Differential induction of glucocorticoid-dependent apoptosis in murine lymphoid subpopulations in vivo following coexposure to lipopolysaccharide and vomitoxin (deoxynivalenol)

Differential induction of glucocorticoid-dependent apoptosis in murinelymphoid subpopulations in vivo following coexposure to

lipopolysaccharide and vomitoxin (deoxynivalenol)

Zahidul Islam,a Louis E. King,c Pamela J. Fraker,a,c and James J. Pestkaa,b,d,*a Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, USA

b Department of Microbiology and Molecular Genetics, Michigan State University, East Lansing, MI 48824, USAc Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824, USA

d Institute of Environmental Toxicology, Michigan State University, East Lansing, MI 48824, USA

Received 16 July 2002; accepted 6 November 2002

Abstract

Lipopolysaccharide (LPS) and vomitoxin (VT) synergistically induce glucocorticoid- mediated apoptotic cell death in lymphoid tissuesof the mouse. Based on the known effects of glucocorticoids, it was hypothesized that the combined exposure to LPS and VT targetsimmature lymphocyte populations. To test this hypothesis, we quantified the effects of VT and LPS on apoptosis induction in T lymphocytesubsets in thymus and B lymphocyte subsets in Peyer’s patches and bone marrow. Flow cytometry revealed that a single dose of LPS (0.1mg/kg body wt ip) together with VT (12.5 mg/kg body wt po) promoted apoptosis of immature (CD4�CD8�, CD4�CD8�) and mature(CD4�CD8�) thymocytes at 12 h with a subsequent reduction of these populations being detectable at 24 h. RU 486, a glucocorticoidreceptor antagonist, significantly abrogated apoptosis in CD4�CD8�, CD4�CD8�, and CD4�CD8� subsets and also prevented loss in cellnumbers. In Peyer’s patches, mature-B lymphocytes (B220�IgM�IgD�) underwent apoptosis and, in bone marrow, pro/pre-B lymphocytes(B220�IgM�IgD�) and mature-B lymphocytes (B220�IgM�IgD�) underwent apoptosis at 12 h after toxin coexposure. RU 486 blockedLPS � VT-induced apoptosis of the aforementioned subsets in Peyer patches and bone marrow at 12 h. Taken together, these data suggestthat LPS can interact with VT in mice to induce the glucocorticoid-driven apoptotic loss of immature thymocytes and cytotoxic Tlymphocytes in thymus, mature-B lymphocytes in Peyer’s patch, and pro/pre-B lymphocytes and mature-B lymphocytes in bone marrowin mice.© 2003 Elsevier Science (USA). All rights reserved.

Keywords: Lipopolysaccharide; Deoxynivalenol; Lymphocytes; Apoptosis; Vomitoxin; RU 486; Thymus; Peyer’s patch; Bone marrow; CD4; CD8; B220;IgM; IgD

Introduction

Lipopolysaccharide (LPS), an active component of en-dotoxin, is a constituent of the outer membrane of gram-negative bacterial cell walls. LPS has been extensively stud-ied as a mediator of inflammation and is a majorcontributing factor to bacterial pathogenesis (Hewett andRoth, 1993). Exposure to low doses of LPS initiates a

modest and noninjurious inflammatory response in experi-mental animals, but their organs may be sensitized to injuryby a number of xenobiotic agents (Yee et al., 2000). Forexample, low-dose exposure to LPS in mice can render theliver more sensitive to T-2 toxin, aflatoxin B1, carbon tet-rachloride, galactosamine, ethanol, and allyl alcohol. LPSalso potentiates mercury-induced nephrotoxicity in themouse (Rumbeiha et al., 2000). These and other examplessuggest that humans exposed to low doses of LPS mayexperience increased susceptibility to the toxicity of certainchemicals.

Vomitoxin (VT or deoxynivalenol), a trichothecene my-

* Corresponding author. 234 G.M.Trout Building, Michigan State Uni-versity, East Lansing, MI 48824-1224. Fax: �1-517-353-8963.

E-mail address: [email protected] (J.J. Pestka).

R

Available online at www.sciencedirect.com

Toxicology and Applied Pharmacology 187 (2003) 69–79 www.elsevier.com/locate/taap

0041-008X/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved.doi:10.1016/S0041-008X(02)00031-5

cotoxin produced by Fusarium graminearum, commonlycontaminates human and animal dietary staples such aswheat, corn, and barley (Rotter et al., 1996). Since VT isrecalcitrant to milling and processing, it could possibly bepresent at ppm levels in grain-based food products con-sumed by humans. Besides exposure from foods, trichoth-ecenes can also be encountered by humans via inhalation.Trichothecenes are present in spores of molds that grow oncellulose-based building materials (Bondy and Pestka,2000). Hallmarks of experimental and accidental exposureto high doses of trichothecenes include rapid diminution oflymphoid tissue and lymphopenia that precede death viacirculatory shock-like syndrome (Bondy and Pestka, 2000).B and T lymphocytes appear to be sensitive to relatively lowlevels of the trichothecene T-2 (Corrier and Wagner, 1988).Immature T lymphocytes (Islam et al., 1998) and B lym-phocyte precursor cells (Holladay et al., 1995) also repre-sent sensitive targets of T-2 toxin exposure.

Trichothecene immunotoxicity is synergistically en-hanced by low-dose LPS exposure in mice (Tai and Pestka,1988; Taylor et al., 1991). Pronounced thymic and spleniclymphocyte depletion was characteristically observed inthese studies. Coexposure to subtoxic doses of LPS and VTwere found to markedly induce apoptotic cell death inlymphoid tissues, most notably in the thymus, Peyer’spatches, bone marrow, and spleen of mice (Zhou et al.,1999, 2000). Relatedly, we have recently shown that LPS �VT-induced lymphocyte apoptosis is glucocorticoid depen-dent (Islam et al., 2002).

It has previously been reported that both immature T andB lymphocytes are extraordinarily sensitive to glucocorti-coid-mediated cell death (Ashwell et al., 2000; Laakko andFraker, 2002). Dexamethasone also induces apoptosis inmouse and human mature T lymphocytes (Brunetti et al.1995; Miglioratti et al. 1994). The purpose of this study wasto test the hypothesis that LPS and VT coexposure targetsimmature lymphocyte populations. To test this hypothesis,we quantified specific subpopulations in lymphoid organsthat were sensitive to apoptosis induction following LPSand VT coexposure. The experiments focused on the quan-titation of apoptotic subpopulations in thymus, Peyer’s

patches, and bone marrow using multicolor flow cytometryand included T and B cells at various stages of differentia-tion (Fig. 1). The results suggested that coexposure of LPSand VT induced the glucocorticoid-dependent apoptotic lossof immature thymocytes and cytotoxic T lymphocytes inthymus, mature-B lymphocytes in Peyer’s patch, and pro/pre-B lymphocytes and mature-B lymphocytes in bone mar-row of mice.

Materials and methods

Animals

Male B6C3F1 (C57B1/6J�C3H/HeJ) mice (7 weeks)were obtained from Charles River (Portage, MI). Mice werehoused three per cage under a 12-h light/dark cycle andprovided standard rodent chow and water ad libitum. Allanimal handling was conducted in accordance with recom-mendations established by the National Institutes of Health.Experiments were designed to minimize the number ofanimals required to adequately test the proposed hypothesisand were approved by the Michigan State University Lab-oratory Animal Research Committee.

