wereinduced, in part, by DON-induced p38 activation.

DON induces p38 phosphorylation in PBMC

The effects of DON exposure for 30 min on p38 phos-phorylation in PBMC were assessed by flow cytometry. Amarked upward shift in PE fluorescence corresponding toincreased p38 phosphorylation was observed with increasingDON concentrations (Fig. 6A). Percentages of phospho-p38+

PBMC were significantly increased compared to vehiclecontrol in cultures from all seven individuals tested at 100and 500 ng/ml DON (Fig. 6B). Background levels of phos-pho-p38+ expression were consistent within individuals overthree different blood draws. Cultures from different indivi-duals varied with regard to both basal p38 phosphorylationand response to DON. Although increases in mean fluores-

Fig. 6. DON induces p38 phosphorylation in PBMC. PBMC were treated withDON (500 ng/ml) for 30 min and p38 phosphorylation measured by flowcytometry using PE-labeled anti-phospho-p38 antibody. (A) Representativeflow cytometric histograms from two individuals out of seven used in thisstudy. (B) Percent phospho-p38 positive cells and (C) mean fluorescenceintensities were calculated from the histograms. Data are means ± SEM(n = 6–9) pooled from three separate blood draws. Bars without same letterswithin an individual are significantly different (P b 0.05).

cence intensity in phospho-p38+ cells were similar among allindividuals, percentages of phospho-p38+ PBMC differedamong individuals. Some cultures (S1, S4, S6 and S7) re-sponded to DON concentrations as low as 25 ng/ml, whileother cultures (S2, S3 and S5) did not. Overall, the minimumthreshold for DON's effects on p38 (25 ng/ml) was lowerthan that for IL-8 (500 ng/ml) or IL-6 (100 ng/ml) (data notshown), suggesting this to be a highly sensitive indicator oftoxin exposure.

Emetine and anisomycin also induce p38 phosphorylation inPBMC

The translational inhibitors emetine and anisomycin havebeen previously observed to induce p38 phosphorylation inU937 cells (Zhou et al., 2003a). When compared to DON(500 ng/ml) in cultures at equitoxic concentrations (based oncapacity to inhibit translation), emetine (50 ng/ml) and aniso-mycin (15 ng/ml) induced similar degrees of p38 phosphoryla-tion in PBMC, suggesting that the three chemicals might actthrough a common mechanism (Fig. 7A). Phospho-p38 activa-tion by emetine and anisomycin followed the responsivitypatterns among individuals to that observed for DON (Fig. 7B).

p38 activation occurs in CD14+ monocyte population ofPBMC

Dual color flow cytometry was utilized to determine ifmonocytes were indeed the specific cell population responsi-ble for increased phospho-p38 expression. DON-induced p38phosphorylation in PBMC from individuals that were more(S4) or less (S2) sensitive relative to DON threshold occurredexclusively in the CD14+ monocyte population (Fig. 8A).Individuals S4 and S5 responded to a greater extent (∼3-fold) than did individuals S2 and S7 (∼1-fold) (Fig. 8B).Lower DON thresholds for p38 activation appeared to corre-spond with higher numbers of CD14+ cells in a PBMCculture (Fig. 8A).

DON does not activate TLRs

Since monocytes could also potentially be activated by DONthrough TLRs, the capacity of this toxin to function as a ligandfor these receptors was assessed by measuring SEAP expres-sion in HEK293 cells stably transfected with different humanTLRs. Positive controls for TLR2, 3, 4, 5, 7, 8 and 9 inducedrobust enzyme expression in this model, however, DON at 250ng/ml was devoid of activity (Fig. 9). Thus, activation ofmonocytes by DON might not involve direct activation ofTLRs.