Reagents and antibodies

LPS derived from Escherchia coli serotype O111:B4with an activity of 1,500,000 EU/mg, deoxynivalenol (VT),RU 486 (mifepristone), propidium iodide (PI), and 4�,6-diamidino-2-phenylindole (DAPI) were purchased fromSigma Chemical Co. (St. Louis, MO). Fluoroscein isothio-cyanate (FITC) conjugated rat anti-mouse CD4 (clone:H129.19), R- phycoerythrin (R-PE)-conjugated rat anti-mouse CD8a (clone: 53-6.7), Cy-Chrome (Cy-C) conju-gated rat anti-mouse B220 (CD45R, clone: RA3-6B2),FITC-conjugated rat anti-mouse IgD (clone: 11-26c.2a),R-PE conjugated IgM (clone: R6-60.2) as well as the iso-type-matched controls for FITC (clone: R35-95), R-PE(clone: R35-95), and Cy-C (clone: R35-95) with unrelated

Fig. 1. Schematic representation of murine B and T lymphocyte development.

70 Z. Islam et al. / Toxicology and Applied Pharmacology 187 (2003) 69–79

specificities were purchased from Pharmingen (San Diego,CA).

Experimental design

LPS was dissolved in tissue culture-grade, endotoxin-free water (Sigma), aliquoted, and stored at �80°C. VT wasalso dissolved in tissue culture-grade, endotoxin-free waterand stored at 4°C. LPS (0.1 mg/kg body wt) was injected ip(250 �l/mouse). VT (12.5 mg/kg body wt) was gavaged po(250 �l/mouse) 5 min after ip injection of LPS. In someexperiments, RU 486 was dissolved in dimethyl sulfoxide(Sigma) and injected ip (20 mg/kg in 50 �l per mouse) 30min before LPS � VT treatment. Food and water werewithdrawn from cages 1 h before toxin administration. In atypical experiment, mice were given vehicle (VH ip) plusVH (po) [VH], LPS (ip) plus VH (po) [LPS], VH (ip) plusVT (po) [VT], or LPS (ip) plus VT (po) [LPS � VT].

Cell preparation and counting

Mice were euthanized by cervical dislocation undermetaflurane anesthesia. Thymus, Peyer’s patches (from in-testine), and spleen were immediately removed and sub-merged into ice- cold Dulbecco’s PBS (Sigma). Single-cellsuspensions were prepared according to the method of Islamet al. (1998). Briefly, cells were released from thymus,Peyer’s patches, and spleen by gently pressing tissuethrough a 100-mesh screen (Collector Tissue Sieve, BellcoGlass, Vineland, NJ) with a glass pestle. Bone marrow cellscollected from femur and splenocytes were treated witherythrocyte lysing buffer (144 mM ammonium chloride and17 mM Tris, pH 7.2) for 5 min at room temperature toremove erythrocytes and then washed twice with PBS. Thecell suspension was passed through a 41-�m nylon sieve(Spectrum Laboratories, Laguna Hills, CA) and cell numberwas determined using a Coulter Particle Counter (Coulter,Miami, FL). Cell viability was assessed by trypan blueexclusion.

Quantitation of apoptosis by flow cytometry

Apoptosis in thymus, Peyer’s patches, bone marrow, andspleen was quantified by flow cytometric cell cycle analysisas described previously (Pestka et al., 1994). Briefly, thy-mus, Peyer’s patch, bone marrow, and spleen cells (2 �106) were resuspended in 0.2 ml PBS. Following addition of0.2 ml heat-inactivated fetal bovine serum, cells were im-mediately fixed by dropwise addition of 1.2 ml ice-cold70% (v/v) ethanol with gentle mixing, and then held at 4°Covernight. Cells were washed and incubated in 1 ml of DNAstaining reagent [PBS containing 50 �g/ml PI, 50 �g/mlRNase A, 0.1 mM EDTA disodium, and 0.1% (v/v) TritonX- 100] at room temperature for 1 h and stored on ice untilanalysis. Cell cycle distribution for single cells was mea-sured with a Becton Dickinson FACS Vantage using Lysis

II software (San Jose, CA). Data from 20,000 cells werecollected in list mode. The 488 line of an argon laser wasused to excite PI and fluorescence was detected at 630 � 15nm. Cells were gated to exclude debris and large cell ag-gregates and examined for DNA fluorescence intensity dis-tribution. Cells in the DNA histogram with hypofluorescentDNA were designated apoptotic. All other cells distributedthemselves in a normal cell cycle profile.

Identification of apoptotic cells by flow cytometry

Identities of apoptotic thymocytes and B-lineage cells ofPeyer’s patch and bone marrow were determined by con-current phenotypic staining and cell cycle analysis. Forlabeling, 2 � 106 cells were washed with staining buffercontaining 1% (w/v) BSA and 0.1% (w/v) NaN3. The pelletwas resuspended in 50 �l of the staining reagent containingeither FITC-anti-CD4 and R- PE-anti-CD8a antibodies (0.5�g each) or Cy-C-anti-B220, FITC-anti-IgD, and R-PE-anti-IgM antibodies (0.5 �g each). Cells for negative con-trol were stained with FITC and R-PE or FITC- R-PE-, andCy-C-conjugated isotype-matched rat monoclonal antibod-ies with unrelated specificities, respectively. After a 45-minincubation at 4°C, the cells were then washed twice with thestaining buffer. The pellet was resuspended in 400 �l ofPBS with 50% heat-inactivated fetal bovine serum (FBS)and fixed by dropwise addition of 1.2 ml ice-cold 70% (v/v)ethanol with gentle mixing, and then held at 4°C overnight.Cells were washed and incubated in 1 ml of DNA stainingreagent (PBS containing 1 �g/ml DAPI, 50 �g/ml RNase A,and 0.1 mM EDTA disodium) at room temperature for 1 hand stored on ice until analysis. All UV lines 360–365 nmwere used to excite DAPI and the fluoroscence was detectedat 450 � 13 nm. FITC, R-PE, and Cy-C were excited withthe 488 nm line of an argon laser with emission detection at530 � 15, 570 � 13, and 670 � 10 nm, respectively. FITC-,R-PE-, or Cy-C-positive cells were examined for cell cycledistribution using a three-step gating procedure that elimi-nated cell debris, cell doublets, and nonspecific antibodybinding. When apoptotic cells were detected on T and B cellsubsets, a minimum of 1500 events was obtained for theleast prevalent population for each subset.

Identification of cell distribution by flow cytometry

Identities of thymocytes and B-lineage cells of Peyer’spatch and bone marrow were determined by phenotypicstaining and cell cycle analysis. For labeling, 2 � 106 cellswere washed with staining buffer containing 1% (w/v) BSAand 0.1% (w/v) NaN3. The pellet was resuspended in 50 �lof the staining reagent containing either FITC-anti-CD4 andR-PE-anti-CD8a antibodies (0.5 �g each) or Cy-C-anti-B220, FITC-anti-IgD, and R-PE-anti-IgM antibodies (0.5�g each). Cells for negative control were stained with FITCand R-PE or FITC-, R-PE- and Cy-C-conjugated isotype-matched rat monoclonal antibodies with unrelated specific-

71Z. Islam et al. / Toxicology and Applied Pharmacology 187 (2003) 69–79

ities, respectively. After a 45- min incubation at 4°C, cellswere washed twice with the staining buffer. The pellet wasresuspended in staining buffer with 2% methanol-free form-aldehyde (Polyscience, Warrington, PA). FITC, R-PE, andCy-C were excited with the 488-nm line of an argon laserwith emission detection at 530 � 15, 570 � 13, and 670 �10 nm, respectively. FITC-, R-PE-, or Cy- C-positive cellswere examined for cell cycle distribution using a three-stepgating procedure that eliminated cell debris, cell doublets,and nonspecific antibody binding. When cells were detectedon T and B cell subsets, a minimum of 1500 events wasobtained for the least prevalent population for each subset.

Statistics

Data were analyzed using Sigma Stat for Windows (Jan-del Scientific, San Rafael, CA). For determination of inter-active effects, values from vehicle-treated mice were sub-tracted from LPS, VT, or LPS � VT groups (Zhou et al.,2000). LPS and VT replicates were then randomly com-bined to calculate an expected mean additive response withvariance. This calculated value was compared to that foractual cotreated samples using the Mann–Whitney RankSum test. For the subset data, a Kruskal–Wallis one-wayANOVA on Ranks was performed. Data sets were consid-ered significantly different at p � 0.05.

Results

LPS and VT cotreatment induces apoptosis in thymus,Peyer’s patches, bone marrow, and spleen

Following treatment of mice with LPS and VT, apo-ptosis in thymus, Peyer’s patches, bone marrow, andspleen was assessed after 12 h by detecting hypodiploidnuclear content in permeabilized cells stained with PI.Distinct hypofluorescent peaks were observed in thymus,Peyer’s patches, bone marrow, and spleen of cotreatedmice while treatment with VH or either toxin alone hadvery little effect (Fig. 2). The percentage of apoptoticcells after exposure to VH, LPS, VT, and LPS � VTwere, respectively, 0.33, 0.44, 0.30, and 4.52% in thy-mus; 1.92, 1.64, 1.21, and 6.30% in Peyer’s patches;0.14, 1.16, 0.18, and 3.50% in bone marrow; and 0.2,0.36, 0.17, and 0.6% in spleen. The expected mean ad-ditive increases over control after exposure to the com-bined toxins were 0.16, 0.2, 0.92, and 0.16% for thymus,Peyer’s patches, bone marrow, and spleen, respectively.Actual experimental responses were 4.2 � 1, 4.38 � 1,3.36 � 0.13, and 0.37 � 0.1% for thymus, Peyer’s patch,bone marrow, and spleen, respectively; which were allsignificantly higher (p � 0.05) than expected additiveresponses for the first three tissues. These results dem-onstrate that coexposure to LPS � VT synergistically

increased apoptosis in thymus, Peyer’s patch, and bonemarrow.

LPS � VT-induced apoptosis and cell depletionin thymocyte subpopulations

The phenotype-specific effects of LPS � VT cotreat-ment on thymocytes were analyzed using combinationsof fluorescently labeled anti-CD4 and anti-CD8 antibod-ies in conjunction with DNA staining. Three color flowcytometric analysis was used to assess apoptotic celldeath in each of four phenotypically identifiable popula-tions within thymocytes as defined by expression ofCD4�CD8�, CD4�CD8�, CD4�CD8�, and CD4�CD8�

in B6C3F1 mice. Effects on total subpopulation numberswere analyzed by two-color flow cytometry.

A distinct increase of apoptosis in CD4�CD8� was de-tected in the LPS group whereas apoptosis was not detect-able in VT group (Fig. 3). However, significantly (p � 0.05)increased apoptosis at 12 h in CD4�CD8� (1.8-fold) wasobserved in LPS � VT treatment groups compared to meanadditive responses of the LPS and VT treatment groups.Two-color flow cytometry showed significant decreases inCD4�CD8� subset numbers in the LPS � VT group at 12and 24 h (Table 1) compared to the VH, LPS, and/or VTgroups. In terms of percentage distribution, these subsetnumbers did not change in any treatment group significantlyafter 12 and 24 h.

In CD4�CD8� cells, apoptosis was undetectable in LPSor VT groups at 12 h (Fig. 3). A significant (p � 0.05)

Fig. 2. Flow cytometric quantitation of apoptotic cells in thymus, Peyer’spatch, bone marrow, and spleen following treatment with LPS � VT. Micewere treated with vehicle (ip and po), LPS (0.1 mg/kg body wt ip) plusvehicle (po), VT (12.5 mg/kg body wt po) plus vehicle (ip) or LPS (0.1mg/kg body wt ip) plus VT (12.5 mg/kg body wt po). Cells from thymus,Peyer’s patches, bone marrow, and spleen were isolated 12 h later, stainedwith propidium iodide, and subjected to flow cytometric analysis. Datarepresent means � SEM (n � 4). Data are representative of three separateexperiments. *Significantly (p � 0.05) different from VH. **Significantlydifferent from mean additive responses.


increase (5.2-fold) in apoptosis was observed in the LPS �VT group at 12 h compared to mean additive responses ofLPS and VT. CD4�CD8� subset numbers in the LPS � VTgroup were also decreased compared to VH, LPS, and/orVT groups at 12 and 24 h, respectively (Table 1). Percent-age distribution of this subset also decreased significantly at12 and 24 h.

In the CD4�CD8� subset, neither increases of apoptosis(Fig. 3) nor effects on cell number (Table 1) were observed,but a significant increase in total percentage was found at 12and 24 h.

In CD4�CD8� cells, LPS alone induced a significantincrease of apoptosis but apoptosis was undetectable inthe VT group at 12 h. A significant (p � 0.05) increase(1.3-fold) in apoptosis at 12 h in the LPS � VT groupcompared to mean additive responses of LPS and VT(Fig. 3). However, no significant changes of total cellnumbers were detected (Table 1) in this subset at 12 and24 h. In terms of percentage distribution, CD4�CD8�

cells increased significantly in LPS � VT group com-pared to VH, LPS, and VT groups at both time points.

Taken together, these data indicate that immature thy-mocytes, including double-positive and double-negativepopulations, as well as mature cytotoxic CD4�CD8� cells,were highly susceptible to apoptosis induction by LPS �VT coexposure.

LPS � VT-induced apoptosis and cell depletionin Peyer’s patch B cell subpopulations

To determine B subsets in Peyer’s patches susceptibleto apoptosis following LPS � VT coexposure, cells weresubjected to flow cytometry following staining with fluo-rescently labeled antibodies and the DNA dye DAPI.Samples were first gated for the B220� cells and DNAcontent and then examined for IgM and IgD positivity.Four-color flow cytograms enabled assessment of toxineffects on apoptosis in the four subsets as defined byexpression of IgM�IgD�, IgM�IgD�, IgM�IgD�, andIgM�IgD�.