Discussion

The capacity of DON to upregulate or downregulate im-mune function is well established (Bondy and Pestka, 2000).An underlying mechanism for these effects might alter chemo-kine and cytokine gene expression, which could disrupt normal

Fig. 8. DON induces p38 phosphorylation CD14+ population of PBMC cul-tures. PBMC were treated with DON (500 ng/ml) for 30 min and stained withFITC-labeled anti CD14 and PE-labeled anti phospho-p38 and then analyzed byflow cytometry. (A) Representative flow cytometric diagram of two individuals.(B) Percent CD14+ and phospho-p38+ double positive cells, which were calcu-lated from the upper right quadrants of flow cytometry histograms. Data aremeans ± SEM (n = 6) pooled from two separate experiments. Bars with differentletters within a subject are significantly different (P b 0.05).

Fig. 9. DON does not activate TLRs. HEK 293 cells stably transfected withhuman TLRs were incubated with vehicle, DON at 250 ng/ml or positive controlfor 24 h and then assessed for NF-κB-driven secreted alkaline phosphatasereporter activity. Data are means ± SEM (n = 4) pooled from two experiments.

Fig. 7. Ribotoxins emetine and anisomycin induce p38 phosphorylation inPBMC. Cells were cultured with emetine (50 ng/ml), anisomycin (15 ng/ml)and DON (500 ng/ml) for 30 min to induce phospho-p38. (A) Representativeflow cytometric histogram of two individuals. (B) Percent phospho-p38 positivecells. Data are means ± SEM (n = 6) pooled from two separate experiments.Bars without same letters within an individual differ (P b 0.05).

241Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

regulation of a wide variety of immune responses in bothpositive and negative manner (Pestka et al., 2004). This studyexplored the relationship between the IL-8 and p38 responses inboth cloned human monocyte cell line and in PBMC culturesfrom healthy individuals. Novel findings reported here werethat (1) p38 is activated in U937 cells by DON, and that thisactivation contributed to IL-8 mRNA and protein expression;(2) PBMC were similarly susceptible to p38-mediated IL-8 ex-pression; (3) p38 signaling also contributed, in part, to theinduction of the proinflammatory cytokines IL-1β and IL-6;(4) the monocyte was the primary cell type associated withDON-induced p38 phosphorylation in PBMC; and (5) p38phosphorylation could be induced in PBMC by other ribotoxicchemicals.

From the perspective of human hazard assessment, theobservations reported here are also important because theysuggest for the first time that PBMC from specific individualsvary relative to their sensitivity to DON. Specifically, four ofseven individuals exhibited greater responsiveness to DON-induced p38 phosphorylation relative to threshold toxin con-centration and magnitude of response. Although variabilitymight occur because of a hormonal or physiological (e.g.,diet, medication, disease) differences, these effects were con-sistent within blood collected from individuals at random times.Notably, cultures with the highest monocyte populationsappeared to be most sensitive to DON. Thus, monocyte levelsmight be critical in determining the degree to which cultures

from different human subjects responded to DON. A caveat tothis interpretation is that only a small number of human sub-jects were tested here. Different responsivity might alternative-ly be explained by genetic polymorphisms among individuals

242 Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

relative to toxin metabolism, cell–target interaction or signalingresponses.

We have previously observed that chemokine and cytokinemRNA expressions are induced by DON in rapid and transientfashion within lymphoid tissue in the mouse. Both MAPKs andtranscription factors associated upstream with these genes areactivated prior to or within these time windows (Pestka et al.,2004). p38 has been found to be essential for DON induction ofinflammatory and chemokine genes in several in vitro models(Moon and Pestka, 2002; Chung et al., 2003). Since IL-8 is abiologically potent molecule (Linevsky et al., 1997; Wu et al.,1997), it is very important that its production be strictly regu-lated by cells. The observed IL-8 hnRNA increases suggest thatDON upregulated IL-8 transcription through p38. A phenotype-dependent combination of two or three transcription factors,usually AP-1, C/EBPβ and/or NF-κB (Wu et al., 1997), isrequired to achieve the highest level of IL-8 transcription ac-tivity. All of these factors can be affected by p38, and we havefurther observed that DON induces activation of these proteinsin vivo and in vitro (Zhou et al., 2003a; Wong et al., 2002).