IgM�IgD� cells from LPS � VT-treated animals orany other groups did not exhibit apoptotic peaks (Fig. 4).The total cell numbers in IgM�IgD� cells decreasedsignificantly in LPS, VT, and LPS � VT groups com-pared to VH at 12 h but, at 24 h, numbers of this subsetdecreased significantly in VT and LPS � VT groups

Fig. 3. Percentage apoptosis of CD4�CD8�, CD4�CD8�, CD4�CD8�,and CD4�CD8� thymus T cells at 12 h following treatment with LPS �VT. Mice were treated as described in the legend to Fig. 1. Thymocyteswere surface stained for CD4 and CD8, DNA stained with DAPI, andanalyzed by three-color flow cytometry. Data (mean � SEM) were pooledfrom four separate experiments (n � 12–16 mice). *Significantly (p �0.05) different from VH. **Significantly different from mean additiveresponses.

Table 1Distribution of CD4�CD8�, CD4�CD8�, CD4�CD8�, and CD4�CD8� thymus T cells at 12 and 24 h following treatment with LPS � VT

Time Group CD4�CD8� CD4�CD8� CD4�CD8� CD4�CD8�

Total cells(�105)

Percent total Total cells(�105)



Percent total

12 h VH 19.8 � 1.1 1.9 � 0.1 931.5 � 51.9 87.6 � 0.6 64.2 � 3.6 6.0 � 0.5 42.0 � 2.3 3.9 � 0.2LPS 16.7 � 0.6 2.1 � 0.3 627.7 � 24.0* 83.1 � 1.6 68.2 � 2.6 9.0 � 0.8 40.0 � 1.5 5.3 � 0.4VT 15.3 � 1.3 1.8 � 0.0 716.9 � 61.0* 85.0 � 1.1 59.8 � 5.1 7.1 � 1.0 46.5 � 3.9 5.5 � 0.1LPS � VT 9.4 � 1.4*†‡ 2.3 � 0.2 253.6 � 36.6*†‡ 63.4 � 5.9*†‡ 82.3 � 9.9 20.6 � 4.9*†‡ 46.4 � 6.7 13.2 � 1.2*†‡

24 h VH 22.1 � 1.1 2.0 � 0.1 937.3 � 46.2 85.4 � 0.7 95.9 � 4.7 8.7 � 0.5 8.8 � 0.4 0.82 � 0.1LPS 16.2 � 0.6 2.1 � 0.1 623.2 � 23.9* 82.5 � 1.0 82.5 � 3.2 10.9 � 0.8 7.9 � 0.3 1.05 � 0.1VT 18.4 � 1.2 2.2 � 0.2 697.8 � 46.0* 82.6 � 0.5 93.0 � 6.1 11.0 � 0.5 10.1 � 0.7 1.19 � 0.1LPS � VT 7.5 � 1.9*†‡ 3.3 � 0.3 120.2 � 31.8*†‡ 52.6 � 2.8*†‡ 79.3 � 20.9 34.7 � 2.0*†‡ 6.6 � 0.5 5.03 � 0.4*†‡

Note. Mice were treated as described in the legend to Fig. 2. At 12 h, cells were surfaced stained for CD4 and CD8, fixed with ethanol, DNA stained withDAPI, and analyzed by three-color flow cytometry for the measurement total and percentage of cell subset numbers. At 24 h, cells were surfaced stained forCD4 and CD8, fixed with 2% formaldehyde, and analyzed by two-color flow cytometry for the measurement of subset numbers. Data (mean � SEM) werepooled from four separate experiments (n � 12–16 mice).

* Significantly different compared to VH, p � 0.05.† Significantly different compared to LPS, p � 0.05.‡ Significantly different compared to VT, p � 0.05.


compared to VH or LPS (Table 2). The percentage dis-tribution of IgM�IgD� cells decreased significantly inVT or LPS � VT groups compared to VH or LPS groupsonly at 24 h.

IgM�IgD� cells also did not exhibit altered apoptosis at12 h (Fig. 4) but total cell numbers in LPS � VT groupdecreased significantly compared to VH, LPS, and VT at24 h (Table 2). Total cell numbers also decreased in the LPSgroup compared to the VH group. The percentage distribu-tion was not changed at 12 and 24 h.

LPS or VT alone had no effects on the induction ofapoptosis in IgM�IgD� from LPS- or VT-treated mice at

12 h (Fig. 4). Significant (p � 0.05) apoptosis induction(5.2-fold) by LPS � VT at 12 h was observed in theIgM�IgD� (9%) cells compared to mean additive responsesof LPS and VT. IgM�IgD� subset numbers did not changein the LPS � VT group compared to the VH, LPS, and VTgroup at 12 h whereas a significant decrease of total subsetnumbers was observed at 24 h compared to VH control(Table 2). Percentages of B220� cells were unaffected byany treatment.

In IgM�IgD� cells, changes in apoptosis at 12 h or cellnumbers at 12 and 24 h were not detected in any of thegroups. Total B220� cells were also recorded in all thegroups at 12 and 24 h but significant changes were found inPeyer’s patches. Overall, the IgM�IgD� cells were sensi-tive to LPS � VT-induced apoptosis whereas the toxincombination did not change this subset’s cell numbers sig-nificantly at 12 and 24 h.

Total B220� cell numbers were not affected significantlyat 12 and 24 h.

LPS � VT-induced apoptosis and cell depletion in bonemarrow B cell subpopulations

To determine which B cell subsets in bone marrowundergo apoptosis during LPS � VT treatment, cellswere analyzed as described for Peyer’s patch cells. LPSalone significantly induced apoptosis after 12 h but VThad no effects on IgM�IgD� cells (Fig. 5). Apoptosiswas increased in IgM�IgD� cells significantly (p � 0.05)in LPS � VT-treated mice at 12 h compared to meanadditive responses of LPS and VT. IgM�IgD� subsetnumbers decreased significantly in the LPS, VT, and LPS� VT groups in relation to VH at 12 h. In terms ofpercentage distribution, IgM�IgD� cells were signifi-cantly decreased in the LPS � VT group compared to the

Fig. 4. Percentage apoptosis of IgM�IgD�, IgM�IgD�, IgM�IgD�, andIgM�IgD� Peyer’s patch B cells at 12 h following treatment with LPS �VT. Peyer’s patches from animals described in Fig. 2 were surface stainedfor B220, IgD, and IgM, DNA stained with DAPI, and analyzed byfour-color flow cytometry. Data were (mean � SEM, n � 4 mice) repre-sentative of four separate experiments. **Significantly (p � 0.05) differentfrom mean additive responses.

Table 2Distribution of IgM�IgD�, IgM�IgD�, IgM�IgD�, and IgM�IgD� Peyer’s patch B cells at 12 and 24 h following treatment with LPS � VT

Time Group IgM�IgD� IgM�IgD� IgM�IgD� IgM�IgD� Total B220�

(�104)Total cells(� 104)

PercentB220�

Total cells(�104)

PercentB220�

Total cells(�104)

PercentB220�

Total cells(�104)

PercentB220�

12 h VH 97.1 � 26.7 19.6 � 1.8 8.4 � 2.3 1.7 � 0.2 40.3 � 6.5 11.0 � 1.4 320.0 � 87.9 64.7 � 3.1 494.4 � 35.8LPS 35.8 � 9.5* 18.0 � 1.3 2.6 � 0.7* 1.3 � 0.2 24.3 � 6.3 9.7 � 0.5 128.9 � 34.2 65.0 � 3.0 198.3 � 52.7*VT 41.4 � 7.9* 10.1 � 0.3 6.2 � 1.2 1.5 � 0.1 49.7 � 9.4 12.1 � 1.2 316.7 � 60.1 76.8 � 1.0 412.3 � 78.3LPS � VT 38.2 � 6.5* 9.6 � 1.0 7.8 � 1.3 1.9 � 0.2 42.0 � 7.2 10.5 � 0.5 303.4 � 52.1 75.9 � 1.1 357.2 � 61.3

24 h VH 171.4 � 10.49 10.1 � 0.8 29.75 � 1.8 1.8 � 0.2 38.0 � 2.3 22.5 � 2.2 1107.9 � 67.8 65.6 � 1.7 1242.0 � 301.0LPS 153.0 � 11.2 13.8 � 1.1 26.5 � 1.9 2.4 � 0.4 30.2 � 2.2 27.3 � 2.1 634.4 � 46.4 57.3 � 1.9 753.1 � 102.5VT 91.1 � 1.6*† 6.0 � 0.6*† 16.7 � 0.3 1.1 � 0.1 28.0 � 0.5 18.6 � 1.7 1115.7 � 19.7 74.0 � 1.6 1125.8 � 355.5LPS � VT 55.8 � 10.6*† 5.0 � 0.3*† 8.4 � 1.6*†‡ 0.8 � 0.1 24.3 � 4.6* 21.9 � 2.4 791.2 � 151.3 71.2 � 2.7 789.7 � 200.1

Note. Mice were treated as described in the legend to Fig. 2. At 12 h, cells were surfaced stained for B220, IgM, and IgD, fixed with ethanol, DNA stainedwith DAPI, and analyzed by four-color flow cytometry for the measurement of total and percentage of cell subset numbers. At 24 h, cells were surfacedstained for B220, IgM, and IgD, fixed with 2% formaldehyde, and analyzed by three-color flow cytometry for the measurement of subset numbers. Data (mean� SEM) were pooled from four separate experiments (n � 12–16 mice).



VH, LPS, or VT groups at 12 h (Table 3). At 24 h,IgM�IgD� subset numbers decreased significantly in theLPS � VT group compared to the VH, LPS, and VTgroups. IgM�IgD� subset numbers also decreased sig-nificantly in the LPS or VT groups compared to the VHgroup. The percentage distribution of this subset in VTand LPS � VT groups significantly suppressed comparedto VH and LPS groups.

Apoptosis induction in IgM�IgD� cells was undetect-able in any of the treatment animals. Total IgM�IgD�

numbers were not affected at 12 h, however, the numbers

decreased significantly in the LPS � VT group compared tothe VH and VT groups at 24 h (Table 3). The percentagedistribution in the VT group decreased significantly at 12 hcompared to the VH group.

The IgM�IgD� subset was the most sensitive to LPS �VT coexposure (Fig. 5), with the combination group havingtwice the apoptotic population of that predicted for meanadditive responses of LPS and VT. LPS alone significantlyinduced apoptosis but VT had no effects in this subset. Thedecrease in the IgM�IgD� subset numbers was also signif-icant in the LPS � VT group in relation to the VH, LPS, andVT groups at 12 and 24 h, respectively (Table 3). Relativeto percentage distribution, LPS, VT, and LPS � VT at 12and 24 h were significantly higher than the VH group. At24 h, the percentage distributions in VT and LPS � VTgroups were significantly higher than VH or LPS groups.

A slight induction of apoptosis was detected inIgM�IgD� cells in the LPS or LPS � VT groups comparedto the VH group but no apoptosis was detected in the VTgroup (Fig. 5). IgM�IgD� cell numbers at 12 and 24 hsignificantly decreased in the LPS � VT group compared tothe VH, LPS or VT groups. LPS alone also caused thesignificant decrease of this subset’s all numbers at 24 hcompared to VH or VT groups. However, the percentagedistribution increased in the LPS � VT group significantlyat 12 h compared to the VH, LPS or VT groups. Thepercentage distributions in the LPS or VT groups at 12 hand the VT group at 24 h also increased significantly com-pared to the VH group (Table 3).

Total B220� cells were markedly decreased in LPS, VT,and LPS � VT groups compared to VH after 12 and 24 h(Table 3). The greatest loss was detected at 24 h in the LPS� VT group, which was significantly greater than the com-bined additive effects of LPS and VT.

Fig. 5. Percentage apoptosis of IgM�IgD�, IgM�IgD�, IgM�IgD�, andIgM�IgD� bone marrow B cells at 12 h following treatment with LPS �VT. Cells from bone marrow were surface stained for B220, IgD, and IgM,DNA stained with DAPI, and analyzed by four-color flow cytometry. Datawere (mean � SEM, n � 4 mice) representative of four separate experi-ments. *Significantly (p � 0.05) different from VH. **Significantly dif-ferent from mean additive responses.

Table 3Distribution of IgM�IgD�, IgM�IgD�, IgM�IgD�, and IgM�IgD� bone marrow B cells at 12 and 24 h following treatment with LPS � VT

Time Group IgM�IgD� IgM�IgD� IgM�IgD� IgM�IgD� Total B220�

(�104)Total cells(�104)

PercentB220�

Total cells(�104)

PercentB220�

Total cells(�104)

PercentB220�

Total cells(�104)

PercentB220�

12 h VH 93.8 � 17.4 66.1 � 4.3 13.7 � 2.5 11.9 � 1.0 5.6 � 0.1 3.8 � 0.3 12.7 � 1.2 9.2 � 0.5 129.2 � 23.9LPS 37.0 � 9.9* 57.5 � 9.0 7.4 � 1.9 9.5 � 0.8 4.5 � 0.1 7.7 � 0.2* 8.0 � 1.2 13.8 � 0.2* 52.8 � 14.1*VT 42.5 � 8.9* 62.2 � 11.8 10.4 � 2.2 5.5 � 1.6*† 4.5 � 0.6 7.9 � 0.8* 9.2 � 1.5 14.3 � 0.2* 70.3 � 14.7*LPS � VT 25.4 � 5.7* 46.7 � 1.7*†‡ 6.3 � 1.4 14.2 � 0.3 2.3 � 0.5*†‡ 9.6 � 0.8* 4.2 � 0.4*†‡ 19.8 � 1.4*†‡ 40.3 � 9.3*‡

24 h VH 71.8 � 11.4 64.5 � 11.2 17.6 � 2.8 15.8 � 0.7 8.5 � 1.3 7.7 � 0.7 13.8 � 2.2 12.4 � 1.0 120.5 � 11.5LPS 41.3 � 4.2* 64.4 � 17.3 8.9 � 0.9* 13.8 � 0.9 6.9 � 0.7 10.9 � 0.9* 7.1 � 0.7*† 11.1 � 1.0 74.5 � 12.4*VT 48.2 � 3.7* 50.0 � 11.3*† 14.1 � 1.1 14.6 � 1.5 15.4 � 1.2 16.1 � 1.7*† 16.7 � 1.3 17.4 � 1.5*† 90.1 � 20.3LPS � VT 10.9 � 3.4*†‡ 45.2 � 2.6*† 5.1 � 1.6*‡ 21.3 � 0.9 4.8 � 1.4*‡ 19.8 � 2.4*† 3.2 � 1.0*†‡ 13.5 � 1.8 25.0 � 4.8*†‡

Note. Mice were treated as described in the legend to Fig. 2. At 12 h, cells were surfaced stained for B220, IgM, and IgD, fixed with ethanol, DNA stainedwith DAPI, and analyzed by four-color flow cytometry for the measurement of total and percentage of cell subset numbers. At 24 h, cells were surfacedstained for B220, IgM, and IgD, fixed with 2% formaldehyde, and analyzed by three-color flow cytometry for the measurement of subset numbers. Data (mean� SEM) were pooled from four separate experiments (n � 12–16 mice).