IL-8 gene expression is also subject to post-transcriptionalregulation. AU-rich elements, binding sites for proteins thatstabilize the mRNA transcript, are present in the IL-8 3′ UTR(Jung et al., 1993; Suswam et al., 2005). A 357-bp elementwithin the IL-8 3′ UTR has been determined to destabilizemRNA from a CAT expression vector. Winzen et al. (2004)found that cells expressing a constitutively active mutant ofMK2 pathway have increased IL-8 mRNA stability, suggestinga critical role for p38. Further study of post-transcriptionalmechanisms for DON-induced IL-8 involving p38 is thuswarranted.

These effective DON concentrations (25 to 500 ng/ml) areconsistent with those seen in lymphoid tissues of mice exposedto the toxin orally at 5 and 25 mg/kg (Azcona-Olivera et al.,1995). It is further notable that although DON concentrations aslow as 25 ng/ml could activate p38 in human PBMC, thethreshold for IL-8 induction was 500 ng/ml DON. Thus, p38activation after 30 min per se was not predictive of IL-8 re-sponse. Further assessment of the kinetics of p38 induction andits relationship to IL-8 expression would therefore be desirable.Trends for IL-6 and IL-1β induction by DON corresponded tothe activation patterns of p38 among human subjects. Culturesthat exhibited higher p38 activation also had correspondinglyhigher induction of IL-6 and IL-1β cytokines. As with IL-8,expression of these inflammatory cytokines was suppressed byp38 inhibitor. Thus, p38 activation also played the critical rolein inflammatory gene upregulation in PBMC.

The mechanisms by which DON specifically stimulates p38activation in monocytes are not completely clear. Althoughmonocytes bear TLRs, DON showed no TLR ligand activityin the reporter assay employed here. HEK 293 cells were usedas reporters here because they are naturally devoid of TLRs andthus can be used to monitor effects of individual TLRs stablycloned into the cell line. A caveat to this approach might be thatsince authentic primary or cloned human monocytes were notused, the reporter system may not fully relate to observedeffects on p38 activation and downstream gene expression.

This could be addressed in the future using siRNA or murineknockout approaches to more directly assess specific involve-ment of TLRs in DON-activated monocytes.

The underlying mechanism for DON-induced p38 activa-tion is likely to involve the ribotoxic stress response whichhas been demonstrated in cells exposed to translational inhi-bitors such as DON, T-2 toxin, anisomycin, ricin and α-sarcin(Iordanov et al., 1997; Pestka et al., 2004; Laskin et al.,2002b). Alteration of 28s rRNA by these agents is postulatedto be an initiation signal for activation of MAPKs. Ourlaboratory previously has demonstrated, in macrophages, thatDON activates both double-stranded RNA-activated proteinkinase (PKR) (Zhou et al., 2003b) and Src-family kinases(SFK) (Zhou et al., 2005), both of which are upstream trans-ducers of MAPK activation. The observation that p38 wasactivated in PBMC by DON, emetine and anisomycin inPBMC in cultures from all individuals tested is consistentwith findings in RAW 264.7 murine macrophages (Zhou etal., 2003b, 2005). Further study is necessary to elucidate themechanisms by which DON activates p38 in mononuclearphagocytes and the linkage with PKR and SFKs.

Acknowledgments

This material is based upon work supported by the U.S.Department of Agriculture, under Agreement No.59-0790-4-119, a cooperative project with the U.S. Wheat and BarleyScab Initiative. Any opinions, findings, conclusions or recom-mendations expressed in this publication are those of theauthors and do not necessarily reflect the view of the U.S.Department of Agriculture. The work was further supportedby Public Health Service Grants ES03553 and DK58833. Wethank Kelly Ross, Dr. Louis King, Sarah Godbehere andAnnette Thelen for the technical assistance and Mary Rosnerfor the manuscript preparation.