RU 486 protects against LPS � VT-induced apoptoticcell death in thymic immature (CD4�CD8� andCD4�CD8�) and cytotoxic T lymphocytes (CD4�CD8�)

Since RU 486, a glucocorticoid receptor antagonist,was previously found to be protective against LPS �VT-induced thymic apoptosis in vivo (Islam et al., 2002),the effect of this antagonist was examined on LPS �VT-induced apoptosis in thymocyte subsets defined bythe expression of CD4�CD8�, CD4�CD8�, CD4�CD8�,and CD4�CD8�. RU 486 (20 mg/kg ip) was injected 30min prior to LPS � VT administration. Apoptosis wasmeasured quantitatively by three-color flow cytometrywith fluorescently labeled anti-CD4 and anti-CD8 anti-bodies and the DNA staining dye DAPI (Fig. 6). Theresults indicated that there was significant (p � 0.05)attenuation in apoptosis by RU 486 in CD4�CD8� (5.6-fold), CD4�CD8� (1.5-fold), and CD4� CD8� (3-fold)subsets in relation to mice treated with the LPS � VTgroup whereas the antagonist alone had no effect in thosesubsets. These data suggest that the capacity of LPS �VT to synergistically induce apoptosis in both immatureand mature cytotoxic T cells was mediated by glucocor-ticoids.

RU 486 protects against LPS � VT-induced apoptoticcell death in Peyer’s patch mature (IgM�IgD�) B cells

The effect of the antagonist RU 486 was also exam-ined on LPS � VT-induced apoptosis in Peyer’s patch Bcell subsets. The results indicated that there was signif-icant (p � 0.05) attenuation (3.3-fold) in apoptosis by RU486 in the IgM�IgD� compared to the LPS � VT group,

whereas the antagonist alone had no effect on this subset(Fig. 7). These data indicate LPS � VT- induced apo-ptosis in Peyer’s patch B cells was mediated by glucocor-ticoids.

RU 486 protects against LPS � VT-induced apoptoticcell death in bone marrow mature (IgM�IgD�)B cells at 12 h

The effect of the antagonist RU 486 on LPS � VT-induced apoptosis in bone marrow B cell subsets was alsoevaluated (Fig. 8). The results indicate that there was partial

Fig. 6. Effects of RU 486 on apoptosis of CD4�CD8�, CD4�CD8�,CD4�CD8�, and CD4�CD8� thymus T cells at 12 h following treatmentwith LPS � VT. Thymocytes were surface stained for CD4 and CD8, DNAstained with DAPI, and analyzed by three-color flow cytometry. Data(mean � SEM) were pooled from three separate experiments (n � 9–12mice). *Significant (p � 0.05) increase compared to VH. §Significantdecrease in apoptotic cells compared to LPS � VT-treated mice.

Fig. 7. Effects of RU 486 on apoptosis of IgM�IgD�, IgM�IgD�,IgM�IgD�, and IgM�IgD� Peyer’s patch B cells at 12 h followingtreatment with LPS � VT. Cells from Peyer’s patches were surface stainedfor B220, IgD, and IgM, DNA stained with DAPI, and analyzed by four-color flow cytometry. Data (mean � SEM) were pooled from three sepa-rate experiments (n � 9–12 mice). *Significant (p � 0.05) increase againstVH. §Significant suppression of apoptotic cells against LPS � VT-treatedmice.

Fig. 8. Effects of RU 486 of IgM�IgD�, IgM�IgD�, IgM�IgD�, andIgM�IgD� bone marrow B cells at 12 h following treatment with LPS �VT. Cells from bone marrow were surface stained for B220, IgD, and IgM,DNA stained with DAPI, and analyzed by four-color flow cytometry. Data(mean � SEM) were pooled from three separate experiments (n � 9–12mice). *Significant (p � 0.05) increase against VH. §Significant suppres-sion of apoptotic cells against LPS � VT-treated mice.


but significant (p � 0.05) attenuation in apoptosis by RU486 in the IgM�IgD� and IgM�IgD� subset compared tothe LPS � VT group, while the antagonist alone had noeffect on this subset. These data verify that the capacity ofLPS � VT to induce apoptosis in bone marrow B cells wasmediated partly by glucocorticoids.

Discussion

Phenotypic analysis of lymphocyte subsets is a usefulstrategy for elucidating selective effects and mechanisms ofimmunotoxic chemicals as well as interactive effects ofmultiple chemicals. The immune system is highly sensitiveto trichothecene mycotoxins (Bondy and Pestka 2000; Islamet al., 1998; Nagata et al., 2001; Ueno 1987) particularly inconjunction with a subtoxic doses of LPS (Tai and Pestka,1988; Taylor et al., 1991). Lymphoid organs are majortargets of apoptotic cell death by the cotreatment of LPS andVT in vivo in B6C3F1 mice (Islam et al., 2002; Zhou et al.,2000). The 12.5 mg/kg VT dose represents approximately1/6 of the LD50 for VT (Ueno 1987) whereas the 0.1 mg/kgLPS dose is 1/25 of the dose required for endotoxic shock inmice (Zhang et al., 1993). These doses of LPS or VT alonegenerally had little to no effect on any of the subsets in theseorgans but combination of these toxins synergistically in-duced apoptosis. Here, LPS � VT coexposure caused themarked induction of apoptosis in the lymphoid organs after12 h and the major subsets undergoing apoptosis generallycorrelated with the loss of T and B cells at 12 and 24 h. Aspredicted in our hypothesis, immature lymphocyte popula-tions were highly susceptible to the combined effects of thetoxins. However, in some cases, mature lymphocytes werealso susceptible to the two toxins.

T cells normally undergo apoptosis during maturation inthe thymus (Kamath et al., 1998). CD4 and CD8 surfacemolecules are critical to positive and negative selection ofthymocytes. Double-negative thymocytes are considered tobe prothymocytes (Thurmond and Gasiewicz, 2000)whereas double-positive cells (prethymocytes) are consid-ered to be the intermediate cells that give rise to the func-tional single-positive cells (Blue et al., 1987) (Fig. 1). Thetrichothecene T-2 toxin induces thymic atrophy at hightoxin doses through the loss of lymphoid progenitor cells(Holladay et al., 1993) and apoptotic depletion is observablein CD4�CD8� and CD4� single-positive cells (Islam et al.1998). Nagata et al. (2001) recently reported that oral ex-posure to T-2 toxin caused apoptosis in thymus, mesentericlymph nodes, and Peyer’s patches of mice. Depression wasnoted in CD4�CD8�, CD4�, and CD8� cells in thymus, aswell as CD4� and CD8� cells in mesenteric lymph nodesand Peyer’s patches.

Our results suggest that coexposure of LPS and VTinduced glucocorticoid-dependent apoptotic loss ofCD4�CD8�, CD4�CD8�, and CD8� cells in thymus, ma-ture-B lymphocytes in Peyer’s patch, and pro/pre-B lym-

phocytes and mature-B lymphocytes in bone marrow inmice. The CD4�CD8� pre-T cell has minimal expression ofBcl-2 and Bcl-XL, leaving the cells particularly prone toapoptotic cell death in the presence of corticosterone (Ash-well et al., 2000). The CD4�CD8� subset accounts forapproximately 85% of the total cells in the thymus (Winotoand Littman, 2002). Although these were greatly depletedfollowing LPS � VT coexposure, apoptosis increased inthis subset by only 2%. The low levels of apoptosis in ourstudy could have occurred due to rapid elimination of thissubset from thymus by phagocytosis prior to detection. Insupport of this contention, it has been previously shown thatglucocorticoid-driven apoptosis is enhanced 50–300%among these subsets when cells are collected from treatedanimals and subsequently cultured for 8 h prior to detectionof apoptosis (King et al. 2002). Spleen cells contain primar-ily mature T and B cells and their membrane composition islikely to be different from the early mature cells found inthymus and bone marrow. Although polyclonal CD4�CD8�

and CD4�CD8� cells may have different survival signals,bcl-2 protein expression on naive CD4�CD8� andCD4�CD8� cells is similar (Ferreira et al. 2000). In someinstances, the percentage apoptosis in CD4�CD8� cells issignificantly higher than CD4�CD8� cells (Delogu et al.,2000). Under the conditions employed herein, it is notknown specifically why the percentage apoptosis inCD4�CD8� cells is higher than CD4�CD8� cells.

B lymphocytes are produced in specialized microenvi-ronments provided by the fetal liver and adult bone marrowof mammals. Cell surface molecules defining early devel-opmental stages of B cells and their progenitors includeB220, CD43, CD24, and BP-1 (Burrows and Cooper, 1997).Expansion of B cells from progenitor cells begins with thepro-B cell (B220�IgM�) and ends with many plasma celland memory B cell progeny (B220�IgM�IgD�) (Fig. 1). Blymphocytes undergo genotypic changes by rearranging V,D, and J gene segments to encode functional antigen recep-tors. The VH gene rearrangement process is initiated inpro-B cells. Completion of productive VH rearrangementresults in the pre-B cell. Subsequent rearrangement permitscell surface expression of conventional IgM receptors on theimmature B cell (B220�IgM�IgD�), which exits the bonemarrow and migrates to the periphery. B220�IgM�IgD�

cells occur prior to plasma cells and thus may be consideredto be an early mature or transitional stage (Carsetti et al.,1995). Previous studies have revealed that T-2 toxin targetsB lymphocyte precursor cells (Holladay et al., 1995) andIgA� cells in Peyer’s patches and spleen (Nagata et al.,2001; Pestka et al., 1994). The results presented hereinsuggest that LPS � VT synergistically induced apoptoticcell death in IgM�IgD�, IgM�IgD� subsets in bone mar-row and IgM�IgD� subsets in Peyer’s patches. LPS � VTcontributed to massive loss of all of the subsets in bonemarrow and Peyer’s patch, indicating that LPS � VT mightbe toxic to the precursor B cells and transitional B cells.Developing B cells in vivo become sensitive to tolerance


induction when they are in transition (IgM�IgD�) from theimmature to the mature stage (Carsetti et al. 1995). Most ofthe transitional B cells, but only few of the mature and of theimmature B cells, express Fas antigen; however, mature Bcells, but not immature and transitional B cells, expressbcl-2 protein. The presence of the Fas antigen and theabsence of bcl-2 protein is compatible with the high sensi-tivity of transitional B cells to negative selection. Thesetransitional B cells might be the target of LPS � VT-induced corticosterone and thus contribute to immune sup-pression. It is generally accepted that the mIgD class ofreceptors on murine resting B cells serves as the principalsensor for antigen (Chaturvedi et al., 2002). This subset wasconstitutively localized to cell membrane raft microdo-mains. Its stimulation resulted in the activation of signalingpathways, which integrated with signals derived from thetransmembrane mIgD receptors. Therefore, depletion ofIgD-B cell subset by LPS � VT may modulate the extent ofstrength of the antigenic stumulus.

It is now well known that CD4�CD8� thymocytes (Ash-well et al., 2000) and pro/pre and immature bone marrow Bcells (Laakko and Fraker, 2002) are highly sensitive toapoptosis by corticosteroids. In our previous study, we haveshown that LPS � VT-induction of generalized lymphoidapoptosis was corticosteroid-dependent (Islam et al., 2002).Consistent with these findings, RU 486 protected againstapoptosis in all sensitive subsets in thymus, Peyer’s patches,and bone marrow.

To summarize, the results of this study demonstrate thatLPS � VT synergistically induced apoptosis in prothymo-cyte (CD4�CD8�), immature thymocyte (CD4�CD8�),and mature thymocyte (CD8�) populations in thymus witha concurrent loss of cell numbers. In Peyer’s patches andbone marrow, B220�IgM�IgD� pro/pre B cells andB220�IgM�IgD� mature B cells underwent apoptosis at12 h after toxin treatments. Since RU 486 blocked apoptosisin general, additional studies will be necessary to furtherdefine how LPS and VT upregulate corticosterone in vivo.Although LPS or VT alone at the low doses used in ourmodel elicit a modest glucocorticoid response, it is gener-ally not sufficient for induction of lymphoid apoptosis (Is-lam et al., 2002). The mechanism by which LPS and VTsynergistically increase and prolong the corticosterone re-sponse is unclear but may involve increased stimulation ofthe hypothalamic–pituitary–adrenal axis. LPS and VT alsosynergistically induce several proinflammatory cytokines(Zhou et al., 1999). These may play critical roles in trig-gering glucocorticoid or increasing the half-life of glucocor-ticoid, thereby driving lymphocyte apoptosis. In a previousstudy, however, we determined that TNF-� and IL-6 werenot requisites for the interactive effects observed betweenthese two toxins in apoptosis or corticosterone induction(Islam et al., 2002). It is possible that IL-1� may play thisrole either by action on the hypothalamus or by directbinding to the adrenal gland, which contains IL-1� recep-tors. These latter possibilities warrant further study.

Acknowledgments

This work was supported by Public Health Services GrantsES 09521 (J.J.P.) and ES 03358 (J.J.P.) and from the NationalInstitute for Environmental Health Sciences and the MichiganState University Agricultural Experiment Station. We thank Dr.Hui-Ren Zhou for assistance and useful advice during this project.

References

Ashwell, J.D., Lu, F.W., Vacchio, M.S., 2000. Glucocorticoids in T celldevelopment and function. Annu. Rev. Immunol. 18, 309–345.

Blue, M.L., Daley, J.F., Levine, H., Craig, K.A., Schlossman, S.F., 1987.Identification and isolation of a T4�T8� cell with high T3 expressionin human thymus: a possible late intermediate in thymocyte differen-tiation. J. Immunol. 139, 1065–1069.

Bondy, G.S., Pestka, J.J., 2000. Immunomodulation by fungal toxins. J.Toxicol. Environ. Health B Crit. Rev. 3, 109–143.

Brunetti, M., Martelli, N., Colasante, A., Piantelli, M., Musiani, P., Aiello,F.B., 1995. Spontaneous and glucocorticoid-induced apoptosis in hu-man mature T lymphocytes. Blood 86, 4199–4205.