References

Akahoshi, T., Endo, H., Kondo, H., Kashiwazaki, S., Kasahara, T., Mukaida, N.,Harada, A., Matsushima, K., 1994. Essential involvement of interleukin-8 inneutrophil recruitment in rabbits with acute experimental arthritis induced bylipopolysaccharide and interleukin-1. Lymphokine Cytokine Res. 13,113–116.

Ammann, H.M., 2003. Is indoor mold contamination a threat to health?J. Environ. Health 66 (Pt. 2), 47–49.

Audigé, A., Yu, Z.R., Frey, B.M., Uehlinger, D.E., Frey, F.J., Vogt, B., 2003.Epithelial sodium channel (ENaC) subunit mRNA and protein expression inrats with puromycin aminonucleoside-induced nephrotic syndrome. Clin.Sci. 104, 389–395.

Azcona-Olivera, J.I., Ouyang, Y., Murtha, J., Chu, F.S., Pestka, J.J., 1995.Induction of cytokine messenger-RNAs in mice after oral-exposure to thetrichothecene vomitoxin (deoxynivalenol)-relationship to toxin distributionand protein-synthesis inhibition. Toxicol. Appl. Pharmacol. 133, 109–120.

Bennett, J.W., Klich, M., 2003. Mycotoxins. Clin. Microbiol. Rev. 16, 497–516.Bondy, G.S., Pestka, J.J., 2000. Immunomodulation by fungal toxins. J. Toxicol.

Environ. Health, Part B. Crit. Rev. 3, 109–143.Cameron, P., Smith, S.J., Giembycz, M.A., Rotondo, D., Plevin, R., 2003.

Verotoxin activates mitogen-activated protein kinase in human peripheral

243Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

blood monocytes: role in apoptosis and proinflammatory cytokine release.Br. J. Pharmacol. 140, 1320–1330.

Chung, Y.J., Zhou, H.R., Pestka, J.J., 2003. Transcriptional and posttranscrip-tional roles for p38 mitogen-activated protein kinase in upregulation ofTNF-alpha expression by deoxynivalenol (vomitoxin). Toxicol. Appl.Pharmacol. 193, 188–201.

Dong, C., Davis, R.J., Flavell, R.A., 2002. MAP kinases in the immuneresponse. Annu. Rev. Immunol. 20, 55–72.

Elson, L.H., Nutman, T.B., Metcalfe, D.D., Prussin, C., 1995. Flow cytometricanalysis for cytokine production identifies T helper 1, T helper 2, and Thelper 0 cells within the human CD4+. J. Immunol. 154, 4294–4301.

Etzel, R.A., 2002. Mycotoxins. JAMA 287, 425–427.Forsell, J.H., Kateley, J.R., Yoshizawa, T., Pestka, J.J., 1985. Inhibition of

mitogen-induced blastogenesis in human lymphocytes by T-2 toxin and itsmetabolites. Appl. Environ. Microbiol. 49, 1523–1526.

Froquet, R., Arnold, F., Batina, P., Parent-Massin, D., 2003. Do trichothecenesreduce viability of circulating blood cells and modify haemostasisparameters? Mycopathologia 156, 349–356.

Hassoun, E.A., Wang, X., 1999. Time- and concentration-dependent productionof superoxide anion, nitric oxide, DNA damage and cellular death by ricin inthe J774A.1 macrophage cells. J. Biochem. Mol. Toxicol. 13, 179–185.

Hassoun, E., Wang, X., 2000. Ricin-induced toxicity in the macrophageJ744A.1 cells: the role of TNF-alpha and the modulation effects of TNF-alpha polyclonal antibody. J. Biochem. Mol. Toxicol. 14, 95–101.

Henghold, W.B., 2004. Other biologic toxin bioweapons: ricin, staphylococcalenterotoxin B, and trichothecene mycotoxins. Dermatol. Clin. 22, 257–262.

Higuchi, S., Tamura, T., Oda, T., 2003. Cross-talk between the pathways leadingto the induction of apoptosis and the secretion of tumor necrosis factor-alphain ricin-treated RAW 264.7 cells. J. Biochem. (Tokyo) 134, 927–933.

Iordanov, M.S., Pribnow, D., Magun, J.L., Dinh, T.H., Pearson, J.A., Chen, S.L.Y., Magun, B.E., 1997. Ribotoxic stress response: activation of the stress-activated protein kinase JNK1 by inhibitors of the peptidyl transferasereaction and by sequence-specific RNA damage to the alpha-sarcin/ricinloop in the 28S rRNA. Mol. Cell. Biol. 17, 3373–3381.

Islam, Z., Pestka, J.J., in press. LPS priming potentiates and prolongsproinflammatory cytokine response to the trichothecene deoxynivalenol inthe mouse. Toxicol. Appl. Pharmacol.

Ji, G.E., Park, S.Y., Wong, S.S., Pestka, J.J., 1998. Modulation of nitric oxide,hydrogen peroxide and cytokine production in a clonal macrophage modelby the trichothecene vomitoxin (deoxynivalenol). Toxicology 125,203–214.

Joffe, A.Z., 1983. Environmental conditions conducive to Fusarium toxinformation causing serious outbreaks in animals and man. Vet. Res.Commun. 7, 187–193.

Jung, T., Schauer, U., Heusser, C., Neumann, C., Rieger, C., 1993. Detection ofintracellular cytokines by flow cytometry. J. Immunol. Methods 159,197–207.

Kinser, S., Jia, Q.S., Li, M.X., Laughter, A., Cornwell, P.D., Corton, J.C.,Pestka, J.J., 2004. Gene expression profiling in spleens of deoxynivalenol-exposed mice: immediate early genes as primary targets. J. Toxicol. Environ.Health, A 67, 1423–1441.

Krutzik, P.O., Nolan, G.P., 2003. Intracellular phospho-protein stainingtechniques for flow cytometry: monitoring single cell signaling events.Cytometry 55A, 61–70.

Laskin, J.D., Heck, D.E., Laskin, D.L., 2002a. The ribotoxic stress response as apotential mechanism for MAP kinase activation in xenobiotic toxicity.Toxicol. Sci. 69, 289–291.

Laskin, J.D., Heck, D.E., Laskin, D.L., 2002b. The ribotoxic stress response as apotential mechanism for MAP kinase activation in xenobiotic toxicity.Toxicol. Sci. 69, 289–291.

Linevsky, J.K., Pothoulakis, C., Keates, S., Warny, M., Keates, A.C., Lamont, J.T., Kelly, C.P., 1997. IL-8 release and neutrophil activation by Clostridiumdifficile toxin-exposed human monocytes. Am. J. Physiol. 273,G1333–G1340.

Lorenzana, R.M., Beasley, V.R., Buck, W.B., Ghent, A.W., 1985. ExperimentalT-2 toxicosis in swine: II. Effect of intravascular T-2 toxin on serumenzymes and biochemistry, blood coagulation, and hematology. Fundam.Appl. Toxicol. 5, 893–901.

Madsen, J.M., 2001. Toxins as weapons of mass destruction-A comparison andcontrast with biological-warfare and chemical-warfare agents. Clin. Lab.Med. 21, 593–605.

Marin, M.L., Murtha, J., Dong, W., Pestka, J.J., 1996. Effects of mycotoxins oncytokine production and proliferation in EL-4 thymoma cells. J. Toxicol.Environ. Health 48, 379–396.

Meky, F.A., Hardie, L.J., Evans, S.W., Wild, C.P., 2001. Deoxynivalenol-induced immunomodulation of human lymphocyte proliferation andcytokine production. Food Chem. Toxicol. 39, 827–836.

Mengeling, W.L., Vorwald, A.C., Cornick, N.A., Lager, K.M., Moon, H.W.,2001. In vitro detection of Shiga toxin using porcine alveolar macrophages.J. Vet. Diag. Inv. 13, 421–424.