Burrows, P.D., Cooper, M.D., 1997. B cell development and differentia-tion. Curr. Opin. Immunol. 9, 239–244.

Carsetti, R., Kohler, G., Lamers, M.C., 1995. Transitional B cells are thetarget of negative selection in the B cell compartment. J. Exp. Med.181, 2129–2140.

Chaturvedi, A., Siddiqui, Z., Bayiroglu, F., Rao, K.V., 2002. A GPI-linkedisoform of the IgD receptor regulates resting B cell activation. Nat.Immunol. 3, 951–957.

Corrier, D.E., Wagner, G.G., 1988. Comparison of the effect of T-2 toxinwith that of dexamethasone or cyclophosphamide on resistance toBabesia microti infection in mice. Am. J. Vet. Res. 49, 2000–2003.

Delogu, G., Moretti, S., Antonucci, A., Marcellini, S., Masciangelo, R.,Famularo, G., Signore, L., De Simone, C., 2000. Apoptosis and surgi-cal trauma: dysregulated expression of death and survival factors onperipheral lymphocytes. Arch. Surg. 135, 1141–1147.

Ferreira, C., Barthlott, T., Garcia, S., Zamoyska, R., Stockinger, B., 2000.Differential survival of naive CD4 and CD8 T cells. J. Immunol. 165,3689–3694.

Hewett, J.A., Roth, R.A., 1993. Hepatic and extrahepatic pathobiology ofbacterial lipopolysaccharides. Pharmacol. Rev. 45, 382–411.

Holladay, S.D., Blaylock, B.L., Comment, C.E., Heindel, J.J., Luster, M.I.,1993. Fetal thymic atrophy after exposure to T-2 toxin: selectivity forlymphoid progenitor cells. Toxicol. Appl. Pharmacol. 121, 8–14.

Holladay, S.D., Smith, B.J., Luster, M.I., 1995. B lymphocyte precursorcells represent sensitive targets of T2 mycotoxin exposure. Toxicol.Appl. Pharmacol. 131, 309–315.

Islam, Z., Moon, Y.S., Zhou, H.R., King, L.E., Fraker, P.J., Pestka, J.J.,2002. Endotoxin potentiation of trichothecene-induced lymphocyteapoptosis is mediated by up-regulation of glucocorticoids. Toxicol.Appl. Pharmacol. 180, 43–55.

Islam, Z., Nagase, M., Yoshizawa, T., Yamauchi, K., Sakato, N., 1998. T-2toxin induces thymic apoptosis in vivo in mice. Toxicol. Appl. Phar-macol. 148, 205–214.

Kamath, A.B., Nagarkatti, P.S., Nagarkatti, M., 1998. Characterization ofphenotypic alterations induced by 2,3,7,8-tetrachlorodibenzo-p-dioxinon thymocytes in vivo and its effect on apoptosis. Toxicol. Appl.Pharmacol. 150, 117–124.

King, L.E., Osati-Ashtiani, F., Fraker, P.J., 2002. Apoptosis plays a distinctrole in the loss of precursor lymphocytes during zinc deficiency inmice. J. Nutr. 132, 974–979.

Laakko, T., Fraker, P., 2002. Rapid changes in the lymphopoietic andgranulopoietic compartments of the marrow caused by stress levels ofcorticosterone. Immunology 105, 111–119.


Miglioratti, G., Nicoletti, I., Nocentini, G., Pagliacci, M.C., Ricardi, C.,1994. Dexamethasone and interleukins modulate T-lymphocytes. Phar-macol. Res. 30, 43–52.

Nagata, T., Suzuki, H., Ishigami, N., Shinozuka, J., Uetsuka, K., Na-kayama, H., Doi, K., 2001. Development of apoptosis and changes inlymphocyte subsets in thymus, mesenteric lymph nodes and Peyer’spatches of mice orally inoculated with T-2 toxin. Exp. Toxicol. Pathol.53, 309–315.

Pestka, J.J., Yan, D., King, L.E., 1994. Flow cytometric analysis of theeffects of in vitro exposure to vomitoxin (deoxynivalenol) on apoptosisin murine T, B and IgA� cells. Food Chem. Toxicol. 32, 1125–1136.

Rotter, B.A., Prelusky, D.B., Pestka, J.J., 1996. Toxicology of deoxyni-valenol (vomitoxin). J. Toxicol. Environ. Health 48, 1–34.

Rumbeiha, W.K., Fitzgerald, S.D., Braselton, W.E., Roth, R.A., Pestka,J.J., Kaneene, J.B., 2000. Augmentation of mercury-induced nephro-toxicity by endotoxin in the mouse. Toxicology 151, 103–111.

Tai, J.H., Pestka, J.J., 1988. Synergistic interaction between the trichoth-ecene T-2 toxin and Salmonella typhimurium lipopolysaccharide inC3H/HeN and C3H/HeJ mice. Toxicol. Lett. 44, 191–200.

Taylor, M.J., Lafarge-Frayssinet, C., Luster, M.I., Frayssinet, C., 1991.Increased endotoxin sensitivity following T-2 toxin treatment is asso-ciated with increased absorption of endotoxin. Toxicol. Appl. Pharma-col. 109, 51–59.

Thurmond, T.S., Gasiewicz, T.A., 2000. A single dose of 2,3,7,8-tetrachlo-rodibenzo-p-dioxin produces a time- and dose-dependent alteration inthe murine bone marrow B-lymphocyte maturation profile. Toxicol.Sci. 58, 88–95.

Ueno, Y., 1987. Trichothecenes in food, in: Korgh, P. (Ed.), Mycotoxins inFood, Academic Press, New York, pp. 123–147.

Winoto, A., Littman, D.R., 2002. Nuclear hormone receptors in T lym-phocytes. Cell 109 (Suppl.), S57–S66.

Yee, S.B., Kinser, S., Hill, D.A., Barton, C.C., Hotchkiss, J.A., Harkema,J.R., Ganey, P.E., Roth, R.A., 2000. Synergistic hepatotoxicity fromcoexposure to bacterial endotoxin and the pyrrolizidine alkaloid mono-crotaline. Toxicol. Appl. Pharmacol. 166, 173–185.

Zhang, Y.H., Takahashi, K., Jiang, G.Z., Kawai, M., Fukada, M., andYokochi, T., 1993. In vivo induction of apoptosis (programmed celldeath) in mouse thymus by administration of lipopolysaccharide. In-fect. Immun. 61, 5044–5048.

Zhou, H.R., Harkema, J.R., Hotchkiss, J.A., Yan, D., Roth, R.A., andPestka, J.J., 2000. Lipopolysaccharide and the trichothecene vomitoxin(deoxynivalenol) synergistically induce apoptosis in murine lymphoidorgans. Toxicol. Sci. 53, 253–263.

Zhou, H.R., Harkema, J.R., Yan, D., and Pestka, J.J., 1999. Amplifiedproinflammatory cytokine expression and toxicity in mice co-exposedto lipopolysaccharide and the trichothecene vomitoxin (deoxynivale-nol). J. Toxicol. Environ. Health 57, 115–136.


Documents

Differential induction of glucocorticoid-dependent apoptosis in murine lymphoid subpopulations in vivo following coexposure to lipopolysaccharide and vomitoxin (deoxynivalenol)