Miller, K., Atkinson, H.A., 1986. The in vitro effects of trichothecenes on theimmune system. Food Chem. Toxicol. 24, 545–549.

Moon, Y., Pestka, J.J., 2002. Vomitoxin-induced cyclooxygenase-2 geneexpression in macrophages mediated by activation of ERK and p38 butnot JNK mitogen-activated protein kinases. Toxicol. Sci. 69, 373–382.

Moon, Y., Uzarski, R., Pestka, J.J., 2003. Relationship of trichothecene structureto COX-2 induction in the macrophage: selective action of Type B (8-keto)trichothecenes. J. Toxicol. Environ. Health, A 66, 1967–1983.

Pestka, J.J., 2003. Deoxynivalenol-induced IgA production and IgA nephrop-athy-aberrant mucosal immune response with systemic repercussions.Toxicol. Lett. 140–141, 287–295.

Pestka, J.J., Forsell, J.H., 1988. Inhibition of human lymphocyte transformationby the macrocyclic trichothecenes roridin A and verrucarin A. Toxicol. Lett.41, 215–222.

Pestka, J.J., Smolinski, A.T., 2005. Deoxynivalenol: toxicology and potentialeffects on humans. J. Toxicol. Environ. Health, Part B. Crit. Rev. 8,39–69.

Pestka, J.J., Zhou, H.R., Moon, Y., Chung, Y.J., 2004. Cellular andmolecular mechanisms for immune modulation by deoxynivalenol andother trichothecenes: unraveling a paradox. Toxicol. Lett. 153, 61–73.

Pestka, J.J., Uzarski, R.L., Islam, Z., 2005. Induction of apoptosis and cytokineproduction in the Jurkat human T cells by deoxynivalenol: role of mitogen-activated protein kinases and comparison to other 8-ketotrichothecenes.Toxicology 206, 207–219.

Rao, K.M.K., 2001. MAP kinase activation in macrophages. J. Leukocyte Biol.69, 3–10.

Revankar, S.G., 2003. Clinical implications of mycotoxins and stachybotrys.Am. J. Med. Sci. 325, 262–274.

Roebuck, K.A., 1999. Regulation of interleukin-8 gene expression. J. InterferonCytokine Res. 19, 429–438.

Rosser, E.M., Morton, S., Ashton, K.S., Cohen, P., Hulme, A.N., 2004.Synthetic anisomycin analogues activating the JNK/SAPK1 and p38/SAPK2 pathways. Org. Biomol. Chem. 2, 142–149.

Shifrin, V.I., Anderson, P., 1999. Trichothecene mycotoxins trigger a ribotoxicstress response that activates c-Jun N-terminal kinase and p38 mitogen-activated protein kinase and induces apoptosis. J. Biol. Chem. 274,13985–13992.

Sudakin, D.L., 2003. Trichothecenes in the environment: relevance to humanhealth. Toxicol. Lett. 143, 97–107.

Sugita-Konishi, Y., Pestka, J.J., 2001. Differential upregulation of TNF-alpha,IL-6, and IL-8 production by deoxynivalenol (vomitoxin) and other 8-ketotrichothecenes in a human macrophage model. J. Toxicol. Environ.Health, A 64, 619–636.

Sun, X.M., Zhang, X.H., Wang, H.Y., Cao, W.J., Yan, X., Zuo, L.F., Wang, J.L.,Wang, F.R., 2002. Effects of sterigmatocystin, deoxynivalenol and aflatoxinG(1) on apoptosis of human peripheral blood lymphocytes in vitro'.Biomed. Environ. Sci. 15, 145–152.

Sundstrom, C., Nilsson, K., 1976. Establishment and characterization of ahuman histiocytic lymphoma cell line (U-937). Int. J. Cancer 17,565–577.

Suswam, E.A., Nabors, L.B., Huang, Y., Yang, X., King, P.H., 2005. IL-1betainduces stabilization of IL-8 mRNA in malignant breast cancer cells via the3′ untranslated region: involvement of divergent RNA-binding factors HuR,KSRP and TIAR. Int. J. Cancer 113, 911–919.

Ueno, Y., 1984. Toxicological features of T-2 toxin and related trichothecenes.Fundam. Appl. Toxicol. 4, S124–S132.

244 Z. Islam et al. / Toxicology and Applied Pharmacology 213 (2006) 235–244

Widmann, C., Gibson, S., Jarpe, M.B., Johnson, G.L., 1999. Mitogen-activatedprotein kinase: conservation of a three-kinase module from yeast to human.Physiol. Rev. 79, 143–180.

Winzen, R., Gowrishankar, G., Bollig, F., Redich, N., Resch, K., Holtmann, H.,2004. Distinct domains of AU-rich elements exert different functions inmRNA destabilization and stabilization by p38 mitogen-activated proteinkinase or HuR. Mol. Cell. Biol. 24, 4835–4847.

Wong, S.S., Zhou, H.R., Marin-Martinez, M.L., Brooks, K., Pestka, J.J., 1998.Modulation of IL-1beta, IL-6 and TNF-alpha secretion and mRNAexpression by the trichothecene vomitoxin in the RAW 264.7 murinemacrophage cell line. Food Chem. Toxicol. 36, 409–419.

Wong, S.S., Zhou, H.R., Pestka, J.J., 2002. Effects of vomitoxin (deoxyniva-lenol) on the binding of transcription factors AP-1, NF-kappaB, and NF-IL6in raw 264.7 macrophage cells. J. Toxicol. Environ. Health, A 65,1161–1180.

Wu, G.D., Lai, E.J., Huang, N., Wen, X., 1997. Oct-1 and CCAAT/enhancer-binding protein (C/EBP) bind to overlapping elements within theinterleukin-8 promoter. The role of Oct-1 as a transcriptional repressor.J. Biol. Chem. 272, 2396–2403.

Yang, G.H., Jarvis, B.B., Chung, Y.J., Pestka, J.J., 2000. Apoptosis induction bythe satratoxins and other trichothecene mycotoxins: relationship to ERK,

p38 MARK, and SAP/JNK activation. Toxicol. Appl. Pharmacol. 164,149–160.

Zapor, M., Fishbain, J.T., 2004. Aerosolized biologic toxins as agents of warfareand terrorism. Respir. Care Clin. N. Am. 10, 111–122.

Zhou, H.R., Yan, D., Pestka, J.J., 1997. Differential cytokine mRNA expressionin mice after oral exposure to the trichothecene vomitoxin (deoxynivalenol):dose response and time course. Toxicol. Appl. Pharmacol. 144, 294–305.

Zhou, H.R., Yan, D., Pestka, J.J., 1998. Induction of cytokine gene expression inmice after repeated and subchronic oral exposure to vomitoxin (deoxyni-valenol): differential toxin-induced hyporesponsiveness and recovery.Toxicol. Appl. Pharmacol. 151, 347–358.

Zhou, H.R., Islam, Z., Pestka, J.J., 2003a. Rapid, sequential activation ofmitogen-activated protein kinases and transcription factors precedesproinflammatory cytokine mRNA expression in spleens of mice exposedto the trichothecene vomitoxin. Toxicol. Sci. 72, 130–142.

Zhou, H.R., Lau, A.S., Pestka, J.J., 2003b. Role of double-stranded RNA-activated protein kinase R (PKR) in deoxynivalenol-induced ribotoxic stressresponse. Toxicol. Sci. 74, 335–344.

Zhou, H.R., Jia, Q.S., Pestka, J.J., 2005. Ribotoxic stress response to thetrichothecene deoxynivalenol in the macrophage involves the Src familykinase Hck. Toxicol. Sci. 85, 916–926.